Chapter 6.

Laplace Transforms 1

§6.1 Laplace transform. Linearity.

First Shifting Theorem(s-Shifting)
Definition Laplace transform
Laplace transform f (t) (t ≥ 0) 7−→ F (s)
∫ ∞
F (s) = L(f ) = e−stf (t)dt.
0

Inverse transform f (t) = L−1(F ) .

Note ( ) ( −1 )
−1
L L(f ) = f and L L (F ) = F
Example 1. Laplace Transform
Let f (t) = 1(t ≥ 0). Find the Laplace transform of f (t).
Sol. ∫ ∞ [ ]∞
1 1
L(f ) = e−stdt = − e−st = (s > 0).
0 s 0 s
Example 2.Laplace Transform of exponential function eat
Let f (t) = eat(t ≥ 0). Find L(f ).
Sol. [ ]∞
∫ ∞ −(s−a)t
e 1
L(eat) = e−steatdt = = (s > a).
0 a − s s − a
0
Theorem 1. Linearity of the Laplace Transform
For any functions f (t) and g(t) whose transforms exist and any constants
a, b, the transform of af (t) + bg(t) exists, and
( ) ( ) ( )
L af + bg = aL f + bL g .

Proof. By the definition,

∫
( ) ∞
L af + bg = e−st[af (t) + bg(t)]dt
0∫ ∫ ∞
∞
= a e−stf (t)dt + b e−stg(t)dt
0( ) ( ) 0
= aL f + bL g .

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 2

Example 3. Hyperbolic functions

Find the Laplace transforms of
cosh at, sinh at.

Sol.
( eat + e−at )
L(cosh at) = L
2
1[ ]
= L(eat) + L(e−at)
2
1[ 1 1 ] s
= + = 2 .
2 s−a s+a s −a 2

Similarly,
( eat − e−at ) a
L(sinh at) = L = .
2 s2 − a2
Example 4. Cosine and Sine
Derive the formulas
s ω
L(cos ωt) = 2 , L(sin ωt) = .
s + ω2 s2 + ω 2

Sol. (a) Solution by Calculus(Integrating

] by parts).
−st ∞
L(cos ωt) = e−s cos ωt 0 − ωs L(sin ωt) = 1s − ωs L(sin ωt)
−st ]∞
L(sin ωt) = e−s sin ωt 0 + ωs L(cos ωt) = ωs L(cos ωt)
L(cos ωt) = 1s − ωs ωs L(cos ωt) ⇒ L(cos ωt) = s2+ω
s
( ) 2

L(sin ωt) = ωs 1s − ωs L(sin ωt) ⇒ L(sin ωt) = s2+ω ω

2

(b) Solution by Transforms using Derivatives.

(c) Solution by Complex methods(eiωt = cos ωt + i sin ωt).

L(eiωt) = s−iω1 s
= s2+ω ω
2 + i s2 +ω 2

L(eiωt) = L(cos ωt + i sin ωt) = L(cos ωt) + iL(sin ωt).

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 3

Note 1. n = 0, 1, 2, · · ·
∫ ∞
L(tn) = e−sttndt
[0 ] ∫
1 −st n ∞ n ∞ −st n−1
= − e t + e t dt
s 0 s 0
n
= L(tn−1)
s
n n−1
= · L(tn−2) = · · ·
s s
n!
= n+1 .
s

2. a: positive real number

∫ ∞
L(ta) = e−sttadt
∫0 ∞ ( )a
−x x dx
= e (setting x = st)
0 ∫ ∞ s s
1
= a+1 e−xxadx
s 0
1
= a+1 Γ(a + 1).
s

Note Gamma function

∫ ∞
Γ(a) = e−xxa−1dx (a > 0)
0

Γ(a + 1) = aΓ(a),

Γ(n + 1) = n!, n = 1, 2, · · · .

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 4

Theorem 2. First Shifting Theorem(s-Shifting)

If f (t) has the transform F (s) (s > k), then eatf (t) has the transform
F (s − a) where s − a > k.
L{eatf (t)} = F (s − a),
eatf (t) = L−1{F (s − a)}.

