Vous êtes sur la page 1sur 736

COMPLETE SOLUTIONS

MANUAL FOR

ZILL'S

A FIRST COURSE IN

DIFFERENTIAL
EQUATIONS
WITH MODELlNG ApPLlCATIONS

7TH EOITION

ANO
ZILL& CULLEN'S

DIFFERENTIAL
EQUATIONS
WITH BOUNDARyVALUE PROBLEMS

5TH EOITION

BROOKS/COLE
THOMSON

LEARNING

Australia Canada Mexico Singapore Spain United Kingdom United States

Table of Contents
1

Introduction to Differential Equations

First-Order Differential Equations

22

Modeling with First-Order Differential Equations

71

Higher-Order Differential Equations

104

Modeling with Higher-Order Differential Equations

194

Series Solutions of Linear Equations

240

The Laplace Transform

308

Systems of Linear First-Order Differential Equations

370

Numerical Solutions of Ordinary Differential Equations

430

10

Plane Autonomous Systems and Stability

458

11

Orthogonal Functions and Pourier Series

491

12

Partial Dfferential Equations and Boundary-Value


Problems in Rectangular Coordinates

538

13

Boundary-Value Problems in Other Coordinate Systems

616

14

Integral Transform Method

654

15

Numerical Solutions of Partial Differential Equations

695

Gamma function

717

Introduction to Matrices

718

Appendix 1
Appendix II

Introduction to Differential Equations


Exercises 1.1

1. Second-order ; linear.

2. Third-order; nonlinear because of (dy/dx)4

+ y2 = 1, we see that it is
1) (dx / dy) + x = O, we see that

3. The differential equation is first-order. Writing it in the form x(dy/dx)

nonlinear in y because of y2 However, writing it in the form (y2 it is linear in x.

4. The differential equation is first-order. Writing it in the form u(dv/du)

+ (1 + u)v =

that it is linear in v. However, writing it in the form (v + uv - ue'U) (du/dv)

ue'U we see

+ u = O, we see that it

is nonlinear in u.
5. Fourth-order; linear
6. Second-order; nonlinear because of cos(r
7. Second-order; nonlinear because of

+ u)

JI + (dy/dx)2

8. Second-order; nonlinear because of 1/ R2


9. Third-order; linear

10. Second-order: nonlinear because of j;2


11. From y

= e=t? we obtain yl = _~e-x/2.

12. Frorn y

= ~ - ~e-20t we obtain dy/dt


dy

&

13. From y

= e3x cos 2x we obtain yl

that y" - 6y'


14. From y

+ 20y =

=-

y" = tanx

+ 13y

+y

= _Cx/2

+ e-x/2

= O'.

= 24e-20t, so that

24e-20t

+ 20 (~

- ~e-20t)

= 24.

= 3e3x cos 2x - 2e3x sin 2x and y" = 5e3T cos 2x - 12e3T sin 2x, so

= O.

+ tan x)
ln(secx + tanx).

cos x ln(sec x

+ cosx

Then 2y'

w.e obtain yl = -1 + sin x ln(sec x

Then y"

+ Y = tanx.
1

+ tan x)

and

Exercises

1. 1

15. Writing in(2X - 1) - In(X - 1)

=t

nd differentiating implicitly .

we obtain
2
dX
2X - 1 dt

1 dX
X - 1 dt

---------=1

( 2X 2-

1)

dX

dt

1- X - 1

---~4----~2--~\--~2--~4-t

=1

2X - 2 - 2X + 1 dX
-------=1
(2X - l)(X - 1) dt
dX

dt

-2
-4

= -(2X

- l)(X

- 1) = (X - 1)(1 - 2X).

Exponentiating both sides of the implicit solution we obtain


2X - 1

---

X-1

Solving et

= et

==:> 2X - 1

O we get t

= Xet

et ==:> (et

- 1)

= (et - 2)X

==:> X

et - 1
-t---.
e-2

or on (ln2, (0).

In2. Thus, the solution is defined on (-oo,ln2)

The graph of the solution defined on (-00, in 2) is dashed, and the graph of the solution defined on
(ln 2, (0) is solid.
16. Implicitly differentiating the solution we obtain
2 dy
- 2x -- - 4xy
dx

+ 2y --dy
dx

= O==:>-

x dy - 2xy dx

==:> 2xy dx

+ (x2

+ y dy

=O

- y)dy = O.

Using the quadratic formula to solve y2 - 2x2y - 1 = Ofor y, we get

V 4x4 + 4) /2 = x2 J.X4+l. Thus, two explicit


are YI = x2 + vx4 + 1 and Y2 = x2 - VX4 + l. Both

-2

solutions are defined on (-00,00). The graph of YI (x) is solid and

-4

y = (2x2
solutions

the graph of Y2 is dashed.


17. Differentiating P

= el et /

(1 + el et) we obtain
dP
(1 + elet) elet
dt

= ~----~----,,-----

elet. elet

(1 + elet)2

Exercses
Substituting into the differential equation, we have
2 {X
2
y' + 2xy = 1 - 2xe-X Jo et dt - 2clxe-x
19. Fromy

= cle 2x+c2xe

weobtain -dy
dx

+ 2xe-X

?
= (2cl+c2)e-x+2c2xe-x

(X 2
Jo el dt + 2clxe-x
d2y
and -0
dx-

1. 1

= l.

= (4Cl+4C2) e 2x +4C2xe 2x ,

so that

20. Frorn y

= qx-l

+ C2x + C3Xlnx + 4x2 we obtain

-dy = -C1X-

dx

+ c2 + c3 + c31n x +

8x,

d2y
-3
-1
dx2 = 2CIX + C3X + 8,
and

so that
3 d3y
x dx3

+ 2x

d2y
dy
dx2 - x dx

+y

= (-6q + 4Cl + Cl + cl)x-l

+ (-C3 + c3)xlnx

+ (-C3 + 2C3 - c2 - c3 + C2)X


+ (16 - 8 + 4)x2

21. (a) From rPl = x2 we obtain rP'l = 2x, so

xrP~ - 2rPl = x(2x) - 2x2 = O.


From rP2 = _x2 we obtain rP2 = -2x,

so

XrP2- 2rP2 = x(-2x)

- 2(-x2)

= O.

Thus, rPl and rP2are solutions of the differential equation on (-00, (0).
(b) Frorn y

-x2
2'
{ z,

22. The function y(x)

does not exist at x

x <O
{-2X.
we obtain y' =
'
x ;:: O
2x,

is not continuous at x

= Osince x~olim

x <O
so that xy' - 2y = O.
x ;:: O
y(x)

= 5 and

lim y(x)

x~o+

= O.

23. (a) The domain of the function, found by solving x +

3;:: O, is [-3, (0).

= -5. Thus, y'(x)

Exercises
(b) From

1. 1
= 1 + (x

+ 3)-1/2 we have .
(y - x)y' - y + X - 2 = [x + 2VX+3 - x][l + (1 + (x - 3)-1/2]

y!

- [X + 2Vx+3] + x - 2
= 2vX+3 + 2 - x - 2Vx+3 + x - 2 = O.

Since y(x) is not differentiable at x

-3, y is a solution of the differential equation on (-3,00).

< 5t < K/2, so 5 tan 5t will be a solution on

24. (a) An interval on which tan 5t is continuous is -n)2


(-K/10,K/1O).

(b) For (1- sin t)-l/2 to be continuous we must have 1- sin t > Oor sin t < 1. Thus, (1- sin t)-l/2
will be a solution on (K/2, 5K/2).

25. (a) From y

= memt.

emt we obtain y'

Then y'

+ 2y =

O implies

+ 2emt = (m + 2)emt = O.

memt

Since emt > O for all t, m = -2. Thus y = e-2t is a solution.

= me'?' and y" = m2emt Then y" - 5y' + 6y = Oimplies

(b ) From y = emt we obtain y'

m2emt _ 5memt

+ 6emt = (m -

2)(m - 3)emt = O.

Since emt > O for all t, m = 2 and m = 3. Thus y = e2t and y = e3t are solutions.
26. (a) From y

= t'" we obtain y' = mtm-1 and y" = m(m - 1)tm-2 Then ty" + 2y' = Oimplies
tm(m - 1)tm-2 + 2mtm-1 = [m(m - 1) + 2m]tm-l = (m2 + m)tm-1
= m(m + 1)tm-1

Since tm-1 > O for t > O, m


(b) From y = t'" we obtain y'

= O and

-1. Thus y

= mtm-1 and y"

= m(m

= O.

1 and 7J =

- 1)tm-2

el

are solutions.

Then t2y" - 7t7J'

+ 15y =

implies
t2m(m

+ 15tm =

- 1)tm-2 - 7tmtm-1

[m(m - 1) - 7m

= (m2

- 8m

+ 15]tm

+ 15)tm

= (m - 3)(m - 5)tm

Since t'" > O for t > O, m = 3 and m = 5. Thus y = t3 and y = t5 are solutions.
27. From x = e-2t + 3t and y = _e-2t + 5e6t we obtain
dx
dt

_2e-2t

+ 18e6t

and

dy = 2e-2t
dt

+ 30e6t

Then
x

+ 3y = (e-2t + 3e6t) + 3( _e-2t + 5t)


= _2e-2t

+ 18e6t =

and

dx
dt

= O.

Exercises

+ 3y

5x

28. From x

+ 3e6t) + 3( _e-2t + 5e6t)

= 5(e-2t
=

1. 1

+ 30e6t

2e-2t

= dy .

dt

= cos 2t + sin 2t + ~et and y = - cos 2t - sin 2t - ~et we obtain


dx
-d = - 2 sin 2t

and

d2x
dt

-2

1t

+ 2 cos 2t + - e

-4cos2t

1
5

+ _et

- 4sin2t

dy.

1t

and

- = 2sm2t - 2cos2t --e


dt
5

and

-2

d2y
= 4 cos 2t
dt

+ 4 Sll12t - - el .
5

Then
4y

+ et = 4(- cos 2t - sin 2t - ~ et) + et


=

-4 cos 2t - 4 sin 2t

+ Se

d2x
dt2

and

4x - et

= 4(cos 2t

+ sin 2t + ~et) - et
d2y
dt2 .

1 t

= 4 cos 2t + 4 sin 2t - Se
29. (y')2

+ 1 = O has no real solution.

30. The only solution of (y')2


31. The first derivative of

+ y2 = Ois y = O, since if y -1 O, y2 > O and (y')2 + y2 ~ y2 > '0.

f(t)

equations are y' = y and y'


32. Any function of the form y

= et is el. The first derivative of


= kV,

differential equation is y" - Y

cet or y

f(t)

t ,is ki. The differential

respectively.

ce:' is its own second derivative.

O. Functions of the form y

c sin t or y

derivatives that are the negatives of themselves. The differential equation is y"

The corresponding
c cos t have second

+ Y = O.

33. Since the nth derivative of 4>(x) must exist if 4>(x) is a solution of the nth order differential equation,
alllower-order derivatives of 4>(x) must exist and be continuous. [Recall that a differentiable function
is continuous.]
34. Solving the system

for

and

C2

Cy(O)

+ C2Y2(0) = 2

Cy~ (O)

+ C2Y2(0)

=O

we get
and

C2

2y~ (O)
= - ---_,.:.~-'-::--_-y (0)Y2(0) -

yi (0)Y2(0)

Exercises

1. 1

Thus, a particular solution is

where we assume that YI(O)Y2(O) - y~(O)'Y2(O) =1- O.


35. For the first-order differential equation integrate f(x).

For the second-order differential equation

+ clt + C2

integrate twice. In the latter case we get y = f(J f(t)dt)dt

36. Solving for yl using the quadratic formula we obtain the two differential equations

so the differential equation cannot be put in the form dy/dt


37. The differential equation yyl-ty

= O has normal form dy/dt

= f(t,y).
= t. These are not equivalent because

y = O is a solution of the first differential equation but not a solution of the second.
38. Differentiating we get yl = CI

+ 3C2t2

Y -

and y" = 6c2t. Then


ty" )
(yll )
2
t + 6t

and the differential equation is t2y" - 3tyl

+ 3y

C2

= y" /6t and

= ty I -

el

= yl - ty" /2, so

1 2 11
3't y

= O.

39. When g(t) = O, y = O is a solution of a linear equation.


40. (a) Solving (10 - 5y)/3x

= O we see that y = 2 is a constant solution.

(b) Solving y2 + 2y - 3 = (y + 3)(y - 1) = O we see that y = -3 and y = 1 are constant solutions.


(e) Since l/(y

- 1) = O has no solutions, the differential equation has no constant solutions.

(d) Setting yl = O we have V" = O and 6y = 10. Thus V = 5/3 is a constant solution.
41. One solution is given by the upper portion of the graph with domain approximately (0,2.6). The
other solution is given by the lower portion of the graph, also with dornain approximately (0,2.6).
42. One solution, with dornain approximately (-00,1.6)

is the portion of the graph in the second

quadrant together with the lower part of the graph in the first quadrant.

A second solution, with

domain approximately (0,1.6) is the upper part of the graph in the first quadrant.
solution, with domain (0,00), is the part of the graph in the fourth quadrant.

The third

Exercises

+ y3)/xy = 3c w
xy(3x2 + 3y2y') _

43. Differentiating (x3

:;J'?J

3x3

obtain

t;

1. 1

x3

+ y3)(xy' + y)

=O

+ 3xy3y' -9'-x3_y_---:xy3y'
1:;:;,y3 _ x4

_ y4
xy3)y'

=O
= -3x3y

+ x3y + y4

44. A tangent line will be vertical whor y' is undefined, or in this case, where x(2y3 - x3) = O. This
gives x

= O and 2y3 = x3 Subst.itut.ing y3 = x3/2 into x3 + y3 = 3xy we get


x3

+ ~x3

= 3x

C;/3

3 3
3
-x = --x
2
21/3

x)

x3 = 22/3x2

Thus, there are vertical tangent Iines at x = O and x = 22/3, or at (O,O) and (22/3,21/3) _ Since
22/3 :::::1.59, the estimates of the dornains in Problem 42 were close,
45. Since q;'(X) > O for all x in J, q;(x) is an increasing function on J. Hence, it can have no relative
extrema on J.
46. (a) When y
(b) When y

= 5, y' = O, so y = 5 is a solution of y' = 5 - y.


> 5, y' < O, and thc solution must be decreasing. When y < 5, y' > O, and the

solution must be increasing. Thus, none of the curves in color can be solutions.
(c)

-5

47. (a) y = Oand y = albo


(b) Since dy/dx
dy/dx

y(a - by)

> O [01' O < y < afb, y = q;(x) is increasing on this interval. Since

< Ofor y < Oor y> albo

.11

= q;(x)

is decreasing on these intervals.

Exercises 1. 1
(c) Using implicit differentiation we compute
~:; = y( -by')

+ y'(a

- by) = y'(a - 2by).

= Owe obtain y = a/2b. Since d2y / dx2 > Ofor O < y < a/2b and d2y / dx2 < O
< y < a/b, the graph of y = cp(x) has a point of infiection at y = a/2b.

Solving d2y / dx2


for a/2b
(d)

48. The family of parabolas is plotted using


ellipses are plotted using

C2

Cl

1, 4, lO. The

= 1, 2, 3. It appears from the figure

that the parabolas and ellipses intersect at right angles. To verify


this we note that the first differential equation can be written in the
form dy/dx = y/2x and the second in the form dy/dx = -2x/y.
Thus, at a point of intersection, the slopes of tangent lines are
negative reciprocals of each other, and the two tangent lines are
perpendicular.
49. In Mathematica

use

Clear[y]
y[x_]:= x Exp[5x] Cos[2x]
y[x]
Y"" [x] - 20 Y'" [x]
50. In Mathematica

158 Y" [x] - 580 Y' [x] + 841 y[x] / / Simplify

use

Clear[y]
y[x_]:= 20 Cos[5 Log[x]]/x - 3 Sin[5 Log[x]]/x
y[x]
x-3 Y'" [x]

2x-2 Y" [x]

20 x Y'[x] - 78 y[x] / / Simplify

Exercses

1.2

Exercises 1.2
1
1. Solving --3 =

--

1 + Cl

_
1

we get Cl = -4. The solution is y =

1 - 4e-t

.11
2. Solvmg 2 = ---

we get q = __e-l. The solution is y = 2 _ e-(t+l) .


1 + cj e
2
3. Using x' = -Cl sint + C2cost we obtain Cl = -1 and C2= 8. The solution is x
4. Using

Xl

5. Using x'

= -Cl sin t + C2cos t we obtain C2= O and -Cl = l.

= -cost + 8sint.
The solution is x = - cos t.

= -Cl sin t + C2cos t we obtain

J3 + -C2
1
1
= -2
2

-Cl
2
1
--Cl
2
.
S o1vmg
we fin d Cl =
6. Using x'

J3
4

J3 = O.
+ -C2
2

an d C2= 4:1 . Th e soluti


ution

IS

x =

J3 cos t + 4:l. sm t .
4

-Cl sin t + C2cos t we obtain

V2

-Cl
2

V2

= V2

+-C2
2

V2 Cl + -V2 C2= 2V2 .

- -

Solving we find Cl = -1 and C2= 3. The solution is x = - cos t + 3 sin t.


7. Frorn the initial conditions we obtain the systern
Cl + C2= 1
Cl - C2= 2.
Solving we get Cl =

~ and C2= -

~. A solution of the initial-value problem is y

~ex - ~e-x.

8. Frorn the initial conditions we obtain the systern


cj e + C2e-l = O
cj e - C2e-l
Solving we get el

= ~ and

C2=

- ~e2

= e.

A solution of the initial-value problern is y = ~ex

- ~e2-x.

9. From the initial conditions we obt.ain

Cle-l -

C2e =

-5.

Solving we get Cl = O and C2= 5e-l. A solution of the initial-value

problern is y

= 5e-x-l.

Exercises

1.2

10. From the initial conditions we obtain

Cl -

Solving we get

Cl

= C2 = O. A solution

11. Two solutions are y


12. Two solutions are y

= Oand y = x3
= Oand y = x2

C2

= O.

of the initial-value problem is y

= O.

(AIso, any constant multiple of x2 is

EL

solution.)

13. Por f(x, y) = y2/3 we have ~f = ~y-l/3 Thus the differential equation will have a unique solution

uy
3
in any rectangular region of the plane where y

i= o.

14. Por f(x, y) = .xy we have ~f = ~ ~. Thus the differential equation will have a unique solution

2y y

uy

in any region where x> Oand y > Oor where x < O and y < O.
15. Por f(x, y) = Ji. we have ~f = ~. Thus the differential equation will have a unique solution in any
x
uy
x
region where x i= o.
af
16. Por f(x, y) = x + y we have ay = l. Thus the differential equation will have a unique solution in
the entire planeo
17. Por f(x, y)

2 2
af
X Y '). Thus the differential equation will have a unique
we have -;:;- =
uy
(4 - y2

x2
--2

4- y

solution in any region where y < -2, -2 < y < 2, or y > 2.


18. Por f(x, y)

x2

af
_3x2y2
we have -;:;- =
2' Thus the differential equation will have a unique
uy
(1 + y3)

--3

1+y

solution in any region where y

i=

-l.

y2

af
2x2y
we have -;:;- =
2' Thus the differental equation will have a unique
x +y
uy
(x2 + y2)
solution in any region not containing (O,O).

19. Por f(x,y)

20. Por f(x, y)

= --

y +x
8f
-2x
we have -;:;- = (
)2'
y-x
uy
y-x

Thus the differential equation will have a unique

solution in any region where y < x or where y > X.


21. The differential equation has a unique solution at (1,4).
22. The differential equation is not guaranteed to have a unique solution at (5,3).
23. The differential equation is not guaranteed to have a unique solution at (2, -3).

10

Exercises

1.2

24. The differential equation is not guaranteed to have a unique solution at (-1,1).

25. (a) A one-parameter family of solutions is y = ex. Since y' = e, xy' = xe = y and y(O) = e 0= O.
(b) Writing the equation in the form y' = y/x we see that R cannot contain any point on the y-axis.
Thus, any rectangular region disjoint from the y-axis and containing (xo, Yo) will determine an
interval around Xo and a unique solution through (xo, yo). Since zo = O in part (a) we are not
guaranteed a unique solution through (O,O).
(e) The piecewise-defined function which satisfies y(O)

= O is not a solution since it is not differ-

entiable at x = O.
d

26. (a) Since dx tan(x

+ e) =

sec2(x

+ e)

= 1 + tan2(x

+ e),

we see that y = tan(x

+ e) satisfies

the

differential equation.
(b) Solving y(O) = tan e = O we obtain e = O and y = tan x.

Since tan x is discontinuous at

x = 7l' /2, the solution is not defined on (-2,2) because it contains 7l' /2.
(e) The largest interval on which the solution can exist is (-7l' /2, 7l'/2).
27. (a) Since

!!_ ( __

dt
equation.

1_)

+e

( t+e )

(b) Solving y(O) = -l/e

t+c

is a solution of the differential

= 1 we obtain e = -1 and y = 1/(1 - t). Solving y(O) = -l/e

we obtain e = 1 and y = -1/(1


of existence of y

= Y , we see that y =

+ t).

= -1

Being sure to inelude t = O, we see that the interval

= 1/(1- t) is (-00,1), while the interval of existence of y = -1/(1 + t) is

(-1,00).
(e) Solving y(O)

= -l/e

= Yo we obtain e = -l/yo
1
Y = - -l/yo

+t =

and
Yo
1 _ yo t '

yO =1 O.

Since we must have -l/yo

+ t =1 O, the

either (-00, l/yo)

> Oor (l/yo, 00) when yo < O.

when yo

(d) By inspection we see that y


28. (a) Differentiating 3x2 - y2

= Ois a

largest interval of existence (which must contain O) is

solution on (-00,00).

= e we get 6x - 2yy' = Oor yy' = 3x.


11

Exercises

1.2

(b) Solving 3x2 - y2

= 3 for y

Y = <P2(X) = -J3(x2

= <P4(X) = -J3(x2

< 00,

-oo<x<-l,
-4

- 1) ,

Only y = <P3(X) satisfies y( -2)

= 2 and y

< 00,

1< x

- 1) ,
- 1) ,

Y = <P3(X) = J3(x2

(e) Setting X

1< x

-1) ,

y = <Pl(X) = J3(x2

we get

-00

< X < -1.

= 3.

= -4 in 3x2 _y2

= e we get 12-16 = -4 = e,

so the explicit solution is


y = -J3x2

+4,

-00

< X < oo.

-4

-2

-4

(d) Setting e = O we have y = V3x and y


29. When x = O and y = ~, y'

-V3x, both

defined on (-00,00).

-1, so the only plausible solution curve is the one with negative slope

at (O, ~ ), or the black curve.


30. The value ofy'
x

= dyjdx

f(x,y)

is determined by f(x,y),

so it cannot be arbitrarily specified at

= xo

31. If the solution is tangent to the x-axis at (xo, O), then y'
these values into y'

+ 2y

=O

when x

= Xo

and y

= O.

Substituting

= 3x - 6 we get O + 0= 3xo - 6 or Xo = 2.

32. We look for the curve containing the point corresponding to the initial conditions.

(a) D

(b) A

(e)

(d)

(e) B

(f) A

33. The theorem guarantees a unique (meaning single) solution through any point. Thus, there cannot
be two distinct solutions through any point.
34. The functions are the same on the interval (0,4), which is all that is required by Theorem 1.1.

12

Exercises

35. Frorn y'

=x-

O, for y

1.3

x - 1, we see that solutions intersecting the line

1 have horizontal tangent lines at the point of intersection. From

y' > O, for y > x - 1, we see that solutions aboye the line y
are increasing. From y' < O, for y < x -

1,

=x - 1

we see that solutions are

decreasing below the line y = x - 1. From y" = O, for y

x, we see

that solutions have possible infiection points on the line y = x. Actually,


y = x is easily seen to be a solution of the differential equation, so the
solutions do not have infiection points. From y" > O for y > x we see
that solutions aboye the line y = x are concave up. From y" < O for y < x we see that solutions
below the line y = x are concave down.

Exercises 1.3

dP
dP
= kP+r'
- = kP-r
dt
'
dt
2. Let b be the rate ofbirths and d the rate of deaths. Then b = k1P and d = k2P. Since d.Pf dt = b-d,
1. -

the differential equation is dP/dt = klP - k2P.


3. Let b be the rate of births and d the rate of deaths. Then b = klP and d
the differential equation is dP / dt

= k2P2. Since dP/dt = b=d,

= kl P - k2P2.

4. Let P(t) be the number of owls present at time t. Then dP/dt

= k(P - 200 + 10t).

5. From the graph we estimate To = 180 and Tm = 75. We observe that when T = 85, dT/dt ~ -1.
Frorn the differential equation we then have
k = dT/dt
T-Tm

-1
85-75

= -0.1.

6. By inspecting the graph we take Tm to be Tm(t) = 80 - 30cos7ft/12.

Then the temperature of the

body at time t is determined by the differential equation

~ =

k [T -

(80 - 30 cos ;2

t) ], t > O.

7. The number of students with the flu is x and the number not infected is 1000 - x, so dxf dt. =
kx(1000 - x).
8. By analogy with differential equation modeling the spread of a disease we assurne that the rate at

which the technological innovation is adopted is proportional to the number of people who have
adopted the innovation and also to the number of people, y( t), who have not yet adopted it. If one
person who has adopted the innovation is introduced into the population then x
dx
dt = kx(n + 1 - x), x(O) = 1.

13.

+y

=n

+ 1 and

Exercises

1.3

9. The rate at which salt is leaving the tank is


(3 gal/min) .

C~O

lb/gal) = 1~ lb/mino

Thus dA/dt = A/lOO.


10. The rate at which salt is entering the tank is
Rl = (3 gal/rnin) . (2 lb/gal) = 6 lb/mino
Since the solution is pumped out at a slower rate, it is accumulating at the rate of (3 - 2)gal/min

1 gal/min. After t minutes there are 300 + t gallons of brine in the tank. The rate at which salt is
leaving is
R2

= (2 gal/min) ( -- A) lb/gal
300 + t

2A
lb/mino
300 + t

The differential equation is


dA = 6 _ 2A
dt
300 + t
11. The volume of water in the tank at time t is V
dh
dt
Using.Ao

_1 dV

Aw dt

= Awh.

_1 (-cAofiih)

The differential equation is then

= _ cAo fiih.

Aw

Aw

= 7r 22)2 = ~ , Aw = 102 = 100, and g = 32, this becomes


dh
dt

= _ C7r/36 J64h = _~ v'h.


100

12. The volume of water in the tank at time t is V

450

~7rr2h = ~Awh. Using the formula from

Problem 11 for the volume of water leaving the tank we see that the differential equation is
dh
dt
Using Ah

= _2_ dV = _2_ ( -CAh fiih)


Aw dt

= 7r(2/12)2 = 7r/36,

= - 3cAh

Aw

Aw

fiih.

= 32, and e = 0.6, this becomes

dh = _ 3(0.6)7r/36
dt
Aw

J64h = _ 0.47rh 1/2.


Aw

To find Aw we let r be the radius of the top of the water. Then r / h


Aw

= 7r(2h/5f

= 47rh2/25. Thus

dh = _ 0.47r h1/2 = -2.5h-3/2


dt
47rh2/25
dq
d2q
13. Since i = dt and L dt2

dq

+ R dt

= E(t) we obtain

di

L dt + Ri = E(t).

14. By Kirchoff's second law we obtain Rdq + el q = E(t).


dt

14

8/20, so r

2h/5 and

Exercises
dv
= -kv
dt

15. From Newton's second law we obtain m-

1.3

+ mg.

16. We have from Archimedes' principIe

upward force of water on barrel = weight of water displaced

(62.4) x (volume of water displaced)

= (62.4)7T(S/2)2y = 15.67Ts2y.
w d2y
--2

It then follows from Newtori's second law that


9

9 dt

-15.67Ts 'Y

= 32 and w is the weight of the barrel in pounds.

17. The net force acting on the mass is

d2x
F = ma = m dt2 = -k(s
Since the condition of equilibrium is mg

+ x) + mg

+ mg

= -kx

- ks.

= ks, the differential equation is


d2x

= +ks.

m dt2

18. From Problem 17, without a damping force, the differential equation is md2x/dt2

= -kx.

With a

damping force proportional to velocity the differential equation becomes


d2x
m dt2

dx

d2x

= -kx - (3 dt

or

m dt2

dx

+ (3 dt + kx = O.

19. Let x(t) denote the height of the top of the chain at time t with the positive direction upward. The

weight of the portion of chain off the ground is W = (x ft) . (1 lb/ft) = x. The mass of the chain is
m = W/g = x/32. The net force is F = 5 - W = 5 - x. By Newton's second law,
d x
dt (32 v) = 5 - x

dv
x dt

or

dx

+ v dt

= 160 - 32x.

Thus, the differential equation is


x-d2x
dt

+ (dx)2
+ 32x =
dt

160.
d

20. The force is the weight of the chain, 2L, so by Newton's second law, -d [mv] = 2L. Since the mass

of the portion of chain off the ground is m = 2(L - x)/g, we have

!!_ 2(L
dt

- x) v] = 2L
9

(L-x)-+vdv
dt

or

Thus, the differential equation is


(L - x)-d2x - (dx)
dt2
dt

15

= Lg.

(dX)
-dt

=Lg.

Exercises

21. From g

1.3
kj R 2 we find k

gR~.0 Using a

d 2 r j dt~0 and the fact that the positive direction is

upward we get
or
22. The gravitational force on mis F = -klVIrmjr2.

u, = r3 M j R3

Since NIr

= 41fr3j3 and M

= 41fR3j3 we have

and

F = -k Mrm = -k r3VImjR3 = -k mM r.
r2
Now from F

r2

R3

= ma = d2r j dt2 we have

23. The differential equation is dA = k(M _ A).

dt
dA
24. The differential equation is - = kl(M - A) - k2A.
dt
25. The differential equation is x'(t) = r - kx(t) where k > O.
26. By the Pythagorean Theorem the slope of the tangent line is y'
27. We see from the figure that 28
y

-x

= tan

+ o: = tt,

o: = tan(7T- 28)

-y

-r====

Js2 - y2

Thus

=-

tan 28 =

2 tan 8
1 _ tan2 8 .

Since the slope of the tangent line is y' = tan 8 we have y j x


or y - y(y')2 = 2xy', which is the quadratic equation y(y')2

= 2y' [1- (y')2]


+ 2xy' - y = O

in yl Using the quadratic formula we get


,

Y =

-2x

J4x2
2y

+ 4y2

-x

Jx2

+ y2
.x

Since dy j dx > O, the differential equation is


dy
dx

- x

+ J x2 + y2

or

28. The differential equation is dPjdt

y ~~ - Jx2

+ y2 + X

= O.

= kP, so from Problem 31 in Exercises 1.1, P = ekt, and a

one-parameter family of solutions is P = ct.


29. The differential equation in (3) is dTjdt

= k(T - Tm).

When the body is cooling, T > Tm, so

< O and k < O. When the body is warrning, T < Ti, so


T - T m < O. Since T is increasing, dT j dt > O and k < O.
T - Tm > O. Since T is decreasing, dTjdt

16

Exercses

30. The differential equation in (8) is dA/dt

at this time and A


for t

> O and

1.3

If A(t) attains a maximum, then dA/dt = O

= 6 - A/I00.

= 600. If A(t) continues to increase without reaching a maximum then

A cannot exceed 600. In this case, if A'(t)

A'(t)

> O

approaches O as t increases to infinity, we

see that A(t) approaches 600 as t increases to infinity.


31. The input rate of brine is ri gal/rnin and the concentration of salt in the infiow is e lb/gal, so the

input rate of salt is riCi lb/mino The output rate of brine is ro gal/min and the concentration of
salt in the outfiow is colb/gal, so the output rate of salt is rocolb/min.

The solution in the tank

< ro). After

is accumulating at arate of (ri - ro) gal/rnin (or decreasing if ri

Vo + (ri - r o)t gallons of brine in the tank, and the output rate of salt is r oA/[Vo

t minutes there are

+ (ri -

ro)t]lb/min.

The differential equation is


dA
roA
- = riCi dt
Vo+(ri-ro)t

32. This differential equation could describe a population that undergoes periodic fiuctuations.
dP

33. (1):

di =

(3):

di =

(6):

dX
-dt

(10):

dT

kP

is linear

(2):

k(T - Tm) is linear

(5):

= k(a - X)(f3 - X) is nonlinear

dh
dt

=-

Ah ~

dA

di
dx
dt

= kx(n + 1 - x) is nonlinear

dA

A
=6- is linear
100

(8): -

dt

d2q

Aw V 2gh is nonlinear

(11):

is linear

= kA

L dt2

dq

+ R dt +

Cq

= E(t)

is linear

dv
(14): m- = mg - kv is linear

(12):

dt

d2s

d2x

ds

(15): m dt2 + k dt = mg is linear


34. From Problem 21, d2r / dt2

= -g

(16): -

dt2

64

- -x = O is linear
L

R2 / r2. Since R is a constant, if r

+ s,

then d2r / dt2

= d2s / dt2

and, usng a Taylor series, we get


d2s
R2
2
dt2 = -g (R + s)2 = -gR (R

+ s)

-2

;:::;-gR

[R

-2

- 2sR

-3

+ ... ] = -g +

2gs
R3

+ ....

Thus, for R much larger than s, the differential equation is approximated by d2s/dt2 = -g.
35. If p is the mass density of the raindrop, then m = pV and
dm = p dV
di
dt

If dr l dt is a constant, then dm/dt


decreasing, k
r = kt] p

< O. Solving

dr/dt

= p !!:_ [~1T3l= P (47rr2 dr) =


dt 3

dt

pS dr .
dt

= kS where p dr l dt = k or dr l dt = k/p. Since the radius is


= k/p we get r = (k/p)t + Co. Since r(O) = ro, Co = ro and

+ ro.
17

Exercises 1.3
d
Frorn Newton's second law, -[mv]
dt
dv
m dt

= mg, where

dm

vis the velocity of the raindrop. Then

or

+vdi = mg

Dividing by 4p1fr3/3 we get


dv

3k

dt

pr

-+-v=g

dv

or

3k/ p

+ k t / P + ro v

-d

= g,

< O.

36. We assume that the plow clears snow at a constant rate of k cubic miles per hour. Let t be the
time in hours after noon, x(t) the depth in miles of the snow at time t, and y(t) the distance the
plow has moved in t hours. Then dy/dt is the velocity of the plow and the assumption gives

dy
wx-=k
dt

where w is the width of the plow. Each side of this equation simply represents the volume of snow
plowed in one hour. Now let to be the number of hours before noon when it started snowing and
let s be the constant rate in miles per hour at which x increases. Then for t

> -to, x

= s(t

The differential equation then becomes

dy
dt

-=---

ws t + to

Integrating we obtain
k

y = - [In (t + to) + e]
ws
where e is a constant. Now when t = 0, y = so e = -In to and

y = ~ In (1

ws

..!.)
.
to

Finally, frorn the fact that when t = 1, Y = 2 and when t = 2, Y

(1 + ~)
Expanding and simplifying gives

ta + to -

= 3, we obtain

(1 + t~f

1 = O. Since to

> 0, we find to ~ 0.618 hours ~

37 minutes. Thus it started snowing at about 11:23 in the morning.

18

+ to).

Chapter

1 Review Exercises

Chapter 1 Review Exercises


= ky

dy
dx
d

2. dx (5 + cle-2x)

d
3. -(q eos kx
dx

= -2cle-2x

+ C2sin kx)

d2
-2 (Cleos kx
dx

+ cle-2x

= -2(5

= -kCl sin kx

+ C2sin kx)

dy
dx

- 5)

= -2(y - 5) or -dy = -2y + 10


dx

+ kC2 eos kx

= -k2cl eos kx - k2c2 sin kx = -k2(Cl eos kx

d2y
d2y
= _k2y
or + k2y = O
dx2
dx2
d
4. - (Cleosh kx + C2sinh kx) = kc sinh kx
dx

+ C2sin kx)

+ C2sinh kx)

d22 (Cleosh kx
dx
d2y
2
dx2 = k y
5. y'

or

= k2 Clcosh kx

+ k2 C2sinh kx

= k2 (ci eosh kx

+ C2sinh kx)

d2y
2
dx2 - k y = O

= C1ex + C2xex + C2ex

y"

+ kC2 eosh kx

y"

+ C2xex + 2C2ex;
= 2(C1eX + C2xex + C2eX) = 2y'

= C1ex

+ Y = 2(C1eX + C2xeX) + 2C2ex

+ CleX eos x + C2ex eos x + C2ex sin x;


el eX sin x + q e" eos x = - 2CleX sin x + 2c2ex eos x;

y" - 2y'

+y

=O

6. y' = -C1ex sin x

y" = -Cl eX eos x - C1eX sin x y" - 2y' = -2clex


7. a,d

eos x - 2c2ex sinx = -2y;

8. e

13. A few solutions are y

y" - 2y'

+ 2y

eos x

+ C2ex eos x + C2ex sin x

=O

10. a,e

9. b
= O, y

+ c2ex

C2ex sin x

11. b

12. a,b,d

= e, and y = eX.

14. Easy solutions to see are y = Oand y = 3.


15. The slope of the tangent line at (x, y) is y', so the differential equation is y' = x2
16. The rate at which the slope changes is dy' / dx

y" +y'

+ y2.

y", so the differential equation is y"

-y'

or

= O.

17. (a) The domain is all real numbers.


(b) Sinee y'

= 2/3xl/3, the solution y

= x2/3 is undefined at x

= O. This funetion is a solution of

the differential equation on (-00, O) and also on (0,00).


18. (a) Differentiating y2 - 2y = x2 -

+ Cwe obtain
19

2yy' - 2y' = 2x - 1 or (2y - 2)y' = 2x - 1.

1 Review Exercises

Chapter

(b) Setting x = O andoy = 1 in the solution we have 1 - 2 = O - O + e or e


of the initial-value problem is y2 - 2y = x2 - x - l.

-l. Thus, a solution

= O by the quadratic formula we get y = (2V4 + 4(x2 - x - 1) )/2

(c) Solving y2-2y-(x2-x-1)

-JX2=X = 1 vx(x - 1). Since x(x - 1) ;:: O for x S O or x ;:: 1, we see that neither

=1

= 1+

vx(x - 1) nor y

= 1 - vx(x - 1) is differentiable at x = O. Thus, both functions are

solutions of the differential equation, but neither is a solution of the initial-value problem.
19. (a)

y
3

-3

(b) When y

= x2 + el. y'

= 2x and (y')2

')

(c) Pasting together x-, x ;::O, and -x

= 4x2. When y

{_x

6V4 +

5( -1)3

x<O

2,
2

, x S O, we get y =

20. The slope of the tangent line is y' 1(-1,4)=

21. Differentiating y

= -x2 + e2, y' = -2x and (y')2 = 4x2.

x,

-.

= 7.

= sin(lnx) we obtain y' = cos(lnx)/x and y" = -[sin(lnx) +cos(lnx)1/x2

2 '"

x y + xy + y

= x 2( -

sin(lnx) + cos(lnx))

+x

cos(lnx)

. (1 )
+ sm n x

Then

= O.

22. Differentiating y = cos(lnx) ln(cos(lnx)) + (lnx) sin(lnx) we obtain

= cos

(1)

nx

1
(Sin(lnx))
(1 ) cos nx
x

=- ln(cos(ln x)) sin(ln x)


x

1 (

+ neos

(1

nx

))(

sin(lnx))
1 cos(lnx)
+ nx
x
x

+ sin(lnx)
x

(ln x) cos(ln x)
x

+~--'---'-----'-

and
y

,,[

= -x ln(cos(lnx)) cos(lnx) + sin(lnx)


x

1
(Sin(lnX))]
(1 ) cos nx
x

1
[
(Sin(lnx))
+ ln( cos(ln x)) sin(ln x)"2 + x (In x) x
x
2

1
"2
x

COS(lnX)] 1
?
x
z-

1[
sin (ln x)
.
="2 - ln( cos(ln x) cos(ln x) +
(1 ) + ln(cos(ln x) sm(ln x)
x
cos nx
- (lnx) sin(lnx) + cos(lnx) - (lnx) cos(lnx)].

20

1
(In x) cos(ln x)"2
x

1 Review Exercises

Chapter
Then
2

x 2 y 1/

+ XY / + y =

+ cos(ln x)
+ (In x)
=

+ sin ((ln X)) + ln(cos ())lnx

-In ( cos(lnx))cos(lnx)

sin2(lnx)
cos(ln x)

. ()lnx
sm

cos lnx

- (In x) cos(ln x) - ln( cos(ln x)) sin(ln x)


cos(ln x)

+ cos(ln x)

ln(cos(ln x))

+ (In x)

sin2(lnx) + cos2(ln x)
cos(ln x)

+ cos ()In x =

sin (In x)

1
cos(ln x)

(This problem is easily done using M athematica.)


23. From the graph we see that estimates for YO and Yl are YO = -3 and Yl
24. The differential equation is
dh
dt

Using Ao

= _ cAo

= O.

fi;h.

Aw

= 11"(1/24)2 = 11"/576, Aw = 11"(2)2 = 411", and g = 32, this becomes


dh = _ c11"/576
dt
411"

J64h

V3

= -mg - {J-mg

=~

288

25. From Newton's second law we obtain


dv
mdt

. ( lnx )
- (lnx ) sm

or

21

vh.

= sec (1)n x

Flrst-Order Differential Equations


Exercises 2.1

1.

7.

2.

8.

\ I \ 1\1 \

....._'"'""

\ I
\ I

\ 1\ 1\ I \
\ I \ lit
\
\1 \ 1II \

\ \ ,-""
\ \ , .._'"'"
\\"
..-~,

\ I \

\ \"

I I I \

\ I
\ I

\ I \ I I I \
\ I \ I I I \

\ I

\I \ I I I \

.........
......... ~
~

.........
.........

22

.-.........

\ I

__

~,
~,

__

<lo ..

"
,

\ 1\1 \ .....- ...


\ 1\1 \ \ , ...
_'"'
\ t \ I \ \ \ \ ..-

,~
,~

,~
,~

......

-_

__

......

Exercises 2. 1
9.

10.

