staffweb.itsligo.ie · contents 1 mathematics review ce1h 2 1.1 partial fractions . . . . . . . . ....
TRANSCRIPT
Mathematics
Leo Creedon
1-st year Civil Engineering (Level 8)
February 23, 2009
Contents
1 Mathematics Review CE1H 21.1 Partial Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . 151.2 Integration using Trigonometric Substitutions and other tricks 22
2 First Order Differential Equations CE1H 272.1 Separable Differential Equations . . . . . . . . . . . . . . . . . 282.2 First Order Linear Differential Equations (The Integrating Fac-
tor Method) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352.3 Homogeneous Differential Equations . . . . . . . . . . . . . . 44
3 Laplace Transforms CE1H 503.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513.2 Proof by Induction . . . . . . . . . . . . . . . . . . . . . . . . . 523.3 Inverse Laplace Transforms . . . . . . . . . . . . . . . . . . . . 583.4 Solving Differential Equations using Laplace Transforms . . . 66
4 Second Order Differential Equations 764.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5 Complex Numbers 93
6 Matrices 1026.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1036.2 Addition and Subtraction of matrices . . . . . . . . . . . . . . . 1046.3 Scalar Multiplication . . . . . . . . . . . . . . . . . . . . . . . . 1066.4 Zero Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
1
CONTENTS 2
6.5 Multiplication of Matrices . . . . . . . . . . . . . . . . . . . . . 1076.6 Transpose of a Matrix . . . . . . . . . . . . . . . . . . . . . . . . 1156.7 Identity Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . 1166.8 The Inverse of a 2× 2 matrix . . . . . . . . . . . . . . . . . . . . 1176.9 The inverse of a 3× 3 matrix . . . . . . . . . . . . . . . . . . . . 1216.10 Solving equations involving Matrices . . . . . . . . . . . . . . 1246.11 Cramer’s Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1346.12 Eigenvalues and Eigenvectors . . . . . . . . . . . . . . . . . . . 139
7 Partial Differentiation 146
Bibliography
Chapter 1
Mathematics Review CE1H
3
CHAPTER 1. MATHEMATICS REVIEW CE1H 4
Exercise 1.1 12 + 1
6 =
Exercise 1.2 12 −
13 =
Exercise 1.3 12.
16 =
Exercise 1.4 12 ÷
16 =
Exercise 1.5 ab + c
d =
Exercise 1.6 ab −
cd =
Exercise 1.7 ab .
cd =
Exercise 1.8 ab ÷
cd =
Exercise 1.9 (ab)c =
Exercise 1.10 (23)2 =
Exercise 1.11 2−3 =
Exercise 1.12 ln(ab) =
Exercise 1.13 ln(ab) =
Exercise 1.14 ln(ab) =
Exercise 1.15 ln(e) =
Exercise 1.16 ln(1) =
CHAPTER 1. MATHEMATICS REVIEW CE1H 5
Exercise 1.17 eln x =
Exercise 1.18 ln(ex) =
Exercise 1.19 e0 =
Exercise 1.20 Draw a rough graph of y = ex:
Exercise 1.21 Draw a rough graph of y = ln(x):
Exercise 1.22 Draw a rough graph of y = x2:
Exercise 1.23 Draw a rough graph of y = x2 + 1:
CHAPTER 1. MATHEMATICS REVIEW CE1H 6
Exercise 1.24 Draw a rough graph of y = (x + 1)2:
Exercise 1.25 ddxx
2 =
Exercise 1.26 ddx(3x6 − 10x5 + x− 1) =
Exercise 1.27 ddx(√
x) =
Exercise 1.28 ddx(5x4 + 3x2 + 10x + 2 +
√x + 3x
32) =
Exercise 1.29 ddx sin x =
Exercise 1.30 ddt(t
−12) =
Exercise 1.31 ddx(ln(x)) =
CHAPTER 1. MATHEMATICS REVIEW CE1H 7
Exercise 1.32 ddx(x cos x) =
Exercise 1.33 ddx(sin x cos x) =
Exercise 1.34 ddx[(3x2 + 5x + 2)(cos x)] =
Exercise 1.35 ddx
(cos x2x
)=
Exercise 1.36 ddx
3x2−7x+25x2+1
=
Exercise 1.37 ddx sin(2x) =
CHAPTER 1. MATHEMATICS REVIEW CE1H 8
Exercise 1.38 ddx
√3x =
Exercise 1.39 ddx[(cos x)100] =
Exercise 1.40 ddx[(3x2 + 5x + 2)−1] =
Exercise 1.41 ddx cos(3x2 + 2x− 10) =
Exercise 1.42 ddx2 tan(
√x) =
Exercise 1.43 ddx
√cos(2x) =
Exercise 1.44 Use implicit differentiation to find dydx, where
xy + sin y = 0:
CHAPTER 1. MATHEMATICS REVIEW CE1H 9
Exercise 1.45 Use implicit differentiation to find dydx, where
y +√
y = x. Hence find the slope of the tangent line to the
curve when y = 4 (i.e. find dydx when y = 4).
Exercise 1.46 ddx ln(5x2 + 2) = ...
Exercise 1.47 Find dydx, where y = (10x2+1)100(2x−1)9
(x3+1)11
(i) by direct differentiation
(ii) by taking ln of both sides and then differentiating (log-
arathmic differentiation)
CHAPTER 1. MATHEMATICS REVIEW CE1H 10
CHAPTER 1. MATHEMATICS REVIEW CE1H 11
Review of Integration:∫ b
a f (x)dx = the area under the curve f (x), between a
and b. Also, if∫
f (x)dx = g(x) + C, (a function) thenddxg(x) = f (x). In fact
∫ b
a f (x)dx = [g(x)]ba = g(b)−g(a),
a number. This is called the Fundamental Theorem
of Calculus.
Exercise 1.48∫
[f (x) + g(x)]dx =
Exercise 1.49∫
kf (x)dx =
Exercise 1.50∫ a
b f (x)dx =
Exercise 1.51∫ a
a f (x) =
Exercise 1.52∫
x2dx =
Exercise 1.53∫ 1
0 x2dx =
Exercise 1.54∫
(3x5 − 6x3 + 2x2 − x + 5)dx =
Exercise 1.55∫ π
20 sin xdx =
Exercise 1.56∫
(cos x + tan x)dx =
CHAPTER 1. MATHEMATICS REVIEW CE1H 12
Exercise 1.57∫ √
sds =
Exercise 1.58∫ e
11xdx =
Exercise 1.59∫
( 1√x
+ 6x2 + 3x4 − x−3 + 2x− 10)dx =
Exercise 1.60 Find I =∫
cos(3x)3dx using the substitu-
tion u = 3x.
Exercise 1.61 Find I =∫ √
3x5 + x(15x4 + 1)dx, using
the substitution u = 3x5 + x.
Exercise 1.62 Find I =∫
(6x2 + sin x)32(12x + cos x)dx
using the substitution u =
CHAPTER 1. MATHEMATICS REVIEW CE1H 13
Exercise 1.63 Find I =∫
sin(3x2 − x)(6x− 1)dx.
Exercise 1.64 Find I =∫ 4
0 ex22xdx.
Exercise 1.65 Find I =∫
sin(2x)dx, by letting u =
Exercise 1.66 Use integration by parts to find I =∫
x sin xdx.
Exercise 1.67 Find I =∫
x cos xdx.
CHAPTER 1. MATHEMATICS REVIEW CE1H 14
Exercise 1.68 Find I =∫ e
1 xlnxdx.
Exercise 1.69 Find I =∫
lnxx dx.
CHAPTER 1. MATHEMATICS REVIEW CE1H 15
Exercise 1.70 Find I =∫
lnxdx.
Exercise 1.71 Find I =∫ 3
22x−1x2−x
dx.
Exercise 1.72 Find I =∫
xexdx.
CHAPTER 1. MATHEMATICS REVIEW CE1H 16
1.1 Partial Fractions
A rational function is the quotient of two polynomi-
als. e.g. f (x) = 3x2+2x−15x4+x+7
and g(x) = 1x are rational
functions.
Given a rational function, we can often rewrite it as
the sum of a few easier fractions (the partial fractions
decomposition of a rational function).
Roots of Quadratic Functions
Given a quadratic function f (x) = ax2 + bx + c, its
roots are x = −b±√
b2−4ac2a .
If b2 − 4ac > 0 then we get two real roots of f (x) =
(x− r1)(x− r2).
If b2 − 4ac = 0 then we get one (repeated) real root
of f (x) = (x− r)2.
CHAPTER 1. MATHEMATICS REVIEW CE1H 17
If b2 − 4ac < 0 then we get no real roots of f (x) (we
get two complex roots).
In this case f (x) has no factorisation and we say that
f (x) is an irreducible quadratic.
Next, Let f (x) be a rational function(
f1(x)f2(x)
). If we
factorise the denominator f2(x) into the product of
distinct linear terms, repeated linear terms and irre-
ducible quatratic terms then we can decompose f (x)
as follows:
CHAPTER 1. MATHEMATICS REVIEW CE1H 18
Case 1: For each distinct linear term (x − r) in the
denominator, the decomposition contains a term Ax−r
where A is an unknown constant.
Case 2: For each repeated linear term (x − r)2 in the
denominator, the decomposition contains terms Ax−r+
B(x−r)2
where A and B are unknown constants.
For each repeated linear term (x− r)3 in the denomi-
nator, the decomposition contains terms Ax−r + B
(x−r)2+
C(x−r)3
, where A, B and C are unknown constants.
Similarly for (x− r)4, (x− r)5, etc.
Case 3: For each irreducible quadratic term ax2+bx+
c in the denominator, the decomposition contains the
term Ax+Bax2+bx+c
where A and B are unknown constants.
CHAPTER 1. MATHEMATICS REVIEW CE1H 19
Example 1.73 The partial fractions decomposition of 2x+3(x−1)(x−2)
is Ax−1 + B
x−2, by Case 1.
Example 1.74 The partial fractions decomposition of x2+3x−1(x−6)(x−3)2
is Ax−6 + B
x−3 + C(x−3)2
, by Case 1 and Case 2.
Example 1.75 The partial fractions decomposition of x2+2x+1(x−4)(x2+1)
is Ax−4 + Bx+C
x2+1, by Case 1 and Case 3.
Example 1.76 The partial fractions decomposition of5x3+3x+1
(x−3)(x+1)3(x−6)2(x2+2)(x2+3)is
Ax−3 +
[B
x+1 + C(x+1)2
+ D(x+1)3
]+[
Ex−6 + F
(x−6)2
]+ Gx+H
x2+2+
Ix+Jx2+3
.
(Note that x2 + 2 and x2 + 3 are irreducible quadratics).
Example 1.77 In Example 1.1 we saw that 2x+3(x−1(x−2) =
Ax−1 + B
x−2. Next, we should find the constants A and B.
Multiply both sides by (x− 1)(x− 2) to get
2x + 3 = Ax−1(x− 1)(x− 2) + B
x−2(x− 1)(x− 2)
⇒ 2x + 3 = A(x− 2) + B(x− 1). ∗
CHAPTER 1. MATHEMATICS REVIEW CE1H 20
Equation ∗ is true for all values of x, so substitute in some
clever values of x.
x = 2 : 2(2) + 3 = A(0) + B(2− 1)
⇒ 7 = B.
x = 1 : 2(1) + 3 = A(1− 2) + B(0)
⇒ 5 = −A ⇒ A = −5.
∴ the partial fractions decomposition is 2x+3(x−1)(x−2) = −5
x−1 +7
x−2.
Example 1.78 In Example 1.2 we had that x2+3x−1(x−6)(x−3)2
=A
x−6 + Bx−3 + C
(x−3)2. Find A, B and C.
Multiply by (x− 6)(x− 3)2 to get
x2 +3x− 1 = A(x− 3)2 +B(x− 6)(x− 3)+C(x− 6) ∗x = 3 : 9 + 9− 1 = A(0) + B(0) + C(−3)
⇒ 17 = −3C ⇒ C = −173 .
x = 6 : 36 + 18− 1 = A(32) + B(0) + C(0)
⇒ 53 = 9A ⇒ A = 539 .
x = 0 : 0 + 0− 1 = A(−3)2 + B(−6)(−3) + C(−6)
⇒ −1 = 9A + 18B − 6C ⇒ −1 = 9539 + 18B − 6
(−173
)⇒ −1 = 53 + 18B + 34 ⇒ −1 = 87 + 18B
⇒ −88 = 18B ⇒ B = −8818 ⇒ B = −44
9 .
∴ the partial fractions decomposition is
CHAPTER 1. MATHEMATICS REVIEW CE1H 21
x2+3x−1(x−6)(x−3)2
=539
x−6 +−449
x−3 +−173
(x−3)2.
Exercise 1.79 In Example 1.3 we had that x2+2x+1(x−4)(x2+1)
=A
x−4 + Bx+Cx2+1
. Find A, B and C.
x2 + 2x + 1 = A(x2 + 1) + (Bx + C)(x− 4) ∗.
x = 4 : 16 + 8 + 1 = A(16 + 1) + 0 ⇒ A = 2517.
x = 0 : 0 + 0 + 1 = A(1) + (B(0) + C)(−4) ⇒
CHAPTER 1. MATHEMATICS REVIEW CE1H 22
∴ we have that x2+2x+1(x−4)(x2+1)
=2517
x−4 +−817 x+ 2
17x2+1
.
CHAPTER 1. MATHEMATICS REVIEW CE1H 23
1.2 Integration using Trigonometric Substitutions
and other tricks
Example 1.80 Prove the formula in the log tables Page
41: ∫1
a2 − x2dx =
1
2aln
∣∣∣∣a + x
a− x
∣∣∣∣ + C.