Proof.
∫ ∫
∞
−(s−a)t
∞
−st
[ ]
F (s − a) = e f (t)dt = e at
e f (t) dt
0 0
= L{eatf (t)}.
If F (s) exists for s > k, then F (s − a) exists for s − a > k.

Example 5. s-Shifting
From Example 4 and the first shifting theorem, we have
s−a
L(eat cos ωt) = ,
(s − a)2 + ω 2
ω
L(eat sin ωt) =
(s − a)2 + ω 2.
Theorem 3. Existence Theorem
If f (t) is piecewise continuous on t ≥ 0 and satisfies the growth restric-
tion |f (t)| ≤ M ekt for some constants k and M , then L(f ) exists for
all s > k.
Proof. Since f (t) is piecewise continuous, e−stf (t) is integrable over fi-
nite interval on t ≥ 0.
∫ ∞ ∫ ∞
|L(f )| = e−stf (t)dt ≤ |f (t)|e−stdt
∫ 0∞ 0
M
≤ M ekte−stdt = (s > k).
0 s−k
2
Note et does not satisfy the growth restriction condition. The conditions
of Theorem 3 are sufficient rather than necessary.
Computational Science & Engineering (CSE) T. Jeong
 Chapter 6. Laplace Transforms 5

§6.2 Transforms of Derivatives and Integrals. ODEs

Theorem 1. Laplace Transform of Derivatives
f (t) : continuous for all t ≥ 0, |f (t)| ≤ M ekt for some k and M
f ′(t) : piecewise continuous on every finite interval on t ≥ 0
=⇒ ∃ L(f ′) when s > k and L(f ′) = sL(f ) − f (0).

Theorem 2. Laplace Transform of the Derivative f (n)

L(f (n)) = snL(f ) − sn−1f (0) − sn−2f ′(0) − · · · − f (n−1)(0)
Note
L(f ′′) = sL(f ′) − f ′(0) = s[sL(f ) − f (0)] − f ′(0)
= s2L(f ) − sf (0) − f ′(0).
Proof of Theorem 1.
∫ ∞ ∫
[ ] ∞
∞
L(f ′) = e−stf ′(t)dt = e−stf (t) 0 + s e−stf (t)dt
0 0
= −f (0) + sL(f ) when s > k.
Example 2.
Derive the formulas
L(cos ωt) = s2+ω
s
2 , L(sin ωt) = s2 +ω 2 .
ω

Sol. Let f (t) = cos ωt. Then f ′(t) = −ω sin ωt and f ′′(t) =
−ω 2 cos ωt = −w2f (t).
L(f ′′) = s2L(f ) − sf (0) − f ′(0) = s2L(f ) − s
= −ω 2L(f )
s
∴ L(cos ωt) = 2 .
s + ω2
1
L(sin ωt) = − L(f ′)
ω
1 ω
= − [sL(f ) − 1] = 2 .
ω s + ω2

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 6

[Laplace Transform of the Integral of a function]

Theorem 3. Laplace Transform of the integral
Let f (t) be piecewise continuous for t ≥ 0 and satisfies |f (t)| ≤ M ekt
for some k and M . Then, for s > 0, s > k, and t > 0,
{∫ t }
1
L f (τ )dτ = F (s)
0 s
thus ∫ t { }
1
f (τ )dτ = L−1 F (s) .
0 s
∫t
Proof. Denote g(t) = 0 f (τ )dτ . Since f (t) is piecewise continuous,
g(t) is continuous and
∫ t ∫ t
M kt
|g(t)| ≤ |f (τ )|dτ ≤ M ekτ dτ = (e − 1)
0 0 k
M kt
≤ e (k > 0).
k
This shows that g(t) satisfies a growth restriction. Also, g ′(t) = f (t),
except at points at which f (t) is discontinuous. Hence g ′(t) is piecewise
continuous on each finite interval. Therefore
L(f ) = L(g ′) = sL(g) − g(0) = sL(g).
Example 3
1 1
Find the inverse of and .
s(s2 + ω 2) s(s2 + ω 2)
ω
Sol. Recall that L(sin ωt) = 2 . Then
s + ω2
{ ( )} ∫ t
1 1 1 1
L−1 = sin ωτ dτ = (1 − cos ωt),
s s2 + ω 2 ω 0 ω2
and integrating this result again, we obtain
{ ( )} ∫ t
1 1 1 ωt − sin ωt
L−1 = 1 − cos ωτ dτ = .
s2 s2 + ω 2 ω2 0 ω3