y
II I 11I 11I I

I I I

1111/111'
I
1111/1'
1/.
111"
1'1'
I

y
111

11

11

111

,.,

II

II1

II1
III
II
II
II
III

~~~~~.f-+~~~~'~'
1X
1""

..... "

I ,
I , 1" I , .._"
I , l' I " .._,
l'
,\ I ",
...

11.

1\ 1, I ",

I I 11I '1

III
11
11

12.

y
II

I l'
"11/111111
"11/111111

I III

1"

J"

11""

13.

1I

"

I .. ,.,."

.,

II
11
11

,,,
'"
,,,

111

1111

1111

III
III
III

1111
111.

11

1111
111I
1111

1II1
'11111
1111"

II

, 1II
111.

1111'
1111'

1111
111I

.. ,,"
"

"
"

1\ 1 1 1\1

'1"
I
I

1 ........

,"

_
_
-----

III
III
III

1'11
I1II
11II

I l' , , l'
1111'111'
I1I1
1111'
11II
111"

'"

1111 111

" I
"
I
11I

,"

"

""""'\

""

'"

.., -1
.. \ - 1
..\-1
..\-1

11
I
I

I I
I
II

," ,
"11

11- \-

11
...

11

11- \11_ \_
\J
11- \11 ... ' _

II
II

,,-

"III I

1-'-

I I

11- \-

111111

11- \-

11-,

, , ,
I
I
I

"

II

1 1-\-

11

= -1, y = O,and y =

1'"
1""

y
_1

I1 - \ - I
I 1- \-1
11-' -.

1 1-'

11-'
11-'

"

15. Writing the differential equation in the form dyJdx = y(llocated at y

I
I

, I

" "'\\'\,
"1
""
", I "'1 ""

-- .. _ .. ""

, I r r r 1"
II 11 I 1'"
I
11 I
I 11 ,
II11 I 1""

1 .........

---------, " " '"


11"

I
I
I
I

"""""

.. ........

"""""
X
---------,\ ""
,,,
"
,,,
1" , , 1" , , ,

,"

11""
11""
\,,,
\ "

.........
_- ......
\ \
--- __ "".. "
----....."
'"'_.._ ...
,'"

14.

y
1111

"

.. I

II
II
II

....

I "1

, I111
I ",,
'11111/111
'1//1
I I11I

l'

11

II
II

y
, 1'"

y)(l

+ y) we see that critical points are

1. The phase portrait is shown below.


-1
y

(a)

(b)

(c)

(d)

23

-+------------~------------~2x

Exercises 2. 1
16. Writing the differential equation in the form dyjdx

are located at y = -1, Y = 0, and y

= y2(1 - y)(l

+ y)

we see that critical points

1. The phase portrait is shown below.

-1

(a)

(b)

iL

-2

-1

2 x

(d)

(c)

-2

-3

-.

-5

17. Solving y2 - 3y = y(y - 3) = Owe obtain the critical points O and 3.

From the phase portrait we see that Ois asymptotically stable and 3 is unstable.
18. Solving y2 - y3

= y2(1 - y)

we obtain the critical points Oand 1.

o
From the phase portrait we see that 1 is asymptotically stable and Ois semi-stable.
19. Solving (y - 2)4

= O we obtain the critical point

2.

From the phase portrait we see that 2 is semi-stable.


20. Solving 10 + 3y - y2 = (5 - y)(2

+ y) = Owe obtain

-2

the critical points -2 and 5.

From the phase portrait we see that 5 is asymptotically stable and -2 is unstable.
21. Solving y2(4 - y2)

= y2(2 - y)(2 + y)
-2

= Owe

obtain the critical points -2, O, and 2.

24

Exercses 2. 1
From the phase portrait we see that 2 is asymptotically stable, Ois semi-stable, and -2 is unstable.
22. Solving y(2 - y)(4 - y) = Owe obtain the critical points 0,2, and 4.

From the phase portrait we see that 2 is asymptotically stable and O and 4 are unstab!e.
23. Solving y ln(y

+ 2) = Owe obtain
-2

the critical points -1 and O.


o

-1

From the phase portrait we see that -1 is asymptotically stable and O is unstab!e.
24. Solving yeY

9y = y(eY

9) = Owe obtain the critical points O and In 9.


o

In 9

From the phase portrait we see that O is asymptotically stable and In 9 is unstable.
25. (a) Writing the differential equation in the form

dvdt = mk (mkg _ v)
we see that a critical point is mg j k.
rng/k

From the phase portrait we see that mg j k is an asymptotical!y stable critica! point. Thus,
limt->oov

= mg j

k.

(b) Writing the differential equation in the form

dvdt = 'm~ (mkg _ v2) = mk ( VT


mg _ v) ( fi'i'Lg+ v)
VT
we see that a critical point is

Jmgjk.
--J rng/k

J mg j k is an asymptotically

From the phase portrait we see that


limt~oo v

stable critical point. Thus,

= Jmgjk.

26. (a) From the phase portrait we see that critica! points are a and (3. Lct X(O)

= Xo.

If Xo

< a, we see that X

-t

a as t

-t

oo. If a < Xo < (3, we see that X

25

-r+

a as t

-r+

oo.

Exercses 2. 1
If Xo

> (3, we see that X(t) increases in an unbounded manner, but more specific behavior of

X (t) as t

---+ 00

is not known.

(b) When a = (3 the phase portrait is as shown.

If Xo

< a, then X(t)

---+

a as t ---+ co. If Xo > a, then X(t) increases in an unbounded manner.

This could happen in a finite amount of time. That is, the phase portrait does not indicate
that X becomes unbounded as t

---+

oo.

(e) When k = 1 and a = (3 the differential equation is dX / dt = (a - X) 2 Separating variables


and integrating we have
dX
(a _ X)2

= dt

--=t+c
a- X
a-X=--

t+c

X=a---.
For X(O)

For X (O)

= a/2

t+c

we obtain
1

X(t)=a-

/
t+2 a

X(t)=a-

/
t- 1a

2a we obtain
1

For Xo > a, X(t) increases without bound up to t


X ---+ a as t
co.
-jo

26

= l/a. For t > l/a, X(t) increases but

Exercises 2. 1

27. Critical points are y = O and y

= c.

y
1111111111
1 r I I 11 I I I I
I 11 J 1I I

I JI

1 11 111

1 1 111 1 11 11

1111

11 1/ 1 11/1
III~IIIII

\""""
\ \ , \ \ \ I \,

""'"
'\\"1'"

\ \ , \ \ \ 1"

"""""

-2.2,

,\

, I \ \ \ I \ \ \
, I \ \ \ I \ \ \

\1 1111111
\1 1111111

1111111111
1111111111

points are y

11

III
II
I 1I II

1"1
I111

/1/11/111/
111111111'
\
\111111\11
\ \ \ 1 \ \ 1 \ \ \

28. Critical

1 111111

\ \ \\ l' \\

""""

1111111111
111
I 11 I 11

III~IIIII
II111
I 11

111111111

11111111

I I

0.5, and

y = 1.7.

0.5

-2.2

1.7

29. At each point on the circle of radius c the lineal elernent has slope c2.
30. (a) When x = Oor y = 4, dy/dx

dy/dx

= -2 so the lineal elernents have slope -2.

= x - 2, so the lineal elernents at (x,3)

When y

= 3 or y

= 5,

or (x,5) have slopes x - 2.

(b) At (O,YO) the solution curve is headed down. As x increases, it will eventually turn around and
head up, but it can never cross y
O. Thus, y cannot approach
31. When y

< ~x2, y'

00

= 4 where a tangent line to a solution curve rnust have slope

as x approaches oo.

= x2 - 2y is positive and the portions of solution

curves "inside" the nullcline parabola are increasing. When y

> ~x2,

y' = x2 - 2y is negative and the portions of the solution curves


"outside" the nullcline parabola are decreasing.
I
I I ,
11/
//11
1111
11 I
I f 1/
II
I

/'.1.

I I I
I , I
I I
I 1
I I
I I I
I I
I I I

J
J
I
JI'
I
III
l'
I

I I
I I
I I
I I
I
I I

-3~~~~~~O~/~/~~~I~I~;~I~I~~~

32. For dxl dt = Oevery real nurnber is a critical point, and hence all critical points are nonisolated.
33. Recall that for dy / dx

f(y)

we are assurning that f and

f'

are continuous functions of y on

sorne interval J. Now suppose that the graph of a nonconstant solution of the differential equation

27

Exercses 2. 1
crosses the line y

= c. If the point of intersection is taken .as an initial condition we have two distinct

solutions of the initial-value problem. This violates uniqueness, so the graph of any nonconstant

f is continuous it can only

solution must lie entirely on one side of any equilibrium solution. Snce

change signs around a point where it is O. But this is a critical point. Thus, f(y) is completely
positive or completely negative in each region R; If y(x) is oscillatory or has a relative extremum,
then it must have a horizontal tangent lne at sorne point (xo, Yo). In this case yO would be a critical
point of the difIerential equation, but we saw aboye that the graph of a nonconstant solution cannot
intersect the graph of the equilibrium solution y = yo.
34. By Problem 33, a solution y(x) of dyjdx
monotone. Since y'(x)

= f(y)

cannot have relative extrema and hence must be

= f(y) > O, y(x) is monotone increasing, and since y(x) is bounded aboye

by C2, limx-+ooy(x) = L, where L ::; C2. We want to show that L = C2. Since L is a horizontal
asymptote of y(x), limx-+ooy'(x)

f(L)

= O. Using the fact that f(y) is contnuous we have

= f( x-+oo
lim y(x)) = Iim
x-oo

But then L is a critical point of f. Since


C2, L = C2

Cl

f(y(x))

= lim

x-oo

y'(x)

= O.

< L ::; C2, and f has no critical points between

Cl

and

35. (a) Assuming the existence of the second derivative, points of infiection of y(x) occur where

y"(x) = O. Frorn dyjdx = g(y) we have d2yjdx2 = g'(y) dyjdx. Thus, the y-coordinate of a
point of infiection can be located by solving g' (y) = O. (Points where dy j dx = Ocorrespond to
constant solutions of the difIerential equation.)
(b) Solving y2_y_6

= (y-3)(y+2)

points. Now d2yjdx2

= Owe see that 3 and -2 are critical

(2y - 1) dyjdx

(2y - l)(y - 3)(y + 2),

so the only possible point of infiection is at y

i,

although the

concavity of solutions can be difIerent on either side of y = -2 and

y = 3. Since y"(x) < O for y < -2 and

i < y < 3, and y"(x)

>O

< y < ~ and y > 3, we see that solution curves are concave
down for y < -2 and i < y < 3 and concave up for -2 < y < i and
y > 3. Points of inflection of solutions of autonomous difIerential

-s

for -2

-5

equations will have the same y-coordinates because between critical points they are horizontal
translates of each other.

28

Exercises 2.2

Exercises 2.2
In many of the following problems
solve for g(y) we exponentiate
implies g(y)

ecef(x).

wc will encounter

Letting

3. Frorn dy
4. From (

el

dx we obtain In

Iyl =

1
6. From - dy = -2x dx we obtain In
y

41n

From

+ 2 + ;)

(y

10. Frorn (2y

11.

dy =
cscy
- cos y = - ~x -

ll. From --

---2-

sec x

+ 5)2 dx

or

+ 2y + In Iyl
2

we obtain

dx or sin y dy

t sin 2x + e

+ 2e3x = e.

dy = (4x

1 - --.
x+e

Ixl + e or y = elx4

dy = x21n x dx we obtain y;

+ 3)2

= x + e or y =

Iyl = -x 2 + e or y = ele-x.

7. From e-2Ydy = e3xdx we obtain 3e-2y

9.

= elef(x)

+ e.

~e-3x

=-

4 cos y

2y

+3 =

cos2x dx

= x33In

Ixl -

(eY

eY

+ 1) 2

dy -

+ 5 + e.

-~(1 + cos 2x) dx we obtam

= 2x + sin 2x + Cl

_ex 3 dx we obtain - (eY


(eX + 1)

+ 1)-1

= ~(eX

y
x
(
2)1/2
(
2)1/2
1/2 dy =
1/2 dx we obtain
1+Y
= 1+x
(1 + y2)
(1 + x2)

15. From

S dS

16. Frorn -Q 1
-70

= k dr we obtain S = er.

dQ

= kdt

we obtain In

IQ -

701 = kt

29

+ e or Q -

70

+ e.

+ 1)-2 + e.

14. FrOl11

+ e.

4x

sin 3x
2
.
2
1
2
12. From 2y dy = - cos3 3x dx = - tan 3x sec 3x dx we obtain y = -'6 sec 3x

13. From

~x3

= qt.

+ e.

To

= eCef(x) which

= ~(x + 1)3 + e.

y-1

Ig(y)1 = f(x) + e.

I = ef(x)+c

t cos 5x + e.

)2 dy = dx we obtain -_-

= -x

5. From - dy
y

of the form In

This yields Ig(y)

ec we obtain g(y)

we obtain y

_e-3x dx we obtain y

y-1

+ 1)2 dx

an expression

both sides of the equation.

1. From dy = sin 5x dx we obtain y = 2. From dy = (x

Exercses

2.2

17. From P _1 p2dP


P
or -P
1-

= (1P + 1 _1)P
t

dP = dt we obtain In IPI-In 11-PI = Hc so that In 1 _P P = Hc


C1et
t
1 + C1e

= C1e . Solving for P we have P =

1
18. From N dN = (tet+2 - 1) dt we obtain In INI = tet+2 - et+2 - t + c.
19. From y - 23dy = x - 1 dx or (1 __ 5_) dy = (1 __ 5_) dx we obtain
y+
x+4
y+3
x+4
y-5Inly+31=x-5Inlx+41+c

20. From y + 1 dy = x + 2 dx or (1 + _2_)


y-1
x-3
y-1

or

dy = (1 + _5_)
x-3

Y +2Inly-ll=x+5Inlx-31+c

21. From x dx =

or

n
1 - y2

(X+4)5
y+3

=C1ex-y.

dx we obtain
(y_1)2 _
x-y
(x _ 3)5 - C1e

dy we obtain ~x2 = sin-1 y + c or y = sin (x2 + e).


2

1
1
eX
1
1
22. From -2 dy =
dx = ()2
dx we obtain -- = tan-1 eX + c or y = .
y
eX + e-x
eX + 1
y
tan-1 eX + c
23. From

-2-1-

dx = 4dt we obtain tan-1 x = 4t + c. Using x(7r/4) = 1 we find c = -37r/4.


x +1
solution of the initial-value problem is tan -1 x = 4t - 3; or x = tan ( 4t _ 3;).

24. From -')--1 dy


y- - 1

The

1 dx or -1( -- 1 - -- 1) dy = -1( -- 1 - -- 1) dx we obtain


.
x2 - 1
2 Y- 1 y + 1
2 x - 1 x + 1

= --

In Iy - 11-ln Iy + 11= In Ix - 11-ln Ix + 11+ In e or y - 1 = c(x - 1) . Using y(2) = 2 we find


y+1
x+1
y-l

c = 1. The solution of the initial-value problem is -y+1

x-l

= --

x+1

or y = x.

1 - x dx = ( 2"
1 - -1) dx we obtain
. In Iyl = -- l' -In Ixl = c or xy = c1e- 1/x. Using
25. From -1 dy = -2y
x
x
x
x
y(-l) = -1 we find C1= e-l. The solution of the initial-value problem is xy = e-1-1/x.
1
26. From 1 _ 2y dy = dt we obtain -~ In 11- 2yl = t + c or 1 - 2y = ee-2t. Using y(O) = 5/2 we find
C1= -4. The solution of the initial-value problem is 1 - 2y = _4e-2t
27. Separating variables and integrating we obtain
dx

Vf=X2

dy

J1- y2

=0

and

30

sin-1x-sin-1y=c.

or y = 2e-2t + ~ .

Exercises 2.2
Setting x = O and y =

V3/2

we obtain e =

-1r

/3. Thus, an implicit solution of the initial-value

problem is sin-l x - sin-l y = 1r/3. Solving for y and using a trigonometric identity we get
1

y = sin ( sin- x
28. From

1r)
+ '3

= x cos

1r
V3 V1=X'2
'31r + Vr.:":
1 - x2 sin '3 = '2 +
2
.
X

-x

dy dx we obtain
1 + (2y)2
- 1 + (x2)2
1

+e

- tan-l 2y = -- tan-l x2
2
2
Using y(l)

= O we find

C}

or

tan-l 2y

+ tan-l

x2 = el.

= 1r/4. The solution of the initial-value problem is

tan-l 2y

+ tan-l

x2

= ~.
4

= 2 and y(x) = -2 satisfy the initial conditions y(O) = 2 and


= -2, respectively. Setting x = ~ and y = 1 in y = 2(1 + eX)/(l - eX) we obtain

29. (a) The equilibrium solutions y(x)


y(O)

1 = 21 + ee
1-ee'

1-

ee

= 2 + 2ee,

and e=--.

-1 = 3ee,

3e

The solution of the corresponding initial-value problem is


y =2

1- lx-l
3
1 + le4x-l

3_ e4x-l
2---;----:3 + e4x-l

(b) Separating variables and integrating yields


1
In Iy -

21 -

In Iy -

1
In Iy

21 -

+ 21 + In el

In Iy

=x

+ 21 + In e = 4x

In I e(y - 2)
y+2

I = 4x

y - 2
4
e-- = e x
y+2
Solving for y we get y = 2(e+X)/(e-X).

The initial condition y(O)

= -2 can be solved for,

= -2. The initial condition y(O) = 2 does not correspond to a value


= 2 is a solution of the initial-value problem.
Setting x = ~ and y = 1 in y = 2(e + X)/(e - X) leads to e = -3e. Thus, a solution of the

yielding e = O and y(x)

of e, and it must simply be recognized that y(x)


initial-value problem is
y =2

~3e + x
-3e - e4x

3 - x-l
= 2 ---:----:3 + e4x-l

31

Exercises

30. From (_1._

2.2

Y - 1

+ -1)
y

.!.x dx

dy =

we obtain ln jy - 11-ln Iyl = ln Ixl + C or y =

1-

1
ClX

Another solution is y = O.

(a) If y(O) = 1 then y = 1.


(b)

If y(O) = O then y = O.

1
(e) If y(1/2) = 1/2 then y = 1 + 2x .

i we obtain

(d) Setting x = 2 and y =


1

1-

1
Thus,y=--=--.
1 + ~x
31. Singular solutions of dy/dx
(eX + e-X)dy/dx

= y2 is y

32. Differentiating ln(x2

Cl (2)

1-

2Cl

= 4,

and

Cl

-2'

2
2 + 3x
= x~

are y = -1 and y = 1. A singular solution of

= O.

+ 10) + cscy

= C we get

2x
x2

dy

+ 10 - coty csey dx = O,

2x
x2

+ 10

_ eosy_l_dy
siny sinydx

'

or

2xsin2ydx

- (x2

+ 10) eosydy

= O.

Writing the differential equation in the form


dy
dx

(x2

2xsin2 y
+ 10) eosy

we see that singular solutions oecur when sin2 y = O, or y = k7r, where k is an integer.

32

Exercises 2.2

33.

The singular solution

1 satisfies the initial-value problem.

y
1. 01

-0.004-0.002

0.0020.004

0.002 0.004

0.98

0.97

34. Separating variables we obtain


---

y-1

=x + c

dy

y-1

and

)2 = dx. Then
y

1.02

x+c-1
x+c

= ----

Setting x = O and y = 1.01 we obtain c = -100. The solution is


y=

0.0040.002

x - 101
.
x-lOO

0.99

0.98

35. Separating variables we obtain

dy

)2

y - 1

+ 0.01

= dx. Then

1
x+c
y=l+-tan--.
10
10
Setting x = O and y = 1 we obtain c = O. The solution is
10tan-110(y-1)=x+c

1.0004

and

1. 0002

1
x
y = 1 + 10 tan 10 .
0.9998

0.9996

36. Separating variables we obtain

dy

)2

y - 1

- 0.01

= dx. Then

1.0004

51n 10Y 1

111

10y- 9

= x + c.
1.0002

Setting x = O and y = 1 we obtain c = 51n 1 = O. The solution is


5 In 11Oy -

111 = x.

0.9998

10y- 9

0.9996

33

Exercises 2.2
37. Separating variables, we have
dy
Y - y3

dy
y(l _ y)(l

+ y) =

(1

1/2

y + 1_ y

1/2

- 1 + )dy

dx.

Integrating, we get
In Iyl -

"2 In 11 - yl - "2 In 11 + yl = x + c.

> 1, this becomes

When y

1
1
lny - -ln(y - 1) - -ln(y
2
2
Letting x
x

= Oand

+ 1) = In

y
~
Vy2 - 1

= x + c.

= 2 we find c = In(2/ V3). Solving for y we get Yl (x) =

2ex /

v'4e2x

3, where

> In( V3/2).

When O < y

< 1 we have
1
In y - -ln(l
2

Letting
-00

1
- y) - -ln(l
2

+ y) = In

x = O and y = ~ we find c = In(l/V3).

y
~
VI - y2

= x + c.

Solving for y we get Y2(X)

= eX/v'e2x + 3, where

< x < oo.

When -1

< y < O we have


1
ln( -y) - -ln(l
2

Letting x = O and y
where -00 < x < oo.
When y

1
- y) - -ln(l
2

= -~ we find c = In(l/V3).

+ y)

= In ~

-y

VI _ y2

=x

+ c.
= -ex/v'e2x+3,

Solving for y we get Y3(X)

< -1 we have
1
ln(-y) - -ln(l2

1
y) - -ln(-l2

y) = In

-y

#-=1 = x + c.

= O and y = -2 we find c = In(2/ V3). Solving for y we get


where x > ln( V3/2).
Letting x

34

Y4(X)

-2ex /

v'4e2x

- 3,

Exercises 2.2
y

SX

2
2

-2

-2

-2

-4

-4

-4

SX

l/(y - 3) _
(y - 3)2 -

dy/dx_
(y-1)2-

-I

38. (a) The second derivative of y is


d2y
dx2

1
(y-3)3'

The solution curve is concave up when d2y/dx2

> O or y > 3,

< Oor y < 3. From the phase


<3

and concave down when d2y/dx2

portrait we see that the solution curve is decreasing when y


and increasing when y

> 3.
-2

(b) Separating variables and integrating we obtain

y
8

(y - 3) dy = dx

'21Y 2 y2 _ 6y

3y = x

+e

+ 9 = 2x + el

(y - 3)2 = 2x

+ el

y = 3 ..j2x

-2

+ el.

The initial condition dictates whether to use the plus or minus signo
When Yl(O) = 4 we have q = 1 and Yl(x) = 3 + ..j2x
When Y2(O)

= 2 we have q = 1 and Y2(X) = 3 - ..j2x

+ l.
+ l.

When Y3(1) = 2 we have q = -1 and Y3(X) = 3 - ..j2x - l.


When Y4(-1)

= 4 we have el

= 3 and Y4(X) = 3 + ..j2x + 3.

39. (a) Separating variables we have 2ydy


y( -2)

= -1 we find e = -1, so y2

= (2x
= x2

+ l)dx.

+x

Integrating gives y2 = x2

- 1 and y = -vx2

root is chosen because of the initial condition.

35

+x

+ x + e.

When

- l. The negative square

Exetcises 2.2
y

(b) The interval of definition appcars to be approximately (-00, -1.65).

-7-

4 -3 -

2"

-1_1
-2
-3
-4
-5

+x (-00, -~ - h/5).

(e) Solving x2

O we get x

- ~ ~J5, so the exact interval of definition is

+ 2e3x

40. (a) From Problem 7 the general solution is 3e-2y


3e-2y

+ 2e3x = 5. Solving for y

we get y

= c. When y(O) = O we find c = 5, so

= -~ In 1(5 -

2e3x).
y

(b) The interval of definition appears to be approximately (-00,0.3).

-1
-2

(e) Solving 1(5 - 2e3x) = Owe get z = 1In(~), so the exact interval of definition is (-00, ~ ln(~)).
41. (a) While Y2(X)

= -V25

- x2 is defined at x

= -5

and x

= 5, y~(x) is not defined at these values,

and so the interval of definition is the open interval (-5,5).


(b) At any point on the x-axis the derivative of y( x) is undefined, so no solution curve can cross
the x-axis. Since

-xlv

is not defined when y

= O, the

initial-value problem has no solution.

42. (a) Separating variables and integrating we obtain x2 - y2 = c. Por c

i= O the

graph is a square

= Oand y = X. Since the


differential equation is not defined for y = O, solutions are y = x, x < Oand y = x, x > O.
The solution for y(a) = a is y = x, x> O; for y(a) = -a is y = -x; for .y( -a) = a is y = -x,
x < O; and for y( -a) = -a is y = x, x < O.
hyperbola centered at the origino All four initial conditions imply c

(b) Since x/y is not defined when y = O, the initial-value problem has no solution.
(e) Setting x

= 1 and y = 2 in x2 - y2 = c we get c = -3, so y2 = x2 + 3 and y(x)

= vx2

+ 3,

where the positive square root is chosen because of the initial condition. The domain is all real
numbers since x2

+ 3 > Ofor all X.


36

Exercises 2.2

43. Separating variables we have dy/(

I+Y2 sin2 y)

= dx

y
3.5

which is not readily integrated (even by a CAS). We


note that dy / dx

2': O for all values of x and y and that

dy/dx = O when y = O and y = 7f, which are equilibrium


solutions.
-6

44. Separating variables we have dy / (.jY

+ y) = dx / ( .x + x).

-4

-2

To integrate

J dt / ( Vi + t)

8 x

we substitute

u2 = t and get

J u+u

du

~2

J l+u
_2_

du = 2ln 11+ ul

+ C = 2ln(1 + vX) + c.

Integrating the separated differential equation we have


2ln(1 +.jY) = 2ln(1
Solving for y we get y

+ vX) + C

or

ln(l +.jY) = ln(l

+ vX) + ln cj.

= [C1(1+.x) - 1J2.

45. We are looking for a function y(x) such that


y2

(~~r
= 1.

Using the positive square root gives


dy
dx

=~

Thus a solution is y = sin(x

~ dy

=:}

v=

= dx

=:}

. -1
sm y

= x + C.

_y2

+ c). If we use the negative square root we obtain

y = sin(c - x)

=-

sin(x - c)

=-

sin(x

+ Cl)'

= -1 are solutions.

Note also that y = 1 and y


46. (a)

(b) Por Ixl > 1 and Iyl > 1 the differential equation is dy / dx = ~
variables and integrating, we obtain
dy
dx
and cosh-l y = cosh-1 x

vX2=l

37

/.JX2=l.
+ c.

Separating

Exercises 2.2
Setting x

= 2 and y = 2 we find c = cosh-1 2 - cosh-1 2 = O and cosh-1 y = cosh'"! x. An

explicit solution is y = x.
47. (a) Separating variables and integrating, we have
(3y2

+ l)dy =

-(8x

+ 5)dx

and

y3

+ Y = -4x2

Using a CAS we show various contours of f(x, y)

- 5x
= y3

+ c.

+y+4x2

5x. The plots shown on [-5,5] x [-5,5J correspond to c-values


of O, 5, 20, 40; 80, and 125.

(b) The value of c corresponding to y(O) = -1 is f(O, -1) = -2;


to y(O)

= 2 is f(0,2)

and to y(-l)

= 10; to y(-l)

= 4 is f(-1,4)

-3 is -31.

48. (a) Separating variables and integrating, we have

(-2y

+ y2)dy

= (x - x2)dx

and
-y

2131213

+ 3'y

= 2"x - 3'x

+ c.

Using a CAS we show sorne contours of f(x, y)


2y3 - 6y2

+ 2x3

= 67;

- 3x2. The plots shown on [- 7, 7] x

[-5,5J correspond to e-vales of -450, -300, -200,


-120, -60, -20, -10, -8.1, -5, -0.8, 20,60, and
120.

38

Exercises 2.3

(b) The value of e corrcsponding to y(O) = ~ is f(O, ~) =


- 247.
The portion of the graph between the dots cor-

responds to the solution curve satisfying the intial condition.

To determine the interval of definition we find

dy/dx

for 2y3 - 6y2

+ 2x3

differentiation we get y'

- 3x2

= -~.

(x - x2)/(y2

Using implicit

-2

- 2y), which

-4

is infinite when y = O and y = 2. Letting y = O in


2y3 - 6y2 + 2x3 - 3x2 = - 247 and using a CAS to solve
for x we get x = -1.13232. Similarly, letting y

-2

= 2, we find x

= 1.71299. The largest interval

of definition is approximately (-1.13232,1. 71299).


(c) The value of e corresponding to y(O) = -2 is f(O, -2) =

-40. The portion of the graph to the right of the dot


corresponds to the solution curve satisfying the initial
condition. To determine the interval of definition we find
dy / dx for 2y3 - 6y2 + 2x3 - 3x2 = -40. Using implicit

-2
-4
-6

= (x - x2)/(y2 - 2y), which


is infinite when y = O and y = 2. Letting y = O in

differentiation we get y'

+ 2x3

2y3 - 6y2

for x we get x

-8
-4

-2

10

- 3x2 = -40 and using a CAS to solve

= -2.29551. The largest interval of definition is approximately (-2.29551, (0).

Exercises 2.3

1. For y' - 5y = O an integrating factor is e-J 5 dx


-00

+ 2y =

= e2x so that

O an integrating factor is eJ2dx

+y

= e3x an integrating factor is eJ dx

= ce-2x for

dx

+ 4y

-00

1 an integrating

factor is eJ 4 dx

-00

x so that

d~ [e4Xy]

= 1x and y =

1 + ce-4x

< x < oo. The transient term is ce-4x.

5. For y' +3x2y = x2 an integrating factor is eJ 3x2 dx


for

d~ [e2Xy] = O and y

= eX so that ..!!:_ [eXy] = e4x and y = -41e3x+ ce?" for

< x < oo. The transient term is ce-x.

4. For y'
for

= O and y = ce5x for

< x < oo. The transient term is ce-2x.

3. For y'
-00

d~ [e-5Xy]

< x < oo.

2. For y'
-00

= e-5x so that

< x < oo. The transient term is ce-x3.


39

ex3 so that ddx [ex3 y] = x2 ex3 and y = i+ce-x3

Exercses 2.3
x3 an integrating factor is eJ 2xdx

+ 2xy =

6. For y'

~x2 - ~ + ee-x2 for

for O < x

< x < oo. The transient term is ce-x2

+ -11y =

7. For y'

-00

2" an integrating factor is e (l/x)dx

< oo.

= eX so that

1
= x so that -d [xy] = -1 and y = -lnx
+ -e

dx

= x2 + 5 an integrating factor is e- J2dx = e-2x so that d~ [e-2Xy] = x2e-2x + 5e-2x


and y = _~x2 - ~x -1} + ee2x for -00 < x < oo.

8. For y' _ 2y

= ~ so that .!!:_ [~y]

9. For y' _ ~y = xsinx an integrating factor is e- J(l/x)dx


x
y = ex - x cos x for O < x < oo.

+ ~y = ~

10. For y'

for O < x

y = ~x3 - ~x
12. For y'_

13. For y'


y

1 eX
=- ?

2 z-

14. For y'

x4 so that

ddx [x4y]

+~
x
y

x6

x4 and

< oo.

+ --ce"
x+1

for O < x

for -1

< x < oo.

= x2ex

factor is eJ[1+(2/x)]dx

< oo. The transient term is

so that

ce:"
x

-2- .

~e-x sin 2x an integrating factor is eJ[1+(l/x)]dx = xe" so that .!!:_ [xeXy]


x
~
1
ee-x
__ e-x cos 2x + -- for O < x < oo. The entire solution is transient.
2x
x

16. For dx
dy
2

4y5 an integrating factor is e- J(4/y)dy = y-4 so that

for O < y
=

< oo.

eY an intezratinz factor is eJ(2/y)dy


o

= y2 so that

-eY + 2" + 2" for O < y < oo. The transient term is
Y

= 3x and y = ~ + ex-2

[x2y]

+ (1 + ~) Y =

+ cy4

dx

= sinx and

dx

ce:"

2y6

eY

for O < x

2x + 3
and y = -x - -x+1

15. For ~~ - ; x

.!!:_

= x an integrating factor is e-f[x/(1+x)]dx = (x+ l)e-X so that .!!:_ [(x + l)e-Xy] =

+ -2x

sin 2x and y

+ ex-4

( 1 + -x2) y = -x2eX an integrating

so that

x2 - 1 an integrating factor is eJ(4/x)dx

x
y
(1 + x)

+ l)e-

x(x

= x2

dx

< oo.

+ ~y

11. For y'

an integrating factor is eJ(2/x)dx

40

2+e

-2- .

:y [y-4x]

4y and

Exercses 2.3

= sec x an integr ating factor is eJ


y = sin x + e cos x for -7f /2 < x < 7f/2.

17. For y' + (tan x)y

tan

x dx

18. For y' + (cot x)y = sec2 x ese x an integrating factor is eJ


and y = sec x + e ese x for O < x

= 2xe-x

19. For y' + x + 2 y


x+1
x2
2x and y = _x+1
20. For y' + _4_
x+2

dr
de
+r

e-x + -x+1

sec e

ce-3x
y = e-3x + -x

for O < x

= (x+1)eX,

y] = sec2 x

so'!!:_ [(x + l)eXy]


dx

--.iy
x
-1

< oo. The entire solution is transient.

= cos e an integrating factor is eJ secadO

(3 +~)

25. For y ,1+ - y


x

e-x for -1 < x

dx

(4t - 2)et2-t and P

and (x - l)y

< 7f/2.

.!!:_ [( sin x)

= 3(x + 2)-1 + c(x + 2)-4 for -2 < x < oo. The entire solution is transient.

22. For ~~ + (2t - l)P

24. For y' +

x dx = sin x so that

Y = ( 5 )2 an integrating factor is eJ[4/(x+2)]dx = (x + 2)4 so that dd [(x + 2)4y]


x+2
x

1 + sine and r(sece + tan ')

23. For y' +

cot

an integrating factor is eJ[(x+2)/(x+1)]dx

x+1

5(x + 2)2 and y


21. For

= sec x so that :x [(sec x) y] = sec2 X and

= sec e+tan

e so that ded

[r(sec e + tan

= e - cose + e for -7f/2 < e < 7f/2 .

4t - 2 an integrating factor is eJ(2t-1)dt

= 2 + cet-t2 for

-00

et2-t so that :t [pet2-t]

< t < oo. The transient term is c-t2.

= e~3X an integrating factor is ef[3+(1/x)]dx = xe3x so that ~ [xe3Xy]


for O < x

e)] =

1 and

< oo. The transient term is ce-3x [.

= x + 1 an integrating factor is eJ[2/(x2-1)Jdx = x - 1 so that

x-1
x(x + 1) + c(x + 1) for -1

x+1

< x < l.

= -1 ex an integrating factor is e J(1/ x )dx


x

< oo. If y(l) = 2 then e = 2 - e and y

1
x

so that

2- e
.
x

.!!:_ [X dx

x+1

1y] - 1

1
-d []xy = e x and y = -e
dx
x

+- e
x

= 2 and x = 2y2

+ cy

= _ex + --

26. For dx _ ~ x = 2y an integrating factor s e- J(1/y)dy = ~ so that


dy
y
y
49
for -00 < y < oo. If y(l) = 5 then e = -49/5 and x = 2y2 -

.!!:_
dy

[~x]
y

s-Y'

27. For di + !.!:. i = E an integrating factor is eJ (R/ L) dt


dt
L
L

i = ~ + ce-Rt/L

for

-00

< t < oo. If i(O)

= eRt/ L so that

= io then e = io

!!_
dt

[ieRt/ L]

E eRt/ L and
L

- E/ R and i = ~ + (io - ~) CRt/L.

28. For dT - kT = -Tmk an integrating factor is eJ(-k)dt = e-kt so that dd [Te-kt] = -Tmke-kt
dt
t
T = Tw. + cekt for -00 < t < oo. If T(O) = To then e = To - Tm and T = Tm + (To - Tm)t.

41

and

Exercises 2.3

= Inx an integrating factor is ef[l/(x+l))dx = x + 1 so that ~[(x + l)y] = ln z


29. For y' + _l_y
x+1
x+1
~
x
x
C
and y = -In x - -+ -for O < x < oo. If y(l) = 10 then C = 21 and y =
x+1
x+1
x+1
x
x
21
--Inx---+--.
x+1
x+1
x+1
= cos2 x an integrating factor is eftanxdx = secx so that ~ [(secx) y] = cosx and
dx
y = sinxcosx + ccosx for -1f/2 < x < 1f/2. If y(O) = -1 thenc = -1 and y = sinxcosx - cosx.

30. For y' + (tanx)y

31. For y' + 2y =

f (x) an integrating factor is e2x so that


ye

2x

={

-2le2x

+ ci ,

C2,

If y(O) = O then
C2

-1/2

Cl

O:::; x < 3;
x> 3.

and for continuity we must have

1"x

= ~- ~so that
O :::; x :::;3;

x> 3.
32. For y' + y =

f (x) an integrating factor is e" so that


O:::;x:::;l;

x>l.
If y(O)

= 1 then

Cl

Oand for continuity we must have

so that

ye

= 2 then

Cl

2e

x>l.

an integrating factor is eX so that


x2

{~ex2
C2,

If y(O)

O :::; x :::;1;

1,
y= { 2el-x-1,
33. For y' + 2xy = f(x)

C2

+Cl,

O:::; x:::;

1;

x> l.

= 3/2 and for continuity we must have

C2

1e + ~ so that
O :::; x :::;1;

x>l.

42

Exercises 2.3
~'
1+ x
- { -x
1 + x2'

,2x
Y +--y1 + x2

34. For

O:::;x:::;l;

1
x

an integrating factor is 1 + x2 so that

( 1 +x
If y(O)

= O then

Cl

= O and

-1

O:::; x:::;

2) y = { ~x2 + ci,
_~x2+C2,

x>l.

for continuity we must have


1
- -

_
y -

C2 =

2 (1, + x2) ,
3
1

2 (1 + x

1;

2) - -,

< x < l'

x>

1 so that
,

l.

35. We need

P(x)dx

2X,

O:::;x:::;l

-21nx,

x>

20

15

An integrating factor is

eJ

P(x)dx =

2x

e ,

10

O:::;x:::;l

{ 1/x2,

and
d [{ ye2x,
dx
y / x2 ,

O:::; x < 1] = { 4xe2x,


x >1
4/ x,

O:::; x < 1
>1

Integrating we get
ye2x,
{ y / x2 ,

Using y(O)

= 3 we find

Cl =

O:::; x :::;1 = { 2xe2x - e2x


x > 1
41n x + C2,

+ Cl,

4. For continuity we must have

2x - 1 + 4e-2x, .
{
Y = 4x21nx + (1 + 4e-2)x2,

36. An integrating factor for y' - 2xy

C2

O:::; x :::;1 .
x > 1

= 2 - 1 + 4e-2 = 1 + 4e-2.

O :::; x :::;1
x >1

= 1 is e-x2. Thus

-d [_X2
e y 1 =e

_x2

dx

2
[X
e-X y = Jo e-t

dt

erf(x)

and
2

Y = eX erf(x)

. 43

+ ce" .

+c

Then

Exercises
Frorn y(l)

2.3
= 1 we get 1= e erf(l)

Y
37. For y'

+ eXy =

+ ce,

so that c = e-l - erf(l). Thus

1 an integrating factor is
d

When y'

= 1 we get e = e, so y

+ eXy

= e-e

= eX

-1

+ eX (erf(x) - erf(l)).

and
t

fox dt

ee y

t
= lox
o dt + c.

+ e 1-e x .

= O we can separate variables and integrate:

dy
y

= _ex dx

In jyj

and

Thus y = Cle-e". From y(O) = 1 we get Cl = e, so y


When y'

Thus

- lee y] = ee
dx
Frorn y(O)

= eX erf(x) + (e-l - erf(l))eX

+ eXy = e"

we can see by inspection that y

_ex

+ c.

= el-e".
= 1 is a solution.

38. We want 4 to be a critical point, so use y' = 4 - y.


39. (a) Al! solutions of the form y = x5ex - x4ex
since 4/ x is discontinuous at x

+ cx4

satisfy the initial condition.

In this case,

= O, the hypotheses of Theorem 1.1 are not satisfied and the

initial-value problem does not have a unique solution.


(b) The differential equation has no solution satisfying y(O) = Yo, yO =j:. O.
(c) In this case, since xo =j:. O, Theorem 1.1 applies and the initial-value problem has a unique
solution given by y
40. On the interval (-3,3)

= x5ex

- x4ex

+ cx4

where c = yo/x~ - xoexQ

+e

XQ

the integrating factor is


efxdxj(x2-9)

= e- fxdxj(9-x2)

= e1In(9-x2)

)9 _ x2

and so

_:!_ [) 9 - x2
dx

41. We want the general solution to be y


approaches O as x

-> 00

y]

=O

and

y = ";9 c-

= 3x - 5 + ce:", (Rather than e-x, any function that

could be used.) Differentiating we get


y' = 3 - ce-x

so the differential equation y'

+y

=3-

(y - 3x

+ 5) = -y + 3x

= 1 is l/e and the right-hand derivative at x = 1 is

1 - l/e. Thus, y is not differentiable at x = 1.


= c/x3

- 2,

= 3x - 2 has solutions asymptotic to the line y = 3x - 5.

42. The left-hand derivative of the function at x


43. (a) Differentiating Yc

"
x2

we get
,
3c
3 c
3
y - -_ - -- _ - --y
e x4 x x3 x

44

Exercises

+ 3y = O. Now

so a differential equation with general solution Yc = c/x3 is xy'


xY~

+ 3yp = x(3x2) + 3(x3)

2.3

= 6x3

so a differential equation with general solution y = c/x3

+ x3

is xy'

+ 3y

= 6x3. This will be a

general solution on (O, (0).


(b ) Since y(l)

y(l)

13

= 13 - 1/13

+ 2/13

O, an initial condition is y(l)

= 3, an initial condition is y(l)

= O. Since

= 3. In each

case the interval of definition is (O, (0). The initial-value problem


xy'

+ 3y = 6x3,

= O has

y(O)

solution y

= x3 for

-00

< x < oo.

-3

(e) The first two initial-value problems in part (b) are not unique. For example, setting y(2) =
23 - 1/23 = 63/8, we see that y(2) = 63/8 is also an initial condition leading to the solution

y = x3 - 1/x3.
44. Since eJ P(x)dx+c

= eJ P(x)dx

C1eJ P(x)dxy

C2

= cleJ

r=.

we would have

C1eJ P(x)dx f(x) dx

and

eJ P(x)dxy

C3

eJ P(x)dx f(x) dx,

which is the same as (6) in the texto


45. We see by inspection that y = O is a solution.
46. The first equation can be solved by separation of variables. We obtain x
we obtain

C1 = Xo

and so x

= xoe-Alt.

cle-Alt.

The second equation then becomes


or

which is linear. An integrating factor is eA2t; Thus


.!!._ [eA2ty 1 = xOA1e-AlteA2t

= XOA1e(A2-Al)t

dt

e A2ty = \ XOA1\ e (A2-Al)t + C2


/\2 - /\1

From y(O) = Yo we obtain

C2

y=

= /\2
\ XOA1
\
- /\1

-Alt

+ C2e -A2t

= (YOA2 - YOA1 - XOA1)/(A2 - Al)' The solution is


XOA1 -Alt
e
A2 - Al

+ YOA2 - YOA1 - XOA1 e -A2t


A2 - Al

45

Frorn

x(O) = Xo

Exercises 2.3
47. (a) Letting y = YI

+ Y2

+ Y2(XO)

we have y(xo) = YI(XO)

+ O= a

=a

and

yl + P(x)y = (YI + Y2)1 + P(X)(YI + Y2)


= (Yi + P(X)YI) + (y; + P(X)Y2) = O + (x) = (x).
Thus y = YI

+ Y2

is a solution of the initial-value problem yl

+ P(x)y

= (x),

y(xo) = a.

(b) By Theorem 1.1 the initial-value problem

yl + P(x)y = h(x)

+ h(x),

(1)

y(xo)=a+,6

has a unique solution. Consider y = YI +Y2 Since y(xo) = YI(XO) +Y2(XO) = a+,6,
the initial condition of (1). Also

yl + P(x)y = (YI + Y2)1 + P(x)(Yl


= (y~

y satisfies

+ Y2)

+ P(X)YI) + (y~ + P(X)Y2) = h(x) + f2(x),

so y satisfies the differential equation in (1). Thus y = Yl +Y2 is the unique solution of (1) and

y(xo) = a +,6.
Since

CIYI

yl + P(x)y

is a solution of yl

+ P(x)y

e-X2 y

ft/2,

Y = eX 2

=_

_x2

(X

Jo

e-t2 dt = _ ft erf(x)
2

and noting that erf(O) = O, we get


--

ft erf(x) + -J1F)
2

CIa,

yl - 2xy = -1 is e-x2. Thus

d [ _x2 1
dx e y =-e

From y(O) =

y(xo)

and C2Y2 is a solution of


= C2f2(X), y(xo) = C2,6, we see by the aboye argument that y(xo) = CIa + c2,6.

48. (a) An integrating factor for

(b)

= Clil(X),

+ c.

ft /2.

Thus

= -ft eX 2 (1 - erf(x)) = -ft eX 2 erfc(x).


2
2

Using Mathematica we find y(2) ~ 0.226339.

49. (a) An integrating factor for


I
2
y +-Y=
x

10sin x
3
x

46

Exercses 2.3

is x2. Thus
d [ 2 1
-xy=l-dx

Osinx
X

x 2 y = 10
y
From y(l)

= Owe get e = -lOSi(l).

la

sin t
-dt

+C

10x-2Si(x)

+ cx-2.

Thus

y = lOx-2Si(x) - lOx-2Si(1) = 10x-2(Si(x) - Si(l)).


y
2

(b)

x
-1

-2
-3
-4
-5

(e) From the graph in part (b) we see that the absolute maximum occurs around x = 1.7. Using
the root-finding capability of a CAS and solving y'(x) = O for x we see that the absolute
maximum is (1.688,1.742).
50. (a) Separating variables and integrating, we have
dy
. 2
Y
= sin x dx
Now, letting t

and

(X
2
In Iyl = Jo sin t dt

+ c.

;tiu we have

rrr2, Jo(y2;; X sin( 2,u2)du,

(X
Jo sin t2 dt = V

tt

so
y

cleJ;sint2dt = CIe#Jo.2!;x

Using S(O) = O and y(O) = 5 we see that

Cl

sin(1ru2/2) du

= 5 and y

(b)

-JO

-5

JO

47

= cle#S(y2;;x).

= 5e#S(y2;;x).

Exercises 2.3
(e) Frorn the graph we see that as x
9.35672. Since limx_oo 5(x)
-~, limx__:_ooy(x) = 5e-.;;s

-+ 00,

y(x) oscillates with decreasing amplitude approaching

= ~, limx_oo y(x) = 5e.;;s

~ 9.357, and since lim.,__

oo

S(x) =

~ 2.672.

(d) Frorn the graph in part (b) we see that the absolute maximum occurs around x
absolute minimum occurs around x = -1.8.

= 1.7 and the

Using the root-finding capability of a CAS and

solving y'(x) = O for x, we see that the absolute mximum is (1.772,12.235) and the absolute
minimum is (-1.772,2.044).

Exercises 2.4
1. Let M = 2x-1

h'(y)

and N = 3y+7 so that My = O = Nx. From fx

= 3y + 7, and h(y)

= ~y2 + 7y. The solution is x2 -

= 2x-1

we obtain f

+ ~y2 + 7y

= x2-x+h(y),

= c.

2. Let NI = 2x + y and N = -x - 6y. Then My = 1 and Nx = -1, so the equation is not exacto

= 5x + 4y and N = 4x - 8y3 so that My = 4 = Nx. From fx = 5x + 4y we obtain


f = ~x2 + 4xy + h(y), h'(y) = _8y3, and h(y) = _2y4. The solution is ~x2 + 4xy - 2y4 = C.
4. Let j1I = siny - ysinx and N = cosx + xcosy - y so that My = cosy - sinx = Nx' From
3. Let M

fx = sin y - y sin x we obtain f

= x sin y + y cos x + h(y),

h'(y)

= -y,

and h(y) =

h2. The solution

is x sin y + y cos x - ~y2 = C.

= 2y2x - 3 and N

5. Let M

1=

= 2yx2 + 4 so that NIy

x2y2 - 3x + h(y), h'(y) = 4, and h(y)

4xy

Nx'

Frorn fx = 2y2x - 3 we obtain

= 4y. The solution is x2y2 - 3x + 4y = C.

6. Let M = 4x3 - 3y sin 3x - y/x2 and N = 2y - l/x + cos 3x so that My


N = 1/x2 - 3sin3x.

= -3 sin 3x - 1/x2 and

The equation is not exacto

7. Let M = x2 - y2 and N

= x2

- 2xy so that My = -2y and Nx

2x - 2y. The equation is not

exacto
8. Let M = 1 + lnx+y/x

1=

and N = -1 +lnx so that NIy = l/x = Nx' From Iy = -1 +lnx we obtain

-y + ylnx + h(y), h'(x) = 1 + lnx, and h(y) = x lnx. The solution is -y + ylnx + xlnx = C.

9. Let M = y3 - y2sinx

- x and N = 3xy2 + 2ycosx

Ix = y3 - y2sinx - x we obtain I = xy3 + y2cosx


solution is xy3 + y2 cos X - ~x2 = C.

ix

11. Let M

= 3y2 - 2ysinx = Nx' From

- ~x2 + h(y), h'(y) = O, and h(y) = O. The

= x3 + y3 and N = 3xy2 so that My = 3y2 = Nx.

10. Let NI

1=

so that My

From Ix = x3 + y3 we obtain

+ xy3 + h(y), h'(y) = O, and h(y) = O. The solution is x4 + xy3 = C.

= ylny - e-xy and N = l/y + xlny so that My = 1 + lny + ye-xy and Nx = lny. The

equation is not exacto

48

Exercises 2.4
12. Let M
obtain

= 3x2y + eY and N =
f = x3y + xeY + h(y),
=

13. Let M
f

= xy

14. Let M

+ 2ex + h(y),

= 1-

3/x

= 1 = Nx. Frorn fx = y - 6x2 = O. The solution is xy - 2x3 -

x so that My

= O,and

h' (y)

y and N = 1 - 3/y

x - 31n Ixl + xy

f =

+ y + xy

+ h(y),

h(y)

+ x so that

h'(y)

= 1=

My

1 - ~, and h(y)

From

Nx'

2xex we obtain
2xex

+ 2ex = c.

= 1-

fx

+y

3/x

y - 31n Iyl. The solution is

- 31n Ixyl = c.

15. Let M

y - 6x2 - 2xex and N

- 2x3 - 2xex

we obtain
x

+ xeY - 2y so that My = 3x2 + eY = Nx' From fx = 3x2y + eY we


h'(y) = -2y, and h(y) = _y2. The solution is x3y + xe" - y2 = c.

x3

1/ (1 + 9x2)

x2y3 -

1/ (1 + 9x2)

we obtain
= c.

and N

= !x3y3

x3y2 so that My

-!arctan(3x)

+ h(y),

3x2y2

Nx'

h'(y) = O, and h(y)

From

= x2y3 -

fx

= O. The solution is

x3y3 - arctan(3x)

16. Let M = -2y and N

so that

5y-2x

My

= -2

+ ~y2

h'(y) = 5y, and h(y) = ~y2. The solution is -2xy

= tanx -

17. Let M

and N

sinxsiny

tan x - sin x sin y we obtain


is In Isec xl
18. Let M

From

fx

+ cos x sin y

= In Isec xl

-2xy+h(y),

= c.

so that

cosxcosy

-2y we obtain f

My

+ cos x sin y + h(y),

-sinxcosy

h' (y) = O, and h(y)

Nx'

From

= O. The

fx

solution

= c.

2y sin x cos x - y

= Nx. From fx

2ysinxcosx

+ 2y2exy2 and

- 1+ 4xy eXY

My

- y

+ 2y2exy2 we obtain

2sinxcosx

h(y) = O. The solution is ySin2 x - xy

+ sin2 x + 4xyexy2 so that

N = -x

+ 2exy2

f =

+ 4yeXY

ysin2 x - xy

= Nx'

+ 2exy2 + h(y),

h'(y)

= O, and

= c.

= 4t3Y-15t2-y
and N = t4+3y2-t
so that My = 4t3-1 = Nt. From ft = 4t3Y-15t2-y
we
obtain f = t4y - 5t3 - ty + h(y), h' (y) = 3y2, and h(y) = y3. The solution is t4y - 5t3 - ty + y3 = c.

19. Let M

20. Let M = l/t+


Nt Frorn

ft =

1/t2-y/
l/t

and h(y) = yeY

(t2

+ 1/t2

+ y2) and

- y/ (t2

N = yeY +t/

= x2+2xy+y2
and N
we obtain f = !x3 + x2y +
!x3

My = (y2 _ t2) / (t2

(f) +

+ y2)2

h(y), h'(y)

= yeY,

eY. The solution is

21. Let M

+ x2y + xy2
+ x2y + xy2

+ y2) so that

+ y2) we obtain f = In Itl- ~ - aretan

In Itl- ~ - aretan

!x3

(t2

= 2xy+x2-1
xy2

h(y),

(f) +

so that
h'(y)

My

yeY

eY

= 2(x+y)

-1, and h(y)

= c.
= Nx' From fx = x2+2xy+y2
= -y. The general solution is

- Y = c. lf y(l) = 1 then c = 4/3 and the solution of the initial-value problem is


- Y=

1.
49

Exercses 2.4

= eX+y

22. Let M

and N

= 2+x+yeY

h(y), h'(y) = 2 + yeY, and h(y)


If y(O)

= 1 then e = 3 and

so that My

= 2y + yeY

= 1 = Nx'

From Ix

= eX+y

we obtain I = eX+xy+

y. The general solution is eX + xy

the solution of the initial-value problem is eX + xy

+ 2y + ue" + 2y + yeY -

eY = c.
eY

= 3.

+ 2t - 5 and N = 6y + 4t - 1 so that My = 4 = Ni, Prom!t = 4y + 2t - 5 we


obtain I = 4ty + t2 - 5t + h(y), h'(y) = 6y - 1, and h(y) = 3y2 - y. The general solution is
4ty + t2 - 5t + 3y2 - y = c. If y( -1) = 2 then e = 8 and the solution of the initial-value problem is
4ty + t2 - 5t + 3y2 - y = 8.

23. Let M = 4y

24. Let M

= t/2y4 and N = (3y2 - t2) /y5 so that My = -2t/y5

3
1=-4
+
h(y), h'(y) = 3' and h(y)
4y
y
t2

3
--2
2y

Ni. From

It

= t/2y4 we obtain

3
. The general solution is -4 --2
4y
2y
t2

then e = -5/4 and the solution of the initial-value problem is ~ - -;.


4y
2y

= c.

If y(l)

=1

= -~

- 3x2y - 2x and N ='2y sinx .; x3 + lny so that My = 2ycosx


- 3x2 = Nx. From
Ix = y2 cos X -3x2y -2x we obtain 1= y2 sinx _x3y _x2 +h(y), h'(y) = In y, and h(y) = y In y-y.

25. Let M = y2 cos X

The general solution is y2 sin x - x3y - x2 + y In y - y = c. If y(O)


of the initial-value problem is y2 sin x - x3y - x2 + y In y - y = O.

= y2 + Y sin x and N = 2xy - cos x-1/

26. Let M
y2

+ ysinx

we obtain

= xy2 - Y cos x

(1

+ h(y),

+ y2)
=

+ 4kxy3

and N

= 3y2 + 40xy3

-1

--2

l+y
1 then e

initial-value problem is xy2 - Y cosx - tan-1 y = -1 27. Equating My = 3y2

so that My = 2y

h'(y) =

solution is xy2 - Y cos x - tan-1 y = c. If y(O)

From

Iy = 2x2y

+ 2xy2

cos x and N

cos x we obtain

I = x2y2

+ sin x = Nx'

From Ix =

'

and h(y) = - tan-1 y. The general

-1 - tt/ 4 and the solution of the

we obtain k

= 10.

28. Equating My = 18xy2 - sin y and N = 4kxy2 - sin y we obtain k


29. Let M = _x2y2 sin x

= e then e = O and the solution

= 9/2.

= 2x2y cos X so that My = _2x2y sin x + 4xy cos x


cos X

+ h(y),

h' (y)

= O, and

h(y)

= O. The

= Nx'

solution of the

differential equation is x2y2 cos x = c.


30. Let M

and N = (y2+2xy-x2)/(y2+2xy+x2)

(x2+2xy_y2)/(x2+2xy+y2)

My = -4xy/(x

+ y)3 = Nx'

h(y), h'(y) = -1, and h(y)

From Ix = (x2

+ 2xy + y2

- 2y2) /(x +y)2 we obtain I

so that
2y2

= x + --

= -y. The solution of the differential equation is x2 + y2 =

x+y
c(x + y).

+ 3x2
+ h(y),

31. We note that (My - Nx)/ N = s.], so an integrating factor is eJ dx/x = x. Let M = 2xy2
and N = 2x2y so that My
h'(y)

= O, and h(y)

= 4xy =

Nx' Frorn Ix

= 2xy2 + 3x2

we obtain I = x2y2

= O. The solution of the differential equation is x2y2

+ x3

= c.

32. We note that (My - Nx)/N = 1, so an integrating factor is eJ dx = e", Let M = xyeX

50

+ x3

+ y2ex + yeX

Exercises 2.4
and N = xex + 2yeX so that My = xex + 2yeX + eX = N.
I = xyeX + y2ex + h(x),
xyeX

+ y2ex

h'(y) = O, and h(y) = O. The solution of the differentia1 equation is

= c.

33. We note that (Nx - My)/M

= 2/y, so an integrating factor is eJ2dy/y = y2. Let M = 6xy3 and

N = 4y3 + 9x2y2 so that My = 18xy2


h' (y)

From Iy = xe" + 2yeX we obtain

Nx'

From Ix = 6xy3 we obtain I = 3x2y3 + h(y),

= 4y3, and h(y) = y4. The solution of the differentia1 equation is 3x2y3 + y4

34. We note that (My - Nx)/N


cosxcscx

cotx,

= -

so an integrating factor is e- Jcotxdx

= cotx and N = (1 + 2/y) sinxcscx

we obtain I = 1n(sinx)+h(y),

h'(y)

equation is 1n(sin x) + y + ln y2

cscx.

Let M

= 1 + 2/y, so that My = 0= Nx. From Ix = cotx

= 1+2/y,

and h(y) = y+1ny2. The solution ofthe differentia1

= c.

= 3, so an

35. We note that (My - Nx)/N

= c.

integrating factor is eJ3dx

= e3x.

Let

= (10 - 6y + e-3x)e3x = 10e3x - 6ye3x + 1

M
and

so that My
h' (y)

= O, and

_6e3x + Nx. From Ix

10e3x - 6ye3x + 1 we obtain I == 130e3x- 2ye3x + x + h(y),

h(y) = O. The solution of the differential equation is

36. We note that (Nx - My)/M

= -3/y,

13 e3x

so an integrating factor is e-3Jdy/y

M = (y2

+ xy3)/y3

= l/y

- 2ye3x + x
= 1/y3.

= c.

Let

+x

and
N = (5y2 - xy + y3 sin y)/y3 = 5/y - x/y2 + sin y,
so that My = _1/y2

= Nx. From Ix = l/y+x

we obtain I

= x/y+~x2+h(y),

h'(y) = 5/y+siny,

and h(y) = 51n Iyl- cos y. The solution of the differentia1 equation is x/y + ~x2 + 51n Iyl- cos y = c.
37. We note that (My - Nx)/N
Let M = x/(4+x2)

- 2x/(4 + x2), so an integrating factor is e-2 J xdx/(4+x2) = 1/(4 + x2).

and N = (x2y+4y)/(4+x2)

= y, so that My = O = Nx. From Ix = x(4+x2)

we obtain I = ~ 1n(4 + x2) + h(y), h'(y) = y, and h(y) = ~y2. The solution of the differentia1
equation is ~ 1n(4 + x2) + ~y2

= c.

38. We note that (My - Nx)/ N = -3/(1 + x), so an integrating factor is e-3 J dX/(l+x) = 1/(1 +x)3.
M = (x2+y2-5)/(1+x)3
From Iy = -y/(l

and N = -(y+xy)/(1+x)3

+ x)2 we obtain I =

= -y/(1+x)2,

-h2/(1 + x)2

+ h(x),

so that My = 2y/(1+x)3
h'(x)

2
+ (1+x)2

2
+ (l+x)

51

+lnI1+xl

= Nx.

= (x2 - 5)/(1 + x)3, and

h(x) = 2/(1 + x)2 + 2/(1 + x) + ln 11+ z]. The solution of the differential equation is
y2
-2(1+x)2

Let

= c.

Exercises 2.4
39. (a) Implicitly differentiating x3
3x2

+ 2x2

dy
dx

+ 2x2y + y2 = c and
+ 4xy + 2y

we obtain

dy = _ 3x2 + 4xy .
dx
2x2 + 2y

dy = O and
dx

+ 3x2)dx + (2y + 2x2)dy = O.


+ 2x2y + y2 = c we find c = 4, and

Separating variables we get (4xy

= O and y = -2 in x3

(b) Setting x

solving for dy/dx

setting x

= y = 1 we also

find c = 4. Thus, both initial conditions determine the same implicit solution.
(e) Solving x3

+ 2x2y + y2 = 4 for y

,.y

we get

and

40. To see that the equations are not equivalent consider dx

.t(x, y)

=y

resulting in y dx

+ x dy = O. A solution

= (x/y)dy

= O. An integrating factor is

of the latter equation is y

= O, but this is not a

solution of the original equation.


41. The explicit solution is y = .r-(3-+-c-o-s2-x-)-/-(-1---x-2-).
Since 3 + cos2 X

> O for al! x we must have

> O or -1 < x < 1. Thus, the interval of definition is (-1, 1).


(a) Since fy = N(x, y) = xexy + 2xy + l/x we obtain f = eXY + xy2 + JI... + h(x) so that fx
1 - x2

42.

y
yexy + y2 - 2

(b) Since fx

+ h'(x).

= M(x, y)

so that fy

Let M(x, y)

+x

(x2

= yl/2x-l/2

= y-l/2xl/2

y
yexy + y2 - 2'

+ y r1

+ ~ (x2 + yr1 + g'(x).

we obtain f

= 2yl/2xl/2 + ~ In Ix2 + yl + g(y)

= y-l/2xl/2

Let N(x,y)

+ ~ (x2 +yr1.

43. First note that

d( /x2 + y2) =

x
/x2

Then x dx

+ y dy

= /

x2

+ y2 dx
x

+ y2

dx

+
/ x2

The left side is the total differential of / x2


Thus / x2

+ y2 = X + c is a

dx

y
'x2

+ y2

dy.

becomes

-===

/ x2

+ y2

+ y2

+ y2

dy

= d( V/ x2 + y2 )

= dx.

and the right side is the total differential of x

solution of the differential equation.

44. To see that the statement is true, write the separable equaton as -g(x)

M = -g(x)

and N

= l/h(y),

+ c.

dx

+ dy/h(y).

Identifying

we see that l\l/y = O = Nx, so the differential equation is exacto

52

Exercises 2.5

Exercises 2.5

l. Letting y = ux we have

+ x(udx + xdu) = O

(x - ux)dx

+ xdu

dx

dx
x

- + du
In Ixl

=O

=O

+u =e

x In Ixl

+y

= ex.

2. Letting y = ux we have

(x

+ ux)

dx

+ x(udx + xdu) = O
+ 2u) dx + x du = O

(1

dx
du
-+--=0
x
1 + 2u
1

In [z] + 2 In 11+ 2ul = e

(1 + 2~) =

x2

x2
3. Letting x

+ 2xy

el

= el.

= vy we have
vy(v dy

+ y dv) + (y

- 2vy) dy

+ (v2

+ 1)

vy dv

- 2v

+ dy = O

vdv

(vIn Iv In

1.::11y

1)2

11- --v -

dy

+ In Iyl = e

x/y - 1

(x - y) In Ix -

53

+ In y
yl -

=e

y = e(x - y).

Exercses

4.. Letting x

2.5

vy we have
y(vdy

+ydv)

- 2(vy
y dv - (v

+ y)

dy

+ 2)

dy = O

~_dy=O
v +2

=O

In Iv + 21- In Iyl = C
In

I~

+ 2-ln Iyl = c
x

5. Letting

+ 2y

= cly2.

= ux we have

(U2X2 + ux2)

+ xdu) = O

dx - x2(udx

u2 dx - xdu = O

1
=c
u
x
ln ]r] + - = c
y
In Ixl

y In

6. Letting

y = ux

+-

Ixl + x

= cy.

we have

(U2X2 + ux2)

dx

+ x2(u

(u2

dx

+ 2u)

dx

dx

-+
x
1
In [z] + 2'ln

lul -

+ x du) = O
+ x du = O
du

=0

u(u+2)

1
2'ln

lu + 21 = c
x2u
--=Cl
u+2

2Y = ci

X ;;

54

(y

;;

+2

Exercises 2.5
7. Letting y

= ux we have

+ x) (u dx + x du)

(ux - x) dx - (ux

+ 1)

(u2

+ x( u + 1) du = O

dx

dx

In Ixl + ~ In

=O

u + 1 du = O
u2 + 1

(u2 + 1) + tan-l u = C

8. Letting y = ux we have
(x

+ 3ux)

dx - (3x

+ ux) (u dx + x du) = O

(u2

1) dx + x(u + 3) du = O

dx
x

u +3
(u-1)(u+1)

In Ixl + 21n lu -

11 -

du

=O

In lu + 11 = C
x(u - 1)2
u+1

Cl

(y - x)2 = Cl(Y

+ x).

9. Letting y = ux we have
-uxdx

+ (x + y'ux)(udx + xdu) = O
(x

+ xy'u)
(u-3/2

du

+ u3/2

dx = O

+ ~) du + d:

=O

_2u-l/2 + In lul + In Ixl = C


In Iy/xl

+ In Ixl =

2NY +

y(1n Iyl - c)2 = 4x.

55

Exercises 2.5
10. Letting

y = ux

we have
(ux

+ Vx2 + u2x2

) dx - x(udx

+ x du) = O

xVI +'u2dx

- x2du = O

dx

du

-x
In JxJ - In

VI + u2

lu + VI + u
u + VI + u

=0
=e

2 =

11. Letting

y = ux

we have
(X3 - u3x3) dx

+ u2x3(udx + xdu)

=O

+ u2xdu = O

dx

dx
2
- +u du

In JxJ

12. Letting

= 2 we find el

y = ux

=O

+ "3u3 = e

3x31n JxJ
Using y(l)

elx

+ y3

elx3.

= 8. The solution of the initial-value problem is 3x31n JxJ

+ y3 = 8x3.

we have
(X2

+ 2u2x2)

dx - ux2(u dx

(1 + u2)

+ x du) = O

dx - ux du =
dx

udu

- 1 +u2

=O

+ u2)

=e

In IxJ - ~ In (1

x2

1 + u2 = el
x4

Using y( -1) = 1 we find el

= 1/2. The

= el

(y2

+ x2)

solution of the initial-value problem is 2x4

56

= y2 + x2.

Exercses 2.5
13. Letting y = ux we have

dx - z e" du = O
dx
ud u=O
--e
x

In Ixl-

eU = e

In Ixl - eY/x = e.
Using y(l)

= O we find

14. Letting x

= vy we have

e = -l. The solution of the initial-value problem is In Ixl

y( v dy

+ y dv) + vy (1n vy

eY/x - l.

- In y - 1) dy = O

+ v In u dy

y dv

=O

~+dy=O
vlnv
y

In 11nIvll

+ In Iyl = e

I~I

= el.

yln

Using y(l)
15. From y

= e we find el = -e.

1
+ -y1 = -yx

x3w = x3

+ e or

and w

The solution ofthe initial-value problem is Ylnl~1

= y 3 we obtain.

y-l we obtain dw
dx
eXw = _~e2x + e or y-l = _~ex + ce:",

e-3x

+y

= xy

+w

= _ex.

and w = y- 3 we obtain -dw - 3w = -3x.

= xe-3x + --e-3x + e or y-3 = x + --

18. From y' -

3
3
3
+ -w
= -.
An integrating factor is x so that

dw
dx

y3 = 1 + cx-3.

16. From y' - Y = eXy2 and w

17. From y I

dx
+ ee3x.

(1 + 2:.)
y = y2 and w = y-l we obtain
x

xe" so that xexw = =xe"

+ eX + e or

1
1 2
19. From y - -y = --2Y
and w
I

= -e.

= y-

y-

= -1 +

dw
dx

An integrating factor is e- 3z so that

(1 + 2:.)
w = -l. An integrating
x

factor is

e
-1 + _e-x.
x

dw

we obtain dt

57

An integrating factor is eX so that

+ -w
t

-2'

An integrating factor is t so that

Exercises 2.5

tw

t
Y

= In t + e or y- = -In t + - _ Writing this in the form - = In x + e, we see that the solution

can also be expressed in the form


,2

20. From y + 3 (

1+ t

2) Y =

et/y

= elX.

2t
4
dw
2t
- 2t
.
2) Y and w = y-3 we obtain - - --2 w = --2
. An mtegrating
3 1+ t
dt
1+ t
1+ t
(

. --21 so that --2w = --21 + e or y-3 = 1 + e ( 1 + t2 ) .


factor IS
l+t
l+t
l+t
,2
3 4
3
. dw
6
9
.
6
21. From y - -y = -2Y and w = y- we obtam - + -w = --2 . An integratmg factor is x so that
x
x
dxx
x
X6W = _*x5 + e or y-3
22. From y' + y

= _*x-l + ex-6. If y(l)

1 then

e=

~ and y-3 = _*x-l + ~x-6.

= y-l/2 and w = y3/2 we obtain dw + ~w = ~. An integrating factor is e3x/2 so that

dx
2
2
e3x/2w = e3x/2 + e or y3/2 = 1 + ee-3x/2. If y(O) = 4 then e = 7 and y3/2 = 1 + 7e-3x/2.
du

= x + y + 1 so that du/dx = 1 + dy/dx. Then - - 1 = u or --2


du = dx. Thus
dx
1 +u
tan-l u = x + e or u = tan(x + e), and x + y + 1 = tan(x + e) or y = tan(x + e) - x - 1.

23. Let u

24. Let u

or u2

du
dx

1-u
or udu = dx. Thus !u2
u

= x + y so that duf d = 1 + dy/dx. Then - - 1 = -= 2x

= 2x +

+ el, and (x + y)2

25. Let u = x + y so that du/dx

el.
du
2
2
Then dx - 1 = tan u or cos udu = dx. Thus

= 1 + dy/dx_

~u+i sin 2u = x+e or 2u+sin2u = 4x+el, and 2(x+y)+sin


2x

=x+e

2(x+y) = 4X+Cl or 2y+sin 2(x+y) =

+ Cl.

26. Let u = x + y so that du/dx

du
dx

. u )/(1 - sin
in uu) we h ave
by (1 - sin
tan u - sec u

=x

= 1 + dy/dx. Then - - 1 = sm u or

= dx. Multiplying

1 - sin
2 ud u = dx or (2sec u - tan u sec u ) du = dx. Th us
cos u

+ C or tan( x + y) - sec(x + y)

27. Let u = y - 2x + 3 so that duf d

1
.
du
1 + smu

=x

+ c.
du

= dy/dx - 2. Then dx + 2 = 2 +

Fu or

1
yU du = dx. Thus

2yU = x + C and 2y1y - 2x + 3 = x + e_


28. Let u = y - x + 5 so that du/dx

_e-u
29. Let u

du

= dy/dx - 1. Then dx + 1

1 + en or e-udu

= dx. Thus

= x + C and _ey-x+5 = x + e.

=x+

y so that duf d

du

= 1 + dy/dx. Then dx - 1 = cos u and

1
1 + cos u

1 - cos u

1 - cos U

= 1 _ cos2 U = sin2 u

58

= ese

1 + cosu

2
u - ese u cot u

du = dx. Now

Exercises
so we have J(csc2 u-cscucotu)du
x + c. Setting x

= O and

= J dx

and - cot u--csc u

= 7r / 4 we obtain

= v2 -

= x+c.

= 3x+2y

so that duf dx

du
Then -d
x

= 3+2dy/dx.

h- 1.

= 3+

Now
u+ 2
5u + 6

1
5

--=-+--so we have

J (~+
5

and

tu + .g ln 125u+ 301 = x + c. Thus

25u

2u
-u +2

5u + 6
u +2
and --6 du
u +2
5u+

= --

+ 30

) du = dx

1
4
5(3x + 2y) + 25 ln 175x + 50y + 301 = x +
Setting x = -1 and y

= -1 we obtain e = t ln 95.

= dx.

25u

+ 30

1. The solution is

csc(x + y) - cot(x + y) = x +
30.Let u

Thus - cot(x+y)+csc(x+y)

2.5

c.

The solution is

144

5(3x + 2y) + 25 ln 175x + 50y + 301 = x + 5ln 95


or

5y - 5x + 2ln 175x + 50y + 301 = 10ln95.


31.

We write the differential equation M(x, y)dx + N(x, y)dy

= O as dy/dx = f(x,

y) where

M(x,y)

f(x, y)

= - N(x, y) .

The function f(x, y) must necessarily be homogeneous of degree O when M and N are homogeneous
of degree a. Since Mis homogeneous of degree a, M(tx, ty) = tO: M(x, y), and letting t

= l/x we

have
1
M(l, y/x) = - M(x, y)
xO:

or

M(x, y)

= xO:M(l,

y/x).

Thus
dy = f(x y) = _ xO:M(l,y/x)
dx
'
xO:N(l, y/x)

= _ M(l,y/x)

N(l,yjx)

= F (J!._).
x

To show that the differential equation also has the form

~~ = G (~)
we use the fact that M(x, y) = yO:M(x/y,
the substitutions u

1). The forms F(y/x)

= y/x or y = ux and v = x/y or x = vy.


59

and G(x/y)

suggest, respectively,

Exercises

32. As x

2.5

----+. -00,

Then, as x

e6x

----+ 00,

----+

O and y

e-6x

33. (a) The substitutions

----+

----+

2x

O and y

= Yl

+u

+ 3.

----+

Now write (1 + eX)/(l

- ee6x) as (e-6x

+ e)/(e-6x

e).

2x - 3.

and
dy
dx

dYl
dx

-=-+-

du
dx

lead to
dYl
du
-;;
+ dx

=P

+ Q(Yl + u) + R(Yl + u)

=P

+ QYl + RYf + Qu + 2YIRu + Ru2

or
du
dx -(Q+2YlR)u=Ru.

This is a Bernoulli equation with n = 2 which can be reduced to the linear equation
dw
dx

+ (Q + 2YlR)w =

-R

by the substitution w = u-l.


(b) Identify P(x) = -4/x2,

Q(x) = -l/x,

and R(x) = 1. Then dw


dx

+ (_2.x +~)
w = -1.
x

+ e or u = [-~x + ex-3

integrating factor is x3 so that x3w = _~x4

1.

An

Thus, y = ~ +u.

34. Write the differential equation in the form x(Y' / y) = In x + In y and let u = In y. Then du] dx = y' / y

and the differential equation becomes x(du/dx)

= ln z

+u

or duf d - u/x = (lnx)/x,

which is

first-order, linear. An integrating factor is e- J dx] = l/x, so that (using integration by parts)

!!._ [2.u]

= ln

dx x

and

x2

~ =_

2. _ ln x

The solution is
In y = -1 - ln x

+ ex

+ e.

e-cx-1
or

= --

Exercises 2.6
1. We identify (x, y) = 2x - 3y

+ 1. Then,

_
for h = 0.1,

Yn+l = Yn + 0.1(2xn - 3Yn + 1) = 0.2xn

+ 0.7Yn + 0.1,

and
y(l.1) ~ Yl = 0.2(1)

+ 0.7(5) + 0.1 = 3.8

y(1.2) ~ Y2 = 0.2(1.1)

+ 0.7(3.8) + 0.1 = 2.98.

60

Exercises 2.6
For h = 0.05
Yn+l = Yn

+ 0.05(2xn

- 3Yn

+ 1) = O.lxn + 0.85Yn + 0.1,

and
y(1.05) ;:::::
Yl = 0.1(1)
y(1.1) ;:::::
Y2

+ 0.85(5) + 0.1 = 4.4

= 0.1(1.05) + 0.85(4.4) + 0.1 = 3.895

y(1.15) ;:::::
Y3 = 0.1(1.1)
y(1.2) ;:::::
Y4
2. We identify f(x, y) = x

+ y2.

+ 0.85(3.895) + 0.1 = 3.47075

= 0.1(1.15) + 0.85(3.47075) + 0.1 = 3.11514

Then, for h = 0.1,

and
y(O.l) ;:::::
Yl

= 0.1(0) + O + 0.1(0)2 = O

y(0.2) ;:::::
Y2

= 0.1(0.1) + O + 0.1(0)2 = 0.01.

For h = 0.05
Yn+l

= Yn + 0.05(xn + y~) = 0.05xn + Yn + 0.05y~,

and
y(0.05) ;:::::
Yl

= 0.D5(0) + O+ 0.05(0)2 = O

y(O.l) ;:::::
Y2 = 0.05(0.05)
y(0.15) ;:::::
Y3
y(0.2) ;:::::
Y4

+ O+ 0.05(0)2 = 0.0025

= 0.05(0.1) + 0.0025 + 0.05(0.0025)2 = 0.0075


= 0.05(0.15) + 0.0075 + 0.05(0.0075)2 = 0.0150.

3. Separating variables and integrating, we have


dy
= dx
y

and

In Iy I

=x

+ c.

Thus y = Clex and, using y(O) = 1, we find c = 1, so y = e" is the solution of the initial-value
problem.

61-

Exercises 2.6
h-O
- 1

h-O
- 05

xn

Yn

0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1. 00

1.0000
1.1000
1.2100
1.3310
1.4641
1.6105
1.7716
1.9487
2.1436
2.3579
2.5937

True
Value
1.0000
1.1052
1.2214
1.3499
1.4918
1.6487
1.8221
2.0138
2.2255
2.4596
2.7183

Abs.
Error

% Rel.

0.0000
0.0052
0.0114
0.0189
0.0277
0.0382
0.0506
0.0650
0.0820
0.1017
0.1245

0.00
0.47
0.93
1.40
1.86
2.32
2.77
3.23
3.68
4.13
4.58

Error

xn

Yn

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00

1.0000
1.0500
1.1025
1.1576
1.2155
1.2763
1.3401
1.4071
1.4775
1.5513
1.6289
1.7103
1.7959
1.8856
1.9799
2.0789
2.1829
2.2920
2.4066
2.5270
2.6533

True
Value
1.0000
1.0513
1.1052
1.1618
1.2214
1.2840
1.3499
1.4191
1.4918
1.5683
1.6487
1.7333
1.8221
1.9155
2.0138
2.1170
2.2255
2.3396
2.4596
2.5857
2.7183

Abs.
Error

% Re1.

0.0000
0.0013
0.0027
0.0042
0.0059
0.0077
0.0098
0.0120
0.0144
0.0170
0.0198
0.0229
0.0263
0.0299
0.0338
0.0381
0.0427
0.0476
0.0530
0.0588
0.0650

0.00
0.12
0.24
0.36
0.48
0.60
0.72
0.84
0.96
1.08
1. 20
1.32
1.44
1.56
1.68
1.80
1.92
2.04
2.15
2.27
2.39

Error

4. Separating variables and integrating, we have

dy
y

Thus y = qeX

and, using y(l)

= 2xdx

and

In Iyl = x2

= 1, we find c = e-1,

+ c.
2

so y = e" -lis the solution of the initial-value

problem.
h-O
- 1

h=O 05

xn

Yn

1.00
1.10
1.20
1.30
1.40
1.50

1.0000
1.2000
1.4640
1.8154
2.2874
2.9278

True
Value
1.0000
1.2337
1.5527
1.9937
2.6117
3.4903

Abs.
Error

% Rel.

Error

xn

Yn

0.0000
0.0337
0.0887
0.1784
0.3243
0.5625

0.00
2.73
5.71
8.95
12.42
16.12

1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50

1.0000
1.1000
1.2155
1.3492
1.5044
1.6849
1.8955
2.1419
2.4311
2.7714
3.1733

62

True
Va1ue
1.0000
1.1079
1.2337
1.3806
1.5527
1.7551
1.9937
2.2762
2.6117
3.0117
3.4903

Abs.
Error

% Rel.

0.0000
0.0079
0.0182
0.0314
0.0483
0.0702
0.0982
0.1343
0.1806
0.2403
0.3171

0.00
0.72
1.47
2.27
3.11
4.00
4.93
5.90
6.92
7.98
9.08

Error

Exercises 2.6
h-O
- 05

h=O.l

5.

h=O.l

Xn

Yn

xn

Yn

0.00
0.10
0.20
0.30
0.40
0.50

0.0000
0.1000
0.1905
0.2731
0.3492
0.4198

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

0.0000
0.0500
0.0976
0.1429
0.1863
0.2278
0.2676
0.3058
0.3427
0.3782
0.4124

h=O.l

h=O 05

Xn

Yn

Xn

Yn

0.00
0.10
0.20
0.30
0.40
0.50

0.5000
0.5250
0.5431
0.5548
0.5613
0.5639

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

0.5000
0.5125
0.5232
0.5322
0.5395
0.5452
0.5496
0.5527
0.5547
0.5559
0.5565

7.

h=O.l

9.

6.

Xn

Yn

Xn

Yn

1.0000
1.0000
1.0191
1.0588
1.1231
1.2194

1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50

1.0000
1.0000
1.0049
1.0147
1.0298
1.0506
1.0775
1.1115
1.1538
1.2057
1.2696

10.

I 1

Eul e r

~
-10

1.0000
1.0500
1.1053
1.1668
1.2360
1.3144
1.4039
1.5070
1.6267
1.7670
1.9332

h=0.05

Xn

Yn

Xn

Yn

0.00
0.10
0.20
0.30
0.40
0.50

1.0000
1.1000
1.2159
1.3505
1.5072
1.6902

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

1.0000
1.0500
1.1039
1.1619
1.2245
1.2921
1.3651
1.4440
1.5293
1.6217
1.7219

h-O
- 05

Xn

Yn

Xn

Yn

0.00
0.10
0.20
0.30
0.40
0.50

0.5000
0.5250
0.5499
0.5747
0.5991
0.6231

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

0.5000
0.5125
0.5250
0.5375
0.5499
0.5623
0.5746
0.5868
0.5989
0.6109
0.6228

ARUnge-KUlta

f\
\
,

!.

h;O.1

A RungC.Kjlt.

Yn

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

1\

Xn

1.0000
1.1000
1.2220
1.3753
1.5735
1.8371

h=O.l

1.00
1.10
1.20
1.30
1.40
1.50

. 1

Yn

0.00
0.10
0.20
0.30
0.40
0.50

h=O.l

8.

h=O 05

11.

h=0.05

Xn

Eulcr

\.
10

63

h=O.05

~RUngC-Kutta

~uler
5

Exercises 2.6
12.
h=O.25
h=O.1

h=O.05
Run g e-Kut t a

Runge-Kutta

Euler
Euler
Eul e r

-2

-2

13. Using separation of variables we find that the solution of the differential equation is y = 1/(1- x2),
which is undefined at x = 1, where the graph has a vertical asymptote.
h=O 1

Euler

h=O 05

Euler

h-O
1

R-K

h=O 05

R-K

Xo

yo

Xo

yo

Xo

yo

Xo

Yo

0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1. 00

1.0000
1.0000
1. 0200
1. 0616
1.1292
1. 2313
1. 3829
1. 6123
1. 9763
2.6012
3.8191

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1. 00

1.0000
1.0000
1.0050
1. 0151
1. 0306
1.0518
1. 0795
1.1144
1.1579
1.2115
1.2776
1.3592
1.4608
1. 5888
1.7529
1. 9679
2.2584
2.6664
3.2708
4.2336
5.9363

0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1. 00

1.0000
1.0101
1.0417
1.0989
1.1905
1. 3333
1. 5625
1.9607
2.7771
5.2388
42.9931

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1. 00

1.0000
1.0025
1.0101
1.0230
1.0417
1.0667
1.0989
1.1396
1.1905
1.2539
1. 3333
1.4337
1. 5625
1.7316
1.9608
2.2857
2.7777
3.6034
5.2609
10.1973
84.0132

Because the actual solution of the differential equation becomes unbounded at x approaches 1, very
small changes in the inputs x will result in large changes in the corresponding outputs y. This can
be expected to have a serious effect on numerical procedures.

64

Chapter 2 Revew Exercses


14. (a)

0.6

.._______

--~.~~

-01

"J4
.o

(b) For y

+ 2xy

= 1 an integrating factor is e

Y = e-x 210X eX22ft


dx

+ ce-x = -

If y(O) = Othen e = Oand y = ft

(e) Using FindRoot in Mathematica

2
d
2
2
x x = e" so that - [eX y 1 = eX and
dx

J2 d

erf (x)

+ ce-x. 2

erf (x).
we see that the solution has a relative minimum at

(-0.924139, -0.541044) and a relative maximum at (0.924139,0.541044).

Chapter 2 Review Exercises


1. Writing the differential equation in the form

is a repeller for k

> O and an attractor for

y' = k(y + Aj k) we see that the critical point - Aj k

< O.

2. Separating variables and integrating we have


dy
y

i dx
x

In y = 41n x

We see that when x = O, y

= O, so the

+ e = In x4 + e

initial-value problem has an infinite number of solutions for

k = Oand no solutions for k =1 O.


3. dy = (y _ 1)2(y _ 3)2
dx
4. dy = y(y _ 2)2(y - 4)
dx

65

Chapter 2 Review Exercises


x

5.

x even

For dx/dt = xn,when

ti

is even, Ois serni-stable; when n is odd, Ois unstable. For daf d: = -xn,when

n is even, O is serni-stable; when n is odd, O is stable.


6. The zero of f occurs at approximately 1.3. Since P'(t)
for P

> 1.3, limt_oo P(t)


y

7. ,._,~"",
----"""
-.",

_.....................

'

= 1.3.

"",.,_"
"",,---.",

,,_

...

_... ..._

----"\,,, "",,--_.
,.-

....."",

","

......
- ...1

8. (a) linear in y, homogeneous, exact


(b) linear in x

(e) separable, exact, linear in x and y


(d) Bernoulli in x

(e) separable
(f) separable, linear in x, Bernoulli

(g) linear in x
(h) homogeneous

(i) Bernoulli
(j) homogeneous, exact, Bernoulli
(k) linear in x and y, exact, separable, homogeneous

(1) exact, linear in y


(m) homogeneous
(n) separable

66

= f(P) > Ofor P < 1.3 and P'(t) = f(P) > O

Chapter 2 Review Exercises


9. Separating variables we obtain

cos x dx =

Y
-2--

y +1

dy

1 + -1sin 2x = -In
1
-x
2
4
2

===}

(2y

+ 1 + c

===}

(2

2x + sin 2x = 21n y + 1 + c.

10. Write the differential equation in the form y In; dx = (x In ; - y) dy. This is a homogeneous
equation, so let x = uy. Then dx = u dy + y du and the differential equation becomes
ylnu(udy+ydu)=(uylnu-y)dy

or

ylnudu=-dy.

Separating variables we obtain


lnudu=_dy

===}

ulnlul-u=-lnlyl+c

===}

x(lnx -lny)

===}

:::'lnl:::.I-:::'=-ln1Y1+c
y
y
y

- x = -yln Iyl + cy.

. 1 equation
.
2 IS Bernou 11'
U'smg w
11. The diff
1 erentia
-dy + --y 2
---y 3x2
1.
dx
6x + 1
6x + 1
dw
6
9x2
-d + -6--w
= ---.
An integrating factor is 6x + 1, so
x
x+1
6x + 1
d
2
[(6x + l)w] 9x
dx
- -

===}

3x3
c
w +
- - 6x + 1 6x + 1

===}

(6x + 1)y3

y 3 we obtai
tam

== -3x3 + c.

(Note: The differential equation is also exact.)


12. Write the differential equation in the form (3y2 + 2x)dx + (4y2 + 6xy)dy = O. Letting M = 3y2 + 2x
and N = 4y2 + 6xy we see that My = 6y = N so the differential equation is exacto From

Ix

= 3y2 + 2x we obtain

so h(y) =

3.

= 3xy2 + x2 + h(y). Then

Iy

= 6xy + h'(y) = 4y2 + 6xy and h'(y) = 4y2

The general solution is

13. Write the equation in the form


1
3
-dQ + -Q
= t lnt.

dt

An integrating factor is e1n t

= t,

so
1

dt [tQ] = t In t

===}

tQ = - 25 t + 5t In t + c

===}

Q = - 25 t + 5t In t + t'

14. Letting u = 2x + Y + 1 we have


du _ 2 dy
dx - + dx'

67

Chapter 2 Review Exercises


and so the given differential equation is transformed into
du
dx

2u

+1
u

Separating variables and integrating we get


u
--du=dx
2u + 1
du = dx
(~_~_1_)
2 2 2u + 1
1
1
"2u- ln 12u + 11 = x

+c

2u - ln 12u + 11 = 2x

+ cj ,

Resubstituting for u gives the solution


4x

+ 2y + 2 -

ln 14x + 2y

+ 31 = 2x + CI

2x

+ 2y + 2 -

ln 14x + 2y

+ 31 = CI.

or

dy
15. Write the equation in the form -d
x

8x
+ zrr:
x +4

2x
x +4

= -2--.

An integrating factor is

x2

+ 4)4 ,so

= 2r2 cos e sin e + r cos e and N = 4r + sin e - 2r cos2 e we see that M; = 4r cos e sin e +
= Ne so the differential equation is exacto From fe = 2r2 cos e sin e + r cos e we obtain
f = _r2cos2e + r sin ' + h(r). Then fr = -2rcos2e + sine + h'(r) = 4r + sine - 2rcos2e and
h'(r) = 4r so h(r) = 2r2. The general solution is

16. Letting M
cos e

_r2 cos2 e + r sin () + 2r2 = c.


d
17. The differential equation has the form dx [(sinx)y] =
y

c/sinx.

7r < x

The initial condition implies c

< 27ris chosen to include x

o.

Integrating we have (sinx)y = c or

= -2sin(71r/6)

= 77r/6.

68

= l. Thus, y = 1/sinx,

where

Chapter 2 Review Exercises


18. Separating variables and integrating we have

d; = _ 2 (t + 1) dt
y

1
-- = -(t

+ 1) 2 + e

The initial condition implies e = -9, so the solution of the initial-value problem is
1
y

t2

+ 2t -

where

< t < 2.

- 4

19. (a) For y < O, y'Y is not a real number.


(b) Separating variables and integrating we have
dy

= dx

y'Y

Letting y(xo) = YO we get e = 2#0


2y'Y = x
Since y'Y> O for y

-=1=

+ 250

and

- xo, so that
- Xo

and

1
= 4(x
+ 2#0

- xo) .

= ~(x + 2#0 - xo) must be positive. Thus, the

O, we see that dy/dx

interval on which the solution is defined is (xo - 2#0,


20. (a) The differential equation is homogeneous and we let y
(x2 - y2) dx
(x2 - u2x2) dx

+ c.

2y'Y = x

+ xy

dy

(0).
= ux.
=

Then

+ ux2(udx + x du) = O
dx

+ uxdu = O
dx
udu =-x
1
'2u2 = ln jxj + e
=

y2
2" = - 2ln j x j

+ q.

The initial condition gives q = 2, so an implicit solution is y2 = x2(2 - 2ln jxj).

69

Chapter 2 Review Exercises

(b) Solving for y in part (a) and being sure that the initial condition is still satisfied, we have y = -V2lxl(l

y
2

- In Ixl)1/2, where

-e:::; x:::; e so that 1-1n [z] ;:::O. The graph ofthis function indicates that the derivative is not defined at x = Oand x = e. Thus,

the solution of the initial-value problem is y = -V2 x(1-1n x)1/2,


for O < x

-2

< e.

21. The graph of Yl (x) is the portion of the closed black curve lying in the fourth quadrant. Its interval
of definition is approximately (0.7,4.3). The graph of Y2(X)

is the portion of the left-hand black

curve lying in the third quadrant. Its interval of definition is (-00, O).
22. The first step of Euler's method gives y(1.1) ~ 9 + 0.1(1
more time gives y(1.2) ~ 9.4 + 0.1(1

+ l.lV9.4) ~

70

+ 3) = 9.4.

9.8373.

Applying Euler's method one

Modeling with First-Order


Differential

Equations

Exercises 3.1
1. Let

P = P(t) be the population at time t, and Po the initiaI population. From dP/dt = kP we
Using P(5) = 2Po we find k = ~ In 2 and P = Poe(ln2)t/5. Setting P(t) = 3Po

obtain P = pot.
we have

3 = e(ln2)t/5

==:::}

In3 = (1n2)t
5

==:::}

t = 5In3 ::::::
7.9 years.
In 2

Setting P(t) = 4Po we have


4 = e(ln2)t/5
2. Setting P =10,000 and t

=3

==:::}

In4 = (ln2)t
5

==:::}

t = 10 years.

in ProbIem 1 we obtain

10,000 = Poe(ln2)3/5

==:::}

Po = 10,000e-O.61n2::::::
6597.5.

Then P(10) = Poe21n2 = 4Po ::::::


26,390.

P = P(t) be the population at time t. From dP/dt = kt and P(O) = Po = 500 we obtain
P = 500t. Usng P(10) = 575 we find k = 1~In 1.15. Then P(30) = 500e31n1.15::::::
760 years ..

3. Let

N = N(t) be the number ofbacteria at time t and No the initiaI number. From dN/dt = kN we
obtain N = Not. Using N(3) = 400 and N(10) = 2000 we find 400 = Noe3k or = (400/No)1/3.
From N(10) = 2000 we then have

4. Let

2000 = y, e10k = y, (400)10/3


O
O No

==:::}

2000 = y,-7/3
40010/3
o

==:::}

y, =
O

= l(t) be the intensity, t the thickness, and 1(0) = 10. If dl/dt


1 = 10t, k = k lnO.25, and 1(15) = 0.0009810.

5. Let 1

2000 )-3/7 ::::::


201.
40010/3

kl and 1(3) = 0.2510 then

d.Sl d: = t S we obtain S = Soert where S(O) = So.


= 5.75% then S(5) = $6665.45.
(b) If S(t) =$10,000 then t = 12years.
(e) S:::::: $6651.82

6. From

(a) If So = $5000 and r

N(t) be the amount of lead at time t. From d.Nf dt = kN and N(O) = 1 we obtan
ekt. Using N(3.3) = 1/2 we find k =
In 1/2. When 90% of the lead has decayed, 0.1 grams

7. Let N

71

Exercises

3. 1

will remain. Setting N(t)

= 0.1 we have

et(1/3.3)ln(1/2) = 0.1

==}

_t_In ~

3.3

= In 0.1

==}

= 3.3 In/o.1 ;:::j10.96 hours.

In 1 2

8. Let N = N(t) be the amount at time t. From dN/dt = kt and N(O) = 100 we obtain N = 100t.
Using N(6) = 97 we find k = tlnO.97. Then N(24) = 100e(1/6)(ln097)24= 100(0.97)4;:::j88.5 mg.
9. Setting

N(t)

= 50

in Problem 8 we obtain
1

50 = 100e t

==}

kt = In 2'

10. (a) The solution of dA/dt = kA is A(t)


half-Iife T

In 1/2

t = (1/6) In 0.97 ;:::j136.5 hours.

==}

= Aoekt. Letting A = ~Ao and solving for t we obtain the

= -(ln2)/k.

(b ) Since k = -(ln2)/T

we have

A(t) = Aoe-(ln2)t/T = AoTt/T.


(e) Writing gAo

= Ao2-t/T as 2-3 = 2-t/T and solving for t we get t

3T. Thus, an initial

amount Ao wil! decay to kAo in three half lives.


11. Assume that

A = Aot and k = -0.00012378. If A(t) = 0.145Ao then t ;:::j15,600years,

12. From Example 3, the amount of carbon present at time t is A(t) = Aoe-O.00012378t.Letting t = 660
and solving for Ao we have A(660) = Aoe-O.0001237(660)
= 0.921553Ao. Thus, approximately 92%
of the original amount of C-14 remained in the cloth as of 1988.
13. Assume that dT/dt

= k(T-10)

so that T

10 + cekt. IfT(O)

and k = 2In(2/3) so that T(l) = 36.670 IfT(t)


14. Assume that dT/dt

= k(T-5)

and c = 59.4611 so that T(O)

so that T

= 150 then

= 70

and T(1/2) = 50 then c


0

= 60

t = 3.06 minutes.

= 5+cekt. IfT(l) = 550 and T(5) = 300 then k = -:!-ln2

= 64.4611.

k(T - 100) so that T = 100 + ce": If T(O) = 200 and T(l) = 220 then
c = -80 and k = In(39/40) so that T(t) = 90 implies t = 82.1 seconds. If T(t) = 98 then
t = 145.7seconds.

15. Assume that dT/dt

16. Using separation of variables to solve dT/dt = k(T -Tm) we get T(t) = Tm +cekt. Using T(O) = 70
we find c = 70 - Tm, so T(t) = Tm + (70 - Tm)ekt. Using the given observations, we obtain

T(~) = Tm + (70 - Tm)ek/2 = 110


T(l) = Tm + (70 - Tm)

72

145.

Exercises
Then

/2

3. 1

= (110 - Tm)/(70 - Tm) and

ek

= (ek/2)2 = (110-Tm)2

= 145-Tm

70 - Tm

70 - Tm

(110 - Tm)2
70 - Tm
12100 - 220Tm

= 145 _ T

+ T'/'n=

10150 - 250Tm

+ T'/'n

Tm = 390.
The temperature in the oven is 390

0.

17. From dA/dt

4 - A/50 we obtain A = 200 + ce-t/50

If A(O) = 30 then c = -170 and A =

200 - 170e-t/50.
18. Frorn dA/dt
19. Frorn dA/dt

= 0- A/50 we obtain A

= ce-t/5D If A(O)

= 30 then c = 30 and A = 30e-t/50.

= 10 - A/100 we obtain A = 1000 + cct/100.

= 1000 - 1000e-t/100.

If A(O) = O then c = -1000 and

At t = 5, A(5) ;:::;
48.77 pounds.

dA
lOA
2A
= 10 = 10 we obtain A = 1000 - 10t
dt
500-(10-5)t
100-t
A(O) = O then c = -110' The tank is empty in 100 minutes.

20. Frorn

dA
21. From -d =3-

00

4A
2A
(
) =3---weobtainA=50+t+c(50+t)-2.
6- 4 t
50 + t

t)2. If

IfA(0)=10then

e = -100,000 and A(30)

+ c(100 -

= 64.38 pounds.

22. (a) Initially the tank contains 300 gallons of solution.

Since brine is pumped in at arate

of

3 gal/rnin and the solution is pumped out at arate of 2 gal/mn, the net change is an increase
of 1 gal/min. Thus, in 100 minutes the tank will contain its capacity of 400 gallons.
(b) The differential equation describing the amount of salt in the tank is A'(t)

=6

- 2A/ (300 + t)

with solution
A(t) = 600 + 2t - (4.95 x 107)(300

+ t)-2,

O < t :::;100,

as noted in the discussion following Example 5 in the texto Thus, the amount of salt in the
tank when it overflows is
A(100) = 800 - (4.95

107)(400)-2

= 490.625

lbs.

(c) When the tank is overflowing the amount of salt in the tank is governed by the differential

73

Exercses

3. 1

equation
dA

di

= (3 gal/min)(2 lb/gal) - (400 lb/gal)(3 gal/min)

A(100) = 490.625.
Solving the equation we obtain A(t)

= 800

+ ee-3tI400.

The initial condition yields

e = -654.947, so that
A(t) = 800 - 654.947e-3tI400
When t = 150, A(150) = 587.37 lbs.
(d) As t

-4

00, the

amount of salt is 800 lbs, which is to be expected since

(400 gal)(2lbs/gal)=

(e)

800 lbs.

8oor---------~~~---600

~----2-00~--4-0~0----6-0-0--t

23. Assume Ldi/dt

+ Ri =

E(t), L

then e = -3/5 and limt~ooi(t)


24. Assume Ldi/dt

+ Ri = E(t),
,;=

'(0)' =
Smee
2

20

we obtai
tain e

= 0.1, R = 50, and E(t) = 50 so that i = ~ + ee-500t. If i(O) = O


= 3/5.

= Eosinwt, and i(O) = io so that

E(t)

EoR
. t
EoLw
t+
-RtIL
L2w2 + R2 smw - L2w2 + R2 cosw
ce
EoLw
= 20. + L2w2
+ R2

25. Assume Riiql dt. + (l/e)q = E(t), R = 200, e


If q(O) = Othen e = -1/100 and i = ~e-50t_
26. Assume Rdq/dt+
and i

(l/e)q

= _200ee-200t.

0.1472 amps. As t

-4

= 10-4, and E(t) = 100 so that q = 1/100 + ee-50t.

= 1000, e = 5 x 10-6, and E(t) = 200. Then q = 10100


+ee-200t
0.4 then e = - 560' q(0.005) = 0.003 coulombs, and i(0.005)

= E(t), R

If i(O) =
00

we have q

-4

10~0'

27. For O~ t ~ 20 the differential equation is 20d'i/dt


:t [etllOi]

= 6etl10 and i = 60 + ele-tilO.

+ 2i =

If i(O) = Othen

For t > 20 the differential equation is 20 di/ dt

120. An integrating factor is etilO, so


el

+ 2i = Oand
74

= -60 and i

i = e2e-tl10.

= 60 - 60e-tllO.

Exercises
At t

= 20

we want C2e-2

= 60 - 60e-2 so that
i(t)=

C2 = 60

3. 1

(e2 - 1). Thus

< 20;

O ::s t
t>20.

60 - 60e-t/la
{ 60(e2-1)e~t/1O,

28. Separating variables we obtain


E

dq
dt
a - q /C = k~l + k-2t

=}

Setting q(O) = qa we find

C2

qI
-C In I Ea - C

(Ea - qo/C)-C
1/k2
k1

' so

(Ea - q/C)-C
(kl + k2t)1/k2

+ k2t[ + Cl

-k In [kl
-2

q)-C
( Ea - C

(
=

=}

qa)-C
Ea - C

q
(
qa) ( kl )
Ea - - - Ea -C C
k + k2t
q = EaC
29. (a) From mdv/dt

= mg-

+ (qa -

kv we obtain v = gm/k+ce-kt/m.

EaC)

k)
( k + lk2t

kl

+ k2t

)-1/k

1/ C k2

1/Ck2

Ifv(O) = va then c = va-gm/k

and

the solution of the initial-value problem is


gm
v=T+
(b) As t

-t 00

va-T

gm)

-kt/m

the limiting velocity is gm/ k.

(e) Frorn ds/dt = v and s(O) = Owe obtain


gm - m ( va - -gm) e-ktf
s = -t
k
k
k

m.

+ c.

30. (a) Integrating d2s/dt2 = -g we get v(t) = ds/dt = -gt


the velocity is v(t) = -32t

+ -m ( va -

-gm) .
k

From v(O) = 300 we find c = 300, so

+ 300.

= O we get s(t) = -16t2 + 300t. The maximum height is


attained when v = O,that is, at ta = 9.375. The maximum height will be s(9.375) = 1406.25 ft.

(b) Integrating again and using s(O)

31. When air resistance is proportional to velocity, the model for the velocity is m dv / dt = -mg

kv (using the fact that the positive direction is upward.) Solving the differential equation using
separation of variables we obtain v(t) = -mg/k
v(t) = _ mg

+ ce:"!":

From v(O)

+ (300 + mg)e-kt/m.
k

75

= 300 we get

Exercises

3. 1

Integrating and using s(O) = Owe find


s(t) = _ mg t
k
Setting k

= 0.0025, m = 16/32

+ m (300 + mg)
k

(1 _ e-kt/m).

= 0.5, and 9 = 32 we have

s(t)

1,340,000 - 6,400t - 1,340,000e-O.005t

and
v(t) = -6,400

+ 6,700e-0005t.

The maximum height is attained when v = O, that is, at ta = 9.162. The maximum height will be
s(9.162) = 1363.79 ft, which is less than the maximum height in part (a).
32. Assuming that air resistance is proportional to velocity and the positive direction is downward, the
model for the velocity is m dv / dt = mg - kv. Using separation of variables to solve this differential
equation we obtain v(t) = mg/k

+ ce-kt/m.

From v(O) = O we get v(t)" = (mg/k)(l

- e-kt/m).

= 5, and 9 = 32 we have v(t) = 320(1 - e-o.lt). Integrating,


we find s(t) = 320t + 3200e-o.lt.
At t = 15, when the parachute opens, v(15) = 248.598 and
s(15) = 5514.02. At this point the value of k changes to k = 10 and the new initial velocity
Letting k

0.5, m

160/32

is Vo = 248.598. Her velocity with the parachute open (with time measured from the instant of

= 16 + 232.598e-2t.

opening) is vp(t)

Integrating, we find sp(t) = 16t - 116.29ge-2t.

seconds after leaving the plane is five seconds after the parachute opens.

Twenty

Her velocity at this

time is vp(5) = 16.0106 ft/sec and she has fallen s(15) + sp(5) = 5514.02 + 79.9947 = 5594.01 ft.
Her terminal velocity is limt->oovp(t) = 16, so she has very nearly reached her terminal velocity
five seconds after the parachute opens. When the parachute opens, the distance to the ground is
15,000 - 5514.02

= 9485.98 ft.

Solving sp(t)

= 9485.98 we get t = 592.874 s = 9.88 mino Thus,

it

will take her approximately 9.88 minutes to reach the ground after her parachute has opened and
a total of (592.874 + 15)/60 = 10.13 minutes after she exits the planeo
33. (a) The differential equation is frst-order, linear.
eJ3bdt/(bt+ro) = (ro + bt)3. Then
!((ro

+ bt)3v]

= g(ro

+ bt)3

and

Letting b = k/ p, the integrating factor is

(ro

+ bt)3v

= :b (ro

+ bt)4 + c.

The solution of the differential equation is v( t) = (g/ 4b) (ro + bt) + c(ro + bt) -3. Using v(O) = O
we find c = -gr~/4b,

so that

9
grt
gP(
k )
v(t) = 4b (ro + bt) - 4b(ro + bt)3 = 4k ro + -}
(b) Integratingdr/dt

= k/pweget

= kt/p+c.

Usingr(O)

76

gpr~
4k(ro + kt/ p)3 .

= ro wehavec = ro,

sor(t)

= kt/p+ro

Exercises

3. 1

= 0.007ft when t = lOs, then solving r(lO) = 0.007 for k/p, we obtain k/p = -0.0003 and

(e) Ifr

r(t) = 0.01 - 0.0003t. Solving r(t) = O we get t = 33.3, so the raindrop will have evaporated
completely at 33.3 seconds.
34. Separating variables we obtain

dP

p = k cos t di =>
If P(O)

= Po then

35. (a) Frorn

= k sin t + e ==> P =

= Po and P = po

el ek

sin

t.

sin t:

dP/dt = (kl - k2)P we obtain P = Poe(kl-k2)t where Po = P(O).

(b) If kl > k2 then P


as t

In IP I

-7

00

as t

-7

oo. If

k: = k2 then P = Po for every t. If kl < k2 then P

-7

oo.

-7

36. The first equation can be solved by separation of variables. We obtain x = ele-Alt.

we obtain el

= Xo

From x(O) = xo

and so x = xoe-Alt. The second equation then becomes


or

which is linear. An integrating factor is eA2t. Thus

Frorn y(O) = YO we obtain

e2

(YOA2 - YOAl - XOAl)/(A2

- A1). The solution is

k(M - A) - k2A = Ofor A we find the equilibrium solution A = kM/(k1 + k2)' From
the phase portrait we see that lim,.....oo A(t) = k1M /(kl + k2).

37. (a) Solving

------~-------+------~~-----A

Since k2

> O, the material will never be completely memorized and the larger k2 is, the less

the amount of material will be memorized over time.

77

Exercses 3. 1

(b) Write the differential equation in the form dA/dt+(kl


k1iVI. Then an integrating factor is e(k+k2)t, and

!!_
dt

[e(k+k2)t

Al

=:}

=:}

= k M e(k +k2)t
1

e(k+k2)tA=
A -- k k1Mk
1

UsingA(O)=Owefindc=A_,

+k2)A =

k1M

k1 + k2
2

e(k+k2)t+c

+ ce-(k+k2)t .

k1M andA=
k1 + k2

kliV~ (1_e-(k+k2)t).
k1 + 2

As t _,

00,

k1M .
k1 + k2

38. (a) Solving r - kx = O for x we find the equilibrium solution x = r/k.


and when x

When x

< r/k, dxf d: > O

> r/k, dx/dt < o. From the phase portrait we see that limt-+oox(t) = r/k.

------__------~------~~------x
r

(b) From dxf d: =


so that x_,

r -

kx and x(O) = Owe obtain x = r j k - (r/k)e-kt

r/k as t _, oo. If x(T) = r/2k then T = (ln2)/k.

39. It is necessary to know the air temperature from the time of death until the medical examiner
arrives. We will assume that the temperature of the air is a constant 65F. By Newton's law of
cooling we then have
dT
- 65)
T(O) = 82.
dt
'
Using linearity or separation of variables we obtain T = 65+ct.
From T(O) = 82 we find c = 17, so

- = k(T

that T = 65 + 17ekt. To find k we need more information so we assume that the body temperature
at t = 2 hours was 75F. Then 75 = 65 + 17e2k and k = -0.2653 and
T(t)

= 65 + 17e-O.2653t.

At the time of death, to, T(to) = 98.6F, so 98.6 = 65 + 17e-O.2653t, which gives t = -2.568. Thus,
the murder took place about 2.568 hours prior to the discovery of the body.

78

Exercses

40. We will assume that the temperature

3.2

of both the room and the cream is 72F, and that the

of the coffee when it is first put on the table is 175F. If we let TJ(t) represent the

temperature

temperature of the coffee in Mr. Jones' cup at time t, thcn


dTI

dt

k(TI

72),

which implies TI = 72 + CI t. At time t = O Mr. Jones adds cream to his coffee which immediately
reduces its temperature by an amount a, so that TI (O)

175 - a. Thus 175 - a

= 5, TI (5) = 72 + (103 - a)k


Now we let T2(t) represent the temperature of the coffee in Mrs. Jones' cupo From T2 = 72 + c2t
and T2(0) = 175 we obtain C2 = 103, so that T2(t) = 72 + 103t. At t = 5, T2(5) = 72 + 103e5k.

which implies

CI =

103 - a, so that TI (t) = 72 + (103 - a)t

= TI (O) = 72 + CI,

Whcn cream is added to Mrs. Jones' coffee the temperature

fact that k

At t

is reduced by an amount a. Using the

< O we have
T2(5) - a

72 + 103e5k

< 72 + 103k

ak

= 72 + (103 - a)e5k = TI (5).

Thus, the temperature

of the coffee in Mr. Jones' cup is hotter.

41. Drop an object from a great height and measure its terminal velocity,

that

Vt

= gm/ k,

so k

Vt.

In Problem 29(b) we saw

= gm/vi.

42. We saw in part (a) of Problem 30 that the ascent time is t = 9.375. To find when the cannonball

hit s the ground we solve s(t)


the time of descent is td

-16t2

+ 300t = O, getting

a total time in flight of t

18.75 - 9.375 = 9.375. The impact velocity is

Vi

= 18.75. Thus,

= v(18.75) = -300,

which has the same magnitude as the initial velocity.


We saw in part (b) of Problem 30 that the ascent time in the case of air resistance is
Solving s(t)

= Owe see that

1,340,000- 6,400t-1,340,OOOe-0.005t

Thus, the descent time is td

= 18.466 - 9.162 = 9.304. The impact velocity is

-290.91, compared to an initial velocity of Vo

Vi

When O < N

9.162.

= v(18.466) =

300.

Exercises 3.2
1. (a) Solving N(1 - 0.0005N)

t =

the total time of flight is 18.466.

= O for N we find the equilibrium solutions N = O and N = 2000.

< 2000, dN/dt > O. From the phase portrait we see that 1imi-<oo.N(t)
o

2000

A graph of the solution is shown in part (b).

79

= 2000.

Exercises 3.2

(b) Separating variables and integrating we have

2000

dN
N(l - 0.0005N)

= (~_

= dt

1
)dN
N - 2000

1500
1000

and

500

In N -ln(N

- 2000)

Solving for N we get N(t)

= t + c.

= 2000+t

/(1

10

+ +t) =

2000et /(1

solving for we find = 1/1999 and so N(t) = 2000et/(1999

dN

di =

N(a - bN) and N(O) = 500 we obtain N = 500b

=50,000 and N(l)

3. From

dP

di

= 1000 we have a

= P (10-1

10-7

p)

20t

+ et). Using N(O) = 1 and


+ et). Then N(lO) = 1833.59,

so 1834 cornpanies are expeeted to adopt the new teehnology when t


2. From

15

= 10.

500a
500b)e-at'

+ (a _

= 0.7033, b = 0.00014, and N

Sinee t~~ N =

50,000

= 1 + 9ge-O.7033t .

500
and P(O) = 5000 we obtain P = 0.0005 + 0.0995e-o.lt so that

P __. 1,000,000 as t __. oo. If P(t) = 500,000 then t = 52.9 months.


4. (a) We have dP/dt