Using partial fractions, 1a2−x2 = 1
(a+x)(a−x) = Aa+x + B
a−x
⇒ 1 = A(a− x) + B(a + x) ∗x = a ⇒ 1 = 0 + B(2a) ⇒ B = 1
2a
x = −a ⇒ 1 = A(2a) + B.0 ⇒ A = 12a
∴ I =∫
1a2−x2dx =
∫ 12a
a+x +12a
a−xdx
= 12a
∫1
a+x + 1a−xdx = 1
2a{ln |a + x| − ln |a− x|} + C
= 12a ln |a+x
a−x| + C as required.
Example 1.81 (Completing the square)
Find I =∫
1x2+4x+2
dx.
I =∫
1(x+2)2−2
dx. Let u = x + 2, so du = dx.
∴ I =∫
1u2−(
√2)2
du = (−1)∫
1(√
2)2−u2du
= (−1) 12√
2ln |
√2+u√2−u
|+ C by the formula from the previous
Example (log tables page 41).
Thus I = −12√
2ln |
√2+x+2√2−x−2
| + C.
Note that we could also have done this example using par-
CHAPTER 1. MATHEMATICS REVIEW CE1H 24
tial fractions. Try this at home.
Example 1.82 Find I =∫
13x2+2x+4
dx.
Note that we cannot do this using partial fractions since
3x2+2x+4 is irreducible, so 13x2+2x+4
is already the partial
fractions decomposition of 13x2+2x+4
(check!).
Instead, try completing the square:
I =∫ 1
3x2+2
3x+43dx = 1
3
∫1
(x+13)2+11
9dx = 1
3
∫1
(x+13)2+
√119
2dx.
Let u = x + 13, so du = dx.
∴ I = 13
∫1
u2+√
119
2du = 13
1√119
tan−1
(u√119
)+ C
= 13
√911 tan−1[
√911(x + 1
3)] + C
= 1√11
tan−1( 3√11
x + 1√11
) + C.
CHAPTER 1. MATHEMATICS REVIEW CE1H 25
Exercise 1.83 Find I =∫
1x2+6x+4
dx.
I =∫
1(x+3)2−5
dx. Let u = x + 3, so du = dx.
∴ I = . . .
Exercise 1.84 Find I =∫
13x2+2x−4
dx.
I =∫ 1
3x2+2
3x−43dx = 1
3
∫1
(x+13)2−13
9dx.
Example 1.85 (Trigonometric Substitution)
Find I =∫ √
1− x2dx.
Recall that sin2 θ + cos2 θ = 1 (log tables page 9).
Let x = sin θ, so dx = cos θdθ.
∴ I =∫ √
1− sin2 θ cos θdθ
=∫ √
cos2 θ cos θdθ (since sin2 θ + cos2 θ = 1)
=∫
cos θ cos θdθ =∫
cos2 θdθ
CHAPTER 1. MATHEMATICS REVIEW CE1H 26
= 12[θ + 1
2 sin(2θ)] + C (log tables page 42)
= 12[sin
−1 x + 12 sin(2 sin−1 x)] + C.
Exercise 1.86 Find I =∫ √
a2 − x2dx, where a is a con-
stant.
Let x = a sin θ. ∴ dx = a cos θdθ.
CHAPTER 1. MATHEMATICS REVIEW CE1H 27
Exercise 1.87 Prove this formula from page 41 of the log
tables: I =∫
1√a2−x2dx = sin−1(x
a) + C.
Chapter 2
First Order Differential EquationsCE1H
28
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 29
2.1 Separable Differential Equations
A differential equation is an equation containing deriv-
atives (we often write D.E. for differential equation).
Example 2.1 dydx = 2xy is a D.E.
We can solve this D.E. (i.e. we can find y in terms of x).
Proceed as follows:1y
dydx = 2x ⇒ 1
ydy = 2xdx ⇒∫
1ydy = 2
∫xdx
⇒ ln|y| = 2x2
2 + C ⇒ ln|y| = x2 + C ⇒ |y| = ex2+C
⇒ y = ±ex2+C .
If a function y = f (x) satisfies a D.E. then it is called
a solution of the D.E.
In Example 2.1, y = ±ex2+C is a solution of the D.E.dydx = 2xy.
To show this for y = ex2+C , we substitute y = ex2+C
into the D.E. to get:ddx(ex2+C) = ex2+C(2x+0) = 2xex2+C = 2xy as required,
so y is a solution of the D.E.
(you can check that y = −ex2+C is also a solution).
Example 2.2 Show that y = x4
16 is a solution of the D.E.dydx = xy
12 .
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 30
Substituting in we get that the L.H.S. (left hand side) of
the D.E. is ddx
(x4
16
)= 1
16ddxx
4 = 1164x
3 = 14x
3.
The R.H.S. of the D.E. becomes x(
x4
16
)12
= x(x4)12
(16)12
= x√
x4√16
=
xx2
4 = x3
4 = L.H.S. as required, so y = x4
16 is a solution of
the D.E.
The order of a D.E. is the order of the highest deriv-
ative in the D.E.
Example 2.3 d2ydx2 + 5
(dydx
)3
− 4y = ex has order 2, and(dydx
)6
+ 3d2ydx2 +
√d3ydx3 = 0 has order 3.
A first order D.E. of the form dydx = g(x)h(y) is called
a separable D.E.
(where g(x) is a function of x and h(y) is a function of
y).
This D.E. can be solved as follows:1
h(y)dydx = g(x) ⇒
∫1
h(y)dy =∫
g(x)dx.
Then integrate and solve for y if possible.
Example 2.4 Solve the separable D.E. dydx = 3x2
sin y .
sin ydy = 3x2dx ⇒∫
sin ydy = 3∫
x2dx
⇒ − cos y = x3 + C ⇒ cos y = −x3 − C
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 31
⇒ cos−1(cos y) = cos−1(−x3−C) ⇒ y = cos−1(−x3−C).
Exercise 2.5 Solve the separable D.E. dydx = −e−3x.
Exercise 2.6 Solve the separable D.E. dydx = (x+1)2, given
that when x = 0, y = 0.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 32
Exercise 2.7 Solve the separable D.E. dydx = (x + 4)50,
given that when x = 0, y = 451
51 .
Exercise 2.8 Solve the separable D.E. dydx = 4y
x .
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 33
Exercise 2.9 Solve the separable D.E. dydx = cos x+2
sin y .
Exercise 2.10 Solve the separable D.E. dydx = 2x
y+1.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 34
Exercise 2.11 Solve the separable D.E. dydx = 1+x
1+y .
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 35
Exercise 2.12 Solve the separable D.E. dydx = y2+xy2
x2y−x2 .
Exercise 2.13 Solve xy dydx = x2+1
y+1 .
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 36
Exercise 2.14 Solve dydx = x2y + x2 + 3y + 3.
2.2 First Order Linear Differential Equations (The
Integrating Factor Method)
Definition 2.15 A D.E. of the form dydx + P (x)y = Q(x)
is called a first order linear D.E.
We can find the general solution of this D.E. as fol-
lows:
1.) Find the integrating factor e∫
P (x)dx.
2.) Multiply the D.E. by the integrating factor (I.F.) to
get:
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 37
e∫
P (x)dx(
dydx + P (x)y
)= e
∫P (x)dxQ(x) ∗
3.) Note (or remember) that the L.H.S. of ∗ equalsddx(ye
∫P (x)dx)
(= ddx(y.I.F.)).
∴ ∗ becomes d(ye∫
P (x)dx) = e∫
P (x)dxQ(x)dx.
4.) Integrate both sides and solve for y.
This algorithm is called the integrating factor method.
Example 2.16 Solve the first order linear differential equa-
tion dydx − 3y = 0 using the integrating factor method.
1.) I.F. = e∫
P (x)dx = e∫−3dx = e−3x
(note that we do not include the arbitrary constant C).
2.) Multiply the D.E. by the I.F. to get
e−3x(dydx − 3y) = 0e−3x ∗
3.) Recall that the L.H.S. of ∗ equals ddx(y.I.F.).
∴ ddx(ye−3x) = 0 ⇒ d(ye−3x) = 0dx.
4.) Integrate and solve for y:∫d(ye−3x) =
∫0dx
⇒ ye−3x = 0 + C
⇒ y = Ce3x.
Exercise 2.17 Use the I.F. Method to solve dydx + 5y = 4.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 38
1.) I.F. = e∫
5dx = e5x.
2.) Multiply by the I.F. to get
e5x(dydx + 5y) = 4e5x. ∗
3.) Recall that the L.H.S. of ∗ equals ddx(y.I.F.),
so ddx(ye5x) = 4e5x, so d(ye5x) = 4e5xdx.
4.) Integrate to get:∫d(ye5x) = 4
∫e5xdx
⇒ ye5x = 45e
5x + C
⇒ y = 45 + C
e5x .
Note that we could have done this question using
separable variables:dydx = 4− 5y ⇒ dy = (4− 5y)dx ⇒
∫1
4−5ydy =∫
dx
⇒ −15 ln |4− 5y| = x + C
(check: let u = 4− 5y).
Now solve for y.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 39
Exercise 2.18 Solve xdydx − 4y = x6ex.
This is not in the correct form for the I.F. Method, so we
divide across by x to get:dydx −
4xy = x5ex. Now we can use the I.F. Method on this
D.E.:
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 40
Exercise 2.19 Solve dydx + 2y = 3.
Note that dydx + 2y = 3 is a separable D.E. so it can be
solved as follows:dydx = 3− 2y ⇒ 1
3−2ydy = dx ⇒ ...
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 41
Exercise 2.20 Solve dydx + 3y = 2e8x.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 42
Exercise 2.21 Solve dydx − 3y = 2e5x.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 43
Exercise 2.22 Use differentiation to show that y = xex +
Cex is a solution of dydx = y + ex.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 44
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 45
2.3 Homogeneous Differential Equations
The polynomial anxn + . . .+a2x
2 +a1x+a0 has degree
n (if an 6= 0).
A D.E. is homogeneous of degree n if each term in
the D.E. is a polynomial of degree n (in the variables
x and y).
Example 2.23 x2dy + y2dx = 0 is a homogeneous D.E.
of degree 2.
(x2 + 6y2)dx + (5x2 − xy)dy = 0 is a homogeneous D.E.
of degree 2.
(x2 + 6y2)dx + (x3 + 8xy2)dy = 0 is not a homogeneous
D.E. (since 2 6= 3).
(x2 + 3y2)dx + (x2 − 2xy)dy = x2 is not a homogeneous
D.E.
Note: To solve a homogeneous D.E., just let y = vx
(where v is a new variable) and you will get a sepa-
rable D.E. which should be easy to solve.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 46
Example 2.24 Solve the homogeneous D.E. (x − y)dx +
xdy = 0.
Let y = vx, so (x− vx)dx + xd(vx) = 0
⇒ (1− v)dx + d(vx) = 0 (dividing by x)
⇒ (1− v)dx + vdx + xdv = 0 (product rule)
⇒ dx− vdx + vdx + xdv = 0 ⇒ dx + xdv = 0
⇒ dx = −xdv ⇒ −1x dx = dv (variables are separated)
⇒∫ −1
x dx =∫
dv ⇒ (−1) ln |x| + C = v.
But y = vx, so v = yx.
∴ − ln |x| + C = yx ⇒ y = −x ln |x| + Cx.
Exercise 2.25 Solve the homogeneous D.E.
ydx + (x + y)dy = 0.
Let y = vx. ∴ vxdx + (x + vx)d(vx) = 0
⇒ vdx + (1 + v)d(vx) = 0
⇒ vdx + (1 + v)(vdx + xdv) = 0 (product rule)
⇒ vdx + vdx + xdv + v2dx + vxdv = 0
⇒ 2vdx + v2dx + xdv + vxdv = 0
⇒ 2vdx + v2dx = −xdv − vxdv
⇒ (2v + v2)dx = (−x− vx)dv
⇒ (2v + v2)dx = −x(1 + v)dv
⇒ −1x dx = 1+v
v2+2vdv (separated)
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 47
⇒ ...
Exercise 2.26 Solve the homogeneous D.E. (x + y)dx −xdy = 0
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 48
Exercise 2.27 Solve the D.E. −ydx + (x +√
xy)dy = 0.
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 49
Exercise 2.28 Solve the D.E. (x + y)dx + xdy = 0
CHAPTER 2. FIRST ORDER DIFFERENTIAL EQUATIONS CE1H 50
Exercise 2.29 Solve the D.E. ydx = 2(x + y)dy.
Chapter 3
Laplace Transforms CE1H
51
CHAPTER 3. LAPLACE TRANSFORMS CE1H 52
3.1 Introduction
We will use Laplace transforms to solve differential
equations.
Definition 3.1 Given a function F (t), its Laplace trans-
form is
L(F (t)) = f (s) =
∫ ∞
0
F (t)e−stdt.
Example 3.2 Let y = f (x) = ex. As x goes to ∞, ex goes
to ∞. (i.e. limx→∞ ex = ∞). As x goes to −∞, ex goes to
0. (i.e. limx→−∞ ex = 0).
Example 3.3 Find L(1).
By the definition L(1) =∫∞
0 1e−stdt = 1−s[e
−st]∞0 = −1s (e−s∞−
e−s0) = −1s (0− 1) (if s > 0) = 1
s .
∴ L(1) = 1s (where s > 0).
Example 3.4 Find L(t).
L(t) =∫∞
0 te−stdt. Use integration by parts: L.A.T.E.
Let u = t and dv = e−stdt, so du = dt and v = −1s e−st.
∴ L(t) = [uv −∫
vdu]∞0 = [t−1s e−st −
∫ −1s e−stdt]∞0
CHAPTER 3. LAPLACE TRANSFORMS CE1H 53
= −1s [∞e−s∞−0e−s0]+1
s
∫∞0 e−stdt = −1
s [0−0]+1s
1−s[e
−st]∞0
= 0− 1s2 [e
−s∞ − e−s0] = − 1s2 [0− 1] = 1
s2 .