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 7

[Initial Value Problem]

y ′′ + ay ′ + by = r(t), y(0) = K0, y ′(0) = K1

Laplace’s method
Step 1. Setting up the subsidiary equation
Let Y = L(y) and R = L(r). Then
[s2Y − sy(0) − y ′(0)] + a[sY − y(0)] + bY = R,
we have the subsidiary equation
(s2 + as + b)Y = (s + a)y(0) + y ′(0) + R
Step 2. Solution of the subsidiary equation by algebra.
Putting
1
Q(s) = 2 ,
s + as + b
this give the solution
Y (s) = [(s + a)y(0) + y ′(0)]Q(s) + R(s)Q(s).

In particular, if y(0) = y ′(0) = 0, then

Y = RQ.
Note
Y L(output)
Q= =
R L(input)

Step 3 Inversion of Y to obtain y.

We obtain the solution of the ODE
y(t) = L−1(Y ).

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 8

Example 4. Initial value problem

Solve y ′′ − y = t, y(0) = 1, y ′(0) = 1.

1
Sol. (s2Y − s − 1) − Y = 1
s2
, thus (s2 − 1)Y = s + 1 + .
s2
( )
1 1 1
Y = + −
s−1 s2 − 1 s2

−1 e3t − e−t
∴ y(t) = L (Y ) = e + sinh t − t =
t
− t.
2
Example 6. Shifted Data Problems
′′
√
Solve y + y = 2t, y( π4 ) = 1
2
π, y ′( π4 ) = 2 − 2.

Sol. Set t̃ = t − π4 . Note that dỹ

dt̃
= dy dt
dt dt̃
= dy
dt
. Then
′′ π π ′
√
ỹ + ỹ = 2(t̃ + ), ỹ(0) = , ỹ (0) = 2 − 2
4 2
where ỹ(t̃) = y(t).
π √ 2 π
s Ỹ −
2
s − (2 − 2) + Ỹ = + 2
.
2 s2 s
Thus,
√
2 π
2
π
2
s 2− 2
Ỹ = + + +
(s2 + 1)s2 s(s2 + 1) s2 + 1 s2 + 1
√
1 1 π 1 s π
s + (2 − 2)
= 2[ 2 − 2 ]+ [ − 2 ]+ 2
.
s s +1 2 s s +1 s2 + 1
π π √
ỹ(t̃) = 2(t̃ − sin t̃) + (1 − cos t̃) + cos t̃ + (2 − 2) sin t̃
2 2
π √
= 2t̃ + − 2 sin t̃.
2
√ π
∴ y(t) = 2t − 2 sin(t − ) = 2t − sin t + cos t.
4

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 9

§6.3 Unit Step Function(Heaviside Function).

Second Shifting Theorem(t−Shifting)
[Unit Step Function]
Definition
{
0, if t < a
u(t − a) = (a ≥ 0)
1, if t > a
unit step function or Heaviside function

Fig. 120. Effects of the unit step function:

(A) Given function. (B) Switching off and on. (C) Shift

Fig. 121. Use of many unit step functions

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 10

Theorem 1. Shifting Theorem

L{f (t − a)u(t − a)} = e−asF (s)

or
f (t − a)u(t − a) = L−1{e−asF (s)}
e−as
Note L{u(t − a)} = .
s
Proof.
∫ ∞ ∫ ∞
e−asF (s) = e−as e−sτ f (τ )dτ = e−s(τ +a)f (τ )dτ
∫ ∞ 0 0

= e−stf (t − a)dt (setting τ + a =)

∫a∞
= e−stf (t − a)u(t − a)dt
0
= L{f (t − a)u(t − a)}.
Example 1.
Write the following function using step functions and find its Laplace trans-
form. 