= P(a - bP) with P(O) = 3.929 million.

Using separation of variables we

obtain
P
(t) = 3.929b

a/b
1 + (a/3.929b - l)e-at

3.929a
- 3.929b)e-at

+ (a

c
1 + (c/3.929 - l)e-at .
At t = 60(1850) the population is 23.192 million, so
23 92 =
.1

or c = 23.192 + 23.192(c/3.929

1+ (c/3.929 _ 1)e-60a

- 1)e-60a. At t = 120(1910)
c

91 972 = --,-..,------:---=;,-.
1 + (c/3.929 - 1)e-120a
or c = 91.972 + 91.972(c/3.929

- 1)(e-60a)2. Combining the two equations for c we get

23.192)/23.192)2 (_c __
( (c- c/3.929
- 1
3.929

1)

= c - 91.972
91.972

or
91.972(3.929)(c - 23.192)2

(23.192)2(c - 91.972)(c - 3.929).

The solution of this quadratic equation is c = 197.274. This in turn gives a = 0.0313. Therefore
P( )
t

197.274

= 1 + 49.21e-o.0313t
80

Exercises 3.2
dX
150
8. From -d = k(150 - X)2, X(O) = O, and X(5) = 10 we obtain X = 150 k
where
t
150 t + 1
k = .000095238. Then X(20) = 33.3 grams and X -; 150 as t -; 00 so that the amount of A -; O
and the amount of B -; Oas t -; oo. If X (t) = 75 then t = 70 minutes.
9. (a) The initial-value problem is dh/dt

h
10

= -8AhVh/Aw,

h(O) = H. Separating variables and integrating we have

!!:!!._

Vh

= _ 8Ah dt

and

Aw

the initial-value problem is

= (AwVH-

Aw

Using h(O) = H we find e


h(t)

2Vh = _ 8Ah t + c.

~--~~~=-~~-500

= 2VH, so the solution of

lhW = (AwVHfor O:S t

4Aht)2/A;

4Aht)/Aw,

where AwVH-

:s AwH/4Ah

(h) Identifying H = 10, Aw = 47l',and Ah = 7l'/576 we have h(t) = t2/331,776 Solving h(t)

= Owe

4Aht ;:: O. Thus,

(fs72 /144)t

+ 10.

see that the tank empties in 576/10 se~onds or 30.36 minutes.

10. To obtain the solution of this differential equation we use h(t) from part (a) of Problem 11 in
Exercises 11.3 with Ah replaced by cAh. Then h(t) = (AwVH-

4cAht)2/A;.

Solving h(t) = O

with c = 0.6 and the values from Problem 11 we see that the tank ernpties in 3035.79 seconds or
50.6 minutes.
11. (a) Separating variables and integrating we have
6h3/2dh

Using h(O)

20 we find c

= -5t

and

12h5/2
5

= -5t + c.

1920.;5, so the solution of the initial-value problem is h(t) =

(800.;5 - i~t)2/5. Solving h(t) = Owe see that the tank empties in 384.;5 seconds or
14.31 minutes.
(b) When the height of the water is h, the radius of the top of the water is r = h tan 30 = h] J3
and Aw

= 7l'h2/3. The differential equation is


dh
dt

Ah r;;:;:.

7l'(2/12)2

= -cAw V2gh = -0.6 7l'h2 /3 v64h = - 5h3/2 .

Separating variables and integrating we have


5h3/2dh
Using h(O)

= -2 dt and 2h5/2 = -2t + c.

= 9 we find c = 486, so the solution

Solving h(t)

= Owe

of the initial-value problem is h(t)

= (243-t)2/5.

see that the tank empties in 24.3 seconds or 4.05 minutes.

12. When the height of the water is h, the radius of the top of the water is g(20 - h) and Aw =
47l'(20- h)2/25. The differential equation is
dh
dt

=_

Ah r;;:;:.
eAw V2gh

=_

7l'(2/12)2
~4h =
0.647l'(20_ h)2 /25 v 64f1

82

_~

Vh

6 (20 _ h)2 .

Exercises 3.2
(b)
Year

Census
Population

Predicted
Population

1790
1800
1810
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950

3.929
5.308
7.240
9.638
12.866
17.069
23.192
31.433
38.558
50.156
62.948
75.996
91.972
105.711
122.775
131.669
150.697

3.929
5.334
7.222
9.746
13.090
17.475
23.143
30.341
39.272
50.044
62.600
76.666
91.739
107.143
122.140
136.068
148.445

Error

Error

0.000
-0.026
0.Q18
-0.108
-0.224
-0.406
0.049
1.092
-0.714
0.112
0.348
-0.670
0.233
-1.432
0.635
-4.399
2.252

0.00
-0.49
0.24
-1.12
-1.74
-2.38
0.21
3.47
-1.85
0.22
0.55
-0.88
0.25
-1.35
0.52
-3.34
1.49

The modeI predicts a population of 159.0 million for 1960 and 167.8 million for 1970. The
census populations for these years were 179.3 and 203.3, respectively. The percentage errors
are 12.8 and 21.2, respectiveIy.
5. (a) The differential equation is dP/dt = P(l-InP),
solution P

which has equilibrium

= e. When Po > e, dP/dt < O, and when Po < e, dP/dt >

o.

= P(l + In P), which has equilibriurn


soIution P = l/e. When Po > l/e, dP/dt > O, and when Po < l/e,
dP/dt < o.

(b) The differential equation is dP / dt

6. From
then

dP
di
C

= P(a-bInP)
a
= -b - In Ro .

7. Let X = X(t)

we obtain

be the amount of

-1
b
In la-bInPI

e at

= t+Cl so that

P = ea e-ce

time t and dX = k(120 - 2X)(150 - X).

dt

-w

. If P(O) = Po

If X(O) = O and

150 - 150e180kt
X(5) = 10 then X = 1 _ 2.5e180kt where k = .0001259, and X(20) = 29.3 grams. Now X ~ 60
as t -;

00,

so that the amount of A -; O and the amount of B -; 30 as t -; oc.

81

Exercises 3.2
14. We solve

dv

dt

'

m- = -mg-kv

v(O) = 300

using separation of variables. This gives

Integrating and using s(O)

= O we find

s(t) = ; In leos

fi

t - tan-l 300{Jg

I +~

In (1

+ 90~~Ok)

Solving v(t) = O we see that ta = 6.60159. The maximum height is s(ta) = 823.84 ft.
15. (a) Let p be the weight density of the water and V the volume of the object. Archimedes' principle
states that the upward buoyant force has magnitude equal to the weight of the water displaced.
Taking the positive direction to be down, the differential equation is

dv
2
m dt = mg - kv -

pv.

(b) Using separation of variables we have


mdv
= dt
(mg - pV) - kv2
.22!:._

Vk

Vk dv
(Vmg - pV)2 - (Vkv)2
1

Vk --===~
Vmg - pV

tanh

-1

Vkv
Vmg - pV

= dt

=t

+ c.

Thus
v ()t =

(e) Since tanh t

->

1 as t

-> 00,

Jmg - pV
k

tan

h (vkmg

the terminal velocity is

16. (a) Writing the equation in the form (x - Jx2

- kpV
m
t

J (mg -

+ y2 )dx + y dy

+ Cl )

pV) / k .

we identify iVI = x - Jx2

+ y2

and

N = y. Since M and N are both homogeneous of degree 1 we use the substitution y = ux. It

84

Exercises

3.2

Separating variables and integrating we have


(20-h)2

-'---=""':- dh

vh

5
dt
6

and

= --

r: 80
800y h - -h
3

3/2

2 5/2
+ -h
5

5
6

= --t

+ e.

Using h(O) = 20 we find e = 2560v5j3, so an implicit solution of the initial-value problem is

To find the time it takes the tank to empty we set h

= O and solve for

t. The tank empties in

1024v5 seconds or 38.16 minutes. Thus, the tank empties more slowly when the base of the cone
is on the bottom.
13. (a) Separating variables we obtain

mdv
= dt
mg - kv2
~
dv
= dt
9 1- (kvjmg)2
y'ri0
Jkjmgdv
-Vkg 1- (Vkvjy'ri0)2

0itanh-1

V kg

=dt

Vkv = t + e
y'ri0

[g"g
tanh -1 Vkv
;;;:n;:; =
- t + el
ymg

Thus the velocity at time t is

v(t) =
Setting t

= O and

'mg tanh ( Vfkg


VT
-:;;;t+ el )

v = Vo we find el = tanh-1(Vkvojy'ri0).

(b) Since tanh t _, 1 as t _,

00,

we have v _, Jmgjk as t _, oo.

(e) Integrating the expression for v(t) in part (a) we obtain

Setting t = O and s = So we find e2 = So -

ln( cosh el).

83

Exercises

3.2

follows that
(x - Vx2
x

+ u2x2)

[(1 - ,jI +

dx
2)

+ ux(udx + xdu)
+ U2]

dx

=O

+ x211.du=

udu

dx

1 + u2 - ..j 1 + u2
udu

x
dx

..JI + u2 (1 - ..JI + u2

Letting w = 1 - ..JI + u2 we have dw = -u du] ,11 + u2 so that


- In

(1 - VI + u

2 ) = In x

+C

1
--==:o:=CIX
1 - ..JI + u2

r:-:-;;

C2

1 - VI -t- u~ = --

SoIving for y2 we have


y2

= 2C2X+ c~ = 4 (~)

(x

+ ~)

which is a famiIy of paraboIas symmetric with respect to the x-axis with vertex at (-c2/2, O)
and focus at the origino
(b) Let u

x2 + y2 so that

Then
dy
1 du
y-=---x
dx
2 dx
and the differentiaI equation can be written in the form
1 du
- - - x
2 dx

= -x+ yU or ~ du = yU.
2 dx

85

Exercises

3.2

Separating variables and integrating we have


du

2y'u = x
.jU=x+c

17. (a) From 2W2

W3

W2(2 - W) = Owe see that W

= O and

= 2 are constant solutions.

(b) Separating variables and using a CAS to integrate we get


dW

V
W 4-2W

=dx

-tanh-1(-2V4-2W)=x+c.

and

Using the facts that the hyperbolic tangent is an odd function and 1 - tanh2 x

= sech2 x we

have
1

2"V4 -

2W

= tanh( -x - c) = - tanh(x + c)

4(4 - 2W) = tanh2(x + c)


1-

2" W = tanh (x + c)
1

2"W

= 1 - tanh2(x

+ c) = sech2(x + c).

Thus, W(x) = 2sech2(x + c).


(e) Letting x

= O and W = 2 we find that sech2(c) = 1 and c = O.

= 102 for r2 we see that r2 = 20h - h2. Combining the rate of input of
with the rate of output due to evaporation, k11"r2 = k11"(20h - h2), we have dV/dt =

18. (a) Solving r2 + (10 - h)2


water, 11",

11" - k11"(20h - h2).

Using V = lO11"h2

-111"h3,

we see also that dV/dt

= (20nh - 11"h2)dh/dt.

Thus,
and

86

dh
dt

1 - 20kh

+ kh2

20h - h2

Exercises

(b) Letting k = 1/100, separating variables and integrating (with


the help of a CAS), we get
100h(h - 20)
(h _ 10)2 dh = dt

and

100(h2 - lOh + 100)


-~-10---h----'= t + c.

3.2

h
10
8

Using h(O) = O we find c = 1000, and solving for h we get


h(t) = 0.005( Jt2

+ 4000t -

t), where the positive square root is

chosen because h 2: O.
2000

(e) The volume of the tank is V = ~7I(1O)3 feet, so at arate of

71

4000 6000 80001000b

cubic feet per minute, the tank

will fill in ~(10)3 ~ 666.67 minutes ~ 11.11 hours.


(d) At 666.67 minutes, the depth of the water is h(666.67) = 5.486 feet. From the graph in (b) we
suspect that limt_oo h(t) = 10, in which case the tank will never completely fill. To prove this
we compute the limit of h(t):
.
11mh(t) = 0.005 lim

t-oo

t-oo

(yt/ 2 + 4000t -

= 0.005 11m
t-oo

19. (a)

t
u
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170

P(t)
';.n~
5.308
7.240
9.638
12.866
17.069
23.192
31.433
38.558
50.156
62.948
75.996
91.972
105.711
122.775
131.669
150.697
179.300

t2 + 4000t - t2
t) = 0.005 lim -./!=.2;===:===::=::=-----:t-oo v t + 4000t + t

4000t
tJ1

+ 4000/t + t

4000

= 0.005 lirn --

Q (t)
u.O.;:,
0.036
0.033
0.033
0.033
0.036
0.036
0.023
0.030
0.026
0.021
0.021
0.015
0.016
0.007
0.014
0.019

87

t-oo 1

+1

= 0.005(2000)

= 10.

Exercses 3.2
(b) The regression line is Q = 0.0348391 - 0.000168222P.
Q

..

0.02
0.015
0.01
0.005

--'-~2 0:;---'-'4
0:;--::":6
0:;--~80;--:1'"70--:1
7072--:1-:-4::-'""0
70
p

(e) The solution of the logistic equation is given in equation (5) in the texto Identifying a
0.0348391 and b

= 0.000168222 we have
P(t) =

aPo
.
bPo + (a - bpo)cat

(d ) With Po = 3.929 the solution becomes

P( ) _
0.136883
t - 0.000660944 + 0.0341781e-o.034839lt.
(e)

175
150
125
100
75
50
25
75

100

125

150

(f) We identify t = 180 with 1970, t = 190 with 1980, and t = 200 with 1990. The model predicts

P(180) = 188.661, P(190) = 193.735, and P(200) = 197.485. The actual population figures
for these years are 203.303, 226.542, and 248.765 millions. As t -; 00, P(t) -; a/b = 207.102.
20. (a) Using a CAS to solve P(l - P)

+ 0.3e-P = Ofor P

we see that P

= 1.09216 is an equilibrium

solution.
(b) Since f(P)

> Ofor O < P < 1.09216, the solution P(t) of


dP/dt

= P(l

is increasing for Po

- P) + 0.3e-P,

P(O)

= Po,

< 1.09216. Since f(P) < Ofor P > 1.09216, the


> 1.09216. Thus P = 1.09216 is

solution P(t) is decreasing for Po


an attractor.

-1
-2

88

Exercises 3.2
p

(e) The curves for the second initial-value problem are thicker. The

= 1. Comparing

equilibrium solution for the logic model is P

1.5

1.09216 and 1, we see that the percentage increase is 9.216%.

1D t

21. To find td we solve


m

dv
2
dt = mg - kv )

using separation of variables. This gives

v(t) =

If

=O

v(O)

ffi

tanh

t.

Integrating and using s(O) = Ogives


m In ( cosh
s(t) = k

yfk9)
-;;:;t

To find the time of descent we solve s(t) = 823.84 and find td = 7.77882. The impact velocity is

V(td) = 182.998, which is positive because the positive direction is downward.


22. (a) Solving

Vt

= Jmg/k

for k we obtain k = mg/v'

dv = mg - -2
mg
dt
Vt

m-

The differential equation then becomes


1v
-dv = 9 ( 1 - 2"

or

dt

2)

Vt

Separating variables and integrating gives


Vt tanh

-1 v

= gt

+ ci.

Vt

The initial condition v(O)

= Oimplies

el

= O, so

gt
v(t) = Vt tanh - .
Vt

We find the distance by integrating:

s(t) =

Vt tanh

The initial condition s(O) = Oimplies

gt dt =

V[

Vt

e2

= O, so

In (cosh gt )

+ e2.

Vt

s(t) = V[ In (cosh gt ) .
9

Vt

In 25 seconds she has fallen ~O,OOO


- 14,800 = 5,200 feet. Solving
5200 = (v;/32) In (cosh 32(25))
Vt

89

Exercises 3.2
for

Vt

gives

271.711 ft/s. Then

Vt ~

v2
gt
s(t) = _J_ln(cosh-)
9

= 2307.08 ln(cosh 0.1l7772t).

Vt

(b) At t = 15, s(15) = 2,542.94 ft and v(15) = s'(15) = 256.287 ft/sec.


23. While the object is in the air its velocity is modeled by the linear differential equation mdv/dt

= 160, k

mg - kv. Using m

t, and 9 = 32, the differential equation

becomes dv/dt + (1/640)v

32. The integrating factor is eJ dt/640 = et/640 and the solution of the differential equation is
et/640v = J 32et/640dt

= 20,480et/640 + c. Using v(O) = O we see that e = -20,480 and v(t) =

20,480 - 20,480e-t/640.

Integrating we get s( t) = 20,480t + 13,107,200e-t/640

e = -13,107,200 and s(t) = -13,107,200+20,480t+13,107,200e-t/640.

+ c. Since s(O)

= O,

To find when the object hits

the liquid we solve s(t) = 500-75 = 425, obtaining ta = 5.16018. The velocity at the time of impact
with the liquid is

Va

m dv/dt = mg-kv2.

= v(ta) = 164.482. When the object is in the liquid its velocity is modeled by

Using m = 160, 9 = 32, and k = 0.1 this becomes dv/dt

= (51,200-v2)/1600.

Separating variables and integrating we have


dv
51,200 - v2
Solving v(O) =

Va

= __!!:!_

and

1600

J2 In 1 v - 160J2
640
v + 160J2

1=

_l_t + c.
1600

= 164.482 we obtain e = -0.00407537. Then, for v < 160J2


v - 160J2I=
v + 160J2

eV2t/5-1.8443 or

_ v - 160J2
v + 160J2

= 226.274,

= eV2t/5-1.8443

Solving for v we get


13964.6 - 2208.2geV2t/5
v( t) = 61.7153 + 9.75937eV2t/5 .
Integrating we find
s(t)
Solving s(O)

= O we

= 226.275t - 1600In(6.3237 + eV2t/5) + c.

see that e = 3185.78, so


s(t) = 3185.78 + 226.275t - 1600In(6.3237 + eV2t/5).

To find when the object hit s the bottom of the tank we solve s(t) = 75, obtaining tb = 0.466273.
The time from when the object is dropped from the helicopter to when it hits the bottom of the
tank is ta + tb = 5.62708 seconds.

90

Exercises 3.3

Exercises 3.3
1. The equation dxf dt.

dx]

-A1dt

= -A1X can be solved by separation of variables. Integrating both sides of

+ c from which

we obtain In [z] = -A1t

C1XO so that x =

xoe-Alt.

= Xo

we find

Substituting this result into the second differential equation we have


dy
-- +A2Y=
dt

which is linear. An integrating factor is

.5._
dt
Y

we get x = C1e-Alt. Using x(O)

= A1XO

eA2t

y]

[eA2t

e(A2-AI)te-A2t

A2 - Al
Using y(O) = O we find c2 = -A1XO/(A2
_
y-,

A1xOe

_A t
1

so that

= A 1x Oe(A2-AI)t + c2

+ C2e-A2t

A1XO
A2 - Al

e-Alt

+ c2e-A2t

- Al). Thus
A1XO (-Alt
e
/\2 - Al

e -A2t) .

Substituting this result into the third differential equation we have


dz = A1A2XO (e-Alt

dt

_ e-A2t)

A2 - Al

Integrating we find

2. We see from the graph that the half-life of A is ap-

proximately 4.7 days. To determine the half-life of B


we use t

x,

y,

20

= 50 as a base, since at this time the amount

of substance A is so small that it contributes very little to substance B. Now we see from the graph that
y(50) ~ 16.2 and y(191)

~ 8.1. Thus, the half-life of

B is approximately 141 days.

~~2~5--~5~0--~7~5--1~0~0--1-2~5--15~0~-

t = 5 days. The amounts of x and z are the same


at about t = 20 days. The amounts of y and z are the same at about t = 147 days. The time when
y and z are the same makes sense because most of A and half of B are gone, so half of e should

3. The amounts of x and y are the same at about

have been formed.

91

Exercses 3.3
4. Suppose that the series is described schematically by W ==> -AlX ==> -A2Y

==> -A3Z where

-Al, -A2, and -A3 are the decay constants for VV, X and Y, respectively, and Z is a stable elemento
Let w(t), x(t), y(t), and z(t) denote the amounts of substances W, X, Y, and Z, respectively. A
model for the radioactive series is
dw
= -AlW
dt
dx
- = AlW - A2X
dt
dy
dt = A2X - A3Y

dz
dt = A3Y
5. The system is
11121
Xl = 2 . 3 + 50 X2 - 50 Xl . 4 = - 25Xl + 50 X2 + 6
I
1
1
1
2
2
x2 = 50XI 4- 50X2 - 50X2' 3 = 25xl - 25x2.

6. Let Xl, X2, and X3 be the amounts of salt in tanks A, B, and


I
1
Xl = 100x2 2-

11
x2 = -Xl'
100

e, respectively,

so that

1
1
3
100Xl 6= 50X2 - 50Xl

1
6 + -X3
100

1
- -X2
100

1
. 2 - -X2
100

I
1
1
1
x3 = 100x2' 5 - 100x3 - 100x3 4=

. 5 = -Xl
50

1
+ -X3
100

- -X2
100

1
1
20X2 - 20X3.

7. (a) A model is
dXl = 3 . X2 _ 2 . Xl
dt
100 - t
100 + t '

Xl(O) = 100

dX2 = 2 . Xl _ 3 . X2
dt
100 + t
100 - t '

X2(0)

= 50.

(b) Since the system is closed, no salt enters or leaves the system and Xl(t) +X2(t) = 100+50 = 150
for all time. Thus Xl = 150 - X2 and the second equation in part (a) becomes
dX2
2(150 - X2)
=
dt
100 + t

3X2

100 - t

300
100 + t

=---

2X2
100 + t

or
dX2

(2
3)
+ 100 + t + 100 - t
which is linear in X2. An integrating factor is

di"

X2 =

300
100 + t '

e2ln(IOO+t)-3ln(IOO-t)
= (100 + t)2(100 - t)-3

92

3X2

100 - t

Exercises 3.3
so

Using integration by parts, we obtain


(100

+ t)2(100

- t)-3X2 = 300 [~(lOO + t)(100 - t)-2 - ~(100 - t)-l

+ e] .

Thus
X2

1
2
1
]
= (100300
+ t)2 [c(100 - t) 3 - "2(100t) + "2(100 + t)(100 - t)
300
3
1
(100 + t)2 [c(100 - t) + t(lOO - t) .

Using X2(0) = 50 we find e = 5j3000.

At t

= 30, X2 = (300j1302)(703c

+ 30

70) ~ 47.4 lbs.

8. A model is

dXI
di

dX2 =
di

1
(4 galjmin)(O lbjgal) - (4 galjmin) ( 200XI
lbjgal )

.
(4 gal/min)

dX3 = (4 galjmm)
.
di

(1200XI
(1150X2

)lbjgal ) lbjgal

. (1150X2 lbjgal )
. . ( 100X3
1 lbjgal )
(4 galjmm)
(4 galjmm)

or
dXI
1
= --Xl
dt
50
dX2
1
= -XIdt
50
dX3
2
- = -X2dt
75
-

2
-X2
75
1
-X3
25

Over a long period of time we would expect Xl, X2, and X3 to approach Obecause the entering pure
water should flush the salt out of all three tanks.
9. From the graph we see that the populations are first egual

at about t = 5.6. The approximate periods of

x,y

and y are

both 45.

5D

93

IDD

Exercises

3.3

10. (a) The population y(t) approaches 10,000, while the population

x(t) approaches extinction.

k,
l~k

1 Jo

10

(b) The population x(t) approaches 5,000, while the population

y(t) approaches extinction.

20

; 1

1
10

(e) The population y(t) approaches 10,000, while the population

x(t) approaches extinction.

(d) The population x(t) approaches 5,000, while the population


y(t) approaches extinction.

1:}::1

II
5

11. (a)

E:
t..

I"'t

20

:1

10

20

10

20

1'" t

1'" t

(b)

1:

1 1

20

(e)

20

1 I"'t
40

20

40

(d)

1 1 1'"

40

In each case the population x(t) approaches 6,000, while the population y(t) approaches 8,000.
12. By Kirchoff's first law we have il = i2 + i3. By Kirchoff's second law, on each loop we have E(t) =
Li~ + R1i2 and E(t) =

u; + R2i3 + ~q

so that q = CRli2 - CR2i3. Then i3

so that the system is


Li;
-Rli;

+ Li~ + Rli2 =
1

E(t)

+ R2i3 + Ci3 = O.
94

= q'

= CRli2 - CR2i3

Exercses
13. By Kirchoff's first law we have il = i:

+ i3.

3.3

Applying Kirchoff's second law to each loop we obtain

di2.

E(t) =

2lRl

E(t) =

2lRl

+ i; di + 22R2

and
.

di3.

+ L2di + 23R3.

Combining the three equations, we obtain the system

di2
..
Lldi + (Rl + R2)22 + Rl23 = E
L2 di3 + Rli2 + (Rl + R3)i3 = E.
dt
14. By Kirchoff's first law we have il = i2

E(t) =

u; + Ri2

and E(t)

+ i3.

By Kirchoff's second law, on each loop we have

= u; + Cq so that q = CRi2. Then

i3

= q'

CRi~ so that system is

Li' + Ri2 = E(t)

15. We first note that s(t)

+ i(t) + r(t)

n. Now the rate of change of the number of susceptible

persons, s(t), is proportional to the number of contacts between the number of people infected and
the number who are susceptible; that is, ds/dt

= -klSi.

We use -kl because s(t) is decreasing.

Next, the rate of change of the number of persons who have recovered is proportional to the number
infected; that is, dr / dt = k2i where k2 is positive since r is increasing. Finally, to obtain di/ dt we
use
d
d

-(s+i+r)=-n=O.
dt
dt

This gives

di
ds
.
.
-dt
= --dr
=
-k22 + klSZ.
dt dt
The system of equations is then

ds
.
= -klS2
dt

-di = - k''22

+ k lS2.
dt
dr
- = k2i.
dt
A reasonable set of initial conditions is i(O) = io, the number of infected people at time O, s(O) =
n - io, and r(O) = O.
16. (a) If we know s(t) and i(t) then we can determine r(t) from s

95

+ i + r = n.

Exercses 3.3
(b) .In this case the system is

ds
= -0.2si
dt

di
O.7z+ 02'
z=:
. sz.
s(O) = 10 - io since r(O) = O and i(t) + s(t) + r(t) = O for
all values of t. Now k2/kl = 0.7/0.2 = 3.5, so we consider initial conditions s(O) = 2, i(O) = 8;
s(O) = 3.4, i(O) = 6.6; s(O) = 7, i(O) = 3; and s(O) = 9, i(O) = 1.

We also note that when i(O)

io,

s.i

s.i

s.i

10

10

10

10

s.i

101

101

We see that an initial susceptible population greater than k2/ kl results in an epidemic in the
sense that the number of infected persons increases to a maximum before decreasing to
the other hand, when s(O)

<

k2/kl,

and there is no epidemic.


17. Since xo > YO > O we have x(t) > y(t) and Y - x < O. Thus dx/dt < O
and dy/dt > O. We conclude that x(t) is decreasing and y(t) is
increasing.

As t

--+ 00

o.

On

the number of infected persons decreases from the start

we expect that x(t)

--+

C and y(t)

where C is a constant common equilibrium concentration.

18. We write the system in the form

dx
= kl(y - x)
dt
dy
dt = k2(X - y),

96

--+

C,

x,y

x(O)

Exercises 3.3
where kl

= KjVA and k2 = KjVB. Letting z(t) = x(t) - y(t) we have


dx

dt-

dy
dt =kl(y-x)-k2(X-y)
dz
dt = kl(-Z)

- k2Z

This is a first-order linear differential equation with solution z(t)

CIe-(k+k2)t.

Now

and

Since y(t) = x(t) - z(t) we have

The initial conditions x(O) = Xo and y(O) = YO imply


and
The solution of the system is
x(t)

= (xo - yo)k1
k1

y(t)

e-(k+k2)t

+ k2

= (YO - XO)k2 e-(k+k2)t


kl

+ k2

+ XOk2 + yok1
k1 + k2
+ XOk2 + YOkl
kl + k2

As t ___.00, x(t) and y(t) approach the common limit

This makes intuitive sense because the limiting concentration is seen to be a weighted average of
the two initial concentrations.

97

3.3

Exercises

19. Since there are initially 25 pounds of salt in tank A and

xI,x2

none in tank B, and since furthermore only pure water is

30

being pumped into tank A, we would expect that Xl(t)

20

would steadily decrease over time. On the other hand,

10

since salt is being added to tank B from tank A, we would


expect X2(t) to increase over time. However, since pure

50

100

water is being added to the system at a constant rate


and a mixed solution is being pumped out of the system, it makes sense that the amount of salt in
both tanks would approach O over time.

Chapter 3 Review Exercises

1. From

dP
di

2. Let A

= A(t) be the volume of C02 at time t. From ddA = 1.2 - ~ and A(O) = 16ft3 we obtain

= 0.018P

and P(O)

= 4 billion we obtain P = 4eO.018t so that P(45) = 8.99 billion.

t
4
A = 4.8 + 11.2e-t/4. Since A(10) = 5.7ft3, the concentration is 0.017%. As t ......00 we have
A ......4.8 ft3 or 0.06%.

< 4, 6 :::;t < 10, and 12 :::;t < 16, no voltage is applied to the heart and E(t) = O. At
the other times the differential equation is dE/ dt = - E / RC. Separating variables, integrating,
and solving for E, we get E = ke-t/RC, subject to E(4) = E(10) = E(16) = 12. These initial
conditions yield, respectively, k = 12e4/RC, k = 12elO/RC, and k = 12e16/RC. Thus
O :::; t < 4, 6:::; t < 10, 12:::; t < 16
O,

3. (a) For O:::; t

E ( t ) --

12e(4-t)/RC
12e(lO-t)/ RC,'
12e(16-t)/ RC,

(b)

4:::;t<6
10 :::;t < 12
16:::; t < 18.

10

\\\
4

4. From V dC/dt = kA(Cs

10 12

16 18 '

C) and C(O) = Co we obtain C = Cs

98

+ (Co - Cs)e-kAt/V,

Chapter 3 Review Exercises


5. (a) The differential equation is
dT

-dt = k[T -

T2 - B(TI - T))

dT
(1 + B)T - (BTI

Separating variables we obtain


_2__BIn 1(1 + B)T - (BTI
1+

+ T2)1= kt + e

Since T(O) = TI we must have


T(t)

(b) Since k

< O, lim (1+B)t

e3

di

40 - 2i

___!!:!__.
10 -

+ T2 .
1+ B

140 - 2il

BTI + T2
l+B
.

variables we obtain

el

= 2/JIO.

= 4 or i

10 the equation for the eurrent becomes 0.2i

=k

we obtain dx = ~

dy = 2k sin e eos e de,


Othen c

dx

= 2k

= q(lO -

Solving for i we get i(t) = 4t -

tt

2,

OS t

t2:10

20,

= Owhen e =

BTI + T2)
l+B

= -In 110 - ti + e or J40 - 2i

'( ) _ { 4t It
-

If x

Then

< 10. For t 2:

[1 + (y')2]

T(t) = BTI + T2 + e (1+B)t.


l+B
3

and

t~~Ts=T2+BTI-B

Sinee i(O) = O we must have

7. From y

= k dt. Then

BTI + T2 + TI - T2 (1+B)t.
l+B
l+B

1tO) ~: + 0.2i = 4. Separating

1
--In
2
O:::;t

+ T2)J.

TI - T2
B and so
1+

t--+oo

(e) SineeTs=T2+B(TI-T),

(1 -

+ T2)

- (BTI

= Oand lim T(t) = BTI

i=ecx:

6. We first solve

= k[(l + B)T

dy. If y

t).

10

20

tt

2,

= 20. Thus

< 10 .

= k sin2 e then

G- ~

eos 2e) de,

and

x = ke -

.
'2k sin
2e + c.

= O.

8. (a) From y = - x - 1+ el e" we obtain y' = y + x so that the differential equation of the orthogonal
. dy = --- 1 or -dx + x = -y. A n integratmg
. f actor 1Se
. y , so
f am1'1y 1S-d
x
y+x
dy
~[eYxl
dy

= -yeY

==:;.

eYx

= -yeY + eY + c

99

==:;.

= -y + 1 + ce-Y.

Chapter 3 Review Exercises

(b) Differentiating the family of curves, we have


,
1
y =-(X+CI)2

1
=-y2'

The differential equation for the family of orthogonal trajectories


is then y'

= y'2.

Separating variables and integrating we get


dy = dx
y2
1

-- = x + e
y

y=---.
x+e
9. From ~~
dy
dt

= klx(a - x) we obtain C~a + a1~ax) dx

= k 2XY we obtam.
lnlyl=

k2 I l+Cle
klln

k ti +c

(k

or

y=C2 l+Cle

v"

10. In tank A the salt input is


(7

!~~)
( + ( !~~)
(:o~
2~~l)

~~l)

= (14 + 1~0x2) ~~n .

The salt output is

(3~) (2~)+ (5~) (2~) = _2_


min

100 gal

min

gal)
( 5 min

(Xl lb)
100 gal

100 gal

25

Xl ~ .
min

In tank B the salt input is


=

1
lb
20 Xl min .

The salt output is

(1~) (2~)+
min

100 gal

(2~) =

(4 gal)
min

100 gal

The system of differential equations is then


dx

dt
dX2

dt

1
2
14 + 100x2 - 25xI

-X-

1
20

1
20

-X2.

100

]_X2~.

20

min

Chapter 3 Related Exercises

Chapter 3 Related Exercises

1. (a) The differential equation is dP / dt = P (5 - P) - 4. Solving P (5 - P) - 4 = O for P we obtain

equilibrium solutions P = 1 and P = 4. The phase portrait is shown below and solution curves
are shown in part (b).

We see that for Po


O

>

4 and 1

< Po <

4 the population approaches 4 as t increases.

For

< Po < 1 the population decreases to O in finite time.

(b) The differentiaI equation is

~ = P(5 - P) -

= _(P2

5P

+ 4) =

-(P - 4)(P - 1).

Separating variables and integrating, we obtain

dP
= -dt
(P - 4)(P - 1)
( ~-~)dP=-dt
P-4
P-1
~ In 1P - 41
3
P-1

= =t + c

P- 4
-3t
P _ 1 = ci e
Setting t

= O and P

= Po we find CI = (Po - 4)/(Po - 1). Solving for P we obtain

P(t) = 4(Po - 1) - (Po - 4)e-3t


(Po - 1) - (Po - 4)e-3t

(e) To find when the population becomes extinct in the case O < Po < 1 we set P = O in
P - 4
--=--e
P -1

Po - 4
Po-1

-3t

from part (a) and solve for t. This gives the time of extinction
t

= _ ~ In 4(Po - 1) .
3

Po - 4

2. (a) Solving P(5 - P) - ~ = O for P we obtain the equilibrium solution P = ~. For P =F ~,


dP / dt

< O. Thus, if Po < ~, the population becomes extinct (otherwise there would be another

equilibrium solution.) Using separation of variables to solve the initial-value problem we get

101

Chapter 3 Related Exercises


P(t) = [4Po + (10Po - 25)t]/[4 + (4Po - 10)t]. To find when the population becomes extinct
for Po < ~ we solve P(t) = for t. We see that the time of extinction is t = 4Po/5(5 - 2Po).

(b) Solving P(5 - P) - 7 = for P we obtain complex roots, so there are no equilibrium solutions.

< for all values of P, the population becomes extinct for any initial condition.

Since dP / dt

Using separation of variables to solve the initial-value problem we get

O-5)

v'3 tan [tan-I (2P v'3


P(t) = "25 + 2
=

Solving P(t)

3. (a) Without

v'3]t
- 2

for t we see that the time of extinction is

harvesting, the population is governed by the

logistic equation dP/dt

P(r - r P] K).

With initial

population Po, the population was shown in Section 3.2


to be

P( )
rPo
t = rPo/K+(r-rPo/K)e-rt
Po

50

100

50

100

Poi K + (1 - Poi K)e-rt .


To find when P(t) = ~K we solve

Po
= ~K
Poi K + (1 - Poi K)e-rt
2

or

2Po
= 1.
Po + (K - Po)e-rt

This gives

= _~ ( Po ) = __1_
(
70,000
) ~ 1 4 rs
r In K _ Po
0.08 In 400,000 _ 70,000
9. y .

(b) Letting K = 400,000, Po = 70,000, and r = 0.08, and


solving

PI =
for h, we get h

K - JK2 - 4Kh/r
2

= Po

= ha = 4620.

102

Chapter 3 Related Exercises


(c) The MSY is obtained when Ea = r /2 = 0.04, and the

yield in this case is EPI = KE(1 - E/r) = 8,000. The


limiting population is K(1 - E/r)

= 200,000.

50

4. (a) Letting 1959 be year O and 1973 year 14, we obtain the

100

400

graph shown.

300
200
100

(b) Taking P(O)

= 1.91/(c 414 . 294) ~ 78.4613 and ex-

10

12

14 t

10

12

14 t

perimenting with the other pararneters, we find that the

400

graph of the logistic function fits the data points reason-

300

ably well for K = 500 and r

.. ..

= 0.11.

(e) The graph is shown with Po = 78.4613, K = 500,


r = 0.11, and h = irK ~ 6.875.

400
300
200

(d) The graph is shown with Po = 78.4613, K

= 500,

r = 0.11, and E = ~r = 0.055.

20

40

60

80

100 t

20

40

60

80

100 t

250
200
150
100
50

103

Hlqher-Order DifferentiaJ Equations


Exercises 4.1

_
Thcn y(O) = CI + C2 = o, y'(O) = CI - C2 = 1 so

1. From y = CleX + C2e-X we find y' = CleX - C2CX.

that CI = 1/2 and C2 = -1/2. The solution is y = ~ex - ~e-x.

2. From y = clx
so that cI

+ C2e-x we find y' = 4clx

= 3/5

+ ge-x.

we find y' = CI + c2(1 + ln z ). Then y(l)

that CI = 3 and C2 = -4.

= c +

Then y(O) = CI + C2 = 1, y'(O) = 4CI - C2 = 2

and C2 = 2/5. The solution is y = ~x

3. From y = CIX+C2xInx
4. From y

- c2e-x.

The solution is y

= 3x - 4x

= CI = 3, y'(l)

= CI +C2 = -1 so

In x.

C2cos x + C3sin x we find y' = -C2 sin x + C3cos x and yl/ = -C2 cos X - C3sin x. Then

Y(7!") = CI - C2 = O, Y'(7!") = -C3 = 2, yl/(7!") = C2 = -1 so that CI = -1, C2 = -1, and C3 = -2. The
solution is y = -1 - cosx - 2sinx.
5. From y = c + C2x2 we find y' = 2C2X. Then y(O) = CI = O, y'(O) = 2C2 . O = O and y'(O) = 1 is not
possible.

Since a2(x)

= x is O at x = O, Theorem

6. In this case we have y(O)


are y

= x2 and y

= CI = O,

4.1 is not violated.

y'(O) = 2C2 . O

= O so

CI

=O

and C2 is arbitrary, Two solutions

= 2x2.

7. From x(O) = Xo = CI we see that x(t) = xocoswt


Then x'(O) = Xl = C2W implies C2 = xjw.

+ C2Sinwt and x'(t)

= -xosinwt

Thus
.
+ -Xl smwt.

x ( t ) = Xo coswt

8. Solving the system

x(to) = CI coswto + C2sinwto

=x

+ C2Wcoswto

x'(to) = -CIW sinwto

= Xl

for Cl and C2 gives


CI =

wXo cos wto - Xl sin wto


w

and

C2 =

Xl cos wto + wXo sin wto


w

Thus

X ()t =

wXo cos wto - Xl sin wto


coswt
w

= xo(coswtcoswto

= Xo cosw(t

+ sinwtsinwto)

- to) + Xl sin w(t - to).


w

104

Xl cos wto + wxo sin wto

w
+ Xl (sinwtcoswto
w

sinwt

- coswtsinwto)

+ C2wcoswt.

Exercises 4. 1
9. Since a2(x) = x - 2 and Xo
10. Since ao(x)

= tan

x and Xo

= Othe problem has a unique solution for

-00

= Othe

=t: /2

problem has a unique solution for

11. We have y(O) = el +cs = O, y'(l) = ele+e2e-l

(e2 -1)

= 1 so that el = e/

< x < 2.
< x < 7r /2.
and e2 = -e/

(e2 -1).

The solution is y = e (ex - e-x) / (e2 - 1).


12. In this case we have y(O)

= el

= 1,

y'(l)

= 2e2

= 6 so that el

= 1 and e2 = 3. The solution is

y=1+3x2.
13. From y

= el eX cos x

+ e2ex sin x we find y'

= eleX( -

sin x + cos x) + e2eX(coSx + sinx).

(a) We have y(O) = el = 1, y'(O) = el + e2 = O so that el


y = e" cosx - eX sinx.

= el = 1, Y(7r)

(b) We have y(O)

= -ele7T

(e) We have y(O) = el = 1, Y(7r/2)


y = eX cosx + e-7T/2ex sinx.
(d) We have y(O)

= el

1 and e2

-l. The solution is

= -1, which is not possible.

= e2e7T/2= 1 so that el = 1 and e2

= O, Y(7r) = -ele7T

= e-7T/2. The solution is

= O so that el = O and e2 is arbitrary. Solutions are

y = e2ex sin x, for any real numbers e2.

= el + e2 + 3 = O, y(l) = el + e2 + 3 = 4, which is not possible.

14. (a) We have y(-l)

(b) We have y(O) = el . O+ e2 . 0+3 = 1, which is not possible.


(e) We have y(O)

= el . O+ e2 . 0+3

e2 = -3 - el. Solutions are


(d) We have y(l)
solution is y
15. Since (-4)x

= 3, y(l) = el + e2 + 3 = O so that el is arbitrary and

y = CIX2 -

(el + 3)x4 + 3.

= el + e2 + 3 = 3, y(2) = 4el + 16e2+ 3 = 15 so that el


= _x2 + x4 + 3.

= -1 and e2 = l. The

+ (3)x2 + (1)(4x - 3x2) = Othe functions are linearly dependent.

16. Since (1)0 + (O)x + (O)eX

= Othe functions are linearly dependent. A similar argument shows that

any set of functions containing f(x)

= Owill be linearly dependent.

17. Since (-1/5)5 + (1) cos2 x + (1) sin2 x = Othe functions are linearly dependent.
18. Since (1) cos 2x + (1)1 + (-2) cos2x = Othe functions are linearly dependent.
19. Since (-4)x

+ (3)(x - 1) + (l)(x + 3) = Othe functions are linearly dependent.

20. From the graphs of h(x)

=2+

x and h(x)

= 2 + [z]

we see that the funetions are linearly independent since


they cannot be multiples of each other.
3X

105

3X

Exercises 4.1
1+ x
21. The functions are linearly independent since W (1 + x,
22. Since (-1/2)eX

+ (1/2)e-X

+ (1) sinh x = Othe

x2)

X,

x2

1 2x

functions are linearly dependent.

23. The functions satisfy the differential equation and are linearly independent since

W (e-3x,e4X)
for

-00

i- O

= 7ex

< x < oo. The general solution is


y

= cle-3x + C2x.

24. The functions satisfy the differential equation and are linearly independent since

=2

W(cosh2x,sinh2x)
for

-00

< x < co. The general solution is

+ C2 sinh 2x.

y = Cl cosh 2x

25. The functions satisfy the differential equation and are linearly independent since
W (eX cos 2x, eX sin 2x)
for

-00

< x < oo. The general solution is y

= 2e2x i- O

= clex cos 2x

+ C2ex sin 2x.

26. The functions satisfy the differential equation and are linearly independent since

W (eX/2, xex/2)
for

-00

= eX

i- O

< x < oo. The general solution is


y = clex/2

+ C2xex/2.

27. The functions satisfy the differential equation and are linearly independent since
W (x3,x4)
for O < x

= x6

i- O

< oo. The general solution is

28. The functions satisfy the differential equation and are linearly independent since

W(cos(lnx),sin(lnx))
for O < x

= l/x

i- O

< oo. The general solution is


y

= Cl cos(lnx) + C2 sin(lnx).
106

= 2 i- O.

Exercises 4. 1
29. The functions satisfy the differential equation and are linearly independent since

= 9x-6 =J. O

W (x,x-2,x-2Inx)
for O < x

< oo. The general solution is


y = CIX

+ C2X-2 + C3x-21n

x.

30. The functions satisfy the differential equation and are linearly independent since

W (1, x, cos x, sin x)


for

-00

< x < oo. The general solution is


=

Cl

+ C2X + C3 cos x + C4 sin x.

= e2x and Y2 = x form a fundamental set of solutions of the homogeneous


equation, and Yp = 6ex is a particular solution of the nonhomogeneous equation.

31. The functions Yl

32. The functions Yl =

COS x

and Y2 = sin x form a fundamental set of solutions of the homogeneous

equation, and Yp = x sin x+ (cos x) ln(cos x) is a particular solution of the nonhomogeneous equation.

33. The functions Yl

equation, and Yp

e2x and Y2

x2e2x

34. The functions Yl = x-l/2


equation, and Yp
35.

+x

- 2 is a particular solution of the nonhomogeneous equation.

and Y2

= lsx2 - ~x

xe2x form a fundamental set of solutions of the homogeneous

x-1

form a fundamental set of solutions of the homogeneous

is a particular solution of the nonhomogeneous equation.

(a) We have y'PI = 6e2x and y"PI' = 12e2x so


" - 6y'PI
YPI

Also, Y~2 = 2x

+ 3 and

+ 5y PI =

12e2x - 36e2x

+ 15e2x =

_ge2x .

Y~2 = 2, so

Y~2 - 6Y~2 + 5YP2 = 2 - 6(2x

+ 3) + 5(x2 + 3x)

= 5x2

+ 3x

- 16.

(b) By the, superposition principle for nonhomogeneous equations a particular solution of y" - 6y' +
5y = 5x2

+ 3x

- 16 - ge2x is Yp

= x2 + 3x + 3e2x. A particular solution of the second equation

is

36.

(a) YPI = 5
(b) YP2 = -2x
(e) Yp = YPI + YP2 = 5 - 2x

(d) Yp = ~YPI - 2YP2 = ~ + 4x


37. (a) Since D2x = O, x and 1 are solutions of y" = O. Since they are linearly independent, the
general solution is y = cj z

+ C2.
107

Exercises
(b)

4. 1

Since D3x2

= o, x2,

= O. Since they

x, and 1 are solutions of ylII

general solution is y = C1x2

are linearly independent, the

+ C2X + C3.

(e) Since D4x3 = O, x3, x2, x, and 1 are solutions of y(4) = O. Since they are linearly independent,

= C1x3 + C2x2 + C3X + C4.


(d ) By part (a), the general solution of y" = O is Yc. = C1X + C2. Since D2x2 = 2! = 2, Yp = x2 is a
particular solution of y" = 2. Thus, the general solution is y = C1X + C2 + x2.
the general solution is y

(e) By part (b), the general solution of y'" = O is Yc. = C1x2

+ C2X + C3.

Since D3x3

= 3! = 6,

yp = x3 is a particular solution of ylll = 6. Thus, the general solution is y = C1x2 +C2X+C3 +x3.
(f) By part (e), the general solution ofy(4) = Ois Yc. = C1x3+C2X2+C3X+C4.

Since D4x4 = 4! = 24,

yp = x4 is a particular solution of y(4) = 24. Thus, the general solution is y


C3X + C4

= cx3 + C2x2 +

+ x4.

38. By the superposition principle, if Y1 = eX and Y2 = e-x are both solutions of a homogeneous linear
differential equation, then so are
1
2(Y1

+ Y2)

eX + e-x
2
= coshx

1
2(Y1 - Y2)

and

39. (a) From the graphs of Y1 = x3 and Y2 = !x!3 we see

eX - e-x
2
= sinhx.
y

that the functions are linearly independent since


they cannot be multiples of each other. It is easily
shown that Y1

= x3 solves x2y" - 4xy' + 6y

3x

= O.

To show that Y2 = !x!3 is a solution let Y2 = x3


for x ~ O and let Y2 = _x3 for x

< O.

(b) If x ~ Othen Y2 = x3 and W(Y1, Y2) =


If x

< O then

92 = -x3

and W(Y1,Y2)

13:

x31

3x2

= 1 x32
3x

= o.

_x312

-3x

= O.

= x2 is zero at x = O.
(e) The functions Y1 = x3 and Y2 = x2 are solutions of x2y" - 4xy' + 6y = O. They are linearly
This does not violate Theorem 4.3 since a2(x)
independent since W (x3, x2) = x4
(d) The function y

=1=

Ofor

-00

< x < oo.

= x3 satisfies y(O) = O and y'(O) = O.

(e) Neither is the general solution since we form a general solution on an interval for which
a2 (x)

=1=

O for every x in the interval.

40. Since ex-3 = e-3eX = (e-Se2)ex

= e-Sex+2,

we see that ex-3 is a constant multiple of ex+2 and the

functions are linearly dependent.

108

Exercises 4.2
41. Since OYI + OY2 +

... + OYk + 1Yk+1 =

O, the set of solutions is linearly dependent.

42. The solutions are linearly dependent. Suppose n of the solutions are linearly independent (if not,

+ 1 solutions is linearly dependent).


Then y = CIYI+ C2Y2 +
+ CnYn is the

then the set of n


Y2, ... , Yn

equation and for sorne choice, ci, c2,

Without loss of generality, let this set be YI,


general solution of the nth-order differential

, c~, of the coefficients Yn+l = ciYI + c2Y2 + ... + c~Yn. But

then the set YI. Y2, ... , Yn, Yn+l is linearly dependent.

Exercises 4.2

In Problems 1-8 we use reduction of order to find a second solution. In Problems 9-16 we use formula
(5) from the texto
1. Define y = u(x )e2x so

y'

= 2ue2x + u' e2x,

Therefore u"

= e2xu" + 4e2xu' + 4e2xu,

y"

and

y" - 4y'

= CIX+ C2. Taking CI = 1 and C2 =

O and u

+ 4y = 4e2xu" = O.

O we see that a second solution is

Y2 = xe2x.
2. Define y

= u(x)xe-X

so

y' = (1 - x)e-Xu

+ xe-xu',

y" = xe-xu"

+ 2(1 -

x)e-Xu'

- (2 - x)e-Xu,

and
y"
If w

= u'

we obtain the frst-order equation w'

Therefore w
3. Define y

+ 2y' + y = e-X(xu" + 2u') = O

or

+~ w = Owhich
x

u"

+ ~ u' = O..
x

has the integrating factor e2 J dx]

= u' = c/x2 and u = ci]. A second solution is Y2 = - xe-x = e-x.


x

= u(x) cos4x so
y'

= -4u

sin 4x

+ u' cos 4x,

y"

= u" cos 4x

- 8u' sin 4x - 16u cos 4x

and
y"

+ 16y =

(cos4x)u"

= O or u" - 8(tan4x)u' = O.

- 8(sin4x)u'

If w = u' we obtain the first-order equation w' - 8(tan 4x)w = O which has the integrating factor
e-SJtan4xdx = cos24x. Now
d
dx [(cos2 4x)w]

= O gives (cos2 4x)w = c.


109

Exercises 4.2
Therefore w = u' = C sec2 4x and u
4. Define y

= C! tan 4x.

= sin 4x.

A second solution is Y2 = tan 4x cos 4x

= u(x) sin 3x so
y' = 3u cos 3x

+ u' sin 3x,

= u" sin 3x + 6u' cos 3x - 9u sin 3x,

y"

and
y"

If w

+ 9y

+ 6(cos 3x)u' = O or u" + 6(cot 3x)u' = O.


equation w' + 6(cot 3x)w = O which has the integrating

= (sin 3x)u"

u' we obtain the first-order


= sin23x. Now

factor

e6Jcot3xdx

d
.
dx[(sm23x)w]
Therefore w = u'

= c csc2

=O

gives

(sin2 3x)w

= c.

3x and u = Cl cot 3x. A second solution is Y2

= cot 3x sin 3x

= cos 3x.

5. Define y = u(x) coshx so


y' = usinhx

+ u' coshx,

y" = u" coshx

+ 2u' sinhx + ucoshx

and
y" - y = (coshx)u"
If w

= u' we obtain the first-order

e2Jta.nhxdx

= cosh/

+ 2(sinhx)u' = O or
equation w' + 2(tanhx)w

= csech2 x and u

6. Define y = u(x)X

+ 2(tanhx)u'

= O.

= O which has the integrating factor

x. Now
d
dx [(cosh2 x)w]

Therefore w = u'

u"

Cl

= O gives (cosh'' x)w

= c.

tanhx. A second solution is Y2 = tanh x cosh x = sinhx.

so

and
y" - 25y

If w

= X(u" + lOu')

= u' we obtain the first-order equation

w'

= O or

u" + lOu' = O.

+ lOw = Owhich

has the integrating factor elO J dx

elOx. Now
..!!:.... [elOXw] = O gives

Therefore w = u' = ce-lOx


7. Define y = u(x)e2x/3

dx
and u = cle-lOx.

elOxw = c.

A second solution is Y2 = e-loxx

so

and
9y" - 12y'

+ 4y = ge2x/3u" = O.
110

= e-5x.

Exercises 4.2
Therefore u"

= O and u =

+ C2.

CIX

Taking q = 1 and

O we see that a second solution is

C2 =

Y2 = xe2x/3.
8. Define y = u(x )ex/3 so

and

+ y'

6y"

- y

=e

(6u"

X/3

If w = u' we obtain the first-order equation

X/6.

+ 5u')

w' +iw

=O

or

u"

+ ~u' = O.

= Owhich has the integrating factor e(S/6)

Now

Therefore w

= u'

d
- [eSx/6w] = O gives x/6w = C.
dx
and u = cle-Sx/6. A second solution is Y2 = e-Sx/6ex/3

= ce-Sx/6

9. Identifying P(x) = -7/x

we have
Y2=X4

e- J -(7/x)dx
x8
dx=x4

~dx=x41nlxl.

A second solution is Y2 = x41n [z].


10. Identifying P(x) = 2/x we have
Y2 = x2
A second solution is Y2
11. Identifying P(x)

e- J(2/x) dx
x4
dx

x-6dx

1
= -"5x-3.

= x-3.

= L] we have
Y2

= x2

= lnx

e- J dx/x
(In z )? dx = lnx

dx
x(lnx)2

= lnx

(1-lnx ) =-1.

A second solution is Y2 = 1.
12. Identifying P(x) = O we have
e- JOdx
=xl/21nx
x(lnx)2

Y2=x1/21nx!

(1__ ) =_x 2.
1/

lnx

A second solution is Y2 = xl/2.


13. Identifying P(x) = -l/x

we have

. !

Y2 = xsm(1nx)

e- J =dx]
.
2'
2
dx = xsm(1nx)
x sin (lnx)

= [xsin(lnx)] [-cot(lnx)]

= -xcos(1nx).

A second solution is Y2 = x cos(ln x).

111

x
2(
x sin lnx
2'

dx

= e-x/2.

J dx =

Exercses 4.2
14. Identifyng P(x) = -3/x
Y2 =

we have

X2

le

cos(lnx)

- J -3dx/x
4

x cos

2(1

nx

dx

= x 2 cos(lnx)

3
X2(

x cos In x

dx

= x2 cos(ln x) tan(ln x) = x2 sin(ln x).


A second solution is Y2 = x2sin(lnx).
15. Identifying P(x) = 2(1+ x)/ (1- 2x - x2) we have

Y2 = (x

+ 1)

= (z + 1)

e- J2(1+x)dx/(1-2x-x2)
(x + 1)2

dx

2x - x2
+ 1)2 dx = (x

+ 1)

l
l 1-

(x

= (x + 1) [__

(x

+ 1)

l [2 +
(x

e1n(1-2x-x2)
(x+ 1)2 dx

1) 2 - 1 dx

2_ - x] = -2 - x2 - x.

x+l

A second solution is Y2 = x2 +
16. Identifying P(x) = -2x/
Y2 =

(1 -

+ 2.

x2) we have

e- J -2xdx/(1-x2)dx

A second solution is Y2 = In 1(1 + x)/(I17. Define y = u(x)e-2x

e-1n(1-x2)dx

1_11 - x2

dx

= ~ In 1 + xI
2

1- x

x)l.

so

+ u'e-2x,

y' = _2ue-2x

y" = u"e-2x _ 4u'e-2x + 4ue-2x

and
y" - 4y

= e-2xu" - 4e-2xu' = O or u" - 4u' = O.

If w = u' we obtain the first order equation w' - 4w = O whch has the integrating factor e-4

J dx =

e-4x. Now
d
dx[e-4Xwj
Therefore w

u' = cx and u

c1x.

observation that a particular solution is Yp


y
18. Define y = u(x)1

=O

gives e-4xw = c.

A second solution is Y2

e-2xe4x = e2x. We see by

= -1/2. The general solution is


1

= cle-2x + C2e2x - -

so
y' = u',

y"

= u"

and

y"

112

+ y' = u" + u'

= O.

Exercises 4.2
If w = u' we obtan the first order equaton w' + w = Owhch has the ntegrating factor eJ dx = e",
Now
d
.
dx [eXw] = O gives eXw = c.
Therefore w

u'

ce-x and u

cle-x.

A second solution is Y2 = 1 . e-x = e-x.

We see by

observation that a particular solution is Yp = x. The general solution is

19. Define y = u(x)eX so

and
y" - 3y'

+ 2y =

eXu" - eXu' = O

If w = u' we obtain the first order equation w'-w

or

u" - u' = O.

= Owhich has the integratng factor e- J dx = e-x.

Now
d
dx [e-Xw] = O

gives

e-xw = c.

= u' = ce" and u = ce". A second solution is Y2 = eXex = e2x. To find a particular
solution we try yp = Ae3x. Then y' = 3Ae3x, y" = 9Ae3x) and 9Ae3x - 3 (3Ae3x) + 2Ae3x = 5e3x.
Therefore w
Thus A

= 5/2

and yp

20. Define y = u(x)eX

= ~e3x.

The general solution is

so

and
y" - 4y'

+ 3y

= eXu" - 2exu'

=O

or

u" - 2u'

= O.

If w = u' we obtain the first order equation w' - 2w = Owhich has the integrating factor e-2 J dx =
e-2x. Now
d
dx [e-2xw] = O gives e-2xw = c.

= ce2x and u = cle2x. A second solution is Y2 = eXe2x = e3x. To find a particular


solution we try yp = ax + b. Then y~ = a, y~ = O, and 0- 4a + 3(ax + b) = 3ax - 4a + 3b = x.
Then 3a = 1 and -4a + 3b = O so a = 1/3 and b = 4/9. A particular solution is YP = ~x + ~ and
Therefore w = u'

the general solution is

113

Exercises 4.2
2L (a) For mI constant, let YI

= emx. Then

Yl

= mIemX and

y't

= myemX Substituting into the

dfferential equation we obtain

+ by + CYI = amremX + bmIemX + cemX

ay~

Thus, YI = emX will be a solution of the differential equation whenever amI

+ bm; + e = o.

Since a quadratic equation always has at least one real or complex root, the differential equation
must have a solution of the form YI

= emx.

(b) Write the differential equation in the form


y"

+ -yb, + -ye = O,
a

and let YI = emx be a solution. Then a second solution is given by


Y2 = emX

= emX

J --e
J

e-bx/a
2mX dx

e-(b/a+2mtlxdx
emxe-(b/a+2m)

bja

+ 2mI

bja

+ 2mI

Thus, when mI =1- -bj2a,


When mI = -bj2a

(mI =1- -bj2a)

e-(b/a+m)
.

a second solution is given by Y2

= em2x where m2 = -bja - mI

a second solution is given by


Y2 = emX

dx = xemx.

(e) The functions


sin x
sinhx

= 2i (etX
=

e -tX)

2(eX - e-X)

are all expressible in terms of exponential functions.


22. We have Yl = 1 and Y't = 0, so xY't - xYl + YI = O-x + x =
differential equation. Letting y = U(X)YI(X) = xu(x) we get
y' = xu'(x)
Then xy" - xY'
substitution w

+y =

x2u"

+ 2xu'

+ u(x)

and

- x2u' - xu

y" = xu"(x)

+ xu =

and YI(X)

= x is a solution of the

+ 2u'(x).

x2u" - (x2 - 2x)u'

O. If we make the

u', the second-order linear differential equation becomes x2w' - (x2 - x)w

114

= 0,

Exercises 4.2
which is separable:
dw =
dx

(1- ~)w
x

dw
-:;;
= ( 1In w

=x

1)

~ dx

+e

- In x
eX

Then u'

= el-

x
el eX/ x and u = el J e" dx / x o To integrate eX/ x we use the series representation for e" o

Thus, a second solution is


Y2 = xu(x) = elX
= el x

= elX

1(

eX
-;- dx

- 1+x
x

el

+ -x

x2
2!

3!

(~
+ 1 + 2.x + 2.x
x
2!
3!

=elx ( lnx+x+2(2!)x

+-

+ ooo) dx
o o

o)dX

1
+ 3(3!)x

4)
( x In x + x 21+31--x
2(2!)

+ --x

+000

+ ooo

3(3!)

An interval of definition is probably (0,00) because of the In x termo


23. (a) We have y' = y" = eX, so
xy" - (x
and y

+ lO)y' + 10y

= xe" - (x

+ 10)eX + IOe"

= O,

= eX is a solution of the differential equation.

(b) By (5) a second solution is


Y2 = Yl

J P(x)dx

X
dx = e

Yl

x+lnx10

= eX

e2x

dx = eX

x=t
eJ

10

dx

X
dx = e

eJ(1+l0/x)dx
e

dx

2
X

x10e-x dx

= eX( -3,628,800 - 3,628,800x - 1,814,400x2 - 604,800x3


- 30,240x5

5,040x6

= -3,628,800 - 3,628,800x

- 30,240x5

5,040x6

720x7

90x8

10x9

720x7

115

90x8

10x9

151,200x4

xl0)e-x

- 1,814,400x2 - 604,800x3
-

xlOo

151,200x4

Exercses 4.2

(e) By Corollary (A) of Theorem 4.2, --

1101
Y2

lO!

L -n! xn

is a solution.

n=O

Exercises 4.3
= Oand m = -1/4 so that y = Cl + C2e-x/4.
2. From m2 - 36 = O we obtain m = 6 and m = -6 so that y = cle6x + C2e-6x.
1. From 4m2

3.

+ m = O we obtain

= 3 and m = -2 so that y = cle3x + C2e-2x.


m = 1 and m = 2 so that y = clex + C2e2x.

From m2 - m - 6 = O we obtain m

+ 2 = O we obtain
m2 + 8m + 16 = O we obtain m = -4 and m = -4 so that
m2 - 10m + 25 = O we obtain m = 5 and m = 5 so that y

4. From m2 - 3m
5. From
6. From

7. From 12m2 - 5m - 2 = O we obtain m

= Cle-4x + c2xe-4x.

+ c2xx.

= clx

= -1/4 and m = 2/3 so that y

= cle-x/4

- 1 = O we obtain m == -2 V5 so that y = cle( -2+v's)x

8. From m2

+ 4m

9. From m2

+ 9 = O we obtain m = 3i and m = -3i so that y = Clcos 3x + C2sin 3x.

+ C2e(-2-v's)x.

+ 1 = O we obtain m = i/vIJ and m = -i/vIJ so that y = Clcosx/vIJ + C2sinx/vIJ.


m2 - 4m + 5 = O we obtain m = 2 iso that y = e2x(cl cos x + C2sin x).
2m2 + 2m + 1 = O we obtain m = -1/2 i/2 so that

10. From 3m2


11. From
12. From

y = e-x/2(cl cos x/2


13. From 3m2

+ 2m + 1 = O we obtain

14. From 2m2

3m

+ 4 = O we obtain
y

From m3 - 4m2 - 5m

m = 3/4

= e3x/4

+ C2sin x/2).

= -1/3 V2i/3 so that

y = e-x/3 (Cl COS J2 x/3

15.

+ C2e2x/3.

+ C2sin J2 x/3) .

V23i/4 so that

(Cl COS V23 x/4

+ C2sin v'23 x/4)

= O we obtain m = O, m = 5, and m = -1 so that

16. From m3 - 1 = O we obtain m = 1 and m = -1/2 vlJi/2 so that


y = qeX
17. From m3 - 5m2

+ e-x/2

+ 3m + 9 = O we obtain

(C2cos J3 x/2
m = -1, m

y = Cle-x

+ C3sin J3 x/2)

= 3, and m

+ C2e3x + C3xe3x.
116

= 3 so that

Exercises 4.3
18. From m3 + 3m2

4m - 12 = O we obtain m

= -2, m = 2, and m = -3 so that

+ C2e2x + C3e-3x.

y = cle-2x
19. From m3 + m2

2 = O we obtain m = 1 and m

= clet + e-t(c2cost

u
20. From m3 - m2

4 = Owe obtain m

= -1

= 2 and m

iso that
+ C3sint).

= -1/2 V7i/2 so that

x = Ce2t + e-t/2 (C2 cos J7 t/2 + C3sin J7 t/2) .


21. From m3 + 3m2 + 3m + 1 = O we obtain m = -1, m = -1, and m = -1 so that
y

= Ce -x + C2xe -x + C3X2e -x .

22. From m3 - 6m2 + 12m - 8 = Owe obtain m = 2, m = 2, and m = 2 so that


y = CIe2x
23. From m4 + m3 + m2

= Owe obtain m

+ C2xe2x + c3x2 e2x.

= O, m

= O, and m = -1/2 V3 i/2 so that

y = CI + C2X + e-x/2 (C3 cos V3x/2


24. From m4 - 2m2 + 1 = Owe obtain m

+ C4sin V3x/2)

= 1, m = 1, m = -1, and m = -1 so that

25. From 16m4 + 24m2 + 9 = Owe obtain m = V3 i/2 and m


y = CIcos V3x/2
26. From m4 - 7m2

= V3 i/2 so that

+ C2sin V3 x/2 + C3Xcos V3 x/2 + C4Xsin V3 x/2.

18 = Owe obtain m

= 3, m = -3,

and m = J2 i so that

y = CIe3x + C2e-3x + C3cos J2 x + C4sin J2 x.


27. From mS + 5m4

2m3

10m2 + m + 5

= Owe obtain m

= -1, m = -1, m

= 1, and m = 1, and

m = -5 so that
U=Cle
28. From 2ms - 7m4 + 12m3 + 8m2
x

+C2re

-r

+C3e

+C4re

= Owe obtain m = O, m

+cse

= O, m

-Sr

= -1/2, and m

= 2 2i so that

= CI+ C28 + C3e-s/2 + e2s(c4 cos 28 + Cssin 28).

29. From m2 + 16 = Owe obtain m


then Cl = 2, C2= -1/2,

-r

= 4i so that y = CICOS 4x + C2sin 4x. If y(O) = 2 and y' (O) = -2

and y= 2cos4x-

~sin4x.

30. From m2 + 1 = O we obtain m = i so that y = CIcose + C2sine. If Y(7f/3) = O and y'(7f/3)


1
V3
V3
1
.
then "2C+ 2C2 = O, -2CI
+ "2C2= 2, so CI = -V3, C2= 1, and y = -V3 cose + sm e.

117

=2

Exercises 4.3
31. From m2 - 4m - 5

and y'(l)
y =

-tel-x

O we obtain m

+ c2es =

2, then cle-l

-1 and m

+ 5c2es =

O, -cle-l

5, so that y

2, so Cl

+ c2x.
= -e/3, C2 =
cle-x

If Y(l) = O
e-s /3, and

+ tx-s.
= O we obtain m = -1/2 and m

32. From 4m2-4m-3

= 5 then

= 1, -~Cl

= 3/2 so that y

= 5, so ci =

= cle-x/2+c2e3x/2.

Ify(O)

=1

-7/4, C2= 11/4, and y

= _e-x/2+ lie3x/2.
33. From m2+m+2 = Owe obtain m = -1/2/7
i/2 so that y = e-x/2 (c cos /7 x/2 + C2sin /7 x/2).
If y(O) = O and y'(O) = O then Cl = O and C2= O so that y = O.
34. From m2 - 2m + 1 = O we obtain m = 1 and m = 1 so that y = cex + c2xex. If y(O) = 5 and
y'(O) = 10 then Cl = 5, Cl + C2= 10 so Cl = 5, C2= 5, and y = 5ex + 5xex.
35. From m3+12m2+36m
= Owe obtain m = O,m = -6, and m = -6 so that y = Cl+c2e-6x+c3xe-6x.
and y'(O)

If y(O) = O, y'(O)

c +C2

+~C2

1, and y"(O) = -7 then


Cl + C2= O,

-6C2

+ C3= 1,

36c2 - 12c3 = -7,

+ gxe-6x.
m = 2, and m =

so ci = 5/36, C2= -5/36, C3= 1/6; and y = 356 - 3s6e-6x


36. Frorn m3

+ 2m2

5m - 6 = Owe obtain m
y

If y(O)

= O, y'(O)

= O, and y"(O)

-1,

-3 so that

= cle-x + C2e2x + c3e-3x.

= 1 then

so Cl = -1/6, C2= 1/15, C3= 1/10, and


1
y = __e-x
6

1
1
+ _e2x
+ _e-3x.
15

10

37. From m2 - 10m + 25


38.

= Owe obtain m = 5 and m = 5 so that y = clx + c2xx. If y(O) = 1 and


y(l) = O then Cl = 1, cles + C2es = O, so Cl = 1, C2= -1, and y = x - xesx.
Frorn m2 + 4 = O we obtain m = 2i so that y = Clcos2x + c2sin2x.
If y(O) = O and Y(7r) = O
then c = O and

39. Frorn m2

y = C2sin

2x.

+ 1 = O we obtain

+ C2sinx.

m = i so that y = Cl COSX

If y'(O) = O and y'(7r/2) = 2

then Cl = -2, C2= O, and y = -2cosx.

+ 2 = O we obtain m = 1 iso
ci = 1 and Y(7r) = e1l'COS7r = =e",

40. Frorn m2 - 2m
Y(7r)

1 then

that y

eX(cl cosx

+ C2sinx).

If y(O)

1 and

Since _e1l' =f 1, the boundary-value problem has

no solution.
41. The auxiliary equation is m2 - 3 = O which has roots -V3

= cleV3x + C2e-V3x.

and V3. By (10) the general solution

= clcoshV3x + c2sinhV3x. For


y = cleV3x + c2e-V3x the initial conditions imply Cl + C2= 1, V3Cl - V3C2 = 5. Solving for Cl + C2
is y

By (11) the general solution is y

118

Exercises

4.3

we find CI = i(3 + 5V3) and C2 = i(3 - 5V3) so y = i(3 + 5V3)ev'3x + i(3 - 5V3)ev'3x. For
y = CIcosh V3x + C2sinh V3x the initial conditions imply CI = 1, V3C2 = 5. Solving for CI and C2

we find CI = 1 and C2= V3 so y = cosh V3x +

iV3 sinh V3x.

42. The auxiliary equation is m2 - 1 = O which has roots -1 and 1. By (10) the general solution is
y

= qeX

+ c2e-x.

By (11) the general solution is y = CIcoshx + C2sinhx. For y

boundary conditions imply CI+C2 = 1, cle-c2e-1

= clex

+ c2e-x the

= O. Solving for CI and C2we find CI = 1/(1+e2)

and C2= e2/ (1+ e2) so y = eX/(1 + e2) + e2e-x /(1 + e2). For y = CIcosh x + C2sinh x the boundary
conditions imply CI = 1, C2= - tanh 1, so y = cosh x - (tanh 1) sinh x.
43. The auxiliary equation should have two positive roots, so that the solution has the form y =
cl1X + c22X. Thus, the differential equation is (f).
44. The auxiliary equation should have one positive and one negative root, so that the solution has the

= cl1X + c2e-k2X. Thus, the differential equation is (a).


45. The auxiliary equation should have a pair of complex roots a bi where
form y

has the form

eax (CIcos bx

< O, so that the solution

+ C2sin bx). Thus, the differential equation is (e).

46. The auxiliary equation should have a repeated negative root, so that the solution has the form

= cle-x + C2xe-x. Thus, the differential equation is (c).


47. The differential equation should have the form y" + k2y = O where k
y

= 1 so that the period of the

solution is 27r. Thus, the differential equation is (d).


48. The differential equation should have the form y" + k2y

= O where k

= 2 so that the period of the

solution is n. Thus, the differential equation is (b).


49. (a) The auxiliary equation is m2 - 64/L = O which has roots 8/JI.
of the differential equation is x = CIcosh(8t/JI)
(b) Setting x(O)

= Xo and x'(O)

Thus, the general solution

+ C2sinh(8t/JI).

= O we have CI = xo, 8C2/JI

= O. Solving for CI and C2we get

CI = Xo and C2= O, so x(t) = xocosh(8t/JI).


(e) When L = 20 and Xo

= 1, x(t)

= cosh(4tv's).

The chain will last touch the peg when

x(t) = 10. Solving x(t) = 10 for t we get tI = tv'scosh-I10::::::


chain at this instant is X'(tl) = 12il75
50. Both -e[l]

1.67326. The velocity of the

::::::
17.7989 ft/s.

and CI represent arbitrary constants, and each may take on any real value.

51. Since (m- 4) (m+5)2

= m3 +6m2

-15m

-100 the differential equation is ylll +6y" -15y' -100y

= O.

The differential equation is not unique since any constant multiple of the left-hand side of the
differential equation would lead to the auxiliary roots.
52. A third root must be m3 = 3 - i and the auxiliary equation is

(m + 21) [m -

(3 + i)][m - (3 - i)] = (m +

1 (m2
2)

119

11
- 6x + 10) = m3 - 2m2

+ 7m + 5.

Exercises 4.3
T4e differential equation is
11

y'" - -;y" + 7y' + 5y = O.


53. From the solution YI = e-4x cos x we conclude that mI = -4 + i and m2 = -4 - i are roots of the
auxiliary equation. Hence another solution must be
m3

+ 6m2 + m -

34 by [m - (-4

+ i)][m

Y2 =

- ( -4 - i) 1 = m2

e-4x sin x. Now dividing the polynomial

+ 8m + 17 gives m

- 2. Therefore m3 = 2

is the third root of the auxiliary equation, and the general solution of the differential equation is

54. Since l/x

-+

O as x

-+ 00,

to the solutions of y"

+y

we would expect the solutions of y"

+ (l/x)y' + y

= Oto behave similar

O; that is, like sin x and cos x for large values of x.

Solutions of

xy" + y' + xy = Oare obtained using an ODE solver and are shown below with the indicated initial
conditions.
y

y
1

1
0.5

-0.5
-1

-1

y(l)

= O,

y'(l)

=2

y(l)

= 2,

y'(l)

=O

55. Factoring the difference of two squares we obtain


m4

+ 1 = (m 2 + 1) 2 -

2m2 = (m2

+ 1 - V2 m) (m 2 + 1 + V2 m)

Using the quadratic formula on each factor we get m

= O.

= V2/2 V2i/2. The solution of the

differential equation is

y(x)

= eV2x/2

(CI cos

V; x +

56. (a) The auxiliary equation m2

C2 sin

V; x) +

+ bm + C = Ohas

e-V2x/2

solutions m

(C3

cos

V;

+ C4 sin

V; x) .

= (-b Jb2 - 4c )/2. If b :::;O, then

> O, and the solution cannot approach


O as x -+ oo. Thus, for the solution to approach O we must have b > O. Now, if c < O then
Jb2 - 4c > b and -b + Jb2 - 4c > O. Thus y(x) cannot approach O. Finally, if e > O then
Jb2 - 4c < b and -b Jb2 - 4c < O. In this case the solution has terms of the form ef3x where
(3 < O. Therefore y(x) -+ O as x -+ 00 if and only if b > Oand e > O.
(b) If b2 - 4c > O, then y = cIemX + C2em2x and the only solution satisfying y(O) = O, y(l) = Ois
the solution will contain a term of the form ef3x for (3

= O.
120

Exercises 4.3
If b2 - 4c

= o, then

= eemx + C2temx.

Again, the only solution satisfying y(O)

= o, y(l) = O

is y = O.
If b2

4c

Now y(O)

< O then

= O implies

CI = O and
y

= c2e-bx/2 sin 4c - b2 x.

If we are to have a nontrivial solution, the condition y(l)


4c - b2

O implies ~

= nm or

= n27l'2 for n a positive integer.

57. The auxiliary equation is m2 + A = O and we consider three cases.


Case I When A

O the general solution of the differential equation is y = CI + C2X. The boundary

conditions imply O = y(O)

= CI and

= y( 7l'/2)

= C27l'
/2, so that c

= C2 =

O and the problem

possesses only the trivial solution.

< O the general solution of the differential equation is y = cleV-"Xx+ C2e-V-"Xx,


or alternatively, y = CIcosh H
x + C2sinh H
x. Again, y(O) = O implies CI = O so y =
c2sinhHx.
The second boundary condition implies O = Y(7l'/2) = c2sinhH7l'/2
or C2= O. In
Case II When A

this case also, the problem possesses only the trivial solution.
Case III

When A

> O the general solution of the differential equation is y

C2sin Vf.. z: In this case also, y(O)

= O yields

condition implies O = C2sin Vf.. 7l'/2.

Vf.. = 2k for k

CI = O, so that y

When

Vf.. 7l'/2

= C2sin Vf.. x.

= CICOS Vf.. x

The second boundary

is an integer multiple of

tt ,

that is, when

a nonzero integer, the problem wiJI have nontrivial solutions. Thus, for A = 4k2 the

boundary-value problem wiJI have nontrivial solutions y


On the other hand, when

Vf.. is not

= C2sin 2kx, where k is a nonzero integer.

an even integer, the boundary-value problem will have only the

trivial solution.
58. Applying integration by parts twice we have
J eax f(x) dx

= ~ eax f(x) - ~ J eax f'(x) dx


= ~eaxf(x)
a

- ~ [~eaxf'(x)
a a

- ~Jeax!"(x)dx]
a

= ~ eax f(x) - 2_ eax f'(x) + 2_ J eax f"(x) dx.


a

a2

a2

dx

= ~ eax f(x) -

Collecting the integrals we get


J eax (f(x)

:2

!"(x))

121

:2

eax f'(x).

4.3

Exercises

In order for the technique to work we need to have

eax (f(x)

:2

f"(x))

dx

eax f(x) dx

or
f(x)

- ~ f"(x)
a

= kf(x),

where k =1- O. This is the second-order differential equation


f"(x)

If k

< 1, k

+ a2(k

- l)f(x)

= O.

=1- O, the solution of the differential equation is a pair of exponential functions, in which

case the original integrand is an exponential function and does not require integration by parts for
its evaluation. Similarly, if k

= 1, f" (x) = O and

f (x) has the form f (x)

ax

+ b.

In this case a

single application of integration by parts will suffice. Finally, if k > 1, the solution of the differential
equation is
f(x)

= Clcos av'k=!

+ C2sin a..Jk=l

x,

and we see that the technique will work for linear combinations of cos ax and sin ax.
59. Using a CAS to solve the auxiliary equation m3 - 6m2
m2

= -0.270534,

= 0.658675, and m3 = 5.61186. The general solution is

60. Using a CAS to solve the auxiliary equation 6.11m3


ml

+ 2m + 1 we find ml

+ 8.59m2 + 7.93m + 0.778 = Owe find

= -0.110241, m2 = -0.647826 + 0.857532i, and m3 = -0.647826 - 0.857532i. The general

solution is
y = Cle-O.l1024lx+ e-O.647826x

(C2

cos 0.857532x

61. Using a CAS to solve the auxiliary equation 3.15m4


ml

-1.74806, m2

- 5.34m2

+ C3sin 0.857532x).
+ 6.33m

- 2.03

= O we find

= 0.501219, m3 = 0.62342 + 0.588965i, and m4 = 0.62342 - 0.588965i. The

general solution is
y = Cle -1.74806x

+ c2eO.50l2l9x+ eO.62342x (C3 cos 0.588965x + C4sin 0.588965x).

62. Using a CAS to solve the auxiliary equation m4


m2

= 1/2 - /3i/2,

m3 = -1/2

+ V7i/2,

+ 2m2

and m4

- m + 2 = O we find ml = 1/2 + /3
= -1/2 - V7i/2. The general solution is

122

i/2,

Exercises

+ 3m3 - 16m2 + 15m - 4 = O we obtain m = -4, m = ~, m = 1, and m = 1, so that


+ c2ex/2 + C3ex + C4xex. If y(O) = -2, y'(O) = 6, y"(O) = 3, and y"'(O) = ~, then

63. From 2m4


y

4..4

= cle-4x

CI + C2 + C3 = -2
1

-4CI

+ "2C2+ C3+ C4 = 6

16cI

+ C2 + C3+ 2C4 = 3

-64cI
_

116

_ 918

+ SC2 + C3+ 3C4 = "2 '

58

so CI - -'75 ' C2 - -"""3 ' C3- 25 ' C4- -"5 ' an

4 -4x
116 x/2
y = - 75 e
- Te
64.

From m4 - 3m3
y

+ 3m2

- m

= O we obtain m

= CI + C2ex + c3xex + C4x2ex.

918 x
58 x
25 e - 5xe

= O, m = 1, m = 1, and m

If y(O) = O, y'(O)

so CI = 2, C2 = -2, C3 = 2, C4 = -1/2,

= O,

= 1 so that

y"(O) = 1, and y"'(O) = 1 then

and

Exercises 4.4
1. From m2

yp

+ 3m + 2 = O we find

mI = -1 and m2 = -2. Then Yc = cle-x

+ C2e-2x

+ C2e-2x + 3.

y = cle~x
2. From 4m2

+ 9 = O we find mI = -~

i and m2 = ~i. Then Yc = CI COS ~x + C2sin ~x and we assume

yp = A. Substituting into the differential equation we obtain 9A = 15. Then A


3
y = CI COS "2 x
3.

and we assume

= A. Substituting into the differential equation we obtain 2A = 6. Then A = 3, yp = 3 and

From m2 - 10m

+ 25

. 3

= ~, yp

= ~ and

+ C2SIn "2 x + "3 .

= O we find mI = m2 = 5. Then Yc = clx

+ c2xx

and we assume

yp = Ax + B. Substituting into the differential equation we obtain 25A = 30 and -lOA


- 6 ' B --:5 6 ' yp -- 6:5x + 6:5' an d
Th en A -:5

123

+ 25B

= 3.

Exercises 4.4
4. From m2

+ m - 6 = O we find m1 = -3 and m2 = 2. Then Yc =

Ax

-3x

y=C1e

+C2e

2x

1
--x--

+ C2e2x and we assume


= 2 and A - 6B = O. Then

C1e-3x

+ B.

Substituting into the differential equation we obtain -6A


1 YP -- -"J1x - 18'
1 an d
A -- -3'1 B-- -18'

Yp

18

+ m + 1 = O we find m1 = m2 = O. Then Yc = C1e-2x + C2xe-2x and we assume


Yp = Ax2 + Bx + C. Substituting into the differential equation we obtain A = 1, 2A + B = -2,
and ~A + B + C = O. Then A = 1, B = -4, C = ;, Yp = x2 - 4x +;, and

5. From tm2

y = C1e-2x

6. Frorn m2 - 8m

+ c2xe-2x + x2

+ 2"'

- 4x

+ 20 = Owe find m1 = 2 + 4i and m2 = 2 - 4i. Then Yc = e2x(c1 cos4x + C2sin4x)


+ Bx + C + (Dx + E)ex. Substituting into the differential equation we

and we assume Yp = A.x2


obtain

2A - 8B

+ 20C

13E = O

-6D+
-16A

=O

+ 20B

=O

13D = -26
20A = 100.

Then A = 5, B

i,

= 4, C =
y

D = -2, E

-H, Yp = 5x2 + 4x + i + (-2x

+ C2sin4x) + 5x2 + 4x + ~~+ ( -2x

e2x(c1 cos 4x

iD eX and

- ~~) eX.

7. Frorn m2 + 3 = O we find m1 = Y3i and m2 = -Y3i.


Then Yc = c1cosY3x + C2sin Y3x
and we assume Yp = (Ax2 + Bx + C)e3x. Substituting into the differential equation we obtain

+ 6B + 12C = O, 12A + 12B =


Yp = ( -4x2 + 4x - 1) e3x and
2A

+B

Then A

= Clcos Y3 x + C2sin Y3 x + ( -4x2 + 4x

8. Frorn 4m2 - 4m - 3 = Owe find ml


YP = A cos 2x

O, and 12A = -48.

= -4,

= 4, C = -j,

- ~) e3x.

= ~ and m2 = -~ . Then Yc = cle3x/2 + C2e-x/2 and we assume

sin 2x. Substituting into the differential equation we obtain -19 - 8B = 1 and

8A - 19B = O. Then A

-lis,

y=cle

=-

4~5 '

YP =

- 41iscos 2x

3x/2 +C2e -x/2 --cos


19

425

124

4~5

2x--sm 8.
425

sin 2x, and


2x.

Exercises 4.4
= clex + C2 and we assume Yp = Ax.
Substituting into the differential equation we obtain -A = -3. Then A = 3, Yp = 3x and

9. From m2 - m = O we find mI = 1 and m2 = O. Then Yc

y =

Cl e"

+ C2 + 3x.
+

10. From m2

2m

O we find

+ Ex + Cxe-2x.

Yp = Ax2

mI

and

-2

m2 =

O. Then Yc

4A = 2, and -2C = -l. Then A = ~, E = 2, C = ~, Yp = ~x2 + 2x


Y = cle-2x

C2

and we assume

+ 2E =

5,

+ ~xe-2x, and

+ C2 + _x2 + 2x + _xe-2x.

+ ~ = O we find mI = m2 = ~.
+ Ex2ex/2. Substituting into the differential

11. From m2 - m

Yp = A

cle-2x

Substituting into the differential equation we obtain 2A

Then Yc

= Clex/2 + C2xex/2 and we assume


= 3 and 2E = 1. Then

equation we obtain ~A

A = 12, E = ~, Yp = 12 +~x2ex/2, and

12. From m2 - 16

Yp = Axx.
and

13. From

m2

= O we find

mI

+4

= O we find mI = 2i and m2 = -2i.

Yp = Ax cos 2x

+ Ex

-4A = 3. Then A

sin 2x. Substituting

= -~,

E = O, Yp
y

14. From m2

Yp = (Ax3

= 4 and m2 = -4.

Then Yc = clx

+ C2e-4x

and we assume

Substituting into the differential equation we obtain 8A = 2. Then A = ~, YP = ~xx

Then Yc =

Cl COS2x

C2 sin

2x and we assume

into the differential equation we obtain 4E = O and

= -ixcos2x,

ci COS2x

and

+ C2sm 2x - '4 x cos 2x.

+ 4 = O we find mI = 2i and m2 = -2i. Then


+ Ex2 + Cx) cos 2x + (Dx3 + Ex2 + Fx) sin 2x.

we obtain
2E

+ 4F

6A+8E

=O

=O

12D = O
-4C

+ 2E

-8E+6D

=-3
=O

-12A = l.

125

Yc = Cl COS2x

+ C2 sin 2x

and we assume

Substituting into the differential equation

Exercises 4.4

Then A

=-

/2 ' B

0, C

= ~ , D = 0,

-h, F = 0, YP = ( -112x3 + ~~x) cos 2x + l6x2 sin 2x,

and

+ C2sin 2x + ( -112x3 + ~~x)

y = C cos 2x
15. From m2
(Ax2

+ 1=

+ Bx)

cos x

we find mi

= i and

+ (Cx2 + Dx)

m2

+ 2C =

into the differential

O. Then A

-4,

YP = -lOx

14 3

+ 75x +

into the differential

5x
= ce

1 4

+ C2-

10x

we obtain -20A

equation

and 2G - 5D = 6. Then A = -l~'


53 2 697
d
250x - 625x, an

0, C

0, D

= ~, YP =

= O. Then Yc = cle5x + C2 and we assume YP =

we find mi = 5 and m2
Substituting

= -~, B =

= -1,

6B - 10G
1 4

we obtain 4C = 0,

equation

1 2
l.
2'x cosx+2xsmx.

Y=C1COSX+C2sinx16. From m2 - 5m =
Ax4+Bx3+Cx2+Dx.

2x.

= -i. Then Yc = Cl COS X + C2sin x and we assume YP =

sin x. Substituting

2A + 2D = 0, -4A = 2, and -2B


- ~x2 COS X + ~x sin x, and

+ 116x2sin

cos 2x

14 3

53

= 25io',D = -~~~,

697

+ 75x + 250x

i:,

= 2, 12A-15B

- 625x.

17. Prom m2 - 2m + 5 = we find m = 1+ 2i and m2 = 1- 2i. Then Yc = eX (Cl COS 2x + C2sin 2x) and
we assume YP = Axex cos 2x + Bxe" sin 2x. Substituting
into the differential equation we obtain
4B = 1 and -4A

= O. Then A

= 0, B =

i, YP = ixex

y = eX(c cos 2x
18. From m2 - 2m

+2

+ 2B

= 1 and - 2A

+ C2sin 2x) + xex

we find mI = 1 + i and m2

and we assume YP = Ae2x cos x

sin 2x, and

+B

+ Be2x

= 1-

sin x. Substituting

sin 2x.
i. Then Yc

= eX(c cosx + C2sinx)

into the differential

equation

we obtain

= -3. Then A = ~, B = - i ' YP = ~e2x cos x - ie2x sin x and

y = eX(C cos x

7 2
+ C2sin. x) + Se
X cos x

l.
- Se 2x sin x.

19. Prom m2 + 2m + 1 = we find mI = m2 = -l. Then Yc = ce-x + C2xe-x and we assume


YP = A cos x + B sin x + C cos 2x + D sin 2x. Substituting into the differential equation we obtain
2B

= 0, -2A = 1, -3G + 4D = 3,

YP =

-"21 cos

x -

9 COS 2 X
25

and -4G - 3D
2
2
25 sin x, an d

Y = ce20.

From m2

+ 2m

assume YP = A

- 24 =

+ C2xe-

2'1 cos x

we find m = -6

+ (Bx2 + Cx)x.

Substituting

= O.

Then

A = -~, B = 0, C = -295' D =

9
- 25 cos 2x

and m2

4.

12

+ 25 sin 2x.

Then

into the differential

126

~,

Yc

equation

cle-6x

+ C2x

we obtain -24A

and we
= 16,

Exercises 4.4
2E+10C
and

= -2, and 20E = -1. Then A = -~, E =


y = CIe

21. From m3 - 6m2

-6x

-io'

2 (120 x

+ C2e4x - 3" -

C=
2

-lgo' Yp = _~_(iox2

+ I~OX)x,

19)

4x
100 x e .

= O we find mI = m2 = O and m3 = 6. Then Yc = CI + C2X+ c3x and we assume

Yp = Ax2 + E cos x + C sin x. Substituting into the differential equation we obtain -12A
6E - C
and

= -1, and E + 6C = O. Then A = -~, E =


6x
y = CI + C2X+ C3e

22. From m3 -2m2 -4m+8

= O. Then A = ~, E

1 2
6
1 .
x - 37 cos x + 37 sin x.

-"4

= cle2x+c2xe2x+c3e-2x

Substituting into the differential equation we obtain 24A = 6

= -136' YP = (~x3 - 136X2)


e2x, and

= Cle2x +

3,

= 3\' Yp = _~x2 - 367cosx+ 3\ sinx,

= Owe find mI = m2 = 2 and m3 = -2. Then Yc

and we assume yp = (Ax3 + Ex2)e2x.


and 6A + 8E

-:f7,

C2xe2x + C3e-2x + (~x3 _ 2.x2) e2x.


4
16

= O we find mI = m2 = m3 = 1. Then Yc = Clex + C2xex + C3x2ex and


we assume Yp = Ax + E + Cx3ex. Substituting into the differential equation we obtain -A = 1,

23. From m3 - 3m2 + 3m - 1


3A - E

= O, and

6C = -4. Then A = -1, E


y

= CIe

+ C2xex + C3X2xe

= Owe find mI

24. From m3-m2-4m+4

and we assume Yp = A+Exex+Cxe2x.


-3E

= -3,

= 1, m2

C =

-, Yp = -x

- 3 - ~x3ex, and

23x e .
- x - 3 - 3"x

= 2, and m3 = -2. Then Yc = clex+c2e2x+c3e-2x

Substituting into the differential equation we obtain 4A = 5,

= -1, and 4C = 1. Then A = ~, E

t, C = ~, Yp = ~+ txex

+ ~xe2x, and

5 1
1
y = clex + C2e2x + C3e-2x + - + -xex + _xe2x
4

25. From m4 + 2m2 + 1 = O we find mI = m3 = i and m2 = m4 = -i. Then Yc = CIcosx + C2sinx +


C3Xcos x + C4Xsin x and we assume Yp = Ax2 + Ex + C. Substituting into the differential equation
we obtain A

= 1, E

= -2, and 4A + C

= 1. Then A = 1, E

= -2, C

= -3, Yp = x2 - 2x - 3, and

y = C cosx + c2sinx + C3Xcosx + c4xsinx + x2 - 2x - 3.


26.

From m4 - m2

= O we find mI

= m2 = O, m3 = 1, and m4

and we assume Yp = Ax3 + Ex2 + (Cx2 + Dx)e-x.


obtain -6A

= 4, -2E

O, lOC - 2D

= -1. Then

Yc

= CI+ C2X+ C3ex+ C4e-x

Substituting into the differential equation we

= 0, and -4C = 2. Then A = -~, E = O, C = -~,

D = -~, Yp = _~x3 - Gx2 + ~x) e-x, and

127

Exercises

4.4

27. We have Ye
we find A

= Cl COS2x + C2sin 2x and

= - ~.

we assume Yp

= A.

Substituting into the differential equation

Thus y = ClCOS2x + C2sin 2x - ~. From the initial conditions we obtain c

and C2= j2, so y

= j2 sin 2x - ~ .

28. We have Ye = cle-2x + c2ex/2 and we assume Yp = Ax2 + Ex + C. Substituting

equation we find A

=O

-7, E

= -19, and

From the initial conditions we obtain Cl =


1 -2x
y = -Se

e=

-37. Thus y

= cle-2x

into the differential

+ c2ex/2 - 7x2 - 19x - 37.

-! and C2= l~6 , so

186 x/2
+5
e -

7x - 19x - 37.

= cle-x/5+c2 and we assume Yp = Ax2+Ex. Substituting into the differental equation


we find A = -3 and E = 30. Thus y = cle-x/5 + C2- 3x2 + 30x. From the initial conditions we

29. We have Ye

obtain Cl = 200 and C2= -200, so


y

= 200-x/5

200 - 3x2

+ 30x.

30. We have Ye = cle-2x + c2xe-2x and we assume YP = (Ax3 + Ex2)e-2x.

differential equation we find A = ~ and E = ~. Thus y

Substituting

= cle-2x + C2xe-2x +

(ix3

into the

+ ~x2) e-2x.

From the initial conditions we obtain Cl = 2 and C2= 9, so


y

= 2e-2x + 9xe-2x + (~x3 + ~x2) e-2x.

= e-2x(cl COSX+C2sinx) and we assume YP = Ae-4x. Substituting into the differential


equation we find A = 5. Thus y = e- 2x(ClCOSX + C2sin x) + 7e-4x. From the initial conditions we

31. We have Ye

obtain Cl = -10 and C2= 9, so


y = e-2X(-10cosx + 9sinx + 7e-4x).
32. We have Ye

= Cl coshx

+ C2sinhx and we assume Yp

= Ax cosh x + Ex

sinh z. Substituting into the

differential equation we find A = O and E = ~. Thus


Y

c coshx + c2sinhx + '2xsinhx.

From the initial conditions we obtain Cl = 2 and C2= 12, so


y

= 2coshx + 12sinhx + ~xsinhx.

33. We have x = Cl coswt + C2sinwt and we assume xp

= Atcoswt + Etsinwt.

Substituting into the

differential equation we find A = -Fo/2w and E = O. Thus x = Clcoswt+C2 sinwt-(Fo/2w)tcoswt.


From the initial conditions we obtain Cl = O and C2= Fo/2w2, so
x

= (Fo/2w2)sinwt - (Fo/2w)tcoswt.
128

Exercises 4.4
34. We have x = q cos wt
into the differential

+ C2sin wt

equation
X

+B

and we assume xp = A cos ,t

we find

A = Fo/(w2

= C1cos wt

sin ,t, where , =J w. Substituting

B = O. Thus

- ,2) and

Fo

+ c2 sin wt + (2w -,

2) cos ,t.

From the initial conditions we obtain Cl = Fo/(w2 _,2) and C2 = O, so


Fo
Fo
X

35. We have Yc
differential

Cl

= (w 2 -,'

+ C2ex + C3xex
we find A

equation

2, B

and

+ Bx2ex + c-.

Ax

Substituting

into the

C = ~. Thus

+ C3xe x + 2X + 9xex + 2x

12X 2ex +"2e


1 5x

into the differential

equation

we find A

= C1e-2x + eX(c2 cos j3 x + C3sin

1
2

+ _x.

- 12x2ex

+ eX(c2 cos j3 X + C3sin j3 x)

36. We have Yc = c1e-2x

-12,

2) cos-rz.

-,

we obtain C1 = 11, C2 = -11, and C3 = 9, so

y = 11 - llr.

Substituting

+ (2w

and we assume yp

y = C1 + C2ex
Frorn the initial conditions

2) coswt

and we assume

yp = Ax

t ' B = - ~, and

J3 x) + 4"x -

= ~.

+ B + Cxe-2x.
Thus

8" + "3xe-2X.

Fr om thee iun't'la 1 con diti


. C1 = -12
23 ' C2 = -"24'
59 an d C3 = 72
17 v;:;3
1 ions we o b tam
o , so
2x +e x (59
r:
17 r: :
--cosv3x+-v3smv3x
24
72

y=--e- 23
12

37. We have Yc = C1COS X + C2sin x and we assume yp = A 2 + Bx

equation

we find A

1, B

and y(l)

= O we obtain

= O, and

1
5
+-x--+-xe-.
4
8

;:;)

+ C.

C = -1. Thus y = C1COSX

Substituting

2x

3
into the differential

+ C2sinx + x2 - 1.

From y(O)

=5

C1- 1 = 5
(cos 1)C1 + sin(1)c2 = O.
Solving this system we find C1 = 6 and C2 = -6 cot 1. The solution of the boundary-value problem
is

y = 6 cos x - 6( cot 1) sin x


38. We have Yc = eX(c1 cosx
equation

we find A

+ C2sin x)

+ x2

and we assume yp = Ax

1 and B = O. Thus y

- 1.

+ B.

Substituting

= eX(c1 COSX+C2 sinx)+x.

we obtain
C1 = O

129

into the differential

From y(O)

= O and Y(7r) = 7r

Exercises 4.4
Solving this system we find Cl = O and C2 is any real number.

The solution

of the boundary-valus

problem is

39. We have Yc

Cl COS2x

into the differential

+ C2sin 2x

and we assume Yp

we find A

equation

O and B =

we have y = C3cos 2x

[O,7r/2]. On (7r/2,00)

=A

cos x

1 Thus

+ C4sin 2x.

+B

sin x on [O,7r/2]. Substituting

+ C2sin 2x + 1sin x

y = q cos 2x

on

Frorn y(O) = 1 and y'(O) = 2 we obtain

q =1
1

Solving this system


Now continuity

we find Cl

1 and

1= -C3

Hence C3 =

. Continuity

-i = -2C4

differential
have y

= eX(q

cos3x

={

cos 3x

Solving this system,

7r

+ c2sin3x)

+ C4sin3x).

is

O:::; x :::;tt /2

x> 7r/2.

Frorn y(O) = O and y'(O)

= -2,

+ 3q

~.

Thus y

+ C2sin 3x) + 2 on
=

into the

[O,ttJ. On (7r,00)' we

O we obtain

= O.

= eX(

-2 cos3x

+ ~sin3x) + 2 on

[O,7r].

= 7r implies

+ ~Sin37r) + 2 = e1l'(c3cos37r + c4sin37r)

or C3 = -2e-1l'(1

+ e"}.

continuity

of y' at 7r implies

230e1l'sin 37r = e1l'[(c3 + 3C4) cos 37r + (-3C3


or -c3e1l' - 3C4e1l'= O. Since C3 = -2e-1l'(1
problem

problem

and we assurne yp = A on. [O,7rJ. Substituting

we find q = -2 and C2 =

of y at x

+ 2e1l' = -c3e1l'

initial-value

on [O,7r/2J.

+ ~sin 2x + 1sin x,
~ cos 2x + ~sin2x,

e1l'(-2cos37r
or 2

cos 2x

Now, continuity

+ 1sin x

'3 cos 2' = - 2C3 sin 7r + 2C4 COS7r

we find A = 2. Thus, y = e" (Cl cos 3x

equation

= eX(c3

2x

of y' at x = 7r/2 implies

Then C4= ~ and the solution of the initial-value

y(x)
40. We have Yc

+ ~sin

'65 sin. 7r + '31 sin. 2'7r = C3cos 7r + C4sin. 7r

- 2 sin 7r + '3 cos 7r +


or

= cos 2x

of y at x = 7r/2 implies
COS7r +

or -1 +

'3 + 2C2 = 2.
C2 = ~. Thus y

+ e")

+ C4) sin 37rJ

we have C4 = ~e-1l'(1

+ e"}.

is

y(x) =

ex (- 2 cos 3x
{ (1

+ sin 3x),

+ e1l')eX-1l' (-2

cos 3x

130

+ ~ sin 3x),

z > x,

The solution of the

Exercises

41. (a) From Yp = Ax

we find Y~ = Akx

and Y~ = Ak2x.

Substituting

4.4

into the differential

equation we get
aAk2ekx
so (ak2

+ bk + c)A =

(b) From Yp = Axx

+ bAkx + cAx

= (ak2

+ bk + c)A

+ bm + e = O, A

1. Since k is not a root of am2

we find Y~ = Akxx

= ekx,

= lj(ak2

+ Ax and Y~ = Ak2xx + 2Akx.

+ bk + e).

Substituting into

the differential equation we get


aAk2xx

+ 2aAkekx + bAkxx + bAx + cAxekx


= (ak2 + bk + c)Ax

+ (2ak + b)Aekx

+ (2ak + b)A

= (O)AxX

= (2ak + b)AX = x

+ bk + e = O because k is a root of the auxiliary equation. Now, the roots of the


auxiliary equation are -bj2a Jb2 - 4ac, and since k is a root of multiplicity one, k =J:. -bj2a
and 2ak + b =J:. o. Thus (2ak + b)A = 1 and A = lj(2ak + b).
where ak2

(e) If k is a root of multiplicity two, then, as we saw in part (b), k = -bj2a


From Yp = Ax2x we find Y~ == Akx2x + 2Axx
Substituting into the differential equation, we get
aAk2x2ekx

and Y~ = Ak2x2x

and 2ak

+ 4Akxx

+ b = O.
= 2Ax.

+ 4aAkxx + 2aAx + bAkx2ekx + 2bAxekx + cAx2x


= (ak2 + bk + c)Ax2ekx + 2(2ak + b)Axekx + 2aAx
= (0)Ax2ekx

+ 2(0)Axekx + 2aAekx = 2aAekx =

ekx.

Since the differential equation is second-order, a =J:. O and A = lj(2a).


42. Using the double-angle formula for the cosine, we have
sin x cos 2x = sinx(cos2 x - sin2 x) = sinx(l-

2sin2 x) = sinx - 2sin3 x.

Since sin x is a solution of the related homogeneous differential equation we look for a particular
solution of the form YP = Ax sin x

+ Ex

cos x

+ e sin3 x.

Substituting into the differential equation

we obtain
2a cos x

+ (6c -

2b) sinx - 8csin3 x = sinx - 2sin3 x.

Equating coefficients we find a = O, e = ~, and b = ~. Thus, a particular solution is


1

YP = -x cos x

43. (a)

f(t) = et sin t. We see that Yp --*

(b ) f(t)

= e-t. We see that Yp --*

00

00

as t

as t

1 . 3
+ -4
sm x.

--* 00

--* 00

and Yp --* O as t

and Yp --*

131

00

as t

--*

--*

-oo.

-oo.

Exercises 4.4
(e) f(t)

= sin 2t. We see that Yp is sinusoidal.

(d) f(t)

= 1. We see that Yp is constant and simply translates Yc vertically.

= e2x(Cl cos 2x + C2 sin 2x). We assume a particular solution of


= (Ax3 + Bx2 + Cx)e2x cos 2x + (Dx3 + Ex2 + F)e2x sin 2x. Substituting into the

44. The complementary function is Yc


the form Yp

differential equation and using a CAS to simplify yields

[12Dx2 + (6A + SE)x + (2B + 4F)Je2x cos 2x


+ [-12Ax2 + (-SB + 6D)x + (-4C + 2E)]e2x sin 2x
= (2x2 - 3x)e2x cos 2x

+ (lOx2 -

X -

1)e2Xsin2x.

This gives the system of equations

12D = 2,

6A+SE

-12A = 10,

-SB+6D=-1,

i,

2B

= -3,

from which we find A = -~, B =


C = ~, D
solution of the differential equation is

+ 4F

= 0,

-4C+2E=-1,

= ~, E =

i, and

F =

-k.

Thus, a particular

YP = ( _~x3 + ~x2 + ~x )e2X cos 2x + (~x3 + ~x2 - ~x )e2X sin 2x.


45. The complementary function is Yc =

Cl

cos X

+ C2 sin x + C3X

COS X

+ C4X

sin x. We assume a particular

solution of the form Yp = Ax2 cos X + Bx3 sin x. Substituting into the differential equation and using
a CAS to simplify yields
(-Sa
This implies -Sa

+ 24b) cos x + 3bx sin x = 2 cos x - 3x sin x.

+ 24b = 2 and -24b =

-3. Thus b = ~, a = ~, and Yp = ~x2 cos X

Exercises 4.5
1.

(9D2 - 4)y = (3D - 2)(3D + 2)y = sinx

2. (D2 - 5)y = (D - V5)(D + V5)y = x2 - 2x


3. (D2 - 4D - 12)y = (D - 6)(D + 2)y = x - 6
4. (2D2 - 3D - 2)y

= (2D + l)(D

- 2)y

+ ~x3 sin x.

=1

5. (D3 + lOD2 + 25D)y = D(D + 5)2y = e"


6. (D3 + 4D)y = D(D2 + 4)y = eX cos 2x
7. (D3 + 2D2 - 13D + lO)y = (D - l)(D - 2)(D + 5)y = xe":"
8. (D3 + 4D2 + 3D)y = D(D + l)(D + 3)y = x2 COS X
132

3x

Exercises 4.5

+ 8D)y

9. (D4

10. (D4 - 8D2


11. D4y

+ 2)(D2

= D(D

+ 16)y =

(D - 2)2(D

+ 4)y

=4

+ 2)2y =

(x3 - 2x)X

D4(lOx3 - 2x) = D3(30x2 - 2) = D2(60x) = D(60) = O

= 8Dex/2

12. (2D - l)y = (2D - 1)4ex/2


13. (D-2)(D+5)(e2x+3e-SX)

+ 64)(2cos8x

14. (D2

- 2D

= 4ex/2

- 4ex/2

- 4ex/2 = O

= (D-2)(2e2x-15e-sx+5e2x+15e-SX)

- 5sin8x) = D(-16sin8x
= -128 cos 8x

- 40 cos8x)

= (D-2)7e2x

+ 64(2cos8x

+ 320 sin 8x + 128 cos 8x

= 14e2x-14e2x

- 5sin8x)
- 320 sin 8x = O

15. D4 because of x3

16. DS because of x4

17. D(D - 2) because of 1 and e2x

18. D2(D - 6)2 because of x and xe6x

19. D2

+ 4 because of cos 2x

21. D3(D2
22. D2(D2
23. (D

20. D(D2

+ 16) because o x2 and sin 4x


+ 1)(D2 + 25) because of x, sin x,

+ l)(D

+ 1) because of 1 and sinx

and cos 5x

- 1)3 because of e-x and x2ex

24. D(D - l)(D - 2) because of 1, eX, and e2x


25. D(D2 - 2D

+ 5)

because of 1 and eX cos 2x

+ 2D + 2) (D2

26. (D2

- 4D

+ 5) because

of e-x sin x and e2x cos x

27. 1, x, x2, x3, x4


28. D2

+ 4D = D(D + 4); 1, e-4x

29. e6x, e-3x/2

30. D2 - 9D - 36 = (D - 12)(D

+ 3);

e12x, e-3x

31. cos y'5 x, sin y'5 x

+ 10 = D2 - 2(3)D + (32 + 12); e3x cosx,


10D2 + 25D = D(D - 5)2; 1, x, xx

32. D2 - 6D
33. D3 -

e3x sinx

35. Applying D to the differential equation we obtain

D(D2

9)y

= O.

Then
y = ..._,___..,
ce3x + C2e -3x +
Yc

133

C3

=O

Exercises 4.5
and YP = A. Substituting Yp into the differential equation yields -9A = 54 or A = -6. The general
solution is

36. Applying D to the differential equation we obtain

D(2D2 -7D + 5)y = O.


Then
Yc

and YP

A. Substituting Yp into the differential equation yields 5A

-29 or A

-29/5.

The

general solution is

37. Applying D to the differential equation we obtain

D(D2 + D)y

= D2(D + l)y = O.

Then
Yc

and yp = Ax. Substituting YP into the differential equation yields A = 3. The general solution is
y

= Cl + c2e-3x + 3x.

38. Applying D to the dfferential equation we obtain

Then
Yc

and Yp = Ax. Substituting YP into the differential equation yields A

39. Applying D2 to the differential equation we obtain

Then
Yc

134

10. The general solution is

Exercises 4.5
and Yp = Ax + B. Substituting Yp into the differential equation yields 4Ax
Equating coefficients gives
4A

+ (4A + 4B)

= 2x

+ 6.

=2

4A +4B = 6.
Then A = 1/2, B = 1, and the general solution is

40. Applying D2 to the differential equation we obtain

Then
Yc

and Yp = Ax2

+ Bx.

Substituting Yp into the differential equation yields 6Ax

+ (2A + 3B) = 4x -

Equating coefficients gives


6A=4
2A+3B

= -5.

Then A = 2/3, B = -19/9, and the general solution is


Y = q + C2e

-3x

+ 'Jx

19

2
-

gx.

41. Applying D3 to the differential equation we obtain

Then
Y

= C! + C2X + C3e-x + C4X4 + CSX3 + C6X 2


Yc

and Yp = Ax4

+ Bx3 + Cx2.

Substituting Yp into the differential equation yields


12Ax2

+ (24A + 6B)x + (6B + 2C)

Equating coefficients gives


12A = 8
24A+6B
6B

+ 2C
135

= O.

= 8x2.

5.

Exercises 4.5
Then A = 2/3, B

= -8/3, C = 8, and the general solution is


y = ci

+ C2X + c3e-x + ~x4


3

- ~x3
3

+ 8x2.

42. Applying D4 to the differential equation we obtain

Then
!jc

and YP = Ax3

+ Bx2 + Cx + D.
Ax3

+ (B

- 6A)x2

Substituting Yp nto the differential equation yields

+ (6A

- 4B

+ C)x + (2B

- 2C

+ D) = x3 + 4x.

Equatng coefficients gives


A=1
B - 6A
6A - 4B+C
2B - 2C
Then A

= 1, B

+D

=O
=4
= O.

= 6, C = 22, D = 32 , and the general solution is

= clex + C2xex + x3 + 6x2 + 22x + 32.

43. Applying D - 4 to the differential equation we obtain


(D - 4)(D2 - D - 12)y = (D - 4)2(D

+ 3)y

= O.

Then
!jc

and YP = Axe4x.

Substituting

Yp into the differential equation yields 7Ax

coefficients gives A = 1/7. The general solution is

44. Applying D - 6 to the differential equation we obtain


(D - 6)(D2

+ 2D + 2)y

Then
!jc

136

= O.

=:

x.