∴ L(t) = 1s2 where s > 0.
Exercise 3.5 Find L(t2).
L(t2) =∫∞
0 t2e−stdt. Integration by parts. L.A.T.E.
Let u = t2 and dv = e−stdt. ∴ du = 2tdt and v = −1s e−st.
Thus L(t2) = [t2−1s e−st −
∫ −1s e−st2tdt]∞0
= [−t2
s e−st + 2s
∫te−stdt]∞0
= [−∞2
s e−s∞ − 02
s e−s0] + 2s
∫∞0 te−stdt
= [0−0]+2sL(t) (by the definition of the Laplace transform)
= 2s
1s2 (by the previous example)
= 2s3 .
∴ L(t2) = 2s3 (where s > 0).
3.2 Proof by Induction
If you want to prove that a statement is true for all
the positive whole numbers, then it is sufficient to:
i) Prove that the statement is true for the whole num-
ber n = 1.
ii) Prove that if the statement is true for any partic-
ular positive whole number n, then the statement is
CHAPTER 3. LAPLACE TRANSFORMS CE1H 54
true for the whole number n + 1.
This technique is called proof by induction.
Example 3.6 Use proof by induction to prove that∑ni=1 i = n(n+1)
2 , for n = 1, 2, 3, . . .
Proof:
i) The statement is true for n = 1, since this becomes∑1i=1 i = 1(1+1)
2 , i.e. 1 = 1.22 .
ii) Assume that the statement is true for a particular posi-
tive whole number n (i.e.∑n
i=1 i = n(n+1)2 ).
Now we will use this to prove that the statement is true
for n + 1 (i.e.∑n+1
i=1 i = (n+1)(n+2)2 ).∑n+1
i=1 i = (∑n
i=1 i) + (n + 1) = n(n+1)2 + (n + 1)
(by the assumption)
= n(n+1)2 + 2(n+1)
2 = n(n+1)+2(n+1)2 = (n+1)(n+2)
2 .
∴∑n+1
i=1 i = (n+1)(n+2)2 , completing the induction and com-
pleting the proof.
Example 3.7 L(tn) = n!sn+1 , for n = 1, 2, 3, . . .
Proof: Use induction:
i) The statement is true for n = 1
i.e. L(t1) = 1!s1+1 , i.e. L(t) = 1
s2 . This was proved in a
previous example.
CHAPTER 3. LAPLACE TRANSFORMS CE1H 55
ii) Assume that the statement is true for a particular posi-
tive whole number n (i.e. L(tn) = n!sn+1).
Next we will prove that the statement is true for n + 1
(i.e. L(tn+1) = (n+1)!sn+2 ).
L(tn+1) =∫∞
0 tn+1e−stdt. Use integration by parts: L.A.T.E.
Let u = tn+1 and dv = e−stdt. ∴ du = (n + 1)tndt and
v = −1s e−st.
∴ L(tn+1) = [uv −∫
vdu]∞0
= [tn+1−1s e−st −
∫ −1s e−st(n + 1)tndt]∞0
= [−tn+1
s e−st + n+1s
∫tne−stdt]∞0
= [−∞n+1
s e−s∞ − −0n+1
s e−s0] + n+1s
∫∞0 tne−stdt
= [0− 0] + n+1s L(tn)
(by the definition of the Laplace transform)
∴ L(tn+1) = n+1s L(tn) = n+1
sn!
sn+1
(by the inductive hypothesis) = (n+1)!sn+2 , as required.
This completes the induction and the proof.
Exercise 3.8 Find L(eat), where a is a constant.
L(eat) =∫∞
0 eate−stdt =∫∞
0 eat−stdt =∫∞
0 e(a−s)tdt
= 1a−s[e
(a−s)t]∞0 = 1a−s[e
(a−s)∞ − e(a−s)0]
= 1a−s[0− 1] (if a− s < 0)
= −1a−s = 1
s−a = L(eat) (where a < s).
CHAPTER 3. LAPLACE TRANSFORMS CE1H 56
Example 3.9 Find L(sin(wt)), where w is a constant.
L(sin(wt)) =∫∞
0 sin(wt)e−stdt.
Use integration by parts: L.A.T.E.
Let u = sin(wt) and dv = e−stdt.
∴ L(sin(wt)) = [uv −∫
vdu]∞0
= [sin(wt)−1s e−st −
∫ −1s e−st cos(wt)wdt]∞0
= [−e−st
s sin(wt) + ws
∫cos(wt)e−stdt]∞0
= [−e−st
s sin(wt)]∞0 + ws
∫∞0 cos(wt)e−stdt
= [−e−∞
s sin(∞)− e0
s sin 0] + ws L(cos(wt))
= [0− 0] + ws L(cos(wt)).
∴ L(sin(wt)) = ws L(cos(wt)) = w
s
∫∞0 cos(wt)e−stdt.
Integration by parts: L.A.T.E. Let u = cos(wt) and dv =
e−stdt, so du = −w sin(wt)dt and v = −1s e−st.
∴ L(sin(wt)) = ws [uv −
∫vdu]∞0
= ws [cos(wt)−1
s e−st −∫ −1
s e−st(−w) sin(wt)dt]∞0
= ws
{[−e−st
s cos(wt)]∞0 − ws
∫∞0 sin(wt)e−stdt
}= w
s
{[−e−∞
s cos(∞)− −e0
s cos 0]− ws L(sin(wt))
}ws
{[0 + 1
s ]−ws L(sin(wt))
}= w
s2 − w2
s2 L(sin(wt)).
∴ L(sin(wt)) = ws2 − w2
s2 L(sin(wt)).
Now solve for L(sin(wt)):
L(sin(wt)) + w2
s2 L(sin(wt)) = ws2
CHAPTER 3. LAPLACE TRANSFORMS CE1H 57
⇒ L(sin(wt))[1 + w2
s2 ] = ws2
⇒ L(sin(wt)) =ws2
1 + w2
s2
=ws2
s2+w2
s2
=w
s2
s2
s2 + w2=
w
s2 + w2.
This is Formula 5.
Linear Operations
Let θ be an operation sending functions to functions.
Example 3.10 ddxf (x) = f ′(x), so differentiation is an
operation sending functions to functions.∫f (x)dx is a function, so integration is an operation.
L(F (t)) = f (s), so the Laplace transform is an operation.
An operation θ is said to be a linear operation if
θ(kf (x)) = kθ(f (x)) and
θ(f (x) + g(x)) = θ(f (x)) + θ(g(x)),
where f (x) and g(x) are functions and k is a constant.
Example 3.11 ddx is a linear operation since d
dx(kf (x)) =
k ddx(f (x)) and d
dx(f (x) + g(x)) = ddx(f (x)) + d
dx(g(x)).
Integration is a linear operation since∫kf (x)dx = k
∫f (x)dx and
CHAPTER 3. LAPLACE TRANSFORMS CE1H 58∫[f (x) + g(x)]dx =
∫f (x)dx +
∫g(x)dx.
Also, the Laplace transform is a linear operation since
L(kF (t)) =∫∞
0 kF (t)e−stdt = k∫∞
0 F (t)e−stdt = kL(F (t)).
Also, L(F (t) + G(t)) =∫∞
0 (F (t) + G(t))e−stdt
=∫∞
0 (F (t)e−st + G(t)e−st)dt
=∫∞
0 F (t)e−stdt +∫∞
0 G(t)e−stdt = L(F (t)) + L(G(t))
as required.
Example 3.12 Find L(2t + 1). By linearity of L we get
L(2t + 1) = L(2t) + L(1) = 2L(t) + L(1) = 2 1s2 + 1
s
(by Formulae 2 and 1).
∴ L(2t + 1) = 2s2 + 1
s .
Exercise 3.13 Find L(2t2 + 3t + 4).
Exercise 3.14 L(6t4 + 5t3 − 7t2 + t− 10 + 8 sin(2t))
CHAPTER 3. LAPLACE TRANSFORMS CE1H 59
Exercise 3.15 L(3t5+cos(4t)+sin t+e4t+te2t+et cos(3t))
3.3 Inverse Laplace Transforms
Let L(F (t)) = f (s). We say that the Laplace trans-
form of F (t) is f (s). We can also write that the in-
verse Laplace transform of f (s) is F (t) and we write
L−1(f (s)) = F (t).
∴ L−1(L(F (t))) = F (t) and L(L−1(f (s))) = f (s).
We can now use the table of Laplace transforms (back-
wards) to find inverse Laplace transforms.
Example 3.16 L−1(
1s
)= 1 (Formula 1)
CHAPTER 3. LAPLACE TRANSFORMS CE1H 60
L−1(
1s2
)= t
L−1(
2!s2+1
)= t2
L−1(
n!sn+1
)= tn
L−1(
ws2+w2
)= sin(wt)
Note: L−1 is a linear operation.
i.e. L−1(f (s) + g(s)) = L−1(f (s)) + L−1(g(s)) and
L−1(kf (s)) = kL−1(f (s)).
Example 3.17 Find L−1( 1s4).
We know that L−1( 3!s3+1) = t3 (by Formula 3).
∴ L−1( 6s4) = t3 ⇒ 6L−1( 1
s4) = t3
(since L−1 is a linear operation) ⇒ L−1( 1s4) = t3
6 .
Exercise 3.18 Find L−1( 1s6).
Exercise 3.19 Find L−1( 1s2+9
).
CHAPTER 3. LAPLACE TRANSFORMS CE1H 61
Exercise 3.20 Find L−1( 2ss2+9
).
Exercise 3.21 Find L−1( 3s+6s2+16
).
Exercise 3.22 Find L−1( 2s+7s2+25
).
Exercise 3.23 Find L−1( 4s+33s2+12
).
CHAPTER 3. LAPLACE TRANSFORMS CE1H 62
= 43 cos(2t) + 1
2 sin(2t).
Check: L(43 cos(2t) + 1
2 sin(2t)) =
Theorem 3.24 (Formula 15) L(tF (t)) = − ddsL(F (t)).
Proof: Omit.
Example 3.25 L(t sin t) = − ddsL(sin t) (Formula 15)
= − dds(
1s2+1
) = − dds[(s
2 + 1)−1]
= −(−1)(s2 + 1)−2 dds(s
2 + 1) = (s2 + 1)−2(2s) = 2s(s2+1)2
.
(Note: You could also have used Formula 10).
Exercise 3.26 Find L(t cos(2t)) using Formula 15.
CHAPTER 3. LAPLACE TRANSFORMS CE1H 63
L(t cos(2t)) = − ddsL(cos(2t)) (Formula 15)
= − dds(
ss2+4
) (Formula 6) = −[(s2+4)−s2s(s2+4)2
] (Quotient Rule)
= −[s2+4−2s2
(s2+4)2] = −[ −s2+4
(s2+4)2] = s2−4
(s2+4)2.
(Note: You could also have used Formula 11).
Exercise 3.27 Use Formula 15 to prove Formula 9.
L(teat) =
Theorem 3.28 (First Shifting Theorem (Formula 14))
L(eatF (t)) = f (s− a) = [L(F (t))]s→s−a
(i.e. it is L(F (t)), with s replaced by s− a).
Proof: L(eatF (t)) =∫∞
0 e−steatF (t)dt =∫∞
0 e−(s−a)tF (t)dt.
Recall that L(F (t)) =∫∞
0 e−stF (t)dt, so replacing s with
s− a gives us L(eatF (t)) as required.
Example 3.29 Use Formula 14 to find L(e4tt3).
L(e4tt3) = [L(t3)]s→s−4 (Formula 14)
= [ 3!s3+1 ]s→s−4 (Formula 3) = [ 6
s4 ]s→s−4 = 6(s−4)4
.
Exercise 3.30 L(e6t cos(4t)) =
CHAPTER 3. LAPLACE TRANSFORMS CE1H 64
Exercise 3.31 L(e7tt9) =
Example 3.32 Previously we proved Formula 9
(L(teat) = 1(s−a)2
) using Formula 15. Here we prove For-
mula 9 again, this time using Formula 14:
L(teat) = L(eatt) = [L(t)]s→s−a (Formula 14)
= [ 1s2 ]s→s−a (Formula 2) = 1
(s−a)2.
Exercise 3.33 L(e8tt cos(2t)) =
Definition 3.34 Intuitively, a function is continuous if
it can be drawn without removing the pen from the page.
CHAPTER 3. LAPLACE TRANSFORMS CE1H 65
Theorem 3.35 (Formula 12) Let F (t) and F ′(t) be con-
tinuous functions for t > 0.
Then L(F ′(t)) = sL(F (t))− F (0).
Proof: L(F ′(t)) =∫∞
0 e−stF ′(t)dt.
Use integration by parts: let u = e−st and dv = F ′(t)dt.
∴ du = −se−stdt and v =∫
F ′(t)dt = F (t).
∴ L(F ′(t)) = [uv−∫
vdu]∞0 = [e−stF (t)−∫
F (t)(−s)e−stdt]∞0
= [e−stF (t)]∞0 + s[∫
F (t)e−stdt]∞0
= [e−s∞F (∞)− e0F (0)] + s∫∞
0 e−stF (t)dt
= [0 − F (0)] + sL(F (t)) (by the definition of the Laplace
transform)
= sL(F (t))− F (0).
Example 3.36 Find L( ddt(sin(t)).
By Formula 12, this equals sL(sin(t)− sin 0
(here F (t) = sin t)
= s 1s2+12 − 0 (Formula 5)
= ss2+1
.
Alternativley, L( ddt(sin t)) = L(cos t) = s
s2+1(Formula 6).