 2, 0 < t < 1
1 2
f (t) = t , 1 < t < π2 .
 2
 cos t, t > π2
Sol. In terms of unit step functions,
[ ]
1 π
f (t) = 2[1 − u(t − 1)] + t2 u(t − 1) − u(t − )
[ ]2 2
π
+ cos t u(t − ).
2
First,
2 2
L(2[1 − u(t − 1)]) = − e−s, (i)
s s

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 11

{ } { }
1 1
L t2u(t − 1) = L [(t − 1) + 1]2u(t − 1)
2 {[2 ] }
1 1
= L (t − 1)2 + (t − 1) + u(t − 1)
( 2 ) 2
1 1 1 −s
= + + e . (ii)
s3 s2 2s
Similarly,
{ } ( 2)
1 π 1 π π − πs
L t2u(t − ) = 3
+ 2
+ e 2 (iii)
2 2 s 2s 8s
{ } { }
π π π
L cos t u(t − ) = L − sin(t − )u(t − )
2 2 2
1 πs
= − 2 e− 2 (iv)
s +1
∴ L(f ) = (i) + (ii) + (iii) + (iv) = · · · .
Note
L{f (t)u(t − a)} = L{f ((t + a) − a)u(t − a)}

= e−asL{f (t + a)}.
Ex.
1 1
L{ t2u(t − 1)} = e−sL{ (t + 1)2}
2 2
1
= e−sL{ (t2 + 2t + 1)}
2
−s 1 1 1
= e ( + + ).
s3 s2 2s

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 12

Example 2. Inverse Transform

Find the inverse transform f (t) of
e−s e−2s e−3s
F (s) = + + .
s2 + π2 s2 + π2 (s + 2)2

Sol. Note that

1 1
L−1( )= sin(πt),
s2 + π 2 π
1 1
L−1( ) = t, L−1
( ) = te−2t
.
s2 (s + 2)2

Using the formula

L−1{e−asF (s)} = f (t − a)u(t − a),
we have
1 1
f (t) = sin[π(t − 1)]u(t − 1) + sin[π(t − 2)]u(t − 2)
π π

+(t − 3)e−2(t−3)u(t − 3)


 0, 0<t<1







 − π1 sin πt, 1<t<2
= .



 0, 2<t<3






(t − 3)e−2(t−3), t>3

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 13

Example 3. Response of an RC-circuit to single rectangular wave

Find the current i(t) in the RC-circuit in Fig. if a single rectangular wave
with voltage V0 is applied.

Fig. 124. RC-circuit, electromotive force v(t), and current

Sol. Input V0[u(t − a) − u(t − b)].

model
∫
1 t
Ri(t) + i(τ )dτ = V0[u(t − a) − u(t − b)].
C 0
Subsidiary equation
V0
I(s) V0 −as −bs −as
RI(s)+ = [e −e ] ⇒ I(s) = R
1 (e −e−bs).
Cs s s+ RC

Note that
V0
−1 V0 t
L ( R
1 )= e− RC .
s+ RC
R
Thus
V0 t−a t−b
i(t) = [e− RC u(t − a) − e− RC u(t − b)]
R


 0, 0<t<a
V0 − RCt−a
= e , a<t<b .

 V0 − RC
t
R
a b
e [e RC − e RC ], t>b
R

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 14

§6.4 Dirac’s Delta Function

{ 1
, a≤t≤a+k
Let fk (t − a) = k .
0, otherwise

Fig. 132. The function fk (t − a)

Note
1. In mechanics, the integral of a force acting over [a, a + k] is called the
impulse of the force.
The impulse of fk (t − a) is
∫ ∞ ∫ a+k
1
fk (t − a)dt = dt = 1
0 a k
2.
δ(t − a) = lim fk (t − a)
k→0
Dirac delta function or unit impulse function
{ ∫ ∞
∞ if t = a
3. δ(t − a) = , δ(t − a)dt = 1.
0 otherwise 0

4.
L{δ(t − a)} = lim L{fk (t − a)}
k→0
1
= lim L{ [u(t − a) − u(t − (a + k))]}
k→0 k
1 −as
= lim [e − e−(a+k)s]
k→0 ks

= e−as.