Equating

Exercises 4.5
and Yp = Ax.

Substituting

coefficients gives A

Yp into the differential equation yields 50Ax

= 5x.

Equating

= 1/10. The general solution is


y = e -x

(Cl COS X

+ C2 sin x) + lOe 6x.

45. Applying D(D - 1) to the differential equation we obtain


D(D - 1)(D2 - 2D - 3)y

= D(D - l)(D + l)(D - 3)y =

O.

Then
Yc

and yp = Aex

+ B.

Substituting Yp into the differential equation yields -4Aex

Equating coefficients gives A

= -1

y
46. Applying D2(D

+ 2) to

= 3. The general solution


= cle3x + C2e-x - eX + 3.

and B

is

the differential equation we obtain

D2(D

+ 2)(D2 + 6D + 8)y

+ 2)2(D + 4)y

= D2(D

= O.

Then
Yc

and YP = Axe-2x

+ Bx + C.

Substituting Yp into the differential equation yields

2Ae-2x

+ 8Bx + (6B + 8C)

= 3e-2x

+ 2x.

Equating coefficients gives


2A= 3
8B

6B+8C=
Then A = 3/2, B = 1/4, C

O.

= -3/16 , and the general solution is


3 -2x
y = cre- 2x + C2e- 4x + -2xe

47. Applying D2

+ 1 to

+ -x1 - -.3
4

16

the differential equation we obtain


(D2

+ 1)(D2 + 25)y =

O.

Then
y

= Cl cos 5x + C2 sin 5x + C3 cos x + C4 sin x


Yc

137

- 3B = 4ex - 9.

Exercises

4.5

and Yp = A cos x. + B sin x. Substituting Yp into the differential equation yields


24A cosx

+ 24B sinx = 6 sin x.

Equating coefficients gives A = O and B = 1/4. The general solution is


y = Cl

+ 1) to

48. Applying D(D2

COS 5x

+ C2 sin 5x + 4 sin x.

the differential equation we obtain

D(D2 + 1)(D2 + 4)y = O.


Then
y = Cl

COS 2x

+ C2 sin 2x + C3 cos X + C4 sin x + Cs


Yc

+ B sin x + C.

and Yp = A ~os x

Substituting Yp into the differential equation yields

3A cos x

= 4/3,

Equating coefficients gives A


y

+ 3B sin x + 4C = 4 cos x + 3 sin x

= -2.

B = 1, and C

- 8.

The general solution is

= Cl COS 2x + C2 sin 2x + ~ cos x + sin x

- 2.

49. Applying (D - 4)2 to the differential equation we obtain

Then
Yc

and Yp = Axx

+ Bx.

Substituting Yp into the differential equation yields


49Axx

+ (14A + 49B)e4x

= -xx.

Equating coefficients gives


49A = -1
14A
Then A = -1/49,

+ 49B

= O.

B = 2/343, and the general solution is


y

= cle-3x + C2xe-3x

-xx
49

50. Applying D2(D - 1)2 to the differential equation we obtain

138

+ _e4x
343

Exercises
Then
Yc

and Yp = AxeX

+ Be + Cx + D.

Substituting Yp into the differentialequation yields

+ (5A

-6Axex

- 6B)eX

10Cx

+ (3C

-6A

=1

- 10D)

= xe" + x.

Equating coefficients gives

5A - 6B

=1

-10C

3C - lOD = O.
Then A

= -1/6, B = -5/36, C = -1/10, D


= ce2x

+C2e

-Sx

= -3/100, and the general solution is

1
-"6xe

5
-36e

1
3
-10x-100'

51. Applying D(D - 1)3 to the differential equation we obtain


D(D - 1)3(D2 - l)y

= D(D

- 1)4(D

+ l)y = O.

Then
y = cl e x

+ C2e-x + C3X3e x + C4X2e x + csxe x + C6

'----v---'
Yc

and yp = Ax3ex

+ Bx2ex + Cxe" + D.
6Ax2

Substituting yp into the differential equation yields

+ (6A + 4B)xe

+ (2B + 2C)e

+ 5.

D = x2ex

Equating coefficients gives


6A = 1
6A+ 4B
2B

=O

+ 2C = O
-D= 5.

Then A = 1/6, B

= -1/4, C = 1/4, D = -5, and the general solution is


y = cj e

52. Applying (D

+ 1)3 to the

1 3 x
+ C2e-x + "6x
e -

1 2 x
4x e

1 x
+ 4xe

differential equation we obtain


(D

+ 1)3(D2 + 2D + l)y
139

= (D

+ l)Sy

= O.

4.5

Exercises

4.5

Then
Yc

+ Bx3e-x + Cx2e-x.

and Yp = Ax4e-x

Substituting Yp into the differential equation yields

+ 6Bxe-X + 2Ce-X = x2e-x.

12Ax2e-x
Equating coefficients gives A

1~'

B = O, and C

+ 2 to

2D

general solution is

1
+ c~xe-x
+ _x
4e-x
~
12

Y = c 1e-x

53. Applying D2

= O. The

the differential equation we obtain


(D2

2D

+ 2)(D2

+ 5)y =

2D

O.

Then
y

= eX(C1cos 2x + C2 sin 2x) + eX( C3 cos x + C4 sin x)


Yc

and Yp = Aex cos x

+ Be"

sin x. Substituting Yp into the differential equation yields


3Aex cos x

+ 3Bex

sin x = eX sin x.

Equating coefficients gives A = O and B = 1/3. The general solution is


y = eX(c1 cos 2x
54. Applying D2

+ 10 to

2D
(D2

2D

+ C2 sin 2x) + ~ex

sin x.

the differential equation we obtain

+ 10)

(D2

+D +

y = (D2

+ 10)

2D

(D

~f
y

= O.

Then
Yc

and Yp = Aex cos 3x

+ Be"

sin 3x. Substituting Yp into the differential equation yields

(9B - 27A/4)eX cos 3x - (9A

+ 27B /4)eX

sin 3x

= _ex cos 3x + eX sin 3x.

Equating coefficients gives


27
--A+9B=-1
4

-9A - 27 B
4

1.

Then A = -4/225, B = -28/225, and the general solution is


Y

= c e-x/2 + c xe-x/2
1

4
_ex
225

140

28
cos 3x - _ex
225

sin 3x.

Exercises 4.5
55. Applying D2

+ 25 to

the differential equation we obtain


(D2 + 25)(D2

+ 25) = (D2 + 25)2 = O.

Then
y = Cl cos 5x

+ C2 sin 5x + C3X

cos 5x

+ C4X

COS 5x

Yc

and Yp = Ax cos 5x

+ Bx

sin 5x. Substituting Yp into the differential equation yields


10B cos 5x - lOA sin 5x = 20 sin 5x.

Equating coefficients gives A

= -2 and B = O. The general solution is


=

y
56. Applying D2

+ 1 to

Cl COS 5x

+ C2 sin 5x -

2x cos 5x.

the differential equation we obtain


(D2

+ 1)(D2 + 1) = (D2 + 1)2 = O.

Then
y = Cl

COS X

+ C2 sin x + C3X

cos x

+ C4X

COS X

Yc

and Yp = Ax cos x

+ Bx

sin x. Substituting Yp into the differential equation yields


2B cos x - 2A sin x = 4 cos x - sin x.

Equating coefficients gives A = 1/2 and B = 2. The general solution is


y = Cl
57. Applying (D2

+ 1)2 to

COS X

. x + '2x
1 cos x
+ C2 sin

- 2'
x sm x.

the differential equation we obtain


(D2

+ 1)2(D2 + D + 1)

= O.

Then
y = e.-x/2

[Cl

/3 x
cos 2

/3 x ] + C3 cos x + C4 sin. x + C5X


+ C2 Slll. 2

cos x

.x
+ C6X sin

Yc

and Yp = A cos x+B sin z-l-Cz cos x+ Dx sinx. Substituting YPinto the differential equation yields
(B

+ C + 2D) cos x + Dx

cos x

+ (-A -

2C

+ D) sin x -

Equating coefficients gives

=O

B+C+2D

D=O

-A -2C+D

=O

-C = 1.
141

Cx sin x = x sin x.

Exercises 4.5
Then A

= 2, B = 1, C = -1, and D = 0, and the general solution is


y = e -xj2

58. Writing cos2x

[Cl

=!(1 +cos2x)

/3

cos -

. /3] + ')~

+ C2 sm -

+ sm x

cos x

- x cos x.

and applying D(D2 +4) to the differential equation we obtain


D(D2

+ 4)(D2 + 4)

COS 2x

+ C2 sin 2x + C3Xcos 2x + qx

= D(D2

+ 4)2

= O.

Then
y = Cl

sin 2x

+ C5

Yc

and YP = Ax cos 2x

+ Bx

+ C.

sin 2x

-4A sin 2x

Substituting Yp into the differential cquation yields

+ 4B

cos 2x

+ 4C = 2" + 2" cos 2x.

= 0, B = 1/8, and C = 1/8. The general solution is

Equating coefficients gives A

Cl COS 2x

+ C2 sin 2x + "8x

+ "8'

sin 2x

59. Applying D3 to the differential equation we obtain


D3(D3

+ 8D2)

= D5(D

+ 8)

= O.

Then
Yc

and Yp = Ax2

+ Bx3 + Cx4.
16A

Substituting Yp into the differential equation yields

+ 6B + (48B + 24C)x + 96Cx2

=2

+ 9x

- 6x2.

Equating coefficients gives

+ 6B

16A

=2

+ 24C =

48B

96C = -6.
Then A

= 11/256, B = 7/32, and C = -1/16, and the general solution is


Y=

11 ')

+ C2X + C3e- + 256x~ + 32x

+ 1) to the
1)2(D + 1)(D3

60. Applying D(D - 1)2(D


D(D -

Sx

Cl

1 4
- 16x .

differential equation we obtain


- D2

+D

- 1) = D(D - 1)3(D

+ 1)(D2 + 1)

Then
y=cle

+C2COsx+c3smx

+C4+cSe

Yc

142

-x

+C6xe

2 x

+C7X e

= O.

Exercises 4.5
and

YP

= A + Be-X + Cxe" + Dx2ex. Substituting

Yp

into the differential equation yields

4DxeX + (2C + 4D)eX - 4Be-X - A = xe" - e-x + 7.


Equating coefficients gives

4D = 1
2C

+ 4D = O
-4B = -1
-A=

Then A = -7, B

1/4, C

y=qe

= -1/2,
x

and D

= 1/4,

7.

and the general solution is

.
+c2COsx+c3smx-7+e

1 -x

-"2xe

+x

e.

61. Applying D2 (D - 1) to the differential equation we obtain

Then
Yc

and

YP = A

+ Bx + Cx3ex.

Substituting

YP

into the differential equation yields

(-A + 3B) - Ex + 6Cex = 16 - x + e",


Equating coefficients gives

-A+3B=16
-B=

-1

6C = 1.
Then A = -13, B = 1, and

e = 1/6,

and the general solution is

62. Writing (eX + e-x)2 = 2 + e2x + e-2x and applying D(D - 2)(D

+ 2) to

the differential equation we

obtain

D(D - 2)(D + 2)(2D3

3D2 - 3D + 2) = D(D - 2)2(D + 2)(D + 1)(2D - 1) = O.

Then
Yc

143

Exercises 4.5
and YP = A

+ Bxe2x + Ce-2x.

Substituting.p,

2A

+ 9Be2x

into the differential equation yields

= 2 + e2x + e-2x.

- 20Ce-2x

Equating coefficients gives A = 1, B = 1/9, and C = -1/20. The general solution is

63. Applying D(D - 1) to the differential equation we obtain

Then
Yc

and YP = 'Ax2

+ Bx2ex.

Substituting Yp into the differential equation yields 2A

Equating coefficients gives A

= 1/2 and

= 1/2. The

+ 2Bex

= 1 + eX.

general solution is

64. Applying D3(D - 2) to the differential equation we obtain

Then
Yc

and YP = Ax2

+ Bx3 + Cx4 + Dxe2x.


(-8A

+ 24C)

Substituting YP into the differential equation yields


- 24Bx - 48Cx2

+ 16De2x

= 5x2 - e2x.

Equating coefficients gives


-8A

+ 24C

=O

-24B

=O

-48C

=5

16D = -1.
Then A

= -5/16, B = O,C = -5/48, and D = -1/16, and the general solution is


2x

y = q + C2X + C3e

-2x

+ C4e

144

5
- 16x

5
- 48x

1 2x
- 16xe

Exercises 4.5

65. The complementary function is Yc

+ C2e-8x.

cle8x

Using D to annihilate 16 we find yp

A.

Substituting Yp into the differential equation we obtain -64A = 16. Thus A = -1/4 and
y = Cle

8x

+ C2e-8x

1
- 4

The initial conditions imply


5

+ C2 = "4

ci

8Cl - 8C2 = O.
Thus e = C2 = 5/8 and
5
y = _e8x
8

66. The complementary function is Yc = Cl

5
+ _e-8x
8

+ C2e-x.

1
4

Using D2 to annihilate x we find yp = Ax

Substituting Yp into the differential equation we obtain (A

+ 2B) + 2Bx

+ Bx2.

= x. Thus A = -1 and

B = 1/2, and
1
= Cl + C2e-x - x + -x
2
y' = -C2e-x - 1 + x.
Y

The initial conditions imply


e+c2=1

= 1.

-C2
Thus

ci

= 2 and C2 = -1, and


Y =2- e

67. The complementary function is Yc

-x

1 2
x + "2x
.

= Cl +C2e5x. Using D2 to annihilate

Substituting YP into the differential equation we obtain (-5A+2B)-10Bx


and

B= -1/10, and
Y

= Cl + C2e5x + -x9 - -x1

1
= 5c2e5x + -9 - -x.

25

25

The initial conditions imply


Cl

+ C2 = O
41
5C2 = 125

145

10

x-2

we find yp
= -2+x.

= Ax+Bx2.

Thus A = 9/25

Exercises 4.5
Thus CI = -41/125

and C2= 41/125, and


41

41

5x

= -125 + 125e

9
1 2
+ 25x - 10x .

68. The complementary function is Yc = clex + C2e-6x.


Yp = Ae2x. Substituting

Using D -

:2 to annihilate

lOe2x we find

Yp into the differential equation we obtain 8Ae2x = lOe2x. Thus A = 5/4

and

The initial conditions imply


c
CI -

+ C2= -4
6C2

-"2.

Thus CI = -3/7 and C2= 5/28, and


3
y = __ex
7

5
+ _e-6x
28

5
+ _e2x
4

69. The complementary function is Yc = CICOSX+ C2sinx.


8cos2x

- 4sinx

we find Yp = Axcosx

+ Bxsinx

Using (D2 + 1)(D2 + 4) to annihilate

+ Ccos2x

+ Dsin2x.

differential equation we obtain 2B cos x - 3C cos 2x - 2A sin x - 3D sin 2x


A

= 2, B = O, C

= -8/3,
y

and D

Substituting Yp into the

= 8 cos 2x

- 4 sin x. Thus

= O, and

= CIcos x + C2sin x +

y ,. =-csmx+c2coSX+

2x cos x - ~ cos 2x

2 cosx-

2xsmx+3sm
16 . 2x.

The initial conditions imply


8
3

c2+-=-1

-CI - 7f = O.
Thus Cl =

+t:

and C2= -11/3, and


y

-7fCOSX- -11.smx + 2x cosx - -8 cos 2x.


3
3

70. The complementary function is Yc = Cl + C2ex + C3xex. Using D(D - 1)2 to annihilate
xe" + 5 we find Yp = Ax + Bx2ex + Cx3ex.

Substituting Yp into the differential equation

146

Exercises 4.5
we obtain A + (2B + 6C)eX + 6Cxex = xeX + 5. Thus A = 5, B = -1/2,
y

= Cl +

and C = 1/6, and

1
1
c2ex + C3xex + 5x - _x2ex + _x3ex

y' = C2ex + c3(xeX + eX) + 5 - xe" + ~x3eX


1
1
y" = C2ex + c3(xeX + 2eX) - eX - xe" + _x2ex + _x3ex.

The initial conditions imply

C2+ C3+ 5 = 2

Thus q

= 8, C2=

-6, and C3= 3, and

71. The complementary function is Yc = e2x(cl cos 2x + C2sin 2x). Using D4 to annihilate x3 we

find yp = A + Ex + Cx2 + Dx3. Substituting Yp into the differential equation we obtain


(8A - 4B + 2C) + (8B - 8C + 6D)x + (8C - 12D)x2 + 8Dx3 = x3. Thus A = O, B = 3/32,
C = 3/16, and D = 1/8, and
y=e2x(clcOs2x+c2sin2x)+

3
32x+

3
1
16x2+gX3

3
3
3
y' = e2x [cl(2cos2x - 2sin2x) + c2(2cos2x + 2sin2x)] + 32 + gX + gX2.
The initial conditions imply

Thus Cl = 2, C2= -3/64,

and
y

72. The complementary

3
64

3
3
+ _x2
32
16

= e2X(2cos2x - - sin2x) + -x

function is Yc = Cl + C2X + C3x2 + c4ex.

1
+ _x3.
8
Using D2(D - 1) to annihilate

x + eX we find yp = Ax3 + Bx4 + Cxe". Substituting yp into the differential equation we obtain

147

Exercses 4.5
(-6A + 24B) - 24Bx + CeX = x + eX. Thus A = -1/6, B = -1/24, and C = 1, and
y = Cl

+ C2X + c3X2 + c4e - -x1314- -x


X

I
x
12
13
y = C2 + 2C3X + C4e - 2'x - '6x

12
y "2= c3 + C4ex - x - 2'x

24.

+ xe x

+ ex + xe x

+ 2xe + xe x .

The initial conditions imply

2
Thus

Cl

= 2, C2 = 1, C3 = O, and

C4

+ C4 = O.

-2, and
x

y = 2 + x - 2e -

'61x 3 -

1 4
24 x

+ xe x .

73. To see in this case that the factors of L do not commute consider the operators

and (D

+ 4)(xD -

(xD - 1)(D + 4)

1). Applying the operators to the function x we find

(xD - 1)(D + 4)x = (xD2 + 4xD - D - 4)x

+ 4xDx - Dx - 4x

= xD2x
= x(O)

+ 4x(1) -

1-

4x = -1

and

(D + 4)(xD - 1)x = (D + 4)(xDx - x)


=

(D + 4)(x 1 - x) = O.

Thus, the operators are not the same,

Exercises 4.6

The particular solution, Yp = UlYl + U2Y2, in the following problems can take on a variety of forms,
especially where trigonometric functions are involved. The validity of a particular form can best be
checked by substituting it back into the differential equation.

148

Exercses 4.6
1. The auxiliary equation is m2

+ 1 = O,

so Yc

W=
Identifying

f (x)

= Cl COS X + C2 sin x
cosx

sinx
cosx

- sinx

and

I =1.

= sec x we obtain
I

Ul

=-

U~

sin x sec x

=-

cos x sec x

tanx

1.

Then

Ul

= ln I cos z],

U2

= x, and
y = Cl

2. The auxiliary equation is m2

COS X

+ C2 sin x + cos x In I cos z] + x sin x.

+ 1 = O, so Yc = Cl COS X + C2 sin x
W=

Identifying

f (x)

xl

cos
I sinx

sin
cosx

and

l.

= tan x we obtain
l.

ul
I

U2

= -smxtanx

cos2 X _. 1
= cosx - secx
cosx

= smx.

Then u 1 = sin x - ln I sec x


y

+ tan z], U2 = - cos x, and


= el cos x + C2 sin x + cos x (sin x - ln I sec :r + tan

3. The auxiliary equation is m2

+ 1 = O, so Yc =
W=

Identifying

f (x)

= sin x

Cl COS X

+ e2

xl) - cos x sin x.

sin x and

cosx

sinx =1.
cos z

I -sinx

we obtain
U~
I

U2

Then

= - sin2 x
.

= cos x sin x.

1
1
u 1 = - sin 2x - - x
4
2
U2

1.
1
sm x cos x - - x
2
2

=-

-2"1 cos2 x.

and
Y = el cos x
= el cos x

.
1.
2
+ C2 sm
x + - sm x cos x
222

+ e2 sin

1
x - 2"xcos x.

149

1
12.
- -x cos x - - cos x sm x

4.6

Exercises

4. The auxiliary equation is m2

1 = O, so Yc

cosx

w=
Identifying

f (x)

= sec x tan

= Cl COSX + C2sin x

= - tan2

= - sinx(secxtanx)

U2 = cosx(secxtanx)
Ul

=x

- tanx,

U2

= -In I cos z],

y = Cl COSX

5. The auxiliary equation is m2

+ C2sin x +

1 = O, so Yc

= tanx.

x cos x - sin x - sin x In I cos xl


x cos x - sin x In I cos x l

Cl COSX + C2sin x

cosx

w-

sinx
cosx

I -SInX

I =l.

=, -

SIn X COS x

U2 = cos3 x = cos x (1 Ul =

and

= cos2 x we obtain
Ul

Then

x = 1 - sec2 x

and

= Cl cos X + C3sin x +

Identifying f(x)

I =l.

x we obtain

U~

Then

sinx
cosx

I -sinx

and

sin 2 x) .

icos3 x, U2 = sinx - isin3 x, and


y = Cl cos x

14

.2

+ C2SIn X + '3 cos x + SIn X -

Cl COSX + C2sin x + '3

(cos2 X

1.4

'3 SIn X

sin2 x) (cos2 X

1. 2
+ '3 SIn X.

= Cl COSX + C2sin x + '3 cos2 X + '3 sin x

ClCOSX+C2SInX+'3

6. The auxiliary equation is m2

1 = O, so Yc = Cl COSX

w=
Identifying f(x)

= sec2

cos x

+ C2 sin

sinx I =l.
cos x

I -sinx

x we obtain
I

u ---1-

sinx
cos? x

U2 = sec x.
150

x and

sin2 x)

sin2 x

Exercises 4.6
Then

U =

1
---cosx
= -secx

U2 = In I sec x

+ tan

z]

and
y = c cos x

cos x

7. The auxiliary equation is m2

Identifying f(x)

+ sin x In I sec x + tan

+ C2 sin x

- cos x sec x

+ C2 sin x

- 1 + sin x In I sec x

1 = 0, so Yc

= cex + c2e-x

+ tan x l.

and

= coshx = !(e-X + eX) we obtain


,

1 2x

1
+-4

u = -e

,
1 1 2x
u2 = -- --e

Then
U

1 -2x
1
= --e
+-x

1 2x 1
U2 = --e --x

and
y = cie
= c3e

+ c2e -x

1 -x
- Se

1 (Xe
+ C4e-x + x

1
+ xe

1 x
1 -x
- Se - xe

- e -X)

= C3ex + C4e-x + 2"xsinhx.


8. The auxiIiary equation is m2

1 = 0, so Yc

= cex + C2e-x and

e-x
= -2.
-e-x
Identifying

f (x)

= sinh 2x we obtain
1

u~ = __e-3x

1
+ _ex

,
1 -x
1 3x
u2 = -e
--e

151

z]

Exercises 4.6
Then
Ul

1 -3x
12

1 x
+-e

= -e

1 -x
1 3x
--e
--e

U2 =

12

and

= cex + C2e-x + 3" sinh2x.


9. The auxiliary equation is m2

W =

Identifying f(x)

+ c2e-2x

4 = O, so Yc = ce2x

I 2e2x
e2x

I = -4.

e-2x
_2e-2x

= e2x/x we obtain u = 1/4x and

and

u2 = _x /4x.

Then

Ul

= -lnlxl,
4

U2 = --

X -dt

4 xo t

and
xQ

10. The auxiliary equation is m2

Identifying f(x)

+ c2e-3x

9 = O, so Yc = cle3x

I ;::

Ul

=-

I = -6.

_~~~3X

= 9x/e3x we obtain u = ~xe-6x and


1 -6x

24 e

U2 = -X

U2 =
1

- xe

and

152

and

-~x.

-6x

Then

> O.

Exercises

11. The auxiliary equation is m2

+ 3m + 2 =

W
Identifying f(x)

= 1/(1

+ eX)

(m

+ l)(m + 2) =

~:~x

_~~~2X

0, so Yc

= cle-x + C2e-2x and

= _e-3x.

we obtain
eX
u, - __
1 - 1 + eX
e2x

Then

Ul

= ln(l + eX),
y

U2 = ln(l

eX

= - 1 + eX = 1 + eX

u2

+ eX)

- e

- eX, and

= Cle-x + C2e-2x + e-x ln(l + eX) + e-2x ln(l + eX) - e-x


= C3e-x + C2e-2x + (1 + e-X)e-X ln(l + eX).

12. The auxiliary equation is m2 - 2m

+1=

w=
Identifying f(x)

= eX /

(1 + x2)

we obtain
,

u' 2 -

Ul

= -!In (1 + x2),
y

xexex
e2x (1+ x2)

+ 3m + 2 =
=

(m

z.

+ l)(m + 2) = 0, so Yc = cle-x + C2e-2x

1_e-x
e-x

e-2x
_2e-2x

1--

e-3x

= sin e" we obtain

,
u2
Ul = -

eXeX
1
-_e2x (1 + x2) - 1 + x2'

= clex + c2xex - ~ex In (1 + x2) + xe" tan-1

vV

Then

+ x2

= -1

U2 = tan-1 x, and

13. The auxiliary equation is m2

Identifying f(x)

= 0, so Yc = qeX + C2xex and

I eX

ul = -

Then

(m - 1)2
eX

e-x sin e"

= _e-3x

= _e2x sin e"

cos e", U2 = e" cosx - sin e", and


y = cle-x

+ c2e-2x

- e-x cos e"

153

+ e-x

cos e" - e-2x sin e"

and

4.6

Exercises 4.6

+1=

14. The auxiliary equation is m2 - 2m

(m - 1)2

= O, so Yc = clet + c2tet and

= jet

et
Identifying f(t)

= tan-l t we obtain
I

=-

Ul

tetettan-lt
2t
e

Then
Ul

= -ttan

1 + t2

= --2-tan-

-1

t
t +"2

U2 = t tan -1t - ~ In ( 1 + t2)


and

= clet + C3tet + ~et [(t2 15. The auxiliary equation is m2

(1 + t2)].

t -In

+ 2m + 1 = (m + 1)2 = O, so Yc = qe-t + c2te-t and

w=
Identifying f(t) = e-tInt

1) tan-l
-t

_e-t

t e -t
-te-t

+ e-t

I -e -2t
-

we obtain

Then
ul

= --t1 2 In t + -t1 2
2

U2 = z ln t - t
and
Y = qe-t
= Clet

1
1
2
+ c2te-t - _t2e-t
In t + _t2e-t + t2e-t In t - t e:'
2

+ c2te-t + ~t2e-t

16. The auxiliary equation is 2m2

In t - ~t2e-t
4'

+ 2m + 1 = O, so Yc = e-x/2(
154

cl

cos x /2

+ C2 sin x /2) and

Exercises 4.6

w=

e-x/2 cos:!:
2

I _le-x/2

cos :!:- le-x/2

222

Identifying f(x)

=-

u2 = -

e-x/2 sin(x/2)2.jX
/2
e-x

= -4e

e-x/2 cos(x/2)2JX
/2
e-x

= 4e

Then
Ul

= -4

U2 = 4

y = e-x / 2 ( Cl cos -x
2

= jexsecx

+ 6 = O, so Yc = e" (Cl

6m

I e" cos x

+ C2 sin x)

e X smx

eX cos x

- e" sin x

= jx,

y = clex cosx

(eX sinx)(eX secx)/3


e2x

=-

u2 =

(eX cosx)(eX secx)/3


e2x

and

2x

+ e" sin x = e

= - -3 tan x
1

:3'

and
l'

+ C2ex cosx + :31n(cosx)eX

18. The auxiliary equation is 4m2 - 4m

+ 1 = (2m

w-= tex/2~

cos X

sin -x x et/2Jt cos -t dt.


2 xo
2

+ 4e-x/2

we obtain
I

U2

et/2Jt cos - dt
xo
2

eX cos x

ul

= j ln(cosx),

x/2 r:
x
v x cos -.
2

- 4e-x/2 cos -x x et/2Jt sin -t dt


2 xo
2

w=

x/2 r:: . x
v x sm 2

+ C2 sin -X)

17. The auxiliary equation is 3m2

Identifying f(x)

et/2Jt sin - dt
2

and

Ul

1 -x
Ze .

Xo

Then

= 2.jX we obtain
u}

Identifying f(x)

e-x/2 sin f2
I
cos :!:- lex/2 sin:!:

le-x/2
sin :!:
2
2

ex/2 .

I !ex/2

cosx

- 1)2 = O, so Yc
xex/2
!xex/2 + ex/2

we obtain

155

+ :3xeX

sinx.

= Clex/2 + C2xex/2

I = e",

and

Exercises 4.6
Then
1 (
2)3/2
= 12 1 - x

Ul

x r:--; l.
1
U2 = -V 1 - x- + - sin " x

and

19. The auxiliary equation is 4m2 - 1 = (2m - 1)(2m


W =

+ 1) = O, so Yc = clex/2 + C2e-x/2 and

I !eex/22 _!ee='? I = -1
2
.
x/

Identifying f(x)
U2 = -xex/4+

x/

= xex/2/4 we obtain ui = x[A and ui = -xex/4.

Then

Ul

= x2/8 and

eX/4. Thus

+ c2e-x/2 + ~x2ex/2

y = clex/2

_ ~xex/2

+ ~ex/2
4

1
1
= C3ex/2 + C2e-x/2 + _x2ex/2
__ xex/2

and
yf = ~c3eX/2 _ ~c2e-X/2

+ ~x2ex/2
16

+ ~xex/2
8

_ ~ex/2.
4

The initial conditions imply


C3

1
1
- -C2
224

-C3

Thus

C3

= 3/4 and

C2

=1

C2

1
-

= O.

= 1/4, and

20. The auxiliary equation is 2m2

+m -

1 = (2m - l)(m
ex/2

W =
Identifying f(x)

I !ex/2

= (x + 1)/2 we obtain

156

+ 1) = O, so Yc = clex/2 + C2e-x

and

Exercises 4.6
Then
Ul

_e-x/2

U2

--xe

1
3

Gx - 2)

Thus
-x- 2

x/2 +C2e -x

y=cle

and
,

x/2

y =-Cle

-x

-c2e

-.

The initial conditions imply


Cl -

C2 -

2= 1

C2 -

1
-Cl

Thus

Cl

= 8/3 and

C2

= 1/3, and

y =
21. The auxiliary equation is m2

+ 2m -

8 x/2
1 -x
3"e
+ 3"e
- x - 2.

8 = (m X

2e-2x - e-x

+ 4) = O, so Yc = cle2x + C2e-4x

2)(m

I=

e-4x
- 4 e -4x

w = 12e::
Identifying f(x)

O.

_6e-2x.

we obtain
,

ul =

1 -4x
1 -3x
-e
- -e

1 3x

u2= --e
6

1 2x

--e
3

Then

Thus
y = cle2x

= cle2x

+ C2e-4x

_ ~e-2x

+ C2e-4x

__ e-2x

+ ~e-x

12

and
y' = 2ee2x _ 4C2e-4x

18

1
+ _e-x

1
+ _e-2x
2

157

__

1
e-x.
9

+ ~e-x
18

_ ~e-2x

and

Exercises 4.6
The initial conditions imply
q

5
36 = 1

C2 -

7
2q - 4C2 + 18 = O.
Thus

q = 25/36

C2 = 4/9,

and

and
25
y = _e2x
36

22. The auxiliary equation is m2 - 4m

Identifying f(x)

4
+ _e-4x
9

__ 1e-2x
4

+ 4 = (m - 2)2

1
+ _e-x
9

= 0, so Yc = qe2x

+ C2xe2x and

= (12x2 - 6x) e2x we obtain

u = 6x2

- 12x3

U2 = 12x2 - 6x.
Then

= 2x3

3x4

U2 = 4x3

3x2.

Ul

Thus
y

q e2x

+ C2xe2x + (2x3

3x4) e2x

+ (4x3

3x2) xe2x

3x2)

+ 2e2x

= qe2x + C2xe2x + e2x (x4 _ x3)


and
y'

= 2qe2x + C2 (2xe2x + e2x) + e2x ( 4x3

(x4 _ x3)

The initial conditions imply

=1

ci

2q
Thus

q =

1 and

C2 =

+ C2 = O.

-2, and

y = e2x _ 2xe2x

+ e2x (x4 _ x3) = e2x (x4 - x3

2x

1) .

23. Write the equation in the form


y"
and identify f(x)

= x-1/2. From

Yl

+ .!:.y' +
x

(1 __4x21_) y = x-1/2

= x-1/2cosx

and Y2

158

= x-1/2sinx

we compute

Exercises 4.6

Now
I

= COSX,

.
= smx

so

Ul

u; = cosx

so

U2 = sinx.

COSX

+ c2x-1/2

sinx

+ x-1/2

COS X

+ C2X-l/2

sin x

+ x-l/2.

Ul

and

Thus
y = CIX-1/2
=

ClX-1/2

cos2 X

+ x-l/2

sin2

24. Write the equation in the form


11

y
and identify f(x)

= sec(lnx)jx2.

sec(lnx)

+ ~Y + x2Y

From

Yl

= --x""'2--'-

= cos(lnx) and

sin(ln x)

cos(ln x)

w=

= sin(lnx) we compute

Y2

sin (In x)

cos(ln x)

1
x

Now
I

ul

=-

tan(ln x)
x

so

Ul

= In I cos(lnx)'

and
I

u2

= -1 so
x

U2 = In x.

Thus, a particular solution is


Yp = cos(lnx) In I cos(lnx)1

25. The auxiliary equation is m3

+m

= m(m2
1

w=

+ (lnx)

+ 1) = O, so Yc
cosx

-sinx

- cosx

159

sin(lnx).

= Cl

+ C2 COSX + C3 sinx

sinx
cosx
-sinx

=1.

and

Exercises

4.6

Identifying f(x)

= tanx we obt~in

cosx

sinx

- sinx

cosx

tan x

- cos x

- sin x

sinx

cosx

tanx

- sinx

= - sinx

cosx

u~ = W3 = O
O

-sinx

O
tan x

-cosx

= tanx

= - sin x tan x

cos2 X - 1
= cos x - sec x.
cosx

Then
ln I cos z]

Ul

U2

= cosx

U3

= sin x - in I sec x + tan z]

and
y

= Cl + C2 cos X + C3 sin x - in I cos z] + cos2 X

+ sin 2 x
= C4

for

-00

- sin x in I sec x

+ C2 COSX + C3 sinx

+ tan z]

ln I cos rl

>

sinx in I secx

+ tan

z]

< x < oo.

26. The auxiliary equation is m3

+ 4m = m (m2 + 4) = O, so Yc = Cl + C2 COS 2x + C3 sin 2x


1
W

cos 2x

= O -2sin2x
O -4 cos 2x

160

sin2x
2cos2x
-4 sin 2x

= 8.

and

Exercses
Identifying

f (x) = sec 2x

we obtain

,1
u1

4.6

= -W
1 =
8

1
-8

cos 2x

sin 2x

-2 sin 2x

2 cos 2x

sec 2x

-4 cos 2x

-4 sin 2x

sin2x

2 cos 2x

O sec 2x
,1
u3 = 8"W3

1
4

-4 sin 2x

1
cos 2x
1
8" O -2sin2x
O -4 cos 2x

= -4 sec 2x

= -- tan 2x.

sec 2x

Then
U1

1
= 8"In I sec 2x

U2

U3

1
= 8"lnlcos2xl

+ tan

2xl

-'4x

and
Y = CI + C2 cos 2x
for

+t:

/4

<

+ C3 sin 2x + 8"In I sec 2x + tan

2x I -

'4 x cos 2x + 8"sin 2x In I cos 2x I

<.7f /4.

27. The auxiliary equation is 3m2

6m+ 30 = O,which has roots 1 + 3i, so Yc

= eX(e cos 3x+

+ 30y = 15 sin x, which


+ B sin x and substituting

We consider first the differential equation 3y" - 6y'


undetermined coefficients. Letting

YPl

= A cos x

C2

sin 3x).

can be solved using


into the differential

equation we get
(27A - 6B)cosx

+ (6a+

27b) sinx = 15sinx.

Then
27A - 6B = O and

6a

+ 27b = 15,

so A = ~ and B = 197 . Thus, YPl = (7 cos X + f.7 sin x. Next, we consider the differential equation
3y" - 6y' + 30y, for which a particular solution YP2 can be found using variation of parameters. The
Wronskian is

W=
Identifying f(x)

eX sin 3x
I = 3e2x.
3ex cos 3x + eX sin 3x

eX cos 3x

I eX cos 3x - 3ex sin 3x

= ~ex tan x we obtain

, = -'9sm
l.
3xtan 3x

u1

and

161

u2

l.

= '9 sin 3x.

Exercses

4.6

Then
Ul

. -3X) - In ( cos -3x + sm


. -3X)]
= -1.sm 3x + -1 [(In cos -3x - sm
27

27

1
u') = -- cos3x.
27
Thus
YP2

21eX7 cos3x [In( cos 3; _ sin 3;) _ In (cos 3; + sin 3;) ]

and the general solution of the original differential equation is

= (m-1)2 = 0, which has repeated root 1, so Yc = clex+c2xex.

28. The auxiliary equation is m2-2m+1

We consider first the differential equation y" - 2y' + y = 4x2 - 3, which can be solved using
undetermined coefficients. Letting

= Ax2 + Bx +

YPl

and substituting into the differential

equation we get

= 4x2 - 3.

AX2 + (-4A + B)x + (2A - 2B + e)


Then
A = 4,

-4A + B = 0,

and

2A - 2B +

e = -3,

so A = 4, B = 16, and e = 21. Thus, YPl = 4x2 + 16x + 21. Next we consider the differential
equation y" - 2y' + y = x-1ex, for which a particular solution YP2 can be found using variation of
parameters. The Wronskian is

w=
Identifying f(x)

= eX/x

we obtain u~

eX
xe"
eX xe" + eX

I = e2x

-1 and u2 = l/x.

Then

Ul

= -x and U2 = lnx, so that

and the general solution of the original differential equation is


y

= Yc +

YPl

YP2

= clex + c2xex + 4x2 + 16x + 21 - xe" + xe" lnx.

29. The interval of definition for Problem 1 is (-7r /2, 7r/2), for Problem 7 is (-00,00),

is (0,00), and for Problem 18 is (-1,1).


y
for -7r/2

for Problem 9

In Problem 24 the general solution is

= cj costln z ) +c2sin(lnx) + cos(lnx) In I cos(lnx)1 + (Inx)sin(1nx)

< lnx < 7r/2 or e-1r/2 < x < e1r/2. The bounds on lnx are due to the presence of sec(lnx)

in the differential equation.


30. We are given that Yl = x2 is a solution of x4y" + x3y' - 4x2y = O. To find a second solution we use

reduction of order. Let y = x2u(x). Then the product rule gives


y' = x2u' + 2xu

and

y"

162

= x2u" + 4xu' + 2u,

Exercises 4.7
so
X4y"
Letting w

= u',

this becomes xw'


dw

Thus, w = x-5

+ x3y' - 4x2y = x5(xu" + 5u')


+ 5w = O. Separating variables
5

= -- dx

In Iwl

and

solution of the homogeneous differential equation is Yc

and integrating we have

+ c.

= -5lnx

A second solution is then Y2

and u = -tx-4.

= O.

= x2x-4

= 1/x2, and the general

= CX2 + C2/x2. To find a particular solution,

YP' we use variation of parameters. The Wronskian is


'(IV =

I x2 1/x2 I = _. .
2x

Identifying f(x)
U2 =

-t lnx,

tam u =;jX
1/ X4 we obtai

_2/x3

-5

an d' u2 = -;jX

-1

Then u

i -4
-I6x
an d

so
Y

1
= - -x16

x 2 - -1 (ln x )x -2
4

-4

1
= - -x16

-2 - -1_2
x In x.
4

The general solution is


y = cx

C2
1
1
+ -x2
- -- -4x2
16x2

In x.

Exercises 4.7

1. The auxiliary equation is m2 - m - 2 = (m


2. The auxiliary equation is 4m2 - 4m

+1=

+ l)(m

- 2) = Oso that y

= CIX-l + C2x2.

(2m - 1)2 = O so that y = Clxl/2

+ C2xl/2Inx.

3. The auxiliary equation is m2 = Oso that y = CI + c2ln x.


4. The auxiliary equation is m2 - 4m = m(m - 4) = Oso that y = CI + C2x4.

5. The auxiliary equation is m2 + 4 = Oso that y

= CIcos(2ln x) + C2sin(2ln x).


6. The auxiliary equation is m2 + 4m + 3 = (m + l)(m + 3) = Oso that y = CIX-l + C2X-3.

+ C2x2+v'6.
is m2 + 2m - 4 = O so that y = CjX-I+v's + C2X-I-v's.
is 25m2 + 1 = Oso that y = Ccos (i In x) + C2(! In x)

7. The auxiliary equation is m2 - 4m - 2 = Oso that y = clx2-v'6


8. The auxiliary equation

9. The auxiliary equation

+ 1) = O so that y = CIX/2 + C2X-1/2


11. The auxiliary equation is m2 + 4m + 4 = (m + 2)2 = O so that y = CIX-2 + c2x-2Inx.
12. The auxiliary equation is m2 + 7m + 6 = (m + l)(m + 6) = Oso that y = CIX-l + C2X-6.
10. The auxiliary equation is 4m2 - 1 = (2m - 1)(2m

13. The auxiliary equation is 3m2+3m+

1 = Oso that y

163

= x-I/2

[CIcos

(V;

In x)

+ C2sin

(V;

lnx ) ].

Exercses 4.7
14. The auxiliary equation is m2 - 8m+ 41 = O so that y

= x4

[C1cos(51nx) + C2sin(51nx)]:

15. Assuming that y = xm and substituting into the differential equation we obtain
m(m - l)(m - 2) - 6 = m3

3m2 + 2m - 6 = (m - 3)(m2 + 2)

= O.

Thus
y
16. Assuming that y

= C1x3 + C2cos (hin x) + C3sin ( h In x) .

= xm and substituting into the differentiaI equation we obtain

m(m -l)(m

- 2) + m - 1 = m3

3m2 + 3m - 1 = (m - 1)3 = O.

Thus

17. Assuming that y

= Xffi and substituting into the differential equation we obtain


= m4 - 7m2 + 6m = m(m - l)(m - 2)(m + 3) = O.

m(m - l)(m - 2)(m - 3) + 6m(m - l)(m - 2)


Thus

18. Assuming that y

m(m-1)(m-2)(m-

Xffi

and substituting into the differential equation we obtain

3) +6m(m-1)(m-2)

+3m+ 1 = m4 +2m2+ 1 = (m2 + 1)2 = O.

+9m(m-1)

Thus
y

= C1cos(lnx) + C2sin(lnx) + c3In x cos(ln x) + c4ln x sin(ln x).

19. The auxiliary equation is m2 - 5m

= m(m

- 5)

= O so that

Yc

= C1+

C2x5 and

x5 = 5x 4 .
W(l, x 5 ) = O1 5x4
Identifying f(x) = x3 we obtain

= -ix4 and U2 = 1/5x. Then u1 = -Ax5,

51515

y=C1+C2X

-25x

+Sx

515

lnx=c1+c3X

+Sx

U2 = ilnx,

and

lnx.

20. The auxiliary equation is 2m2 + 3m + 1 = (2m + l)(m + 1) = O so that Yc = C1X-1 + C2X-1/2 and
W(x-1
Identifying f(x) = ~-

x-1/2) =

-1
X
_x-2

we obtain u = x -

-1/2
X
_~x-3/2

I = _x-5/2.
1
2

x2 and U2 = x3/2 - x1/2. Then U1 = ~x2 -1x3,

U2 = ~x5/2 - ~x3/2, and


Y

1
1 2 2 2 2
= C1x-1 + C2x-1/2 + -2x
- -3x + -5x - -x
3

164

-1

= C1x

-1/2
1
1 2
+ C2x
- -x + -x .
6
15

4.7

Exercises
21. The auxiliary equation is m2 - 2m

+1=

(m - 1)2

w (x, x ln x)
Identifying f(x)
and

= 2/x

we obtain

ul

= -21nx/x

y = C1X

22. The auxiliary equation is m2 - 3m

Identifying f(x)

= x2ex we obtain

I1

= x

ul

x ln x = x.
1 + ln x

and U2

+ C2X ln x

+ 2 = (m

= O so that Yc = ClX + C2X ln x and

- x(ln x)2

- l)(m - 2)

= -x2ex

= z. Then

+ 2x(ln

Ul

U2

= 21nx,

x)2

= Oso that Yc

and U2 = xe".

= -(lnx?,

Then

= ClX

Ul

+ C2x2

and

= _x2ex + 2xex

2ex,

U2 = xe" - eX, and

23. The auxiliary equation is m2

+ 2m

y = Cl + C2X-2

= m(m

and

+ 2)

= O, so that

y' = -2C2X-3.

The initial conditions imply


-10

-2C2 = 4.
Thus, Cl = 2, C2 = -2, and y

= 2 - 2x-2.
-20
y

24. The auxiliary equation is m2 - 6m


y

= ClX2 + C2x4

+ 8 = (m
and

y'

- 2)(m - 4) = O, so that

2ClX

+ 4C2X3.

The initial conditions imply


4 x

-4

4Cl + 32c2 = O.
Thus, Cl = 16, C2 = -2, and y

= 16x2

20

2x4.

30

165

Exercses

4.7

25. The auxiliary

equation

+ 1 = 0,

is m2

so that

y
3

+ C2sin(lnx)

y = C cos(lnx)
and

. (ln z )
y , = -CI ~1 sm
x
The initial conditions

+ C2-1 cos(in
x

lOO"

50

x ).

-3

irnply CI = 1 and C2 = 2. Thus

+ 2sin(lnx).

y = cos(inx)

26. The auxiliary equation

+ 4 = (m

is m2 - 4m

Y=Clx2+c2x2inx

and

- 2)2 = 0, so that

y'=2CIX+C2(x+2xinx).

imply CI = 5 and C2 + 10

The initial conditions

= 3.

Thus y = 5x2 - 7x2lnx.


-10

-20

-30

27. The auxiliary equation is m2

= O so that

Yc = CI + c2in x and

15

W(l, ln z) = 11 in/x I = ~.
O 1 x
x
Identifying

f(x)

UI = tx2 - !x2inx,

equation

so that Yc = CIX2

UI

f(x)

u2 = x.

10

Then

+8

in x = c

C2+ ~ =

12

+ C2in x + - x
4

-!. Thus,

= (m - 2)(m - 4)

CI =

~, C2 = -1,

0,

y
0.05

and

= I x2
2x

X43]
4x

= 8x4 we obtain

= -x4, U2 = 2x2,

12

+ -x

t = 1 and

is m2 - 6m

+ C2x4
W

Identifying

- - x in x

+ t x2.

in x

28. The auxiliary

1212

+ -x

imply CI +

The initial conditions

= ~-

ln z and

U2 = ~x2, and

y = CI + C2in x

and y

u~ = +z

1 we obtain

= 2x5.

u~ = -4x3

and y = CX2

and

+ C2X4 + x6.

u2 = 4x.

Then

The initial conditions

166

1 x
-1

imply

Exercises 4.7

1
-Cl
4

1
1
=-16
64
1
3
Cl + 2"C2 = -16 .

1
1
d
Th us Cl = 16'
C2= -"2' an y

+ -C2

1 2 1 4
= l6X
-"2X
+ X 6.

29. Substituting into the differential equation we obtain

d2y
dt2

+8

The auxiliary equation is m2 + 8m - 20 = (m


y

dy
dt - 20y = O.

+ 10)(m

- 2) = O so that

= cle-lOt + C2e2t = qx-lO + C2x2.

30. Substituting into the differential equation we obtain

d2y
dy
dt2 - 10 dt

+ 25y

= O.

The auxiliary equation is m2 - 10m + 25 = (m - 5)2 = O so that


y

= clt + C2tt = C1X5+ c2x51n x.

31. Substituting into the differential equation we obtain

d2y

dy

-dt2 + 9+ 8y
dt
The auxiliary equation is m2

= e2t

+ 9 + 8 = (m + l)(m + 8)

= O so that

undetermined coefficients we try YP = Ae2t. This leads to 30Ae2t


y=Cle-

+C2e-t+30e

1 2t

=C1x

-1

+C2X

Yc = cle-t

= e2t, so that A
-8

1
+30x.

+ C2e-8t. Using
= 1/30 and

32. Substituting into the differential equation we obtain


d2y
dy
dt2 - 5 dt
The auxiliary equation is m2 - 5m

+6

undetermined coefficients we try YP

+ 6y = 2t.

= (m - 2)(m - 3) = O so that Yc = cle2t

+ C2e3t. Using

+ B. This leads to (-5A + 6B) + 6At

= 2t, so that

At

A = 1/3, B = 5/18, and


y

= cle2t + c2e3t + 3"t + 18 = C1X2+ C2x3 + 3"lnx + lB'

33. Substituting into the differential equation we obtain

d2y
dy
dt2 - 4 dt

+ 13y
167

=4

+ 3e .

Exercses

4. 7

The auxiliary equation is m 2- 4m + 13 = Oso that Yc = e2t (ci cos 3t + C2sin 3t). Using undetermined
coefficients we try Yp = A+Bet.

This leads to 13A+ 10Bet

= 4 +3et, so that A = 4/13, B = 3/10,

and
.
Y=e2t(clcos3t+c2sm3t)+
=X2[Clcos(3lnx)

4
3
13 + 10et

+c2sin(3lnx)1

4
3
+ 13 + 10x.

34. From

it follows that

Substituting into the differential equation we obtain


-d3y - 3 -d2y + 2 -dy - 3 (d2y
- -dY) + 6 -dy - 6y = 3 + 3t
dt3
dt2
dt
dt2
dt
dt
or
d3y
d2y
dt3 - 6 dt2
The auxiliary equation is m3-6m2+11m-6

dy

+ 11 dt

- 6y

= 3 + 3t.

= (m-1)(m-2)(m-3)

= Oso that Yc = clet+c2e2t+

C3e3t. Using undetermined coefficients we try YP = A + Bt. This leads to (11B - 6A) - 6Bt = 3 + 3t,
so that A

= -17/12, B = -1/2, and


t
2t
3t 17 1
2
3
17 1
y = cle + C2e + C3e - - - -t = C1X+ C2X + C3X - - - -lnx.
12 2
12 2

168

Exercises 4.7
In the next two problems we use the substitution t = -x since the initial conditions are on the interval
(-00, O). In this case
dy
dy dx
dt = dx dt =

dy
dx

and

35. The differential equation and initial conditions become


d2y
4t2-+y=0
dt2
The auxiliary equation is 4m2 - 4m
y = cl tl/2

+ c2t1/21n

y(t)
'

+ 1 = (2m

I
t=l

=2,

- 1)2 = O, so that

+ C2

y' = ~CICl/2

and

It=l = -4.

y'(t)

(Cl/2 + ~cl/2In t) .

The initial conditions imply CJ = 2 and 1 + C2= -4. Thus


y

= 2tl/2 - 5tl/2ln t

2(_x)1/2 - 5(_x)1/2In( -x),

< O.

36. The differential equation and initial conditions become


t2 ~:; - 4t ~~ + 6y = O;

y(t)

It=2 =

y'(t)

8,

t=2

= O.

+ 6 = (m - 2)(m - 3) = O, so that
= clt2 + C2t3 and y' = 2clt + 3c2t2.

The auxiliary equation is m2 - 5m


y
The initial conditions imply

4Cl + 12c2 = O
from which we find c

= 6 and

C2= -2. Thus


y

37. Letting u = x

+ 2 we obtain

ddy
x

d2y
dx2

= 6t2 -

2t3

= 6x2 + 2x3,

< O.

dy and, using the chain rule,


du
d (dY)
dx du

d2y du
= du2 dx

d2y
du2(1)

d2y
du2

Substituting into the differential equation we obtain


2 d2y
dy
u -+u-+y=O.
du2
du
The auxiliary equation is m2 + 1 = O so that

= C1cos(ln u) + C2sin (In u)

= c cos[ ln(x

169

+ 2)) + C2sin[ ln(x + 2)).

Exercises

4.7

38. If 1 - i is a root of the auxiliary equation then so is 1 + i, and the auxiliary equation is

(m - 2)[m - (1 + i)][m - (1 - i)]


We need m3 -4m2+6m-4

= m3 -

to have the form m(m-1)(m-2)

last expression and equating coefficients we get b

-1, c

4m2 + 6m - 4

= O.

+bm(m-1)

+cm+d. Expanding this

-4. Thus, the differential

3, and d

equation is

+ 3xy'

X3y'" - x2y"
39. For x2y"

- 4y = O.

= O the auxiliary equation is m(m - 1) = O and the general solution is y = Cl + C2X. The

initial conditions imply Cl = YO and C2 = Yl, so Y

Yo + Y1X. The initial conditions are satisfied

for all real values of YO and Yl.

= (m - l)(m - 2) = O and the


general solution is Y = C1X + c2x2. The initial condition y(O) = Yo implies O = Yo and the condition
For x2y" - 2xy' + 2y = O the auxiliary equation is m2 - 3m + 2

y'(O) = Yl implies Cl = Yl. Thus, the initial conditions are satisfied for YO = O and for all real values
of Yl.
For x2y" - 4xy' + 6y = O the auxiliary equation is m2 - 5m + 6

= (m - 2)(m - 3) = O and the

= C1X2 + c2x3. The initial conditions imply y(O)


the initial conditions are satisfied only for YO = Yl = O.
general solution is Y

40. The function y(x)

-v'xcos(lnx)

is defined for x>

for kan integer or where x = e1r/2+krr.

= O = YO and y' (O) = O. Thus,

O and has x-intercepts where lnx = 7r/2 + kr:

Solving 7r/2 + ktc

= 0.5 we get k::::; -0.34, so e1r/2+k1r < 0.5

for all negative integers and the graph has infinitely many x-intercepts in (0,0.5).
41. The auxiliary equation is 2m(m - l)(m - 2) - 10.98m(m - 1) + 8.5m + 1.3

= O, so that ml

-0.053299, m2 = 1.81164, m3 = 6.73166, and


Y

= C1X-O.053299
+ C2X1.81164 + C3x6.73l66.

42. The auxiliary equation is m(m - l)(m - 2) + 4m(m - 1) + 5m - 9 = O, so that ml


the two complex roots are -1.20409
Y

2.22291i.

1.40819 and

The general solution of the differential equation is

= ClX1.408l9+ x-1.20409[c2cos(2.22291ln x) + C3sin(2.22291ln x)].

43. The auxiliary equation is m(m - l)(m - 2)(m - 3) + 6m(m - l)(m - 2) + 3m(m -1) - 3m + 4

so that ml

= O,

= m2 = v'2 and m3 = m4 = -v'2. The general solution of the differential equation is


y = C1XV2 + c2xV2lnx

+ C3X-v'2+ c4x-v'2ln x.

44. The auxiliary equation is m(m-1)(m-2)(m-3)-6m(m-1)(m-2)

so that ml = m2

+33m(m-1)

-105m+ 169 = O,

= 3 + 2i and m3 = m4 = 3 - 2i. The general solution of the differential equation

is
Y

= X3[Cl cos(2ln x) + C2sin(2ln x)] + x3ln X[C3 cos(2ln x) + C4sin(2ln x)].


170

Exercises 4.8

Exercises 4.8
1. From

Dx = 2x - y and Dy = x we obtain y

= 2x -

Dx, Dy = 2Dx - D2x, and (D2 - 2D + 1)x = O.

Then
X

= Clet +

)t + C2tet .
C2tet an d Y = (Cl - C2e

2. From Dx = 4x + 7y and Dy = x - 2y we obtain y = ~Dx - ~x, Dy

(D2 - 2D - 15)x = O. Then


1

5t

Y = "7cle

-3t

- C2e

3. From Dx = -y + t and Dy = x - t we obtain y = t - Dx, Dy = 1 - D2x, and (D2 + 1)x = 1 + t.


Then

x = Cl COS t + C2sin t + 1 + t
and

y = Clsin t - C2cos t + t - 1.
4. Frorn Dx - 4y

= 1 and

x + Dy

= 2 we obtain y = iDx

-..t,

Dy

= iD2x,

and (D2 + 1)x

= 2.

Then

x = Cl COS t + C2sin t + 2
and

1
1.
1.
1
y = C2cost - Cl smt - Cl smt - .

(D2 + 2)y = O we obtain y = ~(D2 + 5)x, D2y = ~(D4 + 5D2)x,

5. Frorn (D2 + 5)x - 2y = O and -2x+


and (D2 + 1)(D2 + 6)x = O. Then

x = Cl COS t + C2sin t + C3cos v'6 t + C4sin v'6 t


and

y = 2Cl COS t + 2C2sin t - ~C3 cos v'6 t - ~C4 sin v'6 t.

-i - i(D

6. Frorn (D + 1)x + (D - 1)y = 2 and 3x + (D + 2)y = -1 we obtain x

Dx = -i(D2

+ 2D)y, and (D2 + 5)y = -7. Then


y

and

= Cl COS /5 t + C2sin /5 t - ~

( 32

/5)
x = --Cl - -C2
3

cosv'5t+

171

(/5
-Cl
3

2).

- -C2 smv'5t+
3

-.
5

+ 2)y,

Exercises 4.8
(D2

+ 4)(D

+ 2)x

- 2)(D

= -3et.

Then

and
.

Y = -e cos 2t 8. From (D2

(D - 5)(D2

+ 5)x + Dy = O and
+ 4)y = O. Then

2t

C2 SIn

+ l)x + (D

(D

+ C3e2t + C4e -2t

- 4)y

1 t
Se
.

= O we obtain (D - 5)(D2 + 4)x = O and

x = cle5t

+ C2 cos 2t + C3 sin 2t

y = c4e5t

+ C5 cos 2t + C6 sin 2t.

and

Substituting into (D
(6e
so that

C4

+ l)x + (D -

4)y = Ogives

+ c4)e5t + (C2 + 2C3 -

-6Cl,

= ~C3,

C5

C6

4C5 + 2C6) cos 2t

+ D2y =

2C5 - 4C6) sin 2t = O

= -~C2, and
-t

y = -6cle"
9. Frorn Dx

+ (-2C2 + C3 -

+ 2'C3 cos 2t

+ l)x + (D

e3t and (D

- l)y

- 2'C2 sin 2t.

= 4e3t we obtain D(D2 + l)x = 34e3t and

D(D2 + l)y = _8e3t. Then


y = e

+ C2 SIn. t + C3 cos t -

and
x
Substituting into (D

+ l)x + (D

(C4 so that

C4

ci)

4 3t

15e
17

C4

+ C5 sin t + C6 cos t + 15e3t.

- l)y = 4e3t gives

+ (C5 -

C6 -

+ (C6 + C5 + C2 -

C3 -

C2) sin t

.
cos t + C3 SIn t

C3) cos t

=O

= cj , C5 = C3, C6 = -C2, and


x = Cl

10. Frorn D2x - Dy = t and (D + 3)x


D(D + l)(D + 3)y = -1 - 3t. Then

C2

+ (D + 3)y =

and

172

17 3t
15e

2 we obtain D(D

+ l)(D + 3)x = 1 + 3t

and

Exercises
Substituting into (D

+ 3)x + (D + 3)y = 2 and D2x - Dy = t


3(Cl

+ C4) + 2(C2 + c5)e-t =

4.8

gives
2

and

Y = -q

-t

C2e - 3c3e

-3t

+ t - "21t 2 .

(D2 - l)x - y = O and (D - l)x + Dy = Owe obtain y = (D2 - l)x, Dy = (D3 - D)x, and
(D - 1)(D2 + D + l)x = O. Then

11. Frorn

J3 + C3sin
. =:'
J3]
= cie t + e-t/2 [C2cos 2t

and
y =

(-~C2
2

J3 C3) e-t/2cos J3t + (J3 C2 -

12. Frorn (2D2-D-1)x-(2D+1)y


and (2D

+ l)(D + l)y

= 1 and

(D-1)x+Dy

~C3) e-t/2 sin J3t.


2
2

= -1 we obtain

(2D+1)(D-1)(D+1)x

= -1

= -2. Then

x = cle-t/2 + C2e-t + c3et + 1


and

y = C4e-t/2 + C5e-t - 2 .
Substituting into (D - l)x

+ Dy

= -1 gives

( -~Cl - ~C4) e-t/2


so that C4 = -3Cl,

+ (-2C2

- c5)e-t

=O

C5 = -2C2, and
y = - 3Cle -t/2 - 2 C2e-t - 2 .

13. From (2D-5)x+Dy

= et and

(D-1)x+Dy

= 5et we obtain

Dy = (5-2D)x+

and (4-D)x

4et.

Then

and Dy

= -3q t + 5et so that


y=--Cle3
4

14. From Dx+Dy

= et and

and D2(D - l)x

4t +C2+5e.

(_D2+D+ l)x+y = Owe obtain y = (D2-D-1)x,

= et. Then

and

173

Dy = (D3 _D2 -D)x,

Exercises

4.8
y = -Cl - C2- C2t- C3et - tet + et.

15. Multiplying the first equation by D + 1 and the second equation by D2 + 1 and subtracting we
obtain (D4 - D2)x

1. Then
X

= Cl + C2t + C3et + C4e-t -"2t.1 2

Multiplying the first equation by D + 1 and subtracting we obtain D2(D + 1)y = 1. Then
12

Y = C5+ C6t+ c-e" -"2t .


Substituting into (D - 1)x + (D2 + 1)y = 1 gives
C2+ C6)t = 1

(-Cl + C2+cs -1) + (-2C4 + 2c7)e-t + (-1-

so that C5= Cl - C2+ 2, C6= C2+ 1, and C7= C4. The solution of the system is
.X

= Cl + C2t +

12

-t

C3e + C4e -"2t

y = (Cl - C2+ 2) + (C2+ 1)t + C4e-t - ~t2.


16. From D2x - 2(D2 + D)y = sin t and

+ Dy

we obtain x

= -Dy,

D2x

D(D2 + 2D + 2)y = - sin t. Then


1

y = Cl + C2e-t cos t + C3e-t sin t + 5" cos t + 5" sin t


and

17. From Dx

1
2
x = (C2+ C3)e - t sin t + (C2- C3)e - t cos t + 5" sin t - 5" cos t.

= y, Dy

= z. and

Dz = x we obtain x = D2y = D3x so that (D - 1)(D2 + D + 1)x = O,

. Tt
J3 + C3cos TtJ3 ] ,
x = cj et + e -t/2 [C2sin
y = Clet +
and

z=qe

t+

(1 J3)
(1-"2c2+Tc3J3)
--C2 - -C3
22

. -t
J3 +
e -t/2 sm
2
. J3
e -t/2 smTt+

1)
(J3
-TC2-"2C3 1)
(J3
-C2
22

-t/2
- -C3 ecos-t

J3
2'

J3
e -t/2 cosTt.

18. From Dx + z = et, (D - 1)x + Dy + Dz = O, and x + 2y + Dz = et we obtain z = -Dx + ,


Dz = -D2x + et, and the system (_D2 + D - 1)x + Dy = _et and (_D2 + 1)x + 2y = O. Then

y = 1(D2 - 1)x, Dy = 1D(D2 - 1)x, and (D - 2)(D2 + 1)x = -2et so that


x = Cle2t
y =

+ C2COS t + C3sin t +

3 2t
.
"2Cle - C2cost - C3smt,

and

174

et,

Exercises

=-

2Cl e2t - C3 cos t

4.8

+ C2 sin t.

19. Write the system in the form

Dx - 6y = O
x - Dy + z = O
x

+y

- Dz = O.

Multiplying the second equation by D and adding to the third equation we obtain

(D + l)x - (D2 - l)y = O. Eliminating y between this equation and Dx - 6y = Owe find
(D3

D - 6D - 6)x = (D + l)(D + 2)(D - 3)x = o.

Thus
x = cle-t

+ C2e-2t + C3e3t,

and, successively substituting into the first and second equations, we get
1 -t
1 -2t
Y = -"6cle - 3"c2e
5
6

-t

Z = --Cle

1
3

-C2e

-2t

3t

+ "2c3e

1 3t
+ -C3e
2

20. Write the system in the form

(D+l)x-z=O
(D

+ l)y

(D2+D-l)z

= O. Eliminating

+ 1 and

z =O

+ Dz

x - y

Multiplying the third equation by D

= O.

adding to the second equation we obtain (D

z between this equation and (D+l)x-z

+ l)x +

= Owe find

D(D+l)2x

so that Dy

= _(D2 + 5D)x.

= O.

Thus
x

Cl

+ C2e-t + c3te-t,

and, successively substituting into the first and third equations, we get
y = Cl

+ (C2 -

+ c3te-t

c3)e-t

= Oand 4x-

(D+ l)y = Owe obtain y = -(D+5)x


Then 4x + (D2 + 5D)x + (D + 5)x = O and (D + 3)2x = O. Thus

21. From (D +5)x+y

and
Y

=-

(2 Cl

+ C2) e -3t
175

2 C2t e -3t

Exercises
Using x(1)

4.8

O and y(l)

1 we obtain

or
Cl
2Cl

Thus

Cl

= e3 and

22. From Dx - y
Then _!(D2

C2

+ C2 = O

+ 3C2 =

_e3.

= _e3. The solution of the initial value problem is

= -1 and
- 2D)y

3x

+ (D

- 2)y

= e-3t+3

_ te-3t+3

_e-3t+3

= Owe obtain

= y - 1 and (D2 - 2D + 3)y

y = et

( Cl COS

+ 2te-3t+3.
x = -!(D

- 2)y so that Dx = _!(D2

- 2D)y.

= 3. Thus

v2 t + C2 sin v2 t) + 1

and
x = ~et

[( ci -

v2 C2) cos v2 t + (v2 ci + C2) sin v2 t 1 + ~.

Using x(O) = y(O) = O we obtain


ci

+1

=O

~ (c - v2 C2) + ~ = O.
Thus

Cl

= -1 and

C2

V2/2. The

='

solution of the initial value problem is

= et(2 - - cos vr:2 t - -V2.sm v;:;)


2 t +-2

= et

(-

cos v2 t

V2.)sin v2 t + 1.
+2

23. Equating Newton's law with the net forces in the x- and y-directions gives m d2X2 = Oand m d2Y2=
dt
dt
-mg, respectively. From mD2x = Owe obtain x(t) = clt+C2, and frorn mD2y = -mg or D2y = -g
we obtain y(t) = _~gt2

+ C3t + ca.

24. From Newton's second law in the x-direction we have


d2x
1dx
dx
mdt2 = -kcose =
dt = -lcl dt .

+:

In the y-direction we have


~y
m-

dt2

.
1~
-mg - ksme = -mg - k-- = -mg
v dt

176

-Icl-.~

dt

Exercises 4.9
From mD2x + IclDx = Owe have D(mD + Icl)x = Oso that (mD + Icl)x = Cl. This is a first-order
linear equation. An integrating factor is ellcldt/melclt/m so that

!!_ [eic1t/mx]= c
dt

elclt/m
1

and e!cltx = (clm/lcl)elclt/m+C2. The general solution ofthis equation is x(t)

= c3+c2elclt/m. From

(mD2 + IclD)y = -mg we have D(mD + lel)y = -mg so that (mD + Icl)y = -mgt
first-order linear equation with integrating factor elclt/m. Thus
d
- [elclt/my]= (-mgt
dt

and

+ Cl This is a

+ cl)elclt/m

y(t) = - m 9 t + m 9 + C3+ c2e-lclt/m.


[e]
c2
25. Multiplying the first equation by D + 1 and the second equation by D we obtain

D(D + l)x - 2D(D + l)y = 2t + t2


D(D + l)x - 2D(D + l)y = O.
This leads to 2t + t2 = O, so the system has no solution.

Exercises 4.9

= eX, so

1. We have y~ = yI

(yn2 = (ex)2 = e2x = yf.


Also,

Y2 = -

sin x and y~ =

cosx, so

(y~) 2 =
However, if y

(y")2

=1=-

f = cos2x = y~.

qYl + C2Y2, we have (y")2 = (cleX - C2cos x)2 and y2 = (ci eX + C2cos xf

y~

= yI = O, so
YlYl"

Y2 = 2x

1 . O = O = -1()2
O
2

However, if y

= -21(Yl')2 .

and y~ = 2, so

Y2Y2" = x 2(2 )
~(y/)2

cos x

y2.

2. We have

Also,

(-

= ~[Cl . O+

= 2x 2 = '21(2x )2 = '21(Y2')2

qYl + c2Y2, we have yy" = (el . 1 + C2x2)(Cl. 0+ 2C2)

c2(2x)]2

= 2c~x2.

Thus yy"

=1=-

~(y/)2.

177

Thus

Exercses 4.9
3. Let u :::;:
y' so that u' = y", The equation becomes u' = -u - 1 which is separable. Thus
du

-2--

+1

= -dx ==> tan-l u = -x + Cl ==> y' = tan(cl - x) ==> y

du
l+u

--2

= 1 + u2. Separating variables we obtain

= y". The equation becomes u'

4. Let u = y' so that u'

= dx ==> tan-

u = x + Cl ==>

= tan(x

= In 1COS(Cl- x)1 + c2

==> y = +In 1cos(x + cl)1 + C2

cj )

= y' so that u' = y", The equation becomes x2u' + u2 = O. Separating variables we obtain

5. Let u

du
u2

dx

= - x2 ==>

-;

= ;; + Cl =

ClX+ 1
X

1
1
Y = 2" In IClx + 11- -x

cl

Cl

1 (

==>

= - Cl

+ l/Cl

1 (1

= Cl qx + 1 - 1

+ C2

,
"du
.
6. Let u = y so that y = u dy' The equation becomes (y

+ 1)u du
dy =

u . Separating variables we

obtain
du

dy
y+1

==> In lul = In Iy + 11+ ln c ==> u = Cl(y + 1)


dy
==> - = Cl(y
dx

+ 1)

dy
==> -= Cldx
y +1

==> In Iy + 11 = ClX+ C2 ==> Y + 1 =

C3eCx.

3
= y , so that y "du= u dy' The equation becomes u du
dy + 2yu = O. Separating variables we

7. Let u
obtain

du
2"
u

+ 2y dy = O==>-

+ y2 = C

==> u=
==>

8. Let u

= y , so that

"du

= u dy'

dy
du = - ==> u
y2
q

If Cl = O, then y dy

= -dx

y2

1
Y

) dy

qy - 1

,1

==>y=---

+ ci

y2

+ ci

"3 y + qy = x + C2

.
2 du
The equaton becomes y u dy

= -- + Cl ==> Y

==> _!:_ (1 +

= u.

Separating variables we obtain

qy - 1
Y
==>
dy
Y
ClY- 1

= dx

= dx (for Cl f. O) ==> _!:_y+ 12 In Iy - 11 = x + C2.


Cl

and another solution is

h2 = -x

178

+ c2.

cl

Exercises 4.9
9. (a)

Iy
l.

-10

(b ) Let u

du
= y ' so that y 11 = u du
dy . The equation becomes u dy + yu = O. Separating variables we

obtain

When x

= 0, y =

1 and y'

dy
1
-=--y
dx

When x

= 0, y

=
1

-1 so -1

= -~ + Cl and

dy

1
= -- dx

===?

--

===?

Y = tan ( -~x

y2

+1

= 1 so 1 = tan e, and C2

Cl

tan-

===?

+ C2)

= -~. Then
1

y = --2x

+ C2

= 1[/4. The solution of the initial-value problem is

The graph is shown in part (a).


(e) The interval of defintion is -1[/2

< 1[/4 - x/2 < 1[/2 or -1[/2 < x < 31[/2.

179

Exercises 4.9

10. Let u = y' so that u' = y". The equation becomes (u')2
which results in u' =

~.

+ u2 =

To solve u' = ~

we

separate variables:
du
Jl="U2

==> sin-l u = x + el ==> u = sin(x + c)

= dx

-27t

27t

==> y' = sin(x + el).


When x
Thus

7r ,

j3

(7r)

j3

-"67r .

= "2 ' y = 2"" ' so 2"" = sin "2 + c and el =


y' = sin

When x =

7r
"2'

(x -~)
(7r

7r)

(x - ~) + e2.

= 2"' so 2" = - cos "2 -"6 + e2 = - 2" + e2 and

initial-value problem is y = 1 - cos


To solve u'

==> y = - cos

= -~

(x - ~).

e2 = l. The solution of the

we separate variables:

du

-v'l=U2 = -dx

==> cos-l u =

+ el

==> u = cos(x + el) ==> y' = cos(x + ci).


When x

7r ,

j3

(7r)

j3

= "2 ' y = ""2 ' so ""2 = cos "2 + el and


y'

= cos (x

- i)

==> y

el

7r

= - 3". Thus

= sin (x

. (7r
= 7r
"2 ' y = 2"1 ' so 2"1 = sm
"2 - 3"7r) +e2 = 2"1+cz and

When x

problem is y = sin

- i) + e2

e2

..

..

= O. The solution of the initial-value

(x - i)1

11. Let u = y' so that u' = y". The equation becomes u' - -u = _u3, which is Bernoulli. Using
W

=U

.,
-

2
2
.
.
. 2
we obtam - + -w = - - _An mtegratmg factor lS x , so
dx
x
x
d 2 1
2
2
el
dx[x w = -2x ==> x w = -x + c ==> w = -1 + x2
.

dw

-2

==>

==>

dy
dx

ci

= -1 + 2"

x
Vel - x2

==> u=

Vel - x2

==> y = -

J el -

x2

+ e2

==> el - x2 = (e2 _ y) 2 ==> x2 + (e2 - y)2 = el-

180

Exercses

12. Let u

y' so that u'

y". The equation becomes u' - ~u

dw
su bstituti
itution w = u- 1 we obtai
tain -d

d
-[xw]
dx

= -x

==> w

1
= --X+-C
2

u2, which is Bernoulli. Using the

. factor IS
. x, so
= -.1 An mtegratmg

1
+ -w

1
1
==> x
u

4.9

Cl -

x2

2x

==> u =

2x

Cl -

x2

==> y

= -In ICl

x2 +C2.

In Problems 13-16 the thinner curve is obtained using a numerical solver, while the thicker curve is the
graph of the Taylor polynomiaJ.
13. We look for a solution of the form
1
1
y(x) = y(O) + y'(O) + 2Y"(0) + 3TY"'(O)
From y"(x)

+ 4Ty(4)

(x)

+ 5iy(5)

(x).

40

= x + y2 we compute
y'''(x)
y(4)(x)

= 1 + 2yy'
= 2yy" + 2(y')2

y(5)(x) = 2yy'"
Using y(O)

+ 6y'y".

= 1 and y'(O) = 1 we find


y"(O) = 1,

y"'(O) = 3,

y(4)(0) = 4,

y(5)(0) = 12.

An approximate solution is
y(x)

14. We look for a solution of the forrn


1
1
y(x) = y(O) + y'(O) + -2Y"(0) + --y"'(O)
3.

Frorn y"(x)

1
10

= 1 +x + _x2 + _x3 + _x4 + _x5.

1
+ --y(4)
4.

(x)

1 (
+ -51Y
5)(x).

= 1 - y2 we compute
y"'(x) = -2yy'

Using y(O)

y(4)(x) = -2yy"

_ 2(y')2

y(5)(x) = -2yy'"

- 6y'y".

= 2 and y'(O)

y"(O) = -3,

= 3 we find

y"'(O) = -12,

-5

y(4)(0)

= -6,

y(5)(0)

102.
-10

181

Exercises 4.9
An approximate solution is
3
y(X) = 2 + 3x - -x
2
15. We look for a solution of the form
1
1
y(x) = y(O) + y'(O) + "2Y"(O) + 3iylll(O)

3
1
- 2x - -x
4

+ -x17
20

.
y

4T1y(4) (x)

5T1y(5)(x).

= x2 + y2 - 2y' we compute

From y"(X)

+ 2yy'

ylll(X) = 2x

= 1 and y'(O)

y"(O)

= -1,

+ 2yy"

- 2ylll

20

- 2y(4l.

= 1 we find

y'"(O) = 4,

30

+ 2yy'"

y(5) (x) = 6y'y"

40

- 2y"

y(4)(x) = 2 + 2(y')2

Using y(O)

10

= -6,

y(4)(0)

y(5)(0)

= 14.

An approximate solution is
12231475
y(x) = 1 + x - -x + -x

+ -x

- -x

60

16. We look for a solution of the form


y(x)

= y(O) +y' (O) + ~y" (O) + ~ylll (O) + ~y(4) (x) + ~y(5) (x) + ~y(6) (x).

From y"(X)

ylll(X)
y(4)(x)

2
3!
eY we compute

4!

5!

61

= eYy'

= eY(y')2 + eYy"

y(5)(x) = eY(y')3

+ 3eYy'y" + eYylll

y(6)(x) = eY(y')4

+ 6eY(y')2y" + 3eY(y")2 + 4eYy'ylll + eYy(4).

Using y(O) = Oand y'(O) = -1 we find


y"

(O) = 1,

10

-2
-4

y 111(0) =-,1

An approximate solution is
y(x)

17. We need to solve

(1 + u2) 3/2

= u' or

-x

+ -x1213-14-x1516+ -x
2

[1 + (y')2f/2
(1 + u2) 3/2

= y".

12

+ -x

24

Let u

+ -x

45

= y' so that u' = y".

The equation becomes

= ~~. Separating variables and using the substitution

182

tan

Exercises 4.9
we have
(

du
1 + u2)3/2 = dx

==:;.

==:;.

18. Let u =

dx

di

==:;.

==:;.

y'

sec2 e
(

)3/'> de = x

e -

1 + tan2

cos e de = x
y'
1+

sin e = x

==:;.

==:;.

sec2 e
sec3 e de = x

==:;.

~
[]

(y')2

=x

x2

= x ==:;. (y')2 = x2 1+ (y')2 = -1--2

- x

vIl=X2

(for x

> O)

==:;.

=-~

du
_k2
The equation becomes u dx = ~.

d2x
du
so that dt2 = u dx

Separating variables we

obtain
k2
udu=--2dx
x
When t

==:;.

-u
2

k2
x

=-+c

k2

k2

Xo

Xo

= O, x = Xo and v = O so O = - + e and e = --.


~V2

= k2

(~

2.)
Xo

Then

dx
~/xo--o
- x
-=-kv2
dt
xXo

and

Separating variables we have


dx = kv2 dt

- ~

V~

==:;.

Po
kV"2

t=-~

J/

x
xo-x

Using Mathematica to integrate we obtain

t = - -1 ro- [-x
k

(xo - x) - -Xo tan -1 (xo - 2x)


2
2x

[V

=-1 ro
-x(xo-x)+-tan-

Xo
2

Xo - 2x
2Jx(xo - x)

dx.

{x!;]
-Xo - x

1.

19.
2

zp'

20 '
-2

d2x
For dt2

xl = o

+ sin x =

-2

xl

O the motion appears to be periodic with amplitude 1 when

amplitude and period are larger for larger magnitudes of

183

Xl.

Xl

O. The

Exercises 4.9

lO'
-1

d2x

For dt2

dx
dt

+ sin x = Othe

lO'

10 '
-1

xl = O

-1

xl= I

x l = -2.5

motion appears to be periodic with decreasing amplitude. The dx I dt

term could be said to have a damping effect.


20. When y = sin x, y'

= cos x,

y" = - sin x, and


(y")2 _ y2 = sin2 x - sin2 x = O.

From (y")2 - y2 = O we have y" = y, which can be treated as two linear equations. Since linear
combinations of solutions of linear homogeneous differential equations are also solutions, we see that
y

clex

+ C2e-x and

C3 cos x

+ C4 sin x must satisfy the differential equation. However, linear

combinations that involve both exponential and trigonometric functions will not be solutions since
the differential equation is not linear and each type of function satisfies a different linear differential
equation that is part of the original dfferential equation.
21. Letting u

= y", separating variables, and integrating we have

f1+2
1i- u~,
VI du+ u2

du _
dx - V

and

- d
- x,

sinh-l u = x

+ CI.

Then
u = y"

= sinh(x + CI),

y' = cosh(x

+ CI) + C2,

and

= sinh(x + e) + C2x + C3.

Chapter 4 Review Exercises


1. Y

=O

2. Since Yc

= cI + C2e-x, a particular solution for y"

- Y = 1 + e" is yp

= A + Bxe".

3. True
4. True
5. They are linearly independent over (-00,00)
6. (a) Since h(x)

= 21nx

and linearly dependent over (0,00).

= 2f(x), the functions are linearly dependent.


184

Chapter 4 Review Exercises


(b) Since xn+l is not a constant multiple of z", the functions are linearly independent.
(c) Since x

+ 1 is not

(d) Since h(x)

a constant multiple of x, the functions are linearly independent.

= cosxcos(7f/2) - sinxsin(7f/2) = -sinx

= -fz(x),

the functions are linearly

dependent.
(e) Since h(x)

= O fz(x), the functions are linearly dependent.

(f) Since 2x is not a constant multiple of 2, the functions are linearly independent.
(g) Since 3(x2)
(h) Since xex+l

+ 2(1 + 0(4x

x2) - (2 + x2) = O, the functions are linearly dependent.


- 5)eX

exe" = O, the functions are linearly dependent.

7. (a) The auxiliary equation is (m - 3)(m


equation is ylll

+ y"

- 17y'

+ 5)(m

- 1) = m3

+ m2 -

17m

+ 15 = O,so the

differential

+ 15y = O.

(b) The form of the auxiliary equation is


m(m - l)(m - 2) + bm(m - 1) + cm
Since (m - 3)(m

+ 5)(m

- 1) = m3

+ d = m3 + (b -

+ m2 -

17m

+ 15 =

3)m2

+ (c - b + 2)m + d = O.

O, we have b - 3 = 1, c - b + 2 = -17,

and d = 15. Thus, b = 4 and c = -15, so the differential equation is ylll + 4y" -15y'

+ 15y

= O.

8. Variation of pararneters will work for all choices of g(x), although the integral involved may not
always be able to be expressed in terms of elementary functions.

The method of undetermined

coefficients will work for the functions in (b), (c), and (e).
9. From m2 - 2m - 2 = O we obtain m = 1 y'3 so that
y
10. From 2m2

11. From m3

+ 2m + 3 = Owe obtain

+ 10m2 + 25m

Cl e(l+V3)x

+ c2e(l-v'3)x.

m = -1/2 V5/2 so that

= Owe obtain m = O, m = -5, and m = -5 so that

12. From 2m3

+ 9m2 + 12m + 5 = Owe obtain

13. From 3m3

+ 10m2 + 15m + 4 = Owe obtain


y = cle-x/3

m = -1, m

+ e-3x/2

= -1, and m = -5/2 so that

m = -1/3 and m = -3/2 0/2


(c2 cos

185

V;

+ C3 sin

V;

x) .

so that

Chapter 4 Review Exercises


14. From 2m4

+ 3m3 + 2m2 + 6m -

4 = O we obtain m

= 1/2, m = -2, and m = v'2 i so that

+ C2e-2x + C3cos h x + C4sin h x.

y = crex/2

15. Applying D4 to the differential equation we obtain D4(D2 - 3D


y

VTI
2

. VTI)
2

+ C2sin --x

= e3x/2 Clcos --x

+ 5) = O. Then

+ C3+ C4X + C5x2 + C6x3

Yc

and Yp

= A + Bx + Cx2 + Dx3.
(5A - 3B

Substituting YP into the differential equation yields

+ 2C) + (5B

+ 6D)x + (5C

- 6C

- 9D)x2

Equating coefficients gives A = -222/625, B = 46/125, C


solution is
3x/,)
222
Y = e - ClCOS--x
+ C2Sin--x
- 2
2
625

(VTI

VTI)

+ 5Dx3 =

36/25, and D

36
+ -x46 + -x
125

+ 4x3.

-2x

25

16. Applying (D - 1)3 to the differential equation we obtain (D - 1)3(D - 2D

4
+ -x
5

4/5. The general


3
.

+ 1) = (D

Then
Yc

and YP = Ax2ex

+ Bx3ex + Cx4ex.

Substituting Yp into the differential equation yields

+ 6Bxex + 2Aex = x2ex.

12Cx2ex
Equating coefficients gives A = O, B

= O,

and C = 1/12. The general solution is


1
+ _x
124ex

Y = C 1eX + C2xe"

17. Applying D(D2

+ 1) to

the differential equation we obtain

+ 1)(D3

D(D2

+ 6D) = D2(D2 + l)(D

- 5D2

- 2)(D - 3)

= O.

Then
Yc

and YP = Ax

+B

cos x

+ C sin x.

6A

Substituting Yp into the differential equation yields

+ (5B + 5C)

cosx

Equating coefficients gives A = 4/3, B


Y = Cl + c2e

+ (-5B

+5C) sinx = 8 + 2sinx.

= -1/5, and C

2x

3x

+ C3e +"3x
186

= 1/5. The general solution is

1 .

"5 cos x + 5" sin x.

- 1)5 = O.

Chapter 4 Review Exercises


18. Applying D to the differential equation we obtain D(D3

D2) = D3(D - 1) = O. Then

Yc

= Ax2. Substituting Yp into the differential equation yields -2A


gives A = -3. The general solution is
and Yp

y = Cl
19. The auxiliary equation is m2

+ C2X + C3ex

- 3x2.

2m+ 2 = [m - (1 + i)][m - (1- i)]

= 0, so Yc = clex sinx + C2ex cos x

and

w=
Identifying

eX sinx

I e" cos x

+ eX sin x

(eX cos x) (eX tan x)


2
-e X

=-

,
(eX sinx)(eX tan x)
u2 =
2
-e X
Ul

eX cosx
-exsinx+excosx

2x
= -e

=-

cosx,

U2

y = Cl eX sin x
=

Cl eX

sin x

= sinx

-In

= Slnx
sin2x

= ---cosx = cosx

I secx + tan z],

and

+ C2ex cos x

- eX cos x In I sec x

1 = 0, so Yc = qeX

+ tan x l.

+ C2e-x

e -X

-e-x
Identifying f(x)

- secx.

+ C2ex cos x - e" sin x cos x + eX sin x cos x - eX cos x In I sec x + tan z]

20. The auxiliary equation is m2

= 2ex /(eX

+ e-X)
eX

1-

and

I = -2.

we obtain
1

u' -

+ e-x

eX
- ----0'-- 1 + e2x

,
e2x
e3x
X
eX
u2 = - eX + e-X = - 1 + e2x = -e + 1 + e2x'
Then

Ul

f (x) = eX tan x we obtain


ul

Then

= 6. Equating coefficients

= tan-l eX, U2 = =e" + tan-l eX, and

21. The auxiliary equation is 6m2

m - 1=
y

so that

= qXl/2 + C2X-l/3.
187

Chapter 4 Review Exercises


22. The auxiliary equation is 2m3 + 13m2 + 24m + 9 = (m + 3)2(m + 1/2) = O so that
Y=ClX
23. The auxiliary equation is m2 - 5m + 6
x4 - x2lnx so that

-3

+C2X

-3

Inx+4:x

(m - 2)(m - 3)

O and a particular solution is Yp

y = ClX2 + C2x3 + x4 - x2lnx.


24. The auxiliary equation is m2 - 2m + 1 = (m _1)2
y

= Oand

= Clx+c2xlnx+

a particular solution is YP = ~x3 so that

1
4:x3.

25. (a) The auxiliary equation is m2 + w2 = O, so Yc = Clcoswt + C2sinwt.


A cos oi + B sin at and

When w

i-

a, Yp =

y = Clcoswt + C2sinwt + A cos at + B sin cd:


When w = a, Yp = Ateos at + Bt sin cd: and
y = Clcos wt + C2sin wt + At cos at + Bt sin cd:
(b) The auxiliary equation is m2 - w2 = O, so Yc = clewt + c2e-wt.
y = clewt

When w

i- a, YP = Al<t and

+ c2e-wt + Aeat.

When w = a, Yp = Ateat and

26. (a) If y = sinx is a solution then so is y = cosx and m2 + 1 is a factor of the auxiliary equation
m4 + 2m3 + 11m2 + 2m + 10 = o. Dividing by m2 + 1 we get m2 + 2m + lO, which has roots
-1

3i. The general solution of the differential equation is


y = Cl COS X + C2sin x + e-x (C3 COS 3x + C4sin 3x).

(b) The auxiliary equation is m(m + 1) = m2 + m = O, so the associated homogeneous differential


equation is y" + y' = O. Letting y

= Cl + C2e-x + ~x2 - x and computing y" + y' we get x.

Thus, the differential equation is y" + y' = x.


27. (a) The auxiliary equation is m4 - 2m2 + 1

= (m2 - 1)2 = O, so the general solution of the

differential equation is
y

= Clsinhx

+ C2coshx

+ C3Xsinhx + C4Xcosh x.

(b ) Since both sinh x and x sinh x are solutions of the associated homogeneous differential equation,
a particular solution of y(4) - 2y" + y = sinhx has the form YP = Ax2 sinhx + Bx2 coshx.

188

Chapter 4 Review Exercises


28. Since

yi

= 1 and yI = O, x2yi' - (x2 + 2x)yi + (x + 2)Yl

= _x2 - 2x

+ x2 + 2x = O, and

Yl

= x is

a solution of the associated homogeneous equation. Using the method of reduction of order, we let
y

= ux. Then y' = xu' + u and y" = xu" + z, so


x2y" - (x2 + 2x)y' + (x + 2)y = x3u" + 2x2u' - x3u' - 2x2u' - x2u - 2xu + x2u + 2xu
= x3u" - x3u' = x3 (u" - u').

To find a second solution of the homogeneous equation we note that u


u" - u' = O. Thus, Yc =

CX

+ C2xex.

= eX is a solution of

To find a particular solution we set x3(u" - u') = x3 so

that u" - u' = 1. This differential equation has a particular solution of the form Ax. Substituting,
we find A = -1, so a particular solution of the original differential equation is YP = _x2 and the
general solution is y

= cx

+ C2xex

29. The auxiliary equation is m2


Y(7f/2)

- x2.

2m + 2 = Oso that m = 1 i and y = eX(cl cos x

= Oand Y(7f) = -1 we obtain

30. The auxiliary equation is m2

= e-7r and

Cl

+ 2m + 1 =

C2

= O. Thus, y = eX-7r

+ C2 sin x).

Setting

cosx.

+ 1)2 = O, so that y = cle-x + C2xe-x. Setting


C2e = Oand -Cl + C2 = O. Thus c = C2 and y = ce-x + cxe:"

y( -1) = Oand y'(O) = Owe get cj e -

(m

is a solution of the boundary-value problem for any real number c.


31. The auxiliary equation is m2 - 1 = (m - l)(m
Assuming YP

+ 1) =

O so that m

1 and y

clex

+ c2e-x.

Ax + B + e sin x and substituting into the differential equation we find A = -1,

= O, and e = --!.

Thus YP = -x - -!sinx and


Y=

Cl

+ C2e-x -

.
"21 sin
x.

x -

Setting y(O) = 2 and y'(O) = 3 we obtain


Cl

Cl -

Solving this system we find

Cl

and

li

C2 -

-i.

3
= 3.
2
The solution of the initial-value problem is
-

13 x 5 -x
1 .
= -e
- -e - x - - smx.

32. The auxiliary equation is m2

C2

+ C2 = 2

442

+ 1 = O, so Yc = Cl cos X + C2 sin x and


W=

cosx

sin x =1.
cosx

-sinx

Identifying f(x) = sec3 x we obtain


,

ul

= -smxsec

x =

sin x

---3COS X

u; = cos x sec3 x = sec2 x.

189

Chapter 4 Review Exercises


Then
Ul

1
1
= --sec
2 cos-' X
2

= ----

u2 = tanx.
Thus
Y=

cr cos x + C2 sin
X

= Cl
=

and

2'1 cos x sec2.x + sin X tan x

2' sec x +

COS X

C3 COS X

,.

Yp =

-C3

+ C2 sin X

1-

COS2 X

cos x

+ C2 Slll X + 2'1 sec x.

sin x

+ C2 cos z + 2' sec x tan

x.

The initial conditions imply


1

C3

+ 2'

=1

C2 =-.

Thus

C3

= C2 = 1/2 and
Y =

33. Let u

1.

2' cos x + 2' sin X + 2' sec x.


du

= y' so that u' = y". The equation becomes u dx = 4x. Separating variables we obtain
u du

When x

= 4x dx

===:;,.

1
_u2 = 2x2 +
2

= 1, y' = u = 2, so 4 = 4 + C2 and

C2

C1 ===:;,.

u2 = 4x2

+ C2.

= O. Then

or

dy = -2x
dx

When x = 1, y = 5, so 5 = 1 + C3 and 5 = -1 + C4. Thus C3 = 4 and C4 = 6. We have y = x2 + 4


and y = -x2 + 6. Note however that when y = -x2 + 6, y' = -2x and y'(l) = -2 =f 2. Thus, the
solution of the initial-value problem is y
34. Let u

=y

so that y

"du

= u -.dy

= x2 + 4.

.
du
The equation becomes 2u dy

2u dy = 3y2 dy

===:;,.

190

= 3y 2 .

u2 = y3 + Cl.

Separating variables we obtain

Chapter 4 Review Exercises


When x =

o, y

= 1 so

= y3

==}

(~~)

==}

_2y-l/2

u2

When x

1 = 1 + Cl and Cl

= 1 and y' = u

= O, y = 1, so 1

however that when y = (x

y3

==}

=X

+ C2

2 =

~~

= O. Then

= y3/2

==}

==}

Y = (x

y-3/2 dy

= dx

+ C2)2 .

4
4
4
Note
= 2" ==} C2 = 2. Thus, y = (x + 2)2 and y =
(x
2)2 .
c2
4
,
8
+ 2)2 , y = - (x + 2)3 and y'(O) = -1 =11. Thus, the solution of the

initial-value problem is y = (x ~ 2)2 .


35. (a) The auxiliary equation is 12m4

+ 64m3 + 59m2

- 23m -12

= Oand has roots -4, -3/2,

-1/3,

and 1/2. The general solution is

(b) The system of equations is


Cl + C2 + C3 + C4 = -1

3
1
-4e - -C2 - -C3

+ -C4

232

16c
-64cl

=2

+ C2 + 9C3 + 4C4 =
27
8

- -C2'-

1
27

-C3

+ -C4
= O.
8

. a CAS we fin d Cl -- - 495


73 ' C2 -- 35'
109 C3 -- - 3726
257 . Th e so1u t'ron of th e
Usmg
385 ' an d C4 -- 45
initial-value problem is

191

Chapter 4 Review Exercises

36.

Consider xy"

+ y' =

O and look fr a solution of the form y = z?'.

'f

Substituting into the differential equation we have


xy"

+ y'

= m(m - l)xm-l

-1

+ mxm-l

= m2x.
-2

Thus, the general solution of xy"

+ y'

particular solution of xy"

+ y'

= O is Yc = Cl + C2lnx. To find a

-IX

we use variation of parameters.

The Wronskian is

Identifying f(x) = _x-l/2


,

x-l/2Inx

= y'Xlnx

l/x

so that
ul

= x3/2(

-5

we obtain

ul =

Then yp

-q

~~:I=~

w=l~

-3

= x 3/2(2-Inx
393

~ In x - ~) - ~x3/2In x

is y = Cl + c2Inx - ~x3/2

u; =

and

4)

- -

= - ~x3/2

and

u2

_x-l/2

= -VI,

s.]

= --x2

3/2 .

and the general solution of the differential equation

The initial conditions are y(l) = O and y'(l) = O. These imply that

Cl = ~ and C2= ~. The soIution of the initiaI-value problem is y = ~+ ~In x - ~x3/2.


37. From (D - 2)x

+ (D

- 2)y = 1 and Dx

+ (2D

- l)y = 3 we obtain (D - l)(D - 2)y = -6 and

Dx = 3 - (2D - l)y. Then

Substituting into (D - 2)x

+ (D

- 2)y = 1 gives C3= 5/2 so that


t
3 2t
x = -C2e - -Cle
2

38. From (D - 2)x - y

=t-

2 and -3x

x = cr e

+ (D

- 4)y

5
+ -.
2

= -4t we obtain (D - l)(D - 5)x = 9 - 8t. Then

8
3
+ C2e5t - -t
- 5

25

and
y = (D - 2)x - t
39. From (D - 2)x - y

+ 2 = -cle

5t

16

11

+ 3C2e + 25 + 25 t.

= _et and -3x + (D - 4)y = -7et we obtain (D - l)(D - 5)x = -4et so that

Then

192

Chapter 4 Review Exercises


40. From

= sin2t

(D+2)x+(D+l)y

and 5x+(D+3)y

= cos2t

Then
y
and

Cl

cos t
1

. t - 3'
2 cos 2t + 3'
7.sm 2t
+ C2 sin
1

x = -"5(D + 3)y + "5 cos 2t

193

we obtain (D2+5)y

= 2cos2t-7sin2t.

Modeling with Higher-Order


Differential

Equations

Exercises 5.1

----_

+ 16x = O we obtain

1. From ~x"

x = ct cos s)2 t

+ C2 sin s)2 t

so that the period of motion is 21f/s)2 = V21f/S seconds.


2. Frorn 20x"

+ kx

= O we obtain

2V'S0. t + C2 sin ~2V'S0. t

x = ct cos ~

and k = 320 N/m. If SOx" + 320x = O then x = ct cos 2t

so that the frequency 2/1f = ~[kf51f


C2 sin 2t

so that the frequency is 2/21f

3. Frorn ~x"

+ 72x = O, x(O) =

4. From ~x"

+ 72x

5. Frorn ix"

+ 40x = O, x(O) =

(a) x(1f/12)
(b) x'

= 1/1f vibrations/second.

-1/4, and x'(O)

= Owe obtain x = -~ cos4V6t.

== 2 we obtain x == ~ sin 4V6 t.

= O, x(O) = O, and x'(O)

1/2, and x'(O) == Owe obtain x

= ~cos St.

= -1/4, x(1f/S) = -1/2, x(1f/6) == -1/4, x(1f/S) = 1/2, x(91f/32) = V2/4.

= 4 ft/s directed downward.


= (2n + 1)1f/16 for n = 0,1,2, ....

-4 sin St so that x'(31f/16)

(e) If x = ~cosSt = Othen t


6. Frorn 50x"
7. From 20x"

+ 200x = O, x(O) = O, and x' (O) = -10 we obtain x = -5


+ 20x = O, x(O) = O, and x'(O) = -10 we obtain x = -10

sin 2t and x' = -10 cos 2t.


sin t and x'

= -lOcost.

(a) The 20 kg mass has the larger amplitude.

= -5V2 mis, x'(1f/2) = O mis; 50 kg: x'(1f/4) = O mis, x'(1f/2) = 10 mis


(e) If -5 sin 2t = -10 sin t then 2 sin t( cos t -1) = Oso that t = n1f for n = O, 1, 2, ... , placing both
(b) 20 kg: x'(1f/4)

masses at the equilibrium position.

The 50 kg mass is moving upward; the 20 kg mass is

moving upward when n is even and downward when n is odd,


8. Frorn x"

+ 16x =

O, x(O)

= -1,

and x'(O)

x = -cos4t

= -2

we obtain

J5

-'- 2'sin4t = Tcos(4t

194

- 3.6).

Exercises 5. 1
The period is 7f/2 seconds and the amplitude is .)5/2 feet. In 47f seconds it will make 8 complete
vbrations.
9. From

;tX" + x

= O, x(O) = 1/2, and x'(O) = 3/2 we obtain


x

10. From 1.6x"

+ 40x

= O, x(O)

= -1/3,

and x'(O)

x = - ~ cos 5t
3

If x = 5/24 then t

+ 200x

11. From 2x"

(a) x =

Vl3.sm(2t + 0.588).

3.

= "2 cos 2t + 4" sm 2t = -4= 5/4

we obtain

+ ~sin 5t = ~
4

12

sin(5t

+ 0.927).

t (~+ 0.927 + 2n7f) and t = t e: + 0.927 + 2n7f)

for n

= 0,1,2, ....

= -2/3, and x'(O) = 5 we obtain

= O, x(O)

-cos 10t + ~sin 10t = ~sin(10t

- 0.927).

(b) The amplitude is 5/6 ft and the period is 27f/10 = 7f/5

(e) 37f = 7fk/5 and k = 15 cycles.


(d) If x = O and the weight is moving downward for the second time, then 10t - 0.927 = 27f or
t=0.721s.

(e) If x' =

235

cos(10t - 0.927)

n = O, 1, 2, '"

= O then 10t - 0.927 = 7f/2 + n7f or t

= (2n

+ 1)7f/20 + 0.0927 for

(f) x(3) = -0.597 ft

(g) x'(3) = -5.814 ft/s


(h) x"(3) = 59.702 ft/s2
(i) If x

= O then

(j) If x = 5/12 then


(k)

If x = 5/12 and

12. From x"

+ 9x

+ n7f) for n = O, 1,2, ... and x'(t) = 2; ft/s.


t = ro (7f/6 + 0.927 + 2n7f) and t = lo (57f/6 + 0.927 + 2n7f) for n
x' < O then t = ro (57f/6 + 0.927 + 2n7f) for n = O, 1,2, ....

t = fo(0.927

= O, x(O) = -1, and x'(O)


X

and x' = 2V3 cos(3t


n = 1,2,3, ....
13. From k1 = 40 and k2

= - cos 3t -

+ 47f/3).

If x'

= -V3 we obtain
V3 sin
. 3t
""3

= 3 then t

2.

= V3 sin 37 +

= -77f /18

47f)
3

+ 2n7f/3

and t = -7f /2

+ 2n7f/3

for

= 120 we compute the efIective spring constant k = 4(40)(120)/160 = 120.

Now, m = 20/32 so k/m = 120(32)/20 = 192 and x"


obtain x(t)

= O, 1, 2, ....

+ 192x

= O. Using x(O) = O and x'(O) = 2 we

:.fJ sin 8V3 t.

14. Let m denote the mass in slugs of the first weight. Let k1 and k2 be the spring constants and
k = 4k1k2/(k1

+ k2)

the efIective spring constant of the system. Now, the numerical value of the

195

Exercises 5. 1
first weight is W

= mg = 32m, so
32m

= klG)

and

32m

= k2G)

From these equations we find 2k1 = 3k2. The given period of the combined system is 27r/w
so w

= 30.

Since the mass of an 8-pound weight is 1/4 slug, we have from w2


302

.s:
1/4

= 4k

or

= 7r/15,

= k/m

k = 225.

We now have the system of equations

Solving the second equation for k1 and substituting in the first equation, we obtain
4(3k2/2)k2
3k2/2 + k2

12k = 12k2
5k2
5

= 225.

Thus, k2 = 375/4 and k1 = 1125/8. Finally, the value of the first weight is
W =.32m = k1 = 1125/8
3
3

= 375 ~ 46.88 lb.


8

15. For large values of t the differential equation is approximated by x" = O. The solution of this
equation is the linear function x = clt

+ C2.

Thus, for large time, the restoring force will have

decayed to the point where the spring is incapable of returning the mass, and the spring will simply
keep on stretching.
16. As t beco mes larger the spring constant increases; that is, the spring is stiffening. It would seem
that the oscllations would become periodic and the spring would oscllate more rapidly. It is likely
that the amplitudes of the oscillations would decrease as t increases.
17. (a) above

(b) heading upward

18. (a) below

(b) from rest

19. (a) below

(b) heading upward

20. (a) above

(b) heading downward

iX" + x' + 2x = O, x(O) = -1, and X'(O) = 8 we obtain x = 4te-4t - e-4t and x' =
8e-4t - 16te-4t. If x = O then t = 1/4 second. If x' = O then t = 1/2 second and the extreme
displacement is x = e-2 feet.

21. From

196

Exercises 5. 1

= O, x(O) = O, and XI(O) = 5 we obtain x = 5te-2v'2t and


x' = 5e-2v'2t (1- 2V2t).
If Xl = O then t = V2/4 seeond and the extreme displacement is

22. From ~x" + V2XI + 2x


X = 5V2e-1/4

feet.

23.

(a) From

24.

(b) From x" + x' +


(a) x = ~e-8t ( 4t

x" + 10xl + 16x


16x

= O, x(O) = 1, and XI(O) = O we obtain x = 1e-2t - ~e-8t.

= O, x(O) = 1, and XI(O) = -12 then x = _e-2t

- 1) is never zero; the extreme displacernent is x(O)

+ ie-8t.

= 1 meter.

~e-8t (5 - 2t) = O when t = ~In a ::::: 0.153 seeond; if x' = te-8t (t -10)
t = ~In 10 :::::0.384 seeond and the extreme displacernent is x = -0.232 meter.

(b) x =
25.

(a) From
(b) x

O.lx" + O.4xl + 2x

= e-2t

V; [- jg

eos4t - ~

(e) If x = O then 4t + 4.25


t = 1.294 seeonds.
26.

(a) From
(b) x

= O, x(O) = -1, and x' (O) = O we obtain


sin4t] =

= 27r,37r, 47r, ...

V;

(~

eos4t + ~

e-2t [- eos 4t - ~ sin 4t].

e-2t sin(4t + 4.25).

so that the first time heading upward is

ix" + x' + 5x = O, x(O) = 1/2, and Xl (O)

= e-2t ~

= O then

sin4t) = ~e-2t

= 1 we obtain x
sin (4t +

= e-2t (~eos 4t + ~ sin 4t).

).

(e) If x = O then 4t+7r /4 = 7r,27r,37r, ... so that the times heading downward are t = (7 +8n)7r /16
for n=O, 1, 2, ....
(d)

1
\
\
\

.5

,,

--_-----"'2
-.5

/
/

-1

27. From

156 x"

+ 3xl+ 5x = Owe find that the roots of the auxiliary equation are m = - ~13 h/432 - 25 .

> Othen 13> 5/2.


(b) If 4132- 25 = O then 13= 5/2.
Ce) If 4132- 25 < Othen O < 13< 5/2.
(a) If 4132- 25

197

Exercises 5. 1
28. From 0.75x"
m=

+ (3x! + 6x = Oand

- ~(3 ~

(3

18 and

(32 -

= e-2(3t/3 (C1 cosh ~J(32 -

If x(O) = O and x'(O) = -2 then


29. If ~x"

+ ~x' + 6x

C1 = O and C2 =

= 10cos3t, x(O)
Xc

and xp =

13 ( cos 3t

> 3vZ we find that the roots of the auxiliary equation are

+ sin 3t)

18t

+ C2 sinh ~J(32

-3/

(32 -

18.

(V47

= e-t/2 C1 cos -2- t

V47 t )
+ C2 sin -2-

so that the equation of motion is


10
+ -(cos3t

.
+ sm3t).

= 12cos 2t+3 sin 2t, x(O) = -1, and x'(O) = 5 then

Xc

= e-t(c1 cos 2t+C2 sin 2t)

= 3 sin 2t so that the equation of motion is

and xp

x = e-t cos 2t
x

(b)

18t] .

= -2, and x'(O) = Othen

V47 t - --64. sm --V47)t


x = e-t /2 (4-- cos -3
2
3V47
2
30. (a) If x" +2x' +5x

+ 3 sin 2t.
x

(e)

steady-state

-3 transJ.ent
.

-3

31. From x" + 8x' + 16x = 8 sin 4t, x(O) = O, and x' (O)
xp = - cos 4t so that the equation of motion is

= O we obtain

Xc

= C1e-4t + c2te-4t and

C1 e-4t

1
x = _e-4t

32.

+ te-4t

As t

-> 00

= 625e-4t(24 + 100t) - 625e-t(24cos4t + 7sin4t).

the displacement x

->

O.

33. From 2x" + 32x = 68e-2t cos 4t, x(O)


xp = ~e-2t cos4t - 2e-2t sin4t so that
x

and

1
- - cos4t.

From x" + 8x' + 16x = e-tsin4t, x(O) = O, and x'(O) = O we obtain


24 -t cos 4t - 62Se
7 -t sin
. 4t so th a t
xp = -625e
x

+ c2te-4t

Xc

= O, and x'(O) = O we obtain x =

1
-21 cos 4t + 49 sin 4t + 2e-2t

198

cos 4t - 2e-2t sin 4t.

C1 cos 4t

+ C2 sin 4t

and

Exercises 5. 1
34. Since x = ~ sin(4t - 0.219) -

sin(4t - 2.897), the amplitude approaches v85/4 as t _, oo.

~e-2t

35. (a) By Hooke's law the external force is F(t) = kh(t) so that mx" + f3x' + kx = kh(t).
(b)

From ~x" + 2x' + 4x


and xp

= 20 cos t, x(O) = O, and x'(O) = O we obtain

e-2t(

el cos 2t + e2sin 2t)

= ~~cos t + ~ sin t so that


x = e-

36. (a) From 100x" + 1600x


and xp =

2 (
t

56
72)
-13 cos 2t - 13 sin 2t

56
32 .
+ 13 cos t + 13 SIn t.

= 1600sin8t, x(O) = O, and x'(O) = O we obtain

Xc

= el cos4t + e2sin4t

-l sin 8t so that
x

(b) If x

Xc

. 4t - -l. SIn 8t.


= -2 SIn
3

= lsin4t(2 - 2cos4t) = Othen t = mr/4 for n = 0,1,2, ....

(e) If x' = ~ cos 4t- ~ cos 8t = ~(l-cos 4t)(1 +2 cos 4t) = Othen t = 1f /3+mr /2 and t = 1f /6+n1f /2
for n = O, 1, 2, ... at the extreme values. Note: There are many other values of t for which

x' = O.
(d ) x(1f/6 + n1f/2) = V3/2 cm. and x(1f/3 + n1f/2) = -V3/2cm.
(e)

x
1

37. From x" + 4x

= -5 sin 2t + 3 cos 2t, x(O) = -1, and x'(O) = 1 we obtain

Xc

= el cos 2t + e2sin 2t,

xp = ~t sin 2t + ~t cos 2t, and


3.
x = - cos 2t - -1 sin 2t + -t
844

SIn

38. From x" +9x = 5 sin3t, x(O) = 2, and x'(O) = Owe obtain

2t + -t
5 cos 2t.
Xc

= el cos3t+e2 sin3t,

xp = -~tcos3t,

and
5.

x = 2 cos 3t + 18 SIn 3t - 6't cos 3t.


39. (a) Frorn
xp =

x'' + w2x = Focos"'(t, x(O) = O, and x'(O)


(Fo cos "'(t)/

(w2

= O we obtain

",(2)so that

x =

Fo

w -"'(

2 cos wt +

199

Fo

w -"'(

2 cos "'(t.

Xc

= el coswt + e2sinwt and

Exercises 5. 1

(b)

lim
"(-+w

Fo

2
2(cos 1t - coswt) = lim
w - 1
"(-+w

40. From x" + w2x = Focoswt,


xp

= (Fot/2w)sinwt

- Fot sin ryt


-

21

Fo

= -t sinwt.
2w

x(O) = 0, and x'(O) =

= (Fot/2w)sinwt

so that x

we obtain Xc = Cl coswt + C2sinwt and

and

lim FOtsinwt = FOtsinwt.


2w
2w
41. (a) From cos( u - v) = cos u cos v + sin u sin v and cos( u + v) = cos u cos v - sin u sin v we obtain
"(-+w

sin u sin v = ~[cos(u - v) - cos(u + v)]. Letting u

~(1 - w)t and v = ~(1 + w)t, the result

follows.
(b) If 10 = 1(1 - w) then 1

;::::o

w so that x = (Fo/2q) sin l sin 1t.

42. See the article "Distinguished Oscillations of a Forced Harmonic Oscillator" by T.C. Procter in The
College Mathematics Journal, March, 1995. In this article the author illustrates that for Fo = 1,
>. = 0.01, 1

22/9, and w

2 the system exhibits beats oscillations on the interval [0,97r], but

that this phenomenon is transient as t .......oo.

-1

43. (a) The general solution of the homogeneous equation is

xc(t) = cle-,\t cos( jw2 - >.2t) + c2e-,\t sin( jw2 - >.2t)


= Ae-,\t

where A

sin[jw2 - >.2t + 4>],

V cy + c , sin 4> = ci] A, and cos 4> = C2/A.

Fo(w2 - 12).
xp(t) = (w2-12)2+4>.212

sm1t+

Now

Fo( -2>'1)
(W2-12)2+4>.212

where

Fo( -2>'1)

and

200

.
COS1t= Asm(1t+B),

Exercises 5. 1

= O then v

+ 2).2 -

= O so that. v = Oor , = VW2 - 2).2. The first derivative


test shows that 9 has a maximum value at , = vw2 - 2).2. The maximum value of gis

(b) If g'(I)

(,2

W2)

,1 = vw
and the resonance curve grows without bound at ,1 =

(c) We identify w2

= k/m =

4, ).

{3/2, and

2 - 2).2

= J4 -

{32/2. As {3---; O,

,1 ---;

2. That is, the system approaches pure

resonance.

(3

yI

2.00
1. 00
0.75
0.50
0.25

1. 41
1. 87
1. 93
1. 97
1. 99

0.58
1. 03
1. 36
2.02
4.01

,t

44. (a) For n = 2, sin2


when

'1 = w /2.

1(1- cos2,t).

The system is in pure resonance when 2,d27r = w/27r, or

(b) Note that

,t

sin3

= sin ,t sin2 ,t

= '2 [sin ,t - sin ,t cos 2,t].

Now
sin(A

+ B) + sin(A

- B) = 2sinAcosB

so
sin ,t cos 2,t = ~[sin 3,t - sin ,t]
and
sin3

,t

'4 sin ,t

'4 sin 3,t.

Thus
x"

+ w 23.
x = '4 sm v t - 'l'3t
4 sm

,.

The frequency of free vibration is w/27r. Thus, when ,d27r = w/27r or


3'2/27r = w /27r or 3,2 = w or

,3

,1

= w /3, the system will be in pure resonance.

201

= w, and when

Exercises
(e)

5. 1
yl=1/2

yl=l

y2=1I3

10

n=3

10

-5
-10

-10

+ 2q' + 100q = O we obtain q(t) = e-20t(cr cos 40t + C2 sin 40t). The initial conditions
q(O) = 5 and q'(O) = O imply cr = 5 and C2 = 5/2. Thus

45. Solving 2~q"

q(t)

= e-20t (5cos40t+

~sin40t)

:::::)25

+ 25/4e-20tsin(40t

+ 1.1071)

and q(O.Ol) :::::4.5676 coulombs. The charge is zero Ior the first time when 40t

+ 0.4636 =

7f

or

t::::: 0.0509 second.

46. Solving

h" + 20q' + 300q


= O imply

and q'(O)

= O we obtain q(t) = ce-2t + C2e-60t. The initial conditions q(O) = 4


= 6 and C2 = -2. Thus
q(t) = 6e-20t _ 2e-60t.

= O we find Ot = 1/3 which implies t < O. Therefore the charge is never O.


Solving %q" + 10q' + 30q = 300 we obtain q( t) = e-3t (C cos 3t + C2 sin 3t) + 10. The initial
Setting q

47.

q(O) = q'(O) = O imply

conditions

= C2 = -10. Thus

q(t)=10-10e-3t(cos3t+sin3t)

and

i(t)=60e3tsin3t.

Solving i( t) = O we see that the maximum charge occurs when t

= 7f /3

and q( 7f /3) :::::10.432.

+ 100q' + 2500q = 30 we obtain q(t) = ce-50t + c2te-50t + 0.012. The initial conditions
q(O) = O and q'(O) = 2 imply C = -0.012 and C2 = 1.4. Thus

48. Solving q"

q(t) = -0.012e-50t

+ 1.4te-50t + 0.012 and i(t)

Solving i(t) = Owe see that the maximum charge occurs when t

= 2e-50t
=

- 70te-50t

1/35 and q(1/35) :::::0.01871.

+ 2q' + 4q = Owe obtain Yc = e-t (cos v'3 t + sin v'3 t). The steady-state charge has the
form Yp = A cos t + B sin t. Substituting into the differential equation we find

49. Solving q"

(3A
Thus, A

+ 2B) cos t + (3B - 2A) sin t = 50 cos t.

= 150/13 and B = 100/13. The steady-state charge is


qp(t)

150

100

13cos t + 13 sin t

and the steady-state current is

202

Exercises 5. 1
50. From

Ea
ip(t) = Z
and Z

= JX2 + R2

(RZ sn-rz - ZX cos,t )

we see that the amplitude of ip(t) is

A = jE'6R2
E'6X2 = Ea. /R2 X2 = Ea
\ Z4 + Z4
Z2 V
+
Z .
51. Thc differential equation is ~q" + 20q' + 1000q

= 100sin t. To use Example 10 in the text we identify

Ea = 100 and , = 60. Then


111
X = L, - ~ = 2(60) - 0.001(60) ~ 13.3333,

Z = ,j X2 + R2 =

,j X2 + 400 ~

24.0370,

and

Eo = 100 ~ 4 1603.

From Problem 50, then

+ cp)

ip(t) ~ 4.1603(60t

where sin

angle. Now

-X/Z

and coscp = R/Z. Thus tan

~ -0.6667 and cp is a fourth quadrant

-X/R

cp ~ -0.5880 and
ip(t) ~ 4.160~(60t - 0.5880).

52. Solving

h" + 20q' + 1000q = O we obtain

qc(t) = (ci cos 40t

+ B cos 60t + e sin 40t + Deos

has the form qp(t) = A sin 60t

+ C2 sin 40t).

The steady-state charge

40t. Substituting

into the differential

equation we find
(-1600A

- 2400B) sin 60t

+ (400C
= 200 sin 60t

Equating coefficients we obtain A

+ (2400A

- 1600B) cos 60t

- 1600D) sin 40t

+ (1600C + 400D)

cos40t

+ 400 cos 40t.

-1/26,

-3/52,

e = 4/17,

and D

state charge is
qp(t)

1
3
26 sin 60t - 52 cos 60t

=-

4
17 sin 40t

1
17 cos 40t

and the steady-state current is


30
cos60t
13

ip(t) = --

45
sin60t
13

+-

203

160
40
cos40t - - sin40t.
17
17

+-

1/17. The steady-

Exercses 5. 1
53. Solving ~q"

+ 10q' + 100q =

150 we obtain q(t) = e-lOt(cl cos 10t

conditions q(O) = 1 and q'(O) = O imply Cl = C2= -1/2.

+ C2sin 10t) + 3/2-.

The initial

Thus

q(t) = -~e-lOt(cos 10t + sin 10t) + ~.


2

As t

-+ 00,

q(t)

-+

3/2.

where Z = JX2 + R2 and


X = L"( - l/C"(. Since Eo is constant the amplitude will be a maximum when Z is a minimum.
Since R is constant, Z will be a minimum when X = O. Solving L"( - l/C"( = O for "( we obtain
"(= 1/..JLC. The maximum amplitude will be Eo/ R.

54. By Problem 50 the amplitude of the steady-state current is Eo/Z,

55. By Problem 50 the amplitude of the steady-state current is Eo/Z,


X

L"( -

l/C"(.

+ R2

and

Since Eo is constant the amplitude will be a maximum when Z is a minimum.

Since R is constant, Z will be a mnimum when X.


C

where Z = JX2

O. Solving L"( - 1/ c"( = O for C we obtain

= 1/L,,(2.

56. Solving O.lq"

+ 10q =

Asin"(t + Bcos"(t.

100sin"(t we obtain q(t)

c cos

10t + C2sin 10t + qp(t) where qp(t) =

Substituting qp(t) into the differential equation we find


(100 - "(2)A sin "(t + (100 - "(2)B cos "(t = 100sin "(t.

Equating coefficients we obtain A


initial conditions q(O)

= 100/(100

= q'(O) = O imply
q(t) =

- ,,(2)and B

Cl = O and C2=

= O. Thus, qp(t) =

100 2 sin "(t. The


100 - "(
-10"(/(100 - "(2). The charge is

102(lOsin"(t-"(sin10t)
100 - "(

and the current is

i(t) =

100"(
2(cos"(t-coslOt).
100 - "(

57. In an L-C series circuit there is no resistor, so the differential equation is


d2q

L dt2 + C q = E(t).
Then q(t) = Clcos (t/..JLC) +C2 sin (t/..JLC)
qp(t) into the differential equation we find

(~-

+qp(t) where qp(t) = A sin "(t+ B cos "(t. Substituting

L"(2) Asin"(t + (~ - L"(2) Bcos"(t = Eocos"(t.

Equating coefficients we obtain A = O and B

= EoC/(l - LC"(2). Thus, the charge is