Exercise 3.37 Use Formula 12 to find L( ddt(t sin t)).
CHAPTER 3. LAPLACE TRANSFORMS CE1H 66
Theorem 3.38 (Formula 13) Let F (t), F ′(t) and F ′′(t)
be continuous functions for t > 0.
Then L(F ′′(t)) = s2L(F (t))− sF (0)− F ′(0).
Proof: L(F ′′(t)) =∫∞
0 e−stF ′′(t)dt. Use integration by
parts: let u = e−st and dv = F ′′(t)dt.
∴ du = −se−stdt and v =∫
F ′′(t)dt = F ′(t).
∴ L(F ′′(t)) = [uv −∫
vdu]∞0
= [e−stF ′(t)−∫
F ′(t)(−s)e−stdt]∞0
= [e−stF ′(t)]∞0 + s∫∞
0 e−stF ′(t)dt
= [e−s∞F ′(∞)− e0F ′(0)] + sL(F ′(t))
= [0− F ′(0)] + s[sL(F (t))− F (0)] (Formula 12)
= −F ′(0) + s2L(F (t))− sF (0)
= s2L(F (t))− sF (0)− F ′(0) as required.
Example 3.39 Find L( d2
dt2(tet)) using Formula 13.
L( d2
dt2(tet)) = s2L(tet)− s(0e0)− [ d
dt(tet)]0
CHAPTER 3. LAPLACE TRANSFORMS CE1H 67
= s2 1(s−1)2
− 0− [tet + et1]0 (by Formula 9)
= s2
(s−1)2− [0e0 + e0] = s2
(s−1)2− 1 = s2
(s−1)2− (s−1)2
(s−1)2
= s2−(s−1)2
(s−1)2= s2−(s2−2s+1)
(s−1)2= 2s−1
(s−1)2.
Alternativley, L( d2
dt2(tet)) = L( d
dt(tet + et))
= L(tet + et + et) = L(tet) + 2L(et) = 1(s−1)2
+ 2 1s−1
(Formulae 9 and 4)
= 1(s−1)2
+ 2 s−1(s−1)2
= 1+2(s−1)(s−1)2
= 2s−1(s−1)2
, as before.
Exercise 3.40 Use Formula 13 to find L(F ′′(t)), where
F (t) = cos t.
L(F ′′(t)) = s2L(F (t))− sF (0)− F ′(0) =
3.4 Solving Differential Equations using Laplace
Transforms
To solve a differential equation using the Laplace trans-
form, we:
1: Take the Laplace transform of both sides of the
CHAPTER 3. LAPLACE TRANSFORMS CE1H 68
D.E. (using Formulae 12 and 13).
2: Solve this new equation to find L(y) in terms of s.
3: Find y by taking the inverse Laplace transform of
both sides (usually we need partial fractions for this).
Example 3.41 Solve the D.E. dydt−2y = 6 cos(2t)−6 sin(2t)
using Laplace transforms, given that when t = 0, y = 0
(i.e. y(0) = 0).
1: L(dydt )− 2L(y) = 6L(cos(2t))− 6L(sin(2t)).
Now use Formulae 12, 6 and 5 to take Laplace transforms
to get:
[sL(y)− y(0)]− 2L(y) = 6 ss2+22 − 6 2
s2+22
2: Solve to find L(y):
sL(y)− 0− 2L(y) = 6(s−2)s2+22
⇒ (s− 2)L(y) = 6(s−2)s2+22 ⇒ L(y) = 6
s2+22
3: Taking the inverse Laplace of both sides we get:
L−1(L(y)) = L−1( 6s2+22) ⇒ y = 3L−1( 2
s2+22)
⇒ y = 3 sin(2t) (by Formula 5).
Exercise 3.42 Use Laplace transforms to solve the D.E.dydt + 4y = e−4t, given that y(0) = 2 (i.e. when t = 0,
y = 2).
CHAPTER 3. LAPLACE TRANSFORMS CE1H 69
Exercise 3.43 Use Laplace transforms to solve the D.E.dydt − y = 1, given that y(0) = 0.
CHAPTER 3. LAPLACE TRANSFORMS CE1H 70
∴ y = −1 + et.
Alternativley, we could solve this D.E. using separable
variables:dydt − y = 1 ⇒
Lastly, we could solve the D.E. using the Integrating Fac-
tor Method:dydt − y = 1 ⇒
CHAPTER 3. LAPLACE TRANSFORMS CE1H 71
Exercise 3.44 Use Laplace transforms to solve the D.E.
3dydt + 2y = e6t, given that y(0) = 0.
CHAPTER 3. LAPLACE TRANSFORMS CE1H 72
⇒ y = 120e
6t − 120e
−23t.
Example 3.45 Use Laplace transforms to solve the D.E.d2ydt2
= y, given that y(0) = 1 and y′(0) = 2.
1: L(d2ydt2
) = L(y) ⇒ [s2L(y) − sy(0) − y′(0)] = L(y)
(Formula 13)
⇒ s2L(y)− s1− 2 = L(y).
2: ∴ s2L(y)− L(y) = s + 2 ⇒ (s2 − 1)L(y) = s + 2
⇒ L(y) = s+2s2−1
.
3: L(y) = s+2(s+1)(s−1) = A
s+1 + Bs−1.
∴ s + 2 = A(s− 1) + B(s + 1).
s = 1 : 1 + 2 = A0 + B2 ⇒ 3 = 2B ⇒ B = 32.
s = −1 : −1+2 = A(−2)+B0 ⇒ 1 = −2A ⇒ A = −12
∴ L(y) =−1
2s+1 +
32
s−1
⇒ y = L−1(L(y)) = −12L
−1( 1s+1) + 3
2L−1( 1
s−1)
⇒ y = −12e−1t + 3
2e1t (Formula 4).
Exercise 3.46 Use Laplace transforms to solve d2ydt2
= t,
given that y(0) = 1 and y′(0) = 2.
CHAPTER 3. LAPLACE TRANSFORMS CE1H 73
⇒ y = t3
6 + 2t + 1.
Alternativley, we could solve this D.E. by integrating twice:d2ydt2
= t ⇒
Exercise 3.47 Solve the D.E. d2ydt2
+ 3dydt + 7y = 2et, given
CHAPTER 3. LAPLACE TRANSFORMS CE1H 74
that y(0) = 1 and y′(0) = −1.
1:
⇒ L(y) = s2+s(s−1)(s2+3s+7)
= As−1 + Bs+C
s2+3s+7
CHAPTER 3. LAPLACE TRANSFORMS CE1H 75
∴ L(y) =211
s−1 +911s+14
11s2+3s+7
⇒ y = 211e
1t + 111L
−1( 9s+14(s+3
2)2+(−(32)2+7)
)
= 2et
11 + 911
{L−1
(s+3
2
(s−(−32))2+(
√192 )2
)+ L−1
(2818−
2718
(s−(−32))2+(
√192 )2
)}= 2et
11 + 911
{e−
32t cos(
√192 t) + 1
181
(√
192 )
L−1[
√192
(s−(−32 ))2+(
√192 )2
]
}= 2et
11 + 911
{e−
32t cos(
√192 t) + 1
182√19
e−32t sin(
√192 t)}
= 2et
11 + 911e
−32t{
cos(√
192 t) + 1
9√
19sin(
√192 t)}
= y.
CHAPTER 3. LAPLACE TRANSFORMS CE1H 76
Summary of First Order Differential Equations:
To solve a first order D.E. we try the following four
methods (in this order):
1: If it is separable then separate, integrate and solve
for y.
2: If it is homogeneous, substitute y = vx. This will
lead to a separable D.E.
3: If it is linear (i.e. it is of the form dydx + P (x)y =
Q(x)) then use the Integrating Factor Method (mul-
tiply the D.E. by I.F. = e∫
P (x)dx and recall that this
equals ddx(yI.F.). This leads to a separable D.E.).
4: Otherwise use the Laplace transform (find L of
both sides using the Formula sheet, solve for L(y),
decompose using partial fractions and then find y =
L−1(L(y)) ).
Chapter 4
Second Order DifferentialEquations
77
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 78
4.1 Introduction
Consider a differential equation of the form
ad2ydx2 + bdy
dx + cy = 0, where a, b and c are constants.
From this D.E. we define the auxilliary equation
am2 + bm + c = 0.
We can solve this using the quadratic formula
m = −b±√
b2−4ac2a .
Case 1: b2 − 4ac > 0
Here we have two real roots of the auxilliary equa-
tion: m1 = −b−√
b2−4ac2a and m2 = −b+
√b2−4ac
2a .
This gives us the following general solution of the
D.E.: y = Aem1x+Bem2x, where A and B are unknown
constants.
Proof: We will show that y = Aem1x + Bem2x is a solu-
tion of the D.E. ad2ydx2 + bdy
dx + cy = 0.dydx = d
dx(Aem1x + Bem2x) = Am1em1x + Bm2e
m2x.
∴ d2ydx2 = Am2
1em1x + Bm2
2em2x.
Thus the L.H.S. of the D.E. becomes
a[Am21e
m1x+Bm22e
m2x]+b[Am1em1x+Bm2e
m2x]+c[Aem1x+
Bem2x]
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 79
= Aem1x(am21 + bm1 + c) + Bem2x(am2
2 + bm2 + c)
= Aem1x0 + Bem2x0 (since m1 and m2 are roots of the
quadratic am2 + bm + c = 0)
= 0 + 0 = 0.
∴ L.H.S. =0 as required, so y = Aem1x + Bem2x is a so-
lution of the D.E. ad2ydx2 + bdy
dx + cy = 0.
Example 4.1 Solve the D.E. d2ydx2 + 5dy
dx + 6y = 0.
The auxilliary equation here is m2 + 5m + 6 = 0.
∴ the roots are m =−5±
√52−4(1)(6)
2(1) = −5±√
25−242 = −5±1
2 .
Thus we have two real roots m1 = −5+12 = −2 and m2 =
−5−12 = −3. Thus Case 1 gives us that the general solution
of the D.E. is y = Aem1x + Bem2x = Ae−2x + Be−3x.
Example 4.2 Solve the D.E. d2ydx2 − 7dy
dx + 12y = 0.
The auxilliary equation is m2 − 7m + 12 = 0.
This has roots m =−(−7)±
√(−7)2−4(12)
2(1) = 7±√
49−482 =
7±12 ⇒ m1 = 7+1
2 = 4 and m2 = 7−12 = 3.
Thus we have two real roots, so Case 1 applies, so y =
Aem1x + Bem2x ⇒ y = Ae4x + Be3x.
Exercise 4.3 Solve the D.E. d2ydx2 + 2dy
dx − 3y = 0.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 80
Case 2: b2 − 4ac = 0 in the auxilliary equation.
This gives one repeated real root m = −b±√
02a = −b
2a .
This gives us the following solution of the D.E.:
y = Aemx + Bxemx.
Proof: Omit.
Example 4.4 Solve the D.E. d2ydx2 + 6dy
dx + 9y = 0.
Auxilliary equation: m2+6m+9 = 0 ⇒ m =−6±
√62−4(9)
2 =−6±
√36−362 = −6±0
2 = −3 = m. Thus we have a repeated
real root m = −3, so Case 2 applies.
∴ y = Aemx + Bxemx = Ae−3x + Bxe−3x.
Exercise 4.5 Solve the D.E. d2ydx2 + 10dy
dx + 25y = 0.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 81
Exercise 4.6 Solve the D.E. d2ydx2 + 8dy
dx + 16y = 0.
Recall: i is a number (called an imaginary number)
such that i2 = −1, so i =√−1. A number of the form
α+ iβ is called a complex number. α is called the real
part of α + iβ and β is called the imaginary part of
α + iβ. We can draw an Argand diagram of the com-
plex number 2 + 3i.
Case 3: b2 − 4ac < 0 in the auxilliary equation.
This gives us two complex roots: α± iβ.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 82
In this case the general solution of the D.E. is
y = eαx[A cos(βx) + B sin(βx)].
Proof: Omit.
Example 4.7 Solve the D.E. d2ydx2 + 4dy
dx + 9y = 0.
m2 + 4m + 9 = 0 ⇒ m =−4±
√42−4(9)
2 = −4±√
16−362 =
−4±√−20
2 = −4±√
4√−5
2 = −4±2√−5
2 = −2 ±√−5 = −2 ±
√−1√
5 = −2± i√
5. This equals α± iβ. ∴ α = −2 and
β =√
5, so by Case 3, y = eαx[A cos(βx) + B sin(βx)] =
e−2x[A cos(√
5x) + B sin(√
5x)].
Exercise 4.8 Solve the D.E. d2ydx2 − 2dy
dx + 10y = 0.
Exercise 4.9 Solve the D.E. d2ydx2 + 16y = 0.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 83
Exercise 4.10 Solve the D.E. d2ydx2 − 12dy
dx + 36y = 0.
Example 4.11 Solve the D.E. d2ydx2 + 2dy
dx − 3y = 0, given
that y(0) = 1 and y′(0) = 2.
m2 + 2m − 3 = 0 ⇒ m =−2±
√4−4(−3)
2 = −2±√
162 =
−2±42 = −1 ± 2 ⇒ we have two real roots m1 = 1 and
m2 = −3. ∴ y = Ae1x + Be−3x.
Next we can find the (arbitrary) constants A and B using
the initial conditions y(0) = 1 and y′(0) = 2.
y(0) = 1 and y = Ae1x + Be−3x imply 1 = Ae0 + Be−3(0)
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 84
⇒ 1 = A + B.
Also, dydx = Aex − 3Be−3x and y′(0) = 2
imply 2 = Ae0 − 3Be−3(0) ⇒ 2 = A− 3B. Next we solve
the simultaneous equations
A + B = 1
A− 3B = 2
⇒ 0 + 4B = −1 (subtracting) ⇒ B = −14.