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 15

Example 2. Hammer-blow Response of a Mass-Spring System

Find the response of the damped mass-spring system(see Sec. 2.8) with a
unit impulse at time t = 1.
y ′′ + 3y ′ + 2y = δ(t − 1), y(0) = 0, y ′(0) = 0.

Sol. Subsidiary equation s2Y + 3sY + 2Y = e−s

and ( )
1 1 1
Y = e−s = − e−s.
(s + 1)(s + 2) s+1 s+2
Note that
1 1
L−1( − ) = e−t − e−2t = f (t).
s+1 s+2
Hence
{( ) }
1 1
y(t) = L−1 − e−s
s+1 s+2

= f (t − 1)u(t − 1)

= (e−(t−1) − e−2(t−1))u(t − 1)
{
0, 0≤t<1
= .
e−(t−1) − e−2(t−1), t>1

Fig. 134. Response to a hammerblow

Note Replacing δ(t − 1) by u(t − 1) − u(t − 2), y(t) ?(Example 1)

Computational Science & Engineering (CSE) T. Jeong
 Chapter 6. Laplace Transforms 16

[ More on Partial Fractions ]

F (s)
ODE =⇒ Y (s) =
G(s)
Consider polynomials F (s) and G(s).
[degree of F (s)< degree of G(s).(why!)]
The quotient of polynomials can be expressed by a sum of partial fractions.

F (s)
Y (s) =
G(s)
F (s)
=
(s − a) · · · (s − b)n · · · (s2 + As + B)m · · ·
a1
=
s−a

+···

bn bn−1 b1
+ + + ··· +
(s − b)n (s − b)n−1 s−b

+···

cms + dm cm−1s + dm−1

+ +
(s2 + As + B)m (s2 + As + B)m−1
c1s + d1
+··· +
s2 + As + B

+··· .
Then we can obtain the inverse transform y(t) of Y (s).

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 17

Example 4. Damped forced vibration

Solve the IVP for a damped mass-spring system acted upon by a sinusoidal
force for some time interval {
10 sin 2t, 0 < t < π
y ′′ + 2y ′ + 2y = r(t), r(t) = ,
0, t>π
y(0) = 1, y ′(0) = −5.
Sol. Note that
L(r) = L(10 sin 2t[1 − u(t − π)])
20
= 2 − e−πs L(10 sin(2(t + π)))
s +4
20
= 2 (1 − e−πs).
s +4
Subsidiary equation
20
(s2Y − s + 5) + 2(sY − 1) + 2Y = 2 (1 − e−πs ),
s +4
(s2 + 2s + 2)Y = 20
s2 +4
(1 − e−πs ) + (s − 3),
20 s−3
Y = (1 − e−πs) + .
(s2 + 4)(s2 + 2s + 2) 2
|s + {z
2s + 2}
| {z }
III
| I
{z }
II

As + B Cs + D −2s − 2 2(s + 1) + 4
I= + 2 = 2 +
s2 + 4 s + 2s + 2 s +4 (s + 1)2 + 1
=⇒ L−1{I} = −2 cos 2t − sin 2t + e−t(2 cos t + 4 sin t).
Since II = I × e−πs ,
L−1{II} = [−2 cos(2(t − π)) − sin(2(t − π))]u(t − π)
+e−(t−π)[2 cos(t − π) + 4 sin(t − π)]u(t − π).
{ }
(s + 1) − 4
L−1{III} = L−1 = e−t
(cos t − 4 sin t).
(s + 1)2 + 1
∴ y(t) = L−1{I} + L−1{II} + L−1{III} = · · · .

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 18

§6.5 Convolution. Integral Equations

∫ t
convolution of f and g : (f ∗ g)(t) = f (τ )g(t − τ )dτ
0

Theorem 1. Convolution Theorem

If L(f ) = F (s) and L(g) = G(s), then
L(f ∗ g) = F (s)G(s)
or L−1{F (s)G(s)} = (f ∗ g)(t).
Proof. Setting p = t − τ , we have
∫ ∞
G(s) = e−spg(p)dp
∫0 ∞
= e−s(t−τ )g(t − τ )dt
τ ∫
∞
= e sτ
e−stg(t − τ )dt.
τ