1
1
EoC
q(t) = Cl COS ..JLC t + C2sin ..JLC t + 1 _ LC"(2 cos vr.

204

Exercises
The initial conditions q(O) = qo and q'(O) = io imply
The current is

Cl

= qo - EoC/(l-

.
Cl.
1
C2
1
EoC,.
2(t) = - ..JLC sin ..JLC t + ..JLC cos ..JLC t - 1 _ LC,2

= 20 COS ..JLC t -..JLC

EOC).
qO- 1_ LC,2

SIn

LC,2) and

t - 1_

58. When the circuit is in resonance the form of qp(t) is qp(t) = At cos kt

= io..JLC.

,t
EoC,.
LC,2

1
SIn..;re

C2

5.2

+ Bt

SIn

,t.

sin kt where k = 1/..JLC .

Substituting qp(t) into the differential equation we find


2
Eo
q~ + k q = -2kA sin kt + 2kB cos kt = Leos kt.

Equating coefficients we obtain A

= O and B

= Eo/2kL.

The charge is

q(t) = C! cos kt + C2 sin kt + 2~~ t sin kt.


The initial conditions q(O) = qo and q'(O) = io irnply C!

= qo and

C2

= io/k. The current is

i(t) = -ek sin kt + C2kcos kt + 2~~ (kt cos kt + sin kt)


EO
) ..
Eo
= ( 2kL - qok SIn kt + 20 cos kt + 2L t cos kt.

Exercises 5.2

1. (a) The general solution is


(
YX)=C!+C2X+C3X

The boundary conditions are y(O)


conditions give C! = Oand

C2 =

Wo

+C4X + 24EIX .

= O, y'(O) = O, y"(L) = O, y"'(L) = O. The first two

O. The conditions at x = L give the system


Wo

2C3+ 6C4L + 2EIL

=O

WO

6C4+ EIL = O.
Solving, we obtain

C3

= woL2/4EI

and

C4

= -woL/6EI.

y(x) = ~(6L2x2
24EI

205

The defiection is

- 4Lx3 + x4).

Exercises 5.2
(b)

~~------------x
20.40.60.81
1

2. (a) The general solution is


(
YX)=e+C2X+C3X

+C4X

Wo

+ 24EIX

The boundary conditions are y(O) = O,y"(O) = O,y(L) = O,y"(L)


give e

= O and

C3

= O. .The conditions at x

C2L

C2

Wo

Wo

+ C4L + 24EIL
6C4L

Solving, we obtain

= L give the

= WOL3/24EI

and

C4

+ 2EIL

system

=O
= O.

= -woL/12EI.

y(x) = 2:;I(L3x

- 2Lx3

= O. The first two conditions

The defiection is

+ x4).

(b)
+-------------~x
0.2 0.4 0.6 0.8

1 y

3. (a) The general soluton is


2

y (x ) = C + C2X + C3X

Wo

+ C4X + 24EIX

The boundary conditions are y(O) = O,y'(O) = O,y(L) = O,y"(L)


give Cl = O and C2 = O. The conditions at x = L give the system
C3L2

+ C4L3 + 2:;IL4
Wo

2C3

+ 6C4L + 2EIL
206

=O
2

= O.

= O. The first two condtons

Exercises 5.2
Solving, we obtain

C3

= woL2/16EJ

and

C4

= -5woL/48EJ.

y(x) = ~(3L2x2
48EJ

- 5Lx3

The defiection is

+ 2x4).

(b)

4. (a) The general solution is


y(x)

Cl

el

o, y'(O)

1r

sm LX'

= O,y"(L) = O. The first two conditions


= Oand e2 = -woL3 / EJ1r3. The conditions at X = L give the system

The boundary conditions are y(O) =


give

wOL4.

+ C2X + C3X + C4X + EJ1r4

C3L

Solving, we obtain

C3

wO

+ C4L + El7r3L

= 3woL2/2El7r3

woL
y(x) = 2El7r3

= O,y(L)

(2-2L

and
X

=O

2e3

+ 6C4L = O.

C4

= -woL/2El7r3.

+ 3Lx

The defiection is

2 - X 3 + ----;2L3
1r)
sin LX

(b)

1 Y
(e) Using a CAS we find the maximum defiection to be 0.270806 when
5. (a) The general solution is

207

= 0.572536.

Exercises 5.2
The boundary conditions are y(O) = o, y"(O) = O, y(L)
give C1= Oand

C3

= O,y"(L) = O. The

first two conditions

= O. The conditions at x = L give the system


3
Wo C2L + C4L + 120EIL" = O

Wo
6C4L + 6EIL
Solving, we obtain

C2

= 7woL4/360EI

and

C4

= O.

= -woL2/36EI.

The defiection is

y(x) = 36~~I(7L'x - lOL2x3 + 3x5).


(b)

(e) Using a CAS we find the maximum defiection to be 0.234799 when x = 0.51933.
6.

(a) Ymax

wOL4

= y(L) = 8EI

(b) Replacing both L and x by L/2 in y(x) we obtain woL4/128EI,

which is 1/16 ofthe maximum

defiection when the length of the beam is L.

5woL4
(e) Ymax = y(L/2) = 384EI
(d) The maximum defiection in Example 1 is y(L/2) = (wo/24EI)L4/16
1/5 of the maximum displacement of the beam in Problem 2.

WOL4/384EI, which is

7. The general solution of the differential equation is

{P
(P Wo 2 woEI
El x + C2 sinh V El x + 2P x + ---;r .

y = C1cosh V
Setting y(O) = O we obtain

Cl

-woEI / p2, so that

woEI
{P
. (P wo 2 woEI
y = ----;r cosh V El x + C2 sinh V El x + 2P x + ---;r'
Setting y' (L) = O we find
C2

(fu

w~~I sinh

fu

L - W;L) /

208

fu fu
cosh

L.

Exercises 5.2
8. The general solution of the differential equation is
y

/P

. /P

Wo

woEI

= el cos V El x + e2sin V El x + 2P x + ---p2 .

Setting y(O) = O we obtain el = -woEI / p2, so that

/P
. /P
El x + cz sin V El x +

'WoEI

y = - ---:>2 cos V
Setting y' (L)

:s: O the

'WoEI

= O we find
e2 =

9. For A

Wo

2P x + --p2 .

/P woEI. /P WOL)
(- V El ----p2 Slll V El L - P

only solution of the boundary-value

V/P
El cos V/P
El L.

problem is y = O. For A

> O we have

y = el cos V>-x + e2sin V>-x.


Now y(O) = O implies el = O, so

gives
V>-n=nn

or

A=n2,

n=1,2,3,

....

The eigenvalues n2 correspond to the eigenfunctions sin nx for n = 1, 2, 3, ....


10. For A :::;O the only solution of the boundary-value problem is y

= O. For A > O we have

y = el cos V>-x + e2sin V>-x.


Now y(O)

= O implies el = O, so
y

() = e2sin V>- = O

gives
,n

VA=nn

or

A=16n,

n=1,2,3,

....

The eigenvalues 16n2 correspond to the eigenfunctions sin 4nx for n = 1, 2, 3, ....
11. For A

:s: O the

only solution of the boundary-value problem is y = O. For A


y = el cos V>-x + e2sin V>-x.

Now

y'(x) = -el V>- sinV>-x+e2V>-cosV>-X


and y'(O) = O implies e2 = O, so
y( L) = el cos V>-L = O

209

> Owe have

Exercises 5.2
gives
r:
VA

L =

(2n - 1)7f
2
or

A=

(2n - 1)27f2

' n = 1, 2, 3, ....

4L2

.
2
.
(2n - 1)7f
The eigenvalues (2n - 1) 7f2/4L2 correspond to the eigenfunctions cos
L
x for
n

= 1,2,3, '" .

12. For A:::; O the only solution of the boundary-value problem is y


y

Now y(O)

= O. For A > O we have

= el cos,f).. x + e2sin ,f).. x.

= O implies CI = O, so
y' (~)

= e2,f).. cos,f)..

=O

gives
,f).. ~ = (2n;

1)7f or

A = (2n - 1)2, n = 1, 2, 3, ....

The eigenvalues (2n - 1)2 correspond to the eigenfunctions sin(2n - l)x.

< O the only solution of the boundary-value problem is y = O. For A = O we have y = CIX +cz.
= CI and y'(O) = O implies el = O. Then y = e2 and y'(7f) = O. Thus, A = O is an eigenvalue
with corresponding eigenfunction y = 1.

13. For A

Now y'

For A > O we have


y

= el cos,f).. x + e2sin ,f).. x.

Now
y' (x) = -el ,f)..sin ,f)..x
and y'(O)

+ e2,f).. cos ,f).. x

= O implies e2 = O, so

gives

J>: 7f = n7f

or

A = n2, n

= 1, 2,

3, .. , .

The eigenvalues n2 correspond to the eigenfunctions cos nx for n

= O, 1, 2, ....

14. For A :::;O the only solution of the boundary-value problem is y = O. For A > O we have
y

= el cos J>: x + cz sin J>: x.

Now y( -7f) = Y(7f) = O implies


el cos J>: n - e2sin ,f).. n = O
el cos J>: n + e2sin ,f).. n = O.

210

(1)

Exercises

5.2

This homogeneous system will have a nontrivial solution when


cos ~ tt
1\
I cos V A tt

sin ~ tt I
.
1\
= 2 sin ~
sin V A tt

1\

tt COS V

1\

A tt = sin 2v A tt = O.

Then
n2
A = 4;

or

n = 1, 2, 3, ....

When n = 2k - 1 is odd, the eigenvalues are (2k - 1)2/4. Since cos(2k - 1)7f/2 = O and
sin(2k - 1)7f/2 =1- O, we see from either equation in (1) that C2 = O. Thus, the eigenfunctions
corresponding to the eigenvalues (2k - 1f /4 are y = cos(2k - l)x /2 for k = 1, 2, 3, ....

Similarly,

when n = 2k is even, the eigenvalues are k2 with corresponding eigenfunctions y = sin kx for
k

= 1, 2, 3, ....

15. The auxiliary equation has solutions

m=~(-2J4-4(A+1))
For A

=-1!=1.

< O we have
y = e -x

(q cosh R x + C2sinh R x) .

The boundary conditions imply


y(O)

y(5) =

ci

sinh

C2e-5

5R =

so C1= C2= O and the only solution of the boundary-value problem is y = O.


For A = O we have

and the only solution of the boundary-value problem is y = O.


For A

> O we have
y = e -x

(C1

COS ~

or

A =

X + C2sin ~

x) .

Now y(O) = O implies q = O, so


gives
n27f2

5~
The eigenvalues n27f2/25
16. For A

= nit

---:25'

n = 1,2,3, ....

n7f
correspond to the eigenfunctions e-x sin 5x for n

< -1 the only solution of the boundary-value problem is

y = O. For A = -1 we have

y = CIX+ C2. Now y' = C1and y'(O) = O implies C1= O. Then y = C2and y'(l)
is an eigenvalue with corresponding eigenfunction y

211

1.

1, 2, 3, ....

= O. Thus, A = -1

Exercises 5.2
For A

> -1 we.have
y = ClCOS~x+c2sin~x.

Now
y' = -cl~sin~x+c2Jr+lcosJr+lx
and y'(O) = O implies C2= O, so
y'(l)

= -clVI+!sinJr+l

gives
~

= n1f

or

/\ = n21f2 -1,

n = 1,2,3, ....

The eigenvalues n21f2 - 1 correspond to the eigenfunctions cos titcx for n = O, 1, 2, ....
17. For A = O the only solution of the boundary-value problem is y = O. For A

=f. O we have

y = ClCOSAX + C2sin AX.


Now y(O)

=O

implies Cl = O, so
y(L) = c2sinAL

gives
AL = tit: or

A=

=O

n1f

L'

ti

= 1,2,3, ....

The eigenvalues n1f/ L correspond to the eigenfunctions sin


18. For A = O the only solution of the boundary-value

n1f
LX

for

ti

= 1, 2, 3, ....

problem is y = O. For A

=f. O we have

y = ClCOSAX + C2sin AX.


Now y(O) = O implies Cl = O, so
gives
.
(2n - 1)1f
31fA=

or

A=--,

2n - 1

n=l,2,3,

The eigenvalues (2n - 1)/6 correspond to the eigenfunctions sin 2n;


19. For A

....

1x for n = 1, 2, 3, ....

> O a general solution of the given differential equation is


y

= q cos(~

lnx) + c2sin(~

lnx).

Since In 1 = O, the boundary condition y(l) = O implies Cl = O. Therefore


y = c2sin(~
Using In e"

= 1f we find that y (e")

lnx).

= O implies

C2Sin~1f

212

=O

Exercises 5.2
or v0."'Jr= n'Jr, n = 1, 2, 3, ....

The eigenvalues and eigenfunctions are, in turn,

A = n2,

= 1, 2, 3,

...

and

For A ::; O the only solution of the boundary-value

= sin (n In x).

problem is y = O.

To obtain the self-adjoint form we note that the integrating factor is (1/x2)ef

dx]

l/x.

That is,

the self-adjoint form is


A = O.
- d [xy '] + -y
dx
x
Identifying the weight function p(x) = l/x we can then write the orthogonality relation

r
)1
20.

e"

1
-sin(nlnx)sin(mlnx)dx
x

For A = O the general solution is y

m =J n.

Cl + C2lnx. Now y' = C2/ z, so y' (e-l)

C2e

O implies

= O gives Cl = O. Thus y(x) = O.

C2= O. Then y = Cl and y(l)


For A < O, y

= O,

= C1X-vY + C2XvY.

The initial conditions give C2 = cle2vY

and Cl = O, so that

C2= O and y (x) = O.


For A > O, y

= Clcos( v0."In x) + C2sin( v0."In x). From y(l)

y = c2sin(v0."lnx).

Now y'

c2(v0."/x)cos(v0."lnx),

cos v0."= O or A = (2n - 1)2'Jr2/4for n

so y'(e-l)

1, 2, 3, ....

= c2ev0."cosv0." = O implies

The corresponding eigenfunctions are

2n - 1
y = sin ( -2-'Jrlnx
21. For A = O the general solution is y = el +C2Inx.

= O we obtain Cl = O and

Now y' = C2/X, so y'(l)

= C2 = O and y

el.

Since y'(e2) = O for any Cl we see that y(x) = 1 is an eigenfunction corresponding to the eigenvalue
A = O.
For A < O, y

= C1X-vY + C2XvY. The initial conditions imply el = C2= O, so y(x) = O.


For A > O, y = Clcos( v0."In x) + C2sin] v0."In x). Now
,

~.

Y = -cl-sm(VA

and y'(l)

= e2v0." = O implies C2 = O.

A = n2'Jr2/4 for n = 1, 2, 3, ....

r:

v0."
r:
lnx) +C2-COS(VA lnx),
x

Finally,

y'(e2)

= -(elv0."/e2)sin(2v0.")

The corresponding eigenfunctions are


y = cos (~'Jr In x ) .

22.

For A> 1/4 a general solution of the given differential equation is


y

C1X-

1/'">

-cos

(/4A 2 - 1)lnx

+ C2X- 1/2 sin

213

(/2

1 lnx. )

O implies

Exercises 5.2
Since In 1 = O, the boundary condition Y(l) = Oimplies cr= O. Therefore
y = C2X-l/2sin (~

lnx).

Using In e2 = 2 we find that y( e2) = O implies


c2e-lsin(V4A-1)
or

J4I-=1 = n-lr, n = 1, 2, 3, ....


A=

(n27r2 + 1) /4,

=0

The eigenvalues and eigenfunctions are, in turn,


n = 1,2,3,

...

For A < O the only solution of the boundary-value

and

= x-l/2sin

(n27rlnx).

problem is y = O.

For A = 1/4 a general solution of the differential equation is


y = ClX-l/2
From y(l)

O we obtain Cl

+ C2x-l/2In

x.

= C2x-l/2In x. From y(e2) = O we obtain 2c2e-l = O or

O, so y

C2= O. Thus, there are no eigenvalues and eigenfunctions in this case.


To obtain the self-adjoint form we note that the integrating factor is

= (1/x2).

(1/x2)eJ(2/x)dx

x2 = 1.

That is, the self-adjoint form is


d~

[x2y'] + Ay

= O.

Identifying the weight function p(x) = 1 we can then write the orthogonality relation

le
or

i .x-l/2 sin

(~7r

he x-l sin (~7r

In

x) x-l/2 sin (n;

lnx) sin

lnx)

(n27rlnx) dx = O,

23. If restraints are put on the column at x = L/4, x = L/2, and x

criticalload

dx = O,
m

m i=

n,

i= n.

= 3L/4, then the

will be P4.

24. (a) The general solution of the dfferential equation is

= Cl COS

ffi

+ C2sin

214

ffi

+ O.

Exercises 5.2

= o, the initial conditions are y(O) = y'(O) = O. If = O


this implies that CI = C2= O and y(x) = O. That is, there is no deflection.
Since the column is embedded at x
(b)

If =1- O, the initial conditions give, in turn,

q =

- and C2= O. Then

In order to satisfy the condition y(L) = we must have


or
This gives )P/El

L = mr/2 for n = 1, 2, 3, ....

cos

--:;

L = O.

The smallest value of Pn, the Euler load, is

then

25. The general solution is

I"'if-

y = CICOS
From y(O)

+ C2sin

I"'if-

= O we obtain CI = O. Setting y(L) = O we find )pw2/T

critical speeds are

Wn

= mrVT /

Lv-P,

n = 1, 2, 3, ....
n1l'

y (x) = C2sin T x,

x.

L = mr, n = 1, 2, 3, ....

n = 1, 2, 3, .. . ,

where C2=1- O.
26. (a) When T(x) = x2 the given differential equation is the Cauchy-Euler equation

+ 2xy' + pw2y = O.

x2y"

The solutions of the auxiliary equation


m( m - 1)

+ 2m + pw2 = m2 + m + pw2 = O

are
mI =

/
-"21 - "21y4pw

- h,

when pw2 > 0.25. Thus


y = CIX-I/2 COS(A
In x)
where A = )4pw2

+ C2X-I/2sin(A In x)

1/2. Applying y(l) = O gives CI = O and consequently


y = C2X-I/2sin(Alnx).

215

Thus,

The corresponding deflection curves are

Exercises 5.2

An

y(e)

The condition

= n7r, n = 1, 2,3,

O requires

....

O. We obtain

a nontrivial

solution

when

But

An
Solving for

c2e-l/2sinA

= J4pw~

- 1j2

= n7r.

gives

Wn

The corresponding

solutions

are

Yn(x) = C2X-1/2 sin(n7r lnx).


(b)

1
n=l

n=3
x

eX

-1

27. The auxiliary

1
n=2

conditions

-1

equation

is m2

u(a) = Uo and u(b)

+m =
= Ul

-1

+ 1) =

m(m

O so that

+C2

yield the system cla-l

_ (UO - Ul) b
b
a
- a

Cl -

an

u(r)

= uo,

+ C2.

clr-1

The boundary

= Ul.

clb-l +C2

Solving gives

_ ulb - uoa
b
.
- a

C2 -

Thus

UO - Ul) ab
ulb - uoa
ur() = (
-+
b-a
r
b-a
28.

The auxiliary

equation

is m2 = O so that u(r)

and u(b) = Ul yield the system Cl

+ C21n a

+ C21n r. The
Cl + C21n b = Ul.

= Cl

= uo,

- uolnb
ln(ajb)

ullna
C =

Thus

u(r) = ullna - Uo lnb


ln(ajb)

+ Uo -

and

.
boundary

conditions

u(a) ~ uo

Solving gives

Uo - Ul
C2= ln(ajb)'

Ul 1
_ Uo ln(rjb) - ulln(rja)
nr ln(ajb)

ln(ajb)

29. (a) The general solution ofthe differential equation is y = Cl cOS4X+C2 sin4x.

From yo = y(O) = Cl

+ c2sin4x.

any solution must

we see that y
satisfy yo

= yocos4x

Yl We also see that when

boundary-value

= Y(7rj2) = yO we see that


yO = Yl, Y = yO cos 4x + C2sin 4x

From Yl

problem for any choice of C2. Thus, the boundary-value

a unique solution for any choice of yO and Yl.


(b)

Whenever

(e) When yo

yO = Yl there are infinitely many sol utions.

=f. Yl there

will be no solutions.

216

is a solution of the

problem does not have

Exercises 5.2
(d) The boundary-value problem will have the trivial solution when YO
will not be unique.
30.

(a)

Yl

O. This solution

COS 4x + C2 sin 4x. From 1 = y(O) = Cl


+ C2 sin 4x. From 1 = y(L) = cos 4L + c2 sin 4L we see that C2 =

The general solution of the differential equation is Y = Cl


we see that Y = cos 4x

(1 - cos 4L) / sin 4L. Thus,

Y = cos 4x
will be a unique solution when sin 4L

(1-.sm4CO~4L) sin 4x

i- O; that is, when L i- kt: /4 where k = 1, 2, 3, ....

(b) There will be infinitely many solutions when sin 4L = O and 1 - cos 4L

O; that is, when

L = br/2 where k = 1, 2, 3, ....


(e) There will be no solution when sin 4L

i- O and 1 - cos 4L i- O; that is, when L

= br /4 where

k = 1, 3, 5, ....
(d) There can be no trivial solution since it would fail to satisfy the boundary conditions.
31. (a) A solution curve has the sarne y-coordinate at both ends of the interval

[-7r,

1 and the tangent

7r

lines at the endpoints of the interval are parallel.


(b)

For A

or

=O

2Cl7r

= O is Y = ClX + C2.

the solution of y"

= O. Thus,

Cl

From the first boundary condition we have

= O and Y = C2. This constant solution is seen to satisfy the boundary-

value problem.
For A

< O we have Y

= Cl cosh AX

y(

-7r)

+ C2 sinh AX.

In this case the first boundary condition gives

= Cl cosh( -An)
= Cl cosh A7r -

+ C2 sinh( C2

sinh A7r

= Y(7r) = Cl cosh A7r

or

2C2

sinh A7r = O. Thus

a similar fashion that

Cl

C2

= O and Y = Cl cosh AX.

= O. Thus, for A

A7r)

+ C2 sinh A7r
The second boundary condition implies in

< O, the only solution of the boundary-value problem

is Y = O.
For A

> O we have Y

= Cl

COS AX

y(

-7r)

+ C2 sin AX.
=

Cl

The first boundary condition implies

cos( - A7r)

= Cl COS A7r

= Y(7r)

= Cl

+ C2 sin( C2

sin A7r

COS A7r

217

An)

+ C2 sin A7r

Exercises 5.2
or 2C2 sin A71" = O. Similarly, the second boundary condition implies 2eA sin A71"
C2

= O the solution is y = O. However, if ci =1- O or

must be an integer, n. Therefore, for Cl and

C2

C2

= O. If e =

=1- O, then sin A71" = O, which implies that A

not both O, y = e cos nx+c2 sin nx is a nontrivial

solution of the boundary-value problem. Since cos( -nx)

= cos nx and sin( -nx)

= - sin nx,

we may assume without loss of generality that the eigenvalues are An = n, for n a positive
integer. The corresponding eigenfunctions are Yn = cos nx and Yn

= sin nx.

(e)

-3

y = 2 sin 3x

y = sin 4x - 2 cos 3x

> O the general solution is y = Cl COS V).. X + C2 sin V).. x. Setting y(O)
so that y = C2 sin V).. x. The boundary condition y(l) + y'(l) = O implies

32. (a) For A

C2 sin

roots.

C2

=1- O, this equation is equivalent to tan

Since An =

< O. For A

general solution y
Cl

= O,

J).. + C2V).. COS V).. = O.

(b) We see from the graph that tan x = -x has infinitely many
Xn

Cl

V).. = -V)... Thus, the eigenvalues are


x;;, n = 1, 2, 3, ... , where the Xn are the consecutive positive roots of tan V).. = -V)...

Taking
An

= O we find

C2

= O, so y

x;;,

there are no new eigenvalues when

= O, the differential equation y"

tan x
5

= O has

= ClX + C2. The boundary conditions imply


= o.

(e) Using a CAS we find that the first four nonnegative roots of tan x = -x are approximately
2.02876,4.91318,7.97867, and 11.0855. The corresponding eigenvalues are 4.11586,24.1393,
63.6591, and 122.889, with eigenfunctions sin(2.02876x), sin( 4.91318x), sin(7.97867x), and
sin(11.0855x).

218

Exercises 5.3

Exercises 5.3
1. The period corresponding to x(O)

_
= 1 is approxi-

= 1, x'(O)

mately 5.6. The period corresponding to x(O) = 1/2, x'(O) = -1


is approximately 6.2.

-2

2. The solutions are not periodic.

x
10
8
6
4

2.___ --+------------t
-2

3. The period corresponding to x(O) = 1, x'(O)

= 1 is approx-

imately 5.8. The second initial-value problem does not have

10
8

a periodic solution.

6
4

-2

4. Both solutions have periods o approximately 6.3.

219

Exercises 5.3
5. Frorn the graph we see that [xII ~ 1.2.

.:c

3
2

-1

6. Frorn the graphs we see that the interval is approximately


(-0.8,1.1).

-1

7. Since
xeO.01x
for small values of

X,

a linearization is d2X2

.:c

8.

= x[l + O.Olx + 2! (0.01x)2 + ...


dt

+X

J ~ x

= O.

-3

For x(O)

= 1 and x'(O) = 1 the oscillations are symmetric about the line x = O with amplitude

slightly greater than 1.


For x(O) = -2 and x'(O) = 0.5 the oscillations are symmetric about the line x = -2 with small
amplitude.
For x(O) =

vI2 and

x'(O) = 1 the oscillations are symmetric about the line x = Owith amplitude a

220

Exercises 5.3
little greater than 2.
For x(O) = 2 and x' (O)

0.5 the oscillations are symrnetric about the line x

2 with small

amplitude.
For x(O) = -2 and

Xl

(O) = Othere is no oscillation; the solution is constant.

For x(O) = -.;2 and XI(O) = -1 the oscillations are symmetric about the line x = Owith amplitude
a little greater than 2.
9. This is a damped hard spring, so al! solutions should be oscillatory with x -) O as t -) oo.

10. This is a damped soft spring, so so we expect no oscillatory


solutions.

x
5
4
3
2

-Jo
-2

221

Exercises

11. .

5.3

k l,

0.01

15

k1

-3

-15
x

kl

20

k1

100

-2

-2

-3

-3

When kl is very small the effect of the nonlinearity is greatly diminished, and the system is close
to pure resonance.
12. (a)

= -0.000471

-0.000472

The system appears to be oscillatory for -0.000471 :::;kl

< O and nonoscillatory for

. :::;-0.000472.
(b)

-0.078

= -0.079

The system appears to be oscillatory for -0.077 :::;k1 < O and nonoscillatory for k1 :::;0.078.

222

Exercses 5.3

13. For,\2 -

W2

> O we choose ,\

2 and w

1 with

= 1 and x'(O) = 2. For,\2 - w2 < Owe choose


,\ = 1/3 and w = 1 with x(O) = -2 and z'(O) = 4.

x(O)

In both cases the motion corresponds to the overdamped and underdamped

cases for spring/rnass

systems.
-3

14. (a) Setting dy/dt

= v, the differential equation in (13) becomes duf d: = _gR2/y2.

chain rule, dv/dt

= (dv/dy)(dy/dt)

= vdv/dt,

so vdv/dy

= _gR2/y2.

But, by the

Separating variables

and integrating we obtain


vdv = -gR
Setting v

2dy

y2

and

gR2
y

=-+c.

= va and y = R we find e = -gR + ~V6 and


R2

v2 = 2g(b) As y --;

1 2
-v

00

- 2gR+v6.

we assume that v --; 0+. Then v6

= 2gR

and va

= V29R.

(e) Using 9 = 32 ft/s and R = 4000(5280) ft we find


va
(d)

VD

= V2(32)(4000)(5280) ::::::
36765.2 ft/s::::::25067 mi/hr.

= V2(0.165)(32)(1080):::::: 7760 ft/s::::::5291 mi/hr

15. (a) Intuitively, one might expect that only half of a 10-pound chain could be lifted by a 5-pound
force.
(b) Since x = O when t

= O, and v = dx/dt = V160 - 64x/3, we have v(O) = v'I65:::::: 12.65 ft/s.

(e) Since x should always be positive, we solve x(t) = O,


getting t = O and t

~j5fi ::::::
2.3717.

Since the

x
10
8

graph o x(t) is a parabola, the maximum value occurs

at t- = ~j5fi. (This can also be obtained by solving

x'(t)
x(tm)

= O.) At this time the height of the chain is


::::::
7.5

0.5

ft. This is higher than predicted because

1.5

2.5t

of the momentum generated by the force. When the chain is 5 feet high it still has a positive
velocity of about 7.3 ft/s, which keeps it going higher for a while.
16. (a) Setting dxf d:

= v, the differential equation becomes (L-x)dv/dt-v2

rule, duf d: = (dv/dx)(dx/dt)

= vdv/dx,

so (L - x)vdv/dx

223.

- v2

= Lg. But, by the chain


= Lg. Separating variables

Exercises

5.3

and integrating we obtain


vIl

v
so VV2

+ Lg

L dV=-L-dx
9
-x

and

= c/(L - X). When X

-21n(v2+Lg)=-ln(L-x)+lnc,

= O,V =

O, and c

Ly'Eg.

Solving for V and simplifying

we get
dx
dt

= V(X)

V Lg(2Lx

- X2)
L - X
.

Again, separating variables and integrating we obtain


L-x
-=====:==dx
Lg(2Lx - x2)

= dt

Since z (O) = O, we have

=L

x(t)

= O and

V 12 -

y'2Lx - x2
y'Eg
=t+cr

and

y'2Lx - x2/y'Eg

= t.

Lgt2

= dx = -=VL=L=g=t=

and

v(t)

Solving for X we get

dt

/L - gt2

(b) The chain will be completely on the ground when x(t) = L or t =

;;.

(e) The predicted velocity of the upper end of the chain when it hits the ground is infinity.
17. (a) The weight of x feet of the chain is 2x, so the corresponding mass is m

= 2x/32 = x/16.

The

only force acting on the chain is the weight of the portion of the chain hanging over the edge
of the platform. Thus, by Newton's second law,
!!:_(mv)
dt
and x dv/dt
xvdv/dx

= !!:_ (.3:_v) =

+ v2 =

dt

16

32x.

]_(x
16

dv
dt

+ V dX) = ]_ (x
dt

16

Now, by the chain rule, dv/dt

dv
dt

+ v2)

= 2x

(dv/dx) (dx/dt)

v dv/dx,

so

dv

= O. This

is

+ v2 = 32x.

(b) We separate variables and write the differential equation as (v2 - 32x) dx

+ xv

not an exact form, but J.L(x) = x is an integrating factor. Multiplying by x we get


(xv2 - 32x2)dx

+ x2vdv =

O. This form is the total differential of u

3i x3 = c. Letting

implicit solution is ~x2v2 -

3 and v

= O we find c = -288. Solving for

v we get
dx
8y'x3 - 27
=v =
dt
vf3x

3 ::; x ::;8.

(e) Separating variables and integrating we obtain

x
y' 3

x - 27

dx =

8
M dt
y3

and

(X

13

224

~x2v2 - 2:fx3, so an

s
ds = ~ t + c.
y's3 - 27
vf3

Exercses 5.3

= O and
t = J3 (X
s
8 13 vs3 -

Since x = 3 when t = O, we see that e

ds.
27
We want to find t when x = 7. Using a CAS we find t(7) = 0.576 seconds.
18. (a) There are two forces acting on the chain as it falls from the platform. One is the force due to
gravity on the portion of the chain hanging over the edge of the platform.

This is FI = 2x.

The second is due to the motion of the portion of the chain stretched out on the platform. By
Newtori's second law this is
Po = _r!_[
]
2
dt mv

= _r!_[(8-X)2
dt

32

8 - x dv
1 dx
dt - 16v dt

=~

] = _r!_[~v]
dt
16
1 [

dv

2]

= 16 (8 - x) dt - v .

d
From dt [mv] = FI - F2 we have
_r!_ [2X v]
dt 32

dv
x-+v
dt

16

[(8 _ x) dvdt _ v2]


dv
dt

=32x-(8-x)-+v

dv
x dt

= 32x - 8 - + x -

dv
dt

dv
8 dt

= 32x.

dv
dt

By the chain rule, dv/dt

(b ) Integrating v dv

= 2x _ 2_

(dv/dx)(dx/dt)
= v dv/dx, so
dv
dv
dv
8 dt = 8v dx = 32x and v dx

= 4x.

= 4x dx we get ~v2 = 2x2 +e. Since v = O when x

v2 = 4x2 - 36 and v =

V4x2

= 3, we have e

= -18. Then

- 36. Using v = dx] dt, separating variables, and integrating we

obtain
dx

v'X2.=9

= 2 dt

and

x
cosh-I'3

= 2t + q.

+ el). Since x = 3 when t = O, we have cosh el


el = O. Thus, x(t) = 3 cosh2t. Differentiating, we find v(t) = daf d: = 6sinh2t.
Solving for x we get x(t) = 3 cosh(2t

= 1 and

(e) To find when the back end of the chain will leave the platform we solve x (t) = 3 cosh 2t =
8. This gives tI = ~cosh-l
6sinh(cosh-1

~)

= 2V55

~ ~

0.8184 seconds.

~ 14.83 ft/s.

225

The velocity at this instant is V(tl)

Exercises 5.3
(d) Replacing 8 with L and 32 with g in part (a) we have Ldv/dt
dv
L dt
.12

Integrating we get:zv
for v we find

dv

= Lv dx = gx

dv
v dx

and

= gx. Then

= LX.
g

v(x)=

/!!'_x2-!!'_x5.
L
L
Then the velocity at which the end of the chan leaves the edge of the platform s

v(L)

j f(p - x5)

19. Let (x, y) be the coordinates of 82 on the curve C. The slope at (x, y) is then
dy/dx

= (Vlt

- y)/(O - x) = (y - vlt)/X

or

xy' - y

Differentiating with respect to x and usng r = VdV2 gves


xy "

+ y"-

dt
y = -vI dx

xy

"

dt ds
ds dx

= -Vl--

-JI + (y')2)

xy" = -VI }:_(


V2
xy"

= rJ1 + (y')2.

Lettng u = y' and separating variables, we obtain


x du = rJ1
dx
du
~

+ u2

'!._dx
x

snh -1 u = r In x

+ In e =

ln(cz")

u = snhln cz"]

!(ex

At t = O, dy/dx

If r > 1 or r

= 2L x + e. Setting x = Xo and v = O, we find e = - 2L xo Solving

1_) .

dy =
T __
dx
2
cz"
O and x = a, so O = ca" -l/eaT. Thus e = l/aT and

< 1, integrating gives


y _- -a [ -- 1
2 l+r

(X)l+T
a

- -- 1
1-r

226

(X)l-T]
a

+ el

= -vlt.

Exercses 5.3
When t =

o, y

+ r) -

= O and x = a, so 0= (a/2)[1/(1

1/(1 - r)l

+ CI.

Thus C1= ar/(l - r2) and

(~)l+r
__1 -1 r (~)l-r]
+ ~.
a
a
1 - r2

y = 9:_[_1
2 1 +r

To see if the paths ever interseet we first note that if r


In other words, S2 always lags behind

SI.

Next. if r

> 1, then

< 1, then

VI

> V2 and y_,

VI

< V2 and y

00

as x _, 0+.

= ar/(l-

r2) when

x = O. In other words, when the submarine's speed is greater than the ship's, their paths will
interseet at the point (O,ar/(l-

r2)).

Finally, if r = 1, then integration gives

y = ~ [x2 _ ~ In x] + C2.
2 2a a
When t
and

= O, y = Oand x = a, so O = (1/2)[a/2
y=~[X2

2 2a

Sinee

u : 00 as x

-~lnx]

- (l/a) In al + C2 Thus

= -(1/2)[a/2

=~[~(x2-a2)+~ln~].
2 2a
a

-~[9:_-~lna]
2 2 a

C2

_, 0+, S2 will never eateh up with

- (l/a) In al

SI.

20. (a) Let (r, e) denote the polar eoordinates of the destroyer

SI.

When

(9, O)to (3, O) it stands to reason, sinee S2 travels half as fast as

SI,

SI

travels the 6 miles from

that the polar eoordinates

of S2 are (3, e2), where e2 is unknown. In other words, the distanees of the ships from (O,O)
are the same and r(t)
SI

= 15t then gives the radial distanee of both ships. This is neeessary if

is to intereept S2.

(b) The differential of are length in polar eoordinates is (ds)2 = (r def

Using ds/dt

= 30 and

dr/dt

= 15 then

gives
2

900

= 225t2 ( ~~) + 225

675 = 225t2 (~;)

de
dt

-=-

e(t) =
When r

= 3, e = O, so C = -y'3
e(t) =

y'3
t

v'3 In t + C = v'3 In ;5 + c.

In(1/5) and

v'3 (In ~15 -ln~) 5


227

v'3 In:3 .

+ (dr )2, so that

Exercises 5.3
Thus r = 3eB/v'3, whose graph is a logarithmic spiral.
(c) The time for SI to go from (9, O) to (3, O) =
for sorne angle

i hour.

Now SI must intercept the path of S2

/3, where O < /3 < 27l'. At the time of interception t2 we have 15t2

= 3e/3/v'3 or

t = e/3/v'3/5. The total time is then


1

t =-

21. Since (dx/dt)2

1
1
+ _e/3/v3
< -(1 + e27r/v3).
;;

;;

is always positive, it is necessary to use Idx/dtl(dx/dt)

in order to account for the

fact that the motion is oscillatory and the velocity (or its square) should be negative when the
spring is contracting.
22. (a) The approximation is accurate to two decimal places for

el = 0.3, and

accurate to one decimal place for

el = 0.6.
0.8
0.6
0.4
0.2
0.20.40.60.8

(h)

0.3r---~
0.2--

0.11-----_
+--'0"."""2
---"-0-'.
40-

e
o.

O.

O.

o.

o.

o.

O.

0.1

O.
.2

0.4

0.3

1.21.4

O.
t

The thinner curves are solutions of the nonlinear differential equation, while the thicker curves
are solutions of the linear differential equation.
23. (a) Write the differential equation as

d2

dt2 +w2sine

= O,

where w2 = g/e. To test for differences between the


earth and the moon we take e = 3, B(O) = 1, and
e/(O) = 2. Using 9 = 32 on the earth and 9 = 5.5 on
the moon we obtain the graphs shown in the figure.

-2

Comparing the apparent periods of the graphs, we see that the pendulum oscillates faster on
the earth than on the moon.

228

Exercises

5.3

(b) The amplitude is greater on the moon than on the earth.


(e) The linear model is

d2e
dt2 + w2e = O,
where w2 = g/e. When
and

e' (O) = 2, the

= 32,

e = 3, e(O)

= 1,

general solution is

e(t) = cos 3.266t + 0.612 sin 3.266t.


When g

-2

5.5 the general solution is

e(t) = cos 1.354t + 1.477sin1.354t.


As in the nonlinear case, the pendulum oscillates faster on the earth than on the moon and
still has greater amplitude on the moon.
24. (a) The general solution of
d2

dt2 + e = O

is e(t)

= Cl cost + C2sint. From e(O) = 1r/12 and e'(O) = -1/3 we find


e(t) = (1r/12)cost - (1/3)sint.

Setting e(t)

= O we have tant = 1r/4 which implies tI

(b) We set e(t) = e(O)

+ e'(O)t + ~e"(O)t + iell/(O)t + ...

= tan-I(1r/4)

::::;:
0.66577.

and use e"(t) = - sine(t) together with

e(O) = 1r/12 and e'(O) = -1/3. Then


e"(O) = - sin(1r/12) =

-h (J3 -1)/4

and

= h (J3 + 1)/12.

ell/(O) = - cos e(O) . e' (O) = - cos(1r/12)( -1/3)


Thus

e(t) = _:__ ~t _
12

(e) Setting 1r/12 - t/3

J2 (J3 -

1) t2 +

J2 (J3 + 1) t3 + .. '.

= O we obtain

tI

72

= 1r/4::::;:0.785398.

(d) Setting

_:_-~t-

J2(J3-1)t2=0

12 3
8
and using the positive root we obtain tI ::::;:
0.63088.

(e) Setting
_:_ _ ~ t _
12

J2 (J3 -

1) t2 +

J2 (J3 + 1) t3 = O
72

229

Exercises 5.3
we find tl ::::::
0.661973 to be the first positive root.

(f) From the output we see that y(t) is an interpolating function on


the intervalO:::; t :::;5, whose graph is shown. The positive root
of y(t) = Onear t = 1 is tl = 0.666404.

5
-0.4

(g) To find the next two positive roots we change the interval used in

0.4

NDSolve and Plot from {t,O,5} to {t,O,lO}. We see from the


graph that the second and third positive roots are near 4 and 7,
respectively. Replacing {t,l} in FindRoot

with {t,4} and then

{t,7} we obtain t2 = 3.84411 and t3 = 7.0218.


25. From the table below we see that the pendulum first passes the vertical position between 1.7 and
1.8 seconds. To refine our estimate of tl we estimate the solution of the differential equation on
[1.7,1.8] using a step size of h = 0.01. From the resulting table we see that tl is between 1.76 and
1.77 seconds. Repeating the process with h = 0.001 we conclude that tl ::::::1.767. Then the period of
the pendulum is approximately 4tl = 7.068. The error when using tl = 271"is 7.068 - 6.283 = 0.785
and the percentage relative error is (0.785/7.068)100 = 11.1.
h=O.Ol

h=O.l
tn

en

tn

0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
1.10
1.20
1.30
1.40
1.50
1.60
1.70
1.80
1.90
2.00

0.78540
0.78523
0.78407
0.78092
0.77482
0.76482
0.75004
0.72962
0.70275
0.66872
0.62687
0.57660
0.51744
0.44895
0.37085
0.28289
0.18497
0.07706
-0.04076
-0.16831
-0.30531

1.70
1.71
1.72
1.73
1.74
1.75
1.76
1.77
1. 78
1.79
1. 80

0.07706
0.06572
0.05428
0.04275
0.03111
0.01938
0.00755
-0.00438
-0.01641
-0.02854
-0.04076

h=O 001
1. 763
1. 764
1.765
1.766
1.767
1.768
1.769
1.770

0.00398
0.00279
0.00160
0.00040
-0.00079
-0.00199
-0.00318
-0.00438

230

en

Chapter 5 Review Exercises

Chapter 5 Review Exercises

1. 8 ft., since k = 4.
2. 271'/5,since

~X"

+ 6.25x

= O.

3. 5/4 m, since x = -cos4t+

~sin4t.

4. True
5. False; since an external force may existo
6. False
7. overdamped
8. From x(O) = (V2/2)sin<t> = -1/2

we see that sin

so <t>is an angle in the third or

= -1/V2,

fourth quadrant. Since x'(t) = V2cos(2t+<t, x'(O) = V2cos<t>= 1 and cos<t>> O. Thus <t>
is in the
fourth quadrant and <t>= -71'/4.
9. The period of a spring mass system is given by T

= 271'/w where w2 = k/m = kg/W,

where k is the

spring constant, W is the weight of the mass attached to the spring, and 9 is the acceleration due
to gravity. Thus, the period of oscillation is T = (271'/..jJg)JW.
is W, then (271'/..jJg)JW

= 3 and

(271'/..jJg)y'W

If the weight of the original mass

- 8 = 2. Dividing, we get JW/JW

- 8 = 3/2

or W = ~(W - 8). Solving for W we find that the weight of the original mass was 14.4 pounds.
10. (a) Solving

iX"

+ 6x

= Osubject to x(O) = 1 and x'(O) = -4 we obtain


x = cos4t - sin4t = V2sin (4t + 371'/4).

(b) The amplitude is V2, period is 71'/2,and frequency is 2/71'.

= 1 then t = n71'/2 and t = -71'/8 + n71'/2 for n = 1, 2, 3, ....


(d ) If x = O then t = 71'/16+ n71'/4 for n = O, 1, 2, .... The motion is upward for neven and
(e) If x

downward for n. odd.

(e) x'(371'/16) = O
(f) If x' = O then 4t + 371'/4= 71'/2+ n71' or t = 371'/16+ n71'.
11. Frorn

~X"

+ ~x' + 2x

= O, x(O)

= 1/3, and

x'(O)

= Owe obtain

e-2t - !e-4t.

12. From x" + /3x' + 64x = Owe see that oscillatory motion results if /32

- 256 < O or O ::; 1/31 < 16.


13. From mx" + 4x' + 2x = Owe see that non-oscillatory motion results if 16 - 8m 2': Oor O < m ::; 2.
14. From

~X"

+ x' + x = O, x(O) = 4, and x'(O) = 2 we obtain x = 4e-2t + 10te-2t.

t = 1/10, so that the maximum displacement is x = 5e-O.2

:::::

If x'(t) = O, then

4.094.

= cos"(t + sin-yz in the form x" + 634x = Bcos"(t + Bsin"(t we identify .x = O and
w2 = 64/3. The system is in a state of pure resonance when "( = w = j6473 = B/-/3.

15. Writing ~x" + ~x

231 .

Chapter 5 Review Exercises


16. Clearly xp = A/w2 suffices.
17. From kx" + Xl + 3x = e-t, x(O) = 2, and X/(O) = Owe obtain Xc = e-4t (c cos 2v2 t + C2sin 2v2
8 -t
17e
,an

xp =

t),

d
26
2Sv2
x=e-4t ( -cos2v'2t+--sin2v'2t
17
17

) +_e8 t.
17

18. (a) Let k be the effective spring constant and Xl and X2 the elongation of springs kl and k2. The
restoring forces satisfy klXI = k2X2 so X2

= (kjk2)XI.

From k(XI + X2) = klXI we have

k (Xl + ~~X2) = klXI

k ( k2 ~ kl )

= kl

k = klk2
kl + k2
1
1
1
k = kl + k2 .
(b) From kl = 2W and k2 = 4W we find l/k = 1/2W +1/4W = 3/4W. Then k = 4W/3 = 4m9/3.
The differential equation mx" + kx = Othen becomes x" + (49/3)x = O. The solution is

x(t) = c cos 2~

t + C2sin 2~

t.

The initial conditions x(O) = 1 and X/(O) = 2/3 imply CI = 1 and C2 = 1/.,f39.
(e) To compute the maximum speed of the weight we compute
Xl (t)

= 100sin50t, q(O)

19. From q" + 104q


qp

= 2~ sin 2~ t + ~ cos 2~ t and IXI (t) I = 4 ~ + ~ = ~ 39 + 1 .


ql(t) = O we obtain qc = ccos100t+

= O, and

A sin 50t, and


A sin 50t,

(a) q =

-1~O sin

(b) i =

-~ cos 100t + 1cos 50t, and

(e) q

100t +

= O when sin50t(1-

cos50t)

= Oor t = mr/50

for n

= 0,1,2,

....

20. (a) By Kirchoff's second law,

d2q
dq
1
L dt2 + R dt + e q = E(t).
Using ql(t)

= i(t)

we can write the differential equation in the form


di

L dt + Ri +

e1 q = E (t).

232

C2sin 100t,

Chapter 5 Review Exercises


Then differentiating we obtain
d2i
L dt2

+ Ri(t) + (1/C)q(t)

(b ) Frorn Li'(t)

+R

di
dt

+ ei = E

(t).

E(t) we find Li'(O) + Ri(O) + (1/C)q(0)

E(O) or Li'(O) +

Rio + (1/C)qo = E(O). Solving for i'(O) we get


i'(O)
21. For .A
Cl

>

[E(O) - b qo - Rio] .

the general solution is y =

Cl COS

J>: x + C2 sin J>: x.

cos 21rJ>: + C2 sin 21rJ>: , so the condition y(O)


=

ci

which is true when


y'

Now y(O)

Cl

and y(21r)

= y(21r) implies

cos 21rJ>: + C2 sin 21rJ>:

ci

J>: = n or .A = n2 for n = 1, 2, 3, .... Since


= - J>: Cl sin J>: x + J>: C2 cos J>: x = -nCl sin nx + nC2cos nx,

we see that y'(O) = nC2 = y'(21r) for n = 1, 2, 3, ....

Thus, the eigenvalues are n2 for n =

1, 2, 3, ... , with corresponding eigenfunctions cos nx and sin nx. When X = O, the general solution
is y = ClX + C2 and the corresponding eigenfunction is y = 1.
For .A

< O the general solution is y

y(21r) =

Cl

Cl

cosh H x

+ C2 sinh 21rH,

cosh 21rH

are no eigenvalues corresponding to .A

so y(O)

+ c2e-wt.

In this case y(O) =

can only be valid for .A

Cl

= O. Thus,

and

there

< O.

22. (a) The differential equation is d2r/dt2-w2r


re = clewt

+ C2 sinh H x.

= y(21r)

= -gsinwt.

The auxiliary equation is m2_w2 = O,so

+ Bcoswt.

A particular solution has the form rp = Asinwt

Substituting

= - 9 sin wt. Thus, B =


= (g/2w2) sinwt. The general solution of the differential equation

into the differential equation we find - 2Aw2 sin wt - 2Bw2 cos wt


O, A = g/2w2,

and "

is r(t)

= clewt + C2e-wt + (g/2w2) sinwt. The initial conditions imply

g/2w -

WCl

+ WC2 = Vo.
C1 = (2w2ro

Solving for

+ 2wvo

Cl

and

- g)/4w2

C2

we get

and

C2

Cl

+ C2

= ro and

= (2w2ro - 2wvo + g)/4w2,

so that
r ()t =

2w2ro

+ 2wvo
4w2

- 9 wt 2w2ro - 2wvo
e +
4w2

+9

-wt

+ 2w2 SIn wt.

(b) The bead will exhibit simple harmonic motion when the exponential terms are missing. Solving
q = O, C2 = O for ro and vo we find ro = Oand vo = g/2w.

To find the minimum length of rod that will accommodate simple harmonic motion we determine the amplitude of r(t) and double it. Thus L = g/w2.

233

Chapter 5 Review Exercises


(e) As t increases, ewt approaches infinity and e-wt approaches O. Since sin wt is bounded, the
distance, r(t), of the bead from the pivot point increases without bound and the distance of
the bead from P will eventually exceed L/2.

(d)

14

(e) For each vo we want to find the smallest value of t for which r(t) = 20. Whether we look for

r(t) = -20 or r(t) = 20 is determined by looking at: the graphs in part (d). The total times
that the bead stays on the rod is shown in the table below.
Vo

When vo

10

16.1

15

17

-20

-20

-20

20

20

1.55007

2.35494

3.43088

6.11627

4.22339

= 16 the bead never leaves the rod.

Chapter 5 Related Exercises


1. (a) The auxiliary equation is m 2 + 4

= Oso Xc = el cos 2t + e2 sin 2t. Letting xp = A sin 4t + B cos 4t

and substituting into the differential equation, we get -12A sin 4t - 12B cos 4t = sin 4t. Thus,

= - l~ , B = O, and xp = - f2 sin 4t. The general solution of the differential equation is


x = el cos 2t + e2 sin 2t - l~ sin 4t. The initial conditions imply el = Oand e2 = ~ + 1- ). Thus
A

(o:

x(t)

~(o:
+~)
sin2t 2
3
= ~(o: +~)sin2t 2
3
=

~ sin4t
12
~(2sin2tcos2t)
12
tt

O<t<-.
- 2
For x"

+x =

sin 4t the auxiliary equation is m2

+1=

O, so

Xc

= el cos t

+ e2 sin t.

Letting

= A sin 4t + B cos 4t and substituting into the differential equation we get -15A sin 4t 15B cos 4t = sin 4t. Thus, A = -l5 ' B = O, and xp = -l5 sin 4t. The general solution of

xp

234

Chapter 5 Related Exercises


the differential equation is x = Cl COSt +
x'(7r/2)

= -(ex +~) imply

Cl

= ex+

C2 sin

and

C2

t-

!5

sin 4t. The initial conditions x( tt /2)

= O,

= O. Thus

x(t) = (ex+~) cost - 115sin4t


= (ex + ~) cos t - 2_ (2 sin 2t cos 2t)
5
15
= (ex+ ~) cost - 2_(2sintcostcos2t)

= cos t [( ex+

D-

15

1~ sin t cos 2t ] ,

37r

tt

-<t<-.
2 - - 2

(b ) The velocity at the start of the second cycle is x' (37r/2) = ex+
(c) Using results from part (a), x =
x(37r/2)

= O and x'(37r/2)

Cl

= ex+

cos 2t +

125'

C2

sin 2t -

which imply

Cl

125 .

112sin 4t.

O and

C2

The initial conditions are now


= -~(ex +

fs).

Thus

2.)

x (t) = - ~ (ex +
sin 2t - 2_ sin 4t
2
15
12
= -~(ex+

175)sin2t-

= sin2t[-~(ex+

For t 2: 27r we have x =

Cl

COSt +

C2

:2(2sin2tcos2t)

2.)
- ~coS2t]
15
6
sin t -

15

'

A sin 4t, as in part

x(27r) = O and x'(27r) = -(ex + ~), which imply

x(t) = -(ex +~)

37r

-2 <
- t<
- 27r.

Cl

= O and

C2

(a). The initial conditions are

= -(ex +

sin t - 2_ sin4t
15

= -(ex +~)
sint - 2_(2sin2tcos2t)
15
15
= - (ex+ ~) sin t - ~ (2 sin t cos t cos 2t)
15
15
= sin t[- (ex + ~) - 2_ cos t cos 2t]
15
15
'
(d)

-2

-2

ex=l

ex=2

235

..

l~).

Thus

Chapter 5 Related Exercses


2. (a) We have x = el cos t

+ e2 sin t

and we let xp = At cos t

+ Bt

sin t.

Substituting into the

differential equation we get 2B cos t - 2A sin t = cos t. Thus A = O, B = -~ , and xp = ~t sin t.


The general solution of the differential equation is x(t)

= el cos t + e2 sin t + ~t sin t. The initial

conditions imply el = Oand e2 = O,so the solution of the initial-value problem is x( t) = ~t sin t.
(b) We have

Xc

= el cos t + e2 sin t and we let xp

A cos 2t + B sin 2t. Substituting into the

= cos 2t. Thus A = - ~, B = O, and xp =


-1 cos 2t. The general solution of the differential equation is x(t) = el cos t + e2 sin t - ~ cos 2t.
The initial conditions imply c = ~ and e2 = O, so the solution of the initial-value problem is
x(t) = ~cost - 1cos2t.
differential equation we get -3A cos 2t - 3B sin t

3. (a) In the first case the differential equations are x"

+X

= sin 4t when

X ::::: O and

x"

+ 4x

= sin 4t

< O. Since the initial velocity is positive, we solve first for x :::::o. The solution of
x" + x = sin 4t is x( t) = el cos t + C2 sin t - t\ sin 4t, as seen in Problem 1(a). The initial
conditions x(O) = Oand x'(O) = 1 imply el = Oand c2 = ~~
, so the solution of the initial-value
when x

problem is
19

x(t) = 15 sint - 15 sin4t


= ~:sin t

19
15

= - sin t
=

- 1~(2 sin 2t cos 2t)


2.
15

- - (2 sin t COS t cos 2t)

19
4
)
(sint) ( 15 - 15 costcos2t ,

0S;tS;7L

The first positive value of t for which x(t) is zero is t = 7r, at which time x'(7r) =
so x

< O immediately after t

x(7r) = O, x'(7r)

= tt. The new initial-value problem is then

-i~.The solution

of this initial-value problem is

x (t) = - ~sin 2t - 1~ sin 4t = (sin 2t) ( - ~ - ~ cos 2t) ,


The next value of t for which x(t)

= O is t

7r S; t S; 3; .

= 37r/2, at which time

x'(37r/2) = ~~ > O, so

> O immediately after t = 37r/2. The next initial-value problem is then x" + x
x(37r/2) = O, x' (37r/2) = ~~. The solution of this initial-value problem is
x

17

-n < O,

x" + 4x = sin 4t,

x(t) = 15 cost - 15 sin4t = (cost)

( 17
4
)
15 - 15 sintcos2t,

37r
2'"
S; t

S;

= sin4t,

57r
2""

= O is t = 57r/2, at which time x'(57r/2) = -~ < O, so


x < O irnmediately after t = 57r/2. The next initial-value problem is then x" + 4x = sin 4t,
x(57r/2) = O, x'(57r/2) = -~. The solution of this initial-value problem is

The next value of t for which x(t)

x(t) = _!_ sin 2t - _!_ sin 4t


15

12

(sin 2t) (_!_


15

236

- ~ cos 2t),
6

57r < t < b


2 - -

Chapter

= O. In this case,

where b is the value of t for which x(t)

5 Related Exercises

i cos 2t is O for t = 8.84514,

115 -

which is between 51i/2 and 31i. Thus, for the first two cycles,
(sin t)(

i~- 1~ cos t cos 2t),

(sin 2t) (-

x(t) =

O::;t::;1i

* - i cos 2t),

1i <
t
-

17
4
(cos t) (15
- 15
sin t cos 2t) ,

(sin 2t) (

< 37r2
<
t < 57r
- 2

37r

5; < t ::;8.84514.

fs - t cos 2t),

The first part of the next cycle is the solution of the initial-value problem x" + x = sin 4t,

x(8.84514) = O, x'(8.84514) = 0.28. This solution is approxirnately x(t) = -0.013 cost 0.109 sin t - 0.067 sin 4t and is positive on (8.84514,9.42478) with amplitude 0.042.

= sin 4t when x 2: O and x" + 4x =


< O. Since the initial velocity is positive we solve first for x 2: O. [The following

In the second case, the differential equations are x" + 64x


sin 4t when x

computations are done with the aid of a CAS having a differential equation solver and graphing
capability.] The solution of x"

+ 64x = sin4t,

1(.

x(O) = O, x'(O) = 1 is

11.)

x(t) = 48 sm4t+2'"sm8t
where

TI

= 0.415458 is found using the root-finding capability of the CASo The next initial-

value problem is x" + 4x = sin4t,

X(T1)

= O, X'(T1)

-0.909091. The solution is

x(t) = 0.403 cos 2t - 0.255 sin 2t - 0.083 sin 4t,


where

T2

2.1403. The next initial-value

TI::;

t::;

T2,

problem is x" + 64x = sin4t,

X(T2)

O, X'(T2)

= O, X'(T3)

l.16206. The solution is

x( t) = 0.153 cos 8t + 0.021 sin 4t - 0.008 sin 8t,


where

T3 =

T2::;

t ::; T3,

2.53479. The next initial-value problem is x" + 4x = sin4t,

X(T3)

-1. 28086. The solution is

x(t) = -0.737cos2t
where

T4

- 0.217sin2t - 0.083sin4t,

T3::;

t::;

T4,

= 4.0453.

In the third case, the differential equations are x" + 36x


sin 4t when x
of x" + 36x

< O.

= sin4t,

sin 4t when x 2: O and x" + 25x

Since the initial velocity is positive, we solve first for

x(O) = O, x'(O) = 1 is
x(t) = 2_ sin 4t +
20

2.
sin 6t,
15
237

O::; t

< TI,

x 2: O. The solution

Chapter 5 Related Exercises


where

TI = 0.571447.

Subsequent solutions are

x(t) = 0.1l5cos5t + 0.1l1sin4t + 0.093sin5t,


x(t) = -0.068 cos 6t + 0.05sin4t+
x(t)

0.082 sin 6t,

= 0.159 cos 5t + 0.111 sin 4t +

0.072 sin 6t,

Tl
T2
T3

S;
S;
S;

t S;
t S;
t S;

T2
T3
T4

= 1.24203
= 1.73555

= 2.21355.

-1

-1

b = 1,

b = 64,

a =4

b = 36,

a =4

a = 25

In each case it appears that x(t) has increasing amplitude as t increases.


(c) When a

= 0.5

the following graphs are obtained.


x

10
-1

-1

b = 1,

a =4

= 64,

a =4

= 36,

= 25

= 36,

= 25

When a = 3 the following graphs are obtained.

-1

-2

-3

1,

a =4

b = 64,

a =4

In both situations, it appears that the amplitude decreases for b = 1 and a = 4, but increases
in the other two cases.

238

Chapter

5 Related Exercses

4. (a) The solutions are obtained numcrically in CAS and plotted below in part (b). Since x'(O) =
1

> O, in each case the first differential equation used is x" + !3x' + 4x

= sin 4t, followed by

x" + !3x' + x = sin 4t, and alternating thereafter.


(b)
2

-2

x
1

-1

-2

!3 = 0.01

!3 = 0.5

!3 = 0.1

As t increases, the amplitude appears to increase for

!3 = 0.5.

239

!3

= 0.01 and decrease for

!3 = 0.1 and

Series Solutions of Linear Equations


Exercises 6.1

1.

u1m lan+ll
-= u1m 12n+lxn+l/(n+1)1
n---+ooan
n---+oo
2nxn [ti

u1m _-2n I1x = 211


X
n---+oo
n+ 1

The series is absolutely convergent for 21xl

< 1 or Ixl < 1/2. At x = -1/2, the series

f:

n=l

(_l)n
n

00 1

I:- is the

converges by the alternating series test. At x = 1/2, the series

harmonic series which

n=l n
diverges. Thus, the given series converges on [-1/2,1/2).
2.

u1m lan+ll
-n---+ooan

= l'1m !100n+l(x+7)n+l/(n+1)!!
n---+oo

100n(x + 7)n/n!

u1m _-100 Ix + 7 1= O
n---+oo
n + 1

The series is absolutely convergent on (-00,00).


ak 1
3. lim _____
k---+ooak
l

I=

1
(x - 5)k+ / lOk+1
1
1
lim
= lim -Ix - 51= -Ix
k---+oo (x-5)k/10k
k---+0010
10

The series is absolutely convergent for 110Ix series


00
I:(-l)k
k=l
4.

00 (_l)k (-10)k
2::
10k
k=l

51 < 1, Ix -

51 < 10, or on (-5,15).

00
2::
1 diverges by the k-th term test.

- 51

At x

= 15, the series

k=l

At x = -5, the
00 (-1)k10k
2::
10k

k=l

diverges by the k-th term test. Thus, the series converges on (-5,15).

lim jak+1j= lim I(k\~t(x-)~)k+ll=


k---+ooak
k---+oo
.x - 1

lim(k+1)lx-11=00,
k---+oo

x=l-l

The radius of convergence is O and the series converges only for x = l.


5. sinxcosx

= (x - ~3 + 1x250_ 5~:0 + ... ) (1-

6. e-x cos x = ( 1 - x +

7.

1
cosx

x2
x3
'2
- "6
+

1
2

1-~+ir4
,

4
.

--fu-+'"
6.

x4
24 - .,.

x; + ~: _

1-

x2
'2
+

;6

x4)
24 - '"

x2
5x4
61x6
= 1 + -2 + -41 + -61 + ...
..

240

0 + ... ) = x- 2~3 + 21~5_ ~~~+ ...

= 1- x +

x3
x4
3
- "6
+

...

Exercises
Since cos( 7r/2) = cos(
1-x
1 3
= - - -x
2+x
2 4

+ -x3

8. --

/2)

T;

- -x
16

= 0, the
3

00

2ncnxn-1

(-7r

7r/2).

-2, the series converges on (-2,2).

00

00

00

+ 6cnxn+l = 21 CIXO + 2nenxn-l

n=l

/2,

+ ...

Since the function is undefined at x


9.

series converges on

6. 1

n=O

+ 6enxn+l

n=2
._,__..._.,
k=n-l

00

n=O
~
k=n+l
00

= 2Cl+ 2(k + 1)Ck+lXk + 6Ck_lXk


k=l

k=l

00

= 2Cl+ [2(k + 1)Ck+l + 6Ck_l]Xk


k=l
00

10.

n(n

00

- 1)enxn

+ 2 n(n

n=2

00

- 1)enxn-2

+ 3 nenxn

n=2

n=l
00

00

00

= 22 1C2XO+ 2 . 3 . 2C3X' + 31 CIX' + n(n - 1)enxn +2 n(n - 1)cnxn-2 +3 nenxn


n=2

n=4
k=n

00

= 4C2 + (12c3

n=2
k=n

'----v--'

k=n-2
00

00

+ (12c3 + 3Cl)X + k(k - 1)CkXk + 2 (k + 2)(k + 1)Ck+2Xk + 3 kCkXk


n=2

n=2

n=2

00

= 4C2 + (3Cl+ 12c3)X + ([k(k - 1)+ 3k]Ck + 2(k + 2)(k + 1)Ck+2)Xk


n=2
00

= 4C2 + (3Cl+ 12c3)X + [k(k + 2)Ck + 2(k + 1)(k + 2)Ck+21xk


n=2
00

11. y' =

00

(_l)n+lxn-l,
n=l

y" =

(-1t+1(n

- 1)xn-2

n=2
00

(x

+ 1)Y" + y'

= (x

00

+ 1) (_l)n+l(n - 1)xn-2 + (_l)n+lxn-l


~l

n~
00

00

00

(_l)n+l(n- 1)xn-1 + (-1t+1(n


n=2

n=2
00

= -xo+xo+

- 1)xn-2

+ (-1t+1xn-l
n=l

00

(_l)n+l(n_1)xn-l+
n=2
k=n-l

(_l)n+l(n_1)xn-2+
n=3
k=n-2

241

00

(_l)n+lxn-l
n=2

'----v---'

k=n-l

Exercises

6. 1
00

00

00

+ (_1)k+3(k

= (_1)k+2kxk
k=l

+ 1)xk + (_1)k+2xk

k=l

k=l

00

[(_1)k+2k

_ (_1)k+2k

+ (_1)k+2]

_ (_1)k+2

xk = O

k=l

,_ ~
12. y - ~

n=l

'"
xy

(-1)n2n
2n-l
22n(n!)2 x
,

" _ ~

y - ~

n=l

(-1)n2n(2n
- 1) 2n-2
22n(n!)2
X

(-1)n2n(2n
- 1) 2n-l
~
22n(n!)2
X
+ ~l

_ ~
+ y + xy - ~l

(-l)n2n
2n-l
~
22n(n!)2 x
+~

k=n

13. Substituting y

y" - xy

k=n

(_1)n
2n+l
22n(n!)2 x
k=n+l

L~o cnxn into the differential equation we have


00

n(n

00

- 1)Cnxn-2 -

n=2

Cnxn+l

00

n=O
k=n+l

00

+ 1)Ck+2Xk

k=O

'-.r---'

k=n-2

(k + 2)(k

00

= 2C2 + [(k

+ 2)(k + 1)Ck+2 - Ck_tlxk = O.

k=l

Thus
C2= O

(k

+ 2)(k + 1)Ck+2 -

Ck+2 = (k

+ 2)(k + 1) Ck-l,

Ck-l = O

and
1

Choosing co = 1 and q = Owe find


1
6

C3

=-

C4

= Cs = O

C6 -

1
180

242

k = 1,2,3, ....

Ck_lXk
k= 1

Exercises
and so on. For

= O and

Co

Cl

= 1 we obtain
C3

=O

C5

= C6 = O

C7

= 504

and so on. Thus, two solutions are


1 3
1
= 1 + -x
+ -x

Yl

14. Subst.ituting y =

180

+ ...

+ x2y

Y2 =

1 4
1 7
+ -x
+ -x
+ ....
12
504

I.:~=O Cnxn into the differential equation we have


00

y"

and

.L n(n

00

_ l)cnxn-2

n=2

+ .L cnxn+2

00

n=O
k=n+2

00

+ 2)(k + l)ck+2Xk + .L Ck_2Xk

k=O

'-v----'

k=n-2

.L (k

k=2

00

= 2C2 + 6C3X +

.L [(k + 2)(k

+ I)Ck+2 + Ck_2]Xk

= O.

k=2
Thus
C2

(k
and

C3

=O

+ 2)(k + I)Ck+2 + Ck-2


1

Ck+2 = _ (k
Choosing

Co

+ 2)(k + 1) Ck-2,

= 1 and CI = Owe find


C4

= _-

12

CS=C6=C7=0

1
C8 = 672

and so on. For

Co

= O and

Cl

= 1 we obt ain
C4 =

Cs

= _-

20

C6=C7=C8=0
cg

1
= 1440

243

=O

k = 2,3,4, ....

6. 1

Exercises

6. 1

and so on. Thus, two solutions are


Yl
15. Substituting y

1 4
1
= 1 - -x
+ -x

12

672

- ...

Y2 = x - -x
.
20

and

1 9
+ --x
1440

= .L~=o cnxn into the differential equation we have


00

y" - 2xy'

+y

n(n

00

- 1)cnxn-2

- 2

n=2

ncnxn
n=l
~

k=n-2

k=n

00

(k
k=O

+ 2)(k + 1)Ck+2Xk -

00

+ cnxn
n=O
k=n

'---v-"

00

00

k=l

k=O

kCkXk + Ckxk

00

= 2C2 + Co +

[(k + 2)(k

+ 1)Ck+2 -

(2k - l)Ck]Xk = O.

k=l

Thus
2C2 + Co = O
(k

+ 2)(k + 1)Ck+2 - (2k - l)Ck = O

and
e

= --Co2

2k - 1
Ck+2 = (k
Choosing

Co

= 1 and

CI

+ 2)(k + 1) ci,

k=1,2,3,

= Owe find
C2

1
=-2

C3

= Cs = C7 = ... = O
1

C4 =--

8
7

C6=-and so on. For

Co

= O and

Cl

336

= 1 we obtain
C2

= C4 = C6 = ... = O
1

C3 =-

6
1

C5

= 24

C7

1
= 112

244

....

Exercises
and so on. Thus, two solutions are
Yl

= 1 - -x1

1
- -x
8

--x

7
336

and

16. Substituting y = ~=o cnxn into the differential equation we have


y" - xy'

+ 2y =

n(n

ncnxn

- 1)cnxn-2 -

n=2

+ 2 cnxn

n=l

k=n

00

kCkXk + 2 Ckxk

+ 2)(k + 1)Ck+2Xk -

= (k
k=O

n=O
k=n

'--v--'

'----v-----'

k=n-2

k=O

k=l
00

= 2C2 + 2co + [(k

+ 2)(k + 1)Ck+2 -

(k - 2)qJxk

k=l

Thus
2C2 + 2co = O
(k

+ 2)(k + 1)Ck+2 -

(k - 2)Ck = O

and
C2 =

-Co
k-2

q+2 = (k

Choosing

Co

1 and

C}

+ 2)(k + 1) q,

k=1,2,3,

....

= O we find
C2 =-1

Por

Co

= O and

Cl

C3

C4

=O

C6

= Cs = ClO = ... = O.

Cs

= C7

= ...

=O

= 1 we obtain
C2

C4

= C6 = ... = O

1
6
1
C5=-120
C3 =--

and so on. Thus, two solutions are


Yl = 1 - x2

and

1
Y2 = x - -x
6

245

--x

1
120

= O.

6. 1

Exercises 6. 1
y = ;~=o Cnxn into the differential equation we have

17. Substituting

00

y"

00

L n(n

+ x2y' + xy =

n=2

+L

ncnxn+l

n=l

"-----v---'

k=n-2

k=n+l

00

00

+L

- 1)cnxn-2

Cnxn+l
n=O
~
k=n+l

00

L (k + 2)(k

+ 1)Ck+2Xk + L (k

k=O

00

- 1)Ck_1Xk

k=2

+L

00

= 2C2 + (6C3 + Co)x

+ L [(k + 2)(k + 1)Ck+2 + kCk_l]xk = O.


k=2

Thus
C2

= 06C3 + Co

=O

+ 2)(k + 1)Ck+2 + kCk-l

(k

=O

and
1

C3

= --co
6

Ck+2 = - (k
Choosing

Co

+ 2)(k + 1) Ck-l,

k = 2,3,4, ....

= 1 and q = Owe find

and so on. Por

Co

= O and

Cl

C3

1
=--

C4

= C5 = O

C6

=-

c3

=O

C4

1
=--

C5

= C6 = O

1
45

= 1 we obtain

C7=-

252

and so on. Thus, two solutions are


1 3

Yl = 1- -x
6

1 6
+-x

45

Ck_1Xk

k=l

and

246

1
Y2 = x - -x
6

5
232

+-x

Exercises 6. 1
18. Substituting y = 2:;;":=0 cnxn into the differential equation we have

= n(n

y" + 2xy' + 2y =

- 1)Cnxll-2 + 2

= nc.ix"

n=l

n=2

.____,

k=n-2

k=n

oo

+2

00

= Cnxn

n=O
k=n

'--v-'"

oc

(k + 2)(k + 1)Ck+2Xk + 2
kCkXk + 2
Ckxk
k=O
k=l
k=O
oo

= 2C2 + 2co

+ [(k

+ 2)(k + 1)q+2 + 2(k + l)qlxk

k=l
Thus
2C2 + 2co

=O

(k + 2)(k + 1)Ck+2 + 2(k + l)ck = O


and
C2 =

Ck+2 =
Choosing

co

= 1 and

C1

-Co
-

+ 2 Ck,

k = 1,2,3, ....

= Owe find
C2 =-1

C3

= Cs = C7 = ... = O

C4

C6

1
=--

and so on. For

Co

= O and

C1

"2

= 1 we obtain
= C6 = ... = O

C2

C3

=--

C4

2
3

4
e5 =-

15

8
c-- -_
c
105
r:

and so on. Thus, two solutions are


Y1 = 1 - x 21+41-x6

- -x

+ ...
247

= O.

Exercises 6.1
19. Substituting y

= ~='o Cnxn into the differential equation we have


00

(x - 1)y"

+ y'

00

= L

n(n - 1)cnxn-l
n=,2
k=n-l

00

- L n(n - 1)cnxn-2
n=2
k=n-2

00

+L

ncnxn-l

n=l

k=n-l

00

= L (k

+ 1)kck+lXk

00

+ 2)(k + 1)q+2Xk + L

- L (k
k=O

k=l

(k
k=O

+ 1)Ck+lXk

00

= -2C2

+ Cl + L[(k + 1)kck+l

+ 2)(k + 1)Ck+2 + (k + 1)Ck+lJXk

- (k

= O.

k=l

Thus

+ Cl

-2C2
(k

+ 1)2Ck+l

=O

+ 2)(k + 1)Ck+2 = O

- (k

and

k+1
Ck+2=k+2ck+l'
Choosing

Co

= 1 and

Cl

= Owe find

C2

k=1,2,3,

= C3 = C4 = ... = O. For Co = O and


1

C2

....

= 2'

C3

1 we obtain

3'

C4

= 4'

and so on. Thus, two solutions are


and

Yl = 1
20. Substituting y

Y2

1 2 1 3 1 4
= x + -x
+ -x + -x + ....
2

= ~=o cnxn into the differential equation we have


00

(x

+ 2)y" + xy'

00

- y = L n(n - 1)cnxn-l
n=2
k=n-l
00

= L(k

+L

00

2n(n - 1)cnxn-2

n=2
k=n-2

00

+L

ncnxn - L Cnxn
n=l
n=O
~~
k=n
k=n

00

00

+ l)kck+lXk + L

2(k
k=O

k=l

+ 2)(k + 1)Ck+2Xk +

00

kCkXk - L Ckxk
k=l
k=O

00

= 4C2 -

Co

+L

[(k

+ 1)kck+l + 2(k + 2)(k + 1)Ck+2 + (k

- 1)cklxk = O.

k=l

Thus
4C2 (k

Co

=O

+ 1)kCk+l + 2(k + 2)(k + 1)Ck+2 + (k


248

- 1)Ck = O,

1, 2, 3, ...

Exercises

6. 1

and
C2 =

-ca
4

(k

+ l)kck+l + (k

ck+2=Choosing

Ca

1 and
Cl

and so on. For

ca

Cl

- l)Ck

2(k+2)(k+1)

k = 1, 2, 3, ....

= Owe find

= O,

= O and

C3
Cl

=-

c-0--480

= O,

C4

24'

1 we obtain
C2 = O
C3

=O

C4

= C5

= C6 = ... = O.

Tbus, two solutions are


and

Y2 = ClX.

21. Substituting y = L~=a cnxn into the differential equation we have


00

y" - (x

+ l)y'

- y

L n(n

00

- 1)cnxn-2 -

n=2

n=l

k=n

00

L (k + 2)(k

'---v----"

k=n-2
=

00

L ncnxn

L ncnxn-l
n=l

'-----v--'

k=n-l

00

+ 1)Ck+2Xk

k=a

00

L Cnxn
n=a
k=n

00

L kCkXk ...,...L (k + l)Ck+lXk

k= 1

k=a

00

L Ckxk

k=a

00

= 2C2 -

Cl -

Ca

+ L [(k + 2)(k + 1)Ck+2 -

(k

+ l)Ck+l

k=l
Thus

(k

+ 2)(k + 1)Ck+2 -

and
Cl

C2= -2Ck+2 =
Choosing

ca =

1 and

Cl

(k - l)(Ck+l

+ Ck) = O

+ ca

Ck+l + Ck
k + 2 Ck,

= 2,3,4,

....

= O we find

C2 =

2'

C3

6'

249

C4

=6

- (k

+ l)Ck]Xk = O.

Exercses 6. 1
and so on-, Por

Co

= O and c = 1 we obtain

C3

2'

C4

1
=4

and so on. Thus, two solutions are


1

u: = 1 + 2x + 6x + 6x + ...

and

Y2

= x + 2x + 2x + 4x + ....

22. Substituting y = :L~=o cnxn into the differential equation we have


(x2

+ 1) y"

00

- 6y

00

00

= L n(n - l)enxn + L n(n - 1)cnxn-2


n=2

- 6 L

n=2
k=n-2

k=n
00

cnxn
n=O
'--v---'
k=n

00

00

= L k(k - l)CkXk + L(k


k=2

+ 2)(k + 1)Ck+2Xk -

k=O
00

= 2C2 - 6co + (6C3 - 6Cl)X + L [(k2 - k -

6L

Ckxk
k=O

6) Ck + (k + 2)(k + 1)Ck+2] xk

k=2
Thus
2C2 - 6co = O

(k - 3)(k

+ 2)Ck + (k + 2)(k + 1)Ck+2 = O

and
C2

= 3co

C3

= Cl

k-3
Ck+2 = - k + 1 Ck,
Choosing

Co

= 1 and

Cl

k = 2,3,4, ....

= O we find
C2 = 3
C3

= Cs = C7 = ... = O

C4

=1

C6

1
=--

250

= O.

Exercises 6. 1
and so on. For

Co

= O and

ci

1 we obtain

C4

C3 =

C7

C2

C5

= C6 = ...

=O

= O.

Cg

= ...

Thus, two solutions are


Yl = 1 + 3x

+x4 -

1 6

+ ...

-x
5

and

Y2 = X

+ X 3.

23. Substituting y = L~=o Cnxn into the differential equation we have


00

(x2

+ 2)yll + 3xy'

00

2:

- y =

00

+ 22:

n(n - l)cnxn

n=2

n=2
k=n

00

2:

00

+ 22:

(k
k=O

k=2

00

nCnxn -

n= 1
~
k=n

k=n-2

00

k(k -l)CkXk

+ 32:

n(n - 1)Cnxn-2

+ 2)(k + 1)Ck+2xk + 32:

(4C2 - co)

2:

Ckxk

k=O

k=l

+ (12c3 + 2Cl)X + 2:

+ 2)(k + 1)ck+2 + (k2 + 2k

[2(k

k=2
Thus
4C2 12c3
2(k

Co =

+ 2c

=O

+ 2)(k + 1)Ck+2 + (k2 + 2k

1)

Ck =

and
1

C2

= -Co

C3

Ck+2 =
Choosing

Co

1 and

Cl

1
--Cl

6
k2

+ 2k

C2

1
=4

C3

- 1

- 2(k + 2)(k + 1) Ck,

k =

= Owe find

C5

C4 =--

C7

= ...

96
251

=O

cnxn
n=O
'-v----'
k=n

00

kCkXk -

00

2:

2,3,4,....

1) Ck] xk = O.

Exercises

6. 1

and so on. Por

Co

= O and

CI =

1 we obtain
C2=C4=C6==0
1

C3 =--

6
7

Cs

120

and so on. Thus, two solutions are


YI

= 1 + 4: x - 96 x + ...

1
Y2 = x - -x
6

and

7
+-x
120

24. Substituting y = ~=a cnxn into the differential equation we have


(x2 -

1) y" + xy'

00

- y =

n(n

00

n(n

- l)cnxn -

n=2

+ ncnxn

n=2

n=l

k=n

k(k

~~

k=n-2

CXJ

00

- 1)cnxn-2

00

-l)CkXk

cnxn
n=O
k=n

00

+ 2)(k + I)Ck+2Xk + kCkXk

- (k
k=O

k=2

k=n

00

k=l

00

Ckxk
k=O

00

[-(k

= (-2C2 - ca) - 6C3X +

+ 2)(k + 1)Ck+2 + (k2 -1)

k=2
Thus
-2C2 -

-(k

Co = O

+ 2)(k + 1)Ck+2 + (k

and

ca

= 1 and

CI = Owe

+ I)Ck

C2

= --Ca

C3

=O

k-l
Ck+2 = k + 2 Ck,
Choosing

- l)(k

= 2,3,4,

find
1

C2 =--

C3

= Cs =

C7

= ... = O

C4 =--

252

....

=O

Ck] xk

= O.

Exercses
and so on. For co

= Oand C} = 1 we obtain
C2

= C4

C6

= ... = O

C3

= Cs

C7

= ...

= O.

Thus, two solutions are

Yl
25. Substituting y =

1 2
1 4
1 - -x --x

and

Y2

= x.

I:~=o c,.,xn into the differential equation we have


00

(x - l)y" - xy'

00

+ y = L n(n - 1)cnxn-1

n=2
k=n-l
00

00

00

L(k+1)kck+1Xkk=l

-2C2

00

L n(n - 1)c,.,xn-2 - L ncnXn


n=2
n= 1
'-v--'
k=n- 2
k=n
00

L(k+2)(k+1)Ck+2Xkk=O

+ L cnxn
n=O
k=n

'---v-"'

00

LkckXk+
k=l

LCkXk
k=O

00

+ Co + L[-(k

+ 2)(k + 1)Ck+2 + (k + l)kck+l

- (k - l)Ck]Xk

k=l
Thus

+ Co =

-2C2
-(k

+ 2)(k + 1)Ck+2 + (k

- l)kck+l - (k - l)Ck = O

and
C2

1
= "2co

kCk+1
(k - l)Ck
ck+2 = k+2 - (k+2)(k+
1)'
Choosing

Co =

1 and

C1

Co

= 1,2,3, ....

= Owe find
C2

and so on. For

= O and

Cl

="2'

C4 = O
6'
C2 = C3 = C4 = ... = O. Thus,
C3

= 1 we obtain

Y = el ( 1 +"2x

1 3
+ 6x
+... ) + e2X

and
y' =
The initial conditions imply
y =

el =

el

(x

+ ~x2 + ...) + C2.

-2 and Cz = 6, so

-2 (1 + ~x2 + ~x3 + ...) + 6x = 8x 253

2ex

= O.

6. 1

Exercises 6. 1
26. Substituting y = L~=O cnxn into the differential equation we have

(x+1)y"

- (2 - x)y'

+y

00

L n(n

00

n=2

00

L n(n

- l)cnxn-l

- 1)cnxn-2

n=2

k=n-2

00

00

L(k + l)kck+lxk

k= 1

n=l

k=n-l

00

L ncnxn-l

- 2

L ncnxn

n=O
k=n

n=l
'-..---'

k=n-l

'--....---'

k=n

00

L(k + 2)(k

+ 1)Ck+2Xk - 2 L(k + l)Ck+lXk +

k=O

00

k=O

L cnxn

00

00

k= 1

k=O

L kCkXk + L Ckxk

00

= 2C2 -

+ Co +

2Cl

L[(k + 2)(k

+ 1)ck+2 - (k + l)Ck+l + (k + l)Ck]Xk = O.

k=l

Thus
2C2 - 2Cl

(k

+ Co = O

+ 2)(k + 1)Ck+2 - (k + l)Ck+l + (k + l)Ck

=O

and
C2 =

2'co

ci -

Ck+2 = k

+ 2 Ck+l

+ 2 Ck,

= 1,2,3, ....

Choosing eo = 1 and el = O we find


C2 =

and so on. For eo = O and

Cl =

-2"

C4

1
12

=-

1 we obtain
C2 =

1,

C3

= O,

e4

1
=-4

and so on. Thus,


y

el

1213-14-x
( 1 - -x
2
6

y'

el

(-x

+ -x

+...

12

+ e2 ( x + x 21-4 -x

+ ... )

and

The initial conditions imply


y =

- ~x2
2
el

+ ~x3 + ... ) + e2
3

= 2 and

2 (1 - ~x2
2
')

~x3
6
1
3

= 2 - x - 2x- - -x

e2 =

(1 + 2x

- x3

+ ... ) .

-1, so

+ "'!_x4 + ... ) - (x + x2 _ ~x4 + ... )


12

+ -x5

12

+ ....

254

Exercses
27. Substituting y

= 2:;::='=0 Cnxn into the differential equation we have


00

+ 8y

y" - 2xy'

00

=L

n(n - 1)Cnxn-2

- 2L

n=2

00

nCnxn

+8 L

cnxn
n=O
~
k=n

n=l

'-----'

k=n-2

k=n

00

00

= L (k

+ 2)(k + 1)Ck+2Xk

- 2L

k=D

00

+8L

kCkXk

Ckxk

k=O

k=l
00

= 2C2 + 8co

+ L[(k + 2)(k + 1)Ck+2 + (8 -

2k)Ck]Xk = O.

k=l

Thus
2C2 + 8co
(k

=O

+ 2)(k + 1)Ck+2 + (8

- 2k)Ck = O

and
C2 =

-4co
2k - 8

Ck+2 = (k
Choosing

Co

= 1 and

Cl

+ 2)(k + 1) Ck,

k = 1,2,3, ....

= Owe find
C2 =-4

Por

Co

= O and

Cl

C3

C4

4
=-

C6

= ca =

C5

C7

= ... =

3
ClO

= ... = O.

= 1 we obtain

C5 =-

10

and so on. Thus,


Y = el

( 1-

and
Y , = el ( -8x

4 x 4)
4x 2 + '3

+ e2 ( x

16 x 3) + e2 ( 1 +3

255

1
- x 3 + 10x

1
3x +:2x

+ ... )

4 +... ) .

6. 1

Exercises 6. 1
The initial conditions imply Cl = 3 and C2 = O, so
y
28. Substituting

= 3 (1-4x2

+ ~x4) =

+4x4.

L~=a cnxn into the differential equation we have


00

(x2

3 -12x2

+ l)y" + 2xy' =

L n(n

00

- l)cnxn

n=2

+L

00

n(n - 1)cnxn-2

n=2

k=n

L k(k

2ncnxn

n=l

'--v--'

k=n-2

00

+L

k=n

00

00

k=a

k=l

+ L (k + 2)(k + 1)Ck+2Xk + L

- l)ckxk

k=2

2kCkXk

00

= 2C2 + (6C3 + 2Cl)X

+ L [k(k + l)Ck + (k + 2)(k + 1)Ck+21xk = O.


k=2

Thus

k(k

+ l)Ck + (k + 2)(k + 1)Ck+2

=O

and

Ck+2 =
Choosing ca = 1 and

Cl

= Owe find

C3

- k + 2 Ck, k = 2, 3, 4, ....
= C4 = Cs
C3

= ... = O. Por ca = O and

1
=-_

1
Cs =--;:

C7

=_
7

and so on. Thus

and
Y

, = Cl (1 -

x2 + x4 - x6

256

+... ) .

Cl

= 1 we obtain

Exercises
The initial conditions

imply ca = O and

Cl

71x 7 + ....

y = :Z:::~=acnxn into the differential equation we have

29. Substituting

Y 11

= 1, so

31x 3 +"51x

Y =x -

+ (sism x ) y

= ~
L..- n ( n - 1) Cnx n-2

n=2

~ [2C2 + 6C3X
= 2C2 + (6C3

+ (x

1
- -x
6

3 + -x1

s _...

) ( ca + ClX + C2X2)+ ...

120

+ 12c4x2 + 20csx3 + ...] + [cax + ClX2 + (C2

+ ca)x + (12c4 + Cl)X 2 + ( 20cs + C2 -

1 ) x3
"6ca

- ~ca) x3

+ ... = O.

Thus

6C3 + ca = O

+ ci

12c4
20cs

and

=O

1
-Ca = O
6

+ C2 -

C2 = O
C3

1
= --Ca

C4

1
= -12Cl

1
Cs = - 20C2
Choosing

ca

1 and

Cl

= O we

1
120ca.

find
C4

and so on. For ca

= O and

Cl

C2 =

and so

011.

6. 1

Thus, two solutions

Yl

1 3
1 - -x
6

1
Cs = 120

= O,

1 we obtain

O,

C3

= O,

+ ...

1
C4

= -12'

Cs = O

are

1
+ -x
120

and

257

Y2 =

X -

-x + ....
12

+ ...]

Exercises 6. 1
30. Substituting y = I:;:;O=o
cnxn into the differential equation we have
00

y"

+ eXy'

- y =

I:n(n

- 1)Cnxn-2

n=2

(1 + x + ~x2
2

= [2C2 + 6C3X

+ ~x3 + ...)

(Cl

+ 2C2X + 3C3X2 + 4C4X3 + ...) -

+ 12c4x2 + 20C5X3 + ...]

+ [Cl + (2C2 + cl)x + (3C3 + 2C2 + ~Cl)


= (2C2

+ ci

- ce)

x2

+ ...] - [co + ClX + C2x2 + ...]

+ (6C3 + 2C2)X + (12C4 + 3C3 + C2 + ~Cl)

x2

+ ... = O.

Thus
2C2 + Cl

Co

=O

6C3 + 2C2 = O
1
12c4 + 3C3 + C2 + 2Cl = O
and
C2 =

-Co -

-Cl

1
--C2
3
1
C4 = --C3
4
C3 =

Choosing

Co =

1 and

Cl

+ -C2
12

1
24

-Cl

= Owe find
1

C2 =

and so on. For

f Cnxn

n=O

Co = O and Cl =

2'

C4 = O

1 we obtain
1

C3

= 6'

1
24

C4 =--

and so on. Thus, two solutions are


1 2 1 3
Yl = 1 + -x - -x + ...
2
6

and

1
Y2 = x - -x
2

258

+ -1x 3
6

1 4
- - x
24

+ ....

Exercses
31. Substituting y

= :;:o=o Cnxn into the first differential equation leads to


00

y" - xy =

00

L n(n

- 1)Cnxn-2 -

n=2

L Cnxn+l
n=O
~
k=n+l

k=n-2

00

L (k + 2)(k

00

+ 1)Ck+2Xk

k=O

L Ck_lXk
k=l

00

= 2C2 +

L [(k + 2)(k

+ 1)Ck+2 - Ck_l]Xk =

l.

k=l
Thus

(k

+ 2)(k + 1)Ck+2 -

Ck-l

=O

and
1

C2 =-

k = 1,2,3, ....

Ck+2= (k+2)(k+1)'

Let Co and Cl be arbitrary and iterate to find


1

C2 =

2'

C3 = '6CO
1

C4 = 12cl
1

Cs = 20C2 = 40
and so on. The solution is
Y

1213141

= Co + ClX + 2'X + '6COX + 12 ClX +


13) +...
= Co ( 1 + '6x

Substituting y =

40 Cs +

14) +...
+ Cl ( X + 12x

...

ls + ....
+ '212x + 40x

:;:O=o cnxn into the second differential equation leads to


00

y" - 4xy' - 4y =

L n(n

00

- 1)Cnxn-2 -

n=2

L 4ncnxn
n=l

...._,..._.

k=n-2

259

k=n

00

L 4Cnxn
n=O
k=n

'-----v-'

6. 1

Exercses

6. 1

00

00

+ 2)(k + 1)Ck+2Xk

= (k

00

4kckXk

k=O

4CkXk

k=l

k=O

00

= 2C2 - 4co

+ [(k

+ 2)(k + 1)Ck+2

+ l)cklxk

- 4(k

k=l

=e

= 1 + L..... -k'x

k=l

k
o

Thus
2C2 - 4co = 1
(k

+ 2)(k + 1)Ck+2 -

4(k

+ l)Ck =

1
k!

and
C2 =

"2 + 2co
1

Ck+2 = (k

+ 2)! + k + 2 ck,

= 1, 2, 3,

Let Co and ct be arbitrary and iterate to find


1

C2 =

"2 + 2co
1
3!

4
3

1 4
+ -ct
3! 3

C3

= - + -ct

C4

=-

C5

C6

= 6! + "6C4 =

C7

=-

1 4
1
1
13
+
C2 = - + - + 2co = 4! 4
4! 2
4!
1

51 + 5"C3
1

= 5! + 5 3!
o

16

+ 15c1

4 13 8
4! + "6co

51 + 6
1

+ 2co

4 17
o

64

17

16

261

51 + 15C

51 + 3co
409

64

7! + "7C5 = 7! + 7.5! + 105Cl = T! + 105c1

and so on. The solution is

260

Exercises

Y = Ca + CIX + ( -1 + 2ca ) x 2
2

(261
6!

+ ~CO)X6 +
3

4
= Ca [ 1 + 2x 2 + 2x 4 + -x

(409
7!
6

+ -x21 2 + -x3!1
32. We identify P(x)
x4/5! -"',

= Oand

for [z]

+ -x13
4!

Q(x)

1 + -CI
4)
+ ( "1
3. 3

X 3 - (13
,+
4.

+ ~CI)Xi

+ (17
,+
5.

+ ...] + CI [ X + -x4 3 + -x16 5 + -x64


5
261
+ -x17
+
-x
5!
6!

= sin x [.

16)
-CI
15

X5

+ ".

105

2ca ) x 1

6. 1

15

+ -x409
7!

105

+".

+" ..

The Taylor series representation for sin x/x is 1- x2 /3! +

< oo. Thus, Q(x) is analytic at x

O and x = O is an ordinary point o the

differential equation.
33. The differential equation xy" = O has a singular point at x = O. It also has two solutions, YI = 1
and Y2
34.

If x

= z, that

are analytic at x

> O and y > O, then y"

= O.
< O and the graph o a solution curve is concave down. Thus,

= -xy

whatever portion o a solution curve lies in the first quadrant is concave clown. When x
y < O, y"

= -xy

> O, so whatever portion o a solution curve lies in the fourth quadrant is concave

up.
35. (a) Substituting y = L~=a Cnxn into the differential equation we have
00

y"

> O and

+ xy' + y

L n(n

00

- 1)cnxn-2

n=2

n=l

k=n-2
00

00

+ L ncnxn + L Cnxn

L (k + 2)(k

k=a

n=a
k=n

'---v-'

k=n
00

00

k=l

k=

+ 1)Ck+2Xk + L kCkXk + L Ckxk


00

= (2C2 + ca) + L [(k + 2)(k + 1)Ck+2 +" (k + l)Cklxk = O.


k=l

Thus
2C2 + Ca = O
(k

+ 2)(k + 1)Ck+2 + (k + l)Ck

and

1
C2 = --Ca
2

Choosing Ca = 1 and CI = Owe find

261

=O

Exercises 6. 1

C2 =--

C3

and so on. For

Co

= Oand

Cl

= Cs = C7 = ... = O

= 1 we obtain
C2=C4=C6==O
C3

= --

1
2
=-3
3!

es = -~( -~) = 5~3 =


C7

_~(\!2)

=_

\!2

6.;!. 2

and so on. Thus, two solutions are


~

= k=O
L......t

Yl

(_l)k
2k
2k . k' x

and

4
2

4
2

N=2

4
2 N=10

N=6

4 x

-4

-4

-4

-4

262

Exercises
(e)

yl

y2

4/

2'

6. 1

-2

-2

-4

-4

The graphs of Yl and Y2 obtained from a numerical solver are shown. We see that the partial
sum representations indicate the even and odd natures of the solution, but don't really give a
very accurate representation of the true solution. Increasing N to about 20 gives a much more
accurate representation on [-4,4].
(d) From e" = 2:.'{'=oxk/k!

we see that e-x2/2 = 2:.~o(-x2/2)k/k!

= Lk=O(-1)kx2k/2kk!.

Frorn

(5) of Section 4.2 we have

= (

~ (_l)k
6
2kk!

2k) (~
x

6 (2k + 11)2kk!

2k+l)
x

36. (a) We have


y"

+ (cos x)y

= 2C2 + 6C3X

+ 12qx2 + 20csx3 + 30C6X4 + 42c7XS


2

+ (1 - .:._ +.:._ - .:._+ ... )(co + CJx + C2x2 + C3x3 + C4x4 + csxS + ... )
2!

(2C2

4!

6!

+ ce) + (6C3 + Cl)X + (12q + C2 - ~co )x2 + (20cs + C3 - ~CJ)x3

+ (30C6 + C4 + ~co
24

- ~C2)x4
2

263

+ (42c7 + Cs + ~Cl - ~C3)XS + ....


24

Exercses 6. 1
Then
and
which gives

C6

= -co/80 and

C7

= -19cr!5040.

Thus

Yl(X) = 1 -

2X + 12X

1
- 80X

Y2(X) = X -

6X + 30X

19
- 5040X

+ ...

and
7

+ ....

(b) From part (a) the general solution of the differential equation is 'Y =

y(O) = Cl + C2' 0=
Y = Yl

(e)

and y/(O) =

Cl

+ Y2

= 1+ x

-"2x

Cl .
2

CIYl

6x + 12x + 30x

1
- 80x

19
- 5040x

+"'.

4
2

x
-2
-4

(d)

Then

O+ C2 = C2, so the solution of the initial-value problem is

+ C2Y2.

-6

4 -2

-2

-2

-4

-4

264

Exercises 6.2

Exercises 6.2

1. Irregular singular point: x = O

2. Regular singular points: x = O, -3

= 3; regular

3. Irregular singular point: x

singular point: x

-3

4. Irregular singular point: x = 1; regular singular point: x = O


5. Regular singular points: x = O, 2i
6. Irregular singular point: x = 5; regular singular point: x = O
7. Regular singular points: x = -3, 2
8. Regular singular points: x = O, i

= O;regular singular points: x = 2, 5

9. Irregular singular point: x

= -1;

10. Irregular singular point: x

regular singular points: x

= 0,3

11. Writing the differential equation in the form

we see that Xo

= 1 and

Xo

, +--y=O
x
y " +--y 5
x-1
x+1
-1 are regular singular points. For Xo

= 1 the differential equation

can be put in the form


(x - 1)2y"

+ 5(x

- l)Y'

In this case p(x) = 5 and q(x) = x(x - 1)2/(x

x(x - 1)2

Y = O.
x+1
1). For Xo = -1 the differential equation can be

put in the form


(x

+ 1)2y" + 5(x + 1) x + 1 y' + x(x + l)y = O.

x-1
In this case p(x) = (x + l)/(x - 1) and q(x) = x(x

+ 1).

12. Writing the differep.tial equation in the form


x+3
+ --

y' + 7xy = O
x
we see that Xo = O is a regular singular point. Multiplying by x2, the differential equation can be
y"

put in the form


x2y"

+ x(x + 3)y' + 7x3y

= O.

= x + 3 and q(x) = 7x3.


We identify P(x) = 5/3x + 1 and Q(x) = -1/3x2, so that p(x) = xP(x)
x2Q(x) = -j. Then ao = ~, be = -j ,and the indicial equation is
We identify p(x)

13.

r(r - 1)

+ ~r
3

- ~ = r2 + ~r - ~ = ~(3r2
3
3
3 3

265

+ 2r - 1) = ~(3r
3

- l)(r

~+ x

and q(x) =

+ 1) = O.

Exercises 6.2
The indicial roots are

k and

-1. Since these do not differ by an integer we expect to find two series

solutions using the method of Frobenius,

= l/x and Q(x) = lO/x, so that p(x) = xP(x)

14. We identify P(x)

= 1 and q(x) = x2Q(x)

= 10x.

Then ao = 1, bo = O, and the indicial equation is

+r

r(r - 1)

= r2 = O.

The indicial roots are Oand O. Since these are equal, we expect the method of Frobenius to yield a
single series solution.
15. Substituting y

= L~=o cnxn+,.

into the differential equation and collecting terms, we obtain


00

2xy" - y'

+ 2y

+ L [2(k + r

= (2r2 - 3r) cox,.-l

- l)(k

+ r)ck

- (k

+ r)ck + 2Ck_l]Xk+,.-1

k=l

which implies
2r2 - 3r

= r(2r - 3)

=O

and
(k
The indicial roots are r = Oand r

+ r)(2k + 2r

- 3)Ck + 2Ck-l

= O.

= 3/2. For r = Othe recurrence relation is

Ck =

2Ck-l

- k(2k _ 3)'

k = 1,2,3, ... ,

and
= 2co,

Cl

For r

= 3/2

C2

= -2co,

the recurrence relation is


2Ck-l
Ck = - (2k + 3)k '

and
ci

= -Sco,

C2 =

= 1,2,3,

... ,

2
35 Co,

The general solution on (0,00) is


Y

4 3 +... ) + C2X 3/2 ( 1 - -x


2 + -x2
= Cl ( 1 + 2x - 2x 2 + -x
9

- -x4
945

L~=o cnxn+,. into the differential equation and collecting


+ 5y' + xy = (2r2 + 3r) cox,.-l + (2r2 + 7r + 5) cx,.

16. Substituting y =
2xy"

35

+... ) .

terms, we obtain

00

+ L [2(k + r)(k + r
k=2
= O,

266

- l)Ck

+ 5(k + r)ck + Ck_2]Xk+,.-1

O,

Exercises 6.2
which implies
2r2

+ 3r = r(2r + 3) = O,

(2r2 + 7r + 5)

= O,

Cl

and

+ r)(2k + 2r + 3)Ck + Ck-2 = O.


and r = O, so Cl = O . Por r = -3/2

(k
The indicial roots are r = -3/2

Ck-2
Ck = - (2k - 3)k '

the recurrence relation is

k = 2,3,4, ... ,

and
1

C2

C3 = O,

= -"2co,

Por r = Othe recurrence relation is


Ck =

Ck-2
- k(2k + 3) , k

= 2,3,4, ... ,

and
1

C3 = O,

C2 = ~ 14co,

C4 =

1
616co.

The general solution on (0,00) is


Y
17. Substituting y
4xy"

Clx

1
1 - -x
2

= :;~=o Cnxn+r

+ ~yl + Y =
2

-3/2

+ -x1

40

+... ) + C2 ( 1 -

-x1
14

+ -x1

616

+... ) .

into the differential equation and collecting terms, we obtain

(4r2 - ~r) coxr-l


2

f [4(k + r)(k + r - l)Ck + ~(k2 + r)ck +

k=l

=0,
which implies
4r2 - ~r = r (4r -

+ r)(8k + 8r

+ Ck-l

=O

and
1
"2(k

- 7)q

= O.

The indicial roots are r = O and r = 7/8. Por r = Othe recurrence relation is
Ck =

2Ck_l

k(8k _ 7)'

1,2,3, ... ,

and
Cl

C3= --Co

= -2co,

459

267

Ck-l]

xk+r-l

Exercises

6.2

For r = 7/8 the recurrence relation is


2Ck-l

ck=-(8k+7)k'

1,2,3, ... ,

and
2

Cl

= -15co,

345co,

C2 =

C3

4
= - 32,085 Co

The general solution on (0,00) is


Y = Gl ( 1 - 2x
18. Substituting y

2
+ -x
9

- -x4
459

+... ) + G2X7/8

1 - -x2
15

+ -x2

345

- --x 4
32,085

+... ) .

= :~=o Cnxn+r into the differential equation and collecting terms, we obtain

2x2y" - xy'

+ (x2 + 1) y

= (2r2 - 3r

+ 1) coxr + (2r2 + r)

ClXr+l

00

L [2(k + r)(k

+r -

+ r)ck + Ck + Ck_2]Xk+r

l)Ck - (k

k=2

= 0,
which implies
2r2 - 3r

+ 1=

(2r - l)(r - 1) = 0,

(2r2 +r)

Cl =

+ 2r -

+ l]Ck + Ck-2 = O.

0,

and
[(k + r)(2k

The indicial roots are r = 1/2 and r = 1, so

3)

= O. For r = 1/2 the recurrence relation is

Cl

k=2,3,4,

... ,

and
1

C2

= -"6co,

C3

= 0,

1
C4

= 168 Co

For r = 1 the recurrence relation is


Ck-2
Ck = -k(2k+1)'

k = 2,3,4, ... ,

and
C3

= 0,

The general solution on (0,00) is


Y = Glx

1/2 (

1
1 - -x
6

+ -x1

168

+... ) + G2X (11 268

-x
10

+ -x1

360

+... ) .

Exercises 6.2
19. Substituting y
3xy"

= L:;:O;o Cr,xn+r into the differential equation and collecting terrns, we obtain

+ (2 -

x)y' - y

(3r2

r) coxr-l

00

+ 2:J3(k + r

+ r)ck + 2(k + r)ck

- l)(k

- (k

+ r)ck_1Jxk+r-l

k=l

= O,

which implies
3r2

r = r(3r -1) = O

and
(k
The indicial roots are r

+ r)(3k + 3r

- l)Ck - (k

+ r)ck-l

= O.

= Oand r = 1/3. For r = Othe recurrence relation is


Ck-l
Ck = (3k - 1) )

k=1,2,3,

... ,

and
1

ci

For r

= 2'co,

= 1/3 the recurrence relation is


k

Ck-l

Ck=

3k'

2 3

'=1"

, ... ,

and
1

C]

= '3co,

C2 =

1
18co,

C3 =

1
162co.

The general solution on (0,00) is


Y

1
Cl ( 1 + -x
2

20. Substituting y =

+ -x1

10

+ -x1

80

+... ) + C2X 1/3

1
1 + -x
3

+ -x1

18

+ -x1

162

+".

L:;:O=o Cr,xn+r into the differential equation and collecting terms, we obtain

= O,
which implies

and
[(k

+ r)(k + r

- 1)

269

~J

q-

ck-l

= O.

Exercses 6.2
The indicial roots are r = 2/3 and r = 1/3. For r = 2/3 the recurrence relation is
3Ck-l
Ck = 3k2 + k '

k = 1,2,3, ... ,

and

3
Cl = 4co,

C3

= 560co.

For r = 1/3 the recurrence relation is


3Ck-l
Ck = 3k2 _ k'

k = 1,2,3, ... ,

and
3
Cl = 2'co,

C3

= 160co.

The general solution on (O,(0) is


Y = Clx

2/3 ( 1

+ -x3 + -x9

56

+ -x9

560

+... ) + C2X 1/3 ( 1 + -x3 + -x9

20

+ -x9

160

+... ) .

21. Substituting y = :L~""ocnxn+r into the differential equation and collecting terms, we obtain
00

2xy" - (3

+ 2x)y' + y

= (2r2 - 5r) coxr-l

+ "L[2(k + r)(k + r

- l)Ck

k=l
- 2(k

+r

5r = r(2r - 5)

=O

- 3(k

+ r)ck

- l)Ck-l

+ ck_dxk+r-l

=0,
which implies
2r2

and
(k
The indicial roots are r

+ r)(2k + 2r

- 5)Ck - (2k

= O and r = 5/2. For r

+ 2r

- 3)Ck-l

= Othe recurrence relation is

(2k - 3)Ck-l
Ck = k(2k _ 5) ,

k = 1,2,3, ... ,

and
1

Cl = 3co,
For r = 5/2 the recurrence relation is
Ck =

= O.

2(k + l)Ck-l
k(2k + 5) ,

and

270

k = 1,2,3,. " ,

Exercises

6.2

The general solution on (O,(0) is


Y

= Gl ( 1 + -x1
3

22. Substituting y

1 2 - -x
1 3
- -x
6

+... ) + G2X 5/2

4
1 + -x
7

+ -x4

21

+ -x32

693

+... ) .

= ~=a cnxn+r into the differentia! equation and collecting terms, we obtain

f:

[(k

+ r)(k + r

- l)Ck

+ (k + r)ck

iq +

Ck-2]

xk+r

k=2

=0,
which implies
r2 - ~ = (r
(r2

D D

= O,

(r -

+ 2r + ~) ci

= O,

and
[(k

The indicia! roots are r

= -2/3

and r
Ck =

+ r)2

- ~] q + Ck-2

= 2/3,

so

= O.

= O. For r = -2/3

9Ck-2
3k(3k - 4) ,

the recurrence relation is

k = 2,3,4, ... ,

and
C3

= O,

C4 =

-Ca
128

For r = 2/3 the recurrence relation is


Ck =

3k(3k

+ 4) ,

k = 2,3,4, ... ,

and
C3

= O,

- --Ca
1280
.
,

C4 -

The general so!ution on (O,(0) is


Y

3 2 + -x9
Glx -2/3 ( 1 - -x
4

128

+... ) + G2X 2/3


271

1 - -x3
20

+ --x9

1,280

+... ) .

Exercises

6.2

23. Substituting y = L~=O cnxn+r into the differential equation and collecting terms, we obtain
9x2y"

+ 9x2y' + 2y

= (9r2 - 9r

+ 2) Coxr

00

+ r)(k + r - l)Ck + 2Ck + 9(k +r - 1)Ck_1Jxk+r

+ ~[9(k
k=l

= O,

which implies
9r 2

9r

+2 =

(3r - 1)( 3r - 2)

=O

and
[9(k

+ r)(k + r - 1) + 2lck + 9(k + r = 2/3. For r

The indicial roots are r = 1/3 and r

and

ci

For r

= O.

= 1/3 the recurrence relation is

(3k - 2)Ck-l
k(3k - 1) ,

Ck = -

l)Ck-l

= -2"co,

= 1,2,3, ... ,

C3

7
120

= --Co.

= 2/3 the recurrence relation is


(3k - l)Ck-l
k(3k+1)
,

Ck=and

k = 1,2,3, ... ,

ci =

-2"co,

C3

1
21

= --Co

The general solution on (0,00) is


Y = Clx
24. Substituting

2x2y"

1/3 (

1
1 - -x
2

1 2 - -x7
+ -x
5

120

+...

+ C2X 2/3 ( 1- -x1 + -x5


2

28

- -x1
21

+...

= L~=O cnxn+r into the differential equation and collecting terms, we obtain

+ 3xy' + (2x -l)y

= (2r2

+r

-1) coxr

00

+ ~ [2(k + r)(k + r - l)ck + 3(k + r)ck - Ck + 2Ck_1Jxk+r


k=l

= O,
which implies
2r2

+r

- 1 = (2r - l)(r

+ 1)

=O

and
[(k

+ r)(2k + 2r + 1) - l]Ck + 2Ck_l


272

= O.

Exercises 6.2
The indicial roots are r

= -1

= 1/2. For r = -1

and r
Ck =

2Ck-l

k(2k _ 3)'

the recurrence relation is

= 1,2,3, ...

and
ci

= 2co,

C2

= -2co,

For r = 1/2 the recurrence relation is


2Ck-l
- k(2k + 3) , k

Ck =
and

= 1,2,3, ... ,

ci

= -"5co,

C3 = --Co

945

The general solution on (O,(0) is


Y

= Clx -1 (

25. Substituting y
xy"

4
1 + 2x - 2x 2 + -x
9

+... ) + C2X 1/2 ( 1 -

2
-x
5

+ -x2

35

- -x4
945

+ ....)

= ~=o Cnxn+r into the differential equation and collecting terms, we obtain

+ 2y'

- xy = (r2

+ r)

+ (r2 + 3r + 2) Cl xr

coxr-l

00

+ :L [(k + r)(k + r -

l)Ck

+ 2(k + r)ck

- ck_21xk+r-l

k=2

= 0,

which implies
r2

+ r = r(r + 1) = O,

(r2

+ 3r + 2)

Cl

= O,

and
(k
The indicial roots are

rl

= O and r2

+ r)(k + r + l)Ck

= -1,

so

= O. For

Cl

Ck-2

Ck = k(k

- Ck-2 = O.

+ 1) ,

rl

k=2,3,4,

and
1

C2 =

3fco

C3 = es = C7 =

C2n = (2n

... = O

+ l)!co.
273

= Othe
... ,

recurrence relation is

Exercises

For

1"2

6.2

= -1 the recurrence relation is


Ck-2
Ck = k(k - 1) ,

k = 2,3,4, ... ,

and

C3

= Cs = C7 = ...

C4

4TCQ

1
C2n = (2n)! Co .
The general solution on (0,00) is
Y=

e ~

2n

l;:O(2n+1)!x

[e1;:0~

(2n

= -;;;

1
x

+ l)!x

1
;:O(2n)!x

-1 ~

2X

2n+ 1

ef

+ 2 n=O

2n

1 2n]
(2n)!x

= -[el smh x + e2cosh z].


26. Substituting y = z.:::;:o=o
cnxn+r into the differential equation and collecting terms, we obtain
x2y"

+ xy' + ( x2

- ~) y = (1"2 - ~) cQxr

+ (1"2 + 21" + ~) C1Xr+1

f [(k + 1")(k +

1" -

l)Ck

+ (k + 1")ck -

~2

~Ck

+ Ck-2]

xk+r

=0,
which implies

(1"2 + 21"
and

The indicial roots are

C1

= O,

[(k+T)--4 .) 1] Ck+Ck-2=0.
1"1

= 1/2 and

1"2

= -1/2, so

Cl

Ck-2
Ck = - k (k + 1) ,

274

= O. For

1"1 =

= 2,3,4, ... ,

1/2 the recurrence relation is

Exercses

and
1

C2

= --CO
3!

C3

Cs

C7

= ... = O

(_1)n
C2n = (2n + 1)! en.
For r2

= -1/2 the recurrence relation is


Ck-2

k = 2,3,4, ... ,

q=-k(k-1)'
and
C2 =
C3

1
--cO

21

= Cs

C7

= ...

=O

1
C4 =

4!cO

(_1)n
C2n = -(2n.)1 Ca
The general solution on (0,00) is
y

CIxl/2

f (2n(_1)n+ 1)!x2n + C2X-I/2 f (_1)n


x2n
(2n)!

n=O

= CIX-I/2

n=O

f (2n(_l)n+ 1)!x2n+1 + C2

X-I/2

n=O

f (_l)n
(2n)!

x2n

n=O

= x-I/2[CI sinx + C2 cosx].


27. Substituting y =

I::~=o Cnxn+r into the differential equation and collecting terms, we obtain
00

xy" - xy'

+ y = (r2 - r) coxr-l +

.L [(k + r + l)(k

+ r)ck+1 - (k + r)ck + q]xk+r = O

k=O

which implies
r2 - r = r(r - 1) = O
and

(k + r + l)(k + r)ck+1 - (k + r - l)Ck = O.


The indicial roots are

rl

= 1 and r2 = O. For

rl

= 1 the recurrence relation is

kci.
Ck+1 = (k+2)(k+l)'

k=0,1,2, ... ,

275

6.2

Exercises 6.2
and one solution is YI = CoX. A second solution is
Y2 = X

J -J

-dx

e x2

dx = x

J :2
x

(2_x2 +.!. +.!.2 +!..x3!

= xJ

1 (
dx = x J x2 1 + x

+ !"x2 + ...) dx

1 ')
+ -x1
2
12

= x ln x - 1 + -x-

=x

4!

+ -x1

72

+ 2x2 + 3!x3 + ...


[_.!.+ lnx
x

+.!.x
2

dx

+ 2_x2 + 2_x3 + ...]


12

72

+ ....

The general solution on (0,00) is

28. Substituting y
y"

= I:~="o cnxn+r into the differential equation and collecting terms, we obtain
3

+ -;yl

- 2y = (r2

+ 2r)

coxr-2

+ (r2 + 4r + 3)

CIXr-1

00

L [(k + r)(k

+r

- l)Ck

+ 3(k + r)ck

- 2ck_21xk+r-2

k=,,2
= O,

which implies

+ 2r

r2

(r2
(k
The indicial roots are TI = O and

+ 2) = O

= r(r

+ 4r + 3)

CI

+ r)(k + r + 2)Ck -

r2 =

2Ck-2 = O.

-2, so c = O. For

2Ck-2
Ck = k(k + 2)'

rl = O the

recurrence relation is

k = 2,3,4, ... ,

and
1

C2

= ::tco

C3= C5

= C7= ... = O

1
C4

= 48co

1
1,152

C6= --Co

The result is
Yl

1214+
1 -x
= Co( 1 + -x
4

48

276

+ --C6
+
1,152

oo'

Exercises 6.2
A second solution is
Y2 = YI J

= YI

= YI J

e- J(3jx)dx
2
dx = YI
YI

dx
x3 (1 + Ix2
+
~x4
2
48

+ _!_x6
+ ...) = YI
576

(2_x3 - .2_2x + _2'_x48

~X3
576

+y

= -~YIlnx

[__

dx
x3

(1 + Ix2
+ _'!_x4
+ ...)
4
48

+ ...) = YI

+ _2'_x2 _

1_

2x2

[ __

1 (
1 2
x3 1 - -2x

1_ - ~

2x2

lnx

7 4
19 6
)
+ 48
x + 576x + ...

+ _2'_x2 96

~x4
2,304

+ ...]

+ ...] .

~x4
2,304

96

The general solution on (0,00) is

29. Substituting y = :;;:o=o

Cr,xn+r

into the differential equation and collecting terms, we obtain


00

xY"

+ (1-

x)Y' - y

= T2cOXr-1 + L[(k

+ T)(k + T -

l)Ck

+ (k + T)Ck -

(k

+ T)ck_Ilxk+r-1

k=O
which implies

T2 = O

The indicial roots are

and

TI =

T2 = O and the recurrence relation is


Ck =

Ck-I
k
'

k = 1, 2, 3, ....

One solution is
YI = Co ( 1 + x

+ -x21213+ IX
+... )
3.

= coe

A second solution is
Y2 = YI

e- J(ljx-l)dx
e2x

= e xJ1( -

1- x

dx = eX

+ -x1213 2

= eX[l nx - x + -x1

J eXIx
J 1
e2x dx = e" -;;e-xdx

-x
3!

2 - --x1
2 2
3 .31

+ ... )

dx

= e xJ(l - - 1 + -x1 - -x12 + ... ) dx


x

3!

3 + ... ] = e xl nx - e x~(_l)n+1
L...,
x n.
n=1 n n!

The general solution on (0,00) is

277

= O,

Exercises 6.2
30. Substituting y

= I:~=,QCnxn+r into the differential equation and collecting terrns, we obtain


00

xY"

+ y' + y

+ [(k + r)(k + r

= r2cQxr-1

- 1)Ck + (k

+ r)Ck + Ck_l]Xk+r-l = O

k=l

= O and

which implies r2

The indicial roots are

"i

= T2 = O and the recurrence relation is


Ck-l
Ck = - k2'
k = 1,2,3, ....

One solution is
Yl

= ca

( 1- x

+ 22x

1
- (3!)2x

A second solution is
Y2

e- f(l/x)dx

= u.

= Yl

J x ( 1 - 2x + 2x2 -

dx

Yl

J
u: J (~ +

1(
5
;; 1+2x+2'x

= Yl lnx

= y1ln x

+ (4!)2x

2 + ~x

ca

i: (-1)n
(n!)2

n
x .

dx

(1 - x + ix2

- 3~x3

+ ...)

35