∴ A + B = 1 ⇒ A− 14 = 1 ⇒ A = 5
4.
Thus the solution of the D.E. is
y = Aex + Be−3x = 54e
x − 14e−3x.
Exercise 4.12 Solve the D.E. 2d2ydx2 + 6dy
dx + 5y = 0, given
that y(0) = 1 and y′(0) = −1.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 85
Exercise 4.13 Solve the D.E. 4d2ydx2 − 12dy
dx + 9y = 0, given
that y(0) = 1 and y′(0) = 2.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 86
Summary:
To solve the D.E. ad2ydx2 + bdy
dx + cy = 0, find the roots of
the auxilliary equation am2 + bm + c = 0. If you get:
1: Two different real roots m1 and m2 then the gen-
eral solution is y = Aem1x + Bem2x.
2: One repeated different real root m then the gen-
eral solution is y = Aemx + Bxemx.
3: Two complex roots α ± iβ then the general solu-
tion is y = eαx[A cos(βx) + B sin(βx)].
Next we learn how to solve a D.E. of the form
ad2ydx2 + bdy
dx + cy = f (x) ∗First solve the D.E. ad2y
dx2 + bdydx + cy = 0 to get the com-
plementary function (= yc)
Then find a particular solution (= yp) of ∗.
Then the genaral solution of ∗ is of the from
y = yc + yp.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 87
R.H.S. yp
a A1
ax + b A1x + B1
ax2 + bx + c A1x2 + B1x + C1
aebx A1ebx
a sin(bx) A1 sin(bx) + B1 cos(bx)
a cos(bx) A1 sin(bx) + B1 cos(bx)
Example 4.14 Solve d2ydx2 − 5dy
dx + 6y = 24 ∗.
Auxilliary equation: m2 − 5m + 6 = 0 ⇒ m1 = 2 and
m2 = 3. ∴ C.F. = yc = Ae2x + Be3x.
Next we find yp. R.H.S. of ∗ = 24 ⇒ yp = C.
Now yp = C is a particular solution of the D.E. ∗, sod2
dx2C−5 ddxC+6C = 24 = 0−0+6C = 24 ⇒ C = 4 = yp.
Thus the general solution of ∗ is
y = yc + yp = Ae2x + Be3x + 4.
Example 4.15 Solve d2ydx2 + 14dy
dx + 49y = 4e5x ∗m2 + 14m + 49 = 0 ⇒ (m + 7)(m + 7) = 0
⇒ repeated real root m = −7. Thus by Case 2,
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 88
yc = Aemx + Bxemx = Ae−7x + Bxe−7x.
yp = Ce5x is a particular solution of ∗ (by the table).
Next find C by substituting yp = Ce5x into ∗:d2
dx2(Ce5x) + 14 ddx(Ce5x) + 49(Ce5x) = 4e5x
⇒ 25Ce5x + 14C5e5x + 49Ce5x = 4e5x
⇒ 25C + 70C + 49C = 4 ⇒ 144C = 4 ⇒ C = 136.
∴ yp = 136e
5x.
Thus the general solution of ∗ is
y = yp + yc = 136e
5x + Ae−7x + Bxe−7x.
Example 4.16 If the R.H.S. of ∗ is 3 cos(x), write down
yp: yp = A1 sin x + B1 cos x
If R.H.S. = 2e7x then yp = Ce7x.
If R.H.S. = 3 sinh x = 312(e
x− e−x then yp = Cex +De−x.
If R.H.S. = 2x2 − 7 then yp = A1x2 + B1x + C1.
If R.H.S. = x + 2ex then yp = A1x + B1 + C1ex.
Example 4.17 Solve d2ydx2 − 5dy
dx + 6y = 2 sin x ∗ ∗Note that the L.H.S. of ∗∗ =L.H.S. of ∗ from a previous
example, so they have the same yc.
∴ yc = Ae2x + Be3x (again).
R.H.S.= 2 sin x, so by the table, yp = C cos x + D sin x.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 89
This is a solution of ∗∗, sod2
dx2(C cos x+D sin x)−5 ddx(C cos x+D sin x)+6(C cos x+
D sin x) = 2 sin x
⇒ ddx(−C sin x+D cos x)−5(−C sin x+D cos x+6(C cos x+
D sin x) = 2 sin x
⇒ −C cos x−D sin x + 5C sin x− 5D sin x + 6C cos x +
6D sin x = 2 sin x
⇒ cos x(−C−5D+6C)+sin x(−D+5C +6D) = 2 sin x
⇒ (5C − 5D) cos x + (5C + 5D) sin x = 2 sin x
⇒ 5C − 5D = 0 and 5C + 5D = 2
(by comparing coefficients).
∴ 5C − 5D = 0
5C + 5D = 2
⇒ 10C + 0D = 2 (by adding)
⇒ C = 15. Also, 5C − 5D = 0, so 51
5 − 5D = 0 ⇒1− 5D = 0 ⇒ D = 1
5
∴ yp = 15 cos x + 1
5 sin x.
Also, yc = Ae2x + Be3x.
Thus the general solution is
y = yp + yc = 15 cos x + 1
5 sin x + Ae2x + Be3x.
Exercise 4.18 Solve d2ydx2 + 6dy
dx + 10y = 2 sin(2x) ∗
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 90
To find yp, we see that 2 sin(2x) is not exactly on our table,
but we can work it out by replacing x with 2x.
∴ yp = C cos(2x) + D sin(2x).
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 91
Thus the general solution of the D.E. is y = yp + yc =−215 cos(2x) + 1
15 sin(2x) + e−3x(A cos x + B sin x).
Exercise 4.19 Solve the initial value problem (I.V.P.)d2ydx2 + 3dy
dx − 11y = 0, given that y(0) = 1 and y′(0) = 2.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 92
Exercise 4.20 Solve the I.V.P. d2ydx2 + 2dy
dx + 3y = 0, given
that y(0) = 2 and y′′(0) = 3.
CHAPTER 4. SECOND ORDER DIFFERENTIAL EQUATIONS 93
Chapter 5
Complex Numbers
94
CHAPTER 5. COMPLEX NUMBERS 95
Let z = a + ib, where a and b are real numbers and
i2 = −1. z is called a complex number. a is called the
real part of z and b is called the complex part of z.
Example 5.1 Let z1 = a1 + ib1 and z2 = a2 + ib2. Then
z1+z2 = (a1+a2)+i(b1+b2), z1−z2 = (a1−a2)+i(b1−b2),
z1z2 = (a1+ib1)(a2+ib2) = a1a2+ia1b2+ia2b1+i2b1b2 =
(a1a2 − b1b2) + i(a1b2 + a2b1) (since i2 = −1).
Exercise 5.2 Let z1 = 1+2i and z2 = 3−4i. Find z1+z2,
z1 − z2 and z1z2.
Exercise 5.3 Let z1 = 3+5i and z2 = 6−2i. Find z1+z2,
z1 − z2 and z1z2.
CHAPTER 5. COMPLEX NUMBERS 96
Definition 5.4 Let z = a + ib. Then z̄ = a− ib is called
the conjugate (or complex conjugate) of z.
We can draw a graph of a complex number on a
diagram called an Argand diagram.
So to get a complex conjugate, we just reflect it through
the real (horizontal) axis on its Argand diagram.
Write Re(z) for the real part of z and write Im(z) for
the imaginary part of z.
So Re(a + ib) = a and Im(a + ib) = b.
The length (or mudulus) of a complex number z =
CHAPTER 5. COMPLEX NUMBERS 97
a + ib is |z| = |a + ib| =√
a2 + b2
(by Pythagoras, |z|2 = a2 + b2, so |z| =√
a2 + b2).
Exercise 5.5 Draw on an Argand diagram the set of com-
plex numbers z where Re(z) = 2.
Exercise 5.6 Draw on an Argand diagram the set of com-
plex numbers z where Im(z) = −3.
CHAPTER 5. COMPLEX NUMBERS 98
Exercise 5.7 Draw on an Argand diagram the set of com-
plex numbers z where |z| = 1.
Exercise 5.8 Draw on an Argand diagram the set of com-
plex numbers z where |z + 2− i| = 2.
Next we learn to divide by complex numbers. To
divide by a complex number, multiply above and be-
low by the complex conjugate of the denominator
and then simplify (i.e. z1z2
= z1z2
z̄2z̄2
= . . .)
Example 5.9 Let z1 = 1 + 2i and z2 = 3− 4i.
CHAPTER 5. COMPLEX NUMBERS 99
Then z1z2
= 1+2i3−4i
z̄2z̄2
= (1+2i)(3+4i)(3−4i)(3+4i) = 3+4i+6i+8i2
9+12i−12i+i2(−4)4
= −5+10i9+(−1)(−16) = −5+10i
25 = −525 + 10
25i = −15 + 2
5i.
Exercise 5.10 Let z1 = 5− 3i and z2 = 2 + 4i. Find z1z2
.
The Polar Form of a Complex Number
Let z = a + ib. Thus r =√
a2 + b2 is the length (or
modulus) of z (r = |z|). θ = tan−1 ba is the argument of
z (= arg z). (From the graph, tan θ = ba, so θ = tan−1 b
a.)
Exercise 5.11 Let z = 2 + 5i. Find |z| and arg z.
CHAPTER 5. COMPLEX NUMBERS 100
Note that from the diagram cos θ = ar , so a = r cos θ.
Also, sin θ = br(
oppositehypothenuse), so b = r sin θ. Thus we can
rewrite z = a + ib as z = a + ib = r cos θ + ir sin θ =
r(cos θ + i sin θ) = z. This is called the polar form of
the complex number z.
Exercise 5.12 Find the polar form of z = 4 + 3i.
Exercise 5.13 Find the standard form of the complex num-
ber z if z has length 10 and argument π4 .
Theorem 5.14 (Euler’s Theorem) eiθ = cos θ + i sin θ
Proof: Omit.
CHAPTER 5. COMPLEX NUMBERS 101
Example 5.15 eiπ = −1.
Indeed, eiπ = cos π + i sin π = −1 + i0 = −1.
Next we can use Euler’s Theorem to prove
Theorem 5.16 (DeMoivre’s Theorem) einθ = cos(nθ)+
i sin(nθ) = (cos θ + i sin θ)n.
Proof: einθ = (eiθ)n = (cos θ + i sin θ)n (by Euler’s Theo-
rem). Also,
einθ = ei(nθ) = cos(nθ) + i sin(nθ) (by Euler’s Theorem).
Example 5.17 Let z = 1 +√
3i. Write z in polar form,
find z15 and write your answer in standard (Cartesian)
form.
r = |z| =
√12 +
√3
2=√
1 + 3 = 2.
arg z = θ = tan−1√
31 = π
3 .
∴ z = r(cos θ + i sin θ) = reiθ = 2eiπ3 .
∴ z15 = (2eiπ3 )15 = 215(eiπ
3 )15 = 32, 768e(iπ315) = 32, 768ei5π.
In standard form this is 32, 768(cos(5π) + i sin(5π))
(by DeMoivre’s Theorem) = 32, 768(−1 + i0) = −32, 768.
Exercise 5.18 Let z =√
3 + i. Find z100.
CHAPTER 5. COMPLEX NUMBERS 102
Chapter 6
Matrices
103
CHAPTER 6. MATRICES 104
6.1 Introduction
Definition 6.1 A matrix is a rectangular array of num-
bers.
Example 6.2
(2 5
1 3
)is a matrix.
Example 6.3
3 1 2
2 1 0
0 −5 1
is a matrix.
Example 6.4
(2 1 2
3 5 2
)is a matrix.
Note : A matrix depends on the number of rows and
columns that it contains.
Example 6.5 Example 6.2 is a 2× 2 matrix, Example 6.3
is a 3× 3 matrix and Example 6.4 is a 2× 3 matrix.
Exercise 6.6 What types of matrices are the following :
(i)
(2
1
)(ii)
(2 3)
(iii)
3 1
5 1
2 3
(iv)
3 3 1 0
0 5 1 0
5 1 −1 7
CHAPTER 6. MATRICES 105
6.2 Addition and Subtraction of matrices
We can only add or subtract matrices of the same
type. When two matrices are of the same type, we
add or subtract componentwise.
Example 6.7 Let A =
(2 2
5 5
)and B =
(1 5
3 −5
). Then
A + B =
(2 + 1 2 + 5
5 + 3 5 + (−5)
)=
(3 7
8 0
).
Example 6.8 Let A =
2 5 3
−1 0 1
3 2 1
and B =
1 1 5
2 −1 3
7 8 9
.
Then
A−B =
2− 1 5− 1 3− 5
−1− 2 0− (−1) 1− 3
3− 7 2− 8 1− 9
=
1 4 −2
−3 1 −2
−4 −6 −8
.
Exercise 6.9 Let A =
(2 5
3 1
), B =
(5 3
−1 2
), C =(
2 5
3 1
)and D =
(2 1 5
3 3 2
). Calculate where possible:
CHAPTER 6. MATRICES 106
(i) A + B.
(ii) A−B.
(iii) B − C.
(iv) B + A.
(v) D + A.
(vi) A− C.
Exercise 6.10 Let A =
2 5 3
1 1 1
0 1 1
, B =
3 5 2
5 1 0
0 0 1
and
C =
6 0 3
1 0 0
0 2 0
. Calculate :
(i) A + B.
(ii) A− C.
(iii) B + A.
CHAPTER 6. MATRICES 107
6.3 Scalar Multiplication
When we are multiplying a matrix by scalar (e.g. a
real or complex number), we simply multiply every
entry in the matrix by the scalar.