The relation gives

∫ ∞ ∫ ∞
−sτ
F (s)G(s) = e f (τ )dτ G(s) = e−sτ G(s)f (τ )dτ
∫0 ∞ [ ∫ ∞ 0 ]
= f (τ ) e−stg(t − τ )dt dτ
∫0 ∞ [∫ ∞ τ ]
= e−stf (τ )g(t − τ )dt dτ.
0 τ

Fig. 141. Region of integration in the tτ -plane

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 19

By reversing the order of integration, we have

∫ ∞ [∫ t ]
F (s)G(s) = e−stf (τ )g(t − τ )dτ dt
∫0 ∞ 0 [∫ t ]
= e−st f (τ )g(t − τ )dτ dt
∫0 ∞ 0

= e−st(f ∗ g)(t)dt = L{(f ∗ g)(t)}.

0

Note L(f g) ̸= L(f )L(g)

Ex. Let f (t) = et and g(t) = 1. Then f g = f and

1 1
L(f ) = and L(g) = .
s−1 s
Thus
1 1
L(f )L(g) = ̸= L(f g) = L(f ) = .
s(s − 1) s−1
Example 2. Convolution
1
Let H(s) = (s2+ω 2 )2 . Find h(t) = L−1(H(s)).
{ sin ωt }
Sol. Note that L ω
= 1
s2 +ω 2
.
Thus
sin ωt sin ωt
h(t) = ∗
ω∫ ω
1 t
= sin ωτ sin ω(t − τ )dτ
ω2
∫
0
t
1
= [− cos ωt + cos ω(2τ − t)] dτ
2ω 2 [0
[ ]t ]
1 sin ω(2τ − t)
= −t cos ωt +
2ω 2 2ω 0
[ ]
1 sin ωt
= −t cos ωt + .
2ω 2 ω

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 20

[Properties]
f ∗g = g∗f
f ∗ (g1 + g2) = f ∗ g1 + f ∗ g2
f ∗ (g ∗ v) = (f ∗ g) ∗ v
f ∗0 = 0∗f =0
Example 3.
Show that (a) f ∗ 1 ̸= f in general, and (b) f ∗ f ≥ 0 may not hold.
∫t
Sol. (a) t ∗ 1 = 0 τ · 1 τ = 12 t2 ̸= t.

(b) sin t ∗ sin t = − 12 t cos t + 12 sin t. (see Example 2.)

[ Application to Nonhomogeneous Linear ODEs]

y ′′ + ay ′ + by = r(t)

subsidiary equation of the ODE

Y = [(s + a)y(0) + y ′(0)]Q(s) + R(s)Q(s)
where R(s) = L(r) and Q(s) = 1/(s2 + as + b) ≡ L(q).

If y(0) = 0 and y ′(0) = 0, then

Y = RQ.
Thus the solution is
y(t) = (q ∗ r)(t) [or (r ∗ q)(t)]
∫ t
= q(t − τ )r(τ )dτ.
0

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 21

Example 5. Response of a damped vibrating system to a single square wave

Using convolution, determine the response of the damped mass-spring system
modeled by
y ′′ + 3y ′ + 2y = r(t),
{
1, 1 < t < 2
where r(t) = , y(0) = y ′(0) = 0.
0, otherwise
Sol. Since y(0) = y ′(0) = 0, we have
Y = RQ,
where R = L(r) and
1 1 1
Q = L(q) = = − .
s2 + 3s + 2 s+1 s+2
Note that q(t) = e−t − e−2t. Thus
∫ t
y(t) = (q ∗ r)(t) = q(t − τ )r(τ )dτ
∫0 t
= {e−(t−τ ) − e−2(t−τ )}r(τ )dτ.
0

(i) t < 1 : y(t) = 0.