+ 288x4

5
gX3

23
+gX

- ...

4
)
+ 677
288x + ...

dx

+ 23x2 + 677 x3 + ...) dx

288

27

+ --x677

<1

1,152

5 2 23
+ -x
+ -x

( 2x

-...

dx

5 2 23
+ 2x + -x
+ -x

+ Yl

J
X

=Yl
=

2
Yl

27

+ --x677

+ ... ]
4

1,152

+... ) .

The general solution on (0,00) is

31. Substituting y

= I:~=a

cnxn+r into the differential equation and collecting terrns, we obtain


00

xy"

+ (x

- 6)y' - 3y = (r2 - 7r)caxr-1

+ [(k + r)(k + r k=l

- 6(k

+ r)ck

- 3Ck_dxk+r-1

which implies
r2

7r

= r(r
278

- 7)

=O

O,

1)Ck + (k

+r -

1)Ck-l

Exercises 6.2
and
(k + r)(k + r - 7)Ck + (k + r - 4)Ck-1 = 00
The indicial roots are r1

= 7 and r2 = 00 For r1 = 7 the recurren ce relation is


(k+ 7)kck

+ (k + 3)Ck-1

=0,

k=l,2,3,ooo,

or
k
Taking co

= 1,2,3,

=f. O we obtain
1

C1 = --Co

C2

5
18cO

1
C3 = -6"co,
and so on. Thus, the indicial root r1 = 7 yields a single solution.
relation is
k(k - 7)Ck

+ (k

- 4)Ck-1 = O,

Now, for r2 = O the recurrence

k = 1,2,3,

Then
-6C1 - 3co = O

-12c3-C2=0
-12c4 + OC3 = O

===}

C4

=O

+ C4 = O

===}

Cs

=O

=O

===}

C6

=O

OC7 + 3C6 = O

===}

C7 is arbitrary

-lOcs

-6C6 + 2cs

and
k-4
Ck = - k(k _ 7) Ck-1,

279

k = 8,9, lO,

Exercises 6.2
Taking Co 1= O and C7 = Owe obtain
1

Cl = --Co

2
1
C2= lOcO
1

C3 = --Co
120
C4

Taking

= es = C6 = ... = O.

Co = O and C7 1= O we obtain

ca = --C7
2

Cg =

ClO

1
= - 36c7'

In this case we obtain the two solutions


1
1 2
1 3
Yl = 1 - "2x + 10x - 120x
32. Substituting y

= ~=o

x(x - l)y"

+ 3y' -

cnxn+r

5
36c7

Y2 = X

and

1 a
-x
2

+ -x5

36

1
- -x
36

10

+ ....

into the differential equation and collecting terms, we obtain

2y
00

= (4r - r2) coxr-l

+ [(k + r

- l)(k

+r

- 12)Ck_l - (k

+ r)(k + r

k=l

= O,

which implies
4r - r2 = r( 4 - r)

and
-(k
The indicial roots are

+ r)(k + r
rl

= 4 and
-(k

- 4)Ck + [(k
r2

+r

- l)(k

= O. For rl = 4 the

+T

or
k=1,2,3,.

280

2) - 2]Ck-l

= O.

recurrence relation is

+ 4)kck + [(k + 3)(k + 2) ck=(k+1)Ck_l,

2]Ck-l = O

- l)Ck

Exercises 6.2
Taking

Co =1=

Owe obtain
CI =

and so on. Thus, the indicial root

2co

C2

= 3!co

C3

= 4!co,

= 4 yields a single solution. For

TI

T2

= O the recurren ce relation

is
-k(k

- 4)Ck

+ k(k

- 3)Ck-1

= O, k = 1,2,3, ... ,

or
-(k-4)Ck+(k-3)Ck_1

=0,

k=

1,2,3, ....

Then
3CI -

2co = O

2C2 - CI = O
C3

+ OC2

OC4

=O

+ C3 = O

=:>

C3 =

=:>

C4

is arbitrary

and
Ck =
Taking

Co =1=

O and

C4 =

(k-3)Ck_1
c-4

k = 5,6, 7, ....

Owe obtain
2

C2

= -Co

3
1
= "3co

C3 =

Taking

Co =

O and

C4 =1=

C4 = C5 = ... = O.

Owe obtain
CI = C2 = C3 =

In this case we obtain the two solutions


2
1 2
YI = 1 + "3x + "3x
and
33. (a) Frorn t

= l/x we have dt/dx

= _1/x2

5364
Y2 = x 42+
x7

= _t2

Then

281

+ ....

Exercises 6.2

dy
dx

dy di
dt dx

_=

-_

dy
dt

-t -

and

Now
x4 ~d2
dx

+ Ay

1 (d2
t4 __J!_2
dt

+ 2t3 ddY ) + Ay

= 4"

d2 + -2 _}j_
d + Ay = O
-i
dt
t dt

becomes
d2y

t -d
(b) Substituting y
d2y
t -d 2

dy

+ 2 -dt + >..ty

= O.

= L~=Ocntn+r into the differential equation and collecting terms, we obtain


dy

+ 2 - + Aty = (r2 + r)cotr-l


dt

+ (r2 + 3r + 2)Cltr

00

+ L [(k + r)(k + r -

l)Ck

+ 2(k + r)ck + ACk_2]tk+r-1

k=2
= O,

which implies
r2

+ r = r(r + 1)

(r2

+ 3r + 2)

= O,

CI = O,

and
(k

+ r)(k + r + l)Ck + ACk-2

The indicial roots are r 1 = Oand


Ck =

r2 =

-1, so

ACk-2
k( k + 1) ,

CI = O.

For

C3

O.

rl = Othe

k = 2,3,4, ... ,

and
C2 =

A
3!

--ca

= C5 = C7 = ...

282

=O

recurrence relation is

Exercises
For

r2 =

6.2

-1 the recurrence relation is


.ACk-2
Ck = - k( k - 1) ,

= 2,3,4, ... ,

and
.A

C2

= --ca
2!

C3

= Cs =

C7

= ... = O

The general solution on (0,00) is

~[CI
t

f (2n(-lt+ 1)! (V>:t)2n+1 + C2 f (_l)n


(2n)!

n=

(V>:t)2n]

n=

= ~[CI sin v>: t + C2 cos v>: t l.


t

(e) Using t = l/x, the solution of the original equation is


v>:
sin -

y(X)CIX
34.

+ C2X

v>:
cos -.
x

(a) Frorn the boundary conditions y(a) = O, y(b) = O we find


v>:

CI sin -

+ C2 cos -v>:
a

v>:

= O

v>:

CI sin -b- + C2 cos -b- = O.


Since this is a homogeneous system of linear equations, it will have nontrivial solutions if
. v>:

Slll-

.~

sm-

v>:
cosa

~
cosb

. v>:
v>:
v>: . ~
= sm - cos - cos - Sllla
b
a
b

=sin(~

- ~)

283

=sinv>:C~ba)

=0.

Exercises 6.2
This will be the case if

v0.

or

n ,'\ab

=b_

nAab

= L'

n = 1, 2, ... ,

or, if

The criticalloads
we have

are then Pn = n27f2(a/b)2 E/o/ L2.


y

= Clx[sin

v0. _
x

= C3X [Sin

and

Using C2

sin( v0./a) cos v0.]


cos( v0./ a)
x

v0. cos v0. _ cos v0. sin v0.]


x

(1 1) = C4xsin--n7fab ( 1-

Yn(x) = C3xsin--n7fab - - L

(b) When n

= 1,

= 11, and

= -Cl sin( v0./a)/ cos(v0./a)

1, we have,

-a) .
x

for C4 = 1,
Yl(x) = x sin 1.l7r

(1 - ~) .

35. Express the differential equation in standard form:

ylII + P(x)y" + Q(x)y' + R(x)y = O.


Suppose Xo is a singular point of the differential equation. Then we say that Xo is a regular singular
point if (x - xo)P(x),

(x - xo)2Q(x), and (x - xo)3R(x) are analytic at x = xo.

36. Substituting y = 2:~=o cnxn+r into the first differential equation and collecting terms, we obtain
00

x3y" + y = coxr +

L [Ck + (k +

7" -

l)(k + 7"

2)Ck_lJxk+r

k=l

It follows that Co

= Oand
Ck =

-(k + 7"

l)(k + 7"

The only solution we obtain is y(x) = O.

284

2)Ck_l'

= O.

Exercses
Substituting y

6.3

= L~=O Cnxn+r into the second differential equation and collecting terms, we obtain
00

x2y"

+ (3x

- l)y'

+y

+ 2: [(k + r + 1)2Ck -

= -rco

(k

+ r + l)Ck+dxk+r

= 0,

k=O
which implies

+ r + 1)2ck

(k
If en

-rco

- (k

+ r + l)Ck+l = O.

= 0, then the solution of the differential equation is y

= O. Thus, we take r = 0, from which

we obtain
Ck+l = (k+ l)Ck,
Letting

Co

is then y

= 1 we get

Cl

= 2,

= L~=O(n + l)!xn,

C2

= 3!,

C3

k = 0,1,2,

....

= 4!, and so on. The solution of the differential equation

which converges only at x

37. We write the differential equation in the form x2y"

= O.

+ (b/a)xy' + (c/a)y

= O and identify ao

= b/a

and bo = e] a as in (12) in the texto Then the indicial equation is


r(r - 1)

+ -a r + -a =

or

ar2

+ bxy' + cy

which is also the auxiliary equation ofax2y"

+ (b -

a)r

= O.

Exercises 6.3
1. Since v2

= 1/9 the general solution is y = clJl/3(X)

+ C2Ll/3(X).

3. Since v2

= 25/4 the general solution is y

+ c2 Yl (x).
= qJS/2(x)
+ c2Ls/2(x).

4. Since v2

= qJl/4(X)' + c2L1/4(X).

5. Since v2

2. Since v2 = 1 the general solution is y = Cl JI (x)

6. Since v2
7. Since v2
8. Since v2

1/16 the general solution is y

+ C2Yo (x).
= 4 the general solution is y = clJ2(X) + C2Y2(X).
= 2 the general solution is y = qh(3x)
+ C2Y2(3x).
= 1/4 the general solution is y = qJl/2(6x)
+ C2J_l/2(6x).

9. If y = x-l/2v(x)

+ c = 0,

the general solution is y = q Jo (x)

then

'21x-3/2v(x),

y' = x-l/2v'(x)

y" = X-l/2v"(x)

_ x-3/2v'(x)

285

+ ~x-S/2v(x),
4

Exercises

6.3

and
x2y"

+ 2xy' + x 2x2y

= x3/2v"

+ X 1/2v' + ( x 2x3/2

- ~X

-1/2) v.

Multiplying by xl/2 we obtain


X2V" +xv'
whose solution is
10. From y

= xnJn(x)

+ ()..2x2

+ c2J_1/2()..X).

= c1J1/2()..X)

-~)

V = O,

Then y = cx-1/2J1/2()..X)

+ C2X-1/2J_l/2()..X).

we find
and

Substituting into the differential equation, we have


xn+l J~

+ 2nxn

+ n(n

J~

= xn-l

- 1)xn-1 Jn

.O

+ (1 -

2n)(xn J~

+ nxn-1

Jn)

+ xn+l

Jn

(sinee Jn s a solution of Bessel's equation)

= O.

Therefore, xn Jn is a solution of the original equation.


11. From y = x-n Jn we find
,
y=x

-nJ' n-nx

-n-lJ

Substituting into the differential equation, we have


xy"

+ (1 + 2n)y' + xy

= x-n-1

[x2J~

= x-n-1

+ xJ~ + (x2

(sinee Jn is a solution of Bessel's equation)

= O.

Therefore, x-n Jn is a solution of the original equation.


12. From y = Vx Jv()..x) we find
y'

= )..v'xJ~()"x)

- n2) Jn]

+ ~x-l/2Jv()..x)

and

286

Exercises 6.3
Substituting into the differential equation, we have

x2y" + (A2X2 - v2 +

y = JX [A2x2J~(AX) + AxJ~(AX) + (A2x2 - v2) Jv(AX)]

JXo

(since Jn is a solution of Bessel's equation)

= o.

Therefore,

JX

Jv(AX) is a solution of the original equation.

13. From Problem 10 with n = 1/2 we find y

= x1/2J1/2(X).

Frorn Problem 11 with n

= xJ1(x).

Frorn Problem 11 with n

= -1/2

we find

y = x1/2J_1/2(X).
14. From Problem 10 with n = 1 we find y

y = xJ-l(X)

From Problem 11 with n = 1 we find

-X-1j_1(X).

JX Jo(2x).
we find y = JX J3/2(X)

16. Frorn Problem 12 with A = 2 and v = O we find y


17. From Problem 12 with A = 1 and v = 3/2

and y =

JX J-3/2(X).

= 3 we find y = x3 J3 (x). From Problem 11 with n

18. From Problem 10 with n

-1 we find

-xJl(X).

15. From Problem 10 with n = -1 we find y = X-1j_1(X).

y = X-1Jl(X)

= -3 we find

= X3j_3(X) = -x3]s(x).

19. (a) The recurrence relation follows from

fa
00

-vJv(x)

+ xJv-l(X)

=-

=_

(-l)nv
(x)2n+v
+ v + n) '2

(::.)2n+v +

00
(-l)nv
fan!r(l+v+n)

+ v) (::.)2n+v _

(-1)n(2n

- n=O
L n.Ir( 1 + v + n )

00

00

(_l)n

(x)2n+v-l

+ x ~o nlr(v + n) '2

nlf(l

00
(_l)n(v
fan!f(l+v+n)

+ n)

.2 (::') (::.)2n+v-l
2
2

- xJv(x).

(b) The formula in part (a) is a linear first-order differential equation in Jv(x).

An integrating

factor for this equation is z", so

ddx[xvJv(x)] = XVJv-l(X).
20. Subtracting the formula in part (a) of Problem 19 from the formula in Example 4 we obtain
0= 2vJv(x)

- xJv+1(x) - xJv-1(x)

or

21. Letting v = 1 in (15) in the text we have

xJo(x)

dx[xJl(X)]

so

287

2vJv(x) = xJv+l(X) + xJv-l(X).

Exercises 6.3
22. From (14) we obtain Jb(x) = -JI (x), and from (15) we obtain J(x) = Ll(X).

LI(X)

= -JI

(x).

23. Since
1

r ( 1 - "2 + n

(2n - 1)!

(n _ 1)!22n-I

we obtain

24. (a) By Problem 20, with

1/

= 1/2, we obtain

JIj2(X) = xJ3j2(X) + xJ_Ij2(X)

hj2(X)=Y;;; [2 (SinX
-;--cosx
with

1/

= -1/2

we obtain -J_1j2(X)

1/

= xJ1j2(X) + xJ_3j2(X)

J_3j2(X) =
and with

-ff

so that

so that

(CO;X + sinx);

= 3/2 we obtain 3J3j2(X) = xJSj2(X) + xJ1j2(X) so that

(h)

-0.5
-1

-1
y

'1

o.5

-0.5
-1

'1

L\ v ~
C>.-3/2C">
5 'Cl10

""B

0.5 /""'\

20 x

v C>.
~ 5/2

C">

25. Letting

we have

dx = dx ds = dx [~ {k (_~) e-o.tj2] = dx (_
dt
ds dt
dt a. Y -;;;,
2
ds

288

(k e-o.tj2)

Y -;;;,

Thus Jb(x)

Exercises

and
d2x = .!!:._ (dX) = dx (::
dt2
dt dt
ds 2

(k e-at/2) + .!!:._

Y -:;

(dX)
ds

dt

(_

(k e-at/2)

Y -:;

= dx (:: (k e-at/2) + d2x ds (_ (k e-at/2)


ds

Y -:;

Y -:;

ds2 dt

= dx (:: (k e-at/2) + d2x (!5_ e-at) .


ds

Y -:;

ds2

Then
m -d2x

dt2

+ ke-atx

= ke-at -d2x

ds2

+ -ma
2

If
-

e-at /2 -dx
dt

+ ke-a

t x = O.

Multplying by 22/ a2m we have


22 k e-at d2x
a2 m
ds2

+ ~ {k ="? dx + .2_ !5_ e-atx


a

Y -:;

di

=O

a2 m

or, since s = (2!a)Nme-a.t/2,


2 d2x
s ds2

dx

+ s ds + s

x = O.

26. Differentiating y = x1/2w (ax3/2) with respect to ax3/2 we obtain


y' = x1/2w' (~ax3/2) ax1/2

+ ~x-1/2w

(~ax3/2)

and

+ aw'

y" = axw" (~ax3/2) ax1/2

+ ~aw'

Gax3/2)

(~ax3/2)

- ~x-3/2w Gax3/2)

Then, after combining terms and simplifying, we have


y"

+ a2xy

= a [~x3/2wll

+ ~w' + (ax3/2

- 4a~3/2) w] = O.

Letting t = ~ax3/2 or ax3/2 = ~t this differential equation becomes


~ ~ [t2wlI(t)

+ tw'(t) + (t2

- ~) w(t)]

= O,

t O.

27. The general solution of Bessel's equation is

Thus, the general solution of Airy's equation for x


Y

=x

1/2w

> O is

(2"3ax 3/2) = C1X1/2J1/3 (2s=3/2) + C2X1/2J-1/3 (2"3ax 3/2) .


289

6.3

Exercises 6.3
28. Setting y = ft JI (2ft)

and differentiating we obtain

y' = ft J~(2ft)

2e
2yx

le JI (2ft)
2yx

= J{(2y'X)

le JI (2y'X)
2yx

and
"

"(

"(

1,

y = JI 2y X ) 2y'X + 2y'X JI (2y X ) 2ft


=

ftJ1

el,

2yx)+2xJ1(2yx)-

1
(e
- 4x3/2 JI 2y X )

4x3/2Jl(2yx).

Substituting into the differential equation and letting t = 2y'X we have

xy" + y = y'X J~'(2y'X) + ~J{(2Vx)


2

Jx

t2 [t2"
JI

[XJ~'(2y'X) +

le h(2y'X)

4yx

V; J;(2y'X)

+ (x -

+ y'X Jl(2y'X)

JI (2y'X)]

1) (]

(t) + 4t, JI (t) + (t2 - 4 JI t)

2..
[ t2 J;' (t) + tJU t) + (t2 2t

1) JI

(t )l.

Since h(t) is a solution of t2y" + ty' + (t2 - l)y = O, we see that the last express ion aboye is Oand
y = ft JI (2ft) is a solution of xy" + y = O.
29. (a) Using the expressions for the two linearly independent power series solutions, Yl(X) and Y2(X),
given in the text we obtain

P6(X) = 116(231x6 - 315x4 + 105x2 - 5)


and

P7(X) = 1~ (429x7 - 693x5 + 315x3 - 35x) .


(b) H(x)

satisfies

(1- x2)

y" - 2xy' + 42y = Oand P7(X) satisfies

30. The recurrence relation can be written


2k + 1
k
Pk+l (x) = -k-- XPk(X) - -k- Pk-l (x),
+1
+1

290

(1- x2)

y" - 2xy' + 56y = O.

k = 2, 3, 4, ....

Exercises

k = 6: P7(X) = 13x (231 x6 _ 315x4 + 105x2 _ ~) _ ~ (63 x5 _ 35x3


7
16
16
16
16
6 8
4
429 7 693 5 315 3 35
= -x
- -x
+ -x
--x
16
16
16
16

6.3

15X)
8

31. If x = cos then


dy
. dy
= - smede
dx'

d2y
. 2 d2y
dy
de2 = sm e dx2 - cos dx '
and
d2y
sine de2

+ cose dy
de + n(n + l)(sine)y

[(

= sine

2 ) d2y
dy
1- cos e dx2 - 2cose dx

+ n(n + l)y ] = O.

That is,
d2y
( 1 - x 2) -dx2
32. Frorn

y = u/vx

dy
2xdx

+ n(n + l)y

= O.

we find
2xu' - u
y' = -2-x3;C-/""'2-

and

Then
2 "
x y

+ xy +

(2
2)
4x5/2u" - 4x3/2u'
x - u y=
4x

+ 3xl/2u

2xu' - u
2x1/2

(x2;2)u
xl/2

= x3/2u" + ~x-l/2u + x3/2u - ;2X-l/2u = O.


4

Multiplying by xl/2 we obtain


x 2"u

1 + x 2u + -u
4

2u =

or
d2u
dx2

+ (1-

;2

-1/4)
x2

u= O.

For large values of x this equation can be approximated by u"

+u

= O whose solutions are sin x

and cos x. Thus, we expect the solutions to oscillate with increasing frequency for larger x.
33. Rolle's theorem states that for a differentiable function f(x),
exists a number e between a and b such that f'(c)
JO(x) = h(x).
for which h(x)

for which f(a)

= O. Frorn Problem 20 with

f(b) = O, there
;

= O we have

Thus, if a and b are successive zeros of Jo(x), then there exists a e between a and b

= J6(x) = o.
291

Exercises 6.3

34. Since

!J; f(x) dx! ::; J; If(x)1 dx, we have


11n(x)l::; ~ rlcos(xsint-nt)ldt::;

Kh

35. (a) We identify m

= 4, k = 1, and

o:

= 0.1.

r 1dt=

K)O

l.

Then

x(t) = cllo(lOe-o.05t) + c2Yo(10e-005t)


and

x'(t) = -0.5Cll(10e-0.05t) - 0.5e2Y(lOe-0.05t).


Now x(O)

= 1 and x'(O) = -1/2 imply


cllo(lO)

+ c2Yo(10) =

ql(lO)

+ c2Y(10) =

1.

Using Cramer's rule we obtain


Y(10) - Yo(10)
el = lo(10)Y(1O) - l(10)Yo(1O)
and
10(10) - lb(10)
C2= lo(10)Y(10) - lb(lO)Yo(lO) .
Using Y = -Yr and l = -11 and Table 6.1 we find el

= -4.7860 and C2= -3.1803. Thus

x(t) = -4.78601o(lOe-o.05t) - 3.1803Yo(lOe-0.05t).


(b)

10
5

-5

50

36. (a) Identifying o:

100

150

200

1, the general solution of x" + ttx


x(t) = Clxl/2 11/3 (~x3/2)

Solving the system x(O.l)


(b)

= O is

+ C2xl/2]_1/3 (~x3/2) .

= 1, x'(O.l) = -~ we find

-1

292

= -0.809264 and C2= 0.782397.

Exercises 6.3
37. (a) Letting t

= L - x, the boundary-value problem beeomes

d2
dt2

+ ohe = O,

e/(o) = O,

e(L) = O,

= Og/E l. This is Airy's differential equation, so by Problem 27 its solution is

where 0:2

= cltl/2Jl/3

+ C2tl/2J_l/3

Go:t3/2)

Go:t3/2)

= CJel(t) + C2e2(t).

(b) Looking at the series forms of el and e2 we see that el (O)


boundary eondition e/rO)

= o implies

= O, and

el

e(t) = C2VtJ-l/3

=1=

O, while e2(0) = o. Thus, the

so

Go:t3/2).

From e(L) = O we have


c2VLJ-l/3
so either

C2 =

O, in whieh

case

GO:L3/2)

= O,

e(t) = O, or J_l/3(~o:L3/2)

= O. The eolumn will just start

to bend when L is the length eorresponding to the smallest positive zero of J-l/3.
Mathematica,

the first positive root of Ll/3(X)

is Xl ~ 1.86635. Thus ~o:L3/2

Using

= 1.86635

implies
L =
(

3(1.86635)
)
20:

2/3

= [ 9El (1.86635)2 ] 1/3

40g

= [9(2.6

X 107)7l'(0.05)4/4(1.86635)2] 1/3
4(0.28)7l'(0.05)2

38. (a) Writing the differential equation in the form xy"

+ (PL/M)y

76.9 in.
= O, we identify A

PL/M.

From Problem 28 the solution of this differential equation is


y = Cl.xJl
Now

h(O)

+ C2.xYl

(2VPLX/M)

(2VPLX/M).

= O, so y(O) = O implies C2 = O and

y = Cl.xJl
(b) From y(L) = Owe have y = JI (2L..fPM)

(2VPLX/M).
= O. The first positive zero of JI is 3.8317 so, solving

2LV PI! M = 3.8317, we find PI = 3.6705M / L2. Therefore,


Yl(X) = Cl.xJl

2 J3.6705X
L

) = Cl.xJl

293

(3.8317)
-L-.x

Exercises 6.3

(e) For

el

= 1 and L = 1 the graph of Y1 = y'X J1 (3.8317 y'X)

is shown.
0.2
0.1
x

0.2

39.

Zeros and differences of Jo, J1, and

0.4

0.6

0.8

Jo

.Jo

JI

.J1

J2

.J2

2.4048
5.5201
8.6537
11.7915
14.9309

3.1153
3.1336
3.1378
3.1394

0.0000
3.8317
7.0156
10.1735
13.3237

3.8317
3.1839
3.1579
3.1502

0.0000
5.1356
8.4172
11.6198
14.7960

5.1356
3.2816
3.2026
3.1762

Successive zeros of Jn for n = O, 1, 2 appear to be approximately equally spaced for larger values
of x. Furthermore, this spacing appears to be approximately
can be used to determine that successive roots of Jo near x

it .

To check this conjecture a CAS

= 200 are 200.277156 and 203.418739,

with difference 3.141583. This is consistent with the observation in Problem 32 that the frequency
of the solutions for large x is approximately 1/27r.
40. (a) Using a CAS we find
1 d1

P1(x) =

"2 dx1

P2(x)

1d22
212
222! dx2 (x - 1) = "2(3x - 1)

P3(X)

1 d3
2
3
233! dx3 (x - 1)

= "2(5x - 3x)

P4(x)

1 d4
244! dx4 (x2 - 1)4

= S(35x4

(x - 1) = X

d5

30x2

+ 3)

P5(X) = 255' dx5 (x2 _1)5 = S(63x5 - 70x3 + 15x)


1d62

616

P6(X) = 266! dx6 (x - 1) = 16 (231x - 315x + 105x - 5)


1

d7

P7(X) = 277' dx7 (x - 1) = 16 (429x - 693x + 315x - 35x)


294

Exercises 6.3

(b)

P1
1

P2
1

0.5
-1

0.5

1 x

-1

-0.5
-1
P5
1

-1

-1
P7
1

P6
1

0.5

0.5

-0.5
-1

-1

-1

(e) Zeros of Legendre polynomials

H(x) : O
P2(X) : 0.57735
P3(X) : O, 0.77460
P4(X) : 0.33998, 0.86115
P5(X) : O, 0.53847, 0.90618
P6(X) : 0.23862, 0.66121, 0.93247
P7(X) : O, 0.40585, 0.74153 ,0.94911
PlO(X):

0.14887, 0.43340, 0.67941, 0.86506, 0.097391

The zeros of any Legendre polynomial are in the interval (-1, 1) and are symmetric with respect
to O.

295

Chapter 6 Review Exerc;ses

Chapter 6 Review Exercises

1. The interval of convergence is centered at 4. Since the series converges at -2, it converges at least
on the interval [-2,10).

Since it diverges at 13, it converges at most on the interval [-5,13). Thus,

at -7 it does not converge, at O and 7 it does converge, and at 10 and 11 it might converge.
2. We have
x3

sin x

x-

f ()x =--=
cosx

3. Solving x2 - 2x

+ 10 =O we obtain

radius of convergence is
4. Setting 1 - sin x

1-

x5

6 + 120 x

'2 + 24 -

x3
=x+-+-+

2x5

...

J11, which

x =1

11 - J111 = J11-

...

15

.. .

are singular points. Thus, the minimum

1.

= O we see the singular points closest to O are -311'/2 and 11'/2. Thus, the minimum

radius of convergence is 11'/2.


5. The differential equation (x3

+ y' + y

x2)y"

= O has a regular singular point at x = 1 and an

irregular singular point at x = 1.


6. The differential equation (x - l)(x

+ 3)y" + y = O has

regular singular points at x = 1 and x

= -3.

7. Substituting y = I:~o cnxn+r into the differential equation we obtain


00

2xy"

+ y' + y =

(2r2 - r) coxr-l

+ L [2(k + r)(k + r -

l)Ck + (k

+ r)ck + ck_llxk+r-l = O

k=l

which implies
2r2 - r = r(2r - 1) = O
and
(k

+ r)(2k + 2r -

The indicial roots are r = O and r = 1/2. For


Ck =

l)Ck + ck-l

= O the recurrence relation is

- k(2k - 1) , k= 1,2,3, ... ,

so
Cl

= O.

= -Co,

C3

= --Co

90

For r = 1/2 the recurrence relation is


k(2k

+ 1) ,

= 1,2,3, ... ,

so
C3

296

1
630

= --Co

Chapter 6 Review Exercises


Two linearly independent solutions are
1

( l-x+6x

Y1=Cl

-90X1

+ ... )

and
Y2 = C2X

1/2 (

'31x + 301 x 2 -

1-

1 3
630x +...

8. Substituting y = L~=o cnxn into the differential equation we have


00

y" - xy' - y

00

00

L n(n - 1)cnxn-2 - L nenxn - L cnxn


n=2
n=l
n=O
~
"'--"
k=n-2
k=n
k=n
00

00

+ 2)(k + 1)Ck+2Xk -

= L(k
k=O

00

L kCkXk - L Ckxk
k=l
k=O

00

= 2C2 - Co

+ L[(k + 2)(k + 1)Ck+2 k=l

Thus
2C2 - Co = O
(k

+ 2)(k + 1)Ck+2 -

(k

+ l)Ck = O

and
1

C2

ck+2
Choosing ca = 1 and

C1

= '2co

= k + 2 Ck,

k = 1, 2, 3, ....

= Owe find

1
2

C2

=-

C3

= C5

C4

=-

C6

=-

= C7

= ... = O

1
8
1

48

297

(k

+ l)CkJXk = O.

Chapter 6 Review Exercises


and so on. For

ca = O and Cl =

1 we obtain
C2 = C4 = C6 = ... = O

1
3
1
=15

C3 = -

C5

1
= 105

C7

and so on. Thus, two solutions are


Yl

= 1 + 2x + S"X + 48 x + ...

and
Y2=X+3X

1
+15x

1
+105x

+ ....

= L~=a Cnxn into the differential equation we obtain

9. Substituting y

00

+ 3y =

(x - l)y"

(-2C2

+ 3ca) + L (k

- l)(k -

2)Ck-l

k(k - l)Ck

+ 3Ck_2]xk-2 = O

k=3
which implies

C2 = 3ca/2

and

ce=
Choosing

ca =

1 and

Cl

(k - l)(k - 2)Ck-l
k(k-1)

k = 3,4,5, ....

= O we find
3

C2 =

and so on. For

+ 3Ck-2

ca = O and c

2'

C3 =

2'

C3 =

2'

5
8

C4 =-

= 1 we obtain
C2

= O,

C4 = -

and so on. Thus, two solutions are


Yl

301354

= Cl ( 1 + -x~ + -x
2
2

+ -x + ....)
8

and
Y2 = C2 (x
10. Substituting y

+ ~x3 + ~x4 + ...) .

= L~=a cnxn into the differential equation we obtain


00

y" - x2y'

+ xy = 2C2 + (5C3 + ca)x + L [(k + 3)(k + 2)Ck+3


k=l

which implies

C2 = O, C3 = -ca/5,

and
k-1

k = 1,2,3,

298

....

- (k - l)cklxk+l

=O

Chapter 6 Review Exercises


Choosing

Co

= 1 and

and so on. For

co

Cl

= Owe find

= O and

Cl

C3

1
=--

C4

= C7 = ClO = ... = O

Cs

= Cs = Cll

C6

1
=-_

= ...

=O

90

= 1 we obtain
C3

= C6 = Cg = ... = O

C4

C7

Cs

Cs

= Cll = ...

ClO

= ... = O
=O

and so on. Thus, two solutions are


Yl = Co
11. Substituting y

= :;~=oCnxn+r

(1 - ~x3 -

:ox6 - ... )

and

Y2 = ClX.

into the differential equation, we obtain


00

xy" - (x + 2)y' + 2y = (r2

3r)coxr-l

.L [(k + r)(k

+ r - 3)Ck

k=1

- (k

+r -

3)Ck_Ilxk+r-l

= O,

which implies

r2

3r = r(r - 3) = O

and
3)Ck-1 =

(k + r)(k + r - 3)Ck - (k + r The indicial roots are

rl

O.

= 3 and r2 = O. For r2 = O the recurrence relation is

k(k - 3)Ck - (k -

3)Ck-l

= O,

= 1, 2, 3, ....

Then
Cl -

OC3 -

Co

OC2

=O

=O

==>

C3

is arbitrary

and
k = 4,5,6,

299

....

Chapter 6 Review Exercises


Taking Co =j:. O and C3 = Owe obtain

C2 = 2"CO
C3 = C4 = Cs

= ... = O.

1
C4 = C3

6
4!C3

Taking Co = O and C3 =j:. Owe obtain

Cs

1
6
5 . 4 C3 = 5! C3

1
C6 = ~C3

= 6!C3,

and so on. In this case we obtain the two solutions


Yl

1 2

= 1 + x + 2"x

and
Y2 = x 3 + -x6 4 + -x6
4!
5!

+ -x6 6 + ... = 6e x - 6 ( 1 + x + -x1 2) .


6!

12. Substituting y = ~=o cnxn into the differential equation we have


(cosx)y "

+y

(11 - -x 2 + -x1 4 - -x1 6 +


2
24
720

... ) (2C2 + 6C3X + 12c4X 2 + 20C5X 3 + 30C6X4 + ... )

[2C2 + 6C3X + (12c4 - C2)X2 + (20cs - 3C3)X3

+ [co + C1X + C2x2 + C3x3 + C4x4 + ...

= O.
Thus
Co + 2C2 = O

300

+ (30C6 - 6C4 + 112C2) x4 + ... ]

Chapter 6 Review Exercises


20cs - 2C3 = O
1
30C6 - 5C4 + 12c2 = O
and

c? = --Co

C3

= -"6Cl

C4

=O

Cs =

C6 =

Choosing

Co

= 1 and

Cl

Co

1
1
-C4 - -C2
6
360

= Owe find

= O and

-"2'

Cl

= 1 we find

C2

and so on. Por

1
10 C3

C2=0,

C3

= O,

C3=-"6'

C4

= O,

= O,

Cs

C4=0,

Cs

= - 60'

es

=O

and so on. Thus, two solutions are


Yl
13. Substituting y

1 2
1
= 1 - -x
+ -x

+...

= x - -1x 3 - -x1

2
720
6
L~=O Cnxn into the differential equation we have
00

y"

Y2

and

+ xy' + 2y

2:= n(n

00

- 1)Cnxn-2

+ 2:=

n=2
00

2:= (k + 2)(k

k=O

60

+ ....

00

ncnxn

+ 2 2:= Cnxn

n=l

k=n-2

n=O
k=n

'--v---"

k=n
00

00

k=l

k=O

+ 1)Ck+2Xk + 2:= kCkXk + 2 2:=

Ckxk

00

= 2C2 + 2co

+ 2:= [(k + 2)(k + 1)Ck+2 + (k + 2)Cklxk


k=l

Thus
2C2 + 2co = O
(k

+ 2)(k + 1)Ck+2 + (k + 2)Ck =

and
C2

= -Co

ck+2 = - k

+ 1 Ck,
301

k = 1,2,3, ....

= O.

Chapter 6 Review Exercises


Choosing ca = 1 and

Cl

= O we find
C2

= -1

C3

C4

1
=-

C6

=--

Cs

C7

= ... =

C6

= ... =

15

and so on. For ca = O and q = 1 we obtain


C2

C4

C3 =--

Cs = -

8
1

C7=

--

48

and so on. Thus, the general solution is


Y

ca ( 1 - x 21416)
+ 3"x - 15x

+... + cl

(131S
X -"2x

+ B"x

17
- 48x

+ ...)

and
Y,

= ca (

-2x

+ -x4332S -

Setting y(O) = 3 and y'(O)


initial-value problem is

= -2 we find

y = 3 - 2x - 3x
14. Substituting y

+... ) + Cl

-x

Co

3254
1 - -x
2

= 3 and

Cl

16
+ x + x 41S-:x

23

+ -x
8

= -2.

- 5"x

+ ...)

- -x76
48

Therefore, the solution of the

+ 2417 x + ....

= L~=a cnxn into the differential equation we have


00

(x

+ 2)y" + 3y =

L n(n -

CXJ

+2

l)cnxn-l

n=2

L n(n

+3

n=2
k=n-l

L (k + l)kck+lXk

k=l

L cnxn
n=O
k=n

'---v----'

k=n-2

00

00

- 1)cnxn-2

CXJ

CXJ

+ 2 L (k + 2)(k + 1)Ck+2Xk + 3 L

Ckxk
k=O

k=O

CXJ

= 4C2 + 3co + L [(k + 1)kck+l + 2(k + 2)(k + 1)Ck+2 + 3Cklxk


k=l

Thus
4C2

(k

+ l)kq+l

+ 3ea

=O

+ 2(k + 2)(k + 1)Ck+2 + 3q


302

=O

= O.

Chapter 6 Review Exercises


and
C2

3
4

--Co

k
Ck+2 = -

Choosing

Co

1 and

Cl

2(k

+ 2) Ck+l

2(k

k = 1,2,3, ....

+ 2)(k + 1) cu,

= Owe find
3

C2

=-4

C3

=-

1
8
1

C4 =-

16
9
320

Cs=-and so on. For

Co

= O and

Cl

=1

we obtain

C3

=-4

C4 =-

16

es = O
and so on. Thus, the general solution is

co

3213
( 1 -:x + S"X

+ 1614x -

9
320x

s)+...

+ Cl X -:x13 + 1614x + ... )

and
y,
Setting y(O)

21394)
= ce (33
-2"x
+ SX +:x - 64x +...

= O and y'(O) = 1 we find

Co

= O and

+ Cl

(31213
-:x

+:x +... ) .

= 1. Therefore, the solution ofthe initial-value

Cl

problem is
Y

1 3
= x - -x
+ -x1

16

+ ....

15. Writing the differential equation in the form


y

11

(1- cos x ) y , + xy
x

= O,

and noting that

is analytic at x

= O, we conclude

that x

= Ois an

ordinary point of the differential equation.

303

Chapter 6 Review Exercises


16. Writing the differential equation in the form

y" +(eX - x1 - x )y

=O

and noting that


x
2
2
x
x2
...
eX - 1 - x
x
3 18 270
we see that x = O is a singular point of the differential equation. Since

----=---+-+-_

?(
X)
x: eX _ 1 _ x

2x2
x4
3
+ x3
18 + 270 -

= 2x -

... ,

we conclude that x = O is a regular singular point.


17. Substituting y =

I.:~=Ocnxn

into the differential equation we have

00

y"

+ x2y' + 2xy

2: n(n

00

n=2

2: (k + 2)(k

k=O

Cnxn+l
n=O
k=n+ 1

n=l

k=n00

00

+ 2: ncnxn+l + 22:

- 1)cnxn-2

~~
2

k=n+ 1
00

+ 1)Ck+2xk + 2: (k

00

+ 22: Ck_lXk

- l)Ck_lXk

k=2

k=l

00

= 2C2 + (6C3 + 2co)x + 2:[(k + 2)(k + 1)Ck+2 + (k + l)Ck_lJXk = 5 - 2x + 10x3.


k=2
Thus
6C3 + 2co=-2

20C5

(k
and

+ 4C2 = 10

+ 2)(k + 1)Ck+2 + (k + l)Ck-l = O,

= 4,5,6, ... ,

5
C2 = 2
1

C3=--CO--

C4

1
--Cl

C5 = ~ -

~C2 = ~ - ~
1

ck+2 = - k

+ 2 Ck-l
304

G)

=O

Chapter 6 Review Exercises


Using the recurrence relation, we find
1
1
C6 = -6"C3 = 3.6 (co

+ 1) = 32.

1
2! co + 32.2'

1
1
C7 = --C4 = -Cl
7
47
Cs = Cll = C14 = ... = O
1
1
1
C9 = -gC6 = - 33. 3! co - 33.3'
ClO =

1
10C7 =

1
4 . 7 . 10 Cl

C12 = - 12 C9 = 34 . 4! Co + 34 . 4!
Cl3 =

1
1
13 Co = 4. 7 . 10 . 13 Cl

and so on. Thus


1 3
1
[ l-'3x +32.2!x

y=co

+ci [x-'4x

1 4

1
-33.3!x

+ 4.7x

+ [~x2 _ ~x3
2
3
18. (a) From y

1
+34.4!x

1
- 4.7.10x

+ _1_x6

10

__

32 . 2!

12

- ...
1

+ 4.7.10.13x

1_x9

33 . 3!

+ _1_x12
34 . 4!

1 du
u dx

= -- - we obtain

Then dy / dx

= x2 + y2 becomes
-;

1d u

dx2

+ u2

()2dudx

=x

+ u2

(b) If u = xl/2w( ~x2) then


u' = x3/2w' (~x2)

+ ~x-l/2w

and

305

(~x2)

( dudx )2 '

13

]
-oo.

_ ... ].

Chapter 6 Review Exercises


so

u" + x2u
Letting t

= xl/2

[x2w" Gx2)

+ 2w' Gx2)

+ (x2 - ~x-2) W Gx2)]

= O.

~x2 we have

Vii [2tw"(t) + 2w'(t) +

(2t - 4 \t)

W(t)] = O

or

t2w"(t) + tw'(t) + (t2 - 116)w(t) = O.


This is Bessel's equation with

1/

= 1/4, so

(e) We have
y = _]:,du = _

u dx

xl/2w(t)

..!!:_xl/2w(t)

dx

= __ 1_

[xl/2dW dt + ~x-l/2w]
dt dx
2

= __ 1_

[x3/2dW + _1_ w]
dt
2xl/2

xl/2w
xl/2w

=-

1 [2x 2 di
dW]+ w = - 2xw
1 [dW
2xw
4t di + w ] .

Now

4t

dw

di + w = 4t dt [C)l/4(t) + C2J_l/4(t)] + clJl/4(t) + C2J-l/4(t)


= 4t [Cl (J_3/4(t) - :t Jl/4(t))
+ C)l/4(t) + C2J_l/4(t)
= 4CltJ_3/4(t) - 4C2th/4(t)

so

306

+ C2( -

:/-1/4(t)

- h/4(t))]

Chapter 6 Revew Exercses


Letting e

= el / e2 we have

19. Let

1
Y2 = -x[ln(1
2

+ x)

- ln(1 - x)]-

so that
Y2,1[1
= -x
2

1
+ -- 1] + -[ln(1
+ x)

-l+x

l-x

-ln(1 - x)]

and
"
1 [1
Y2 = 2"x - (1 + x)2
1[

+ (1 -

1]
x)2

1[

1
1+ x

+ 2"

1]

1]
1- x

1[

+ 2"

1
1+ x

1]
x

+ 1-

= 2"x - (1 + x)2 + (1 - x)2 + 1 + x + 1 - x'


Then

+ x)y~ - 2xy~ + 2m = o,

(1 - x)(l
20. (a) By the binomial theorem we have
[1

+ (t2 -

2xt

(b) Letting x

)] -1/2

= 1 - 2"1 (t 2 -

2xt

+ '83 (t 2 -

= 1 in (1- 2xt+t2)-1/2,

we have

+ t2)-1/2

=-

(1 - 2t

= (1 - t)-l

2xt

)2

+ ... = 1 + xt + 2"1 (23x

= 1+ t

1-t

+ t2 + t3 + ",

- 1) t 2 + ....

(Itl < 1)

n=O
From part (a) in the text we have
00

:L Pn(l)tn

00

= (1 - 2t

+ t2)-1/2 = :L e

n=O
n=O
Equating the coefficients of corresponding terms in the two series, we see that Pn(l) = 1.
Similarly, letting x = -1 we have
(1 + 2t

+ t2)-1/2

= (1 + t)-l

1
l+t

=-

= 1- t

:L Pn(-I)tn,

00

= :L(-I)ntn
n=O

00

n=O

so that Pn( -1) = (-I)n,

307

+ t2

- 3t3

+ ".

(Itl < 1)

The Laplace Transform


Exercises 7.1

-s -

(2

4e-8tdt = _~e-st
s

= -;e
2.

o'i'{f(t)} =

Jo

1
= ( - -e
s

1 (
1
-; - O- -;e

6.

s2

f(

s>O
s>O

_~(e-2S - 1),
s

-1 ) - -1(O- e -8)

s2

= (-~e-stsint

1 ( 1 - e -8) , s>O
= -s2

_ -2-1-e-stcost)
s +1

s +1

+ -2-1-e-stsint)

o'i'{f(t)} = t'" (cost)e-8tdt = (-~e-stcost


J1</2
8 +1
(O

+ _1_e-1<S/2)
82

+1

= __
82

1_e-1<s/2,

+1

+1

11<
O

s> O

100
1</2

8> O

t)={O,O<t<l
t, i > 1

o'i'{f(t)} =

l te-si dt =
OO

8.

1
- -; ,

(O + _2_1_e-1<s) - (O - -2-1_) = _2_1_(e-1<S + 1),


8+1
s+l
8+1

=O-

7.

12 =

-8

-s - -e
1 -s) - ( 0--;-2 - -1) = -1( 1 - 3e-S) + -2 ( 1 - e -S) ,
S
s2 S
S
s2

o'i'{f(t)} = r(sint)e-stdt
Jo
=

-s) = -;e2

-s - -es21 -s) - ( O-

2 -s - -e
2
= ( --e
5.

__1te-st __ 1)e-st
8

82

100 = _1 e-s + -1 e-s,


I

s2

O,
0< t < 1
f(t) = { 2t _ 2, t> 1

o'i'{f(t)} = 2 (00 (t _ 1)e-8t dt = 2 (-~(t


JI
S

_ l)e-st _ ~ e-sI) 100 = 2 e-s,


s2
I
s2

308

8> O

s>O

Exercses 7. 1
1 - i,

:t'{f(t)}

10. f(t) =

12. :t'{f(t)}

13. :t'{f(t)}

0< t < 1

t> O

9. f(t) = { O
,

(1(1 _ t)e-st di = (-~(1 _ t)e-st + ~ e-st)

Jo

O,

O<t<a

e,

a < t < b;

O,

11=

82

~ e-s

82

8> O

82

-5
= ( 00 e-2t-5e-stdt = e-5 ( 00 e-(s+2)tdt = __e_e-(s+2)t
Jo

+ ~ - ~,

Jo

+2

= (00 te(4-s)tdt = (_1_te(4-S)t

(00 tte-stdt

Jo

Jo

4- 8

100-5= _e_,

e(4-S)t)

(4 - 8)2

8> -2

+2

100
o

1
= (4-8)2'

14. :t'{f(t)}

1000 t2e-2te-st

>4
di =

1000t2e-(s+2)tdt

= ( __ 1_t2e-(S+2)t _
8+2
15. :t'{f(t)}

1000 e-t(sin

= (

-(8+1)
(5+1)2+1

2
te-(s+2)t _
2
e-(S+2)t) 100=
2
(8+2)2
(8+2)3
O (8+2)3'

t)e-stdt =

1000(sin t)e-(s+l)tdt

e-(s+l)tsint-

- (8 + 1)2 + 1 - 82 + 28 + 2 '

1 - ,e(1-s)t
8
( (1-8)2+1
1-8

100

e-(S+l)tcost)

(8+1)2+1

cos t

8> -1

1
(1-8)2+1

8- 1
52 _ 25 + 2'

>1
309

e(1-s)t sin t )

100
o

8> -2

Exercises 7. 1
17. Y{f(t)}

18. Y{f(t)}

= 10':;<) t(cost)e-stdt
=

st
[( --2-1
8 +

82 - 1
(82+1)2'

28). ?
(82+ 1)-

(smt)e -st]

28
(82+1)2'

28) 2 (cost)e-st(82+1)

(8t-2--+

82- 12) (sint)e-st ] 00


(82+1)
O

8 +1

8>0

= 24~

20.

= 5!

Y{t5}

86

7 3
22. Y{7t + 3} = 2" + 8
8

2
2
6
3
23. Y {t + 6t - 3} = - + - - 83 82 8

24. Y{-4t

3!
2
3
1
25. Y{t3 + 3t2 + 3t + 1} = """"4 + 33" + 2" +8
8
8
8

26. Y{8t

t}

1
1
= - + -8 8- 4

29. Y{l +2e2t +e4t}

28. Y{t

121
+-8 8-2
8-4

= - + --

33. Y {sinhkt}

+ 16t + 9} = -4-

-12t

8 - k

CP
-..L-

cosht} = Y {e-t et ~ e-t}

2
+6t-1}

{ 2t -2+e

= 2

8 + 25

8
2
8 - k

= 2

1}_

{1
1
-e 2t - - - -1
2
2
2(8- 2) 28

= y {~+ ~e-2t} = 218+ 2(8~ 2)

310

3!
84

2 6 1
83 82 8

= 8--12-+---

-2t }=----+-1
8-2

3D. Ye

37. Y {sin2t cos2t} = y {~sin4t} = 2 2


2
8 + 16

2
16 9
+ - +83 82 8

-9t
2
1
5
- e
+ 5} = - - +83 8 + 9 8

34. Y {coshkt}

= 2k 2

(/){et sm
. h t} _
(/){ et et - e-t}
35. ..L-..L2

.82
32. Y {cos5t + sin 2t}

31. Y{ 4t2- 5sm3t} = 4-2 - 5-- 3


83
82+ 9

36. Y{e-t

8>0

4
10
21. Y {4t - lO} = - - 82 8

27. Y {1+

00

= 1000 t(sint)e-stdt

= [( --2--t
8 +1

19. Y{2t4}

82-1) 2 (cost)e-st + (t-2-1


(82+ 1)
8 +

2
8

1
8+2

+ -28 +4

Exercises

38. ~ {cos t}

=~

{1"2 + "21cos 2t } =

lIs

28

39. (a) Using integration by parts for

(b) Let u = st so that du

Q:

+ "2

82

+4

> O,

= 8 dt. Then
Q:

= r(1/2)

40. (a) ~{cl/2}

81/2

41. Identifying

For n = 1,

7. 1

> -l.

f!_

V-;

f(t) = tn we have f'(t) = ntn-1, n = 1, 2, 3, .. , . Then, since f(O) = O,

~{t}=-~{1}=2'

For n = 2,

For n

= 3,

42. Let F(t)

= t1/3.

Then F(t) is of exponential order, but f(t)

t = O and hence is not of exponential order.

311

F'(t)

iC2/3 is unbounded near

Exercises 7. 1

Let

f (t)

2tet cos et

d
2
sin et . This function is not' of exponential order, but we can show that
dt

=:_

its Laplace transform exists. Using integration by parts we have


:;t {2tet

cos et 2 }

= 1000 e-st

(00

= s)o

-d sin et 2) dt
dt
2

lim [ e-st sin et 2


a->oo

laO + s loaO e-st

sin et 2 dt ]

= s :;t{sin et }.

e-st sin et dt

Since sin et2 is continuous and of exponential order, :;t {sin et2} exists, and therefore :;t {2tet2 cos et2}
exists.
43. The relation will be valid when s is greater than the maximum of C1 and
44. Since et is an increasing function and t2

C2.

> In M + ct for N! > O we have et2 > e1n M+ct = N!ect for
2

t sufficiently large and for any c. Thus, et is not of exponential order.


45. By part (c) of Theorem 7.1
:;t {e(a+ib)t} =

1
s - (a + ib)

1
(s-a)+ib
(s - a) - ib (s - a)

s - a + ib
(s-a)2+b2

+ ib

By Euler's formula, eie = cos () + i sin (), so


:;t {e(a+ib)t} = :;t {eateibt.}:;t {eat (cos bt
= :;t { eat cos bt}

+ i :;t { eat sin bt}

s- a
= (s - a)2

+ i sin bt)}

+ b2 + i (s

b
- a)2

+ b2

Equating real and imaginary parts we get


at
:;t { ecos
46. We want f(ax

bt} = (
s-a

S -

r-+a b')-

+ (3y) = af(x) + (3f(y)


m(ax

and

:;t{eatsinbt}

= (

s-a

~2

+ b2'

or

+ (3y) + b = a(mx + b) + (3(my + b) = m(ax + (3y) + (a + (3)b

for all real numbers a and (3. Taking a = (3 = 1 we see that b = 2b, so b = O. Thus, f(x) = mx
will be a linear transformation when b

= O.

312

+b

Exercises

Exercises 7.2

3. ~

-1 {1
48}
s2 - s5

4!}
4
= ~ -1 {1s2 - 48
24 . S5 = t - 2t

6. ~-1{(S+32f}=~-1{~+42.2+2.
s
s
s

23}=1+4t+2t2
s

7. ~-1{12_~+_1_}=t_1+e2t
s
s
s- 2

9.

~-l

10.

{_l_}
4s + 1

~-l

~-l

{_l_}
_
5s - 2 -

~-l

13.

~-l

14.

~-l

1 } = ~e-t/4
4

{~.

4 s + 1/4

1 }_- ~e2t/5
5

{~ .

5 s - 2/5

4s } = ~-l
4s2 + 1

1 }=
4s2 + 1

{~ .

~-l

15. ~-1{2s2-6}=~-1{2_2s
s+9

s2 + 1/4

= cos ~t

1/2 } = ~ sin ~t
2 s2 + 1/4
2
2

s+9

7.2

-2.~}=2COS3t-2sin3t
s+9

313

Exercises

7.2

/2}

{s.

/2-

S + 1 }
16. X' -1{ ~2
= X' -1 -2-- + 1! -2-- = cosvr:
2t + - sin vr:
2t
s +
s + 2 v2 s + 2
2

17.

18.
19.
20.

1 } = X'-1
X'-1 { S + 1 } = X'-1
s2 _ 4s

_1_}
{_~ . ~ + ~ . _1_} = _~ + ~e4t
4 s
4 s- 4
4 4
X'-1 {
S
} = X'-1 {~ . _1_ + ~ . _1_} = ~et + ~e-3t
s2 + 2s - 3
4 s - 1 4 s +3
4
4
X'-1 {

X'-1 {

s2 + 3s

s-O.l

O)~s
)} = X'-1 {(0.3). _1_
s+0.2.
s-O.l

X'-1 {(s _ vi~s3 +


-1 {

23. X'

26. X'

27.

} _ X'-1 {~ . _1 __ ~ . _1_} _ ~e4t _ ~e-5t


9 s - 4 9 s +5 - 9
9

s2 + S - 20 -

21. X'-1 {(

22.

-1{

X'-1 {

{~ . ~ _ ~ .
= ~ _ ~e-3t
3 s
3 s+3
3 3

)3)}

= X'-1 {s2 ~ 3

(s2+4)(s+2)
28 - 4

-1 {

(s-2)(s-3)(s-6)

=X'

=X'

-1 {

= -4 +
( -1{ -- 1}
28. X'
s4 - 9

=X'

-1 {1
-.----.--

6V3

s+0.2

-)3 . s2~
1

2"s-2-s-3+2"s-6

1
s
1
2
's2+4+'s2+4-'s+2

} _ X'-1 {

(s2 + s)(s2 + 1) -

+ (0.6). _1_}

0.3eo.lt + 0.6e-O.2t

3} = coshv'3t - v'3 sinh v'3t


1

1}

1}

=2'e

2t

3t

-e

+2'e

1
1 .
=cos2t+sm2t-4e

2s - 4 } _ X'-1 {4
3
s
3}
s(s2 + 1)2 -~ + s + 1 + s2 + 1 + s2 + 1

3e-t + cost + 3 sin t

)3

s2 - 3

6V3

V3}

s2 - 3

314

1.
r,;
=-smhv3t--smv3t

6)3

1.

6/3

r:

6t

1 -2t

Exercises

30.

68 + 3}
.:t'-l { (82 + 1)(82 + 4)

= .:t'-l

2}

{8
1
8
1
2 82 + 1 + 82 + 1 - 2 82 + 4 - '2 . 82 + 4
1

= 2 cos t + sin t - 2 cos 2t - '2 sin 2t


31. The Laplace transform of the differential equation is
1

8.:t{y}

- y(O) - .:t{y} = -.
8

Solving for .:t' {y} we obtain


1

.:t'{y} = -- + -.
8

1
1

8-

Thus

32. The Laplace transform of the differential equation is

28.:t'{y} - 2y(0)

= .:t'{y} = O.

Solving for .:t' {y} we obtain


6

.:t'{y} = 28+ 1 = 8+ 1/2


Thus

33. The Laplace transform of the differential equation is

8.:t{y}

- y(O) + 6.:t{y}

1
8_

4.

Solving for .:t' {y} we obtain


1

.:t'{y} = (8-4)(8+6)

2
+ 8+6

= 10' 8-4

19

19
1
+ 10' 8+6'

Thus
Y

4t

= 10e + 10e

-6t

34. The Laplace transform of the differential equation is


8.:t' {y} - .:t' {y}

28

= 82 + 25 .

Solving for .:t'{y} we obtain


2
Y{y} = (8-1)(82+25)

1118
= 13' 8-1

- 1382+25

Thus
1
y = -e
13

1
- -cos5t+
13

315

5
-sin5t.
13

55
+ 13' 82 +25'

7.2

Exercises 7.2
35. The Laplace transform of the differential equation is
S2 ~ {y} - sy( O) - y' (O) + 5 [s ~ {y} - y (O)] + 4 ~ {y} = O.
Solving for ~ {y} we obtain
s+5

~{y}

= s2

+ 5s + 4 = "3 s + 1 -"3 s + 4

Thus
y = ~e-t _ ~e-4t.
3
3
36. The Laplace transform of the differential equation is
s2 ~{y}

_ sy(O) - y'(O) - 4[s ~{y}

_ y(O)] = _6
s-3

3_.
s+l

Solving for ~ {y} we obtain


<o

oL-{y}-

6
- (s - 3)(s2 - 4s) - (s
5 1
2
= _. - - -2 s

s - 3

3
+ 1)(s2 - 4s)

s-5
s2 - 4s

+-=--

3
1
11
1
+ -.
--o
5 s + 1 10 s - 4

- _. --

Thus
y

5
2

3
5

= __ 2e3t __ e-t

11
+ _e4t

10

37. The Laplace transform of the differential equation is


s2~{y}
Solving for ~{y}

- sy(O) +~{y}

~2'

we obtain
~

{y}

+~=~+_2

(s2+1)(s2+2)

s2+1

2_
s2+1

s2+1

s2+2'

Thus
y = lOcos t + 2 sin t -

h sin ht.

38. The Laplace transform of the differential equation is


s2 ~{y}

+ 9~{y}

= -.
s -1

Solving for ~ {y} we obtain


~{y}

1
= (8-1)(s2+9)

11111
= 10' s-l

- 10' s2+9

Thus
y

1 tI.
= -e
- -sm3t
10

30

316

1
- -cos3t.
10

-10'

s
s2 +9'

7.2

Exercises
39. The Laplace transform of the differential equation is
2 [s3.:t'{y} - s2(0) - sy'(O) - y"(O)] +3[s2 .:t'{y}-sy(0)-y'(0)]-3[s

.:t'{y}-y(0)]-2

.:t'{y} = _1_.
s+l

Solving for .:t'{y} we obtain

.:t'
{y}

2s + 3
(s+1)(s-1)(2s+1)(s+2)

= ~ "_1_ + ~
2s+1

_1 __

18s-1

+ ~ _1_

9s+1/2

9s+2"

Thus
1
5 __ 8e-t/2 + _e-2t
1
Y = _e-t
+ _et
2

18

40. The Laplace transform of the differential equation is

s3 .:t'{y} - s2(0) - sy' (O)- y" (O)+ 2[s2 .:t'{y} - sy(O) - y' (0)]- [s .:t'{y} - y(O)]- 2 .:t'{y} = +-9
s

"

Solving for .:t'{y} we obtain

.:t'{}s2+12
y - (s - 1)(s + 1)(s + 2)(s2 + 9)
13
= 60

13

16

s - 1 - 20 s + 1 + 39 s + 2 + 130 s2 + 9 - 65 s2 + 9 .

Thus

13 t

Y = -e

60

41. Let f(t)

= 1 and g(t) = {

13 -t
- -e
20

16 -2t
3
+ -e
+ -cos3t
39

130

1 "
- -sm3t.
65

t > O, t =f. 1
. Then .:t'{f(t)} =.:t'{g(t)}
O, t = 1
1,

= 1, but f(t) =f. g(t).

42. The Laplace transform of the differential equation is

.:t'{}
.:t' { }
s +3
s +3
y +
y = (s + 3)2 + 4 = s2 + 6s + 13 .

Solving for .:t'{y} we obtain

.:t'{ }_
Y

s+3
(s+1)(s2+6s+13)

4 .s +1 - 4

=4'

1 (S+3

s+1-4'

s+l
s2+6s+13

2)

(s + 3)2 + 4 - (s + 3)2 + 4 .

Thus
1

y = _e-t - e-3t cos 2t + _e-3t sin 2t.


4
4
43. For y" - 4y' = 6e3t - 3e-t the transfer function is W(s) = 1/(s2 - 4s). The zero-input response is

yo(t)=.:t'-l{

s-5
}=.:t'-l{~.~_~._l_}=~_~t,
s2 - 4s
4 s
4 s - 4

317

Exercises 7.2
and the zero-state response is

Yl (t) =:L

-1

{6

=:L-l{27.

3}

(3 _ 3)(32 _ 43) - (3 + 1)(32 - 43)

_1
20 3 - 4

2_ + ~. ~ _ ~. _1_}
3- 3 4 3 5 3+ 1

Exercises 7.3
1.

:L {telOt} =

2. :L {-6t}
te
3.

1
(3 - 10)2

1
= (3 + 6)2

:L{t3e-2t} =

4. :L{t10e-7t}
5

3!
(3 + 2)4

lO!
(3 + 7)11

:L{t(et+e2t)2}=:L{te2t+2te3t+tt}=

1
(3-2)2
2

6.

2
(3-3)2
2

-2t

(3_13)2+9

3+2
= (3+2)2+16

8.

:L { e

9.

:L{(1 - et + 3e-4t) cos 5t} =:L {cos 5t - et cos5t + 3e-4t cos 5t}

cos4t

3
- 32+25

11.

1
(3-4)2

:L{ e2t(t - 1)2} =:L {t2e2t - 2te2t + e2t} = (3 _ 2)3 - (3 _ 2)2 + 3 _ 2

7. :L{etsin3t}=

10.

t) } =:L {3tge

( { 3t(
.
:L e 9 - 4t + 10 sm"2
:L-l{

3-1
(3-1)2+25

- 4te

3t

3(3+4)
+ -,-----'---:-,,-(3+4)2+25

t}

3t .
9
4
5
+ lOe sm"2 = 3 _ 3 - (3 _ 3)2 + (3 _ 3)2 + 1/4

1
} _ :L-l{~
2
} _ ~t2e-2t
(3+2)3
2 (3+2)3
- 2

318

7.3

Exercises

{(S ~ 1)4}

= ~-l

(S ~!1)4}

= ~t3et

12.

~-l

13.

~-l {2s - 6s1 + 10} = ~-l {(s - 3)\ + 12}= e3t sin t

14.

~-l{ s2+ 2s1 + 5 }=~-l{~2 (s + 1)22+ 22}=~e-tsin2t


2

15.

~-l{ s2 + 4ss + 5 }=~-l{

16.

~-l{s2 +2s+5
}=~-1{2 (s +(s+3)
_~
5
}=2e-3tcos5t-~e-3tsin5t
6s + 34
3)2 + 52 5 (s + 3)2 + 52
5

17.

18.

19.

20.

{~

(s +22) 2-2
12 2}=e-2tcost-2e-2tsint
(s + 2) + 1
(s + 2) + 1

~-l {(s: 1)2} =~-l {s(:: 0;}=~-l {s ~ 1 - (s: 1)2} = e-t - u:'
~-l {(s-2)2
5s } = ~-1 {5(S - 2) + lO} =~-1 {_5_ +
10 } = 5e2t + 10te2t
(s-2)2
s-2
(s-2)2
2s - 1}
{5; - s21 - S +5 1 - (s +4 1)2 - 2"3 (s +2}1)3 = 5-t-5e -4te
~
s2(s + 1)3 = ~
-1 {

~-1

-t

-1

{(S + 1)2} =
(s+2)4

~-1

}=

1
_
2
+~
3!
(s+2)2
(s+2)3
6(s+2)4

u:? _ t2e-2t

+ ~t3e-2t
6

21. The Laplace transform of the differential equation is

s~{y}
Solving for ~{y}

we obtain ~{y}

= (s

- y(O) + 4~{y}
1

+ 4)2 +
y

= -4'
s+

+ 4' Thus

= te-4t + 2e-4t.

22. The Laplace transform of the differential equation is

s ~{y} - ~{y}
Solving for ~{y}

1
=;

+ (s _ 1)2 .

we obtain

~{y}

s(s - 1) + (8 - 1)3 =

Thus

319

-;

+ -s---1 + -(8---1-=-)3
.

-t

2 -t

-2"t e

Exercises 7.3
23. The Laplace transform of the differential equation is

S2 :t'{y} - sy(O) - y'(O) + 2[s:t'{y}


Solving for :t'{y}

- y(O)] + :t'{y} = O.

we obtain
s

+3

:t'{y} = (s + 1)2 = s + 1 + (s + 1)2 .


Thus

24. The Laplace transform of the differential equation is

s2 :t'{y} - sy(O) - y'(O) - 4 [s:t'{y} - y(O)] + 4:t'{y}


Solving for :t'{y}

we obtain

(s ~ 2)4'

:t'{y} = 2~ (s ~!2)6 . Thus, y = 2~t5e2t.

25. The Laplace transform of the differential equation is


1
s

s2 :t'{y} - sy(O) - y'(O) - 6 [s :t'{y} - y(O)] + 9 :t'{y} = 2"'


Solving for :t' {y} we obtain

:t' y} = 1 + s2
{
s2(s - 3)2

= 2 ~+ ~~ _ 2 _1_ + 10
27 s

9 s2

27 s - 3

1
9 (s - 3)2 .

Thus
2

y=27+9t-27e

2 3t

10 3t

+gte

26. The Laplace transform of the differential equation is

s2 :t'{y} - sy(O) - y'(O) - 4 [s:t'{y} - y(O)] + 4:t'{y}

= 4"'
s

Solvng for :t' {y} we obtain

:t' y = s5 - 4s4 + 6 = ~ ~ + ~~ + ~!
+ ~ ~ + ~ _1 __
{}
s4(S - 2)2
4 s
8 s2
4 s3
4 s4
4 s - 2

13
8

Thus

Y=

3 2

1 2t

+ st + t + t + e -

13 2t

ste

27. The Laplace transform of the differential equation is

S2 :t'{y} - sy(O) - y'(O) - 6 [s:t'{y} - y(O)] + 13 :t'{y} = O.


Solving for :t' {y} we obtain

:t'{y} = - s2 _ 63s+ 13 = -~ (s _ 3~2+ 22 .

320

(s - 2)2 .

Exercises

7.3

Thus
y

3
.
= - 2e3t
SIl1 2t.

28. The Laplace transform of the differential equation is

2[82.:t'{y}

- 8Y(0)] + 20[8 .:t'{y} - y(O)] + 51.:t'{y}

= O.

Solving for .:t'{y} we obtain


.:t"

{y}

48+40
282+208+51

28+20
(8+5)2+1/2

2(8+5)
(8+5)2+1/2

10
(8+5)2+1/2'

Thus

29. The Laplace transform of the differential equation is


2

.:t' {y} - 8y (O) - y' (O) -:- [8.:t' {y} - y (O)] =

8 -

(8 _

1) 2 + 1 .

Solving for .:t' {y} we obtain


.:t'

{y}

=~~_~

1
8(82-28+2)

8-1
+~
282(8-1)2+12(8-1)2+1'

Thus
1 1t
1t.
y = - - -e cos t + -e sm t.
2 2
2
30. The Laplace transform of the differential equation is

82 .:t'{y} - 8Y(0) - y'(O) - 2 [8.:t'{y} - y(O)] + 5.:t'{y}


88

= ~ + 12,

Solving for .:t'{y} we obtain

.:t'{y}

2. ~

482 + 8 + 1 =
+ ~ _.!:_ + -78/25 + 109/25
82(82 - 28 + 5)
25 8 5 82
82 - 28 + 5
7 1 1 1
7
8- 1
25858225(8-1)2+2225(8-1)2+22'

= - - +- - - -

51

+ - ----,..."....---.",-

Thus
7
Y=25

+ -t1
5

7 t
- -e cos 2t
25

51 t .
+ -e
sm 2t.
25

31. Taking the Laplace transform of both sides of the differential equation and letting e = y(O) we

321

Exercses 7.3
obtain

se {y"} + ~ {2y'} + ~ {y}


- sy(O) - y'(O) + 2s ~{y} - 2y(0) + ~{y}

s2~{y}

=O
=O

s2~ {y} - es - 2 + 2s ~{y} - 2e + ~ {y} = O


(s2 + 2s + 1) ~{y}

~{

es + 2e + 2

}_
y

es

(s

=e

+2
+ 1)2

2e

+ 1)2 +

(s

s +1- 1
2e + 2
+ ..,.---=
(s+1)2
(s+l)2
e+2

+ ..,.------,'"
(s+1)2'

= -s+l
Therefore,

To find e we let y(l)

2. Then 2

ee-1

+ (e + 2)e-1 = 2(e + 1)e-1

and e

e - l. Thus

y(t) = (e - l)e-t + (e + l)te-t.


32. Taking the Laplace transform of both sides of the differential equation and letting e

y' (O) we

obtain

~ {y"} + ~ {8y'} + ~ {20y} = O


s2~{y}

- y'(O) + 8s~{y}

s2~{y}

- e+8s~{y}

+ 20~{y}

=O

+20~{y}

=O

(i+8s+20)~{y}=e
~{

} _

y - s2 + 8s + 20 - (s + 4)2 + 4 .

Therefore,
y (t )

<..0-1 {

oL-

s+4

e)2

+4

To find e we let y'(;r) = O. Then O = y'(;r) = ee-4

1l'

= ce -4t

and e

sin 2 t.

= O. Thus, y(t) = O. (Since the

differential equation is homogeneous and both boundary conditions are O, we can see immediately
that y(t)

= O is a solution. We have shown that it is the only solution.)


322

Exercises

= -kx - f3x'. Now m = W/g = 4/32 = ~ slug, and 4 = 2k so

33. Recall from Section 5.1 that mx"

that k = 2 lb/ft. Thus, the differential equation is x"

x(O) = -3/2 and x'(O)

= O.

+ 7x' + 16x

= O. The initial conditions are

The Laplace transform of the differential equation is

82:t'{X} +

~8

+ 78:t'{X} + 21 + 16:t'{x} = O.
2

Solving for :t' {x} we obtain

:t' x _ -38/2 - 21/2 _


{ }-

82 + 78 + 16 -

8 + 7/2

7V15

-2 (8 + 7/2)2 + (V15/2)2

10

Vl5/2
(8 + 7/2)2 + (V15/2)2 .

Thus
3 -7t/2 cos-t
Vl5 - --e
7Vl5
x = --e
2
2
10

-7t/2

. Vl5
sm-t.
2

34. The differential equation is

d2q
dq
dt2 + 20 dt + 200q = 150,

q(O) = q'(O) = O.

The Laplace transform of this equation is


150

82:t'{q} + 208:t'{q} + 200:t'{q} = -.

Solving for :t' {q} we obtain

:t'{ } _

150
q - 8(82 + 208 + 200)

3 1

4' ~ - 4'

8 + 10
3
10
(8 + 10)2 + 102 - 4' (8 + 10) 2 + 102 .

Thus
3 -lOt
3 -lOt .
q(t) = -3 - -e
cos lOt - -e
sm 10t
444

and

i(t) = q'(t) = 15e-10t sin 10t.


35. The differential equation is

d2q
dq
2
Eo
dt2 + 2'\ dt + w q = L'

q(O) = q'(O) = O.

The Laplace transform of this equation is


2
2
Eo 1
8 :t'{q} + 2'\8:t'{q} + w :t'{q} = L ~

or
(82 + 2'\8

+ w2):t'{q}

= ~o ~.

Solving for :t' {q} and using partial fractions we obtain

:t'{q} = Eo (1/W2
L

7.3

(1/w2)8 + 2'\/W2) = Eo (~_


82 + 2'\8

+ w2

323

Lw2

+ 2,\

82 + 2'\8

+ w2

) .

Exercises

For A

7.3

> w we write s2 + 2AS + w2

(s + A)2 - (A2 - w2), so (recalling that w2 = 1/ LC)

1
S+A
A)
,,;t'{q} = EaC ( ~ - (s + A)2 _ (A2 _ w2) - (s + A)2 _ (A2 _ w2)
Thus for A

> w,
q(t) = EaC(l-

For A

= w,

VA2

w2

sinh

VA2 -

w2 t) .

so

A)

s+
(S+A)2+(W2-A2)

= (s + A)2

s2 + 2A + w2

{q}
Thus for A

w2 t -

- (S+A)2+(W2-A2)

< w,

,,;t'

VA2 -

+ A)2 + (w2 - A2),

= (s

1
( ~-

,,;t'{q}=Ea

For A

e-At cosh

< w we write s2 + 2AS +w2

Thus for A

and

= Ea (l/A

Ea
1
L S(S+A)2

l/A
S+A

l/A ) = Ea (~ __ 1_ _
A )
(S+A)2
LA2 s S+A
(S+A)2

= w,

36. The differential equation is

dq

R dt + C q = Ece

-kt

= O.

,q(O)

The Laplace transform of this equation is


R ,,;t'{ q}

+ C ,,;t'{ q}

Ea s + k .

Solving for ,,;t' {q} we obtain

EaC
Ea/ R
,,;t'{q} = (s + k)(RCs + 1) = (s + k)(s + 1/ RC) .
When 1/ RC =1- k we have by partial fractions
::t { }

= Ea (1/ (1/RC - k) _ 1/(1/RC - k))


R

s + 1/RC

s +k

Ea
R

1
1/RC -

Thus

q(
When 1/ RC

=k

t)

EaC
(-kt
-ti Re)
kRC e - e
.

1_

we have

,,;t'{q} =

Ea

(s + k)2 .

324

(_1_ _
k

s +k

s + 1/RC

).

Exercises

7.3

Thus

q(t) = Ea te-kt = Ea te-tiRe.

37 . .:t{ (t - 1)OU(t _ 1)} = e~s

38. .:t{ e2-t OU(t - 2)}

39 .

= .:t {e-(t-2)

OU(t _ 2)}

-2s

= _e _
s+1

e-2s
2e-2s
.:t{tOU (t - 2)} = .:t {(t - 2)OU(t - 2) +20U (t - 2)} = -2 +-s

e-s

40. .:t { (3t + 1)OU(t - 1)} = 3 .:t { (t - 1)OU(t - 1)} + 4.:t { OU(t - 1)} = -2 +-

4e-s

41. .:t { cos 2t OU(t - 71')}

45 . .:t-1 L~-:s1}

46 .

= cos2 (t -

e-s
s(8+1)

48. .:t-1 { 2r2S

)}

s8-1

49. (c)

55.
56.
57.

se-7rS
.:t { cos 2 (t - 71') OU(t - 71')} = 82 + 4

= sin(t - 71') OU(t - 71')

.:t-1 {8:;:S~2}

47 .. .:t-1 {

= .:t-1
= .:t-1

%) OU(t _ %)

{e-S _ e-S}
8
8+1

= OU(t

_ 1) _

{- e-2s _ e-22S + e-2s }


8
8
s-1

50. (e)

51. (f)

e-(t-l)

OU(t - 1)

= _ OU(t _

2) - (t - 2) OU(t - 2) + -2ou (t - 2)

53. (a)

52. (b)

54. (d)

.:t {2 - 4 OU(t - 3)} = ~ _ ~e-3s


s

1 e-4s
e-5s
.:t {1 - OU(t - 4) + OU(t - 5)} = - - +8
S
s
.:t {t2 OU(t - 1) } = .:t { ((t - 1)2+ 2t - 1]OU(t - 1)} = .:t { ((t

325

1)2+ 2(t - 1) - 1]OU(t - 1)}

Exercises

7.3
2
2
= ( -+-+s3 s2

1)_s
e
s

1 e-2s
s2 -~--s-

59. 2{t-tOU(t-2)}=2{t-(t-2)OU(t-2)-2OU(t-2)}=
60. 2{sint-sint

61. 2{f(t)}=2{OU(t-a)-OU(t-b)}=--62. 2{f(t)}

OU(t-27f)} = 2{sint-sin(t-27f)OU(t-27f)}
e-as

e-bs

= 2{OU(t -1) + OU(t -2) + OU(t-3)

= s2+1

e-s

e-2s

2e-2s
e-27rs
- s2+1

e-3s

e-s

+ ...} = -;- + -s- + -s- + ... = ~ 1- e-S

63. The Laplace transforrn of the differential equation is


s2{y}

-y(O) +2{y}

5
s

= _e-s.

Solving for 2 {y} we obtain

2{ } = 5e-S
y

s(s+1)

= 5e-s [~ __

1_]

s+1

Thus
y = 5OU(t- 1) - 5e-(t-l)OU(t -1).
64. The Laplace transforrn of the differential equation is
s 2{y} - y(O) +2 {y}
Solving for 2{y}

= ~ - ~e-s.
s

we obtain
1
2{y}

= s(s

+ 1) -

2e-s
1
1
-8 [1
1]
s(s + 1) = ~ - s + 1 - 2e
~- s +1

Thus

65. The Laplace transform of the differential equation is


s 2{y} - y(O) + 2 2{y} =

1
s-

? -

s +1
e-s-2-.
s

Solving for 2 {y} we obtain


2{y}

1
s2(s+2)

-8

-e

s+1
s2(s+1)

11 1 1
1
=-4~+2s2+4s+2-e

326

-s

[1 1 1 1
1
4~+2s2-4s+2'

1 ]

Exercses

7.3

Thus
y = -~ + ~t + ~e-2t - [~+ ~(t - 1) - ~e-2(t-l)]q(t
4 2
4
4 2
4

- 1).

66. The Laplace transform of the differential equation is

52 .:t'{y} - 5y(O) - y'(O)

+ 4.:t'{y}

-s

e -.
5

Solving for .:t'{y} we obtain


1- 5

-s

.:t'{y} = 5(52+4) -e

1
5(52+4)

11
=4~-4

5
52+4

-2'

2
-s [1 1
1 5
52+4 -e
4~-452+4'

Thus
y = ~ - ~ cos 2t - ~ sin 2t - [~ - ~ cos 2(t - 1)] q(t - 1).
4 4
2
4 4
67. The Laplace transform of the differential equation is

52.:t'{y}

- 5Y(O) - y'(O) + 4.:t'{y}

= e-27rS~1
5

Thus
y = cos 2t + [~sin (t - 27r)- ~ sin 2(t - 27r)]q (t - 27r).
68. The Laplace transform of the differental equation is

52.:t'{y}

e-s

- 5Y(O) - y'(O) - 5 [5.:t'{y} - y(O)] + 6.:t'{y}

= -.
5

Solving for .:t'{y} we obtain


1

.:t'{y} = e-s 5(5 _ 2)(5 - 3) + (5 - 2)(5 - 3)

= e-s

[~ ~ - ~ 5 ~ 2 + ~ 5 ~

3] -

5~ 2 + 5~

_ 1)

+ e3t

3.

Thus
y = [~_ ~e2(t-l)
6 2

+ ~e3(t-l)]q(t
3

_ e2t.

69. The Laplace transform of the differential equation is


-7r5

52 .:t'{y} _ 5y(O) _ y'(O) +.:t'{y}

= _e

-27r5

e _.
5

Solving for .:t'{y} we obtain

.:t'{y} = = [~__
5

5_] _ e-27rs [~5 __ 52 5_]


+
+ 1
52 + 1

_1_.

52 + 1

327

Exercses 7.3
Thus
y

[1 - cos(t - 7r)]OU(t - -r) - [1 - cos(t - 27r)]OU(t - 27r) + sin t.

70. The Laplace transform of the differential equation is

s2 ~{y}

_ sy(O) _ y'(O)

+ 4[s~{y}

_ y(O)]

+ 3~{y}

-28

-48

e _
s

-68

+ _e

++".

Solving for ~ {y} we obtain

~{y} = ~ ~ _ ~ _1_ + ~ _1__ e-28 [~~ _ ~ _1_ + ~ _1_]


3s

2s+1

6s+3

3s

2s+1

+ ~ _1_] + e-68

_ e-48 [~~ _ ~ _1_


3s
2s+1

6s+3

+ ~ _1_]

[~~ _ ~ _1_
3s
2s+1

6s+3

6s+3

Thus
1 1 -t
y = 3" - 2e
- [~_
3

+ "61e -3t

~e-(t-4)

+ ~e-3(t-4)]OU(t

[1

1 -(t-2)

3" - 2e

_ 4)

71. Recall from Section 5.1 that mx" = -kx


so that k

+ "61e -3(t-2)]

(
OUt - 2)

+ [~3

+ f(t).

+ ~e-3(t-6)]OU(t

~e-(t-6)
2

Now m = W/g = 32/32 = 1 slug, and 32 = 2k

= 16 lb/ft. Thus, the differential equation is x" + 16x = f(t).

x(O) = O, x'(O)

= O.

20t,

f(t)

The initial conditions are

Also, since
f(t)

and 20t = 20(t - 5)

- 6).

+ 100 we can

={
O,

O:::; t

<5

t >: 5

write

= 20t - 20tOU(t - 5) = 20t - 20(t - 5)OU(t - 5) -100OU(t

- 5).

The Laplace transform of the differential equation is


s2~{x} + 16~{x}
= 20 20 e-58 __ 100e-58.
s2
s2
S
Solving for ~ {x} we obtain
~{x}

20
_
20
e-58 _
100
e-58
- s2(s2 + 16) s2(s2 + 16)
s(s2 + 16)

=(~.2__~.
4 s2
16

s2

+ 16

)(1_e-58)_(25.~_25.

.5

s2

+ 16

)e-58.

Thus
x(t) = ~t - 156sin4t - [~(t - 5) - 156sin4(t - 5)] OU(t - 5) - [~5 - 245cos4(t
5

25

- 5)] OU(t - 5)

= t - 16 sin 4t - tOU (t - 5) + 16 sin4(t - 5) OU(t - 5) + "4 cos 4(t - 5) OU(t - 5).


328

Exercises 7.3
72. Recall from Section 5.1 that mx"

-kx + f(t).

= W/g = 32/32 = 1 slug, and 32 = 2k


+ 16x = f(t). The initial conditions are

Now m

so that k = 16 lb/ft. Thus, the differential equation is x"


x(O) = O, x' (O) = o. Also, since

< 27T

O:::;t

sin i,

f(t)

={
O,

t 2: 27T

and sin t = sin( t - 27T) we can write

f(t) = sin t - sin(t - 27T)OU(t - 27T).


The Laplace transform of the differential equation is

82X{X}

+ 16x{x}

= -2--1

1
- -2--1e-27rS.
8

Solving for X {x} we obtain

VJ{ X }

oL.

= -1/15

82 + 16

e -27rS

- (82+16)(82+1)

- (82+16)(82+1)

+ 1/15 _ [-1/15 + 1/15] e-27rs.


82 + 1
82 + 16 82 + 1

Thus

x(t) = -~ sin4t + ~ sin t + ~ sin4(t - 27T)OU(t - 27T) - ~ sin(t - 27T)OU(t - 27T)


60

15

60

= { -lo sin 4t + A sin t,

15

O:::;t

< 27T

t 2: 27T.

O,

73. The differential equation is

2.5 dq
dt

+ 12.5q

= 5OU(t

- 3).

The Laplace transform of this equation is

8X{q}

+ 5x{q}

=-

e-3s.

Solving for X {q} we obtain

x{q}

e-3s = (~ . ~ _ ~ .

8(8+5)

5 8

_1_)

5 8+5

Thus

q(t) = ~OU(t - 3) 5

~e-5(t-3)

OU(t - 3).

74. The differcntial equation is


10 ~~ + 10q = 30et

329

30etOU(t - 1.5).

e-3s.

Exercises 7.3
The Laplace transform of this equation is
3
3e1.5
8.;t'{q} - qO +.;t' {q} = 8 _ 1 - 8 _ 1.5 e-1.5s.
Solving for .;t'{q} we obtain
.;t'{q} = (qO _ ~) . _1_ + ~. _1_3e1.5
2
8+ 1 2 8- 1

(-2/5 +~)
e-1.55.
8 + 1 8 - 1.5

Thus

75. (a) The differential equation is

(t _ 37r)OU(t_
37r)
22'

= sin z } cos
~
The Laplace transform of this equation is
di + 10i

1
8.;t'{i} + 10.;t'{i}

i(O) = O.

8e-31rs/2

= 82 + 1 + 82 + 1 .

Solving for .;t'{i} we obtain


.;t'{i} =

1
(82 + 1)(8 + 10)

__ 1_ (_1
- 101 8 + 10

e-31rs/2

(82 + 1)(8 + 10)

8_ + ~)
+ _1_ (~
+ ~
+ _1_)
e-31rs/2
82 + 1 82 + 1
101 8 + 10 82 + 1 82 + 1
.

Thus
1

i(t) = 101 (e-10t-cost+10sint)


+ 1~1 [_10e-10(t-31r/2)
(b)

+ lOcos (t - 3;) + sin (t _ 3;) ]OU(t _ 3;).

0'2f
--+--~""'---"--~..__.=::::--,._~._.__;:",.,_~::::----"'~~~--..__
1

3~

t
6

-0.2
The mximum

value of i(t) is approximately 0.1 at t = 1.7, the minimum s approximately

-0.1 at 4.7.
76.

(a) The differential equation is


50 ~~ + 0.~1 q = Eo[OU(t -1) -OU(t - 3)],
or

330

q(O) = O

Exercises

50 ~;

+ 100q = Eo[VU(t -

q(O) = O.

1) -VU(t - 3)],

The Laplace transform of this equation is

+ 100:t{q} = Eo

50s:t{q}

Ge-s

- ~e-3S).

Solving for :t {q} we obtain

:t

{q}-

_ Eo [ e-s
50 s(s+2)

e-3s

s(s+2)

] _ Eo [~ (~ __ 1 ) e-s _ ~ (~ __ 1 ) e-3S]
- 50 2 s s+2
2 s s+2
.

Thus

(b)

The maximum value of q(t) is approximately 1 at t = 3.


77. The differential equation is
d4y
El dx4 = wo[l - VU(x - Lj2)].
Taking the Laplace transform of both sides and using y(O)

= y'(O) = O we obtain

s4:t {y} _ sy" (O) - ylll(O) = Wo ~


El s

(1 _ e-Ls/2)

Letting y"(O) = Cl and y"'(O) = C2 we have

:t{y} = 2 +
s3

C2

s4

+ Wo ~
El

s5

(1 _ e-Ls/2)

so that
1
= -ClX

y(X)

To find Cl and

C2

2 + -C2X
1 3 + -1 -Wo [4x
6
24 El

(x

L)2 4 VU( x - -L)2 ] .

- -

we compute
y

"(x) = Cl

Wo [2x + C2X + "21 El

( x -"2L)2

1/(
u x -

and
y'" (x) = C2 +

~~ [x - (x -

~)vu ( x -

331

~) ] .

L)]
"2

7.3

Exercises 7.3
Then y" (L)

=;

v" (L)

= Oyields the system.


Cl

+ C2L + -1 -wo [2L - (L)2]


2 El

wo (L)

Solving for

and

Cl

we obtain

C2

Cl

"2

+ El

C2

wOL2 = O
= Cl + C2L + -3S -El

= C2

+ '2 El

1 woL

~wOL2/ El and

C2

=-

= O.

iwoL / E l. Thus

wo (1 2 2
1 3
1 4
1 (
y(x) = El 16L x - 12Lx + 24 x - 24 x

-"2L)4

(
L))
/Jx - "2

78. The differential equation is

d4y

El dx4 = wo[/J(x- L/3) - /J(x- 2L/3)].


Taking the Laplace transform of both sides and using y(O)

Letting y"(O)

= Cl

= y' (O) = O we obtain

s4:t'{y} _ sy"(O) _ ylll(O) = wo ~ (e-Ls/3


El s
and ylll(O) = C2 we have

e-2Ls/3).

:t'{y} = ~ + C2 + wo ~ (e-Ls/3 _ e-2Ls/3)


s3 s4 El s5
so that
1
y(x) = '2ClX
To find

Cl

and

C2

1 3 + 24
1 El
wo [( x - '3
L)4 /Jx
( - '3
L) - ( x - '3
2L)4 /Jx
( - '3
2L)]
+ 6C2X

we compute

and

Then y" (L)

= u" (L) = O yields the system


Cl

+ c2L + -1 -Wo [(2L)2


2 El

- (L)2]
-

Cl

Wo [2L

C2

Solving for q and

C2

we obtain

Cl

Wo (112L 2 x 2 - lS1Lx
y(x) = El

=
3

+ El

twoL2/EI

wOL2
+ C2L + -1 -6 El

L]
'3- '3

and

C2

=O

1 woL

C2

+ '3 El = O.

-~woL/EI.

Thus

1 [( x - '3
L)4 /Jx
( - '3
L) - ( x - 3
2L)4 /Jx
( - '3
2L)])
+ 24

332

Exercises
79. The differential equation is
El d4y = 2wo
dx4
L

[!:_
x +(x - !:) a (x _ !:)] .
2
2
2

Taking the Laplace transform of both sides and using y(O)


s4 .:t {y} - sy" (O) _ ylll (O)

= Cl

Letting y"(O)

= y'(O) = Owe obtain

= 2wo [.!:__ _ ~ + ~ e-LS/2] .


ElL

2s

s2

s2

= C2 we have

and y"'(O)

UJ{ } _--+-+-Cl
C2
2wQ [ L
1
1 -LS/2]
----+-e
s3
s4
ElL
2s5
s6
s6

""-Y
so that
y(X) = ~Clx2
2
=
To find

Cl

and

+ ~C2x3 + 2wQ ['!:__x4__ 1_x5 +


El L 48

120

'21ClX 2 + 6'1C2X3 + 60ElWQ L [5L


2X 4 -

!:)
2

_1_ (X -

120

L)
+ ( X - "2

5 (i'1

+ C2X + 60~~L [30LX2 - 20x3 + 20 (X -~)

=u.

(X _

lJij

!:)]
2

"2L)] .

X-

we compute

C2

y"(x) = Cl

o (X - ~)]

and
Y /1' (X)=C2+60ElL WQ
Then y" (L) = u" (L)
ci

= Oyields the

[60Lx-60x

Cl

and

y(x)

+ c2L + 60El L 30L - 20L + '2L


Wo

Solving for

C2

we obtain

wQL2

= 48ElX

2-

C2

+ 60El L[60L

Cl

= woL2/24El

woL
24El

+60

L)

x-"2

2 o

( x-"2'

L)]

system

[3

WQ

WQ

+ 60ElL

3] =

- 60L
and

[5L

2X

C2

5woL2

Cl

+ C2L + 24El

woL

+ 15L ] =

C2

+ 4El

= -woL/4El.

4 - X5'(+

X-

=O

= O.

Thus
L)5

"2

(
L)]
o X - "2

80. The differential equation is


d4y
El dx4

= wo[l - o (x - L/2)J.

Taking the Laplace transform of both sides and using y(O)

Letting y"(O)

= ci

S4.:t{y} _ sy"(O) - y"'(O) = Wo ~


El s
and y'''(O) = C2 we have
.:t{y} =

3. +
s3

C2

s4

+ Wo ~

El s5

333

= y'(O) = Owe obtain

(1 _ e-Ls/2).

(1- e-Ls/2)

7.3

Exercises 7.3
so that
1 2 + -C2X
1 3 + -1 -Wo [4X y(X) = -C1X
2
6
24 El
To find

Cl

and

C2

= y"(L)
1
-clL
2

ci

( - "2
L)] .
+ C2X+ "21WO[2
El X - ( X -"2L)2 llUx

= Oyields the system

2 +-C2L
1
3 +-1 Wo [4L
6
24 El
C

Solving for

-L)4 llU( X - -L)] .

we compute
y " (x) = Cl

Then y(L)

(X-

and

C2

we obtain

4]

- (L)
2

1
2

= -clL

Wo [2L -"2(L)
+ C2L + "21 El
Cl

1~8

woL2/EI and

2 +-C2L
1
3 +--L5wo
2]

C2

128EI

4 =0

3wo L 2 = O.
= ci + C2L + 8EI

= -lS278

Wo (9
2 2
19 3
1 4
1 (
y(x) = El 256L x - 256Lx + 24x - 24 x

wo~/EI.

-"2L)4

r711(
'u,

Thus

-"2L)) .

81. In order to apply Theorem 7.7 we need the function to have the form f(t-a)llU(t-a).

To accomplish

this rewrite the functions given in the forms shown below.


(a) 2t

+ 1 = 2(t -

1 + 1) + 1 = 2 (t - 1) + 3

(b) et = et-s+s = et-S


(c) cost=-cos(t-7r)
(d) t2-3t=(t-2)2+(t-2)-2
82. (a) Prom Theorem 7.6 we have :t{tekti}

= 1/(8 -

ki)2. Then, using Euler's formula,

:t {tekti} = :t {t cos kt + it sin kt} = :t {t cos kt} + i :t {t sin kt}


=

(8 - ki)2

(8+ki)2
82_k2
.2k8
=
+ Z ---=----,,-,-;:(82 + k2)2
(82 + k2)2
(82 + k2)2 .

Equating real and imaginary parts we have


82 _

:t{tcoskt}

= (82

k2

+ k2)2

and

2k8
:t{t sin kt} = (82 + k2)2'

(b) The Laplace transform of the differential equation is

82 :t {x} + w2 :t {x} =
Solving for :t{x}

we obtain :t{x}

2 8
8

+w

2'

= 8/(82 +w2)2 Thus x


334

= (1/2w)tsinwt.

Exercises

Exercises 7.4

8.::tt {

9.

ie

t}

=s3(s-1)2
s-l
(s+1)[(s-1)2+1J

::t{e-t*etcost}=

(s-2)(s2+1)

10. ::t{e2t*sint}

11. ::t{fot erdT}

= ~::t{et} = S(s~ 1)

12. ::t{t

COSTdT} = ~::t{cost}
t

13. ::t{fo CrcosTdT

14. ::t{l

TSinTdT}

15. ::t{fot Tet-r dT}

= S(S2S+ 1)

s2 ~ 1

1
-t
1
S + 1
= ~ ::t{e
cost} = ~ (s+1)2+1

= ~ ::t{tsint}
= ::t{t}::t{et}

1(
=~ -

7.4

dI)

+1

s(s2+2s+2)

1 - 2s
_
2
ds s2 + 1 = -~ (s2 + 1)2 - (s2 + 1)2

S2(Sl_1)

335

Exercises

7.4

16. ::t{ rtsinrCOS(t-r)dr}

=.:t'{sint}::t{cost}

Jo

17.

::t{t

r Sinrdr}

= -!!:_::t{ rt Sinrdr}

Jo

= - :S .:t'{t

18. ::t{t fat re-Tdr}


19. (a) ::t-1{

(b) ::t-1{

(e) ::t-1{

s(s - 1)
21

s3(s - 1)

re-Tdr}

{l/(S

} = ::t-1

s (s - 1)

Jo

ds

-dd

(~-2_+11_) = s2(s2+1)
3s2 + 1 2

-1)} = Jort eTdr

(S: 1)2) = S2~;: ~)3


= et-1

rt(eT_1)dr=et_t_1

Jo

=::t_1{1/S2(S-1)}
s

20. (a) The result in (4) is

S ss

= - :S

}=::t_1{1/S(S-1)}=
s
}

s
(s2+1)

::t-1{F(s)G(s)}

=f

(t(eT-r_1)dr=et_~t2_t_1

Jo

* g, so identify

2k3
F(s) = (s2 + k2)2

and

4s
G(s) = s2 + k2 .

Then

f(t)

= sinkt

- ktcoskt

g(t) = 4coskt

and

so
::t-1 {(S2S:3;2)2 } = ::t-1 {F(s)G(s)}

=f

*9 =4

= 4 fat (sin kr - kr cos kr) cos k(t

f(r)g(t

- r)dt

- r)dr.

Using a CAS to evaluate the integral we get

::t

-l{

(s2Sk+3S}.
k2)3 . = tsm kt - kt cos kt.

(b) Observe from part (a) that


::t{t(sinkt - ktcoskt)}
and from Theorem 7.8 that ::t{tf(t)}
2k3/(s2+k2)2, so

-F'(s).

336

Sk3s

(s2 + k2)3'

We saw in (5) that ::t{sinkt - ktcoskt}

Exercses 7.4

22.

:f{f(t)}

23.

:f{f(t)}

25.

:f{f(t)}

26. :f{f(t)}

12
(a e-stdt
1 - e- as Jo
1-

= 8 (1 +1e-as )

e- bs Jo(b~bte-stdt = ~8 (b18 _ -be s 1-)


- 1

io

1f

1 - e-1fS o
1 2
1- e- 1fS

e-st sintdt = --. 1

+1

82

+ e-1fsj2 = -- 1 coth- 7f8


- e-1fsj2
82 + 1
2

e1fsj2
e1fsj2

r e-st sin t dt = -21+ 1 . 1- 1e-

Jo

1fS

27. The Laplace transform of the differential equation is


28

8:f{y} + :f{y} = (82 + 1)2'


Solving for :f {y} we obtain
28

:f{y} = (8 + 1)(82 + 1)2 =

Thus
1

Y (t) = - - e- t
2
=

1.1

l.

l.

- - sm t + - cos t + - (sin t - t cos t) + -2 t sm t


2

+ -l' cost

1t

--e-

1
- -tcost
2

+ -l.r sm t.
2

28. The Laplace transform of the differential equation is


2(8 - 1)

8:f{y} - :f{y} = ((82 _ 1)2 + 1)2 .


Solving for :f {y} we obtain
2

:f{y} = ((8 _ 1)2 + 1)2 .


Thus

y = et sin t - tet cos t.


29. The Laplace transform of the differential equation is

822{y}

-2 8 + 1 - 2 82 + 1 + 2 82 + 1 + (82 + 1)2 + (82 + 1)2'

- 8y(O) - y'(O) + 92{y}

337

= ~9'
8

Exercises 7.4
Letting y(O)

= 2 and

y'(O)

;t

= 5 and

solving for ;t{y}

we obtain

= 253 + 552 + 195 - 45 = ~

{y}

(52+9)2

+ _5_
+
5
52+9
(52+9)2'

52+9

Thus

5 ,

1,

Y = 2 cos 3t + "3 sin 3t + i3t sin 3t,


30. The Laplace transform of the differential equation is

52;t{y}

- 5Y(0) -y'(O) +;t{y}

= -2-1-,
5 +1

Solving for ;t {y} we obtain


53 - 52 + 5

.i'{y} = (52 + 1)2 = 52 + 1 - 52 + 1 + (52 + 1)2 '


Thus
y

= cos t - 2' sin t - 2' t cos t,

31. The Laplace transform of the differential equation is


52;t{y}

- 5y(0) - y'(O) + 16;t{y}

=;t {cos 4t - cos 4t Ilfl(t - 7f)}

or

= 1+ 25

(52+ 16);t{y}

+ 16

-e-7fS;t{cos4(t+7f)}

= 1 + 52 + 16 - e-7fS ;t{cos4t}
= 1+

52 + 16

e-7fS

52 + 16

Thus

;t{ }
1
5
y =52+16+(52+16)2-(52+16)2e

-7fS

and

y = ~sin 4t + ~ t sin 4t - ~ (t - 7f)sin 4 (t - 7f)OU (t - 7f),

4
8
8
32. The Laplace transform of the differential equation is

52 ;t {y} - 5y(0) - y' (O) +;t {y} =;t { 1 -Ilfl (t - ~) + sin t OU( t - ~) }
or

(52 + 1) ;t{y}

= 5+ ~ - ~

e-7fs/2 + e-7fS/2;t{ sin (t +~)}

= 5 + ~ - ~e-7fs/2 + e-7fs/2 ;t{cost}


5

+ ~ _ ~ e-7fs/2 + _5_ e-7fs/2


5

52 + 1

338

Exercses 7.4
Thus
.,'i'{ }

s
s2

+ 1 + s(s2 + 1)

S
= s2

+ 1 + ~-

1 (1~ -

= ~-

s
s2

+1-

S)
s2

1
- s(s2

+1

+ 1)

(1~ -

-1rs/2

e-1fs/2

S)
s2

+1
s

(s2

(s2

-1fs/2

+ 1)2 e

+ 1)2
+

e-1fs/2
S

(s2

+ 1)2 e

-1fs/2

-1fs/2

and

(t - ~) ] VU (t - ~) + ~ (t - ~) sin (t - ~) VU (t - ~)
(1 - sin t) VU (t - ~) + ~ (t - ~) cos t VU (t - ~) .

y = 1 - [1 - cos

= 1-

33. The Laplace transform of the differential equation is


2

s .,'i'{y}

2s

+ .,'i'{y}

= (s2

+ 1) + (s2 + 1) 2 .

2s

50

Thus

~{y}

(s2

+ 1)2 + (s2 + 1)3

-50

and, using Problem 20,


Y

34. (a)

= - (sin t - t cos t) + - (t sin t - t2 cos t).

y
1

-1

(b)

y
4
2
1

-2
-4

339

Exercises 7.4
35. The Laplace transform of the given .equation is
~{f}
Solving for ~{f}

+~{t}~{f}

1
= -2--.

we obtain ~{f}

Thus, f(t)

+1

= ~{t}.

= sin t.

36. The Laplace transform of the given equation is

= ~{2t}

~{f}

- 4~{sint}~{f}.

Solving for ~ {f} we obtain

= 2s2 + 2

= ~ 2_

s2(s2+5)

{f}

5s2

+ _8_

V5

5V5s2+5

Thus
28.

= -t +

f (t)

~ sin Vs t.

5v5

37. The Laplace transform of the given equation is

Solving for ~ {f} we obtain

~{

}_

_ ~ _1_ + ~

s2

1
+ ~ 2 _ ~ _1_
4 (s - 1)2 4 (s - 1)3 88+ 1 .

f - (s - 1)3(s + 1) - 8 s - 1

Thus
f(t)

1 t 3 t 12 t
= -e
+ -te + -t e --el_t
8

38. The Laplace transform of the given equation is

~{f}
Solving for ~{f}

+ 2~{cos

t}~{f}

= 4~{ e-t} + ~{sin t}.

we obtain
~{

} _ 4s2 + s + 5 __ 4_ _
7
+4
2
(s + 1)3 - S + 1 (s + 1)2
(s + 1)3.

f -

Thus

39. The Laplace transform of the given equation is

~{f}
Solving for ~{f}

we obtain ~{f}

+~{1}~{f}

1
= --.

s+l

= ~{1}.

Thus, f(t) = e-t

340

Exercses 7.4
40. The Laplace transform of the given equation is

Solving for .:t'{f} we obtain

.:t'{f} - _-

82

+1

+ _-1
82 + 1 '

Thus

f(t) = cost + sin t.


41. The Laplace transforrn of the given equation is

.:t'{f} =.:t' {1}+.:t' {t} - ~ .:t'{t3}.:t'{f},


Solving for .:t'{f} we obtain

+ 1)

.:t'{f} =

82(8
84 +

16 =

83
84 + 16

82
84 + 16

'

Thus

f (t)

= cos

J2 t cosh .J2 t +

1M(sin .J2 t cosh

.J2 t + cos .J2 t sinh .J2 t) ,

2v2

42. The Laplace transform of the given equation is

Solving for .:t'{f} we obtain


82 -

.:t'{f}

1 1

= 2T = 2 82

1 31
12 84

'

Thus
1
2

1 3
.
12

f(t) = -t - -t
43. The Laplace transform of the given equation is
8 .:t' {y}

- y ( O) = .:t' { 1} - .:t' { sin t} - .:t'{ 1} .:t'{y} .

Solving for .:t'{f} we obtain

.:t'

= 83 - 82

{y}

+8 =

8(82+1)2

_1_

+~

82+1

28
2(82+1)2'

Thus
.

1,

Y = sin t - 2 t sin t.
44. The Laplace transform of the given equation is

8.:t'{y} - y(O) + 6.:t'{y} + 9.:t'{1}.:t'{y}


341

.:t'{1}.

Exercises 7.4

Solving for ~{f}

we obtain ~{y}

= (

8+3

)2' Thus, y = te-3t.


i
30

45. The differential equation is


di
0.1 dt

+ 3i + 0.05 lo

20

i(T)dT = 100[)L(t -1) -)L (t - 2)]

10

or

si

.5 3

= 1000[)L(t - 1) -)L (t - 2)],

di + 30i + 200 (t i(T)dT

Jo

-10

-20
where i(O) = O. The Laplace

transform of the differential equation is

.
200
8~{2} - y(O) + 30~{i} + -~{i}

1000

= _(e-S

-3 O

_ e-2S).

Solving for ~ {i} we obtain


~{i}

= 1000e-s

-1000e-2s
82+308+200

~)(e-S
8+108+20

_ e-2s).

= (~-

Thus
i(t) = 100(e-10(t-l) - e-20(t-l)) )L(t - 1) - 100(e-10(t-2) - e-20(t-2))
i

46. The differential equation is


di
1
0.005 dt + i + 0.02

)L(t - 2).

lo i(T)dT
t

= 100[t - (t -l))L(t

r------.,

-1)]
1.5

or
d'
t
__:.+ 200i + 10,000 { i(T)dT
8t
Jo

= 20,000[t - (t - 1) )L(t - 1)],


0.5

where i(O) = O. The Laplace transforrn of the differential equation is


8~{i}

+ 200~{i} + 10,000 ~{i}


8

0.5

1.5

= 20,000( 12- 12e-S).


8

Solving for ~ {i} we obtain


~{i}

20,000 (1 _ e-S) = [~_


2
_
200
] (1 _ e-S)
8(8 + 100)2
8 8 + 100 (8 + 100)2
.

Thus
i(t) = 2 - 2e-100t - 200te-100t

- 2)L(t - 1) + 2e-100(t-l) UiL(t- 1) + 200(t - 1)e-100(t-l) '11 (t - 1).

342

Exercises

47. The differential equation is

di
dt
where i(O)

7.4

i
1

+ i = E(t),

= O, The Laplace transform of this


4

equation is
s:t{i}

-0,5

+ :t{i}

= ~{E(t)},
-1

From Problem 21 we have


:t{E(t)}

1- e-s
= s (1 + e-S )'

Thus
(s

+ 1) :t{i}

1 -e -s
s + e- S)

= (1

and
1 -e -s
~ = s(s + 1)(1 +

:t{"}

= (~ __

1_) (1_ e-S)(l

s+ 1

s
=

e-S)

1 -e -s
s(s+l)l+e-s

( -1s - --s+l1)

(1 - 2e -s

_ e-s

+ 2e -2s

+ e-2s

_ e-3s

- 2e -3s

+ e-4s

+ 2e -4s

_ " ,)

- , , , ),

Therefore
i(t) = [1- 2OU(t -1)
- [e-t

+ 2OU(t - 2) -

+ 2e-(t-l)

+ 2e-(t-3)

2OU(t -'3)

OU(t - 1) - 2e-(t-2)
OU(t- 3) - 2e-(t-4)

+ 2OU(t - 4) -" ,]

OU(t - 2)
OU(t - 4)

00

= 1- e-t +2 2:::(-lt(1-

e-(t-n))OU(t

n=l

343

- n).

+ ... ]

Exercises 7.4

48. The differential equation is

di .
dt + 2 = E(t),

1
0.5

where i(O) = O. The Laplace transforrn of this


equation is
-0.5

+ :t{i}

s:t{i}

:t{E(t)}.
-1

Frorn Problern 23 we have

:t{E(t)}

= -1(1
- - - 1)
= -1 - -1 -. 1
s s eS - 1
s2 s eS - 1

Thus

(s + 1) :t{i}
and

:t{"}
2

= s2(s + 1)

Therefore

1 1
( ---+-s2 S

1 1 1
- --s2 s eS - 1

=-

1
1
s(s+1)es-1

1) - (1
---- 1) (-S
e
S+1
s s +1

i(t) = (t -1 + e-t) - (1- e-(t-l))r7U(t


- (1- e-(t-3))r7U(t
=

(t - 1+ e-t) -

+e

-2s

+e

-38 +e -48

-1) - (1- e-(t-2))r7U(t

- 3) - (1- e-(t-4))r7U(t

- .... )

- 2)

- 4) - ...

00

L (1- e-(t-n))r7U

(t - n).

n=l

49. The differential equation is x"

+ 2x' + lOx = 20f(t),

where f(t) is the rneander function with a

Using the initial conditions x(O) = x'(O) = Oand taking the Laplace transforrn we obtain

(S2 + 2s + 10) :t{x(t)}

= 20 (1 _ e-7rs)__ l_

1 + e-7rS

20

= -(1 - 2e-7r8+ 2e-27rs_ 2e-37rs+ ... )

344

= tt .

Exercses 7.4
Then

.:t'{ (t)}
X

. 20
40
~ ( l)n -n7rS
= 5(52+ 25 + 10) + 5(52+ 25+ 10) ~l e
2
=~-

25 + 4
~ ( )n [4
45+ 8 ] -n7rS
52+25+1O+~
-1
~- 52+25+10 e

=~_2(5+1)+2+4f.
5 (5+1)2+9

(_l)n[~_

(5+1)+1]e-n7rS
(5+1)2+9

11=1

and

x(t) = 2 (1- e-tcos3t

+ 4~(-l)n[l-

- ~e-tsn3t)

e-(t-n7r)

cos3(t - mr)

- ~e-(t-n7J") sn3(t - mr)] lJU(t- mr).


The graph of x(t) on the interval [O,2n) is shown below.
x

-3

50. The differential equation is x"

+ 2x' + x

= 5f(t),

where f(t) is the square wave functon with a = tt .

Using the initial conditions x(O) = x'(O) = O and taking the Laplace transform, we obtain

(52+ 25 + 1).:t'{x(t)} = ~
5

1 + e-7rS

~(1 _ e-7rS + e-27rs _ e-37rs + e-47rs _ ... )


5

Then

.:t'{x(t)} =

5
f (_l)"e-n7rs
5(5 + 1)2 n=O

= 5 f (-1)" (~ __ 1_ _
n=O

5+ 1

1 ) e-n7rs
(5 + 1)2

and
00

x(t) = 5

L (-1)"(1

e-(t-n7r) -

n=O

345

(t - nn)e-(t-n7r))

lJU(t- nn).

Exercises

7.4

The graph of x(t) on the interval [O,4Jr) is shown below.


x

-5
51. f(t)

= -~

.,;t'-l{ ~(ln(s

ds

3)- ln(s + l)l} = _~t .,;t'-l {_ls - 3

l_} = _~t (

+1

t -

e-t)

52. By definition, t *q(t - a) = IJ(t - 7)q(7 - a) dr . We consider separately the cases when a
and when a

> t. When a < t, then


= (t (t _ 7) dr = _ (t - 7) 2 It = ~ (t - a f

(t (t _ 7) q (7 _ a) dr

Jo
When a

> t, then q (7 - a)

l;
7 < t < a and

= O since

(t - 7) q (7 - a) dr = O.

Therefore

hq(t-a)

> a = ~(t-a)2q(t-a).
t <a 2

{i(t-a)2,

O,

53. First method: By the definition of the Laplace transform and integration by parts we have

{l f (7)

.,;t'

dr }

1000 e =st (l f (7) d7)

dt

= _ e-st (t f(7) dr 100 +~ (00 e-st f(t) dt

s Jo

= ~ (00 e-st f(t)

s Jo

dt = F(s) .

Jo

Second Method: Let g(t) = Id f(7) dr ; then g'(t) = f(t) and g(O)

= O. By Theorem

.,;t'{g'(t)} = s .,;t'{g(t)} - g(O) = s .,;t'{g(t)},


so

and

.,;t'{f(t)} = s.,;t'
.,;t'{l

f(7) d7}

{la

f(7) d7} ,

= ~.,;t'{f(t)}

346

F;S) .

7.8

<t

Exercses
54. Let u

7.4

= t - T so that du = dr and

* 9 = 10t f(T)g(t

- T) dr = -

iD f(t

- u)g(u)du = 9 * f.

55. (a) Using Theorem 7.8, the Laplace transforrn of the differential equation is
d

- ds [s2y - sy(O) - y'(O)] + sY - y(O) + ds [sY - y(O)] + nY

= -_

dx

[s2y] + sY + - [sY] + nY
ds

=:
(s - s2)

y) - 2sY + sY + s(:s y) + y + nY

(!

y) + (1 + n - s)Y = O.

Separating variables, we find

dY = 1 + n-s ds = (_n __
y
s2 - S
S - 1
In Y = n ln(s ~ 1) - (1 + n) In s
Y

_
- CI

1 + n) ds

+c

(s - l)n
sl+n

Since the differential equation is homogeneous, any constant multiple of a solution will still
be a solution, so for convenience we take

el

= 1. The following polynornials are solutions of

Laguerre's differential equation:

n=O:
n = 1:
n = 2:
n

= 3:

n = 4:'

(b) Letting f(t)

= tne-t we note that f(k)(O)

= O for k = 1, 2, 3, ... ,

347

n - 1 and f(n)(o) = n!.

Exercises 7.4
Now, by the first translation theorem,

where Y

=::t {Ln(t)}.

Thus

56. (a) The output for the first three lines of the prograrn are

9y[t] + 6y'[t] + y"[t] == t sin[t]


1 - 28 + 9Y
Y-->-

+ 8 Y + 6(-2 + 8Y)

==

28
(1 + 82)2

-11 - 48 - 2282 - 483 - 1184 - 285)


(1 + 82)2(9 + 68 + 82)

The fourth line is the same as the third line with Y

-->

removed. The finalline of output shows

a solution involving complex coefficients of eit and e:". To get the solution in more standard
form write the last line as two lines:
euler = {E~(I t)-> Cos[t]+1
InverseLaplaceTransform

Sin[t], E~(-I t)-> Cos[t]-I

Sin[t]}

[Y,s,t] j .euler j jExpand

We see that the solution is


487

247)

y(t) = ( 250 + 5t e-3t + 250 (13 cos t - 15t cos t - 9 SIn t + 20t sin t) .
(b) The solution is

348

Exercises

7.5

(c) The solution is


q( t)

= 1 - cos t + (6 - 6 cos t) q,(t - 37f)- (4 + 4 cos t) q,(t - 7f).


q

Exercises 7.5

l. The Laplace transform of the differential equation yields

.:t'{y} = ~3e-2S
8-

so that
y

= e3(t-2)q,(t - 2).

2. The Laplace transform of the differential equation yields


2

e-s

8+1

8+1

.:t'{y} = -+so that


y = 2e-t

+ e-(t-l)q, (t -

1).

3. The Laplace transform of the differential equation yields

.:t'{y} = 82~1 (1+e-2TfS)


so that

= sin t + sin t q,(t - 27f).

4. The Laplace transform of the differential equation yields

.:t'{ } = ~
y

82

+ 16

e-2Tfs

so that
y

.
41 sm4(t

- 27f)q,(t - 27f).

5. The Laplace transform of the differential equation yields

.:t'{y} = _1_ (e-Tfs/2 + e-3TfS/2)


82

+1

349

Exercises 7.5
so that
y

= sin (t - ~)OU (t - ~) + sin (t - 3;)OU (t - 321T)


= - cos t

OU(t - ~) + cos t OU(t - ~) .

6. The Laplace transform of the differential equation yields

.s:'{y} = _8_ + _1_(e-271'S


82

+1

82

+1

+ e-471'S)

so that
y

= cos t + sin t [OU(t - 21T) + OU(t - 41T)J .

7. The Laplace transform of the differential equation yields

.s:'{y} =

1 (1 + e-S) = [~~_ ~ _1_] (1 + e-S)


82+ 28
2 8 2 8+ 2

so that
y

1 -2t + [1
1 -2(t-l)]IilI(
= -1 - -e
- - -e
'u t - 1).
2

8. The Laplace transform of the differential equation yields

.s:' y =
{}

8+ 1 +
1
e-2s = ~_1 __
82(8-2)
8(8-2)
48-2

~ ~ _ ~ ~ + [~ _1 __
48
282
28-2

~~] e-2s
28

so that
y = ~e2t _ ~ _ ~t + [~e2(t-2) - ~] OU(t - 2).
4
4 2
2
2
9. The Laplace transform of the differential equation yields

.s:'{ }

1
y = (8+2)2+1e

-271's

so that
y = e-2(t-271') sin tOU(t - 21T).
10. The Laplace transform of the differential equation yields

.s:'{y} =

(8:

1)2e-s

so that
y = (t - l)e-(t-l)OU(t

- 1).

11. The Laplace transform of the differential equation yields


4+8
e-71's + e-371's
.s:'{y} = 82+48+13 + 82+48+13

2
3
8 +2
3
(-71'S
+
+- 1
e
+e -371'S)
3 (8 + 2)2+ 32 (8 + 2)2+ 32 3 (8 + 2)2+ 32

=-

350

Exercses 7.5
so that
y

= ~e-2t sin 3t + e-2t cos 3t + ~e-2(t-7l")sin 3(t - 71fU (t - ?T)


3

+ ~e-2(t-37l")sin3(t

- 3?T)OU(t - 3?T).
3
12. The Laplace transform of the differential equation yields
1

e-2s

+ e-4s

.z:{y} = (s _ 1)2(s - 6) + (s - l)(s - 6)


1
= - 25

s- 1-

'5 (s -

1 1
1)2 + 25 s - 6

[1

+ -'5 s -

1 1 ] (-2S
1 + '5 s _ 6 e

so that
y

_~et

25

_ ~tet + ~e6t
5

+ [_~et-4

25

+ [_~et-2

+ ~e6(t-2)] OU(t - 2)

+ ~e6(t-4)] OU(t - 4).

13. The Laplace transform of the differential equation yields

.z:{}
1 2 "(O)
1 3! /1'(0)
1 Po 3!
y ='2s3Y
+6"s4Y
+6"Els4e
so that

-Ls/2

2'L) 3 OU( x - 2'L) .

1 /1 ) 2 1 /1,( 3 1 Po (
y = '2Y (O x + 6"Y O)x + 6" El X Using y/l(L) = O and ylll(L) = Owe obtain

1::)3 OU(x _ 1::)

y = ~ PoL x2 _ ~ Po x3 + ~ Po (x _
4 El
6 El
6 El
2

Po (!:"x2 _ ~x3)
O < x < !:..
El 4
6
' 2
- { POL2 (~x _ !:___)
4El 2
12'

1:: < x < L.


2 -

14. From Problem 13 we know that


1 /1 ) 2 1 111 ) 3 1 Po (
L ) 3 OU(
L)
y = '2y (O x + 6"Y (O x + 6" El X - 2'
x - 2'
Using y(L)

= Oand

y'(L)

= Owe obtain

1::)3 ou(x _ 1::)

y = ~ PoL x2 _ 2._ Po x3 + ~ Po (x _
16 El
12 El
6 El
2

O<
-x
L

-<x<L.
2 351

<-2

+e

-4s)

Exercises 7.5
15. The Laplace transform of the differential equation yields
1

~{Y}=82+w2
so that y(t) = sinwt. Note that y'(O) = 1, even though the initial condition was y'(O) = O.

Exercises 7.6

1. Taking the Laplace transform of the system gives

8~{X}

+ ~{y}

= -~{x}

8~{Y} -1 = 2~{x}
so that
1

~{x} = (8 -1)(8 + 2) = 3" -8---1 - 3" -8-+-2


and

Then
x

1
_et
3

1
_e-2t
3

and

2
3

1
+ _e2t

Y = _et

2. Taking the Laplace transform of the system gives


1

8~{x}-1=2~{y}+

8-1

8~{Y} - 1 = 8~{x}

- 2"
8

so that
~

{y}

83 + 782 - 8 + 1
8(8 - 1)(82 - 16)

1 1
8
1
173 1
53 1
16 -; - 158 - 1 + % 8 - 4 - 160 8 + 4

and
1
8 t
y = - - -e
16 15

173 4t
+ -e
96

53
-e
160

-4t

Then
x

1,

sy

st1 = st lIt

173 4t
15e + 192e

3. Taking the Laplace transform of the system gives

8~{X}

+ 1 = ~{x} - 2~{y}

8~{Y} - 2 = 5~{x}

352

-~{y}

53

-4t

+ 320e

Exercises 7.6
so that
-8 -

3 s2 + 9

::t{x} = s2 + 9 = - s2 + 9 and

5 .
3

x = - cos 3t - - sm 3t.
Then
1
1
223

7 .

y = -x - -x = 2cos3t - - sm3t.
4. Taking the Laplace transform of the system gives
1
s

(8 + 3) ::t{x} + s::t{y} = (8 -l)::t{x}

+ (8 -l)::t{y}

= 8 -1

so that
5s - 1

11

::t{y} = 38(S _ 1)2 = -3

~+ 3 s -

3 (s -

1+

1
1)2

and

::t{x} _ 1 - 28
11 1 1
1
1
- 3S(8 - 1)2 - 3 ~ - 3 s - 1 - 3 (s - 1)2 .
Then
X

1 t
= -lIt - -e - -te
and
3

Y = --

lIt

4 t
+ -e + -te.
3

5. Taking the Laplace transform of the system gives


1
s

(2s - 2) ::t{x} + 8::t {y} = -

(s - 3)::t{x} + (8 - 3) ::t{y} = -

so that
-s - 3

11

::t{x} = s(8-2)(8-3)=-2~+2s-2and

38 - 1

8-3

11

::t{y} = S(8 _ 2)(s - 3) = -6

~- 2 s -

+ '3 s -

3'

Then
x

5 2t
3t
1 5 2t 8 3t
= -- 1 + -e
- 2e
and y = -- - -e +-e
2

6. Taking the Laplace transform of the system gives

(8 + 1)::t{x} - (8 - l)::t{y} = -1
8::t{X} + (s+2)::t{y}
353

=1

Exercises 7.6
so that

:L{ } _
Y

8 + 1/2 _
8 + 1/2
82 + 8 + 1 - (8 + 1/2)2 + (V3/2)2

and

:L{x} =

-3/2
82 + 8 + 1

-3/2
(8 + 1/2)2 + (V3/2)2'

Then
y =e

_/

t 2

V3
cos -t

. V3
x = e -t/2 Sln-t
2 .

and

7. Taking the Laplace transform of the system gives

+ l):L{x} - :L{y}

(82

=-2

+ (82 + 1) :L{y} =

-:L{x}

so that
-282 - 1

1 1

:L{x} = 84+282 = -"2 82 -"2 82+2


and
x =-

1
2

3.
2V2

+t - -

Then
y =x

11

Slll V 2 t.

+ x = --t1 +
2

3
r:
In sin v 2 t.
2v2

8. Taking the Laplace transform of the system gives

(8 + 1) :L{x} + :L{y} = 1
4:L {x} - (8 + 1) :L {y} = 1
so that

and
-8

+3

8+1
2
(8 + 1)2 + 22 + 2 (8 + 1)2 + 22 .

:L {y} = 82 + 28 + 5
Then
x

= e-t cos 2t + 2e-t sin 2t and y = _e-t cos 2t + 2e-t sin 2t.

9. Adding the equations and then subtracting them gives

d2x
dt2

"2t2+ 2t

d2y
1
= _t2 - 2t.
dt2
2

354

Exercises
Taking the Laplace transform of the system gives
1

:t'{ x}

= 8~

and

1 4'

1 3!

+ 24 s5 + '3 s4

1 4!

1 3!
3 s4

:t'{y}=----24 s5
so that

10. Taking the Laplace transform of the system gives

(s - 4) :t'{x} + s3:t' {y} = -2-1


s

(S + 2) :t'{x} - 2s3:t'{y} = O
so that
4

414

:t'{x} = (s _ 2)(s2 + 1) =
and

2s + 4

:t'{y}

5s-

2-

= s3(s-2)(s2+1)

=;-

81

5 s2 + 1 - 5 s2 + 1
2

s2-2s3+5s-2-5s2+1

Then
4 2t

x = -e
5

- - cos t - - sin t
5
5

and
y = 1 - 2t - 2t

,
+ 51 e2t - 56 cos t + 58 sin
i.

11. Taking the Laplace transform of the system gives

S2:t'{X} + 3(s + l):t'{y} = 2


1

s :t'{x} + 3:t'{y} = (s + 1)2


so that

2s + 1
1
1
1 2
1
:t'{x} = - s3(s + 1) = ; + s2 + '2 s3 - S + 1 '

Then
X

= 1+ t +

= '3te

-t - e

-t

and
1

-t

'3x

11

= '3te

355

-t

+ '3e

+5s2+1'

-t

1
3

7.6

Exercises 7.6
12. Taking the Laplace transform of the system gives
2e-s

(8 - 4) ..:t'{x} + 2..:t'{y} = -3..:t'{x}

1 e-s
= 2 +-8

+ (8+ l)..:t'{y}

so that

(J){ x } =

-1/2
(8-1)(8-2)
1

= [28-1-28-2

+e -s
1

1]

1
(8-1)(8-2)
+e

-s

[1
-8-1

and

..:t'

e-s

8/4-1

=45-1-28-2+e

-8/2+2

-s

{y} = -;- + (5 - 1)(5- 2) + e


1

-s

1]
+ 8-2

(5 - 1)(8- 2)
[l 3 1
~-28-1

1]
+5-2

Then
x = ~et - ~e2t

+ [_et-l + e2(t-ll]

y = ~et _ ~e2t
4
2

+ [1 -

VU (t - 1)

and
~et-l

+ e2(t-ll] au (t - 1).

13. The system is

+ 2(X2 -

X'{

-3XI

X~

-2(X2 - Xl)

Xl)

XI(O)=O
X~

(O) = 1

X2(0) = 1
x;(O)

= O.

Taking the Laplace transform of the system gives

(82 + 5) ..:t'{xl} - 2..:t'{X2} = 1


-2..:t'{xl}

+ (82 + 2) ..:t'{X2}

=8

so that

..:t'

82+ 28+ 2 _ ~ _8_

{ l} - 84+ 782+ 6 - 5 82+ 1

+ ~ _1 __
5 82+ 1

and

356

~ _8_ + _4_ V6
5 82+ 6 5V6 52+ 6

Exercses

:L x
{2}

83 + 58 + 2 = ~ _8_
+ ~ _1_ + ~ _8
(82 + 1)(82 + 6)
5 82 + 1 5 82 + 1 5 82 + 6

7.6

2_ V6
5V6 82 + 6 .

Then
Xl

2
5

+-

1
2
sin t - - cos V6 t
5
5

4
5

2
5

= - cos t

and

1
5

X2 = - cos t + - sin t + - cos V6 t 14. In this system

Xl

4
re sin v'6 t
5y 6
2
;; sin v'6 t.
5y6

and X2 represent displacements of masses mI and m2 from their equilibrium

positions. Since the net forces acting on mI and m2 are

respectively, Newton's second law of motion gives

mlX{ = -kIXI + k2(X2 - xl)


m2X~ = -k2(X2 - Xl) - k3X2
Using kl

= k2 = k3 = 1, mI = m2

= 1,

XI(O) = O, XI(O) = -1, X2(0) = O, and x2(0) = 1, and

taking the Laplace transform of the system, we obtain


(2 + 82):L{Xl}

- :L{X2} =-1

:L{xl} - (2 + 82):L{X2} = -1
so that

Then
Xl

=-

1.
r:
v'3
sin y 3 t

15. (a) By Kirchoff''s first law we have il

E(t)

= Ril

+Lli2 and E(t)

= Ril

and

= iz + i3.

X2 =

1.
v'3
sin h t.

By Kirchoff's second law, on each loop we have

= E(t)

+L2i~ or Lli2+Ri2+Ri3

(b) Taking the Laplace transform of the system


0.01i2

+ 5i2 + 5i3 = 100

0.0125i~ + 5i2 + 5i3 = 100


gives
(8 + 500) :L{i2}

10,000

+ 500:L{i3} = _-

8,000

400:L{i2} + (8 + 400) :L{i3} = --

357

and L2i~+Ri2+Ri3

= E(t).

Exercises 7.6
so that
:t: .
8,000
80 1 80
{Z3} = s2 + 900s = 9" ~ - 9" s

+ 900

Then
.

Z3

(e) il

= i2 + i3

80 80 -900t.
9"
- 9"e

and

Z2

./

= 20 - 0.0025z3 -

Z3

100 100
9
-ge

-900t

= 20 - 20e-900t

16. (a) Taking the Laplace transform of the system

i; + i~ + 10i2 = 120 -

120OU(t - 2)

+ 5i~ + 5i3 = O

-10i;
gives
(s

. + S:t:{Z3}
. = -120 ( 1 + 10) :t:{Z2}

e _)28

+ 5(s + 1) :t:{i3}

-10s:t:{i2}

=O

so that
(LJ

ol.-

{z?J --

120(s

(3s2

+ 1)

+ lls + 10)s

( 1 - e -2s) - [48
- -- 60
- s + 5/3
s+2

+ -12]
s

( 1 - e -28)

and
:t:{i}=
3

3s2

240

+ lIs + 10

=[

(1_e-2S)

240

+ 5/3

_ 240] (1_e-2S)
s +2

Then
iz

= 12 + 48e-St/3 - 60e-2t -. [12 + 48e-S(t-2)/3

i3

= 240e-St/3 - 240e-2t - [240e-S(t-2)/3

- 60e-2(t-2)]

OU(t- 2)

and

(b) il

= i2 + i3 = 12 + 288e-St/3

- 240e-2(t-2)]OU(t

300e-2t - [12 + 288e-S(t-2)/3

- 300e-2(t-2)]

17. Taking the Laplace transform of the system


i;+ 11i2 + 6i3 = 50 sin t
i~ + 6i2 + 6i3 = 50 sin t
gives
(s

+ 11) :t:{z2} + 6:t:{Z3}

6:t:{i2}

+ (s + 6):t:{i3}
358

= s2

50

+1

= _250

- 2).

+1

OU(t - 2)

Exercises
so that

::t i
{2}

50s
(s+2)(s+15)(s2+1)

= _ 20 _1_

13s+2

375 _1_

1469s+15

145 _s_
113s2+1

+~

_1_

113s2+1'

Then

and

i2 = - 20 e-2t
13

375 e-1St
1469

+ 145 cos t + ~
113

113

,
25.
1,/
11 ,
30 -2t
23 = 3sin t - '622- 622 = 13e

250

+ 1469e

sin t
280
- 113 cos t

-1St

810 ,
113 sin t,

18. Taking the Laplace transform of the system

0,5i~ + 50i2 = 60
0,005i~ + i2 - il = O
gives
,

,120

s ::t{2} + 100::t{22} = -200::t{i}

+ (s + 200) ::t{i2}

=O

so that

::t{i } _

24,000

2 - s(s2

s+ 100

cos 100t -

~e-100t

= 6 1

100

S; - S (s + 100)2 + 1002 - S (s + 100)2 + 1002 '

+ 200s + 20,000)

Then
i2 = ~ 555

and

~e-l00t

, =e 0005
,/ + 22
' =S
6
21
u.
22

sin 100t

6 -100t cos 100t ,


- Se

19. Taking the Laplace transform of the system


2i~ + 50i2 = 60
0,005i;

+ i2 -

il = O

gives
,

,60

2S::t{2} + 50::t{22} = -

-200::t{iI}

+ (s + 200) ::t{i2}

=O

so that

::t{' }

6,000

22 = s(s2
61

+ 2008 + 5,000)
6

s + 100

- S; - S (s + 100)2 -

(50)2)2

359

6)2
- -5- (s

50)2

+ 100)2 -

(50)2)2

'

7.6

Exercises 7.6
Then
6 6
6J2.
i2 = - - _e-lOOt cosh 50J2 t - _e-lOOt
5 5
5

sinh 50J2 t

and

20. (a) Using Kirchoff's first law we write il

= i2+i3.

Since i2 = dqjdt we have il-i3

= dqjdt.

Using

Kirchoff's second law and summing the voltage drops across the shorter loop gives
1

E(t) = iRl + Cq,

(1)

so that
.
1
Il = -E(t)

- -q.
RlC

Rl

Then

.
-dq
=
Il dt
and

l.

I3 = -E(t)

- --q - I3
RC

Rl

dq 1
Rl dt + Cq + Ri3

= E(t).

Summing the voltage drops across the longer loop gives

E(t) = ilRl + L ~: + R2i3.


Combining this with (1) we obtain

di3
dt

ilRl + L- + R2i3
or

di3

L- + R2I3
(b) Using L

= iR

+ Cq

Cq = O.

= Rl = R2 = C = 1, E(t) = 50e-tOU(t -1) = 50e-1e-(t-l)OU(t -1), q(O) = i3(0) = O,

and taking the Laplace transform of the system we obtain


(8 + 1) ~{q}

+~

{i3} =

5Ce-1

__

8+1

e-s

(8 + 1)~{i3} - ~ {q} = 0,
so that

and

q(t) = 50e-1e-(t-l)

sin(t - l)OU(t - 1) = 50e-t sin(t - l)OU(t - 1).

360

Exercises

7.6

21. (a) Taking the Laplaee transform of the system

4e~ + e~ + 8el = O
e~ + e~ + 2e2 = O
gives

4 (s2 + 2).:t{el}