Example 6.11 Let A =
(2 2
5 5
)and B =
(1 5
3 −5
). Then
5A + 6B =
(10 10
25 25
)+
(6 30
18 −30
)
=
(16 40
43 −5
)
Exercise 6.12 Let A =
(2 −1
0 −3
)and B =
(1 4
3 −2
).
Find (i) 3A + 5B
(ii) A− 3B
(iii) 2A− 1
2B.
CHAPTER 6. MATRICES 108
6.4 Zero Matrix
Definition 6.13 The Zero Matrix (On) is an n× n ma-
trix where each of the entries is zero.
Example 6.14 O2 =
(0 0
0 0
), O3 =
0 0 0
0 0 0
0 0 0
and O4 =
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
.
The zero matrix has an important property. Let A be
any n× n matrix. Then
A + On = A = On + A .
6.5 Multiplication of Matrices
When multiplying matrices, we have to be very care-
ful since we can only multiply matrices that are con-
formable.
CHAPTER 6. MATRICES 109
Definition 6.15 Let A and B be matrices. Then A and
B are conformable if the number of columns in A are the
same as the number of rows in B.
Note : If two matrices A and B are conformable, this
doesn’t necessarily mean that B and A are conformable.
Example 6.16 A =
(2 5 3
1 2 3
), B =
(2 2
−1 0
), C =
(5 0
0 3
)and D =
3 1 0
0 1 0
0 3 −1
. A is 2 × 3 matrix, B is
2 × 2 matrix , C is 2 × 2 matrix and D is 3 × 3 matrix.
For the matrices A, B, C and D, Which pairs of matrices
are conformable?
• A and B are not conformable, however B and A
are conformable.
• A and C are not conformable, however C and A
are conformable.
• A and D are conformable, however D and A are
not conformable.
• B and C are conformable, also C and B are con-
formable.
CHAPTER 6. MATRICES 110
• C and D are not conformable, also D and C are
not conformable.
Exercise 6.17 A =
(2 2
1 3
), B =
(2 5
5 4
), C =
(2 1 3
3 4 2
),
D =
2 5
2 1
0 2
, E =
2 1 0
2 −1 9
0 3 2
and F =
0 1 0
1 0 2
1 1 3
.
For the matrices A, B, C, D, E and F , Which pairs of
matrices are conformable?
Let A be an n × p matrix and B be a p × m matrix.
Clearly A and B are conformable. When we multi-
ply the matrix A by the matrix B (AB), the resulting
matrix is a n × m matrix. To find the (i, j)th-entry
of AB, we single out row i from matrix A and col-
umn j from matrix B. Multiply the corresponding
CHAPTER 6. MATRICES 111
entries from the row and columns and add the re-
sulting products.
Example 6.18 Let A =
(2 2
1 3
), B =
(2 5
−5 4
).
Calculate AB.
Clearly AB is a 2×2 matrix. Thus we have 4 entries in
AB, one in (1st row, 1st column), one in (1st row, 2nd column),
one in (2nd row, 1st column) and one in (2nd row, 2nd column).
(1st row, 1st column)
Take the 1st row in A and the 1st column in B and
multiply corresponding entries and add these together.
i.e. (2)(2) + (2)(−5) = 4− 10 = −6.
(1st row, 2nd column)
Take the 1st row in A and the 2nd column in B and
multiply corresponding entries and add these together.
i.e. (2)(5) + (2)(4) = 10 + 8 = 18.
CHAPTER 6. MATRICES 112
(2nd row, 1st column)
Take the 2nd row in A and the 1st column in B and
multiply corresponding entries and add these together.
i.e. (1)(2) + (3)(−5) = 2− 15 = −13.
(2nd row, 2nd column)
Take the 2nd row in A and the 2nd column in B
and multiply corresponding entries and add these to-
gether. i.e. (1)(5) + (3)(4) = 5 + 12 = 17.
∴ AB =
(−6 18
−13 17
).
Note that BA =
((2)(2) + (5)(1) (2)(2) + (5)(3)
(−5)(2) + (4)(1) (−5)(2) + (4)(3)
)=(
9 19
−6 2
)6= AB. Thus if A and B are conformable
and B and A are conformable, it is not necessarily
true that AB = BA.
Example 6.19 Let A =
(2 5
3 1
), B =
(3 5 7
−1 3 1
).
CHAPTER 6. MATRICES 113
Calculate AB.
AB =
((2)(3) + (5)(−1) (2)(5) + (5)(3) (2)(7) + (5)(1)
(3)(3) + (1)(−1) (3)(5) + (1)(3) (3)(7) + (1)(1)
)
=
(6− 5 10 + 15 14 + 5
9− 1 15 + 3 21 + 1
)
=
(1 25 19
8 18 22
)
Example 6.20 Let A =
(2 5
1 1
), B =
(3
−2
).
Calculate AB.
AB =
((2)(3) + (5)(−2)
(1)(3) + (1)(−2)
)
=
(6− 10
3− 2
)
=
(−4
1
).
Example 6.21 Let A =
1 5 1
1 −1 7
9 3 2
, B =
5 2 3
1 2 3
1 1 −1
.
Calculate AB. AB =
CHAPTER 6. MATRICES 1141(5) + 5(1) + 1(1) 1(2) + 5(2) + 1(1) 1(3) + 5(3) + 1(−1)
1(5) +−1(1) + 7(1) 1(2) +−1(2) + 7(1) 1(3) +−1(3) + 7(−1)
9(5) + 3(1) + 2(1) 9(2) + 3(2) + 2(1) 9(3) + 3(3) + 2(−1)
=
5 + 5 + 1 2 + 10 + 1 3 + 15− 1
5− 1 + 7 2− 2 + 7 3− 3− 7
45 + 3 + 2 18 + 6 + 2 27 + 9− 2
=
11 13 17
11 7 −7
50 26 34
.
Example 6.22 Let A =
1 5 1
1 −1 7
9 3 −2
, B =
5 2
1 2
1 1
.
Calculate AB. AB =
CHAPTER 6. MATRICES 115
=
1(5) + 5(1) + 1(1) 1(2) + 5(2) + 1(1)
1(5) +−1(1) + 7(1) 1(2) +−1(2) + 7(1)
9(5) + 3(1) +−2(1) 9(2) + 3(2) +−2(1)
=
5 + 5 + 1 2 + 10 + 1
5− 1 + 7 2− 2 + 7
45 + 3− 2 18 + 6− 2
=
11 13
11 7
46 22
.
Exercise 6.23 Let A =
(3
−3
), B =
(1 7
−2 3
), C =
(−2 −1
5 3
), D =
(0 4 3
−6 2 −7
), E =
3 4
2 −7
8 10
, F =
1 −1 2
2 −7 5
8 10 1
, G =
5 0 1
3 −2 3
1 2 −4
and H =
1 0 2 3
2 −7 1 1
8 1 1 −1
.
(a) For the matrices {A, B, C,D,E, F, G, H}, decide which
pairs are conformable.
(b) For each of the pairs of matrices that are conformable,
calculate their product.
CHAPTER 6. MATRICES 116
6.6 Transpose of a Matrix
Definition 6.24 The transpose of a matrix A denoted by
AT is obtained by converting the rows of A into columns
one at a time in sequence.
Example 6.25 Let A =
1 0 2
2 −7 1
8 1 1
. Then AT =
1 2 8
0 −7 1
2 1 1
.
In general for two conformable matrices A and B:
• (AB)T = BTAT .
Also if A is any matrix, then (AT )T = A .
Exercise 6.26 Let A =
(3
−3
), B =
(1 7
−2 3
), C =
(−2 −1
5 3
), D =
(0 4 3
−6 2 −7
), E =
3 4
2 −7
8 10
, F =
1 −1 2
2 −7 5
8 10 1
, G =
5 0 1
3 −2 3
1 2 −4
and H =
1 0 2 3
2 −7 1 1
8 1 1 −1
.
CHAPTER 6. MATRICES 117
(a) For the matrices {A, B, C,D,E, F, G, H}, find the
transpose of each matrix.
(b) Show (BC)T = CTBT .
(c) Show (FG)T = GTF T .
6.7 Identity Matrix
Definition 6.27 The Identity Matrix (In) is an n × n
matrix where each diagonal entry is 1 and each off diagonal
entry is 0.
Example 6.28 I2 =
(1 0
0 1
), I3 =
1 0 0
0 1 0
0 0 1
and I4 =
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
.
Let A be any n× n matrix, then
A · In = A = In · A .
CHAPTER 6. MATRICES 118
6.8 The Inverse of a 2× 2 matrix
The determinant of a matrix is denoted by |A| and
the adjoint matrix by A∗. Let A =
(a1,1 a1,2
a2,1 a2,2
), then
|A| = a1,1a2,2 − a1,2a2,1 and A∗ =
(a2,2 −a1,2
−a2,1 a1,1
).
Example 6.29 Let A =
(1 5
−7 1
)and B =
(5 1
3 −5
).
Investigate if (i) |AB| = |A||B| (ii) (A + B)∗ = A∗ + B∗
(iii) (AB)∗ = B∗A∗.
(i) |A| = (1)(1)− (5)(−7) = 1+35 = 36. |B| = (5)(−5)−(3)(1) = −25 − 3 = −28 and |A||B| = (36)(−28) =
−1008.
AB =
(1 5
−7 1
)(5 1
3 −5
)=
(5 + 15 1− 25
−35 + 3 −7− 5
)=
(20 −24
−32 −12
).
Thus |AB| = (20)(−12) − (−32)(−24) = −1008. There-
fore |AB| = |A||B| .
(ii) A + B =
(6 6
−4 −4
), ∴ (A + B)∗ =
(−4 −6
4 6
).
CHAPTER 6. MATRICES 119
A∗ =
(1 −5
7 1
), B∗ =
(−5 −1
−3 5
). ∴ A∗+B∗ =
(−4 −6
4 6
).
Therefore (A + B)∗ = A∗ + B∗ .
(iii) AB =
(20 −24
−32 −12
). ∴ (AB)∗ =
(−12 24
32 20
).
B∗A∗ =
(−5 −1
−3 5
)(1 −5
7 1
)=
(−5− 7 25− 1
−3 + 35 15 + 5
)=(
−12 24
32 20
). Therefore (AB)∗ = B∗A∗ .
In general if A and B are square matrices of the same
type, then
• |AB| = |A||B|
In general if A and B are matrices of the same type,
then
• (A + B)∗ = A∗ + B∗
In general if A and B are conformable matrices, then
• (AB)∗ = B∗A∗ .
CHAPTER 6. MATRICES 120
Also if A is any matrix, then (A∗)∗ = A .
Exercise 6.30 Let A =
(2 3
1 4
), B =
(−1 −3
3 5
)and
C =
(1 2
3 4
). Show that (i) |AB| = |A||B| (ii) |BC| =
|B||C| (iii) (A + C)∗ = A∗ + C∗ (iv) (A + B)∗ =
A∗ + B∗ (v) (AC)∗ = C∗A∗ (vi) (BC)∗ = C∗B∗
(vii) (A∗)∗ = A (viii) (B∗)∗ = B.
We only try to invert square matrices. In general
not every square matrix is invertible. If a square ma-
trix has an inverse, it is said to be invertible or non−singular otherwise it is said to be non− invertible or
singular. If a matrix has an inverse, we denote it by
A−1. The inverse matrix has the property:
A · A−1 = In = A−1 · A .
In general the formula for the inverse of an n×n ma-
trix is :
A−1 =1
|A|A∗ .
CHAPTER 6. MATRICES 121
Thus
A−1 =1
a1,1a2,2 − a1,2a2,1
(a2,2 −a1,2
−a2,1 a1,1
).
where |A| is the determinant of A and A∗ is the ad-
joint matrix of A. Thus if |A| = 0, A is non−invertible.
Also if |A| 6= 0, A is invertible.
Example 6.31 Let A =
(5 3
4 2
). Find A−1.
|A| = 10 − 12 = −2. Also A∗ =
(2 −3
−4 5
). Therefore
A−1 =1
−2
(2 −3
−4 5
).
Example 6.32 Let A =
(3 2
6 4
).
Find |A|. Does A−1 exist?
|A| = (3)(4)− (2)(6) = 12− 12 = 0.
Therefore A−1 does not exist.
Exercise 6.33 Let A =
(7 3
4 2
), B =
(7 1
3
63 3
), C =
CHAPTER 6. MATRICES 122(1 −1
5 0
)and D =
(2 6
1 3
). For the matrices {A, B, C,D},
calculate (where possible) their inverses.
6.9 The inverse of a 3× 3 matrix
Let A =
a1,1 a1,2 a1,3
a2,1 a2,2 a2,3
a3,1 a3,2 a3,3
. Then |A| = a1,1
∣∣∣∣∣a2,2 a2,3
a3,2 a3,3
∣∣∣∣∣−a1,2
∣∣∣∣∣a2,1 a2,3
a3,1 a3,3
∣∣∣∣∣ + a1,3
∣∣∣∣∣a2,1 a2,2
a3,1 a3,2
∣∣∣∣∣= a1,1[a2,2a3,3−a2,3a3,2]−a1,2[a2,1a3,3−a2,3a3,1]+a1,3[a2,1a3,2−a2,2a3,1] and
A∗ =
+
∣∣∣∣∣a2,2 a2,3
a3,2 a3,3
∣∣∣∣∣ −∣∣∣∣∣a2,1 a2,3
a3,1 a3,3
∣∣∣∣∣ +
∣∣∣∣∣a2,1 a2,2
a3,1 a3,2
∣∣∣∣∣−
∣∣∣∣∣a1,2 a1,3
a3,2 a3,3
∣∣∣∣∣ +
∣∣∣∣∣a1,1 a1,3
a3,1 a3,3
∣∣∣∣∣ −∣∣∣∣∣a1,1 a1,2
a3,1 a3,2
∣∣∣∣∣+
∣∣∣∣∣a1,2 a1,3
a2,2 a2,3
∣∣∣∣∣ −∣∣∣∣∣a1,1 a1,3
a2,1 a2,3
∣∣∣∣∣ +
∣∣∣∣∣a1,1 a1,2
a2,1 a2,2
∣∣∣∣∣
T
.