(ii) 1 < t < 2 :

∫ t
y(t) = (e−(t−τ ) − e−2(t−τ ))dτ
1
1 1
= − e−(t−1) + e−2(t−1).
2 2
(iii) t > 2 :
∫ 2 ( )
y(t) = e−(t−τ ) − e−2(t−τ ) dτ
1
1
= e−(t−2) − e−(t−1) − [e−2(t−2) − e−2(t−1)].
2

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 22

[Integral Equations ]
Example 6. A Volterra Integral Equation of the Second Kind
Solve the Voterra Integral Equation
∫ t
y(t) − y(τ ) sin(t − τ ) dτ = t.
0

Sol. The given equation can be written as a convolution,

y − y ∗ sin t = t,
and we have
1 1 s2 + 1 1 1
Y (s) − Y (s) · = hence Y (s) = = + .
s2 + 1 s2 s4 s2 s4
The solution is
t3
y(t) = t + .
6

Example 7. A Volterra Integral Equation of the Second Kind

Solve the Voterra Integral Equation
∫ t
y(t) − (1 + τ )y(t − τ )dτ = 1 − sinh t.
0

Sol. The given equation can be written as a convolution,

y − (1 + t) ∗ y = 1 − sinh t,
and we have
1 1 1 1 s
Y (s) − ( + 2 )Y (s) = − 2 hence Y (s) = 2 .
s s s s −1 s −1
The solution is
y(t) = cosh t.

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 23

§6.6 Differentiation and Integration of Transforms.

ODEs with Variable Coefficients
[Differentiation of Transforms]
∫ ∞
F (s) = L(f (t)) = e−stf (t)dt
0
∫ ∞
′ dF (s)
=⇒ F (s) = =− e−st[tf (t)]dt.
ds 0

∴ L(tf (t)) = −F ′(s), L−1(F ′(s)) = −tf (t)

Example 1 Differentiation of Transforms

Derive the formulas
2βs s2 − β 2
L(t sin βt) = , L(t sin βt) = .
(s2 + β 2)2 (s2 + β 2)2

Sol.
d
L{t sin βt} = − L(sin βt)
ds
d[ β ]
= −
ds s2 + β 2
2βs
= ,
(s2 + β 2)2

d
L{t cos βt} = − L(cos βt)
ds
d[ s ]
= −
ds s2 + β 2
s2 − β 2
= .
(s2 + β 2)2

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 24

[ Integration of Transforms ]
{ } ∫ ∞ [∫ ∞ ]
f (t) f (t)
L = F (s̃)ds̃, L−1 F (s̃)ds̃ =
t s s t
Proof.
∫ ∞ ∫ ∞ [∫ ∞ ]
F (s̃)ds̃ = e−s̃tf (t)dt ds̃
s s 0
∫ ∞ [∫ ∞ ]
= f (t) e−s̃tds̃ dt
0 s
∫ ∞ f (t)
= e−st dt
0 t
{ }
f (t)
= L .
t

Example 2 Differentiation and Integration of Transforms

{ ( 2
)}
Find the inverse transform of L−1 ln 1 + s2 .
ω

Sol.(Method 1) The derivative of F (s)

( )
ω2
F (s) = ln 1 + 2 = ln(s2 + ω 2) − 2 ln s
s
is
2s 2
F ′(s) = − ,
s2 + ω 2 s
and
−tf (t) = L−1{F ′(s)} = 2 cos ωt − 2.
Hence the inverse of f (t) is
2
f (t) = (1 − cos ωt).
t

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 25

(Method 2)
( )
d ω2 d
− ln 1 + 2 = − [ln(s2 + ω 2) − 2s]
ds s ds
2 2s
= − := G(s)
s s2 + ω 2
∫ ∞( )
ω2
=⇒ G(s̃)ds̃ = ln 1 + 2
s s
Since g(t) = L−1(G) = 2 − 2 cos ωt,
{ ( 2 )} {∫ ∞ }
ω
L−1 ln 1 + 2 = L−1 G(s̃)ds̃
s s
g(t)
=
t
2
= (1 − cos ωt).
t

Example 3. Laguerre’s Equation

Solve ty ′′ + (1 − t)y ′ + ny = 0.