+ s2.:t{e2}

+ (s2 + 2) .:t {e2}

s2 .:t {el}

= 3s
=o

so that

or

Then

123
e2 = - eos ""t - - eos 2t
2

vs

and

e~ = -e~ - 2e2

so that

el =
(b)

1
4

2
y3

3
4

- eos ""t + - eos 2t.

91

92

-1

-2

Mass

m2

has extreme displaeements o greater magnitude.

equilibrium position at about t = 0.87, and mass

m2

Mass

m1

first passes through its

first passes through its equilibrium

position at about t = 0.66. The motion of the pendulums is not periodie sinee eos 2t/
period V3 1[", eos 2t has period
number.

1[",

and the ratio of these periods is

361

J3 , which

J3 has

is not a rational

Exercises 7.6

(e) The Lissajous curve is plotted for O

(d1/

t= 1

:s t :s 30.

t= 2

t= 4

92

81

82

1
2
3'
4
5
6
7
8
9
10

-0.2111
-0.6585
0.4830
-0.1325
-0.4111
0.8327
0.0458
-0.9639
0.3534
0.4370

0.8263
0.6438
-l. 9145
0.1715
1.6951
-0.8662
-0.3186
0.9452
-l. 2741
-0.3502

t= 7

t=

(e) Using a CAS to solve el(t) = e2(t) we see that el = e2 (so that the double
pendulum is straight out) when t is about 0.75 seconds.

362

10

t=0.75

Chapter 7 Review Exercises


(f) To make a movie of the pendulum it is necessary to locate the mass in the plane as a function
of time. Suppose that the upper arm is attached to the origin and that the equilibrium position
lies along the negative y-axis. Then mass mI is at (x, (t), Y1(t)) and mass m2 is at (X2(t), Y2(t)),
where

Xl(t) = 16sinBI(t)

and

Yl(t) = -16cosBl(t)

and

Y2(t) = Yl(t) - 16cosB2(t).

and

X2(t) = Xl(t) + 16sinB2(t)

A reasonable movie can be constructed by letting t range from O to 10 in increments of 0.1


seconds.

Chapter 7 Review Exercises


1.

.:t'{f(t)}

= rlte-stdt+

(X)(2_t)e-stdt=

Jo

3. False; consider f(t)


4. False, since

Jl

22e-s
8

t-I/2.

f(t) = (et)IO = elOt.

5. True, since lims->ooF(8)


6. False; consider

7 . .:t'{e-7t}

128

= 1 i- O. (See Theorem 4.5 in the text.)

f(t) = 1 and g(t) = 1.

= _1_

8+7
8. .:t'{te-7t}

9. .:t'{sin2t} =

(8:

7)2

-i+4
8

10. .:t'{e-3tsin2t}

= (8+3~2+4

2]

.
11. .:t'{t sin
2t} = -- d [-2ds 8 + 4

= (2 48 )2
8

+4

2
12. .:t'{sin2tOU(t - 7T)} = .:t'{sin2(t - 7T)OU(t- 7T)} = -2-e-1fS
8 +4

363

Chapter 7 Revew Exercses

17.

:::t-1{

18. :::t-1
19.

20.

s
} =:::t-1
s2 - lOs + 29

12e-5S}

:::t-1{

(s -

s-5
5)2

+ 22

+~

(s -

2
5)2

+ 22

=tcos2t+~tsin2t

= (t - 5)OU(t - 5)

+ 1f 2 e -S}

(1)-1 { S

oL

(1)-1 {

2
s+1f

L2s2 + n21f2

2e

-s

1f
+ s+7i
2

2 e

-S}

=oL

2
s+1f

= COS 1f (t

- 1) OU(t - 1) + sin 1f (t - 1) OU(t - 1)

} _ ~ !:_:::t-1{

- L2 n1f

21.

:::t{e-5t} exists for s > -5.

22.

:::t{teStf(t)}

23.

:::t{eat f( t - k) OU(t - k)}

24.

:::t{l

n1f/L
}
s2 + (n21f2)/ L2

_l-sin~t

Lnat

= - :sF(S-S).

= e-ks

:::t{ea(t+k) f( t)} = e-ks eak :::t{eat f( t)}

= ~:::t{eatf(t)}

eaTf(T)dT}

F(ss-a),

= e-k(s-a)

F( s - a)

whereas

:::t{eat (t f(T) dT} =:::t { (t f(T) dT} I


= F(s) I
= F(s_- a) .
Jo
Jo
s-->s-a
s s-->s-a
S
a
25.

f(t)OU(t-to)

26.

f(t) - f(t)OU(t - to)

27.

f(t - to) OU(t - to)

28.

f(t) - f(t)ilU(t - to) + f(t)OU(t - ti)

29.

f(t) = t - [(t -1) + l]OU(t - 1) +OU(t - 1) -OU(t - 4) = t - (t - l)OU(t - 1) -qj(t - 4)


:::t{f(t)}

2_ - 2_e-s _ ~e-4s
s2

s2

364

Chapter

:L{etf(t)}
30. f(t)

e-(s-l) __ 1_e-4(s-1)
s-l

(s-l)2

= sint"i1(t - 7T) - sintOU(t - 3T) = - sin(t - 7T)"i1(t- 7T) + sin(t - 37T)"i1(t- 37T)

:L{f(t)}

= 2-

:L{f(t)}
:L{etf(t)}

+ _1_e-37rs
s2 + 1

l_e-7rs
S2 + 1

= __

:L{et f(t)} = _
31. f(t)

(s-1)2

7 Review Exercises

e-7r(s-l) +

(s-1)2+1

2"i1(t - 2) + [(t - 2) + 2]"i1(t- 2)


2
s

=-

e-37r(s-1)

(s-1)2+1

= 2 + (t -

2)au(t - 2)

1
+ 2e-2s
s

_2_

s-l

e-2(s-1)

(s-1)2

32. f(t) = t - t"i1(t -1) + (2 - t)"i1(t -1) - (2 - t)"i1(t - 2) = t - 2(t -l)"i1(t

-1)

+ (t - 2)"i1(t -

<-P{f(t )} = -1 - -e
2 -s +-e1 -2s
s2 s2
s2

eL-

:L{et f(t)} =
1
.
(s-1)2

e-(s-l) +

(s-1)2

e-2(s-1)

(s-1)2

33. Taking the Laplace transform of the differential equation we obtain


5

:L{y} = (s _ 1)2 + 2 (s - 1)3


so that

34. Taking the Laplace transform of the differential equation we obtain


1

:L{y}

(s _ 1)2(s2 - 8s + 20)
6

s-4

= 169 s - 1 + 13 (s - 1)2 - 169 (s - 4)2 + 22 + 338 (s - 4)2 + 22


so that

6
169

_et

1
+ -tet
13

6
169

- _e4t

365

cos 2t

+ _t5.

338

sin 2t.

2)

Chapter 7 Review Exercises


35. Taking.the Laplace transform of thegiven differential equation we obtain

~ {}

s3 + 6s2 + 1
1
-2s
2
-2s
+ l)(s + 5) - s2(s + l)(s + 5) e - s(s + 1)(8 + 5) e

y = 82(S
6

13

= - 25 . --;+ 5" . s2 + "2 . s + 1 - 50 . s + 5

_ (_~ . ~ + ~ . ]_ + ~ . _1
1_ . _1_) e-2s
25 s 5 s2 4 s + 1 100 s + 5
+ _]_.
e-2s
5 5 2 s + 1 10 s + 5

_ (~.~_ ~._1_

_1_)

so that
y = _~ + ~t2
25 5

+ ~e-t _

13e-5t
50

i_OU(t - 2) - ~(t - 2)2OU(t - 2)


25

+ ~e-(t-2)OU(t - 2) - _2_ e-5(t-2JOU(t - 2).


4

100

36. Taking the Laplace transform of the differential equation we obtain

~{y} =

s3 + 2 _ 2 + 2s + s2 e-s
s3(s - 5)
s3(s - 5)

=-

2 1
2 1
1 2
125 --;- 25 s2 - 5" s3

127

+ 125 s -

5-

[37
1 12 1
1 2
- 125 --;- 25 s2 - 5" s3

37

l]_s

+ 125 s _

5 e

so that
y

127
= -- 2 - -t2 - -t1 2 + -e
125

25

125

5t

-- 37)2- -(t
125 25

1 - 1)2 + -e
37
- 1) - -(t
5

125

37. Taking the Laplace transform of the integral equation we obtain


1

1 2

~ {y} = --;+ s2 + "2 s3


so that

38. Taking the Laplace transform of the integral equation we obtain

~{f}
so that f(t)

= 6t.

39. Taking the Laplace transform of the system gives

s~{x}
4~{x}

+~{Y}

1
s

2" + 1

+ s~{Y} = 2
366

1
s

= 6 2"

5(t-1J]

OU(t- 1).

Chapter 7 Review Exercises


so that

.:t'{x}

S2 - 2s + 1
s(s _ 2)(s + 2)

11

= -4 ~ + 8 s -

+ 8 s + 2'

Then
x

119

= -- + _e2t + _e-2t
4

and

Y = -x

9 -2t
1 2t
+ t = -e
- -e + t.

40. Taking the Laplace transforrn of the systern gives


1

s2 .:t'{x}

+ s2 .:t'{y} = s _ 2

2s.:t'{x}

+ s2.:t'{y}

1
=- s _ 2

so that
2
11
1
.:t'{x} = s(s-2)2=2~-2s-2+(s-2)2
and

.:t'

-s - 2 __ ~ ~ _ ~ 2_ + ~ _1_ _

{y} - s2(s - 2)2 -

4 s

2 s2

4 s - 2

(s - 2)2 .

Then
x

1
2

1
2

= - - _e2t + te2t and y

= -- - -t
4
2

3 2t
2t
+ -e
- te .
4

41. The integral equation is


10i + 2 fat i(T) dr = 2t2

+ 2t.

Taking the Laplace transforrn we obtain


.

.:t'{t}=

( 4

s+2
= s2(5s+2)

2)

s3

+ s2 10s+2

45

=-~+

s2

+ 5s+1

=-~+

Thus
i(t) = -9 + 2t

+ ge-t/5.

42. The differential equation is


1 d2q

dq

2 dt2 + 10 dt + 100q =

10 - 10OU(t - 5).

Taking the Laplace transforrn we obtain

.:t'{ }

q == 2 (s2

20

+ 20s + 200)

_ [2_ ~ _ 2_
-

10 s

10 (s

(1

- e

-5S)

s + 10 _ 2_
10
] (1 _
+ 10)2 + 102 10 (s + 10)2 + 102
367

s2+ s+1/5'

e-5S)

Chapter 7 Review Exercises


so that
111
_e-10t cos 10t - _e-10t sin 10t
10 10
10

q(t) = - -

- [~

10

- ~e-10(t-5)

10

cos 10(t - 5) -

~e-10(t-5)

10

sin 10(t - 5)] OU(t- 5).

43. Taking the Laplace transform of the given differential equation we obtain

.:t'{y} = 2wo
ElL

(.!:_.

4885

4! __ 1_ . ~ + _1_ . ~ e-SL/2)
12086
12086

+ Cl . ~ + C2. 3!
283
684

so that

2wo [L 4
1 5
1 (
y = El L 48 x - 120x + 120 x
where y"(O) = Cl and y"'(O)

= C2. Using y"(L)

ci 2
C2 3]
-"2L) 5 OU( x -"2L) +"2
x +"6x

= O and

. Cl = woL2/24El,

y"'(L) = Owe find

C2= -woL/4El.

Hence

44. (a) In this case the boundary conditions are y(O)

= y"(O) = O and Y(7r) = y"(7r) = O. If we let

Cl = y' (O) and C2= y'" (O) then

84.:t'{y} -'- 83y(0) - 82y'(0) - 8Y(0) - y"'(O) + 4.:t'{y} =.:t' {wo/El}


and

.:t'{ } = Cl . ~
y
284+4

+ wo (~_

+ C2. _4_
484+4

8El8

From the table of transforms we get


Cl .
.
c2 .
y = "2(smxcoshx + cosxsmhx) + -(smxcoshx
Using Y(7r)

8- 1
(8-1)2+1

.
- cosxsmhx)

8+ 1
(8+1)2+1

wo

+ 4E/1-

)"
.

cosxcoshx)

= O and y"(7r) = Owe find


wo

Cl = 4El (1 + cosh 7r)csch zr,


Hence
y = 8~/1

+ cosh 7r)csch 7r(sinx coshx + cos x sinh x)

wo
- 8E/1

+ cosh 7r)csch 7r(smxcoshx

wo

- cos x sinh z ) + 4El(1

- cos xcoshx).

(b) In this case the boundary conditions are y(O) = y'(O) = O and Y(7r) = y'(7r) = O. If we let
Cl = y"(O) and C2= y"'(O) then

84 .:t'{y} - iy(O) - 82y'(0) - 8y(0) - y"'(O) + 4.:t'{y} =.:t' {(t - 7f/2)}

368

Chapter 7 Revew Exercses


and

:t{ } =
y

Cl . ~
2 54

From the table of transforms

+ C2. _4_ +
4

Y(7T)

[Sin

= O and

54

+4

Wo . _4_ e-srr/2.

4El

54

cos

. 11X )
SlD

(x - ~) cosh (x - ~) - cos (x - ~)

Y'(7T)

+4

we get

cl.
. h
+ 4"
C2( .
= "2
SlDX SlD
X
SlDX COSh X

+ 4V;l
Using

+4

sinh

(x - ~) ] q (x - ~)

= O we find
Wo sinh

El sinh

11"

Cl=---

Wo cosh

El sinh

11"

C2=----

Hence

wo

sinh!I.

y = 2El ~h
sin

2 sinxsinhx

11"

+ 4V;l [sin

Wo

cosh!I.

El ~h
4
sin

2 (sinxcoshx

11"

- cosxsinhx)

(x - ~) cosh (x - ~) - cos (x - %) sinh (x - %)] ou (x - %) .

369

Systems of Linear First-Order


Differential

____

Equations

Exercises 8.1

1. Let X = (~). Then

x' =
2. Let X

(! -5)
8

= (~). Then

x' = (:
3. Let X ~

G)

4. Let X ~

G)

-7)
o

X.

Then

x' =

4 -9)o

-1

103

o
o

x' ~ ( :
-1

x~ (~)

X.

Then
-1

5 Lec

. X.

Dx

Then
-1

x'~U -Dx+
1
1

(-t;t2)

370

U)

+ .:

Exercises 8. 1

6. Let X ~

7.

Then

dx

di = 4x + 2y + et;
dx

8. dt

9.

G)

= ~IX + 5y -

dx
di
=X

9z - 8e- t;

Y + 2z

+ e-

- 3t;

dz
= -2y + 3z + e5t - 3e-2t
dt
dy
-t
dz
t
dt = 3x - 4y + z + 2e '+ t; dt = -2x + 5y + 6z + 2e- - t

dy
dt

= 4x + y + z + 2e5t;

dx
10. -d =3x-7y+4sint+(t-4)e4t;

dy
-=x+y+8sint+(2t+1)e4t

11. Since
X' = ( -5 ) e-5t
-10

and

we see that
X'=
12. Since
I

X =

5 cos t - 5 sin t ) t
2 cos

t - 4 sin t

and

we see that
X t :=
13. Since

( 3 -4)
( 3 -4)
4 -7

4 -7

X=

XI

-10

-5t

X.

t - 5 sin t)
( -- 22 45) X = (52 cos
cos t - 4 sin t

(-2-2 45)

( -1 1/4) X = (3/2)
= (-1 1/4)
1 -1

et

X.

1 -1

we see that

(-5)

e-3t/2

-3

X,

14. Since

( -12 O1) X = ( -15)


we see that
X' = (_~

371

~) X.

et

+(

4)

-4

te'

Exercises 8. 1
15. Since

we see that

16. Since
X'=

cos,
)
~ sin t - ~ cos t
- cost - sint

we see that

and

1)
(
COS,
)
O X = ~ sin t - ~ cos t
- cos t - sin t
O -1
1

X'~ ( : 1 ~) X
-2 O -1
17. Yes, since W(XI, X2) = _2e-8t =1- Oand X and X2 are linearly independent on

-00

< t < oo.

18. Yes, since W(XI, X2) = 8e2t =1- Oand X and X2 are linearly independent on

-00

< t < oo.

= Oand XI, X2, and X3 are linearly dependent on

-00

< t < oo.

19. No, since W(XI,X2,X3)

20. Yes, since W(XI, X2, X3) = -84e-t


-00

=1- O and Xj , X2, and X3 are linearly independent on

< t < oo.

21. Since

we see that

22. Since

we see that

23. Since

372

Exercises

8. 1

we see that

24. Since

X' =
p

3 cos
O

3t)
and

-3 sin 3t

1
-4

2 3)
2 O X,

-6

1 O

( -1)
4 sin 3t
3

( 3 cos 3t )

-3sin3t

we see that

25. Let

Then

x; ~

(=D ,-" ~AX"

X3~ m'~AX3'
and W(XI,X2,X3)
-00

20

i- O so that

Xj , X2, and X3 form a fundamental set for X' = AX on

< t < oo.

26. Let
1 r ) e v'2t ,
Xl = (
-1- v2

= (

-1 +

j2

) e-v'2t

373

Exercises 8. 1
and

Then

x:2 = ( -2-}2
-}2

and W(X1, X2)


set on

-00

) e-v'2t

= AX 2,

= 2}2 i- Oso that Xp is a particular solution and X and X2 form a fundamental

< t < oo.

Exercises 8.2

l. The system is

x'=
and det(A - >"1) = (>..- 5)(>..+ 1)

( -: _~ I ~)

e ~)x

= O. For >"1 = 5 we obtain

-1 ~2 ~)

so that

K1 ~

e).

For >"2 = -1 we obtain

Then
X = ci

( 21)

2. The system is

x'=

e5t

(-1)

+ C2

e-.t

e ~)x

and det(A - >"1) = (>..- 1)(>" - 4) = O. For >"1 = 1 we obtain

e ~I~)=C~I~)

so that

Kl~(-~)

For >"2 = 4 we obtain


(

-2
1

2 ':
==> (-1
1
-1
O
O o

lO)
Io

374

so that

K2

(11).

Exercises 8.2
Then
X =

elle( -2)

+ e2

(1)1 e .
4t

3. The system is

X' - (-~~2 ~) X
and det(A - Al) = (A - l)(A

+ 3) =

O. For Al = 1 we obtain

-5 21 O):=::;. (-5 210)


( -5/2
1 1O
O O 1O
For A2

so that

Kl

so that

K2 = (21) .

(25)'

-3 we obtain

-1 2 1 O) :=::;. (-1 2 1 O)
( -5/2
5 1O
O O 1O
Then
X =

el ( 52) et + e2 (2)1 e- 3t .

4. The system is

X' =
and det(A - Al)

=
(

For A2

(A

(-5/2
3/4

+ 1)(2A + 7) = O. For Al

2)

-2

= -7/2

1
210):=::;.(-1210)
3/4 3/2 1 O
O O 1O

we obtain

Kl=(-21)'

sothat

= -1 we obtain

( -3/23/4

2 lO):=::;. (-3 410)


1O
O O 1O

K2 = (43)'

so that

-1

Then
X

ci ( - 2)1 e- 7t/2 + e2 ( 4)
3 e-.

5. The system is

X' = (10

-5) X

-12

and det(A - Al)

For A2 =

-10

= (A - 8)(A + 10) = O. For Al = 8 we obtain


-5
O):=::;. (1 -5/210)
( 82 -20
O
O O 1O

we obtain

-1/4 lO)
( 208 -5-2 1 OO):=::;. (1O O O
1

375

so that

Kl = (52)'

Exercises 8.2
Then

6. The system is

X' = (-6 2) X
-3

and det(A - Al)

= A(A + 5) = O. For Al = O we obtain

210)
( -6
-3
1 1 O

(1 -1/310)

==>

so

that

Kl =

(31).

For A2 = -5 we obtain

( -3-1 6210)
I O

==>

(1 -210)
O

so that

K2 =

(21),

Then

7. The system is

XI=(~ ~ -~)X
O 1 -1
and det(A - Al)

= (A - 1)(2- A)(A + 1) = O. For Al = 1, A2 = 2, and A3 = -1 we obtain

. ~
so that

m, m, m'
m m m,-'

x ~ '1

KF

andKF

e"+~

"+02

8. The system is

X' = (~ ~~ :) X

052
and det(A - Al) = (2 - A)(A - 5)(A - 7)

= O. For Al

376

= 2, A2 = 5, and A3 = 7 we obtain

Exercses 8.2
so that

9. We have det(A - Al)

= -(A + l)(A - 3)(A + 2) = O. For Al

= -1, A2

= 3, and A3

= -2 we obtain

so that

10. We have det(A - Al)

= -A(A - l)(A - 2) = O. For Al = O, A2 = 1, and A3 = 2 we obtain

so that

11. We have det(A - Al) = -(A

+ l)(A + 1/2)(A + 3/2)

= O. For Al

= -1, A2

= -1/2,

and A3 = -3/2

we obtain

Kl~(_D, K2~C1D,
so that
X ~ c

(-D

12. We have det(A - Al) = (A - 3)(A

' + '2

+ 5)(6

en

and

,-'/2 + '3

- A) = O. For Al

377

K3~(_D,

(-D

,-31/2

3, A2 = -5, and A3 = 6 we obtain

Exercises 8.2
so that

13. We have det(A - Al) = (A

+ 1/2)(A

- 1/2)

K1=

= O. For Al

(~)

= -1/2

K2 =

and

and A2

= 1/2 we obtain

e) ,

so that

If

X(O)
then

C1

== 2 and

C2

= (:)

= 3.

14. We have det(A - Al)

= (2 - A)(A - 3)(A + 1) = O. Por Al

cn'
m cn,-'

H), m,

KI ~

K, ~

so that

x ~CI (

-~)

,"

+02

X(O) ~
C1

15. X = ci

= -1,

C2

= 5/2, and

(~:~:~~:~)

e8.58979t

C3 =

C1

~~4~;~~:3)

e2.25684t

+ C3

(=~:~~~~~~)
e-O.046632lt

0.359815

0.615458

0.35995

0.0312209

-0.280232

0.262219

0.949058
16. X

and K3 ~

-1/2.

+ C2 (

0.239535

-0.836611
e5.05452t

= -1 we obtain

a" + C3

If

then

= 3, and A3

= 2, A2

+ C2

-0.275304

-0.162664

+ C3

0.195825

0.176045

-0.826218
-0.346439

0.0508861

0.338775

0.31957

0956lt

e-2.92362t

continued

378

Exercises 8.2
0,313235

-0,301294

0,64181

0.466599

0,31754

+C4

+ C5

e2,02882t

e -o, l55338t

0,222136

0,173787

0,0534311

-0,599108

-0,799567

17. (a)

(b) Letting

Cl

1 and

C2

= O we get x = 5e8t, y

2e8t, Eliminating the parameter we find

y = tx, x > O. When Cl = -1 and C2 = Owe find y = tx, x < O. Letting Cl = Oand C2
get x = e-lOt, y = 4e-lOt.
Eliminating the pararneter we find y = 4x, x > O. Letting
and

C2 =

-1 we find y = 4x, x

< O.

(e) The eigenvectors K = (5,2) and K2 = (1,4) are shown in the figure in part (a).

= Owe get x = -2et, y = et.


Eliminating the parameter we find y = -!x, x < O. When
C1 = -1 and C2 = O we find y = -!x, x > O. Letting C1 = O

18. In Problem 2, letting

and

C2

= 1 and

C2

= 1 we get x = t, y = t. Eliminating the parameter

we find y
x

Cl

= x, x > O. When

Cl

= Q and

C2

= -1 we find y = z ,

< O.

= O we get x = - 2e-7tj2,
y = e-7tj2. Elimnatng the parameter we find y = -~x, x < O.
When Cl = -1 and C2 = O we find y = -~x, x > O. Letting
ci = O and C2 = 1 we get x = 4e-t, y = 3e-t. Eliminating the
parameter we find y = ~x, x O, When Cl = Oand C2 = -1 we
find y = ~x, x < O.
In Problem

4,

letting

Cl

1 and

C2

379

= 1 we
Cl

=O

Exercises 8.2
19. We have det(A - ,Al)

= ,A2 = O. For ,Al = Owe obtain


K= (~).

A solution of (A - ,A1I)P

= K is
P = (~)

so that

20. We have det(A - ,Al)

= (,A + 1)2

= O. For ,Al = -1

=K

is

P=
so that
X = ci
21. We have det(A - ,Al)

we obtain

C).

K=
A solution of (A - ,A11)P

(~)J.

+C2 [C)t+

X=q (~)

C)

:'

+ C2

C~5)
[C) u:' + C~5)

e-t]

= (,A - 2)2 = O. For ,Al = 2 we obtain


K=

A solution of (A - ,A11)P

=K

is

C).

P (-1~3)
=

so that

X C)
= C1

22. We have det(A - ,Al)

= (,A - 6)2

e2t

+ C2

= O. For ,Al

[C)

=K

+(

= 6 we obtain

K=
A solution of (A - ,A11)P

te2t

(~) .

is

so that

380

-1~3)

e2t]

Exercises 8.2
23. We have det(A - Al) = (1 - A)(A - 2)2 = O. For Al = 1 we obtain

K~(J

For A2 = 2 we obtain
K, ~

and K3 ~

(J

Then

24. We have det(A - Al) = (A - 8)(A

+ 1)2

= O. For Al = 8 we obtain

K~m
For A2

-1 we obtain

K, ~ ( -:)

and

K3 ~

H)

Then

25. We have det(A - Al)

= -A(5 - A)2 = O. For Al = o we obtain

K~(=n
For A2

= 5 we obtain

K~CD
A solution of (A - All)P

= K is

5/2)

P = 1/;
381

Exercises 8.2
so that

26. We have det(A - Al) = (1 - A)(A - 2? = O. For Al = 1 we obtain

KI~m

For A2

=2

we obtain

K~(-D
A solution of (A - A2I)P = K is

(-n

p~
so that
X ~ '1

27. We have det(A - Al) = -(A

el +02 ( -

D ,"

- 1)3 = O. For Al

+03 [( - ~ )

te"

+(

-n

e"].

= 1 we obtain

K~(D
Solutions of (A - A11)P = K and (A - A11)Q = Pare

p~m Q~n)
m (D m [(D m n) "~l
and

so that
X ~ '1

28. We have det(A - Al)

+02 [

te' +

e'] +O,

(A - 4)3 = O. For Al = 4 we obtain

K~m
382

t;"

te' +

Exercises 8.2
Solutions of (A - )l}l)P = K and (A - )l}l)Q

= Pare

p~m Q~m
and

so that

X~

'1

m [m m e"] [m t;e" m m e"].


t'+

'+'2

29. We have det(A - Al)

+~

t'+

= (A - 4)2 = O. For Al = 4 we obtain


K=

C).

P=

C)

A solution of (A - All)P = K is

so that
X

= el (~) e4t + e2 [

G)

te"

C)

e4t]

If

then el

= -7

and e2 = 13.

30. We have det(A - Al) = -(A

+ l)(A -

1)2 = O. For Al = -1 we obtain

KI~cn

For A2 = 1 we obtain

KF(D andKF(D
so that

x ~ '1 ..

e-l

+ '2

If

X(O)~

m m
m

383

e' + '3

e'.

Exercises 8.2
then

el

= 2,

e2

= 3, and

e3 = 2.

= 2 is an eigenvalue of multiplicity 5. Linearly

31. In this case det(A - Al) = (2 - A)5, and Al


independent eigenvectors are

Kl=

K2=

and

K3=

32. Letting el = 1 and e2 = Owe get x = et, y = e', Eliminating

the parameter we find y = x, x

> O. When

el =

-1 and

e2 =

we find y = x, x < O.

Letting c = 1 and
When c

= -1 and

e2 = Owe

get x = e2t, y = e2t. Eliminating the parameter we find y = x, x

e2 = Owe

find y

= x,

> O.

x < O.

In Problems 33-46 the form of the answer will vary according to the choice of eigenvector. For exarnple,
in Problem 33, if Kl is chosen to be (

1 .) the solution has the form


2-2

cos t
) e4t + e? (
sin t
) e4t .
X = el (
2 cos t + sin t
- 2 sin t - cos t

33. We have det(A - Al)

= A2 -

8A

+ 17 = O. For Al = 4 + i we obtain
384

Exercises

i)

KI= ( 2 +
5 '
so that

x, = (2+5

i)e(4+i)t=

e4t+i(cost+2sint)

(2Cost-sint)
5 cos t

e4t,

5 sin t

Then

X = el (

2 cos t - sin t ) 4t
( 2 sin t + cos t)
4t
e + e2
e ,
5 cos t
5 sin t

34. We have det(A - Al) = A2 + 1 = O, For Al = i we obtain


Kl=

( -l-i)
2

so that
XI=

-1 - i ) it
e = ( sin t - eos t ). +z.' (- cos t - sin t ) ,
2
2 cos t
2 sin t

Then
X
35. We have det(A - Al) = A2

= el (

sin t - cos t )
( - cos t - sin t )
+ e2
'
2 cos t
2 sin t

8A + 17 = O, For Al = 4 + i we obtain
K =

( -l-i)
2

so that

Xl = ( -1 - i ) e (4+ ~')t = ( sin t - cos t ).. e4t + z' ( - sin t - cos t ) e4t,
2
2 cos t
2 sin t
Then
X = el (
36. We have det(A - Al) = A2

sin t - cos t) 4t
( - sin t - cos t ) 4t
e + e2
e ,
2 cos t
2 sin t

lOA + 34 = O, For Al = 5 + 3i we obtain


KI = ( 1 - 3i)
2

so that

Xl = ( 1 - 3i) e (5+3 ~')t = ( cos 3t + 3 sin 3t ) e 5t + z' (Sin 3t - 3 cos 3t ) e 5t ,


2
2 cos 3t
2 cos 3t
Then
X = el (

COs3t+3sin3t)
2 cos 3t

5t

e + e2

385

(Sin3t-3cos3t)
2 cos 3t

5t

e ,

8.2

Exercises 8.2
37. We have det(A - Al)

= A2 + 9 = O. For Al = 3i we obtain
Kl= ( 4 + 3i)
5

so that

Xl =

( 4 +5 3i)

Then

X = el
38. We have det(A - Al)

3it

(4 cos 3t - 3 sin 3t ) + . (4 sin 3t + 3 cos 3t )


2

5 cos 3t

4 cos 3t - 3 sin 3t)

+ e2

5 cos 3t

5 sin 3t

(4 sin 3t + 3 cos 3t ) .
5 sin 3t

= A2 + 2A + 5 = O. For Al = -1 + 2i we obtain
Kl =

e~

2i)

so that

=(2cos2t-2sin2tcos2t)e-t+i

Then
X

el (

2 cos 2t - 2 sin 2t ) -t
e
cos 2t

39. We have det(A - Al) = -A (A2

For A2

=i

+ 1) = O.

2 cos 2t + 2 sin 2t )
.
e-t.
sm2t

+ e2 ( 2 cos 2t + 2 sin 2t)


sin 2t

For Al = O we obtain

we obtain

so that

Then
X = el

1)
O
O

+ e2

(Sin

t)

- sin t

cost

386

+ e3

( - cos t )
c~s t .
smt

-t

Exercises
40. We have det(A - Al)

For A2

= -(A + 3)(A2

2A

+ 5) = O. For

Al

= -3 we obtain

1 + 2i we obtain

so that

X2=
Then

X=C1

( - cos 2t - 2 sin 2t)


-3cos2t
et.

2 cos 2t

O)
-2

41. We have det(A - Al)

For A2

- 2 cos 2t + sin 2t )
3sin2t
et+i

2 sin 2t

( - 2 cos 2t + sin 2t )
e-3t+c2

3sin2t

et+c3

( - cos 2t - 2 sin 2t )
-3cos2t
e',

2 cos 2t

= (1 - A)(A2

2A

2 sin 2t

+ 2) = O. For

Al

= 1 we obtain

= 1 + i we obtain

so that

Then
X = C1

O)
(
2 et +

(1

42. We have det(A - Al) = -(A - 6)(A2

C2

8A

cos t )
s~nt et
- SIn t

+ 20) =

+ C3

( sin t )
COS t
et.
cost

O. For Al = 6 we obtain

387

8.2

Exercises 8.2
For '\2 = 4 + 2i we obtain

en

K2~
so that
X2 = (-~)

e(4+2i)t

= ( Si~2t )

(-C~'2t)

t+i

2 cos 2t

e4t.

2 sm 2t

Then
X=Cl

O)1

(O

43. We have det(A -:-'\1) = (2 - ,\)(,\2

For '\2

-2

+ 3i

e6t+c2

sin 2t )
O
2 cos 2t

e4t+c3

+ 4,\ + 13) = O. For

( - cos 2t )
_o
2 sin 2t

e4t.

'\1 = 2 we obtain

we obtain
4 + 3i)

K2 =

-5
O

so that
X2=

4 -5+ 3i)

.
e(-2+3t)t=

(4 cos-5cos3t
3t - 3 sin 3t)

(4 sin-5sin3t
3t + 3 cos 3t)

e-2t+i

e-2t.

Then
X

=q

28 )
5 e 2t + C2

25

44. We have det(A - '\1)

For '\2

(4 cos 3t5 - 33sin 3t)


-

COS

(
e -2t

+ C3

4 sin 3t5 +. 33tcos 3t)


-

sm
O

= -(,\ + 2)(,\2 + 4) = O. For '\1 = -2 we obtain

= 2i we obtain

388

e.-2t

Exercises
so that
X2 =

-2-2i)
1

e2tt =

(-2Cos2t+2sin2t)
cos 2t
cos 2t

+i

(-2Cos2t-2Si112t)
sin 2t
sin 2t

Then
O)
X

C1

( - 2 cos 2t + 2 sin 2t )
cos 2t

45. We have det(A - "\1)

Por "\2

e-2t + C2

-1

+ C3

( - 2 cos 2t - 2 sin 2t )
sin 2t
.

~~

~~

= (1 - ,,\)(,,\2 + 25) = O. Por "\1

= 1 we obtain

= 5i we obtain
1 + 5i)
1
(
1

K2=
so that
X2 =

1 + 5i )
1

tt

( cos 5t - 5 sin 5t )
cos 5t

+i

( sin 5t + 5 cos 5t)


sin 5t
.
sin 5t

+ C3

( sin 5t + 5 cos 5t )
sin 5t
.
sin 5t

cos 5t

Then
X = C1

25 )
7 et

+ C2

( cos 5t - 5 sin 5t )
cos 5t
cos 5t

If

U)

X(O)~
then

C1

C2

= -1 and

c3

46. We have det(A - "\1) =

= 6.

,,\2 -

10,,\+ 29 = O. Por
K1 =

so that

Xl = ( 1 ) e (5+2i)t =
1 - 2i

"\1 =

5 + 2i we obtain

e~

2i)

cos 2t
) e5t
cos 2t + 2 sin 2t

+ z. (

sin 2t
) e5t .
sin 2t - 2 cos 2t

and
.

cos 2t
)
X = C1 (
cos 2t + 2 sin 2t

5t

+ C3 (

389

sin 2t
)
sin 2t - 2 cos 2t

5t

8.2

Exercises 8.2
If

X(O) = (-:)
then

CI =

-2 and

C2 = 5.

47.

Phase portrait for Problem 37

48. (a) Letting

Xl

= YI, X~ = Y2, X2 = Y3, and X2 = Y4 we have

The corresponding linear system is

YI = Y2

Y2

-10YI

+ 4Y3

Y3 = Y4
y~ = 4YI - 4Y3
or

o
y'=

-10
O

4 O

-4

Y.

Using a CAS, we find eigenvalues v'2i and 2y'3i with corresponding eigenvectors

:r=v'2i/4
1/2
:r=v'2i/2
1

1/2
O

390

:r=v'2/4
i

:r=v'2/2
O

Exercises

and

V3i/3

-2
::::V3i/6

-2

-V3/3
O

V3/6

8.2

Thus
-v'2/4
O
-v'2/2
O

The initial conditions


C3

-i ' and

C4

YI

(O) = O, Y2(0) = 1, Y3(0) = O, and Y4(0) = -1 imply

CI

-g,

C2

= O,

= O. Thus,

(t) = -

y;

sin ht

X2(t) = Y3(t) = -

V;

sin ht

- ~

Xl

(t) =

YI

V;

sin 2v'3t
sin2v'3t.

(b) The second-order system is

X~ = -10XI + 4X2
X~ = 4XI - 4X2
or

X" = [-10
We assume solutions of the form X
and -12,

WI

= .)-(

-2)

= v'2 and

4] X.
4 -4

= V coswt and X
W2

)-(-12)

391

V sinwt. Since the eigenvalues are -2

= 2V3.

The corresponding eigenvectors

Exercises 8.2
are
and
Then, the general solution of the system is
X = ci

cos ht

[~]

+ C2[~]

+ C3[ -~]

sin ht

cos 2v'3t

+ C4[ -~]

sin 2v'3t.

The initial conditions


and

X(O) = [~]
imply

ci =

O, C2= -h/10, C3= O, and C4= -/3/10.


Xl (t)= -

V2 sin ht + /3
10

V2.

Thus
sin 2v'3t

)3.

X2(t) = --smht

- -sm2/3t.
10

49. (a) From det(A - Al) = A(A - 2) = O we get Al = O and A2 = 2. For Al = Owe obtain
1 1
1110
2 we obtain
(

For A2 =

I O)

=:}

(1 1
0010

I O)

so that

(-: _~l~)= (-~~l~)

Then
X=ct

so that

( -1) (1)
1

+C2

1 e.2t

The line y = -x is not a trajectory of the system.


Trajectories are

x =

-Cl

+ C2e2t, Y = Cl + C2e2t or

+ 2Cl. This is a family of lines perpendicular


to the line y = -x. AH of the constant solutions

y =x

of the system do, however, lie on the line y = -x.


(b) From det(A - Al) = A2 = Owe get Al = O and
K = (-~).
A solution of (A - All)P = K is
P = (-~)

392

K 1 = (-11)

K2~ (:)

Exercises

so that

x = C! ( - ~ ) + C2 [ (

- ~ )

All trajectories are parallel to y = -x, but


y = -x is not a trajectory. There are constant
solutions of the system, however, that do lie
on the line y

= -x.

50. The system of differential equations is

and

393

t + ( - ~) ] .

8.2

Exercises 8.2
The general-solution of the system is

C2te2t

+ e e2t

C2e2t

x=

C3e2t
C5te2t

+ C4e2t

C5e2t

=e

e2t

+ C2

O
1

te2t

+ C3

e2t

O
1

+ C4

e2t

+ C5

e2t

te2t

e2t

O
1
= eKle2t

+ C2

Klte2t

e2t

O
O
O
O

+ C3K2e2t + C4K3e2t + C5

K3te2t

e2t

O
1
There are three solutions of the form X
the form X

= Kte2t + Pe2t

Ke2t,

where K is an eigenvector, and two solutions of

See (12) in the texto From (13) and (14) in the text

(A - 2I)K1 = O
and

394

Exercises

This implies
O O O

PI

O O O O O
O O O O O

P2

O O O O 1
O O O O O

P4

P5

O 1

so

P2 =

1 and

P5

= O, while

P3

Choosing

PI, P3, and P4 are arbitrary.

PI

= P3 = P4 = O we

1
O
p=

O
O
O

Therefore

a solution

is

x=

Repeating

O te2t

e2t.

for K3 we find
O
O
p=

O
O
1

so another

solution

is

x=

O te2t

+ 2y)2 = 4 sin2 2t

51. From x = 2 cos 2t - 2 sin 2t, y = - cos 2t we find x


(x

+ 2y

e2t.

= -2 sin 2t. Then

= 4(1 - cos2 2t) = 4 - 4 cos2 2t

=4-

4y2

and
or

395

X2

+ 4xy + 8y2

= 4.

have

8.2

Exercises 8.2
This is arotated conic section and, from the discriminant b2

4ac

= 16 - 32 < O, we see that the

curve is an ellipse.

if3, f3 > O. In Problem 36 the eigenvalues are 5 3i, in


Problem 37 they are 3i, and in Problem 38 they are -1 2i. Frorn Problem 47 we deduce that

52. Suppose the eigenvalues are a

the phase portrait will consist of a family of closed curves when a


origin will be a repellor when a

=O

and spirals when a =1- O. The

> O, and an attractor when a < O.

Exercises 8.3

1. From

X'-

( 3 -3) X+ (4)
2 -2

-1

we obtain

Then
and
so that
U=

q,-lF dt

and
Xp

= <llU

q,-l

-2

= ( e-t

J ( -11) = (-l1t)
= ( -11) t + (-15) .
-11 -10
=

5e-t

and

q,-l

dt

-5e-t

2. From

we obtain

Then

so that
U

q,- I F dt

(-2te-t)
2tet

and

396

3 -t
-e
2
_let

dt = (2te-t
2tet

+ 2e-t)
-

2et

Exercses 8.3
3. From

X' (3 -5) X + ( 1)
=

3/4 -1

et/2

-1

we obtain

Then

so that
U

<p-1F dt

J ( 3 -t)
~e1i

dt

and

Xp = <pU =
4.

-13/2)
( -13/4

te

t/2

+(

(3~4~t-t)

(-15/2)
/
-9 4

t/2

e.

From
X'

= (2

-1)
2

sin 2t )
2 cos2t

we obtain
-sin 2t )

Xc = el (
2 cos 2t

+ e2 (COS

2t ) 2t
e .
2 sin 2t

2t

Then
<p-

_e2t
(

sin 2t

2e2t cos2t

cos 2t)
2e2t sin 2t
e2t

and

so that
U=

<p-1Fdt =

and

Xp = <pU =

J (~
1

<p-1

_le-2t

~e-2t

cos 4t ) dt =

"2 sin 4t

e-2t

cos 2t

1
e-2t sin 2t
4

(i

sin 4t )
1
-;o; cos 4t

-k sin 2.tcos 4t - icos 2t cos 4t)


( icos 2t sin 4t - isin 2t cos 4t

5. From

X'

sin 2t

= (

O 2) X + ( -11)

-1 3

we obtain

X, = elle

( 2)

(1)

+ e2 1

Then
and

-1

<p.

397

et

e 2t .
-t

e
-2t
-e

2t

e .

cos 2t)

Exercises

8.3

so that

and

6. From

X'=(_~

~)X+C~3t)

we obtain

x, = el

et

(~)

+ e2

e)

e2t.

Then
cp -_ (2et t
e

e2t)
2t
e

and

cp-1

e-t)
~-2t

-t
(

_:-2t

so that
U=

cP -1 Fdt --

J(

2e-t - e-4t
_2e-2t + 2e-5t ) dt

( 4)

-2e

-t

1 -4t)
+ ".te

= (e -2t - se
2 -5t

and

7. From

we obtain
X,

el

1 e3t + e2

(-2)

1 e-

Then
4e3t
cP

= ( e

3t

_2e-3t)
-3t
e

so that

3t.

and

cP

-1

-3t

6"e
_ e3t

3t
(2t- e -3t - 3"2e-3t)
3t
)
dt
=
2t e 3t -:3 e3t
6te

6te-

and
Xp

= cpU =

O t + (-4/3)
-4/3 .
( -12)

8. From

398

Exercses 8.3
we obtain
X

= CI

( 4)
1

e3t

+ C2

Then
q,

= (

4e3t
e

3t

so that

and

9. From

we obtain

Then

so that

and

10. From

we obtain

Then

so that

399

(-2)
1

e- 3t .

Exercises 8.3
and

3).

xp = eu = ( -5
11. From

XI _
- (O

( sec t )
O
O X+

-1 )

we obtain
cos t )

+ e2

= el ( sin t

Xc

(sin t )
_ cos t .

Then

<P=

cost
( sin t

sin t )
-cost

<P

and

-1

sin t )
- cost

(cos t
. t
Slll

so that

and
t cos t
Xp

+ sin t In I sec ti)

= <P U = ( t sin t _ cos t In I sec ti'

12. From

( 1 -1) X+ (3)

X' -

et

we obtain

X = el (
c

-sin t ) t
e
cos t

t)
+ e2 ( cos
. te.
sin

Then

<P = (

- sin
') t
cos t

COS t

. t
sin

<P -1 = ( - sin

an

cos t

t)

.
cos
sin t

e -t

so that
U =

J <P

-1F

dt

-3 sin t
3 cos t

+ 3 cos t)
+ 3 sin t

and
Xp = <pU
13. From
I

XII

dt

( -3)

(3 cos t + 3 sin t)
3 sin t - 3 cos t

e.t

(1 -1) X + (c~St) e
sin t

we obtain

X=cr
c

-sin t ) t
e+c2.
cos t

400

cos t )
te.
sin

Exercises 8.3
Then

<I>= (

-sin

t)

cos
sin t

cos t

so that

u=

_]

<I>-] F dt =

J (~)

and

cos
Xp = <I>U = ( .
smt
14. From

XI =

<I>

and

( -

cost) e-t
sin t

dt = (~)

t)

te .

(2 -2) X+-e(1) 1
8 -6

sin
cos t

-2t

we obtain

Then
1 2t+1)

<I>= ( 2

4t

+1

-2t

_]

<I>

and

(-(4t+1)
2-1

2t

+ 2 In t )
_ In t
dt

2t+1)e2t

so that

U =

<I>-] F dt =

J(

and

2t

X
p

= <I>U = (
4t

+ In t -

2t In t)

+ 3In t - 4t In t

_ 2t

15. From

we obtain

Xc = CI ( cos t ) + C2 ( sin t )
- sin t
cos t

Then

cos t
<I>= (
- sin t

sin t )
t
cos t

so that

U =
and

Xp=<I>U=

<I>- I F dt =

and

J (-

tan2
tan t

t ) t+. (-

cos
.
( - sin
t

<I>-

t)
t)

sin
sin t tan t

401

= ( cos t - sin
sin t

dt =

t)

cos t

(t - tan t)
In I sec ti

( sin t ) ln l sec t],


cos t

Exercises 8.3
16. From

we obtain
cos t )
Xc = C1 (
.
-sm t
Then

cp=

cos
( - sin t

U=

t)

sin
cos t

+ C2 ( sin t )
cost

1 (COS t

and

cp- =

O)

dt =

- sin

t)

cost

sin t

so that

and

cp-1F dt =

J(

ese t

In I ese t - cot ti

sin t In I ese t - cot ti.)


Xp = cpU = (
.
costIn I csct - cot r]

17. From

XI =

(1-1/21 2) X+ ( sect
ese t ) e

we obtain
2 sin

Xc = cr (

t)

cost

+ C2 ( 2 cos
.
-smt

t)

e.

Then
cp= (

2 sin t
cos t

2 cos t) t
e
- sin t

Md

cP

-1

= ( ~sin t

~ cos t

cos t) -s
e
- sin t

so that
U =

cP -1 F dt =

J ~
(

J2
) dt = (
2
cos t - tan t
~ In I sin ti - In I sec ti

and

Xp = cpU = ( 33 sin t ) tet +


'2 cos t
18. From
I
X=

(COS1.

t ) In I sin ti + ( - 2 cos
. t ) In I sec r],
- '2 sin t
sm t

(1

we obtain
Xc = C1 (
Then
cp =

(COS t -

sin t
cos t

cos t - sin t )

cost

cos t + sin t )
sin t

+ C2

and

402

( cos t + sin t )
.
sin t

cP-

= (-

sin t

cost

t)

cos + sin
sin t - cost

Exercises
so that

U J cp F dt J (2 cos t + sin t - sec t)


=

-1

dt =

2 sin t - cos t

and
Xp = cpU = (

(2 sin t - cos t - In I sec t + tan ti)


- 2 cos t - sin t

3 sin t cos t - cos2 t - 2 sin2 t + (sin t - cos t) In I sec t


sin2 t - cos2 t - cos t(ln I sec t + tan ti)

+ tan ti)

19. From

we obtain

Then
1

cp =

e2t

-1 e2t

so that

and

20. From

x' =

(~

-:

1 -1

=~) X + ( ~ )
1

2et

we obtain

Then

403

-t
e-2t
-e

-e-2t

8.3

Exercises 8.3
so that
U

.
q>-lF dt =

(te-t +
-2e-t

2) dt = (-te-

- e-t
2e-t

21t e -2t

- t e -2t

+ 2t)

+ ;re
1 -2t

and
Xp=q>U=

-1/2)
-1
( -1/2

21. Frorn

,(3

X=
we obtain

q> (_e
=

-1) X +

-1

e2t)
2t
e

t
4t

(-3/4)
-1
-3/4

t+

(2)2
O

et+

(2)2
2

teto

(4e2t)
4e4t

'

and
X = q>q>-l(O)X(O)+ q>

r t q>-lFds

Jo

= q>.

(O) + (eq>.

2t
2t

+ 2t + 2t -

1)

= (~) te2t + ( -: ) e2t + ( -~) te4t + (~) e4t.


22. From

X' (1 -1) X + (1/l/t t )


=

1 -1

we obtain

q>=(11

l+t)
t

q>_l=(-t
'

l+t)
1

-1

'

and
X = q>q>-l(l)X(l)

+ q>

q>-IFds =

e (-:)
+ e C~t)

= (~)

t - (~)

23. Solving
2- A
3
1 = A2 - 1 = (A - l)(A
+ 1) = O
-1 -2 - A
we obtain eigenvalues Al = -1 and A2 = 1. Corresponding eigenvectors are
1

Thus

404

C)

lnt.

Exercises

8.3

Substituting

into the system yields

-al

from which we obtain

al =

-1 and

X(t)

2b1 = -5,

h = 3. Then

= el (-:

) e-t

+ e2 (-;)

et

+ ( -~) .

24. Solving

6- A

3- A

+ 14

= A2 - 9A

we obtain the eigenvalues Al = 2 and A2

Thus
X, = el

(A - 2)(A - 7)

=O

= 7. Corresponding eigenvectors are

( 1)
-4

e 2t

+ e2

(1)

1 e 7t .

Substituting

into the system yields

4al

+ 3bl

b2

+ 4 = O.
= 6. Substituting these two
give al = -~ and bl = ~. Then

Solving the first two equations simultaneously yields a2 = -2 and b2


values into the second pair of equations and solving for al and b:
X(t) = el ( _~)

e2t + e2 (~)

25. Solving

405

e7t + ( -:) t + (

-J) .

Exercises 8.3
we obtain the eigenvalues

)11

= 3 and

A2

7, 'Corresponding eigenvectors are

Thus

1)

x c = c1 ( -3

+ c2

e3t

(1)

9 e 7t ,

Substituting

X, =

b
( al)

into the system yields


1

from which we obtain

3a1

+ 3'b1

'9a1

+ 5b1 = -10

=3

al = 55/36 and bl = -19/4,


X(t) =

1)
( -3 e3t

el

Then

+ C2 ( 91)

e7t

( 55/36 )
-19/4
et,

26. Solving
-1-A
1

we obtain the eigenvalues Al

=A2+4=0

1- A

-1

= 2i and A2 = -2i, Corresponding eigenvectors are


K _ (
1-

1 + 2i

and

K2

= (

5 ),
1 - 2i

Thus
5 cos 2t
)
Xc = Cl (
cos 2t - 2 sin 2t

+ C2 (

5 sin 2t
)
2 cos 2t + sin 2t

Substituting

into the system yields

-al

+ 5bl + a2 + 1 =
-al

+ b: + oz = O
406

'

Exercises

from which we obtain a2


X(t)

27. Let I =

= -3,

oz

= -2/3,

al

= -1/3,

and

br = 1/3. Then

5cos2t
) + e2 (
5sin2t
) + (-3)
= el (
cos 2t - 2 sin 2t
2 cos 2t + sin 2t
-2/3

C~)

cost

+ (-1/3)

1/3

so that

and

Then
3e-12t )
-e-12t '
U = IJ> - 1Fdt

= (lOe2tSint)

30e12tsint

dt = (2e2t(2sint-cost))
26ge12t(12sint - cost)

and
Ip = IJ> U =
so that

1= (1) ee1

If 1(0) = (~) then el = 2 and e2 =

fg.

332sin t 29
( 276sin t 29
2t

+ e2

76cos t )
29
168cos t
29

3) e- + I .

( -1

12t

28. (a) Solving


1- A -1 1
1- A
1 -1

= A2-

2A = A(A- 2)

= O.

we obtain the eigenvalues Al = O and A2 = 2. Corresponding eigenvectors are

Kl = [~]
Thus

and

[ 1] [-1]

X, = el 1 + e2
If we substitute

407

1 e 2t .

sint.

8.3

Exercises

8.3

into the system, we get

[0] [1 -1]
-1

[al]
bl

[3]
-5

-br+3]
+ bl -

[al
-al

5 '

which cannot be solved for al and bi. The difference in the systems is that in Problem 23 the
homogeneous system does not have a constant solution, whereas in this problem it does.
(b) Trying a solution of the form

fails, so we instead try

Substituting into the system, we obtain


(al(-al
and note that a2
hence b2

-1.

= a2

+ bl - 5) + (-a2 + b2)t = b:

= bz. Adding the equations, we find -2 = a2 + b: = 2a2, so a2 = -1 and


From the first equation, we have al - bl + 3 = -1, and from the second

equation, we have -al


al

bl +3) + (a2 - b2)t

+ bl

and bl = 4. Thus

- 5 = -1. These are both equivalent to al - bl = -4, so we take

29. (a) The eigenvalues are O, 1,3, and 4, with corresponding eigenvectors

and

(b)

e=

( ~6
-4

2et

3e3t

et

e3t

2t

e3t

~e")

<p-l

-e
1
3

_,

1 -4t

-"3e

(e) <P-l(t)F(t) =

408

1 -t

"3e
O
2

"3e

-4t

-"3

-2e-t
2

"3e

-3t

"3
2t

e-

_\ e-3t
1 -4t

"3e

Exercises 8.3

4>(t)C + 4>(t)

J 4>-l(t)F(t)dt

409

Exercises 8.4

Exercises 8.4
1. For A

= (~ ~) we have

A2=(~ ~)(~ ~)=(~

~),

A3=AA2=(~

~)(~ ~)=(~

~),

A =AA3=C

~)(~ ~)=(~

16)'

and so on. In general


Ak

= (~

20k)

for

= 1, 2, 3,

o o o

Thus

t2

t3

2!

3!

l+t+-+-+o.o

(
and

(2t)2

(2t)3

2!

_At=(e-

2. For A

l+t+--+--+

0)
e-2t

3!

= (~ ~) we have

A2=(~ ~)(~ ~)=(~ ~)=l


A3 = AA2 = (~ ~) 1
=A
A (A2)2= 1
A = AA Al = A
= (~

410

~)

) (e
=

Exercises 8.4
and so on. In general
k

{A,
1,

k
k

= 1, 3, 5,
= 2, 4, 6,

o o o
o

o o o

Thus
e

At

A
A2
= 1+ -t
+ -t
1!

2!

= 1+ At

A3 3
+ -t
+ooo
3!

+ 2.It2 + 2.At3 + o o o
2!

3!

= 1 (1 + 2.
t2 + 2.t4 + o o o) + A (t + 2.t3 + 2.t5 + o o o)
21
4!
3!
51
cosh t

sinh t )

= Icosht + Asinht = (
sinh t cosh t
and
e-At = (COSh(-t)
sinh(-t)

sinh( -t))
cosh(-t)

= (

cosh t
-sinht

- sinh t)

cosht

3. For

A=( ~ ~ ~)
-2

-2

-2

A2 = (~
-2

-2

we have

Thus, A 3 = A 4 = A 5

eAt = 1 + At =

o o o

~) (~

-2

-2

-2

~ ~)

O O O

O and

O o1

O) + (
O

O O 1

A~U

O
O
1

') ('+1

tt
t
t
-2t -2t -2t

4. For

we have

~) = (~

-2

n
411

t
-2t

t
t+1
-2t -2t + 1

Exercises 8.4

Thus, A 4

= AS = A 6 = ... = O and
eAt

= 1 + At + -21A 2t2

e 0) e

= o

o o

3t
5t

o
o o
t o

0) +

o
o o o =
~t2 o o

0) ( 1
3t

~t2 + 5t

5. Using the result of Problem 1,

x= ( eot

G~)=

e~t )

ci (~)

+ C2

(~

) .

6. Using the result of Problem 2,

x = (COSht
sinh t

t)

sinh
cosh t

(q)

t)

= c (COSh

+ C2 (sinh

sinh t

C2

t) .

cosh t

7. Using the result of Problem 3,

X =

t +t
-2t

t)
t +t 1
t
-2t -2t + 1

(Cl)
C2

= C!

( t +t 1) +
-2t

C3

8. Using the result of Problem 4,


10
X =

3t
( ~t2 + 5t

9. To solve

412

C2

(t)t

+ 1 + C3 (t)
-2t

-2t

+1

Exercises 8.4

we identify to

= 0, F(s) = (_~), and use the results of Problem 1 and equation (5) in the texto
X(t) = eAtC+eAt

{t e-ASF(s)ds

Jto

) (e- e-2s0) ( 3) ds
{t

= (~

e2t Jo

{t ( 3e-s

e2t Jo

_e-2s

-1

ds

10. To solve

we identify to

= 0, F(s) = (e~t)'

X(t)

= eAtC + eAt

and use the results of Problem 1 and equation (5) in the texto
(t e-ASF(s) ds

i;

413

Exercises 8.4
11. To solve

= O, F( s) = (~),

we identify to

= (COSht

sinh t

(Cl
Cl

(Cl
Cl

sinh t

+ (COSht

(Cl)
C2

cosh t + C2 sinh t)
sinh t + C2 cosh t

= (Cl cosh t
Cl

sinh t)
cosh t

and use the results of Problem 2 and equation (5) in the texto

sinh t)
cosh t

sinh t

+ (COSht

sinh t)
cosh t

sinh t

+ C2 sinh t) + (COSht
+ C2 cosh t
sinh t

cosh t + C2 sinh t)
sinh t + C2 cosh t

cosh S

(t (

Jo

cosh s
sinh s

-'- sinh s

- sinh s) (1) ds
cosh s
1

s)

- sinh
coshs

sinh t) ( sinh s - cosh s )


cosh t
- cosh s + sinh s

+ (COSht

ds

t
O

sinh t) ( sinh t - cosh t )


cosh t
- cosh t + sinh t

sinh t
2

+ C2 sinh t) + (Sinh t Cl sinh t + C2 cosh t


sinh2 t -

= (Cl cosh t

(t (

Jo

cosh t) = ci (COSht)
cosh2 t
sinh t

+ C2 (Sinh t) _

(1)

cosh t

12. To solve
X' =

we identify to

= O, F(s)

(C~Sh t
sinh t
Cl
( Cl

sinh t)
cosh t

sinh t

1) X
O

+ (COSht )
sinh t

t)

COSh
, and use the results of Problem 2 and equation (5) in the texto
smht

(Cl)
C2

cosh t + C2 sinh t )
sinh t + C2 cosh t

= (Cl cosh t
Cl

=. (

(O1

+ (C~Sh t
sinh t

sinh t)
cosh t

(t (

Jo

cosh s
sinh s

( cosh t sinh t) (t (1 ) ds
sinh t cosh t Jo O

+.

+ C2 sinh t) + (COSht
+ C2 cosh t
sinh t

(Os)

cSlonshhtt)

414

Ito

S)

- sinh
coshs

S)

(C~Sh
sinh s

ds

Exercses 8.4

Cl cosh
= ( Cl sinh

t + C2sinh t )
t + C2cosh t +

Cl cosh
= (
Cl sinh

t + C2sinh t ) + ( t cosh t )
t + C2cosh t
t sinh t

(COSh t
sinh t

t) (t)
t
O

sinh
cosh

= Cl

(COSh t )
sinh t

+ C2 ( sinh t ) + t ( cosh t) .
cosh

13. We have

Thus, the solution

of the initial-value

problem is

X=(t:1)_4(t~1)+6(
-2t
-2t

~ ).
-2t + 1

14. We have

X(O) = C3 (~)

+ C4 (~) + (

~3)=

e:~)

= (~) .

Thus, C3 = 7 and C4 = ~, so

x = 7 ( ~ ) et + ~ (~)
S-

15. From sI - A =

4'

e2t

+ ( -~ ) .

-3 ) we find

s+4

3/2
(sI-A)-l

1/2

(S-2 - s+2
-1
1
--+-s-2
s+2

and

The general solution of the system is then

415

3/4 _ 3/4 )

s-2

-1/2

s-2

s+2

3/2

s+2

sinh

Exercises 8.4

16. From sI _ A

3/2)

e (
-1

(S - 4
.-1

2t

+ e

(-1/2)

-2t

+ C2

(3/4)
/

-1 2

(sI-A)-l

(S~3 ~ S~2
s-3

and
eAt

s-2

__2 +_2 )
s-3
s-2

-1

--+-s-3
s-2

2e3t _ e2t
- ( e3t _ e2t
_

The general solution of the system is then

= (e - C2) (~)

(S -1- 5

+ C2

2)
we find
s- 1

17. From sI _ A

2t

C3 (~)

e3t + (-e

e3t + C4

e)

e2t.

9 ) we find
8+1

(sI - A)-l =

_1
s-2
(

3
(s-2)2
1
(s - 2)2

416

+ 2C2)

e)

e2t

(-3/4)
/

32

-2t

Exercises 8.4
and

The general solution of the system is then

18. Frorn sI - A

= (~ s

q (

1+
t

3t) e

2t

+C2

(-9t)
1 - 3t

e 2t .

~12)

we find

(sI - A)-

and

At

(e-

s+l+l
(s+1)2+1
=
-2
(

e-

cos t + t sin t
-2e-t sint

(S+1)2+1)
s+l+l
(s+1)2+1
:' sin t
)
:' cos t - :' sin t .

The general solution of the system is then

X=e

At

C=

(e-

Cl

e-

cos t + t sin t
-2e-t sin t

cos t + sin t) -t
e
- 2 sin t

:' sin t
) (CC21)
:' cos t - e-t sin t

+ C2 (

sin t
cos t - sin t

417

)-t

Exercises 8.4
19. Solving
2-'\
-3

we find eigenvalues '\1

= 3 and

1
6 -,\
'\2

=,\2 _ 8,\ + 15= (,\ - 3)('\ - 5) = O

= 5.

Corresponding eigenvectors are

Then

1 1)
P = (1 3 '

-1

(3/2
-1/2

-1/2)
1/2 '

and

D = (~ ~),

so that

PDP-l = (

2 1).

-3 6
20. Solving

we find eigenvalues '\1

2-'\
1

= 1 and

1
2 -,\
'\2

= 3.

=,\2 _ 4A+ 3 = (,\ - 1)('\ - 3) = O


Corresponding eigenvectors are

Then

P= (-1 1)

p_l=(-1/21/2)

1 1 '

1/2

1/2 '

and

= (~ ~)

so that

21. From equation (3) in the text


etA

= etPDP- = 1 + t(PDP-l) + 2_t2(PDP-l)2 + 2_t3(PDP-l)3 + ...


2!

=P

3!

[1+ tD + i(tD)2 + i(tD)3 + ...] p-l = PetDp-l.


2.

3.

418

Exercises

22. From equation (3) in the text


O

etD =

f) +

(!

O
A2
O

O)
An

e
+2" O
1 2

O
A~

1
3!

+ -t

1 + A2t

+ ~(A2t)2 + ...
O

O
23. From Problems 19, 21, and 22, and equation (1) in the text
X = etAC = PetDp-1C

24. From Problems 20-22 and equation (1) in the text


X

= etAC = PetDp-1C

419

JJ

(,3

01 A~
:

jJ+

8.4

Exercses 8.4
25. If det(sI - A) = O, then s is an eigenvalue of A. Thus sI - A has an inverse if s is not an eigenvalue
of A. For the purposes of the discussion in this section, we take s to be larger than the largest
eigenvalue of A. Under this condition sI - A has an inverse.
26. Since A 3

= 0, A is nilpotent. Since
e At = 1 + At

+ A 2 -t2 + ... + A t -tk + ...

if A is nilpotent and A m = 0, then A k =


eAt = 1 + At
In this problem A 3

k!

2!

for k ::::m and

t2

+ A 2 - + ... + A m-l
2!

tm-l
(m - 1)!

= 0, so
2

eAt = 1+ At + A t; = [~ ~ ~] +
O O 1

t - t2/2
-t
[
-t - t2/2
1-

and the solution of X'

[=~ ~ ~] t + [- ~ ~ O1]2!_
-1

1 1

t + t2/2

~ 1+

-1

O 1

.; t2/2

= AX is
Cl]
C2

X(t) = eAtC = eAt


[

C3

[Cl(1-t-t2/2)+C2t+C3(t+t2/2)]
-Clt + C2 + C3t
Cl(-t_t2/2)+C2t+C3(1+t+t2/2)

27. (a) The following commands can be used in Mathematica:


A={ {4, 2},{3, 3}};

e={ el, e2};


m=MatrixExp[A

t];

soleeExpandjm.c]
Colleet[sol, {el, e2}] / / MatrixForm
The output gives

y(t) =

ci ( _~et

+ ~e6t) + C2 Get + ~e6t)

The eigenvalues are 1 and 6 with corresponding eigenvectors

420

Chapter 8 Review Exercises


so the solution of the system is

X(t) = bl (-~)

et

+ b:

C) e

6t

or
x(t)

= -2blet + b2e6t

y(t) = 3blet
If we replace bl with -iCl

+ iC2

+ b2t.

and b2 with tCl

+ ~C2, we obtain

the solution found using the

matrix exponential.
(b ) x(t) = cle-2tcost-

(Cl +c2)e-2tsint

y(t) = C2e-2t cos t

+ (2Cl + c2)e-2t

sin t

+ C3(-6e-2t + 6e-t)
= c2(4e-2t - 3e-t) + c4(4e-2t - 4e-t)
= CI (e-2t - e-t) + C3(_2e-2t + 3e-t)

28. x(t) = cI(3e-2t - 2e-t)


y(t)
z(t)

w(t) = C2(-3e-2t

+ 3e-t) + C4(_3e-2t + 4e-t)

Chapter 8 Review Exercises


1. If X

=k

( : ), then X'

We see that k
2. Solving for

CI

= O and

= ~.
and

C2

we find

CI = - ~

and

C2 =

i.

3. Since

( -1; -4~ -3~)( -1~)=(1:)=4(~)


-4 -1'

we see that A = 4 is an eigenvalue with eigenvector K3. The corresponding solution is X = K3t.
4. The other eigenvalue is A2

= 1 - 2i with corresponding eigenvector K2

solution is
2t ) t
( sin 2t)
t
X(t) = CI ( cos
.
e + C2
e.
- sin 2t
cos 2t

421

= (_~).

The general

Chapter 8 Review Exercises

5. We have det(A - Al) = (A - 1)2 = O and K = ( _~). A .solution to (A - AI)P = K is P =

e)

so that

6. We have det(A - Al)

= (A + 6)(A + 2)
X

7. We have det(A - Al) = A2

Xl =

= ci

= Oso that

e-6t +e2

(_~)

C)

e-2t.

+ 5 = O. For A = 1 + 2i we obtain

2A

( 1) e
i

(1+2i)t =

(cos 2t) e t
- sin 2t

K = (~) and

+ 2. (sin 2t) e t .
cos 2t

Then

8. We have det(A - Al) = A2

Xl = (

32

= el ( cos 2t)
- sin 2t

2A

+ 2 = O. For A = 1 +

i) e

(l+i)t

sin 2t) t
e.
- cos 2t

+ e? (

we obtain k, = (3 ~

= (3 cos t + sin t) e t + 2. (
2 cos t

i)

and

+ 3 sin t) e.t
2 sin t

- cos t

Then
X
9. We have det(A - Al)

= el (

-(A - 2)(A - 4)(A

10. We have det(A - Al) = -(8


and A2

3 cos t + sin t ) t
e
2 cos t

+ 2)(82 -

+ e2

( -

cos t + 3 sin t ) t
e .
2 sin t

+ 3) = Oso that

28 + 3) = O. The eigenvalues are Al = -2, A2 = 1 + V2i,

= 1 - V2i, with eigenvectors

422

Chapter 8 Review Exercises


Thus

x ~"

en [m ';'v2t - ( ~/2)
~+o, m sinv2t]
e-"

H2

v2t+

H3 [ (

=C1

- 7)
5 e-2t+c2

(COS
...;2t)
-...;2sin...;2tj2

v2t] e'

sin

e'

(Sin
...;2t )
...;2cos...;2tj2 et.

et+c3

cos ...;2t

11. We have

x c = c1

(1)

sin ...;2t

e2t + c2

(4)
1

e4t.

Then

cp = (e2t
O
and

U=

cp-1 F dt =

cp-1 =

4e4t)
e4t '

e-2t
(

_4e-2t)

<t '

2e-2t _ 64te-2t )
( 15e-2t + 32te-2t )
dt =
16te-4t
_e-4t - 4te-4t
'

so that
11 + 16t)
Xp = cpU = (
.
-1 - 4t
12. We have

Xc =
Then

cp=
and

2 cos t
( - sin t

-1

cp F dt

so that
Xp

Cl

t)

2 cos
( - sin t

t)

2 sin
cos t

+ C2 (

cp-1

'

(cost-sect)
sin t

dt =

2 sin
cost

t) e .
t

t - sin t)

(~C~S
~ sm t

(Sint-Inlsect+tantl)
- cos t

- 2 cos In I sect + tan

ti)

= cpU = (
.
- 1 + sin t In I sec t + tan ti
423

e-t,

cost

'

Chapter 8 Review Exercises


13. We have
X = Cl (

cos t + sin t )
2 cos t

+ C2 ( sin t -

cos t )
2 sin t

Then
q,

= (cos t + sin t sin t ~ cos t) , q, -1 =


2 cos t

2 sin t

and

q,-lFdt=

U=

J
= (-~ cos t

~ cos t -

sin t'

- cos t

!cos t - ~ sin t )
!cos t + ~sin t

'

1 sin t - 1 cos t + 1 ese t )


2
2
2
dt
(
J - ~ sin t - ~ cos t + ~ese t

i sin t + ~In I ese t -

i sin t + !In I ese t -

cot ti)
cot ti

'

so that

Xp = q,U = ( -1)
-1

+.(

sin t
) In I ese t - cot ti
sin t + cos t

14. We have

Then
q,-l=

t -2t -e -2t)
-e

-e
t -2t

e-u

e-u

'

and

t-1)
( -1

U=Jq,-lFdt=J

dt=

(lt2-t)
2_t

'

so that
X =q,U=
p

(-1/2)t2e2t+
1/2

(-2)te2t.
1

15. (a) Letting

we note that (A - 2I)K = O implies that 3kl


k2

= 0, k3 = 1 and then

k2

+ 3k2 + 3k3

= 1, k3 = we get

424

= 0, so kl

-(k2

+ k3)'

Choosing

Chapter 8 Review Exercises


respectively. Thus,

are two solutions.


(b) Frorn det(A - Al) = A2(3 - A)

= O we see that

Al

= 3,

and O is an eigenvalue of multiplicity

two. Letting
K = (~~) ,
k3
as in part (a), we note that (A - OI)K = AK
-(k2

+ k3).

Choosing k2

O implies that kl + k2 + k3 = O, so kl

= O, k3 = 1, and then k2 = 1, k3 = O we get

respectively. Since the eigenvector corresponding to Al

the general solution of the system is

16. For X = (~~)et we have X'

= X = IX.

425

= 3 is

Chapter 8 Related Exercises

Chapter .8 Related Exercises

1. (a) In operator notation we have

Xl

(t)

CI COS

- 2X2

=O

2Xl - (D2

+ 2)X2

= O.

+ 6D2 + 4)XI =

Eliminating X2 we have (D4


)3 - v5i and )3

(D2 +4)Xl

+ v5i,

O. The roots of the auxiliary equation are

so

)3 - v5 t +

C2

sin )3 - v5 t +

+ v5 t +

C3 cos )3

C4

sin )3 + v5 t.

Then
1

+ 4)Xl

X2(t) = 2(D2

(t)

= 1 + v5CI COS )3 - v5 t + 1 + v5C2 sin )3 - v5 t


2

1-v5

+ -2-C3
The frequencies are

WI =

/
r:
cos V 3 + V 5 t

)3 - v5 and

W2 =

)3

1-v5
2

C4 sin

/
~
V 3 - v 5 t.

+ v5.

(b) We write the system in the form


X" =

[-4 2] X
2 -2

and assume solutions of the form


X = Vcoswt
Since the eigenvalues are -3

+ v5

and

X = Vsinwt

and -3 - v5,

WI

)3 - v5 and

W2

corresponding eigenvectors are

VI = [(1 + ~)/2]

and

Then, the general solution of the system is


X(t)

=CI

[(1+~)/2]

[(1+~)/2]

cosV3-v5t+C2

2. If K is multiplied by a factor of 10, then

426

sinV3-v5t

)3

+ v5.

The

Chapter 8 Related Exercises

-10

A = M-1K=

-10

A
ol
T

-10

-10

-10

-10
2

-10

-10

The table below shows the eigenvalues


periods

Ai

-10

with the corresponding frequencies

-10

Wi =

r-xi

and

Ti = 21r/ Wi.

-13 .8117 -13.2639 -12.4064 -11.3175 -10.0974


3.7164
1.6907

3.6420
1.7252

3.5223
1.7838

3.1776
1.9773

3.3641
1.8677

-8.8604

-7.7256

-6.8082

-6.2089

-1.5000

2.9766
2.1108

2.7795
2.2605

2.6093
2.4080

2.4918
2.5216

1.2247
5.1302

We see that the frequencies are higher and consequently, the periods are smaller.
3. (a) For this exercise we assume a building of four stories, each with mass 10,000 kg and restoring
force constants each k, = 5,000 kg/s2 Then

1 O
M = 10,000

1 O

-10,000

5,000 -10,000

K=

O O 1 O
O O O 1

5,000

5,000

5,000 -10,000

5,000

5,000 -5,000

and
-1

A=

0.5 O
0.5 -1 0.5
O 0.5 -1
O

0.5

O
O
0.5
-0.5

(b) The table below shows the eigenvalues Ai with the corresponding frequencies Wi =
periods

Ti = 21r/ wl .
A
Ol

T,

-1.7660

-1.1737

-0.5000

-0.0603

1.3289
4.7280

1.0834
5.7998

0.7071
8.8858

0.2456
25.5855

427

r-xi

and

Chapter

8 Related Exercises

Since the periods are all larger than 4, the building should be safe in a period-2 earthquake.
(e) If k is multiplied by 10, then
-100
50

A=

50
-100

50

50

-100

50

50

-50

From the table below we see that the periods are generaUy in the range from 1.5 to 3 and the
building is unsafe from a period-2 earthquake.
A.
w
T

-17.660

-11.737

-5.000

-0.603

4.2024
1.4951

3.4259
1.8340

2.2361
2.8099

0.7766
8.0909

Experimenting, we see that if the building is .made stiffer by a factor of 8, the periods are
entering the danger zone.
A.
w
T

-14.128

-9.389

-4.000

-0.482

3.7588
1.67l6

3.0642
2.0505

2.0000
3.1416

0.6946
9.0459

4. The related homogeneous system is X" = M-1KX = AX, where A is shown in the solution of
Problem 3(a). We assume solutions of the form X
are -1. 7660, -l.1737,

= V cosw and X

= V sinwt. The eigenvalues

-0.5, and -0.0603 with eorresponding frequencies 1.3289, 1.0834, 0.7071,

and 0.2456. The eigenvectors are


-0.4285

0.6565

-0.6565

0.2280

0.5774

0.5774

0.4285

0.5774

0.5774
0.2280

V2=

-0.5774

0.2280
and

-0.4285
V4 =

and the general solution is


X( t) = C1V 1 cos l.3289t

+ C7V4

+ C2V 1 sin 1.3289t + C3V 2 cos l.0834t

cosO.2456t

+ ca V4

sin 0.2456t.

Since
10000cos 3t
O

F(t) =

O
O

428

-0.5774
0.6565

Chapter

8 Related Exercises

to solve the nonhomogeneous system we look for a particular solution of the form
bl

al
a2

Xp =

cos3t

a3
a4

Substituting into X" = AX

+ M-IF(t)

-9al = 1 - al

b2
b3
b4

sin3t.

we obtain

+ 0.5a2
and

Solving these systems gives al

-0.12549,

a2 =

0.00787,

a3 =

0.00049,

a4

0.00003, and

= b: = b3 = o = O. Thus,

bl

-0.12549
X p -

0.00787
cos3t.
-0.00049
0.00003

= O and X' (O) = O, we find Cl = -0.0592, C2 = O, C3 = -0.0839,


= 0.0679, C6 = O, C7 = 0.0255, and C8 = O. Thus, the solution of the initial-value problem

Applying the initial conditions X(O)


C4

= O,

Cs

is
-0.4285
X(t)

-0.0592

-0.6565
0.5774

0.6565
cos 1.3289t - 0.0839

0.2280
0.5774
0.4285

0.2280

0.2280

-0.5774

+ 0.0679

0.5774
O

cos 1.0834t

cos 0.0679t

+ 0.0255

-0.4285
-0.5774
0.6565

0.5774

429

-0.12549
cos 0.2456t

0.00787
-0.00049
0.00003

cos 3t.

Numerical Solutions of
Ordinary Differential Equations
Exercises 9.1

AH tables in this chapter were constructed in a spreadsheet program which does not support subscripts.
Consequently,
1.

2.

h -

x(n)
1.00
1.10
1.20
1.30
1.40
1.50

h -

x(n)
0.00
0.10
0.20
0.30
0.40
0.50

3.

x(n)
0.00
0.10
0.20
0.30
0.40
0.50

Xn

and

0.1
y(n)
5.0000
3.9900
3.2546
2.7236
2.3451
2.0801

0.1
y(n)
2.0000
1.6600
1.4172
1.2541
1.1564
1.1122

0.1
y(n)
0.0000
0.1005
0.2030
0.3098
0.4234
0.5470

Yn

will be indicated as x(n) and y(n), respectively.


h = 0.05
x(n)
y(n)
1.00
5.0000
4.4475
1.05
3.9763
1.10
3.5751
1.15
3.2342
1.20
2.9425
1.25
2.7009
1.30
2.4952
1.35
2.3226
1.40
1.45
2.1786
1.50
2.0592

x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

0.05
y(n)
2.0000
1.8150
1.6571
1.5237
1.4124
1.3212
1.2482
1.1916
1.1499
1.1217
1.1056

h = 0.05
x(n)
y(n)
0.00
0.0000
0.05
0.0501
0.10
0.1004
0.15
0.1512
0.20
0.2028
0.25
0.2554
0.30
0.3095
0.35
0.3652
0.40
0.4230
0.45
0.4832
0.50
0.5465

430

Exercses 9. 1
4.

x(n}
0.00
0.10
0.20
0.30
0.40
0.50

5.

0.1
y(n}
0.0000
0.0952
0.1822
0.2622
0.3363
0.4053

0.1
yen}
0.0000
0.0050
0.0200
0.0451
0.0805
0.1266

0.1
yen}
0.5000
0.5215
0.5362
0.5449
0.5490
0.5503

x(n}
0.00
0.10
0.20
0.30
0.40
0.50

6.

x(n}
0.00
0.10
0.20
0.30
0.40
0.50

7.

0.1
y(n}
1.0000
1.1110
1.2515
1.4361
1.6880
2.0488

x(n}
0.00
0.10
0.20
0.30
0.40
0.50

h -

x(n}
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

- 0.05

x(n}
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

y(n}
0.0000
0.0488
0.0953
0.1397
0.1823
0.2231
0.2623
0.3001
0.3364
0.3715
0.4054

0.05
y(n}
0.0000
0.0013
0.0050
0.0113
0.0200
0.0313
0.0451
0.0615
0.0805
0.1022
0.1266

0.05
yen}
0.5000
0.5116
0.5214
0.5294
0.5359
0.5408
0.5444
0.5469
0.5484
0.5492
0.5495

x(n}
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

x(n}
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

0.05
yen}
1.0000
1.0526
1.1113
1.1775
1.2526
1.3388
1.4387
1.5556
1.6939
1.8598
2.0619

431

Exercises 9. 1
8.

9.

10.

h = 0.1
x(n)
y(n)
0.00
1.0000
0.10
1.1079
0.20
1.2337
0.30
1.3806
0.40
1.5529
0.50
1.7557

x(n)
1.00
1.10
1.20
1.30
1.40
1.50

0.1
yen)
1.0000
1.0095
1.0404
1.0967
1.1866
1.3260

h x(n)
0.00
0.10
0.20
0.30
0.40
0.50

0.1
y(n}
0.5000
0.5250
0.5498
0.5744
0.5986
0.6224

11.

h
x(n)
0.00
0.10
0.20
0.30
0.40
0.50

h = 0.05

x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

x(n)
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1. 35
1.40
1. 45
1. 50

h =

x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

y(n)
1.0000
1.0519
1.1079
1.1684
1.2337
1.3043
1.3807
1.4634
1.5530
1.6503
1.7560

0.05
y(n)
1.0000
1.0024
1.0100
1.0228
1.0414
1.0663
1.0984
1. 1389
1.1895
1.2526
1.3315

0.05
y(n}
0.5000
0.5125
0.5250
0.5374
0.5498
0.5622
0.5744
0.5866
0.5987
0.6106
0.6224

0.1
y(n}
2.0000
2.1220
2.3049
2.5858
3.0378
3.8254

exact
2.0000
2.1230
2.3085
2.5958
3.0650
3.9082

x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

h - 0.05
y(n}
2.0000
2.0553
2.1228
2.2056
2.3075
2.4342
2.5931
2.7953
3.0574
3.4057
3.8840

432

exact
2.0000
2.0554
2.1230
2.2061
2.3085
2.4358
2.5958
2.7997
3.0650
3.4189
3.9082

Exercises
12. (a)

(b)
h=O.l

20

EULER

x{nl

1.00
1.10
1.20
1. 30
1.40

15
10
5

9. 1

IMPROVED
EULER

.v.uu.

_y (nl_

1.0000
1.2000
1.4938
1.9711
2.9060

1.0000
1.2469
1.6668
2.6427
8.7988

--~~====;=~~---x
1.1 1.2 1.3 1.4
13. (a) Using the Euler method we obtain y(O.l)

~ Yl = 1.2.

(b) Using y" = 4e2x we see that the local truncation error is

y"(c) h2 = 4e2c (0.1)2 = 0.02e2c.


2

Since e2x is an increasing function, e2c :::;e2(0.1) = eO.2 for O :::;e :::;0.1. Thus an upper bound
for the local truncation error is 0.02eO.2 = 0.0244.
(e) Since y(O.l)

= eO.2 = 1.2214, the actual error is y(O.l) - Yl =

(d) Using the Euler method with h = 0.05 we obtain y(O.l)

~ Y2

0.0214, which is less than 0.0244.

1.21.

(e) The error in (d) is 1.2214 - 1.21 = 0.0114. With global truncation error O(h), when the step
size is halved we expect the error for h

= 0.05 to be one-half the error when h =

0.1. Comparing

0.0114 with 0.214 we see that this is the case.

14. (a) Using the improved Euler method we obtain y(O.l) ~ Yl


(b) Using y/"

1.22.

= 8e2x we see that the local truncation error is


y"/(c) h3
6

= 8e2c

(0.1)3

= 0.001333e2c.

Since e2x is an increasing function, e2c :::;e2(0.1)= eO.2 for O :::;e:::; 0.1. Thus an upper bound
for the local truncation error is 0.001333eo.2 = 0.001628.
(e) Since y(O.l)

= eO.2 =

l.221403,

the actual error is y(O.l)

- Yl

0.001403 which is less than

0.001628.

(d) Using the improved Euler method with h


(e) The error in (d) is 1.221403 -1.221025

0.05 we obtain y(O.l)

~ Y2 = 1.221025.

= 0.000378. With global truncation error O(h2),

when

the step size is halved we expect the error for h = 0.05 to be one-fourth the error for h = 0.1.
Comparing 0.000378 with 0.001403 we see that this is the case.
15. (a) Using the Euler method we obtain y(O.l) ~ Yl

0.8.

(b) Using y" = 5e-2x we see that the local truncation error is
5e-~c (0.1)2
2

= 0.025e-2c.

433

Exercises 9. 1
Since e-2x is a decreasing function, e-2c :S ea = 1 for O-:S e :S 0.1. Thus an upper bound for
the local truncation error is 0.025(1)
(e) Since y(O.l)

= 0.025.

= 0.8234, the actual error is y(O.l) - Yl = 0.0234, which is less than 0.025.

(d) Using the Euler method with h = 0.05 we obtain y(O.l) ~ Y2 = 0.8125.
(e) The error in (d) is 0.8234 - 0.8125 = 0.0109. With global truncation error O(h), when the
step size is halved we expect the error for h

= 0.05 to be one-half the error when h. = 0.1.

Comparing 0.0109 with 0.0234 we see that this is the case.


16. (a) Using the improved Euler method we obtain y(O.l) ~ Yl
(b) Using

y'" =

= 0.825.

-10e-2x we see that the local truncation error is


lOe-2c (0.1)3 = 0.001667e-2c.

6
Since e-2x is a decreasing function, e-2c :S ea

= 1 for O :S e :S 0.1. Thus an upper bound for

the local truncation error is 0.001667(1) = 0.001667.


(e) Since y(O.l)

= 0.823413, the actual error is y(O.l) -Yl

= 0.001587, which is less than 0.001667.

= 0.05 we obtain y(O.l) ~ Y2 = 0.82378l.


(e) The error in (d) is 10.823413- 0.82371811 = 0.000305. With global truncation error O(h2),
when the step size is halved we expect the error for h = 0.05 to be one-fourth the error when
h = 0.1. Comparing 0.000305 with 0.001587 we see that this is the case.

(d) Using the improved Euler method with h

17. (a) Using y" = 38e-3(x-l)

we see that the local truncation error is


yi/(c) h2 = 38e-3(c-l) h2 = 19h2e-3(c-l).
2
2

is a decreasing function for 1 :S x :S 1.5, e-3(c-l) :S e-3(1-1)

(b) Since e-3(x-l)

1 for 1 :S c:S 1.5

and
y"(c) ~2 :S 19(0.1)2(1) = 0.19.
(e) Using the Euler method with h

= 0.1 we obtain y(1.5) ~ 1.8207. With h = 0.05

W(!

obtain

y(1.5) ~ 1.9424.
(d) Since y(1.5)

2.0532, the error for h = 0.1 is Ea.l = 0.2325, while the error for h

0.05 is

Ea.a5 = 0.1109. With global truncation error O(h) we expect Ea. 1/ Ea.a5 ~ 2. We actually have
Ea. 1/ Ea.a5 = 2.10.
18. (a) Using

y'"

= _114e-3(x-l)

we see that the local truncation error is

434

Exercses 9. 1
(b) Since e-3(x-l)

is a decreasing function for 1 ::; x ::;1.5, e-3(c-l)

::; e-3(1-1) =

1 for 1 ::; e ::; 1.5

and
lylll(C) ~3 ::; 19(0.1)3(1) = 0.019.
(e) Using the improved Euler method with h

= 0.1 we obtain y(1.5) ~ 2.080108. With h

= 0.05

we obtain y(1.5) ~ 2.059166.


(d) Since y(1.5)

= 2.053216, the

error for h

= 0.1 is EO.l = 0.026892, while the

error for h = 0.05 is

Eo.os = 0.005950. With global truncation error O(h2) we expect Eo.1/ Eo.os ~ 4. We actuaJly
have Eo.1/ Eo.os = 4.52.
19. ()a

11
1
Usmg y = - (x + 1)2 we see that the local truncation error is
lyll(C) ~21 =

(e:

1)2 ~2.

(b) Since (x ~ 1)2 is a decreasing function for O::; x::; 0.5, (e: 1)2 ::; (O: 1)2 = 1 for O::; e::; 0.5
and
yll(C) ~21 ::; (1) (0.;)2 = 0.005.
(e) Using the Euler method with h = 0.1 we obtain y(0.5) ~ 0.4198. With h
y(0.5) ~ 0.4124.

= 0.05 we obtain

= 0.4055, the error for h = 0.1 is EO.l = 0.0143, while the error for h = 0.05 is
Eo.os = 0.0069. With global truncation error O(h) we expect Eo.1/ Eo.os ~ 2. We actuaJly have
Eo.1/ Eo.os = 2.06.
.
111
2
20. ( a ) U sing y = (x + 1)3 we see that the local truncaton error is
(d ) Since y(0.5)

111

y (e)

h3

6" =

1
h3
(e + 1)3 3"

(b) Since (x ~ 1)3 is adecreasing function for O::; x::; 0.5, (e: 1)3 ::; (O: 1)3 = 1 for O::; e::; 0.5
and
y"I(C) ~3 ::; (1) (0.;)3
(e) Using the improved Euler method with h

= 0.000333.

= 0.1 we obtain y(0.5) ~ 0.405281. With h = 0.05

we obtain y(0.5) ~ 0.405419.


(d) Since y(0.5)

= 0.405465, the

error for h = 0.1 is EO.l

= 0.000184, while the

error for h

= 0.05 is

Eo.os = 0.000046. With global truncation error O(h2) we expect Eo.1/ EO.05 ~ 4. We actually
have Eo.1/ EO.05 = 3.98.

435

Exercises 9. 1
21. Because Y~+l depends on Yn and is used to determine Yn+l, all of the Y~ cannot be computed at one
time independently of the corresponding 'Yn values. For example, the computation of Y4 involves
the value of Y3.

Exercises 9.2
1.

h = 0.1
y(n}
2.0000
2.1230
2.3085
2.5958
3.0649
3.9078

x(nJ
0.00
0.10
0.20
0.30
0.40
0.50

exact
2.0000
2.1230
2.3085
2.5958
3.0650
3.9082

2. Setting ex = 1/4 we find b

= 2,

= -1 and 3 = 1/4. The resulting second-order Runge-Kutta

method is
Yn+ 1 = Yn - k1
k1

= hf(xn, Yn)

k2 = hf(xn
h = 0.1
x(n)
0.00
0.10
0.20
0.30
0.40
0.50

R-K
y(n)
2.0000
2.1205
2.3006
2.5759
3.0152
3.7693

+ 2k2

+ h/4,

Yn

+ kj4).

IMP EULER
y(n)
2.0000
2.1220
2.3049
2.5858
3.0378
3.8254

3.

x(n)
1. 00
1.10
1.20
1.30
1.40
1.50

v(n)
5.0000
3.9724
3.2284
2.6945
2.3163
2.0533

4.

x(n)
0.00
o . 10
0.20
0.30
0.40
0.50

v(n)
2.0000
1.6562
1.4110
1.2465
1.1480
1.1037

5.

x(n)
0.00
0.10
0.20
0.30
0.40
0.50

v(n)
0.0000
0.1003
0.2027
0.3093
0.4228
0.5463

6.

x(n)
0.00
0.10
0.20
0.30
0.40
0.50

v(n)
1.0000
1.1115
1.2530
1.4397
1.6961
2.0670

7.

x(nJ
0.00
0.10
0.20
0.30
0.40
0.50

v(n)
0.0000
0.0953
0.1823
0.2624
0.3365
0.4055

8.

x(n)
0.00
0.10
0.20
0.30
0.40
0.50

v tn)
0.0000
0.0050
0.0200
0.0451
0.0805
0.1266

436

Exercises
9.

x(n)
0.00
0.10
0.20
0.30
0.40
0.50

v(n)
0.5000
0.5213
0.5358
0.5443
0.5482
0.5493

ll.

x(n)
l. 00
1.10
1.20
1.30
1.40
1.50

1.0000
1.0101
1.0417
1.0989
1.1905
1.3333

(n)

10.

x(n)
0.00
0.10
0.20
0.30
0.40
0.50

y(n)
1.0000
1.1079
1.2337
1.3807
1.5531
1.7561

12.

x(n)
0.00
0.10
0.20
0.30
0.40
0.50

(n)
0.5000
0.5250
0.5498
0.5744
0.5987
0.6225

13. (a) Write the equation in the form


dv
dt
(b)

t n)

0.0
1.0
2.0
3.0
4.0
5.0

= 32 - 0.025v = f(t, v).

v(t)

v(n)
0.0000
25.2570
32.9390
34.9772
35.5503
35.7128

40
30
20

~0---1--~2--~3---4--~5t

(e) Separating variables and using partial fractions we have

--

1 (1

2V32 V32 -

JO.025 v

1)

V32 + JO.025

dv

= dt

and
2

V32~ 0.025 (ln 1V32 + JO.025


.

vi - ln

1V32 -

Since v(O) = O we find e = O. Solving for v we obtain


v(t) =

16v5 (ev'TIt - 1)
ev'TIt + 1

and v(5) ~ 35.7678.

437

JO.025 vi) = t

+ c.

9.2

Exercises

9.2

14. (a) See the table in part (e) of this problem.

(b) From the graph we estimate A(1) ~ 1.68, A(2) ~ 13.2, A(3) ~ 36.8, A(4) ~ 46.9, and
A(5) ~ 48.9.
A(t)

50
40
30
20
10

(e) Let a = 2.128 and (3 = 0.0432. Separating variables we obtain


dA
A(a - (3A)

1(1

-a -A + a

(3)
- (3A

1
-[In A -ln(a
a

dA

- (3A)]

A
In a _ (3A

= dt

= dt
= t +e
= a(t

+ e)

= eo(t+c)

a - (3A
A

= aeo(t+c)

- (3Aeo(t+c)

Thus
aeo(t+c)
a
a
- ....,--------;
- 1+ (3eo(t+c) - (3 + e-o(t+c) - (3 + e-oce-ot

A(t) -

From A(O) = 0.24 we obtain


a
0.24 = (3 + e-OC
so that

e-oc =

ajO.24 - (3 ~ 8.8235 and

438

Exercises 9.2

A( ) ~

2.128

t ~ 0.0432 + 8.8235e-2 128t .

t
A
A
A

(da s)
(observed)
(approximated)
(exact)

15. (a)

h - 0.05
x(n)

y(n)

1.00
1.05
1.10
1.15
1.20
1.25
1. 30
1.35
1.40

1.0000
1.1112
1.2511
1.4348
1.6934
2.1047
2.9560
7.8981
1.06E+15

1
2.78
1.93
1.95

2
13.53
12.50
12.64

3
36.30
36.46
36.63

4
47.50
47.23
47.32

5
49.40
49.00
49.02

(b)

h - 0.1

v(n)

1.0000

20

1.2511

15

1.6934

10

2.9425

--~========~------x

903.0282

1.1

16. (a) Using the fourth-order Runge-Kutta method we obtain y(O.l)

(b) Using

y(5)

(x)

;::::Yl

1.2

1.3

1.4

1.2214.

= 32e2x we see that the local truncation error is


y(5)(c)'!!_

120

= 32e2c (0.1)5
120

= 0.000002667e2c.

Since e2x is an increasing function, e2c :S e2(O.1) = eO.2 for O :S e :S 0.1. Thus an upper bound
for the local truncation error is 0.000002667eo.2
(e) Since y(O.l)

= eO.2

= 0.000003257.

= 1.221402758, the actual error is y(O.l) - tn

= 0.000002758 which is less

than 0.000003257.
(d) Using the Iourth-order Runge-Kutta

formula with h = 0.05 we obtain y(O.l)

;::::Y2 =

1.221402571.
(e) The error in (d) is 1.221402758 - 1.221402571 = 0.000000187. With global truncation error
O(h4), when the step size is halved we expect the error for h = 0.05 to be one-sixteenth the

= 0.1. Comparing 0.000000187 with 0.000002758 we see that this is the case.
(a) Using the fourth-order Runge-Kutta method we obtain y(O.l) ;::::Yl = 0.823416667.
(b) Using y(5) (x) = -40e-2x we see that the local truncation error is
error for h

17.

40e-2c (0.1)5 = 0.000003333.


120
Since e-2x is a decreasing function, e-2c :S

eO

the local truncation error is 0.000003333(1)

= 0.000003333.

439

= 1 for O :S e

:S 0.1. Thus an upper bound for

Exercises

9.2

(e) Since y(O.l) = 0.823413441, the actual error is ly(O.l) - Yll = 0.000003225, which is less than
0.000003333.
(d) Using the fourth-order Runge-Kutta method with h

= 0.05 we obtain y(O.l)

~ Y2

0.823413627.
(e) The error in (d) is 10.823413441- 0.8234136271 = 0.000000185. With global truncation error
O(h4),

when the step size is halved we expect the error for h

error when h

= 0.1. Comparing 0.000000185 with 0.000003225 we see that this is the case.

18. (a) Using y(5) = -1026e-3(x-l)

we see that the local truncation error is

l
(b) Since e-3(x-l)

= 0.05 to be one-sixteenth the

y(5)(C).!i!_1 = 8.55h5e-3(c-l).
120

is a decreasing function for 1 :S x:s 1.5, e-3(c-l) :S e-3(1-1)

= 1 for 1 :S c:S 1.5

and
y(5) (c) 1~50:S 8.55(0.1)5 (1) = 0.0000855.
(e) Using the fourth-order Runge-Kutta method with h = 0.1 we obtain y(1.5) ~ 2.053338827.
With h = 0.05 we obtain y(1.5) ~ 2.053222989.
.
()
24
19. (a) Usmg y 5 =
we see that the local truncation error is
(x + 1)5
(5)

y
(b) Since (x

+ 1)5 is a decreasing

c .!i!_ _
1
h5
() 120 - (c + 1)5 5

1
1
function for O :S x :S 0.5, (c + 1)5 :S (O + 1)5 = 1 for O :S c:S 0,5

and
y(5)(C) ~5 :S (1) (0.;)5

= 0.000002.

(e) Using the fourth-order Runge-Kutta method with h = 0.1 we obtain y(0.5) ~ 0.405465168.
With h = 0.05 we obtain y(0.5) ~ 0.405465111.
20. Each step of the Euler method requires only 1 function evaluation, while each step of the improved
Euler method requires 2 function evaluations - once at (xn, Yn) and again at (Xn+l, y~+l)'

The

second-order Runge-Kutta methods require 2 function evaluations per step, while the fourth-order
Runge-Kutta method requires 4 function evaluations per step. To compare the methods we approximate the solution of y' = (x

+y

- 1)2, y(O) = 2, at x = 0.2 using h = 0.1 for the Runge-Kutta

method, h = 0.05 for the improved Euler method, and h = 0.025 for the Euler method. For each
method a total of 8 function evaluations is required. By comparing with the exact solution we see
that the fourth-order Runge-Kutta method appears to still give the most accurate resulto

440

Exercises

EULER
IMP EUL
R-K
0.025 h = 0.05 h = 0.1
y(n)
y_(n)
y(n)
2.0000
2.0000
2.0250
2.0526
2.0553
2.0830
2.1165
2.1228
2.1230
2.1535
2.1943
2.2056
2.2395
2.2895
2.3075
2.3085

9.2

h
x(n)
0.000
0.025
0.050
0.075
0.100
0.125
0.150
0.175
0.200

exact
2.0000
2.0263
2.0554
2.0875
2.1230
2.1624
2.2061
2.2546
2.3085
y

21. (a) Por y'

+y =

10 sin 3x an integrating factor is eX so that

d
dx [eXy] = 10ex sin 3x ~

eXy

~
When x
3 cos 3x

= O, y

= O, so O = -3

+ 3e-x.

= eX sin 3x

+e

- 3ex cos 3x

y = sin 3x - 3 cos 3x

+ c and c =

+ ce-x.

3. The solution is y = sin 3x -

Using Newton's method we find that x = 1.53235 is the only

positive root in [0,2].


(b) Using the fourth-order Runge-Kutta method with h

= 0.1 we obtain the table ofvalues shown.

These values are used to obtain an interpolating function in Mathematica.


interpolating function is shown. Using Mathematica's
only positive root in [O,2Jis x
x(n)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

y(n)
0.0000
0.1440
0.5448
1.1409
1.8559
2.6049
3.3019
3.8675
4.2356
4.3593
4.2147

x(n)
1.0
1.1
1.2
1.3
1.4
1.5
1'.6
1.7
1.8
1.9
2.0

root finding capability we see that the

= 1.53236.
y

y(n)
4.2147
3.8033
3.1513
2.3076
1.3390
0.3243
-0.6530
-1.5117
-2.1809
-2.6061
-2.7539

441

The graph of the

Exercises 9.3

Exercises 9.3

1. For y' - y = x - 1 an integrating factor is e- J da: = e-x, so that

~[e-Xyl

dx

= (x

- l)e-X

and
y

=e

X(

-xe-x

+ e) =

-x

+ ce".

From y(O) = 1 we find e = 1 and y = -x + e". Comparing exact values with approximations
obtained in Example 1, we find y(0.2)

::::::1.02140276 compared to Yl = 1.02140000, y(O.4) ::::::

1.09182470 compared to Y2 = 1.09181796, y(0.6) ::::::1.22211880 compared to Y3 = 1.22210646, and


y(0.8) ::::::1.42554093 compared to Y4 = 1.42552788.

2. 100 REM ADAMS-BASHFORTHj ADAMS-MOULTON


110 REM METHOD TO SOLVE Y'=FNF(X,Y)
120 REM DEFINE FNF(X,Y) HERE
130 REM GET INPUTS
140 PRINT
150 INPUT "STEP SIZE=", H
160 INPUT "NUMBER OF STEPS (AT LEAST 4)=",N
170 IF N <4 GOTO 160
180 INPUT "XO =" ,X
190 INPUT "YO =" ,Y
200 PRINT
210 REM SET UP TABLE
220 PRINT "X", "Y"
230 PRINT
240 REM COMPUTE 3 ITERATES USING RUNGE-KUTTA
250 DIM Z(4)
260 Z(l)=Y
270 FOR I=1 TO 3
280 Kl=H*FNF(X,Y)
290 K2=H*FNF(X+Hj2,Y
300 K3=H*FNF(X+Hj2,Y

+K1j2)
+K2j2)

310 K4=H*FNF(X+H,Y+K3)
320 Y=Y+(Kl+2*K2+2*K3+K4)j6
330 Z(I+l)+ Y

442

Exercises

9.3

340 X=X+H
350 PRINT X,Y
360 NEXT 1
370 REM COMPUTE REMA1N1NG X AND Y VALUES
380 FOR 1=4 TO N
390 YP= Y+ H*(55*FNF(X,Z( 4))-59*FNF(X-H,Z(3) )+37*FNF(X-2*H,Z(2))
-9*FNF(X-3*H,Z(1)) )/24
400 Y= Y+ H* (9*FNF(X + H,YP)+ 19*FNF(X,Z( 4))-5*FNF(X-H,Z(3))+ FNF(X-2*H,Z(2))) /24
410 X=X+H
420 PRINT X,Y
430 Z(1)=Z(2)
440 Z(2)=Z(3)
450 Z(3)=Z(4)
460 Z(4)=Y
470 NEXT 1
480 END
3.

x (n)
0.00
0.20
0.40
0.60
0.80

4.

x iri)
0.00
0.20
0.40
0.60
0.80

y(n)
1.0000
0.7328
0.6461
0.6585
0.7332
0.7232

y(n)
2.0000
1.4112
1.1483
1.1039
1.2109
1.2049

initia1 condition
Runge-Kutta
Runge-Kutta
Runge-Kutta

predictor
corrector

initia1 condition
Runge-Kutta
Runge-Kutta
Runge-Kutta

predictor
corrector

443

Exercises
5.

x(n)
0.00
0.20
0.40
0.60
0.80
1.00

9.3

y(n)
0.0000
0.2027
0.4228
0.6841
1.0234
1.0297
1.5376
1.5569

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector
predictor
corrector

x(n)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00

6.

x(n)
0.00
0.20
0.40
0.60
0.80
1.00

y(n)
1.0000
1.4414
1.9719
2.6028
3.3483
3.3486
4.2276
4.2280

x (n)

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector
predictor
corrector

0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00

444

'y(n)
0.0000
0.1003
0.2027
0.3093
0.4227
0.4228
0.5462
0.5463
0.6840
0.6842
0.8420
0.8423
1.0292
1.0297
1.2592
1.2603
1.5555
1.5576

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector

y(n)

1.0000
1.2102
1.4414
1.6949
1.9719
1.9719
2.2740
2.2740
2.6028
2.6028
2.9603
2.9603
3.3486
3.3486
3.7703
3.7703
4.2280
4.2280

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector

Exercises 9.3
7.

x(n)
0.00
0.20
0.40
0.60
0.80
1. 00

y(n)
0.0000
0.0026
0.0201
0.0630
0.1362
0.1360
0.2379
0.2385

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector
predictor
corrector

x(n)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1. 00

8.

x tn)
0.00
0.20
0.40
0.60
0.80
1.00

y(n)
1.0000
1.2337
1.5531
1.9961
2.6180
2.6214
3.5151
3.5208

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector
predictor
corrector

x(n)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00

445

y(n)
0.0000
0.0003
0.0026
0.0087
0.0201
0.0200
0.0379
0.0379
0.0630
0.0629
0.0956
0.0956
0.1359
0.1360
0.1837
0.1837
0.2384
0.2384

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector

v tn)
1.0000
1.1079
1.2337
1.3807
1.5530
1.5531
1.7560
1.7561
1.9960
1.9961
2.2811
2.2812
2.6211
2.6213
3.0289
3.0291
3.5203
3.5207

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector

Exercises 9.4

Exercises 9.4
1. The substitution

y'

= u leads to the
Yn+l = Yn

The initial conditions are Yo

= -2

_
iteration formulas

+ hun,

Un+l = Un

+ h(4un

and Uo = 1. Then

u. = Yo + O.luo

= -2

+ 0.1(1) =

-1.9

Ul = Uo + 0.1(4uo - 4yo) = 1 + 0.1(4


Y2 = u. + O.lUl = -1.9

+ 0.1(2.2)

The general solution of the differential equation is y


we find C1

= -2 and

C2 = 5.

- 4Yn).

= -1.68.

= C1e2x + C2xe2x. From the initial conditions

= _2e2x + 5xe2x

Thus y

+ 8) = 2.2

and y(0.2) ;:::;1.4918.

2. The substitution y' = u leads to the iteration formulas

+ hun,

Yn+l = Yn
The initial conditions are yO

=4

Un+

1 = un + h (~ Un - :2 Yn) .

and Uo = 9. Then

Yl

= yo + O.luo

= 4 + 0.1(9)

Ul

= Uo + 0.1 iUo - iYO = 9 + 0.1[2(9) - 2(4)]

(2

Y2 = Yl

+ O.lUl

= 4.9

2)

= 4.9

+ 0.1(10) = 5.9.

The general solution of the Cauchy-Euler differential equation is y


conditions we find C1 = -1 and
3. The substitution

C2 = 5.

= 10

Thus y

= -x + 5x2 and y(1.2)

C1X

+ c2x2.

From the initial

= 6.

y' = u leads to the system


y' = u,

u'

= 4u - 4y.

Using formula (4) in the text with x corresponding to t, y corresponding to x, and u corresponding
to y, we obtain
RUllfLe-Kuttamethod with h=0.2

ml

m2

m3

m4

kl

k2

k3

k4

0.2000

0.4400

0.5280

0.9072

2.4000

3.2800

3.5360

4.8064

ml

m2

m3

m4

kl

k2

k3

k4

0.1000
0.2443

0.1600
0.3298

0.1710
0.3444

0.2452
0.4487

1.2000
1.7099

1.4200
2.0031

1.4520
2.0446

1.7124
2.3900

Runge-Kutta

446

0.00 -2.0000
0.20 -1.4928

1.0000
4.4731

method with h=O.l


x

0.00 -2.0000 1.0000


0.10 -1.8321 2.4427
0.20 -1.4919 4.4753

Exercises
4. The substitution y'

9.4

= u leads to the system


,2
2
u = -u - -y.

y' = u,

x2

Using formula (4) in the text with x corresponding to t, y corresponding to x, and u corresponding
to

y,

we obtain
Runge-Kutta

method with h=0.2

ml

m2

m3

m4

kl

k2

k3

k4

1.8000

2.0000

2.0017

2.1973

2.0000

2.0165

1.9865

1.9950

1.00
1.20

4.0000
6.0001

9.0000
11.0002

Runge-Kutta

method with h=O.l

ml

m2

m3

m4

kl

k2

k3

k4

0.9000
1.0000

0.9500
1.0500

0.9501
1.0501

0.9998
1.0998

1.0000
1.0000

1.0023
1.0019

0.9979
0.9983

0.9996
0.9997

1.00
1.10
1.20

4.0000
4.9500
6.0000

9.0000
10.0000
11.0000

5. The substitution y' = u leads to the system


u'

y' = u,
Using formula (4) in the text with

= 2u - 2y + etcost.

corresponding to x and u corresponding to


Runge-Kutta

ml
0.4000

m2
0.4600

m3
0.4660

m4
0.5320

kl
0.6000

k2
0.6599

k3

k4

0.6599

ml
0.2000
0.2315

m2
0.2150
0.2480

m3
0.2157
0.2487

m4
0.2315
0.2659

kl

k2

0.3000
0.3299

0.3150
0.3446

kl

k2

k3
0.3150
0.3444

6.
ml

10.0000
8.7500
10.1563
13.2617
17.9712

m2

m3

0.0000 12.5000
-2.500013.4375
-4.375017.0703
-6.3672 22.9443
-8.8867 31.3507

m4

0.0000
-20.0000
-28.7500 -5.0000
-40.0000 -8.7500
-55.1758 -12.7344
-75.9326 -17.7734

5.0000
4.3750
5.0781
6.6309
8.9856

447

k4

k3

0.3298
0.3587

we obtain

method with h=0.2


t

0.00
O .7170 0.20

Runge-Kutta

y,

1.0000
1.4640

2.0000
2.6594

method with h=O.l


t

0.00 1.0000
0.10 1.2155
0.20 1.4640

u
2.0000
2.3150
2.6594

Runge-Kutta method with h=O.l


i2
k4
t
il

0.00 0.0000 0.0000


22.5000
-5.0000
0.10 2.5000 3.7500
29.6875
0.20
-10.6250
2.8125 5.7813
-16.0156 40.3516 0.30 2.0703 7.4023
-22.5488 55.3076 0.40 0.6104 9.1919
-31.2024 75.9821 0.50 -1.5619 11.4877

Exercises 9.4

i1,i2
7

i1

i2
2
1

As t -;

00

5t

we see that il(t) -; 6.75 and i2(t) -; 3.4.

7.

Runge-Kutta

method with h=0.2

ml

m2

m3

m4

kl

k2

k3

k4

2.0000

2.2800

2.3160

2.6408

1.2000

1.4000

1.4280

1.6632

0.00
0.20

6.0000
8.3055

2.0000
3.4199

ml

m2

m3

m4

kl

k2

k3

k4

1.0000
1.1494

1.0700
1.2289

1.0745
1.2340

1.1496
1.3193

0.6000
0.7073

0.6500
0.7648

0.6535
0.7688

0.7075
0.8307

Runge-Kutta

x,y

0.511.52t

448

method with h=O.l

0.00 6.0000
0.10 7.0731
0.20 8.3055

y
2.0000
2.6524
3.4199

Exercises

8.

Runge-Kutta
ml

m2

m3

m4

0.6000

0.9400

1.1060

1.7788

kl

k2

k3

k4

method with h=0.2


t

0.00 1.0000
1.40002.06002.39403.72120.202.07853.3382
Runqe-Kutta

ml

m2

m3

m4

kl

k2

k3

k4

0.3000
0.5193

0.3850
0.6582

0.4058
0.6925

0.5219
0.8828

0.7000
1.1291

0.8650
1.4024

0.9068
1.4711

9.4

1.0000

method with h=O.l


t

0.00 1.0000
1.1343 0.10 1.4006
1.84740.202.0845

1.0000
1.8963
3.3502

x,y
50

40 y(t)

30

9.

Runge-Kutta
m1

m2

m3

m4

kl

k2

k3

k4

method with h=0.2


t

0.00 -3.0000
-1.0000 -0.9200 -0.9080 -0.8176 -0.6000 -0.7200 -0.7120 -0.8216 0.20 -3.9123
Runge-Kutta
m1

m2

m3

m4

kl

k2

k3

k4

5.0000
4.2857

method with h=O.l


t

0.00 -3.0000
-0.5000 -0.4800 -0.4785 -0.4571 -0.3000 -0.3300 -0.3290 -0.3579 0.10 -3.4790
-0.4571 -0.4342 -0.4328 -0.4086 -0.3579 -0.3858 -0.3846 -0.4112 0.20 -3.9123

449

5.0000
4.6707
4.2857

Exercises

9.4

x,y

10.

Runqe-Kutta
ml

m2

m3

m4

kl

k2

k3

k4

0.6400

1.2760

1.7028

3.3558

1.3200

1.7720

2.1620

3.5794

Runge-Kutta
ml

m2

m3

m4

kl

k2

k3

k4

0.3200
0.7736

0.4790
1.0862

0.5324
1.1929

0.7816
1.6862

0.6600
1.0117

0.7730
1.2682

0.8218
1.3692

1.0195
1.7996

x,y
2

1.5

0.1

0.2

0.3t

11. Solving for x' and y' we obtain the system

+ y + 5t

x'

y'

= 2x + y - 2t.

-2x

450

method with h=0.2


t

0.00 0.5000
0.20 2.1589

y
0.2000
2.3279

method with h=O.l


t

0.00 0.5000
0.10 1.0207
0.20 2.1904

y
0.2000
1.0115
2.3592

Exercses

Runge-Kutta
mI

m2

m3

m4

kl

-0.8000 -0.5400 -0.6120 -0.3888

0.0000

k2

k3

k4

mI

m2

m3

kl

m4

method with h=0.2


t

0.00
-0.2000 -0.1680 -0.3584 0.20
Runge-Kutta
k2

k3

k4

9.4

1.0000
0.4179

-2.0000
-2.1824

method with h=O.l


t

0.00
-0.4000 -0.3350 -0.3440 -0.2858 0.0000 -0.0500 -0.0460 -0.0934 0.10
-0.2866 -0.2376 -0.2447 -0.2010 -0.0929 -0.1362 -0.1335 -0.1752 0.20

1.0000
0.6594
0.4173

-2.0000
-2.0476
-2.1821

x,y

12. Solving for

X'

4 t

and y' we obtain the system


X

= -y - 3t
2

+ 2t -

Runge-Kutta
mI

m2

m3

m4

-1.1000 -1.0110 -1.0115 -0.9349

kl

k2

k3

k4

1.1000

1.0910

1.0915

1.0949

Runge-Kutta
mI

m2

m3

m4

-0.5500 -0.5270 -0.5271 -0.5056


-0.5056 -0.4857 -0.4857 -0.4673

method with h=0.2


t

0.00
0.20

3.0000
1.9867

-1.0000
0.0933

method with h=O.l

kl

k2

k3

k4

0.5500
0.5456

0.5470
0.5457

0.5471
0.5457

0.5456
0.5473

0.00
0.10
0.20

3.0000
2.4727
1.9867

-1.0000
-0.4527
0.0933

451

Exercises 9.4

x,y

60
40

y(t)

20
4
-20

8 t

x(t)

-40
-60

Exercises 9.5
1. We identify P(x)

= o, Q(x) = 9, f(x)

= O, and h

(2 - 0)/4

0.5. Then the finite difference

equation is
Yi+l

+ 0.25Yi + Yi-l = O.

The solution of the corresponding linear system gives


0.0
4.0000

0.5
-5.6774

2. We identify P(x)

1.0
-2.5807

1.5
6.3226

2.0
1.0000

= O, Q(x) = -1, f(x) = x2, and h = (1 -

0)/4

= 0.25. Then the finite difference

equation is

+ Yi-l

Yi+l - 2.0625Yi

= 0.0625x;'

The solution of the corresponding linear system gives


0.00
0.0000

0.25
-0.0172

3. We identify P(x)

0.50
-0.0316

2, Q(x)

0.75
-0.0324

1. 00
0.0000

= 1, f(x) = 5x, and h = (1 -

0)/5

equation is
1.2Yi+l - 1.96Yi

+ 0.8Yi-l = 0.04(5xi).

The solution of the corresponding linear system gives


y

0.0
0.0000

0.2
-0.2259

0.4
-0.3356

0.6
-0.3308

0.8
-0.2167

1.0
0.0000

452

0.2. Then the finite difference

Exercises

9.5

4. We identify P(x) = -10, Q(x) = 25, f(x) = 1, and h = (1 - 0)/5 = 0.2. Then the finite difference
equation is
-Yi

+ 2Yi-l = 0.04.

The solution of the corresponding linear system gives


0.0
1.0000

0.2
1.9600

0.4
3.8800

0.6
7.7200

0.8
15.4000

1.0
0.0000

5. We identify P(x) = -4, Q(x) = 4, f(x) = (1 + x)e2x, and h = (1 - 0)/6 = 0.1667. Then the finite
difference equation is
0.6667Yi+l - 1.8889Yi+ 1.3333Yi_l = 0.2778(1

+ xi)e2xi.

The solution of the corresponding linear system gives


x

0.0000
3.0000

0.1667
3.3751

0.3333
3.6306

0.5000
3.6448

6. We identify P(x) = 5, Q(x) = O, f(x)

0.6667
3.2355

0.8333
2.1411

1.0000
0.0000

= 4y1x, and h = (2 - 1)/6

= 0.1667. Then the finite

difference equation is
1.4167Yi+l - 2Yi

+ 0.5833Yi_l = 0.2778(4.JXi).

The solution of the corresponding linear system gives


x

1.0000
1.0000

1.1667 1.3333 1.5000 1.6667 1.8333 2.0000


-0.5918 -1.1626 -1.3070 -1.2704 -1.1541 -1.0000

7. We identify P(x) = 3/x, Q(x) = 3/x2, f(x)

= O, and

h = (2 - 1)/8 = 0.125. Then the finite

difference equation is
0.1875)
(
0.0469)
(
0.1875)
Yi+l+ -2+--2- Yi+ 1---.- Yi-l=O.
( 1+--.-X,
xi
x,
The solution of the corresponding linear system gives
1.000
5.0000

1.125
3.8842

1.250
2.9640

8. We identify P(x) = -l/x,

1.375
2.2064

1.500
1.5826

1.625
1.0681

Q(x) = x-2, f(x) = lnx/x2,

1.750
0.6430

1.875
0.2913

2.000
0.0000

and h = (2 - 1)/8 = 0.125. Then the finite

difference equation is
0.0625)
(
- Yi+l + -2
( 1 - --o
xt

0.0156)
+ --2-

xi

(
0.0625)
Yi + 1 + --o - Yi-l

xt

0.0156lnxi.

The solution of the corresponding linear system gives


1.000
0.0000

1.125
1.250
1.375
1.500
1.625
1.750
1.875
2.000
-0.1988 -0.4168 -0.6510 -0.8992 -1.1594 -1.4304 -1.7109 -2.0000

453

Exercises 9.5
9. We identify P(x)

= 1-x,

Q(x)

= x, f(x)

= (1-0)/10

= x, and h

= 0.1. Then the finite difference

equation is
[1 + 0.05(1 - Xi)]Yi+l

+ [-2 + O.OlXi]Yi + [1 -

0.05(1 - Xi)]Yi-l

= O.Olxi.

The solution of the corresponding linear system gives


0.0
0.0000

0.1
0.2660

0.2
0.5097

0.3
0.7357

0.4
0.9471

I
10. We identify P(x)

0.5
1.1465

0.6
1.3353

0.7
1.5149

0.8
1.6855

0.9
1.8474

1.O
2.0000

= x, Q(x) = 1, f(x) = x, and h = (1 - 0)/10 = 0.1. Then the finite difference

equation is

The solution of the corresponding linear system gives


0.0
1.0000

0.1
0.8929

11. We identify P(x)


equation is

0.2
0.7789

0.3
0.6615

= O, Q(x) = -4, f(x)

0.4
0.5440

0.5
0.4296

0.6
0.3216

0.7
: 0.2225

0.8
0.1347

= O, and h = (1 - 0)/8

Yi+l -

2.0625Yi

0.9
0.0601

1.O
O .0000 :

= 0.125. Then the finite difference

+ Yi-l = O.

The solution of the corresponding linear system gives


x
y

0.000
0.0000

0.125
0.3492

12. We identify P(T)

0.250
0.7202

0.375
1.1363

2/T, Q(T) = O, f(T)

0.500
1.6233

0.625
2.2118

= O, and h

0.750
2.9386

= (4 - 1)/6

0.875
3.8490

1.000
5.0000

= 0.5. Then the finite difference

equation is
0.5)
( 1 + G Ui+l - 2Ui

+ (0.5)
1- G

Ui-l

= O.

The solution of the corresponding linear system gives


u

1.0
2.0
3.0
3.5
4.0
1.5
2.5
50.0000 72.2222 83.3333 90.0000 94.4444 97.6190 100.0000

13. (a) The difference equation

(1 +

%Pi)

Yi+l

+ (-2 + h2Qi)Yi +

(1 -

%Pi)

Yi-l

= h2fi

is the same as the one derived on page 434 in the texto The equations are the same because
the derivation was based only on the differential equation, not the boundary conditions. If we

454

Chapter 9 Review Exercises


allow i to range from
Yn-l

o to

n - 1 we obtain n equations in the n + 1 unknowns Y-l, YO, Yl, ... ,

Since Yn is one of the given boundary conditions, it is not an unknown.

(b) Identifying Yo = y(O), Y-l

= y(O -

h), and Yl

= y(O + h) we have from (5) in the tcxt

21h[Yl - Y-d = y'(O) = 1 or

Yl - Y-l = 2h.

The difference equation corresponding to i = O,

(1+

~PO) Yl

+ (-2 + h2Qo)YO +

(1-

~PO) Y-l = h2fo

becomes, with Y-l = Yl - 2h,

(1 +

~Po) Yl

+ (-2 + h2Qo)YO +

(1 - ~po)

(Yl - 2h)

h2 fo

or
2m + (-2

+ h2Qo)yo

= h2 fo

Alternatively, we may simply add the equation Yl -Y-l


obtaining n + 1 equations in the
(e) Using n

ti

+ 2h -

Po.

= 2h to the list of n difference equations

+ 1 unknowns Y-l, YO, Yl, ... , Yn-l

= 5 we obtain

0.0
0.2
0.4
0.6
0.8
1.0
-2.2755 -2.0755 -1.8589 -1.6126 -1.3275 -1.0000

14. Using h = 0.1 and, after shooting a few times, y'(O) = 0.43535 we obtain the following table with
the fourth-order Runge-Kutta method.
0.0
1.00000

0.1
1.04561

0.2
1.09492

0.3
1.14714

0.4
1.20131

0.5
1.25633

0.6
1.31096

0.7
1.36392

0.8
1.41388

0.9
1.45962

Chapter 9 Review Exercises


1. ~
x(n)
1. 00
1.10
1.20
1.30
1.40
1.50

EULER
2.0000
2.1386
2.3097
2.5136
2.7504
3.0201

IMPROVED
EULER
2.0000
2.1549
2.3439
2.5672
2.8246
3.1157

RUNGE
KUTTA

2.0000
2.1556
2.3454
2.5695
2.8278
3.1197

h=0.05
x(n)
1. 00
1. 05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50

EULER
2.0000
2.0693
2.1469
2.2328
2.3272
2.4299
2.5409
2.6604
2.7883
2.9245
3.0690

455

IMPROVED
EULER
2.0000
2.0735
2.1554
2.2459
2.3450
2.4527
2.5689
2.6937
2.8269
2.9686
3.1187

1.0
1.50003

_
RUNGE
KUTTA

2.0000
2.0736
2.1556
2.2462
2.3454
2.4532
2.5695
2.6944
2.8278
2.9696
3.1197

Chapter 9 Review Exercises


2. ~
x(n)
0.00
0.10
0.20
0.30
0.40
0.50

3.

.h=iU.
x(n)
0.50
0.60
0.70
0.80
0.90
1.00

4.

EULER
0.0000
0.1000
0.2010
0.3049
0.4135
0.5279

IMPROVED
EULER
0.0000
0.1005
0.2030
0.3092
0.4207
0.5382

RUNGE
KUTTA
0.0000
0.1003
0.2026
0.3087
0.4201
0.5376