Note that A−1 = 1|A|A
∗.
CHAPTER 6. MATRICES 123
Example 6.34 Let A =
−1 1 2
3 0 5
1 7 2
. Find A−1.
|A| = (−1)
∣∣∣∣∣0 −5
7 2
∣∣∣∣∣− (1)
∣∣∣∣∣3 −5
1 2
∣∣∣∣∣ + (2)
∣∣∣∣∣3 0
1 7
∣∣∣∣∣= (−1)[0(2)−−5(7)]−(1)[3(2)−(−5)(1)]+(2)[3(7)−0(2)]
= −35− 11 + 42 = −4.
A∗ =
+
∣∣∣∣∣0 5
7 2
∣∣∣∣∣ −
∣∣∣∣∣3 5
1 2
∣∣∣∣∣ +
∣∣∣∣∣3 0
1 7
∣∣∣∣∣−
∣∣∣∣∣1 2
7 2
∣∣∣∣∣ +
∣∣∣∣∣−1 2
1 2
∣∣∣∣∣ −∣∣∣∣∣−1 1
1 7
∣∣∣∣∣+
∣∣∣∣∣1 2
0 5
∣∣∣∣∣ −∣∣∣∣∣−1 2
3 5
∣∣∣∣∣ +
∣∣∣∣∣−1 1
3 0
∣∣∣∣∣
T
=
+[(0)(2)− (5)(7)] −[(3)(2)− (5)(1)] +[(3)(7)− (0)(2)]
−[(1)(2)− (2)(7)] +[(−1)(2)− (2)(1)] −[(−1)(7)− (1)(1)]
+[(1)(5)− (0)(2)] −[(−1)(5)− (2)(3)] +[(−1)(0)− (1)(3)]
T
=
−35 −1 21
12 −4 8
5 11 −3
T
=
−35 12 5
−1 −4 11
21 8 −3
.
CHAPTER 6. MATRICES 124
A−1 =1
|A|A∗. Therefore A−1 =
1
−4
−35 12 5
−1 −4 11
21 8 −3
.
Example 6.35 Let A =
−2 7 6
5 1 −2
3 8 4
. Find |A|.
Does A−1 exist ? Explain ?
|A| = (−2)
∣∣∣∣∣1 −2
8 4
∣∣∣∣∣− (7)
∣∣∣∣∣5 −2
3 4
∣∣∣∣∣ + (6)
∣∣∣∣∣5 1
3 8
∣∣∣∣∣= (−2)[(1)(4)−(−2)(8)]−(7)[(5)(4)−(−2)(3)]+(6)[(5)(8)−(3)(1)]
= (−2)[20]− (7)[26] + (6)[37] = −40− 182 + 222 = 0.
∴ |A| = 0, so A−1 doesn’t exist.
Exercise 6.36 Let A =
2 3 4
−5 5 6
7 8 9
, B =
1 2 3
2 4 4
1 2 5
and C =
−2 1 4
3 5 −7
1 6 2
.
(i) Find |A|, |B| and |C|. (ii) For the matrices {A, B, C},
calculate their inverses where possible.
CHAPTER 6. MATRICES 125
6.10 Solving equations involving Matrices
If you want to solve the linear equation ax + b = 0,
you calculate that x = −b
a. However to solve AX =
B where A, X and B are matrices, you cannot say
that X = −B
Asince we cannot divide matrices. There-
fore we must solve equations involving matrices a
different way.
Example 6.37 Let A =
(5 2
1 3
), B =
(7 1
−1 0
), C =(
0 1
−1 1
), D =
(1
−1
)and X be a matrix. Solve the
following matrix equations for X :
(i) AX = B
(ii) XB = C
(iii) BX = D
(iv) ABX = C
(v) BXA = C
(vi) XA = B + 2C.
CHAPTER 6. MATRICES 126
(i) AX = B .
We first pre-multiply (multiply on left) by A−1 on
both sides of the equation.
A−1AX = A−1B
Thus
I2X = A−1B
∴ X = A−1B
We are now ready to calculate X .
X = A−1B = 113
(3 −2
−1 5
)(7 1
−1 0
)= 1
13
(23 3
−12 −1
)= 23
13313
−1213
−113
.
(ii) XB = C .
XB = C
XBB−1 = CB−1 (post-multiply by B−1)
XI2 = CB−1 (BB−1 = I2)
X = CB−1 (XI2 = X)
X = CB−1 =
(0 1
−1 1
)1
1
(0 −1
1 7
)=
(0 1
−1 1
)(0 −1
1 7
)=
1 7
1 8
.
CHAPTER 6. MATRICES 127
(iii) BX = D .
BX = D
B−1BX = B−1D pre-multiply by B−1
I2X = B−1D BB−1 = I2
X = B−1D I2X = X
X = B−1D =
(0 −1
1 7
)(1
−1
)=
(1
−6
).
(iv) ABX = C .
ABX = C
A−1ABX = A−1C pre-multiply by A−1
I2BX = B−1C AA−1 = I2
BX = A−1D I2B = B
B−1BX = B−1A−1C pre-multiply by B−1
I2X = B−1A−1C BB−1 = I2
X = B−1A−1C I2X = X
CHAPTER 6. MATRICES 128
X = B−1A−1C =
(0 −1
1 7
)113
(3 −2
−1 5
)(0 1
−1 1
)=
113
(1 −5
−2 33
)(0 1
−1 1
)= 1
13
(5 −4
−33 29
)=
513
−413
−3313
2913
.
(v) BXA = C .
BXA = C
B−1BXA = B−1C pre-multiply by B−1
I2XA = B−1C
XA = B−1C
XAA−1 = B−1CA−1 post-multiply by A−1
XI2 = B−1CA−1
X = B−1CA−1
X = B−1CA−1 =
(0 −1
1 7
)(0 1
−1 1
)113
(3 −2
−1 5
)=
113
(1 −1
−7 8
)(3 −2
−1 5
)= 1
13
(4 −7
−29 54
)=
413
−713
−2913
5413
.
CHAPTER 6. MATRICES 129
(vi) XA = B + 2C .
XA = B + 2C
XAA−1 = (B + 2C)A−1 post-multiply by A−1
XI2 = (B + 2C)A−1
X = (B + 2C)A−1
X = (B+2C)A−1 =
[(7 1
−1 0
)+
(0 2
−2 2
)]113
(3 −2
−1 5
)=
113
(7 3
−3 2
)(3 −2
−1 5
)= 1
13
(18 1
−11 16
)=
1813
113
−1113
1613
.
Example 6.38 Let A =
−1 1 2
3 0 5
1 7 2
and B =
1
0
−1
.
Solve AX = B for X where X is a matrix.
AX = B
A−1AX = A−1B
I3X = A−1B
X = A−1B
CHAPTER 6. MATRICES 130
X = A−1B = 1−4
−35 12 5
−1 −4 11
21 8 −3
1
0
−1
= 1−4
−40
−12
24
=
10
3
−6
.
Exercise 6.39 Let A =
(7 3
4 2
), B =
(1 −1
5 0
), C =
(1 0
1 3
), D =
(1
−2
), E =
2 3 4
−5 5 6
7 8 9
, F =
1
1
−1
and X be a matrix. Solve the following equations for X :
(i) AX = B
(ii) AX = D
(iii) BX = D
(iv) XAB = C
(v) AXB = C
(vi) AX = B + 2C
(vii) EX = F .
CHAPTER 6. MATRICES 131
Example 6.40 Solve the following simultaneous equations
x + y = 1
x− y = 7
We can rewrite this system of equations as an equa-
tion with matrices as follows:(1 1
1 −1
)(x
y
)=
(1
7
)
Let A =
(1 1
1 −1
), X =
(x
y
)and B =
(1
7
), so we
have AX = B. So solving the original system of
equations is equivalent to solving this matrix equa-
tion for the matrix X .
AX = B
A−1AX = A−1B
I2X = A−1B
X = A−1B
CHAPTER 6. MATRICES 132
X =
(x
y
)= A−1B = 1
−2
(−1 −1
−1 1
)(1
7
)= −1
2
(−8
6
)=(
4
−3
). Therefore x = 4 and y = −3.
Example 6.41 Solve the following simultaneous equations
3x− y = 11
3x− 2y = 13
(3 −1
3 −2
)(x
y
)=
(11
13
)=⇒ AX = B
where A =
(3 −1
3 −2
), X =
(x
y
)and B =
(11
13
).
AX = B
A−1AX = A−1B
I2X = A−1B
X = A−1B
X =
(x
y
)= A−1B = 1
−3
(−2 1
−3 3
)(11
13
)= −1
3
(−9
6
)=(
3
−2
). Therefore x = 3 and y = −2.
CHAPTER 6. MATRICES 133
Example 6.42 Solve the following equations
2x + y + z = 8
5x− 3y + 2z = 3
7x + y + 3z = 20
2 1 1
5 −3 2
7 1 3
x
y
z
=
8
3
20
=⇒ AX = B
where A =
2 1 1
5 −3 2
7 1 3
, X =
x
y
z
and B =
8
3
20
.
AX = B
A−1AX = A−1B
I3X = A−1B
X = A−1B
CHAPTER 6. MATRICES 134
X =
x
y
z
= A−1B = 13
−11 −2 5
−1 −1 1
26 5 −11
8
3
20
=
13
6
9
3
=
2
3
1
. Therefore x = 2 and y = 3 and z = 1.
Exercise 6.43 Solve the following systems of equations
using matrix methods.
(i) 3x− 2y = 7
4x + y = 13
(ii) 2x + 3y = 5
5x− y = −16
(iii) x + y + 2z = 3
4x + 2y + z = 13
2x + y − 2z = 9
(iv) x + y + z = 2
2x + 3y + z = 7
3x− y + 2z = 4
CHAPTER 6. MATRICES 135
6.11 Cramer’s Rule
Suppose we want to solve a system of n linear equa-
tions in n unknowns:
a11x1 + a12x2 + · · · + a1nxn = b1
a21x1 + a22x2 + · · · + a2nxn = b2
...
an1x1 + an2x2 + · · · + annxn = bn
We can rewrite this in matrix form as AX = B, where
A =
a11 a12 . . . a1n
a21 a22 . . . a2n
... ... ...
an1 an2 . . . ann
, X =
x1
x2
...
xn
and B =
b1
b2
...
bn
.
We want to solve AX = B, to find X .
Definition 6.44 Given a matrix A =
a11 a12 . . . a1n
a21 a22 . . . a2n
... ... ...
an1 an2 . . . ann
,
its i, j-cofactor matrix Cij is the matrix you get when you
remove the ith row and the jth column. This gives us an
n− 1× n− 1 matrix.
CHAPTER 6. MATRICES 136
For example C11 =
a22 . . . a2n
...
an2 . . . ann
and
C2n =
a11 a12 . . . a1,n−1
a31 a32 . . . a3,n−1
... ... ...
an1 an2 . . . an,n−1
.
Definition 6.45 Given the matrix A above, its determi-
nant is det(A) = |A| =
(−1)1+1a11det(C11)+(−1)1+2a12det(C12)+· · ·+(−1)1+na1ndet(C1n).
Example 6.46 If A =
(a11 a12
a21 a22
)then
det(A) = a11a22 − a12a21.
Example 6.47 Let A =
1 2 3
4 5 6
7 8 9
. Then det(A) = |A| =
1det(C11)− 2det(C12) + 3det(C13) =
1det
(5 6
8 9
)− 2
(4 6
7 9
)+ 3
(4 5
7 8
)=
1(45−48)−2(36−42)+3(32−35) = −3−2(−6)+3(−3) =
−3 + 12− 9 = 0.
Thus det(A) = 0, so A−1 = A∗
det(A) does not exist.
CHAPTER 6. MATRICES 137
Exercise 6.48 Find det(A), where A =
1 2 −4
1 0 3
2 −1 0
.
det(A) = 1det(C11)− 2det(C12) + (−4)det(C13) =
1(0+3)− 2(0− 6)− 4(−1− 0) = 1(3)− 2(−6)− 4(−1) =
3 + 12 + 4 = 19.
Using determinants we are now ready to state
Cramer’s Rule
Let AX = B be a system of n linear equations in n
unknowns. Let A =
a11 a12 . . . a1n
a21 a22 . . . a2n
... ... ...
an1 an2 . . . ann
, X =
x1
x2
...
xn
and B =
b1
b2
...
bn
. The jth component of X is
xj =det(Bj)
det(A) , where Bj =
a11 a12 . . . b1 . . . a1n
... ... ... ...
an1 an2 . . . bn . . . ann
.
(In Bj, the vector B replaces the jth column of A).
CHAPTER 6. MATRICES 138
Proof: Omit.
Example 6.49 Use Cramer’s Rule to solve:
x1 + 3x2 = 0
2x1 + 4x2 = 6.
Here AX = B, with A =
(1 3
2 4
), X =
(x1
x2
)and B =(
0
6
). Now B1 =
(0 3
6 4
)and B2 =
(1 0
2 6
)(recall that Bj is just A with the vector B replacing the jth
column of A).
∴ x1 = det(B1)det(A) =
det
0 3
6 4
det
1 3
2 4
= 0−18
4−6 = −18−2 = 9 = x1.
Also, x2 = det(B2)det(A) =
det
1 0
2 6
−2 = 6−0
−2 = −3 = x2.
Example 6.50 Use Cramer’s Rule to solve
x + 3y − 2z = 3
2x + 5y + 2z = 5
3x− y + 4z = 8.