Sol. Using
′ d
L(ty ) = − [sY − y(0)]
ds
dY
= −Y − s ,
ds
d
L(ty ′′) = − [s2Y − sy(0) − y ′(0)]
ds
dY
= −2sY − s2 + y(0),
ds
we
[ have ]
dY dY
−2sY − s2 + y(0) + [sY − y(0)] + [Y + s ] + nY = 0.
ds ds

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 26

Simplification gives
dY
(s − s2) + (n + 1 − s)Y = 0
ds
and ( )
dY n+1−s n n+1
=− ds = − ds.
Y s − s2 s−1 s
Integrating, we get
ln Y = n ln(s − 1) − (n + 1) ln s + c∗
and taking c∗ = 0
(s − 1)n
Y (s) = .
sn+1
Note that
n! n!
L(tn) = n+1
, L(tne−t) = .
s (s + 1)n+1
Putting f (t) = tne−t,
L(f (n)) = snL(f ) − sn−1f (0) − sn−2f ′(0) − · · · − f (n−1)(0)

= snL(f )

n!sn
= := F (s).
(s + 1)n+1

Since Y (s) = 1
n!
F (s − 1), we have

−1 et
y(t) = L (Y ) = L−1(F )
n!
et
= f (n)(t)
n!
et dn n −t
= (t e ).
n! dtn
The solution is cy(t). (Laguerre polynomials)
Computational Science & Engineering (CSE) T. Jeong
 Chapter 6. Laplace Transforms 27

§6.7 Systems of ODEs

Consider a first-order linear system with constant coefficients
y1′ = a11y1 + a12y2 + g1(t)
y2′ = a21y1 + a22y2 + g2(t).
Writing
Y1 = L(y1), Y2 = L(y2), G1 = L(g1), G2 = L(g2),
we obtain from the system
sY1 − y1(0) = a11Y1 + a12Y2 + G1(s)
sY2 − y2(0) = a21Y1 + a22Y2 + G2(s).
By solving this system algebraically for
Y1(s) and Y2(s)
and taking the inverse transform, we obtain the solution
y1(t) = L−1(Y1), y2(t) = L−1(Y2).

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 28

Example 1 Mixing Problem Involving Two Tanks

Initially
T1: 100 gal of pure water, T2: 100 gal of water containing 150 lb of salt.
Find the salt contents y1(t) and y2(t) in T1 and T2, respectively.

Fig. 144. Mixing problem

Sol.

 ′
 y1 = − 100 y1 +
8 2
y +6
100 2
, y1(0) = 0, y2(0) = 150.

 y′ =
2
8
y
100 1
− 8
y ,
100 2

So, 

 sY1 = − 100
8
Y1 + 2
Y + 6
100 2 s


 sY − 150 =
2
8
Y
100 1
− 8
Y
100 2
Thus,


 Y1 =
9s+0.48
s(s+0.12)(s+0.04)
= 100
s
− 62.5
s+0.12
− 37.5
s+0.04


Y = 150s2 +12s+0.48
2 s(s+0.12)(s+0.04)
= 100
s
+ 125
s+0.12
− 75
s+0.04
.



 y1 = 100 − 62.5e
−0.12t
− 37.5e−0.04t
∴

 y = 100 + 125e−0.12t − 75e−0.04t
2

Computational Science & Engineering (CSE) T. Jeong

 Chapter 6. Laplace Transforms 29

Example 3. Model of Two Masses on Springs

The mechanical system in Fig.144 consists of two bodies of mass 1 on three
springs of the same spring constant k and of negligibly small masses of the
springs. The model of the physical system is the system of ODEs
y1′′ = −ky1 + k(y2 − y1),
y2′′ = −k(y2 − y1) − ky2.
Here y1 and y2 are the displacements of the bodies from their positions of
static equilibrium.
Determine the solution corresponding to the initial conditions
′
√ ′
√
y1(0) = 1, y2(0) = 1, y1(0) = 3k, y2(0) = − 3k.

Fig. 146. Two masses on springs

√
s Y1 − s − 3k = −kY1 + k(Y2 − Y1)
2
Sol. √
s Y2 − s + 3k = −k(Y2 − Y1) − kY2
2

where L(y1), Y2 = L(y2).

Thus, { √
s 3k
Y1 = s2+k + s2+3k
√
Y2 = − s2+3k
s
s2 +k
3k

{ IVP
Hence the solution of the is √ √
y1 = cos kt + sin 3kt
√ √
y2 = cos kt − sin 3kt.

Computational Science & Engineering (CSE) T. Jeong