h-0.05
x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50

EULER
0.0000
0.0500
0.1001
0.1506
0.2017
0.2537
0.3067
0.3610
0.4167
0.4739
0.5327

IMPROVED
EULER
0.0000
0.0501
0.1004
0.1512
0.2027
0.2552
0.3088
0.3638
0.4202
0.4782
0.5378

RUNGE
KUTTA
0.0000
0.0500
0.1003
0.1511
0.2026
0.2551
0.3087
0.3637
0.4201
0.4781
0.5376

EULER
0.5000
0.6000
0.7095
0.8283
0.9559
1.0921

IMPROVED
EULER
0.5000
0.6048
0.7191
0.8427
0.9752
1.1163

RUNGE
KUTTA
0.5000
0.6049
0.7194
0.8431
0.9757
1.1169

h=0.05
x(n)
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00

EULER
0.5000
0.5500
0.6024
0.6573
0.7144
0.7739
0.8356
0.8996
0.9657
1.0340
1.1044

IMPROVED
EULER
0.5000
0.5512
0.6049
0.6609
0.7193
0.7800
0.8430
0.9082
0.9755
1.0451
1.1168

RUNGE
KUTTA
0.5000
0.5512
0.6049
0.6610
0.7194
0.7801
0.8431
0.9083
0.9757
1.0452
1.1169

EULER
1.0000
1.2000
1.4760
1.8710
2.4643
3.4165

IMPROVED
EULER
1.0000
1.2380
1.5910
2.1524
3.1458
5.2510

RUNGE
KUTTA
1.0000
1.2415
1.6036
2.1909
3.2745
5.8338

h=0.05
x(n)
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50

EULER
1.0000
1.1000
1.2183
1.3595
1.5300
1.7389
1.9988
2.3284
2.7567
3.3296
4.1253

IMPROVED
EULER
1.0000
1.1091
1.2405
1.4010
1.6001
1.8523
2.1799
2.6197
3.2360
4.1528
5.6404

RUNGE
KUTTA
1.0000
1.1095
1.2415
1.4029
1.6036
1.8586
2.1911
2.6401
3.2755
4.2363
5.8446

1EQ_J.

x(n)
1.00
1.10
1.20
1.30
1.40
1.50

5. Using
Yn+l = Yn

+ hun,

YO = 3
Uo

= 1

we obtain (when h = 0.2) Yl = y(0.2) = YO + huo = 3 + (0.2)1 = 3.2. When h = 0.1 we have
Yl

= YO + O.luo = 3 + (0.1)1 = 3.1

Ul = Uo

Y2

Yl

+ 0.1(2xo + l)yo = 1 + 0.1(1)3 = 1.3

+ O.lUl =

3.1 + 0.1(1.3) = 3.23.

456

Chapter 9 Review Exercises


6.

x(n)

y(n)

0.00
0.10
0.20
0.30
0.40

7. Using Xo

2.0000
2.4734
3.1781
4.3925
6.7689
7.0783

initial condition
Runge-Kutta
Runge-Kutta
Runge-Kutta
predictor
corrector

= 1, YO = 2, and h = 0.1 we have


Xl

= Xo + h(xo +yo) = 1 + 0.1(1 + 2) = 1.3

YI = YO

+ h(xo -

X2 =

+ h(XI + YI) =

YO)

= 2 + 0.1(1 - 2) = 1.9

and
Xl

+ h(XI

Y2 = YI

1.3 + 0.1(1.3

+ 1.9) =

- YI) = 1.9 + 0.1(1.3 - 1.9)

1.62

= 1.84.

Thus, x(0.2) ~ 1.62 and y(0.2) ~ 1.84.


8. We identify P(x) = O, Q(x) = 6.55(1

+ x),

= 1, and h

f(x)

= (1 - 0)/10

= 0.1. Then the finite

difference equation is
Yi+l

+ [-2 + 0.0655(1 + Xi)]Yi + Yi-l = 0.001

or
Yi+l

+ (0.0655xi - 1.9345)Yi + Yi-l = 0.001.

The solution of the corresponding linear system gives


0.0
0.0000

0.1
4.1987

0.2
8.1049

0.5
0.4
0.3
11.3840 13.7038 14.7770

0.7
12.5396

457

0.6
14.4083
0.8
9.2847

0.9
4.9450

1. O
0.0000

10

Plane Autonomous Systems


and Stability
Exercises 10.1

1. The corresponding plane autonomous system is


X'

If (x,y)

is a critical point, y =

= y,

y' = -9sinx.

and -9sinx

o.

Therefore x = mf and so the critical points

are (mf, O)for n =: 0, 1,2, ....


2. The corresponding plane autonomous system is

x' = y,
If (x,y)

is a critical point, then y

y' = -2x _ y2.

and so -2x

- y2

= -2x = O. Therefore (0,0) is the sole

critical point.
3. The corresponding plane autonomous system is

X' = y,
If (x, y) is a critical point, y

y' = x2 - y(l - x3).

= and so x2 - y(l - x3) = x2 = O. Therefore (O,O) is the sole critical

point.
4. The corresponding plane autonomous system is
I

If (x, y) is a critical point, y =


sole critical point.

X = y,

y = -4--

1 + x2

x
and so -4 --2
l+x

- 2y.

2(0) = O. Therefore x

and so x = 0,

and so (O,O) is the

5. The corresponding plane autonomous system is


I

X = y,

+ EX3 .

+ EX3 = O. Hence x(-l

If (x,y)

is a critical point, y =

-#.

The critical points are (0,0), (#,0)

and -x

y = -x

and (-#,0).

6. The corresponding plane autonomous system is


x' = y,

y' = -x

458

+ Exlxl.

+ Ex2) =

#,

Exercises

+ Exlxl = x( -1 + Elxl) = O.

If (x, y) is a critical point, y = O and -x


critical points are (O, O), (l/E, O) and (-l/E,
7. From x

+ xy =

into -y - xy

+ y) =

O we have x(l

= O, we obtain

Hence x

10. 1

= O, l/E, -l/E.

The

O).

= O or

O. Therefore x

y = -1. If x = O, then,

= O, Likewise, if y = -1, 1 + x = O or

substituting

= -1. We may conclude

that (O,O) and (-1, -1) are critical points of the system.
8. From y2 - x

= O we have x

= y2. Substituting into x2 -y

= O, we obtain y4 -y

= O or y(y3 -1) = O.

It fo11owsthat y = O, 1 and so (O, O) and (1,1) are the critical points of the system.
9. From x - y = O we have y

= x. Substituting into

3x2 - 4y = O we obtain

3x2 - 4x

= x(3x

- 4)

= O.

It fo11owsthat (O,O) and (4/3,4/3) are the critical points of the system.
10. From x3 - y = O we have y = x3. Substituting into x - y3 = O we obtain x - x9 = O or x(l - x8).

Therefore x

= O, 1, -1 and so the critical points of the system are (O, O), (1,1), and (-1, -1).

+ h = 10.

11. From x(lO - x - ~y) = O we obtain x = O or x


y

= O or

+y =

16. We therefore have four cases. If x

may conclude that y(

-h + 6)

Likewise y(16 - y - x) = O implies that

= O, y

= O or y

= 16. If x + h = lO, we

= O and so y = O, 12. Therefore the critical points of the system are

(O, O), (0,16), (10, O), and (4,12).


= O. It fo11owsthat y = 10, and from -2x + y
y+5
10. Therefore (10,10) is the sole critical point of the system.

12. Adding the two equations we obtain 10 - 15 _y_

10 = Owe may conclude that x


13. From x2eY = O we have x

= O. Since

eX - 1

eO -

= O, the second equation is satisfied for an

arbitrary value of y. Therefore any point of the form (O,y) is a critical point.
14. From sin y

= Owe have

y = mL From eX-Y

1, we may conclude that x - y

= O or x

= y. The

critical points of the system are therefore (mr, mr) for n = O, 1,2, ., ..
15. From x(l - x2 - 3y2) = O we have x = O or x2

+ 3y2 =

1. If x = O, then substituting into

y(3 - x2 - 3y2) gives y(3 - 3y2) = O. Therefore y = O, 1, -1. Likewise x2 = 1 - 3y2 yields 2y = O

so that y

O and x2

1 - 3(0)2

1. The critical points of the system are therefore (O, O), (0,1),

(O, -1), (1, O), and (-1, O).


16. From -x(4 - y2) = O we obtain x = O, y = 2, or y = -2. If x = O, then substituting into 4y(1- x2)
yields y

= O. Likewise y = 2 gives 8(1-x2)

= O or x

= 1, -1. Finally y

= -2 yields

or x = 1, -1. The critical points of the system are therefore (O,O), (1,2), (-1,2),

-8(1-x2)

=O

(1, -2), and

(-1, -2).
17. (a) From Exercises 8.2, Problem 1, x = clt

(b) From X(O)

= (2, -1) it fo11owsthat

Cl

c2e-t and y = 2clt

= O and

459

C2

+ c2e-t.

= 2. Therefore x = -2e-t

and y = 2e-t.

Exercises 10. 1
(e)

y
(-2,2)
2

~
-2

-2

18. (a) From Exercises 8.2, Problem 6, x = Cl + 2C2e-5t and y = 3Cl + C2e-5t.

= (3,4) it follows that Cl = C2=

(b ) From X(O)

(e)

= 1+

2e-5t and y = 3 + e-5t.

4
2

-4

l. Therefore x

-2

/(3,4)

4 x

/~

~
~,-4

19. (a) From Exercises 8.2, Problem 37, x

= cl(4cos3t - 3sin3t) + c2(4sin3t + 3cos3t) and y =

Cl(5 cos 3t) + C2(5 sin 3t). Al! solutions are one periodic with p = 27f/3.
(b) From X( O) = (4, 5) it follows that Cl

= 1 and C2 = O. Therefore x = 4 cos 3t - 3 sin 3t and

y = 5 cos 3t.

(e)
6

(4,5)

6 x

-6

-6

20. (a) From Exercises 8.2, Problem 34, x = Cl(sin t - cos t) + C2(- cos t - sin t) and y = 2Clcos t
2C2sin t. Al! solutions are periodic with p = 27f.
(b ) Frorn X(O)

= (-2,2) it follows that Cl = C2=

l. Therefore x

460

= -2cost and y = 2cost+2sint.

Exercises
(e)

10. 1

y
4

-4

4 x

-4

21. (a) Frorn Exercises 8.2, Problem 35, x


2Cl(cos t) e4t

+ 2c2(sin t) t.

(b) Frorn X(O) = (-1,2)


y = 2(cos t) t.

Cl(sin t - cos t)t

+ C2(-

sin t - cos t)t

and y

Because of the presence of e4t, there are no periodic solutions.

it follows that Cl = 1 and C2 = O. Therefore x = (sin t - cos t)t and

(e)

,-s
-5

-5

22. (a) Frorn Exercises 8.2, Problem 38, x = cle-t(2cos2t


y = c e:' cos 2t

+ C2e-t sin 2t.

C2 =

y
2
(2,1)
-2

+ c2e-t(2cos2t

+ 2sin2t)

and

Because of the presence of e-t, there are no periodic solutions.

(b) From X(O) = (2,1) it follows that Cl = 1 and


y = e-t cos 2t.

(e)

- 2sin2t)

-2

461

O. Therefore x = e-t(2 cos 2t - 2 sin 2t) and

Exercises

10. 1

23. Switching to polar coordinates,


-dr = -1 (dX
x-

dt

+ y -dY)

dt

dt

de
1 (
dx
dt = r2 -y dt

+x

1 -xy - x 2 T 4
= -(
r

dY)
1 2
dt = r2 (y

+ xy

- y 2 T 4 ) = ==T 5

+ xyr 4 + x 2 -

xyr ) = 1.

dr
5
If we use separation of variables on - = ==T we obtain
dt
T

Since X(O) = (4, O), r = 4 and

= ( 4t

+ CI

)1/4

and

e=t+C2.

e = O when t = O. It follows that

C2

= O and

C1

1
= 256' The final

solution may be written as


4

r=
e=t
. -Y1024t + 1 '
and so the solution spirals toward the origin as t increases.
24. Switching to polar coordinates,
-dr = -1 (dX
x-

dt

dt

+ y -dY)
dt

de
1 (
dx
dt = r2 -y dt

+x

= -1(xy - x 2 r 2 - xy

dY)
1
dt = r2 (-y

If we use separation of variables, it follows that


1
r =
and
y'-2t + CI
Since X(O) = (4, O), r = 4 and

- xyr

+ y 2r 2 ) = r 3
- x

+ xyr 2 ) = -1.

e = -t + C2.

e = O when t = O. It follows that

C2

= O and

C1

The final

solution may be written as


4

r
Note that r

-+

00 as t -+ (3\) -.

e = -t.
= -vr.;1=-=3;;';2~t
'

Because O::; t ::;

l2' the

curve is not a spiral.

25. Switching to polar coordinates,

dr
1 (dX
dY)
1
2
2
dt=-;
Xdt+Ydt
=-;[-xy+x(l-r)+xy+y
de
1 (
dx
dt = r2 -y dt

+x

dY)
1 2
2
2
dt = r2 [y - xy(l - r ) + x

(l-r

+ xy(l

)]=r(l-r)

- r )] = 1.

Now dr = r - r3 or dr - r = _r3 is a Bernoulli differential equation. Following the procedure in


dt
dt
2
/
3 dr
/
Section 2.5 of the text, we let w = r : so that w = -2rdt ' Therefore w + 2w = 2, a linear

462

Exercises

first order differential equation. It follows that w

= 1 + cle-2t and so r2 =

1
1 + Cle-

10. 1

2t' The general

solution may be written as


1

e = t + C2.

r = -v!:I:=+=c=l=e ==2~t
'
If X(O) = (1, O), r = 1 and
y

= r sin t

= O. Therefore

e = O when t = O. Therefore

= sin t. This solution generates the cirele r =


Cl

= -3/4,

C2

= O = C2 and so x = r cos t = cos t and


1. If X(O) = (2, O), r = 2 and e = O when
ci

= O and so
r=

e = t.

VI - ~ e-2t

This solution spirals toward the cirele r = 1 as t increases.


26. Switching to polar coordinates,

-dr = -1 (dX
x - + y -dY)
dt

de
dt

dt

1 (
dx
r2 -y dt

dt

+x

= -1 [xy - -x2

dY)
1 [
dt = r2 -y

(4 -

T ) -

xY
+ 7(4
-

xy - -y2 (4 r

2 ]

T )

2
2
xY
2
r ) - x - 7(4 - r)

=-1.

From Example 3, Section 2.2,


r
If X(O) = (1, O), r

cle4t
= 2 11 +
- ele4t

and

= 1 and e = O when t = O. It
r

=2

1- le4t
1+

3
~e4t

e = -t + C2.

follows that
and

e2

= O and

el

= -~.

Therefore

e = -t.

= Owhen t = 3 or t = ln3 and r -+ -2 as t -+ oo. The solution therefore approaches


4
the cirele r = 2. If X(O) = (2, O), it follows that Cl = C2 = O. Therefore r = 2 and e = -t so
Note that r

that the solution generates the cirele r = 2 traversed in the elockwise direction. Note also that the
original system is not defined at (O,O)but the corresponding polar system is defined for r = O. If the
Runge-Kutta method is applied to the original system, the solution corresponding to X(O)
will stall at the origino
27. The system has no critical points, so there are no periodic solutions.
28. Frorn x(6y-l)

= Oand y(2-8x)

= Owe see that (O,O) and (1/4,1/6)

are

critical points. From the graph we see that there are periodic solutions
around (1/4,1/6).

463

= (1, O)

Exercises 10. 1

29. The only critical point is (O,O). There appears to be a single periodic
solution around (O,O).

30. The systern has no critical points, so there are no periodic solutions.

31. If X(t)

= (x(t),y(t))

is a solution,

d
of dx
of dy
- f(x(t) y(t)) = - - + - - = QP - PQ = O
dt
'
ox dt
ay dt
'
using the chain rule. Therefore f(x(t),y(t))
curve of f.

= c for sorne constant c, and the solution lies on a level

Exercises 10.2
1. (a) If X(O)

= Xo lies on the line y = 2x, then X(t) approaches (O, O) along this lineo For all other

initial conditions, X(t) approaches (O,O) frorn the direction determined by the line y = -x/2.
(b)

'-

<;
<,
<,

'-1

,-

<,

'-

-1

2. (a) If X(O)

'Y = -x/2

= Xo lies on the line y = -x, then X(t) becornes unbounded along this lineo For all

other initial conditions, X(t) becornes unbounded and y = -3x/2 serves as an asyrnptote.

464

Exercses 10.2
(b)

,,
\

,,
,
-4

(1,1)

-2
-2

4 x

2
\
v

\'yy3XI21
\,

-4

3. (a) AH solutions are unstable spirals which become unbounded as t increases.


(b)

(1,1)
-5

-5

4. (a) AH solutions are spirals which approach the origino


(b)

-1

-1

5. (a) AH solutions approach (O,O) from the direction specified by the line y = X.

465

Exercises

10.2

(b)

-1

-1

6. (a) AH solutions become unbounded and y = x/2 serves as the asymptote.


(b)

4
3

-4 -3 ___
-2 -t l-"-1

4 x

/':')XI2' ~:
-4

7. (a) If X(O)

= Xo

hes on the line y

= 3x,

then X(t) approaches (O,O) along this lineo For all other

initial conditions, X(t) becomes unbounded and y = x serves as the asymptote.


(b)

3
2

-3 -2 -1

/
1

-1

-2
-3

8. (a) The solutions are ellipses which encircle the origino

466

Exercses 10.2
(b)

3
2
(1,1)

-3

3 x

-2
-3

< 0, we may conclude from Figure 10.18 that (O,O) is a saddle point.

9. Since b. = -41

10. Snce b. = 29 and 7


11. Since b. = -19

= -12, 72 - 4b. > and so from Figure 10.18, (O,O) is a stable node.

< 0, we may conclude from Figure 10.18 that (0,0) is a saddle point.
= -3 and so from Figure 10.18, (O,O) is a stable spral point.

12. Since b. = 1 and 7 = -1, 72 - 4b.

13. Since b. = 1 and 7 = -2, 72 - 4b. =


14. Since b.

= 1 and 7 = 2, 72 -

4b.

and so from Figure 10.18, (O,O) is a degenerate stable node.

= and so from Figure 10.18, (O,O) is a degenerate unstable node.

15. Since b. = 0.01 and 7 = -0.03,72 - 4b.

< and so from Figure 10.18, (O,O) is a stable spiral point.

16. Since b. = 0.0016 and 7 = 0.08, 72 - 4b. =


unstable node.
17. b.

= 1-

J.L2, 7 = 0, and so we need b. = 1 - J.L2

and so from Figure 10.18, (O,O) is a degenerate

> for (O,O) to be a center. Therefore

IJ.LI

< 1.

18. Note that b. = 1 and 7 = J.L. Therefore we need both 7 = J.L < and 72 - 4b. = J.L2 - 4 < for (O,O)
to be a stable spiral point. These two conditions may be written as -2 < J.L < O.
19. Note that b. = J.L + 1 and 7 = J.L + 1 and so 72 - 4b. = (J.L + 1) 2 - 4 (J.L + 1) = (J.L + 1) (J.L - 3). 1t follows

20. 7

2a, b.

<

if and only if -1

< 3. We may conclude that (O,O) will be a saddle point


when J.L < -1. Likewise (O,O) will be an unstable spiral point when 7 = J.L + 1 > and 72 - 4b. < O.
This condition reduces to -1 < J.L < 3.
that 72 - 4b.

<

J.L

a2 + (32 > 0, and 72 - 4b. = -4(3 < O. If a < 0, (O,O) is a stable spiral point. If

a > 0, (O,O) is an unstable spiral point. Therefore (O,O) cannot be a node or saddle point.
-F or X = -A-IF. Since Xp(t) = -A-IF is a particular
solution, it follows from Theorem 8.6 that X(t) = Xc(t)+XI is the general solution to X' = AX+F.
If 7 < and b. > then Xc(t) approaches (O,O)by Theorem 10.1(a). It follows that X(t) approaches

21. AXI + F = O implies that AXI

X as t _, oo.
22. If be < 1, b. = adxy(l
stable node.

- be)

>

and 72 - 4b.

= (ax - dy)2

467

+ 4abedxy > O.

Therefore (O,O) s a

Exercises

10.2

23. (a) The critical point is X

= (-3,4).

(b) From the graph, X appears to be an unstable node or a saddle


point.

(e) Since zx = -1, (O,O) is a saddle point.


24. (a) ThecriticalpointisXI=(-1,-2).

(b) From the graph, X appears to be a stable node or a degenerate


stable node.

(e) Since 7 = -16, ,0,.= 64, and 72

4,0,.= O, (O,O) is a degenerate stable node.

25. (a) The critical point is X = (0.5,2).


(b) Prom the graph, X appears to be an unstable spiral point.

(e) Since 7 = 0.2, ,0,.= 0.03, and 72 - 4,0,.= -0.08, (O,O) is an unstable spiral point.
26. (a) The critical point is X = (1,1).
(b) Frorn the graph, X appears to be a center.

(e) Since 7 = O and ,0,.= 1, (O,O) is a center.

468

Exercises

10.3

Exercises 10.3

1. Switching to polar coordinates,

ddr = ~ (x ddx + y ddY) = ~(ax2 - f]xy + xy2 + f]xy + ay2 - xy2) = ~ ar2
t
r
t
t
r
r
Therefore r

ce'" and so r

-+

Oif and only if a

= aro

< O.

dr

2. The differential equation dt = ar(5 - r) is a logistic differential equation. [See Section 3.2, (4) and

(5).] It follows that


r =

e = -t + e2.

and

1 + qe-50t

If a > O, r -+ 5 as t -+ +00 and so the critical point (O,O) is unstable. If a


and so (O,O) is asymptotically stable.
3. The critical points are x = Oand x = n+ 1. Since g'(x)

g'(n + 1) = -k(n

= k(n+

< O, r

1) - 2kx, g'(O)

-+

Oas t

= k(n+

1)

-+

+00

> Oand

+ 1) < O. Therefore x = O is unstable while x = n + 1 is asymptotically stable.

See Theorem 10.2.


4. Note that x = k is the only critical point since ln(x/k)

-k - kln(x/k),

g'(k) = -k

is not defined at x = O. Since g'(x)

< O. Therefore x = k is an asyrnptotically stable critical point by

Theorem 10.2.
5. The only critical point is T
Theorem 10.2.

= To.

= k, g'(To) = k > O. Therefore T

Since g'(T)

6. The only critical point is v = mg/k.

Now g(v)

=9-

(k/m)v

= To is unstable by

and so g'(v) = -k/m

< O. Therefore

v = mg / k is an asymptotically stable critical point by Theorem 10.2.


7. Critical points occur at x = a, f]. Since g'(x)
k(f]-a).

Since a> f], g'(a)

> Oand so x

k(-a

- f]

+ 2x),

g'(a)

k(a - f]) and g'(f])

= a is unstable. Likewise x = f] is asymptotically stable.

8. Critical points occur at x = a, f], ,. Since

g'(x) = k(a - x)(-f]


g'(a) = -k(f] - a)(I - a)
g'(f])

-,-

2x)

+ k(f]

- x)(I - x)(-l),

< Osince a > f] > . Therefore x = a is asymptotically stable. Similarly

> O and g'(I) < O. Therefore x

= f] is unstable while x =r is asymptotically stable.

9. Critical points occur at P = o.[b, e but not at P = O. Since g'(P)

g'(a/b) = (a/b - e)(-b)


Since a < be, -a + be
asymptotically stable.

= -a

+ be

and

= (a - bP)

+ (P

- e)( -b),

g'(e) = a - be.

> O and a - be < O. Therefore P = a/b is unstable while P

469

e is

Exercises 10.3
10. Since A

> O, the only critical point is A

= K2. Since g'(A) = ~kKA-l/2

- k, g'(K2)

-k/2

< O.

Therefore A = K2 is asymptotically stable.


11. The sole critical point is (1/2,1) and

g'(X) = (-2Y

-2X)
2x - 1 .

2y

= -2 and 6. = 2 so that 72

Computing g'((1/2, 1)) we find that 7

46. = -4

< O. Therefore

(1/2,1) is a stable spiral point.


12. Critical points are (1,0) and (-1,0),

and
, (X) = (2XO -2Y2).

At X = (1, O), 7

= 4, 6. = 4, andso

72

46. = O. We may conclude that (1, O) is unstable but we

are unable to classify this critical point any further. At X = (-1, O), 6. = -4

< Oand so (-1, O) is

a saddle point.
13. y' = 2xy - y
x

= y(2x - 1). Therefore if (x, y) is a critical point, either x

1/2 and y - x2

+ 2 = O implies

that (x,y)

(1/2, -7/4).

= 1/2 or y = O. The case

The case y = O leads to the critical

points (J2, O) and (-J2, O). We next use the Jacobian matrix

g'(X) = (-~~

1)

2x ~

to classify these three critical points. For X = (J2,0) or (-J2,0),


both critical points are saddle points. For X
-13
14. y'

< O. Therefore (1/2,-7/4)

= -y + xy

= y( -1

y = O and 2x - y2

+ x).

= (1/2, -7/4),7

= -1,6.

= -1 and 6.

< O. Therefore

= 7/2 and so 72

46. =

is a stable spiral point.

Therefore if (x, y) is a critical point, either y = O or x = 1. The case

O implies that (x, y)

(O,O). The case x

= 1 leads to the critical points

(1, J2) and (1, -J2). We next use the Jacobian matrix

g'(X) =

(2

-2Y)
x - 1

to classify these critical points. For X = (O,O), 6. = -2


either (1, J2) or (1, -J2), 7

2,6. = 4, and so 72

< O and so (O,O) is a saddle point. For

46. = -12. Therefore (1, J2) and (1, -J2)

are unstable spiral points.


15. Since x2 - y2 = O, y2 = x2 and so x2 - 3x

+ 2 = (x

- l)(x - 2)

= O. It follows that the critical

points are (1,1), (1, -1), (2,2), and (2, -2). We next use the Jacobian

g'(X) =

(-3
2x

470

2Y)
-2y

Exercises 10.3

= (1,1),

to classify these four critical points. For X

Therefore (1,1) is a stable node. For X

= -5, 6 = 2, and so

(1, -1),6

T2 -

46

= 17 > O.

-2 < O and so (1, -1) is a saddle point.

= (2,2), 6 = -4 < O and so we have another saddle point. Finally, if X = (2, -2),

For X

6 = 4, and so

T2 -

= 1,

46 = -15 < O. Therefore (2, - 2) is an unstable spiral point.

= O, y = x or y = -x. The case y = x leads to (4,4) and (-1,1) but the case y = -x
leads to x2 - 3x + 4 = O which has no real solutions. Therefore (4,4) and (-1,1) are the only

16. From y2 - x2

critical points. We next use the Jacobian matrix

x-

g'(X) = ( y

2y

-2x

= (4,4),

to classify these two critical points. For X


(4,4) is an unstable spiral point. For X

= (-1,1),

= 12, 6 = 40, and so T2 - 46 < O. Therefore

T
T

3)

= -3, 6 = 10, and so x2 - 46

< O. It follows

that (-1, -1) is a stable spiral point.


17. Since x'

= -2xy = O, either x = Oor y = O. If x = O, y(l- y2) = Oand so (O,O), (0,1), and (O,-1)

are critical points. The case y = Oleads to x = O. We next use the Jacobian matrix

g'(X)

-2y
( -1 + y

-2X)
1+ x - 3y2

to classify these three critical points. For X = (O,O), T


For X

= (0,1), T =

-4, 6

= 4 and so T2

46

= 1 and

= O. We can

6 = Oand so the test is inconclusive.

conclude that (0,1) is a stable critical

point but we are unable to classify this critical point further in ths borderline case. For X = (O,-1),
6 = -4

< O and so (O,-1) is a saddle point.

18. We found that (O,O), (0,1), (O,-1), (1, O) and (-1, O) were the critical points in Exercise 15,
Section 10.1. The Jacobian is

g'(X) = (
For X

= (O,O), T = 4, 6

(0,1) and (O,-1) give

1 - 3x2 - 3y2
-2xy

= 3 and so T2 - 46 = 4
= -8, 6 = 12, and

therefore stable nodes. For X

= (1, O) or

T2

-6XY)

3 _ x2 _ 9y2 .

> O. Therefore (O,O) is an unstable node. Both


- 46 = 16 > O. These two critical points are

(-1, O), 6 = -4 < O and so saddle points occur.

19. We found the critical points (0,0), (10,0), (0,16) and (4,12) in Exercise 11, Section 10.1. Since the
Jacobian is

g'(X) =

10 - 2x - Iy
2
-y

471

_IX)
2
16 - 2y - x

Exercises

10.3

we may classify the critical points as follows:

Conclusion
unstable node

(O, O)
(10,0)

26

160

-4

(0,16)

-14

-60
-32

(4, 12)

-16

24

36

saddle point
saddle point
stable node

160

20. We found the sole critical point (10,10) in Exercise 12, Section 10.1. The Jacobian is

gl(X) = (
g'((10, 10)) has trace

= -46/15,6.

-22

= 2/15, and

15
- (y + 5)2

72 -

46.

> O. Therefore (O,O) is a stable node.

21. The corresponding plane autonomous system is


el = y,
Since

lel < 11",

yl = (cos e - ~) sin e.

it follows that critical points are (0,0), (11"/3,0) and (-11"/3,0).

The Jacobian matrix

is
gl(X) =

( cos 2e - ~ cos e

~)

and so at (O,0),7 = and 6. = -1/2. Therefore (O,O) is a saddle point. Por X = (11"/3, 0),7 =
and 6. = 3/4. It is not possible to classify either critical point in this borderline case.

22. The corresponding plane autonomous system is

Xl = y,
If (x, y) is a critical point, y =

+ (~_ 3y2)y

yl = -x

and so -x

+ x)

- x2 = -x(l

_ x2.
= O. Therefore (O,O) and (-1, O) are

the only two critical points. We next use the Jacobian matrix
gI (X)

(0

-1 - 2x

1)

"2 - 9y

to classify these critical points. Por X = (O,O), 7 = 1/2, 6. = 1, and

72 -

46.

< O. Therefore (O,O)

is an unstable spiral point. Por X = (-1, 0),7 = 1/2, 6. = -1 and so (-1, O) is a saddle point.
23. The corresponding plane autonomous system is

and the only critical point is (O,O). Since the Jacobian matrix is
gl(X) = ex

+03x2y x3

472

J'

Exercises

10.3

= -1 and 6. = O, and we are unable to classify the critical point in this borderline case.

24. The corresponding plane autonomous system is

4x
- 2y
1 + x2
and the only critical point is (O, O). Since the Jacobian matrix is
I

x = y,

= -2, 6. = 4,72 - 46. =

(-4

10_x2
_:),
(1 + x2)2
-12, and so (O,O) is a stable spiral point.
g'(X)

y = ---

25. In Exercise 5, Section 10.1, we showed that (O, O),

(flTt, O) and (- flTt ' O) are the

critical points.

We will use the Jacobian matrix


I

g (X) =
to classify these three critical points. Por X
this critical point.

Por

(flTt, O),

1)

-1 + 3tx2

= (O, 0),7 = O and 6. = 1 and we are unable to classify

= O and 6. = -2

and so both of these critical points are

saddle points.
26. In Exercise 6, Section 10.1, we showed that (0,0), (l/t,O),

Since Dxxlxl

21xl, the

6.

= (O, O),

are the critical points.

Jacobian matrix is
g (X) =

Por X

and (-1/10,0)

(O2Elxl _ 1 O1) .

= O, 6. = 1 and we are unable to classify this critical point. Por (l/E, O),

= O,

-1, and so both of these critical points are saddle points.

27. The corresponding plane autonomous system is


((3 + 0:2y2)x

x'

y,

1 + 0:2x2

y =-

and the Jacobian matrix is


g'(X)

= (

((3 + o:y2)~0:2x2 - 1)

(1 + 0:2x2)2
Por X = (O, O),
28. From x'
y

= 1/(3.

-(Xx

= O and 6. = (3. Since (3 < O, we may conclude that (O, O) is a saddle point.

+ xy

The case y

= x(

-o: +

y) = O, either x = O or y = 0:. If x = O, then 1 - (3y = O and so

= o: implies that

1 - (30: - x2

= O or x2 = 1 -

0:(3. Since 0:(3

> 1, this equation

has no real solutions. It follows that (0,1/(3) is the unique critical point. Since the Jacobian matrix
is
g'(X)

(-o: +
-2x

473

X)

-(3'

Exercises 10.3
T

= -a - (3 +

1- a(3

73 = -(3 + -(3-

< O and 6.

= a(3 - 1

> O. Therefore (0,1/(3) is a stable critical

point.
29. (a) The graphs of -x+y-x3

= Oand _x_y+y2

in the figure. The Jacobian matrix is

g'(X) =

( -1-

3x2
-1

For X = (O,O), T = -2, 6. = 2,

1)

-1+ 2y
T2 -

= O are shown

46. = -4, and so (O,O) is

a stable spiral point.

(b) For Xj , 6.

= -6.07 < O and so a saddle point occurs at Xj .

30. (a) The corresponding plane autonomous system is


1 3) Y I = E(Y - -y

= y,

X'

3
and so the only critical point is (O,O). Since the Jacobian matrix is
I

g (X) =
T

= E, 6. = 1, and so

T2 -

46. = E2

(O-1

1 ),

E(l _ y2)

4 at the critical point (O,O).

> O, (O,O) is an unstable critical point.


(e) When E < O and T2 - 46. = E2 - 4 < O, (O,O) is a stable spiral point. These two requirements
can be written as -2 < E < O.
(d) When E = O, X" + X = Oand so X = CI COS t + C2 sin t. Therefore all solutions are periodic (with

(b) When

period 271')and so (O,O) is a center.


31.

-2x3
may be solved by separating variables. It follows that y2 + x4 = C. If X(O) =
dx
X
y
(xQ, O) where XQ > O, then C = x~ so that y2 = x~ - x4. Therefore if -XQ < X < XQ, y2 > O and
dy

y'

--

so there are two values of y corresponding to each value of x. Therefore the solution X(t) with
X(O) = (xQ, O) is periodic and so (O,O) is a center.
32.

~ i
dX

x'

~-~may
y

since X(O) = (x(O), x'(O))

be solved by separating variables. It follows that

= (1, O), e = ~. Therefore


y2
x3-3x2+2
233

(x-1)(x2-2x-2)

474

~ ~

- = - - x2
2
3

+ C and

Exercises 10.3
But (x - 1)(x2 - 2x - 2)

> Ofor 1- J3 < x < 1 and so each x in this interval has 2 corresponding

values of y. therefore X(t) is a periodic solution.


33. (a) x'

2xy

O implies that either x

O or y

= O.

and there are no real solutions. If y = 0,1-

If x

= O, then

from 1 - x2

+ y2 = O, y2 = -1

x2 = O and so (1, O) and (-1, O) are critical points.

The Jacobian matrix is

g'(X) = (
and so

T =

O and 6.

=4

at either X

2y
-2x

2X)
2y

(1, O) or (-1, O). We obtain no information about these

critical points in this borderline case.


(b)

1 - x2 + y2
d
or 2xy.JI.. = 1 - x2 + y2. Letting
2xy
dx
dJ.L 1
1
J.L= - , it follows that - = - -1 and so J.L= - - - x + 2c.
dx
x2
X
x
y2
1
Therefore - = -- -x+2c
which can be put in the form
x
x
dy
dx

y'
x'
y2

(x - c)2

+ y2

= c2

l.

The solution curves are shown and so both (1, O) and (-1, O)
-2

are centers.
34. (a)

dy
dx

y'
x'

(b) Let w

-x - y2
X
dy
= - - - y and so y
y
dx

---

+ y = -xy

-1

= yl-n = y2. It follows that dw + 2w = - 2x, a linear first order differential equation
dx

whose solution is

y 2 = w = ce -2x.

(1 x ) .

+ "2 -

Since x(O) = ~ and y(O) = X'(O) = O, 0= c and so


y

="2 1 -

x,

a parabola with vertex at (1/2, O). Therefore the solution X(t)


periodic.
35.

dy
dx

y'
x'

y2
y

and so

x4

x2

"2 = ""4 -"2 + c

y(O) = x' (O) = vo, it follows that

Cl

v5 and so
. 475

or y

x4
"2
-

with X(O) = (1/2, O) is not

+ Cl

Since x(O)

O and

Exercises 10.3

2 (x 2 - 1) 2 + 2v5 - 1
Y ="2x - x + Vo =
2
.
2

The x-intercepts on this graph satisfy


X2

JI - 2v5

= 1

and so we must require that 1 - 2v6 2: O (or Ivo I :S ~ v2


1-

JI - 2v5 and -Xo

) for real

solutions to existo If x5

< x < xo, then (x2 - 1)2 + 2V6 - 1 > Oand so there are two corresponding

values of y. Therefore X(t)

with X(O) = (O,vo) is periodic provided that Ivol :S

1v2.

36. The corresponding plane autonomous system is


x' = y,

y' = X2

- x

+1

and so the critical points must satisfy y = O and


x=
Therefore we must require that

:S ~ for real solutions to existo We will use the Jacobian matrix

g (X)
to attempt to classify ((1 ~)/2,O)
For X

(O

2X _

1)

1 O

when :S 1/4. Note that

= ((1 + vT"=4f)/2, O) and

((1 - vT=4E)/2,

1 vT=4E
2

= O and

6. = =f~.

< 1/4, 6. < O and so a saddle point occurs. For X =

O), 6. 2: O and we are not able to classify this critical point using linearization.

37. The corresponding plane autonomous system is


x'

,
a
f] 3 R
y = -- x - - x - - y
L
L
L

= y,

= q and y = q'. If X = (x, y) is a critica! point, y = Oand -ax - f]x3 = -x(a + {3x2) = O.
> O, a + f]x2 = O has no real solutions and so (O,O) is the only critica! point. Since

where x
If f]

= ( -a _O3f]x2

g' (X)

= -R/ L < O and 6. = a/ L > O. Therefore (O,O) is a stab!e critica! point. If f] < O, (O,O) and
(X,O), where x2 = -a/{3 are critica! points. At X(X,O),
T = -R/L
< O and 6. = -2a/L < O.

Therefore both critical points are sadd!es.


38. If we !et dx/dt

= y, then dy/dt

= -x3

dy
dx

X. From this we obtain the first-order differentia! equation


dy/dt
dx/dt

476

Exercises

10.3

Separating variables and integrating we obtain

=-

y dy

(x3

+ x)

dx

and
1 2

1 4

1 2

"2Y = -:x -"2x + ci.


=

Completing the square we can write the solution as y2


C2 = ~(x5

+ 1)2

+ 1)2 + C2. If X(O) = (xQ,O), then

and so
2
1 (2
y = -"2 x

Note that y

_~(x2

= O when x

)2

1 (2

)2

+ 1 +"2 xQ + 1

x6

+ 2x6 + 1 -

x4

2x2 - 1

= -XQ. In addition, the right-hand side is positive for -XQ

< x < XQ,and


= X(t)

so there are two corresponding values of y for each x between -XQ and XQ.The solution X
that satisfies X(O)
39. (a) Letting x

= (xQ,O) is therefore periodic, and so (O,O) is a center.


and

= Xl

we obtain the system

Xl

= Y and

yl =

1/2 - sinx. Since sinn/6

sin 5n /6 = 1/2 we see that (n /6, O) and (5n /6, O) are critical points of the system.

(b) The Jacobian matrix is


g/(X) =

O 1)
( - cosx
O

and so

Al = gI = ((n/6,0))

(O

= -V3/2

01) and

A2

= gI =

((5n/6,0))

(OV3/2 O1) .

> O and the trace of Al is O, no conclusion can be drawn regarding the critical
point (n/6,O). Since det Aj, < O, we see that (5n/6,0) is a saddle point.
Since det Al

(e) From the system in part (a) we obtain the first-order differential equation
dy
dx

1/2 - sinx
y

Separating variables and integrating we obtain

J
and

y dy =

J (~-

sin x ) dx

1
1
_y2 = -x
2
2

+ cos x + C

y2 = X

+ 2 cos x + C2.

or

477

Exercises

10.3

For Xo near

7r /6,

= (xo, O) then

if X(O)

C2

= -xo - 2 cos Xo and y2 = X + 2 cos x - Xo - 2 cos xo.

Thus, there are two values of y for each x in a sufficiently small interval around
(7r /6,

40.

7r /6.

Therefore

O) is a center.

(a) Writing the system as x'


point. Setting x3 - 2y3

x(x3 - 2y3) and y' = y(2x3 - y3) we see that (O, O) is a critical

= O we have x3 = 2y3 and 2x3 - y3 = 4y3 - y3 = 3y3. Thus, (O, O) is

the only critical point of the system.


(b) From the system we obtain the first-order differential equation
dy
2x3y _ y4
dx = x4 - 2xy3
or

which is homogeneous. If we let y = ux it follows that


(2x4u - x4u4) dx

+ (2x4u3

x4u(1

+ u3)

+ x5(2u3

dx

2u3

- dx
x
-1 dx
x

+ xdu) = O

- x4)(udx

- 1) du
-

=O

+ u (3u + 1) du

+ (1-- - -1 + 2u - 1 )
u +1 u
u2 - u + 1

=O

du = O.

Integrating gives
In Ixl + In lu + 11-ln

lul + In lu2 -

u + 11=

or

x (u: 1) (u + 1)
(y ; x) (~~ _ ~ + 1) =

= C2

2 _ U

(xy

+ x2)(y2

- xy
xy3
x3

478

c2

+ x2)

= c2x2y

+ x4

= c2x2y

+ y2 = 3c3xy.

Cl

Exercises

10.4

(c) We see from the graph that (O,O) is unstable. It is not possible
to classify the critical point as a node, saddle, center, or spiral
point.

Exercises 10.4
= e(o) = !!: and y(O) = e'(O)

1. We are given that x(O)

2g

7f

cos '3

+ e = T + e and

so e

= Wo - T'
y2

Smce
y2

wo.

2g
T cosx

+ e,

Wo

Therefore

2t (cos x - ~

+ 2lg w5)

.
1
1 2
1
1 2
and the x-mtercepts occur where cos x = - - - Wo and so - - - Wo must be greater than -1 for
2

2g

solutions to existo This condition is equivalent to


.

2. (a) Since y

2g

= T cosx + e, x(O)

~
-l (cose - coseo).

When e

= e(O)

Iwol <

= eo and y(O)

2g

1Ff.

2g
= e'(O) = O, e = -Tcoseo

and so y

~
= -eo, y2 = -(cos(
-ea) - cos(eo)) = O. Therefore y
1

= -

&

= O

when e = eo.
(b) Since y = de and e is decreasing between the time when e = eo, t = O, and e = -eo, that is,
t

=T,

dt

de = _
Therefore -dt = de

dt

{Jg

T = _ (T

V 2g

fi9 J cos e VT

cos eo .

1
and so
2g ...jcos e - cos eo

r:

Jo=oo

1
de =
...jcos e - cos eo

(T rOo

V 2g

3. The corresponding plane autonomous system is


x' = y,

f'(x)

y' = -g 1 + [f'(x)J2

479

J-oo ...jcos e - cos eo


(J

- my

de
.

Exercises

10.4

and .

If X

~ (_
f'(X)
_ P_ ) __
(1 + [f'(x)J2)f"(x)
- f'(x)2f'(X)f"(X)
DX
9 1 + [f'(X)J2
my 9
(1 + [f'(x)]2)2

= (X1,Y1) is a critical point, Y1 = O and j'(xl)

therefore
I

g (Xj ) =

4. When {3

(O

-g

f"()

= O. The Jacobian at this critical point is

1)

(3.
-m

Xl

= O the Jacobian matrix is


(-9f~(X1)

which has complex eigenvalues A

~)

= J9f"(X1)i.

The approximating linear system with x' (O)

=O

has solution

x(t)
and period 21T/ J gf"(xl).
5. (a) If f(x)

x2
=

""2'

j'(x)

= x(O)

cos J9f"(X1)

Therefore p ~ 21T/ J gf"(X1)

for the actual solution.

= x and so

dy
dx

y'
x'

X 1
1 + x2 y.

-=-=-g---

We may separate variables to show that y2 = -g lnl l

+ x2) + c.

But x(O) = Xo and y(O)

X'(O) = VD. Therefore c = V5 + 9 lnf l + x5) and so


y

Now

V5 - gln

(1 ++ x~)Xo
1

Therefore, if Ixl :::;[ev5/9(1

= vD2 -

2)

1+ x
1 + Xo

9 In

--2

2: O if and only if x2:::; eV5/9(1 + x5) - 1.

+ X5) - 1J1/2, there are two values of y for a given value of x and so

the solution is periodic.


(b) Since z =
Xmax

x2

""2'

the maximum height occurs at the largest value of x on the cycle. From (a),

= [ev5/9(1

+ x5) - 1]1/2 and so


Zmax

6. (a) If f(x)

= coshx,

X2
max

= _[e
2

(1

vO/9

+ x5) - 1J.

f'(X) = sinhx and [f'(X)J2 + 1 = sinh2 x + 1 = cosh2 x. Therefore


dy
y'
sinhx 1
- - - -g
dx - x' cosh 2 x y .

480

Exercises 10.4
.

We may separate vanables to show that y

= -- 2g + e.

But x(O)

cosh x

Xo and

y(O) = x'(O) = vo.

2
2g
Therefore e = Vo - --and so
cosh Xo
2
2g
29
2
Y - -+vo
cosh x
cosh Xo

Now
2g
cosh x

-_

2g
cosh Xo

- ---

2g cosh Xo
- 2g cosh Xo

2
+ Vo > O if and only if cos11X < --'-___,__---

V6

and the solution to this incquality is an interval [-a,a].

Therefore each x in (-a,a)

has two

corresponding values of y and so the solution is periodic.


(b) Since z
Xmax

= cosh x, the maximum height occurs at the largest value of x on the cycle. Frorn (a),

= a where cosh a =

2g cosh Xo
2
2g - Vo cosh Xo
Zmax

7. If Xm

< Xl < Xn, then F(Xl) > F(xm)

Therefore

2g cosh zr,
2
2g - Vo cosh xo

F(xn).

Letting X = Xl,

G(y) = ~
= F(xm)G(alb)
F(Xl)
F(Xl)

eo

Therefore frorn (2) on page 474, G(y) = F(XI)

< G(alb).

has two solutions YI and Y2 that satisfy

u: < alb < Y2


8. From (1), when y

= alb, Xn is taken on at sorne time t. From (3), if z >

Xn

there is no corresponding

value of y. Therefore the maximum number of predators is Xn and Xn occurs when y = alb.
9. (a) In the Lotka-Volterra Model the average number of predators is die and the average number
of prey is alb. But

Xl = -ax + bxy -

EIX =

-(a + EI)X + bxy

y' = -cxy + dy - E2Y = -exy + (d - (2)y


and so the new critical point in the first quadrant is (di e - E21e, alb

+ EI/b).

(b) The average number of predators di e - E21e has decreased while the average number of prey

alb + EI/b has increased. The fishery science model is consistent with Volterra's principie.
481

Exercises 10.4

10. (a) Solving

+ 0.02y) = O

x( -0.1

y(0.2 - 0.025x) = O
in the first quadrant we obtain the critical

20

40

60

100

80

point (8,5). The graphs are plotted using


x(O) = 7 and y(O)

= 4.

(b) The graph in part (a)


is around 40.

was

obtained using NDSolve

Since x(O)

in M athematica.

= 7, we use the FindRoot

We see that the period

equation solver in Mathematica

to

approximate the solution of x(t) = 7 for t near 40. From this we see that the period is more
closely approximated by t = 44.65.
11. Solving
x(20 - 0.4x - 0.3y)

=O

y(10 - O.ly - 0.3x)

=O

we see that critical points are (O, O), (0,100), (50, O), and (20,40). The Jacobian matrix is
'(

X)=

(0.08(20 - 0.8x - 0.3y)


-0.018y

-0.024x
)
0.06(10 - 0.2y - 0.3x)

and so
Al = g, ((O, O)) = (1.6

A3

= g, ((50, O)) =

(-1.6
O

O)

0.6

-1.2)
-0.3

, ((0,100)) = (-0.8
-1.8

A2

=g

A4

= g'((20, 40))

-0.64

= (
-0.72

O)
-0.6
-0.48)
-0.24

= 6.1 = 0.96 > O, T = 2.2 > O, and T[ -46.1 = 1 > O, we see that (O, O) is an unstable
node. Since det(A2) = 6.2 = 0.48 > O, T = -1.4 < O, and Ti -46.2 = 0.04 > O, we see that (0,100)
is a stable node. Since det(A3) = 6.3 = 0.48 > O, T = -1.9 < O, and Tl- 46.3 = 1.69> O, we see
that (50, O) is a stable node. Since det(A4) = -0.192 < Owe see that (20,40) is a saddle point.
Since det(Al)

12. 6.

= rlr2,

= rl +r2 and T2-46.

= (rl +r2)2 -4rlr2

= (rl-r2)2.

Therefore when rl!

r2, (0,0)

is an unstable node.
13. For X
T2

= (Kl, O), T = -rl +rz (1 - ~~ Q2l) and 6. = -rlr2 (1- ~~ Q2l). Ifwe let e = 1- ~~ Q21,

- 46. = (cr2

+ rl)2

stable node. If Kl

> O. Now if kl > K2 , e


Q2l

< O and so T < O, 6. > O. Therefore (Kl, O) is a

< K2 , e> O and so 6. < O. In this case (Kl, O) is a saddle point.


Q2l

482

Exercises

14.

10.4

(x, f)) is a stable no de if and only if K1 > K2 and K2 > K1. [See Figure 10.38(a) in the text.]
012

021

From Problem 12, (0.0) is an unstable node and from Problem 13, since K1

< K2 , (K1, O) is a
021

saddle point. Finally, when K2

< K1 , (O,K2) is a saddle point. This is Problem 12 with the roles


012

of 1 and 2 interchanged. Therefore (O, O), (K1, O), and (O, K2) are unstable.
K1
15 . 012

< K 2 < K 1021 an d so 012021 >

1. T here fore

/\ = (1 -

t...l

T1T2
012021 )"xy K1K2

< O an d so ('z, y') is

a saddle point.
16. (a) The corresponding plane autonomous system is
X'

= y,

I
-g.
y = -smx--y

(3
ml

and so critical points must satisfy both y = O and sinx = O. Therefore (mf,O)
points.

are critical

(b) The Jacobian matrix

_1(3)
( _E.:osx
Z
ml
has trace

= -!}_z and determinant

.6. =

2
T

E. > O at
9

(32

(O, O). Therefore

4.6. = m2Z2 - 4 T

(32 - 4gZm2

m2Z2

We may conclude that (O, O) is a stable spiral point provided

(32 - 4gZm2

< O or

(3 < 2m.gl.

17. (a) The corresponding plane autonomous system is


x

= y,

(3

y = --yly
m

I --xk

and so a critical point must satisfy both y = O and x = O. Therefore (O, O) is the unique critical
point.
(b) The Jacobian rnatrix is

( _ k _
m

and so

/2IYI)

= O and .6. = k > O. Therefore (O, O) is a center, stable spiral point, or an unstable
m

spiral point. Physical considerations suggest that (O,O) must be asymptotically stable and so
(O,O) must be a stable spiral point.

483

Exercses 10.4
18. (a) The magnitude of the frictional force between the bead and the wire is .t(mgcosB)

for sorne

.t > O. The component of this frictional force in the x-direction is


(.tmg cos B) cos B = .tmg cos2 e.
But
1
cos B = ~=====
+ [f'(x)J2

.tmg

and so .tmg cos B = 1 + [f'(x)J2

/1

It follows from Newton's Second Law that

"
mx

f'(x)

.t
x + mg 1 + [f'(x)]2

(3 ,

1 + [f'(x)J2 -

= -mg

and so

"
x

.t-J'(x)

= 9 1 + [f'(x)J2

(3,
- mx .

(b) A critical point (x, y) must satisfy y = O and f'(x)

(Xl, O) where f'(XI)

= u:

= u: Therefore critical points occur at

The Jacobian matrix of the plane autonomous system is

g'(X) = (g (1 + [f'(x)]2)( - f"(x)) ~ (.t - f'(x))2f'(x)f"(x)


(1 + [f'(x)J2)2

-m~)

and so at a critical point X ,

g'(X) = (-9f~(XI)

-m~)'

1 + .t2
(3
Therefore r = --

occurs. When

< O and

,0.. =

9 f" (x )
~ . When

1+.t

f"(XI)

< O, ,0.. < O and so a saddle point

1"(Xl) > Oand


2
r -4,0..=

(32
-,
m2

-4g--

f"(xI)
< O,
1 + .t2

(Xl, O) is a stable spiral point. This condition may also be written as


(32

19.

< 4gm2 f"(XI) .


1 + .t2

and so using separation of variables, y2 = - (X f(.t) d.t + C or y2 + 2F(x) = C.


2
Jo
We may conclude that for a given value of X there are at most two corresponding values of y. If

ddy

y:

f(x)
Y

(O,O) were a stable spiral point there would exist an x with more than two corresponding values of

f(O) = Ois required for (O,O) to be a critical point of the corresponding


plane autonomous system x' = y, y' = - f(x).

y. Note that the condition

484

Exercises
20. (a) x'

= x( -a + by) = Oimplies that x = Oor y = afb. If x

10.4

= O, then, from

-exy + K y(K - y) = O,
y

= O or K.

Therefore (O,O) and (O,K) are critical points. If fj = o.[b, then

fj [-ex + ~(K - fj)] = O.


The corresponding value of x, x = X, therefore satisfies the equation

ex = K

(K - fj).

(b) The Jacobian matrix is

g'(X) = (
and so at X

-a + by

-ex + K(K - 2y)

-ey

> b' 6. < O in this case. Therefore (O,O) and (O,K) are each saddle
y

=- K

Y < Oand

6. = bcifj

y), we may write the Jacobian matrix

and ci: = ~ (K -

g'((x, fj))
T

-1)' Since

points. Por X = (x, fj) where


as

= (O,0),6. = -aT < O. Por X = (O,K), 6. = n(Kb - a) = -Tb (K

we are given that K

and so

bx
T

>

= (
,
-ey --

bX)
T,

K
O. Therefore (x, y) is a stable critical point and so it is

either a stable node (perhaps degenerate) or a stable spiral point.


(e) Write

using
ci= ~(K-y)=
Therefore

72 -

46.

~Y[(~+4b)Y-4bK].

< O if and only if


4bK
fj <

4bK2

K + 4b = T + 4bK

Note that

4bK2
T+4bK

where K is large, and y =

1< K.

4bK .K::::K
T+4bK

Therefore

72 -

point will resulto

485

46.

< Owhen K is large and a stable spiral

Exercises

10.4

21. The equation

=a

x'
implies that x

o or y

= _1_.

1~

Y x - x

When a

a-1

=x

y -

1) = o

> O, Y = _1_

= (3. On the other hand, if y =

equation for yl, y

e7

> O. If x = O, then from the differential


a-1
1
'1
1
1
_-,
~
= - and so - x - _+ (3 = O.
a-1
l+y
a
a
a-1

It follows that

x=

and if (3(a - 1)

= _a_[(a
-1)(3 -1]
a-1
a-1
1, x > O. Therefore (x,?)) is the unique critical point in the first quadrant. The

>

a ((3 - _1_)

Jacobian matrix is

:x

a y ~ 1- 1

(1 y) 2 )
-x
(
y
- 1
-l+y
(1+y)2
and for X = (x, y), the Jacobian can be written in the form

g/(X) =

(a ~ 1)2 x

).

(a - 1)2 _ 1
a2
It follows that

2
7

and so

+ 1).

= -(6

= _ [( a :21)

Therefore 72 - 46

x + 1] < O,

= (6 + 1f - 46

(a :21) 2 X

= (6 - 1)2 > O. Therefore

(x, y) is a

stable node.
22. Letting y = Xl we obtain the plane autonomous system
Xl

=Y

yl = -8x
Solving x5

+ 8x =

6x3

+ 6x3

x5.

x(x2 - 4)(x2 - 2)

O we see that

critical points are (O, O), (O, - 2), (0,2), (O, -y2), and (O, y2).
The Jacobian matrix is
I

g (X)

-8 + 18x2 _ 5x4

1)

and we see that detg'(Xj) = 5x1_18x2 +8 and the trace of g/{X) is O. Since det(g'((y2,
-8

O))) =

< O, (y2, O) are saddle points. For the other critical points the determinant is positive and

linearization discloses no information. The graph of the phase plane suggests that (O, O) and (2, O)
are centers.

486

Chapter

10 Review Exercises

Chapter 10 Review Exercises

1. True

2. llue
3. a center or a saddle point
4. complex with negative real parts
5. False; there are initial conditions for which lim X(t)

t=ecc:

(O,O).

6. llue
7. False; this is a borderline case. See Figure 10.25 in the texto
8. False; see Figure 10.29 in the texto
9. The system is linear and we identify 6. = -o: and
when 6.

> O and

= o: + 1. Since a critical point will be a center

= Owe see that for o: = -1 critical points will be centers and solutions will be

periodic. Note also that when o: = -1 the system is


Xl

= -x - 2y

yl = X

+ y,

which does have an isolated critical point at (O,O).


10. We identify g(x) = sin x in Theorem 10.2. Then
gl(mr)

= cos nzr < O when

Xl

= mf is a critical point for n an integer and

n is an odd integer. Thus, mr is an asymptotically

stable critical point

when n is an odd integer.


11. Switching to polar coordinates,
1 -xy - x 2 T 3 + xy - y 2 T 3 ) = -r
-dr = -1 (dX
x - + y -dY) = -(

dt

dt

di

-de = -1 ( -y -dx + x -dY) = -1 (y 2 + xyr 3 + x 2 - xyr 3 ) = 1.


dt

r2

di

di

r2

1
Using separation of variables it follows that r = ij3TtCl and
r

= 1 and

e = O,. It

3t + el

follows that

el

--

00,

12. (a) If X(O)

-+

Since X(O)

= (1, O),

= 1, e2 = O, and so
r=

As t

e = t + e2.

ij3t + 1 '

e = t.

O and the solution spirals toward the origino

= Xa lies on the line y = -2x, then X(t) approaches (O,O) along this lineo For all other

initial conditions, X(t) approaches (O,O) from the direction determined by the line y = X.

487

Chapter

10 Review Exercises

(b) If X(O) = Xo les on the lne y = -x, then X(t) approaches (O,O) along this lineo.For all other
initial conditions, X(t) becomes unbounded and y

> Oand so (O,O) is a center.

13. (a)

= O, 6. = 11

(b)

= -2, 6. = 1, 72

14. From x'

= 2x serves as an asymptote.

46.

= Oand so (O,O) is a degenerate stable node.

= x(l + y - 3x) = O,either x = Oor 1 + y - 3x = O. If x

= O, then, from y( 4 - 2x - y)

we obtain y( 4 - y) = O. It follows that (O,O) and (0,4) are critical points.

=O

If 1 + y - 3x = O,

then y(5 - 5x) = O. Therefore (1/3, O) and (1,2) are the remaining critical points. We will use the
Jacobian matrix
g' (X) =

(1 +

y - 6x
-2y

x
)
4 - 2x - 2y

to classify these four critical points. The results are as follows:


X

(O, O)

15. From x

(0,4)

4
-20

(t, O)

la
-3

(1,2)

saddle point
saddle point

10

-15

dx
= -rsine
dt

de
dt

-5

stable spiral point

r coet), y = r sin ' we have


-

dy = reose de
dt
dt
Then r'

Conclusion
unstable node

dr
cose
dt

+-

+ dr

sine.

dt

= or , e' = 1 gives
dx
= -r sine
dt

+ ar cose

. e.
-dy = T cos e + Cl.r sm
dt

= O, which corresponds to X = (O,O), is a critical point. Solving r' = Cl.r we have


= cleat. Thus, when CI. < O, limt->oor(t) = O and (0,0) is a stable eritical point. When a = O,

We see that r
r

r' = Oand r = Cl. In this case (O,O) is a center, which is stable. Therefore, (O,O) is a stable critical
. point for the system when CI. ::; O.
16. The corresponding plane autonomous system is x' = y, y' = t(1 - x2) - x and so the Jaeobian at
the critical point (O,O) is

g/((O,O))

= (_~ ~).

488

10 Review Exercises

Chapter

< Oif and only if -2 < J.-L< 2. We may


therefore conclude that (O,O) is a stable node for J.-L< -2, a stable spiral point for -2 < J.-L< O, an
unstable spiral point for O < J.-L< 2, and an unstable node for J.-L> 2.
Therefore

= J.-L,!:::. = 1 and

72 -

41:::.= J.-L2- 4. Now J.-L2- 4

17. Critical points occur at x = l. Since


g'(X) = -~ e-x/2(x2
2
g' (1)
18.

dy

> O and g' ( -1) < O. Therefore x


y'

-dx = ,x =
x(O)

= Xo

+1

_2xVy2
y

and y(O) = X'(O)

- 4x - 1),

= 1 is unstable and x = -1 is asymptotically stable.

= O. It

_x2

+ c.

But

follows that c = 1 + x5 so that


y2

= (1 + x6 - x2)2 -

l.

> 1 for -Xo < x < Xo and y = O for x

Note that 1 + X5 - x2

[Y2+l =

We may separate variables to show that

= xo.

has two corresponding values of y and so the solution X(t) with X(O)

Each x with -Xo < x < Xo


= (xo, O) is periodic.

19. The corresponding plane autonomous system


Y I = -- (3 Y - -k (s

x' = y,

+ x )3 + 9

and so the Jacobian is


I

g(X)=
Por X = (O,O), 7 = _f}_

O
-~(s+x)2

< O, !:::. = 3k s2 > O. Therefore


m

7 -

(32
12k 2
1
- s = -((3
m2
m
m2

41:::. = -

Therefore (O,O) is a stable node if (32 > 12kms2


where ks3

- 12kms

).

and a stable spiral point provided (32 < 12kms2,

= mg.

20. (a) If (x,y)

is a critical point, y = O and so sinx(w2cosx

= g/w2l.

o.

Either sinx = O (in which

> 1 and so the latter equation has no


real solutions. Therefore (O,O) is the only critical point if w2 < g/l. The Jacobian matrix is
case x

= O) of

COSX

But if w2

g'(X) = (
and so

= -(3/ml

< gil,

- gil)

O
w2 cos 2x -

g/w2l

1:)

T cos x -:%J

< O and 1:::. = gil - w2 > O for X

= (O,O). It follows that

asymptotically stable and so after a small displacement, the pendulum will return to

e' = o.
489

(O,O) is

e = O,

Chapter
(b) If w2

(x,

10 Review Exercises

> gjl, cosx

= gjw2l will have two solutions x =

that satisfy

O) are two additional critical points. If X = (O,O), 6.

saddle point. If X =
6. =

cos x

(x,
-

O), T

-7r

< X < tt . Therefore

= gjl - w2 < O and so (O,O) is a

= -f3jml < O and

w2 cos 2x

3:2 - (2 :':2 - 1) =
w2

Therefore (x, O) and (-x, O) are each stable. When e(O)

w2

3:2 >

= ea, e'(O) =

o and

expect the pendulum to reach one of these two stable equilibrium positions.

490

O.
ea is small we

11

Orthogonal Functions and


Fourier Series
Exercises 11.1

1.

2.

r2 xx2dx

= ~x412

r1 x3(x2

+ l)dx =

1-2

J-1

4-2

= 102(x - l)dx = Gx2

102 eX(xe-X

4.

71"cos x sin 2 x dx = -1 sin 3 x 171"= O


10O
3
o
1-71"/2

6.

571"

~x611 +~x411 =
6
-1
4
-1

3.

- e-X)dx

5. rr/2 xcos2xdx

= ~ (-21 cos2x

+ xSin2x)

1:=0

- x)

171"/2
-71"/2

=O

/4 eX sin x dx = (-21eX sin x _ ~ex cos x) 1571"/4= O

h~

7I"~

7. For m =1- n
rr/2

Jo

sin(2n + l)x sin(2m + l)x dx

= -1 1071"/2 [cos2(n
2 o

1
4 n-m)

= (

- m)x - cos2(n + m + l)x] dx

171"/2

sin2(n-m)x

(.

ti

+ m + 1 )sin2(n+m+1)x

= O.

For m = n

Jorr/2 sin2(2n+1)x

dx =

r= (12 - 21

Jo

1
=-x
2

11[/2
o

1T

=-

491

cos 2(2n + l)x

) dx

1
) sin2(2n+
4 n+1
(2

l)x

11[/2
o

171"/2
o

Exercises 11. 1
so that

II sin(2n
8. For m =1-

+ l)xll =

2"V1f.

ti

rr/2

Jo

cos(2n+ l)x cos(2m+ l)xdx


1 fn7r /2
[cos2(n - m)x + cos 2(n + m + l)x] dx
2 o

=-

1
(
4n-m

17r/2

)sin2(n-m)x

1
(
4n+m+l

17r/2

)sin2(n+m+l)x.

= O.

For m = ti

fo7r/2

cos2(2n + l)x dx = fo7r/2


1

1~2

= -x
2 o

+ ~ cos 2(2n + l)X) dx

1~2

(2
) sin 2(2n + l)x
4 n+l
o

so that

II cos(2n +

l)xll =

2"V1f .

9. For m =1- n

7r

lao

sin nx sinmxdx

la

= -1 7r [cos(n - m)x - cos(n + m)x] dx


2 o
= (

2n-m

17r -

)sin(n-m)x

( 1
2n+m

17r

)sin2(n+m)x

= O.

For m

=n

7r

lco

sin2nxdx

1
] dx
= lo7r [1
- - -cos2nx
o 2 2

so that

II sin nxll

492

1
= -x
2

;_.

7r o

1
-sin2nx
4n

17r=-7r
o

Exercises
10. For m

11. 1

=1p

n?f

lcO sin -xp

mit
P

sin -x

= -1 JP ( cos (77 - m)7r x - cos (n + m)7r x ) dx

dx

20

p
2(n-m)Tt

. (n-m)7r
p

S111

IP -

p
. (n+m)7r
sm
x
2(n+m)7r
p

IP
O

= O.
For m =

ti

(P

Jo

[1'2 - 1

2 n7r
(P
sin -;xdx = Jo

2n7r]
'2cospx
dx

= '2x

I0P

p
2n7r
4nTtsinpx

IP0= p'2

so that

11. For m =1- n


p

loO

lloP ( cos (n -

n7r
mt:
cos - x cos - x dx = p

20

m)7r

x + cos

p
. (n-m)7r
IP
sin
x
2(n-m)7r
p
o

(n

+ m)7r x )
p

dx

p
. (n+m)7r
sin
x
2(n+m)7r
p

= O.
For m = n
p

loo

n7r
cos2 -xdx=
p

loP (1-+-cos-x
1 2n7r)
O

dx=-x

Also
p

loO

nst

1 . cos -x
p

4n7r

IP = O

p
n7r
dx = - sin -x
n7r
p

IP +--sin--x
p

2n7r
p

IP =-.p
O

and

so that

12. For m

=1-

n, we use Problems 11 and 10:

j
j

p
-P
p

-P

n7r
cos-x

m7r
cos-xdx

n7r
sin -x

sin -x

n7r
cos-x

mt:
cos-xdx
p

= 2 lcP

n7r
sin -x

m7r
sin -x

tnat

= 2 JP
dx

=O

dx = O.

Also
nm
j -Pp sin -x
P

tnn
cos -x
P

1
dx = 2

jP ( sin (n -P

493

m)7r
P

x + sin

(71+ m)7r x )
P

dx = O,

IP
O

Exercises 11. 1

-P

-P

mr

1 . cos -x
P
mr

-P

n7r
sin -x
P

dx =

l sm -x
P

and
n7r
cos -x dx
P

dx =

P
-P

dx =

n7r
-Pp sin 2 -x
P

dx = P

-P

-P

and

IP = O,

mr

sin -x
p

mr

-p

ru:
-- p cos -x

IP

mr

= O,

-P

2n7r
p
2n7r
sin -x
dx = --- cos --x
p
4n7r
p

Por m = n
-x
-Pp cos2 n7r
P

-p

(1- + -1
(1 1

cos 2n7r)
-x
p

dx = p,

2n7r)
- - - cos -x

dx = p,

IP
-P

= O.

12dx= 2p

-P

so that

13. Since

CXle-x.2

l. 2xdx = _e-x

-~

OC e-x 2 1
-CXl

210

-e -x21CXl=
O

-00

(4x2 - 2) dx = 2
=2

OC
-CXlx ( 2xe-x2)

dx - 2

O
'

CXl
2
-oc e-x dx

21C CXl 2)
( -xe-x -CXl + -CXle-x dx

- 2 CXl e-x 2 dx
-oc

and

OC e-x 2 . 2x (4x~ -oc


?

2) dx = 4

OC
(
2)
-oc x2 2xe-x

= 4 ( _x2e-x 21CXl
-oc +

the functions are orthogonal.

494

dx - 4

OC
-CXl xe"?2dx

CXl
-CXl xe-x 2)dx

- 4

CXl
-CXl xe?" 2dx

Exercises 11. 1
14. Since

rOO e-X 2 1(1 -

Jo

foOO e-x

1 Gx2

- 2x

x) dx

+ 1) dx

= (x

OO
roo
O - Jo

- l)e-X

= (2X -1-

= 1 + (2

= O,

I~+ fooo

~x2) e-x

- x)e-X

e-xdx

I~+ fooo

e-xdx

e-X(x

- 2)dx

= O,

and

= e=

(1"2X -"2x5
3

+ 3x - 1) 100
o + Joroo e -x

1 + e-x Gx2 - 5x + 3) [ +

=1- 3

+ e-X(3x

- 5) [

-3

fooo

fooo

-23x 2 + 5x

3) d x

e-X(5 - 3x) dx

e-xdx

= O,

the functions are orthogonal.

Ji <PO(X)<Pn(x)dx = O for n = 1, 2, 3, ...

15. By orthogonality
n = 1,2,3, ....

16. Using the facts that <Poand

(o:x

; that is,

Ji <Pn(x)dx = O for

<Pl are orthogonal to <Pn for n > 1, we have

+ (J)<Pn(X)

dx

= o:

x<Pn(x) dx

+ (J l; t

{b

<Pn(X) dx
(b

= o: Ja <Pl(X)<Pn(x) dx + (J Ja <PO(x)<Pn(x) dx
=0:0+{J0=0
for n = 2, 3, 4, ....

17. Using the fact that <Pnand <Pmare orthogsnal for n


lI<Pm(x) + <Pn(x)112 = \Pm(X)

m we have

+ <Pn(x)J2dx b [<p~(x) + 2<Pm(X)<Pn (x) + <p;(x)] dx

= b <p~(x)dx + 2l
= lI<Pm(x)1I2

=1=

<Pm(X)<Pn(x)dx

+ 11 <Pn(x)

112 .

495

+ b

<p;(x) dx

Exercises 11. 1
18. Setting
(2
1-2

0=

(2

+ CX3 + C2X )

13(x)f (X) dx = 1-2 (x2

dx =

16

64

:3 + SC2

and
0= [:
we obtain

= O and

CI

13(x)h(x)

C2

dx

(X3

[22

+ CX4 + C2X5) dx

= 654CI

= -5/12.

19. Since sin nx is an odd function on

[-7r,

7r],

(1, sin nx) = ["Ir"lr sin nx dx = O

= 1 is orthogonal to every member of {sin nx}. Thus {sin nx} is not complete.

and f(x)
20. (h

(b

+ h, 13) = Ja

+ h(x)l13(x)

[h(x)

21. (a) The fundamental period is


(b) The fundamental period is

(b

= Ja h(x)13(x)

dx

27r / 27r

27r / ( 4/ L)

dx

(b

+ Ja

h(x)13(x)

dx

= (h, 13) + (h 13)

l.
= 17r L.

+ sin 2x is 27r.
period of sin 2x + cos 4x is 27r / 2 = tt .
period of sin 3x + cos 4x is 27r since the

(e) The fundamental period of sin x


(d) The fundamental
(e) The fundamental
and

27r /

smallest integer multiples of

4 = 'n/2 that are equal are 3 and 4, respectively.

(f) The fundamental period of f(x)

= 2p/n.

is 27r/(n7r/p)

Exercises 11.2
1.

ao =
an

I.
i" f (x)
7r J-"Ir

1
bn = -

-"Ir

j"lr

-"Ir

7r

f(x)

100

7r n=l

7r

I.
7r

= -(1- cos n-r) =

1
-[1(-ltl
n7r

tit:

sinnx

sinnxdx

=O

(0 -1 dx

J -"Ir

+ I.
7r

r 2dx = 1

Jo

= -1 j"lr f (x) cos nx dx = -1 JO - cos nx dx + -1 J"Ir 2 cos nx dx = O


7r

bn

1 lo"lr

7r

l-(-l)n

-11"

=-

7r

= 1

7r

n7r
f(x)sin-xdx

= - +-

= -1 J"Ir cosnxdx

7r

2. ao = I. j11" f (x) dx
an

I.
i" 1dx
7r Jo

dx =

= -1 j"lr f(x)cos-xdxn7r
7r

-"Ir

= -1 j11" f(x)sinnxdx
7r

-11"

27r /3

7r

-11"

= -1 JO -sinnxdx+
7r

-rr

7r

-1 lo"lr 2sinnxdx
7r

496

3
= -[1(-ltl
n7r

Exercises
1 3 00 1 - (_1)n
f(x) = - + - L
2 7r n=l
n
3. ao = JI f(x) dx
-1

SlD nx

= JO 1 dx +
-1

Jo

x dx

=~
2

an

1
= J I f(x) cosn7rX dx = JO coenxx dx + 101 x cosrcs x dx = 22"[(-lt
-1
-1
O
n 7r

bn

I f(x) sin n7rXdx


J -1

3
f(x)=-4+L

00

[(-1)n-1

n=1

4. ao = I f (x) dx =
an

b-, =

-1

J
J

1x

f(x)cosn7rxdx
-Ion
I

-1

5. ao = ~j7r

[(

f(x)dx

tt

i7r

x2 sin nx dx

f(x) = -7r2

101 x cos nr z dx = ~[(-lt


1
Jor

(_l)n+l

x sin n7rXdx = -'---'-n7r

cos n7rX +

(_1)n+l

[2(-l)n
n2
n=1
00

]
sin n7rX

n7r

r x2dx
1

= ~7r2

i7r

7r

7r

non

tt

cosnx

(x

1 -sinnx
2

x2cosnxdx=-

= -1 ( - -x2 cos nx

+L

-lJ

tt

7r

-7r
7r Jo
1 j7r f(x)cosnxdx=7r -7r

bn = -1

dx = ~
2

=~

n7r

n7r

_l)n - 1
2

1
cosn7rx--sinn7rx

7r

an=-

-1

f (x) sin n7rXdx =

3
00
f(x) = -4 + L
n=1

17r+ -2lo7r

17r -- 2lo7r

7r

x cos nx dx )

+ (7r'_(_l)n+l + 2[(-l)n-1J)]
n

n37r

nOn

sinnx

j7r f(x) dx = ~ JO 7r2 dx + ~ r (7r2 - x2) dx = ~7r2


-7r
n -7r
7r Jo
3
1 j7r f(x)cosnxdx=1 JO 7r2cosnxdx+- 1 lo7r (7r2-x2)cosnxdx
7r =r:
7r -7r
7r O
2
-1 (7r2 - x2 sinnx 17r + -2i7r xsinnxdx ) = 2(-lt+
1

7r

non

7r

bn=- 1 j7r f(x)sinnxdx=7r

-7r

1 JO 7r2sinnxdx+-

1 lo7r (7r2-x2)sinnxdx

7r

Ti

-7r

497

xsinnxdx

2( _l)n
=--2-

2 [(-1t
= -tt (-1)n+ 1+ -3-

6. ao = ~

an=-

-lJ

= JO sin n7rXdx + 101 x sin n7rXdx = -- 1

22
7r

Jo

11.2

tt

- 1J

Exercises 11.2

-1] +.!. (X2 -11"2 cosnx

= ~[(-lt

11"

J; [2n2 (_I)n+l

511"2
f(x) = (3 +

11"

11"

-11"

I.

an = -1

11"

00

I,

j1l"

11"

= 211"
=O

= -1

11"

I,

'(

(3 - 2x) cos nx dx

=O

-11"

4
n

(3-2x)sinnxdx=-(-lt

-11"

= 3 +4 L

=6

-11"

f (x) cos nx dx

n=l

_1)n

--

sin nx

r sin x dx = ~11"

ao = .!.j1l" f (x) dx =.!.

11" Jo

-11"

= -1 j1l" f(x) cosnx dx = -1111" sinx cosnx dx = -1 la1l" [sin(n + l)x + sin(1 - n)x] dx
11"

11"

-11"

1+ (-lt

= 1I"1-n
(
2)
al

-11"

(3 - 2x)dx

= .:

-11"

11"

(-lt]

00

f (x)

sin nx

n 11"

00

11"

bn=11"

(-In])
+ 2[1- n311"

+ -i-[I-

2
L -(
_l)n+1 sin nx

= 11" +

= -1

= ~(-lt

8. ao = .!.j1l" f(x)dx
11" -11"

an

11"

-11"

n=l

9.

+ (11"~(_I)n

r iCOSnXdX)

Jo

= -1 j1l" (x + 11") cos nx dx

f (x) cos nx dx

-11"

11"

-11"

= -1 j1l" f (x) sin nx dx = -2 (-1 t+

f(x)

an

cos nx

= .!.j1l" f (x) dx = !_ j1l" (x + 11") dx

7. ao

111"-

nOn

= 211"

Jor

for n

sin2xdx

211"

= 2,3,4, , , ,

=O

bn = -1
f (x) sin nx dx = -1 la1l" sin x sin nx dx
11" -11"
11" o
1 {11"
= 211" Jo [cos(1 - n)x - cos(1 + n)x] dx = O for n = 2,3,4, , , ,

b: = -21
(1 - cos 2x) dx = ~
11" k
2
f (x)

1
=-

11"

ool+(_I)n

n=211"1 - n

+ - sin x + L

2) cos nx

498

Exercses

10.

11.2

2 j1r/2
2 Jc1r/2
?
f (x) dx = cos x dx = ::
7f -1r/2
7f o
tt

ao = an

= -2 j1r
tt

/2

f(x) cos 2nx dx

-1r/2

= -2J1r

/2

= -2Jc1r

/2

7f

1
cosx cos 2nx dx = 7f

!n

1r/2

[cos(2n - l)x

+ cos(2n + l)x] dx

2(_1)n+1

7f(4n2 - 1)
2 j1r /2
bn = f(x) sin 2nx dx
7f -1r/2

7f

1 Jc1r/2
cos x sin 2nx dx = [sin(2n - l)x
7f O

+ sin(2n + l)x] dx

4n
7f(4n2 - 1)

1 L [2( _1)n+1

4n

00

f (x) = - +
(2
) cos 2nx
7f n=l 7f 4n - 1
11.

0,0

= -21j2

-2

an=-21 j2

-2

f (x) dx = ~ (JO - 2 dx
2 -1

f(x)

12.

ao

= --

;:2
t.
= - t.
=~

an=-21
bn

1
2

+L
00

n=l

Jo

(10

dX) = - ~
2

1
2

(JO [ -2sin-x n7f


2

-1

-2

ten
f(x)cos-xdx=2

1
2

-2

n7f
f(x)sin-xdx=
2

-1
2

ten

la

) =--sin-1

nr:
cos-xdx
2

n7f
sin-xdx
2

nat

=- 3

ru:

n7f
2

( 1-cos- n7f)
2

n7f) . nr: ]
1 - cos sm-x
2
2

2 cos-xdx
n7f
2

n7f
xcos-xdx+
2

O
l
O

+ -3

] dx+

la

\ dX) = ~

(!nI

(la

] dx+

-1

X dx

- 1

(JO [ -2cos-xn7f

1 . n7f
nr:
--n7f
sm - cos -x
2
2

f (x) dx = ~

11

) sin 2nx

1
2

nr:
f(x)cos-xdx=2

nr:
bn=- 1 j2 f(x)sin-xdx=2 -2
2

+ 7f(24n

) =2"2
2 ( cos--1
n7f
n tt
2

n7f
xsin-xdx+
2

2
1

nr:
sin-xdx
2

2 ( sin -n7f + -(
n7f -lt+l )
= -n27f2
2
2
f(x) = -3
8

+ '"
--2 ( cos-nr:
s: n27f2
2
00

n=l

n7f
an=-51 j5 f(x)cos-xdx=-5
5

n7f
-1 ) cos-x+
2

1
5

(JO
-5

n7f
cos-xdx+
5

nr:
-- 2 (.sm-n7f + -(-1)
n27f2
2
2

la (l+x)cos-xdxn7f)
5

499

n+l).

sm-xn7f ]
2

5
=2"2[(-1)n-1]
n 7f

Exercises

11.2

mf
bn=- 1 j5 f(x)sin-xdx=5 -5
5

(JO

1
5

-5

9
Loo' [ --5 [( -1)n -lcos-x+]
n1f
=-+
4
n21f2
5

fx()

t.

2 -2

n1f
f(x)cos-xdx=-

n1f
bn=- 1 j2 f(x)sin-xdx=2 -2

(JO

-2

(JO

-2

la

n1f
(2+x)cos-xdx+

2
O

la

n1f
(2+x)sin-xdx+

2
O

n1f
2cos-xdx
n1f
2sin-xdx

2 [ 1 - ( -1 )n] cos -x
mf + -2 ( -1 )n+l.sm -x
n1f ]
= -32 + Loo [ -2
n1f
2
n=l n21f2

bn =

16. ao =

an

1 j1l'

1 j1l'

f (x) sin nx dx = -

1f

-11'

~ j1l' f (x) dx = ~
1f

1f

-11'

bn =

f(x)

r(eX -

lo

1 j1l'
f (x) cos nx dx
1f -11'

=-

1 j1l'

1f

f (x) sin nx dx

-11'

e1l'-1f-1
21f

+L
00

n=1

eXsin nx dx =

-11'

1) dx

= ~ (e 11'1f

1111'
(eX - 1) cosnx dx

1 lo1l'

(eX - 1) sin nx dx

[e1l'(_l)n_1
2) cosnx
1f 1 + n

')
2 2[1-(-lt]
n 1f
w

) =-(-lt+1
2
n1f

1f - 1)

[e1l'(_l)n - 1]
(
2)
1f 1 + n

1f O

1f O

(_l)nn(e-1I' - e1l')
(
'))
1f 1 + n-

==-

f ()x

1f

5
=-(-lt+1
n1f

5
O

5 ( -1)n+l.sm-xn1f ]
n1f
5

n=l

an=- 1

la (l+x)cos-xdxmf)
5

n1f
sin-xdx+
. 5

1 (ne1l'(_l)n+l
1
2
+n

=-

1f

n
+ ( --2
1+ n

[
e1l'(_l)n+

n
(_l)n
+ --2
+ -- -1)
1+ n
n
n

+ 11 +

(_l)n_1)
n

sin nx

17. The function in Problem 5 is discontinuous at x = 1f, so the corresponding Fourier series converges

500

Exercses 11.2
to 7r2/2 at x = Ti. That is,
7r2 7r2
=- +
2
6

[2( _l)n
L
2
n= 1
n
00

cos n7r+

(7r
?[( -l)n - 1])
]
- (_l)n+l + 3
sin n7r
n
n tt
'

and
~2 = ~ (7r22_ ~2) = 1 + 212+ ;2 + ....
At x

= O the series converges to O and

so
7r2
1
--1--+---+
12 22

1
32

1
42

18. From Problem 17


7r2= ~ (7r2 + 7r2) = ~ (2 + _3_ + _3_ + ... ) = 1 + 2_ + 2_ + ....
8
2 6
12
2
32 52
32 52
19. The function in Problem 7 is continuous at x = tt /2 so
-37r =
2
and

f (7r)
- = 7r+
2

(i

L.._., -2 ( -1) n+l

sin -n7r = 7r+ 2 1 - - + -1 - -1 + ... )


2
3 5 7

n=l n

7r
111
-=1--+---+.
4
357

20. The function in Problem 9 is continuous at x = tt /2 so


7r)
1= f ( 2

1
2

tt

1 1
2
=-+-+----+--_
tt
2 37r
and
or

= - + - + L.._.,

n=2

1 + (_l)n
nt:
cos 7r(1- n2)
2

2
3 . 57r

Ti
2
2
7r=1+-+---+----'"
2 3 35

2
57

Ti
1
1
1
4=2+~-'3.5+57-

501

2
5 . 77r

...

Exercises

11.2

21. (a) Letting ca

= ao/2,
f(x)

Cn = (an - ibn), and C-n


=

L (nT
anCOS -x

ao
2"
+

00

= Co

+L

einrrx/p

+ ibn)/2 we have

nT )
+ bnsin-x

00

n=l

= (an

+ e-inrrx/p
2

an

+ bn

einrrx/p _ e-inrrx/p)
2i

- bn

ieinrrx/p _ ie-inrrx/p)
2

n=l
00

= Co

+L

einrrx/p

+ e-inrrx/p
2

an

n=l

~
=co+L.._.
n=l

(an - ibn inrrx/p


e
2

+ an + ib e -inrrx/p)
2

00

= Co + L (Cneinrrx/p + c_nei(-n)rrx/p)
n=l

00

eneinrrx/p.

n=-oo

(b) Multiplying both sides of the expression in (a) by e-imrrx/p and integrating we obtain

f(x)e-imrrx/pdx

: ( f:

L
00

jP

Cn

n=-oo

ei(n-m)rrx/pdx
ei(n-m)rrx/pdx

L Cn jP

ei(n-m)rrx/pdx

-p

ni'm
=

dx

-p

L Cn jP

ni'm

cneinrrx/pe-imrrx/p)

n=-oo

+ Cm jP

ei(m-n)rrx/pdx

+ Cm jP

dx

-p

-P

L Cn jP
ni'm

-P

ei(n-m)rrx/pdx

+ 2pcm.

-P

Recalling that
eiY=cosy+isiny
we have for n-m

j p ei(n-m)rrx/pdx

an integer and n
=

-P

and

f. m

P
ei(n-m)rrx/p
i(n - m)T

e-iY=cosy-isiny

\P
-P

(ei(n-m)rr _ e-i(n-m)rr)

i(n - m)T
=.(
p ) [cos(n-m)T+isin(n-m)T-cos(n-m)T+isin(n-m)T]
zn-mT
= Q.

502

Exercses

11.3

Thus

j -P f(x)e-im7rx/Pdx = 2pcm
. / "dx.
Cm = -1 jP f(x)e-,m7rx
p

and

2p

22. Identifying f(x)

= e-x

-P

and p = 7r, we have

+ 1)7r

2(in

+ 1)7r

=_

7r
-7r

e-(in+l)x

1
2(in

[e-(in+l)7r

_ e(in+l)7r]

e(in+l)7r _ e-(in+l)7r

= ---;----,---

2(in

e7r(cos n7r

+ 1)7r
+ i sin n7r)

e-7r

(cos n7r - i sin n7r)

==-~------:---..,:-------2(in + 1)7r
cosn7r
2(in + 1)7r

(e7r - e-7r

= -'----;------

(e7r - e-7r)(_I)n
2(in

Thus

f(x) =

P'-7r-

'L
00

(-lt e 2
n=-oo
2(in

+ 1)7r
e

einx.

+ 1)7r

Exercises 11.3

==-sin3x = -f(x), f(x) is an odd function.


cos( -x) = -x cos x = - f(x), f(x) is an odd function.

1. Since f(-x)

= sin(-3x)

2. Since f( -x)

= -x

= (_x)2 - X = x2 - x, f(x) is neither even nor odd.


4. Since f( -x) = (_x)3 + 4x = _(x3 - 4x) = - f(x), f(x) is an odd function.
5. Since f( -x) = el-xl = elxl = f(x), f(x) is an even function.
6. Since f( -x) = e-x - e" = - f(x), f(x) is an odd function.
3. Since f( -x)

< x < 1, f( -x)


For ::;x < 2, f( -x)

(_x)2 = x2 = - f(x), f(x) is an odd function.

7. For

8.

= -x + 5 = f(x),

9. Since f(x)

f(x) is an even function.

is not defined for x < 0, it is neither even nor odd.

503

Exercises 11.3
10. Sincej'( -x)

= I(_x)51 = Ix51= f(:1;), f(x) is an even function.

11. Since f(x) is an odd function, we expand in a sine series:

bn

=~

(1r 1. sin nx dx

~Jo

Thus
f(x)

00

n=l

rtt;

L-

[1 -

= 2. [1n~

(-1 t] .

(-ltl sin nx.

12. Since f(x) is an even function, we expand in a cosine series:


ao =
an =

1dX

=1

n~
2
nt:
cos -x dx = - - sin -2 .
2
nt:

Thus
1
f(x) = - +
2

-2
L
-sinn=l ni:

nt:

00

nt:
cos -x.

13. Since f(x) is an even function, we expand in a cosine series:


ao=an

.:

xdx=~

2 -1 t = -2lo1r x cos nx dx = -2-[(


~

n ~

Thus
~

f (x)

=- +
2

00

1].

L -2-[(-1 t
n=l n ~

- 1]cos nx.

14. Since f(x) is an odd function, we expand in a sine series:


bn = -2lo1r x sin nxdx
~

Thus
00

n=l

2 -lt+1.
= -(

L -(-1t+

f (x) =

sin nx.

15. Since f(x) is an even function, we expand in a cosine series:


l

2
2
x dx = -

ao = 2

inO

an=2

x2
lo l x2cosn~xdx=2 ( -sinn~x
n~

Thus
f(x)

=- +
3

00

1 -- n~2 lol xsinn~xdx )

L 22(-ltcosn~x.
n=l n ~
504

4
= 22"(-lt

n ~

Exercises
16. Since f(x)

11.3

s an odd function, we expand in a sine series:


b., = 2

x2
= 2 ( --cosn7rX
nt:

f 1 x2snmfxdx

)0

11

+ -2101 x coe tin x dx )


tvn

Thus
f(x) =

L (2(-lt+1

17. Since f(x)

n7r

- 1J) sin tinx.

tt

is an even function, we expand in a cosine series:


2101r (7r 2 -

aO = -

tt

an

4
+ 33"[(
_l)n

00

n=l

= -2101r (7r2
7r

2 dx

X )

4 2
= -7r
3

= -2

x2)cosnxdx

(7r2

7r

x2 sinnx 11r+ -2101r xsinnxdx


non
O

4
= 2(-lt+1.
n

Thus
f(x)

_7r2 +
3

18. Since f(x)


bn

4
L 2(-lt+1
00

cosnxdx.

n=l n

is an odd function, we expand in a sine series:

= -2101r x3sinnxdx
7r

211"2
n

= _(_1)n+1

211"2
= -(-1)
ti

= -2

tt

( --cosnx
x3
non

11r

+ -3101r x2cosnxdx

12 107r
xsinnxdx
n 11" O

-2-

n+1 - - 12 ( -- x cosnx 11r


n27r
non

+ -1

io

1r cosnxdx )

= -211"2(-1
n'

)n+1 + -12 ( -1 )n .
n3

Thus
12 ( -1)n). sin nx.
f (x ) = ~L.. (211"2(
-1)n+1 + "3
n=l
n
n
19. Since

f (x) is an odd function, we expand in a sine series:


bn

= -tt2101r
(x+
o

l)sinnxdx

2(7r + 1)
(_l)n+1
n7r

Thus
f(x) =

L
00

n=l

(2(11" + 1) (_l)n+l
n7r

505

+ -2)
n11"

.
smnx.

2
+-.
nr:

Exercises 11.3
20. Since f(x) is an odd function, we expand in a sine series:
b., = 2la1 (x - 1) sin mfX dx

= 2 [la1

x sin mfX dx - la1 sin mfX dX]

]1 = - -2

Ix
1
= 2 [ 22
sin mfX - - cos mfX + - cos n7rX
n 7r
n7r
n7r

n7r

n7r
)
cos - - 1 .
2

Thus
f(x)

00

n=l

n7r

L -sinn7rx.

=-

21. Since f(x) is an even function, we expand in a cosine series:


ao

an =

l
lao
l

x dx

1dx

= -3

n7r

lao xcos -xdx


2

nt:
cos -xdx
2

4
= 22

n--t:

Thus
4 (
nt:
)
nt:
f (x) = -43 + '"L00 -cos - - 1 cos -x.
n
27r2
2
2
n=l

22. Since f(x) is an odd function, we expand in a sine series:


b.;

= -1

7r

i7r
O

127r 7rsin -x
n dx = -24 sin -n7r + -(
2 _l)n+l.
7r
2
n 7r
2
n

n
2

x sin -x dx + .

Thus
f(x)

L (4-sin-

n7r + _(_l)n+
2
2
n

00

n=l n27r

1) sin-x.n
2

23. Since f(x) is an even function, we expand in a cosine series:

io7r sin x dx = -7r4

ao

= -2

an

= -2 j.7r sinx

7r o

7r o

cosnxdx

la

= -1 7r [sin(n + l)x + sin(l - n)x] dx


7r O

7r(1:n2)[1+(-ltl
al

forn=2,3,4,

...

io

= -1 7r sin2xdx = O.
7r O

Thus
f(x)

=~+

7r

f 2[17r(1+-(_l)n]
n

n=2

2)

506

cosnx.

Exercises
24. Since (x) is an even function,

we expand

Exercise 11.2 for the computation

in a cosine series.

[See the solution

of Problem

of the integrals.]

ao = - 2 la1f /2 cos x dx = -4
7r/2 o
7r
2

= -/7r 2

an
Thus

ln1f/2 cosx cos -/-x


rit:

tt 2

2
(x)=-+L

7r

dx =

00

4(_l)n+1

n=l

7r 4n - 1

(2

4(_l)n+1

(2

7r 4n - 1

)cos2nx.

f1/2
25. ao = 2 Jo
1 dx = 1
1/2

an = 2

bn

=2

lnO

1 . sin n 7rx dx
00

n=l

L -2
00

J1/2

(x)

cos -n7r)

-nt:2 sin -nr:2 cosn7rX

+L

26. ao = 2 f1

1
1
1/2

bn = 2

= -2 ( 1 tin

n=l nr:

an=2

tin

1/2

1
(x) = 2
(x)

2
n7r
1 . cos n7rXdx = - sin -

lnO

1/2

( 1- cos- n7r) sin nzrz


2

1 dx = 1
lcosn7rxdx=--sin-

trn

tu:

. n7rXd x = -2 ( cos -n7r + ( -1 )n+1)


1 . sm

ru:

1 L (2-- sin -ntt cosn7rX )


2
nm
2

=- +

00

n=l

(x) = L00 - 2 (n7r


cos n=l ru:

27.

+ (_1)n+1 )

sin n7rX

ao = -4lc1f /2 cosxdx =-4


7r O

7r

4lc1f/2

2lc1f/2

an = -

cos x cos 2nx dx = 7r O

7r O

bn =

4101f
-7r
O

/2

2la1f

cos x sin 2nx dx = 7r O

[cos(2n

/2

[sin(2n

+ l)x + cos(2n
+ l)x + sin(2n
507

- l)x] dx =

- l)x] dx =

4( _l)n

(
2)
7r 1 - 4n

(2

8n

7r 4n - 1

11.3

10 in

Exercises 11.3
2

4( _l)n

00

f(x)

= :;:+ ]; 7r(1 _ 4n2) cos 2nx

f (x)

8n

00

L
n=l7r

(2

) sin 2nx

4n - 1

28. ao = ~ (" sin x dx = ~


tt

an=- 2
tt

Jo

tt

lo'" sinxcosnxdx=-

tt

lo'" [sin(n+1)x-sin(n-1)x]dx=

2[( _l)n
+2) 1] for ti = 2,3,4, ...
(

la

bn = -2 11" sinx sin nx dx = -1 Jc1l" [cos(n - l)x - cos(n


7r O
7r o

+ l)x]

7r1-n

dx = O

for n = 2,3,4, ...

al = ~ (" sin 2x dx = O
7r

Jo

bl = ~ (" sin 2 x
tt

Jo

f(x)

dx = 1

= sin z
2

L (-1)n+1
2
1- n

00

f(x) = - + 7r 7r n=2
29. ao =

~(
7r

an = -2
7r
bn =
f (x)

("/2 X dx

Jo

(lo"'/2
o

+ i"

J"'/2

x cos nx dx

cos na:

(7r - x) dX)

'"

",/2

=~
2

(7r - x) cos nx dx

2 (n7r
= -22 cos -

~ ( ("/2 x sin nx dx + t" (7r - x) sin nx dX) =


tt Jo
J"'/2
00
2 (n7r
= -7r + L
-2- 2cos -

n=l

n 7r

1) 1

+ (-1 t+ -

ti

--j:-tt sin n7r2


ti

cos nx

4
n7r
L -2sin - sin nx
00

f (x) =

n=l

ti

2'"
= - 2'"

30. ao=- 1

tt

tt

(x-7r)dx=-

tt '"

an

tt r

(x - 7r) cos -2n x dx

bn = -1 211"( x 7r 11"

f (x)

7r +

=4

7r

)' sm

4 [ (-1
-2n.-t:

t -cos -n7r]
2

-n x dx = -2 ( -1)n+l - - 4. sm -n7r
2
n
n27r
2

4 [
L
(-1)
n=l n27r
00

- cos -n7r] cos-xn


2
2

508

tt

+ (-1) n+l

1)

Exercises

()x n=l~Z:: (2n- (-1 )n+1 - n4.

n1f)
-2- Slll
tt
2

+ 1,\dX

31. ao = (1 xdx

Jo

an =

. -n X
sm
2

=~
2

101 X COS n:; X dx = n2~2 (COS


1

~1f

1)

1,2.

. n1f
xsm-xdx+

4.

n1f
1sm-xdx=-2

bn=

ino

f (x)

3
00
4 (
n1f
)
nTi
= - + '" -cos - - 1 cos - X
4
e:
n21f2
2
2
n=l

--Slll-

n21f2

32. ao = (11 dx

Jo

~ (4

()

f X = Z::
n=l

n tt

ino

. n1f
l'Slll-xdx+

1,2

3
~
=4+~n21f2
n=l

fx()

f(x)

2 (
n+1
-1)
n1f

1,\2 - x) dx = ~2

Jo

n1f

2sm-+2

2 n+1]' Slll-Xn1f
+ -(-1)
n1f
2

n1f
2

an = (1 1 . cos n1fX dx + 1,2(2 - x) cos n1fX dx

bn=

11.3

tt

n1f
2
(2-x)sin-xdx=-+-22sm2
n1f

(n1f

= 242 (COS ~

cos2+(-1)

n+1)

+ (-1 t+

4.

n1f

1f

1)

n1f
cos2x

00 ( 2
4
n1f)
n1f
= '"
sin sin-x
e: -n1f + -n21f2
2
2

n=l

33. ao

=2

an =.2

bn = 2

+ x)

(\x2

Jo

(x2+x)

dx

=~
3

cos n1fX dx =

2(x2 + x)
11
2 'lo1
sin n1fX - (2x+ 1)sin n1fX dx
n1f
O
n1f O

lo (x2 + x) sin n1fX dx = O

4
+ _[(_l)n

= _(_l)n+1
n1f

f(x)

= -6+

f(x)

= '"
~
n=l

n31f3

00

cosn1fX

-1]

L 22[3(
n=l n tt

00(4 _(_l)n-r) .
n1f

2(x2+x)
n1f

-lt

- 1]cosnsrz

4
+ _[(_l)n
n31f3

1] sin n-rz

509

= 222

2lo1
(2x + 1) costin dx
n1f o

+ -

1f

[3(-l)n-1]

Exercises 11.3

3"4

34. ao = Jo (2x - x ) dx =
2

an

bn

= Jor2 (2x

nt:
x2) cos -xdx
2

la (2x O

2.

tu:

- x )sIn2xdx

lao
1 la

n7l"

bn

27r 2 .
=x SIn nx
7l" O

+ 2:

4
-37l"2

00

n=l

r xdx

=~

4
x 2 cos nx dx = 2
n

7l"

1]cos-x

= -1

f(x)

~7l"2

an

27r

16

1]

n37l"3[1- (-1) ]

7l"

=.!.7l" Jor27rx2 dx =

7l"

n7l"
sin-x

00

n=l

7l"

16
2: 33[1(_l)n]

f(x)

-2 2[(-lt+l

00

n=l

36. ao

= -3 + 2: 2'2[( -lt+l

f(x)

35. ao

(4

dx = --47l"
n
47l")
- -sinnx
n

2cosnx
n

= 7l"

7l" Jo

.:

an = -

7l"

bn

x cos 2nx dx = O

= -217r x sin 2nx dx


7l" O

00
= -7l" + 2:

f(x)

37. ao

(1

--

= -- 1

sin 2nx

n=l

= 2la\X + 1) dx = 3

an=2
.

rl(x+
Jo

1)cos2n7l"xdx=

bn = 2
f(x)

38. ao

la (x + 1) sin 2n7l"xdx
O

3001

=-2

= --

n7l'

2: -

n=l

sin 2n7l"x
n7l"

= 2la2 (2 - x) dx = 2
510

Exercises

an

= 2 102 (2 -

bn

= 2 10

f(x)

x) cos mrx dx

(2 - x) sin mrx dx

= 1+

00

n=l

n7r

=O
=-

n7r

L -sinn7rx

39. We have

bn

= -2lo1r

10
5sin ntdt = -[1(-In
ntt

f(t) =

tt

so that
10[1 - (-l)n]

00

sin nt.

n7r

n=l
00

Substituting the assumption xp(t)

L En sin nt

into the differential equation then gives

n=l

and so

E - 10[1 - (_l)n
n - nst (O
1 - n 2)'

Th
us
10

xp(t)

=-

7r

1 - (_l)n

00

L (
n=l n 10 -

2) SlD nt.

40. We have
b

= -7r2 10O1(1 -

t) sin n nt dt

2
= -n7r

so that
f(t) =

00

n=l

n7r

L-

sin nnt:

00

Substituting the assumption xp(t) =

L En sin nnt

into the differential equation then gives

n=l

and so En

= n7r (O
1

2
2 2)'

- n t:

Thus

2001

xp(t) = -

L n (10 -

7r n=l

511

2 2)
7r

sin ntit.

11.3

Exercises

11.3

41. We have

so that
21f2

f(t) =

(-4)

00

3 + L --;z

cosnt.

n=l

Substituting the assumption

-:fA + f1
00

xp(t) =

Ancosnt

into the differential equation then gives

4' -;

f1 (1

+ 12xp = 6Ao .+

) cos nt = 3
21f2
(-4)
+ J; --;z cos nt

00

An

00

+ 12

-4'n2

1f2

and Ao

9'

42. We have

r1/2

ao = (1/2) Jo

r1/2

f(t)

=- +

an = (1/2) Jo

= 2'

tdt

tcos2n1ftdt

= n21f2 [(_l)n

-1]

so that
1

L
n=l
00

(-1)n-1
n

2 2

cos 2n1ft.

1f

Substituting the assumption

into the differential equation then gives


1
- x~

1
and Ao = 24'

00

+ 12xp = 6Ao + L An(12

- n21f2) cos 2n1ft

n=l

00

=- +L
4

n=l

(-1)n-1
An = n21f2(12 _ n21f2) . Thus

1
xp(t) = 48

+ 1f2

f1
00

(-l)n-1
n2(12 _ n21f2) cos 2n1ft.

512

- 1

(_l)n

2 2

1f

cos 2n1ft

Exercises

11.3

43. (a) The general solution is x(t) = el cos vT6t + e2sin vT6t + xp(t), where
1 - (_l)n .
(
2) sin ni,
7["n=l n 10 - n

10

00

xp(t) = - 2::

The initial condition x(O) = O implies el + xp(O) = O. Since xp(O) = O, we have el = O and

x(t) = e2 sin vT6t + xp(t). Then x'(t) = e2vT6 cos vT6t + x~(t) and x'(O) = Oimplies
10 00 1- (_l)n
e2\l()"+ - 2::
2 cos 0= O.
7["n=l 10 - n
Thus
vT6 00 1- (_l)n
e2 = ---;- 2:: 10 _ n2
n=l
and
10 00 l-(-l)n[l
2::
2
7["n=l 10 - ti

1
]
- sin nt - 117\ sin vT6t .
n
y 10

x(t) = -

(b) The graph is plotted using eight nonzero terms in the series expansion of x(t).
x
4
2

-2

-4

44. (a) The general solution is x(t) = el cos 4V3t + e2sin 4V3t + xp(t), where
7["2

xp(t)=-+162::
18

00

2( 2

n=l n n -

48)cosnt.

The initial condition x(O) = O implies el + xp(O) = 1 or


7["2
el = 1 - xp(O) = 1 - 18 - 16
Now x'(t) = -4V3el sin4V3t+4e2 cos4V3t+x~(t),

E n2(n2 00

48) .

so x'(O) = Oimplies 4e2+X~(0) = o. Since

x~(O) = O, we have e2 = Oand


7["2

00

1)

x(t) = ( 1 - - - 162:: 2( 2
)
18
n=l n n - 48

7["2
00
1
+ 16 2:: 2( 2
8) cosnt
18
n=l n ti - 4

cos4..J3t + -

7["2 (
7["2)
00
1
=-+
1-cos4V3t+162::
2( 2 48)[cosnt-cos4..J3t1.
18
18
n=l n n -

513

Exercises

11.3

(b) The graph is plotted using five nonzero terms in the series expansion of x(t) ..
x
1.5
1
0.5
t

1
-0.5
-1

45. (a) We have

bn = ~ (L woX sin tut x dx = 2wo (-lt+1


L Jo
L
L
mr
so that

w(x)

mr
L -2wo (-lt+ 1 sin -L
x.
00

n=l
rttt

00

(b) If we assume y(x) =

n7r

L En sin L x

then

n=l

and so the differential equation EIy(4)


En

= w(x)

gives

2wo( _l)n+l L 4
EIn57r5

Thus

46. We have

[n7r

2n7r]

2L/3
tvn
2wo
bn = wo sin - x dx = cos - - cos -L L/3
L
tit:
3
3
so that
w(x)
If we assume y(x)

00

2wo [n7r
L
cos n=l nt:
3

n7r

00

L En sin L x

then

n=l

514

2mr]

- cos -

tit:

sin -

x.

Exercses

and so the differential equation Ely(4)

Thus

11.3

= w(x) gives

4 00
n7r
2n7r
_ 2 WoL '" cos 3 - COS-3- . ti n
y (x ) - --5 L
5
sin -L x.
El t: n=l
n

47. We note that w(x)is 2n-periodic and even. With p = tt we find the cosine expansion of

={

f(x)
\Ve have
ao = ~

r f(x)dx

n Jo

an

= -2

=~

r/

n Jo

wodx = Wo

= -2

7r

non

< x < n /2
n/2<x<n.

Wo
0,

f(x) cosnxdx

J7r /2

nn
= -2wo sin -2
.

Wocosnxdx

nn

Thus,
Wo
w(x) = 2

2wo

+-

1 . nn
L
-sm-cosnx.
n=l n
2
00

= Ao/2 + L~=l Ancosnx.

Now we assume a particular solution of the form yp(x)

L~=l Ann4 cos nx and substituting into the differential equation, we obtain
Ely~4)(x)

+ kyp(x) =

kAo
-2-

+ L An(Eln4 + k) cosnx
00

n=l

Wo

2wo

= - +-

1 . nn
L
sm n=l
2
00

cos nx.

ti

Thus
Ao

= Wo

and

_ 2wo sin(nn/2)
n-n
n(Eln4 + k) ,

and
Wo 2wo 00 sin(nn /2)
yp(x) = - + (El 4 k) cosnx.
2k
n n=l n
n + .

48. (a) If f and 9 are even and h(x) = f(x)g(x)


h(-x)

then

= f(-x)g(-x)

= f(x)g(x)

= h(x)

and h is even.
(e) If f is even and 9 is odd and h(x)
h(-x)

= f(x)g(x)

= f(-x)g(-x)

then
= f(x)[-g(x)]

515

-h(x)

Then y~4)(x) =

Exercises 11.3
and h is odd.
(d) Let h(x)

= j(x) g(x) where j and g are even. Then


h( -x) = j( -x) g( -x) = j(x) g(x),

and h is an even function.

(f ) If j is even then

fa

La

j(x) dx =

-10a j( -u) du + Jor j(x) dx = Jol" j(u) du + Jofa j(x) dx

=2

fa j(x)

Jo

dx.

(g) If j is odd then

_aaj(x)

dx = =-

:a

O j(u)

j( -x) dx + loa j(x) dx =

loa j(u)

du + loa j(x) dx

du + loa j(x) dx = O.

jex) is even then j( -x) = j(x). If j(x) is odd then j( -x) = - j(x). Thus, if j(x) is both even
and odd, j(x) = j( -x) = - j(x), and j(x) = O.

49. If

50. For Ely(4l+ky = Othe roots ofthe auxiliary equation are


and m4 = -a - ai, where a = (kj EI)I/4jV2. Thus
Yc

= eCY.x (CI cos a + C2 sin oz) + a -CY.X

We expect y(x) to be bounded as x


bounded as x

-+ -00,

-+ 00,

so we must have

C3

= a+ai,

mI

(C3

so we must have

cos ax
CI

= C4 = O. Thus, Yc

m2

= a-ai,

= C2 = O. We also expect

< 27f.

f (x)

2
1
2

-1

-2
-3

516

y(x) to be

= Oand the solution of the differential

51. The graph is obtained by summing the series from n = 1 to 20. It appears that

X,o<x<7f
-7r,
7r <

-a+ai,

+ C4 sin oz ).

equation in Problem 47 is yp(x).

j(x) = {

m3 =

14

Exercises

= 1 to

52. The graph is obtained by summing the series from n

11.4

10. It appears that

O< x < 1
1 < x < 2.

1- x,
f(x) = { O,
f (x)
1.5
1.25
1
0.75
0.5
0.25

10

-0.25
-0.5

53. (a) The function in Problem 51 is not unique; it could also be


O<x<7r

X,

f(x) =

x=7r

.,

7r

<

<

27r.

(b) The function in Problem 52 is not unique; it could also be


-2 < x < -1

O,
x+l,

-l<x<O

= { -x + 1,

f(x)

O< x

<1

1 < x < 2.

O,

Exercises 11.4
1. For A

:S O the only solution of the boundary-value problem is y = O. For A > O we have


y

CI

cos ~

+ C2 sin ~

x.

Now

y'(x) = -CI ~sin


and y'(O) = Oimplies
y(l)

C2 =

O, so

+ y'(l)

CI(COS~

The eigenvalues are An = x;;' where

v'>..

+ C2~COS~X

~x

Xl,

= O or

~sin~)

X2, X3, ...

cot ~

= V)...

are the consecutive positive solutions of cot V).. =

The corresponding eigenfunctions are cos ~

517

= COS xnx

for n

1, 2, 3, ....

Using a CAS

Exercises

11.4

we find that the first four eigenvalues are 0.7402, )1.7349,41.4388,

and 90.8082 with corresponding

eigenfunctions cos 0.8603x, cos 3.4256x, cos 6.4373x, and cos 9.5293x.
2. For A

< O the only solution of the boundary-value

problem is y = O. For A = O we have y = elX +cz

Now y' = el and the boundary conditions both imply el + e2 = O. Thus, A = O is an eigenvalue
with corresponding eigenfunction YO = x - 1.
For A

> O we have
y = el cos ~ x + e2sin ~

and
y' = -el~sin

~x

+ e2~cos~X.

The boundary conditions imply


el

+ e2~ = O

el cos ~

+ e2sin ~

= O

which gives
-e2~cos~+e2sin~=0

or

tan~=~.

The eigenvalues are An = x;; where Xl, X2, x3, ... are the consecutive positive solutions of
tan /).. = /)...

The corresponding eigenfunctions are

taking e2 = -1 in the first equation of the system.)

.;>:.;;, cos .;>:.;;, x

- sin';>:';;' x (obtained by

Using a CAS we find that the first four

positive eigenvalues are 20.1907, 59.6795, 118.9000, and 197.858 with corresponding eigenfunctions
4.4934 cos 4.4934x-sin 4.4934x, 7.7253 cos 7.7253x-sin 7.7253x, 10.9041 cos 10.9041x-sin 10.9041x,
and 14.0662 cos 14.0662x - sin 14.0662x.
3. For A = O the solution of y" = O is y = elX + e2. The condition y'(O) = O implies el = O, so A = O
is an eigenvalue with corresponding eigenfunction 1.
For A

< Owe have y = el cosh Nx+e2

sin Nx

and y' = el N sinh Nx+e2N

The condi tion y' (O) = O implies e2 = O and so y = el cosh

N x.

cosh Nx.

Now the condition y' (L) = O

implies el = O. Thus y = Oand there are no negative eigenvalues.

> O we have y = el cos ~ x + e2sin /).. x and y' = -el/).. sin ~ x + e2/).. cos ~ x. The
condition y' (O) = O implies e2 = O and so y = el cos /).. x. Now the condition y' (L) = O implies
-el/).. sin /).. L = O. For el -=1= O this condition will hold when ~ L = mr or A = n27f2/ L2, where
For A

n = 1, 2, 3,
n

1, 2, 3,

4. For A

These are the positive eigenvalues with corresponding eigenfunctions cos(n7f/L)x,


.

< Owe have


y =

ci

cosh N x + e2sinh

y' = el N sinh

Nx+
518

Nx

e2N

cosh

N x.

Exercises

11.4

Using the fact that cosh x is an even function and sinh x is odd we have
y(-L)

= clcosh(-~L)

= Clcosh ~

+c2sinh(-~L)
L - C2sinh ~

and
y'(-L)

= cl~sinh(-~L)
= -Cl ~

+c2~cosh(-~L)
L+

sinh ~

cosh ~

C2~

L.

The boundary conditions imply


Clcosh ~

L - C2sinh ~

L = Clcosh

v=>: L + C2sinh v=>: L

or

v=>: L = O

2C2 sinh

and

or

= o.

2Cl~sinh~L
Since
y

R L =J o, Cl = C2= O and

the only solution of the boundary-value problem in this case is

= O.

For A = O we have
y

= ClX

y'
From y(-L)

= Cl

= y(L) we obtain
=ci l.

Then Cl = O and y
For A

+ C2

+ C2= clL + C2

= 1 is an eigenfunction corresponding to the eigenvalue A

> O we have
y = ci
y' = -

COS

ci

,j)._ X + C2sin ,j)._ x

,j)._ sin ,j)._ x +

C2,j)._

cos ,j)._ x.

The first boundary condition implies


ci

cos ,j)._ L - C2sin ,j)._ L = Cl COS ,j)._ L + C2sin ,j)._ L

or
2C2 sin

,j)._ L = O.

519

= O.

Exercises 11.4
Thus, if

el

= O and

e2

i= O,
V).,L=mr

The corresponding eigenfunctions are sin

n; x, for n

= 1, 2,3, ., .. Similarly, the second boundary

condition implies
2q V)., sin V)., L
If

e2 = O and el

= O.

i= O,
n21f2

A = V' n

= 1,2,3, ... ,

n1f

and the corresponding eigenfunctions are cos LX, for n = 1, 2, 3, ....


5. The eigenfunctions are cos A
11cos

where cot

A = Fn.

Thus

>:;. xl1 = la cos >:;. x dx = ~la (1 + cos 2>:;' x)


2

=~

(x + 2y~An sin 2>:;' x)

=~

[1 + 2~

=~

[1 + ~

sin

=~

[1 + ~

(sinFn)

(2 sin

r:

11 = ~

cos

(1 +

~ sin 2>:;')

2y An

>:;.) ]

Fn cot Fn sin >:;.]

6. The eigenfunctions are sin Fn x where tan


11sin

dx

Fn (sin Fn)]

=~

= ~2

(x-

2y

=~

[1-

2~

=~

[1- ~

=~

[1- ~ (- Fn cos

o= ~
2

~sin2Fnx)

x,

11

(2 sin .:
tan

= -An. Thus

>:;. xl1 = la sin Fn x dx = ~ la (1 - cos 2>:;' x)


2

(1 + sin Fn) .

cos

(1-

dx

~sin2Fn)
2y An

>:;.) ]

Fn cos >:;. cos >:;.]


2 -:

520

= ~(

1 + cos Fn) .
2

Exercises
7. (a)

11.4

If,\::; O the initial conditions imply y = O. Por .\ > O the general solution of the Cauchy-Euler
differential equation is y = el cos(.J:\ lnx) + e2sin(.J:\ lnx). The condition y(l)

= O implies

el = O, so that y = e2sin(.J:\ lnx).

n1T,

2, 3, ....

The condition y(5) = O implies .J:\ ln5 =

n = 1,

Thus, the eigenvalues are n21T2/(ln5)2 for n = 1, 2, 3, ... , with corresponding

eigenfunctions sin[(n1T/ In 5) In xJ.


(b) The self-adjoint form is

-d [xy '] + -y
dx
x

= O.

(c) An orthogonali ty relation is


(5 sin (m1T lnx) sin (n1T lnx) dx = O.
In5
ln5

JI

8. (a) The roots of the auxiliary equation m2 + m + ,\ = O are


general solution of the differential equation is el +e2e-x.

i( -1

VI - 4,\). When ,\

= O the

The initial conditions imply el +e2 = O

and el + e2e-2 = O. Since the determinant of the coefficients is not O, the only solution of this
homogeneous system is el = e2 = O, in which case y = O. Similarly, if O
solution is
y' = ele~(-1+v'1-4A)x + e2e~(-I-v'1-4A)x.

< .\ <

!'the general

In this case the initial conditions again imply el = e2 = O, and so y = O. Now, for A

> ~, the

general solution of the differential equation is


y = cle-xj2 cos)4,\

- 1 x + C2e-xj2 sin V4,\ - 1x.

The condition y(O) = O implies el = O so y = c2e-xj2 sin V4,\ - 1x. From


y(2)

= c2e-1

sin 2)4'\ - 1 = O

we see that the eigenvalues are determined by 2V4'\ - 1

= n1T for n

= 1, 2, 3, .. , . Thus,

the eigenvalues are n21T2/42 + 1/4 for n = 1, 2, 3, .. , , with corresponding

eigenfunctions

Cxj2sin~x.

2
(b) The self-adjoint form is

(c) An orthogonality relation is


fo2 eX (e-Xj2 sin ~1T x) (e-Xj2 cos n21Tx) dx = fo2 (sin ~1T x) (cos ~1T x) dx = O.
9. To obtain