CHAPTER 6. MATRICES 139
Here AX = B, where A =
1 3 −2
2 5 2
3 −1 4
, X =
x
y
z
and B =
3
5
8
. Write x = x1, y = x2 and z = x3. Now
B1 =
3 3 −2
5 5 2
8 −1 4
, B2 =
1 3 −2
2 5 2
3 8 4
and
B3 =
1 3 3
2 5 5
3 −1 8
.
∴ x = x1 = det(B1)det(A) =
3det
5 2
−1 4
−3det
5 2
8 4
+(−2)det
5 5
8 −1
1det
5 2
−1 4
−3det
2 2
3 4
+(−2)det
2 5
3 −1
= 3(20+2)−3(20−16)−2(−5−40)
(20−(−2))−3(8−6)−2(−2−15) = 3(22)−3(4)−2(−45)22−3(2)−2(−17) = 66−12+90
22−6+34 =14450 = 72
25 = x1 = x.
Also, y = x2 = det(B2)det(A) =
1det
5 2
8 4
−3det
2 2
3 4
+(−2)det
2 5
3 8
50
CHAPTER 6. MATRICES 140
= (20−16)−3(8−6)−2(16−15)50 = 4−6−2
50 = −450 = −2
25 = x2 = y.
Lastly, we could show that x3 = z = −950 (check).
Exercise 6.51 Use Cramer’s Rule to solve
x + y − 2z = 1
x− y + 3z = 0
2x− y + z = 2.
(You should get that x = 45, y = −5
5 and z = −35 ).
Exercise 6.52 Use Cramer’s Rule to find x
x + 2z = 0
−x + z = 1
2w + x− y + 2z = −1.
3w + x− 2y + z = 3.
(You should get that x = −23).
6.12 Eigenvalues and Eigenvectors
Let A = [aij] be a square n× n matrix.
Let X =
x1
x2
...
xn
6=
0
0...
0
be an n × 1 column vector.
Let λ be a number. Can we have AX = λX ? ∗
CHAPTER 6. MATRICES 141
If there is a number λ and a vector X which satis-
fies ∗, then we say that λ is an eigenvalue of A and
X is an eigenvector of A. These have important ap-
plications for coupled oscillations and vibrations and
image processing.
How do we find the eigenvalues of A ?
We must have AX = λX , so
AX − λX = On×1 = AX − λIX = (A− λI)X = On×1.
This happens when A− λI is not invertible
(otherwise X = (A− λI)−1.O = O)
i.e. when det(A− λI) = O.
Thus to find the eigenvalues of A we must solve
det(A− λI) = O for λ.
Example 6.53 Find the eigenvalues of A =
(4 −1
2 1
).
|A−λI| = |
(4 −1
2 1
)−
(λ 0
0 λ
)| = |
(4− λ −1
2 1− λ
)| =
(4−λ)(1−λ)−(−1)2 = 4−4λ−λ+λ2+2 = λ2−5λ+6 =
0 = (λ− 2)(λ− 3).
∴ λ = 2 and λ = 3 are the eigenvalues of A.
Check: |A−2λI| = |
(4 −1
2 1
)−
(2 0
0 2
)| = |
(2 −1
2 −1
)| =
CHAPTER 6. MATRICES 142
2(−1)− (−1)2 = 0 as required. Thus 2 is an eigenvalue of
A.
Exercise 6.54 Find the eigenvalues of A =
(3 1
−1 2
).
Thus λ = 52 +
√3
2 i and λ = 52 −
√3
2 i are the eigencalues
of A.
Exercise 6.55 Find the eigenvalues of A =
8 −2 0
4 2 1
3 2 −1
.
CHAPTER 6. MATRICES 143
⇒ λ3 − 9λ2 + 12λ + 46 = 0. Using numerical tech-
niques the roots can be shown to be λ = −1.593055586,
λ = 5.296527793− 0.9067081753i and λ = 5.296527793 +
0.9067081753i.
Example 6.56 Find the eigenvectors of A =
(4 −1
2 1
).
Recall that in a previous example we saw that the eigen-
values of A are 2 and 3.
Case 1: λ = 2. Find the eigenvector X associated with the
eigenvector λ = 2.
(A − λI)X = 0, i.e. AX − λIX = 0 ⇒ AX = λX ⇒AX = 2X .
CHAPTER 6. MATRICES 144
∴
(4 −1
2 1
)X = 2X ⇒
(4 −1
2 1
)(x1
x2
)=
(2x1
2x2
)
⇒4x1 − x2 = 2x1
2x1 + x2 = 2x2
⇒2x1 = x2
2x1 = x2
.
Thus the eigenvector X is
(x1
x2
)=
(x1
2x1
)= x1
(1
2
).
Case 2: λ = 3. Find the eigenvector X associated with the
eigenvector λ = 3.
AX = λX ⇒
(4 −1
2 1
)(x1
x2
)=
(3x1
3x2
)
⇒4x1 − x2 = 3x1
2x1 + x2 = 3x2
⇒x1 = x2
x2 = x1
.
Thus the eigenvector X is
(x1
x2
)=
(x1
x1
)= x1
(1
1
).
∴ A =
(4 −1
2 1
)has eigenvalue 2 with associated eigen-
vector
(1
2
)(and its scalar multiples) and has eigenvalue
3 with associated eigenvector
(1
1
)(and its scalar multi-
ples).
CHAPTER 6. MATRICES 145
Exercise 6.57 Find the eigenvalues of A =
(1 2
1 0
)and
their associated eigenvectors.
CHAPTER 6. MATRICES 146
∴ A =
(1 2
1 0
)has eigenvalue 2 with associated eigen-
vector
(2
1
)(and its scalar multiples) and has eigenvalue
−1 with associated eigenvector
(1
−1
)(and its scalar mul-
tiples).
Chapter 7
Partial Differentiation
147
CHAPTER 7. PARTIAL DIFFERENTIATION 148
Example 7.1 A cylinder with radius r and height h has
volume V = πr2h. Thus V depends on r and h, so V
is a function of two variables. If we keep r constant and
increase h, then V will increase. The rate of change of V
with respect to h is written as[
dVdh
]r
or more commonly as∂V∂h . This is called the partial derivative of V with respect
to h.
∴ ∂V∂h = ∂
∂h(πr2h) = πr2 since we treat r as a constant.
Also, ∂V∂r = ∂
∂r(πr2h) = πh ∂∂r(r
2)
(since we treat h as a constant) = πh2r.
Example 7.2 Let z = f (x, y) = x3 + xy − y2.
Note that z is a function from two variables to a third vari-
able, so it represents a surface in three dimensions.∂z∂x = ∂
∂x(x3 + xy − y2) = 3x2 + y − 0 and∂z∂y = ∂
∂y(x3 + xy − y2) = 0 + x− 2y.
∴ ∂∂x(∂z
∂x) = ∂2z∂x2 = 6x and
∂∂y(
∂z∂y) = ∂2z
∂y2 = −2 .
Also, ∂∂x(∂z
∂y) = ∂2z∂x∂y = 1 and
∂∂y(
∂z∂x) = ∂2z
∂y∂x = 1.
Note: for almost all functions, ∂2z∂x∂y = ∂2z
∂y∂x.
Exercise 7.3 Let z = f (x, y) = x cos y − 3√
x2y + x.
CHAPTER 7. PARTIAL DIFFERENTIATION 149
Find ∂z∂x, ∂z
∂y , ∂2z∂x2 , ∂2z
∂y2 , ∂2z∂x∂y and ∂2z
∂y∂x.
Exercise 7.4 Let z = f (x, y) = sin(3xy + ey). Find ∂z∂x
and ∂z∂y .
CHAPTER 7. PARTIAL DIFFERENTIATION 150
Example 7.5 Let z = f (x2 + y2).
Show that x∂z∂y − y ∂z
∂x = 0 ∗∂z∂x = ∂z
∂(x2+y2)∂(x2+y2)
∂x = 2x ∂z∂(x2+y2)
.
Similarly, ∂z∂y = ∂z
∂(x2+y2)∂(x2+y2)
∂y = 2y ∂z∂(x2+y2)
.
Thus the LHS of ∗ equals
x∂z∂y − y ∂z
∂x = 2xy ∂z∂(x2+y2)
− 2xy ∂z∂(x2+y2)
= 0 as required.
Exercise 7.6 Let z = f (yx). Show that x∂z
∂x + y ∂z∂y = 0.
∂z∂x = ∂z
∂(yx)
∂(yx)
∂x = ∂z∂(y
x)y(−1)x−2 = −y
x2∂z
∂(yx)
.∂z∂y = . . .
Exercise 7.7 Let z = f (ax + by), where a and b are con-
stants. Show that b∂z∂x − a∂z
∂y = 0.
CHAPTER 7. PARTIAL DIFFERENTIATION 151
Example 7.8 Let z = 1x2+y2−1
.
Show that x∂z∂x + y ∂z
∂y = −2z(1 + z) ∗∂z∂x = ∂
∂x((x2 + y2 − 1)−1) =
(−1)(x2 + y2 − 1)−2 ∂∂x(x2 + y2 − 1) =
(−1)(x2 + y2 − 1)−2(2x) = −2x(x2+y2−1)−2 = ∂z
∂x.
Also,∂z∂y = ∂
∂y((x2 + y2 − 1)−1) =
(−1)(x2 + y2 − 1)−2 ∂∂y(x
2 + y2 − 1) = −2y(x2+y2−1)−2 = ∂z
∂y .
∴ x∂z∂x + y ∂z
∂y = −2x2
(x2+y2−1)2+ −2y2
(x2+y2−1)2= LHS of ∗.
RHS of ∗ = −2z(1 + z) = −2(x2+y2−1)
(x2+y2−1x2+y2−1
+ 1x2+y2−1
) =
−2(x2+y2−1)
( x2+y2
x2+y2−1) = −2x2−2y2
(x2+y2−1)2.
∴ LHS = RHS of ∗.
Exercise 7.9 Let z = xyx−y .
Show that x∂z∂x + y ∂z
∂y = z ∗
CHAPTER 7. PARTIAL DIFFERENTIATION 152
Show also that x2 ∂2z∂x2 − y2∂2z
∂y2 = 0 ∗ ∗
Next show that z ∂2z∂x∂y = 2∂z
∂x∂z∂y ∗ ∗∗
Definition 7.10 Let z = f (x, y).
Then fx = fx(x, y) = ∂z∂x = zx, fy = fy(x, y) = ∂z
∂y = zy,
CHAPTER 7. PARTIAL DIFFERENTIATION 153
fxx = fxx(x, y) = ∂2z∂x2 = zxx, fyy = fyy(x, y) = ∂2z
∂y2 = zyy,
fxy = fxy(x, y) = ∂2z∂x∂y = zxy and fyx = fyx(x, y) =
∂2z∂y∂x = zyx.
For example fx(1, 2) = ∂z∂x evaluated at x = 1, y = 2.
Example 7.11 Let z = f (x, y) = x3 + 2xy + 2y2.
Find fx(1, 2) and fy(1, 2).
fx = ∂z∂x = 3x2 + 2y, so fx(1, 2) = 3(12) + 2(2) = 7.
fy = ∂z∂y = 2x + 4y, so fy(1, 2) = 2(1) + 4(2) = 10.
Exercise 7.12 Let z = f (x, y) = x sin(xy) + 3.
Find fx(1,π2) and fy(1,
π2).
Exercise 7.13 Let z = f (x, y) = xx2+y2 .
CHAPTER 7. PARTIAL DIFFERENTIATION 154
Find fx(1, 0) and fy(2, 1).
Example 7.14 Let w = f (x, y, z) = x3y2 + 3 sin(yz) + 2.
Find the first order partial derivatives at (1, 0, π2).
fx = ∂w∂x = 3x2y2 + 0 + 0 = 3x2y2.
∴ fx(1, 0,π2) = 3(12)(02) = 0.
fy = ∂w∂y = x32y + 3 cos(yz)z + 0 = 2x3y + 3z cos(yz).
∴ fy(1, 0,π2) = 2(13)(0) + 3π
2 cos(0π2) = 0 + 3π
2 cos(0) = 3π2 .
fz = ∂w∂z = 0 + 3 cos(yz)y + 0 = 3 cos(yz)y.
∴ fz(1, 0,π2) = 3 cos(0π
2)0 = 0.
Note: dydx = 1
dxdy
but ∂y∂x 6=
1∂x∂y
.
Example 7.15 Here we show that d2xdy2 =
−d2y
dx2
(dydx)3
(6= 1
d2y
dx2
).
d2xdy2 = d
dy(dxdy ) = d
dy
(1dydx
)= dx
dyddx
(1dydx
)(by the Chain Rule)
CHAPTER 7. PARTIAL DIFFERENTIATION 155
= dxdy
ddx[(dy
dx)−1] = dxdy (−1)(dy
dx)−2 d2ydx2 = 1
(dydx)
(−1) 1
(dydx)2
d2ydx2 =
−d2y
dx2
(dydx)3
as required.
Example 7.16 Let x = t + 2 sin t and y = cos t.
Find dydx and d2y
dx2 and find their values at t = 0.dydx = dy
dtdtdx (by the Chain Rule) = dy
dt
(1dxdt
)= d
dt(cos t) 1ddt(t+2 sin t)
=
− sin t 11+2 cos t = − sin t
1+2 cos t = dydx.
Thus at t = 0 we get dydx = − sin 0
1+2 cos 0 = −01+2 = 0.
Now d2ydx2 = d
dx(dydx) = d
dx( − sin t1+2 cos t) = d
dt(− sin t
1+2 cos t)dtdx
(by the Cain Rule) = 1
(dxdt )
(1+2 cos t)(− cos t)−(− sin t)(0−2 sin t)(1+2 cos t)2
by the quotient rule.