Topic 3. Complex numbers.

Level 3.
Problem COM3_01.
1
If z1 = 2 − 3i and z2 = 1 + 4i , evaluate (a) z1 + z 2 ; (b) z1 − z 2 ; (c) z1 z 2 ; (d) z12 ; (e) ; (f)
z2
z 2 ÷ z1 ; (g) z12 − z 22 ; (h) z13 − z 23 .

10 11
Answer: (a) 3 + i ; (b) 1 − 7i ; (c) 14 + 5i ; (d) − 5 − 12i ; (e) 1 − 4 i ; (f) − + i ; (g) 10 − 20i ;
17 17 13 13
(h) 1 + 43i .

Solution: (a) z1 + z 2 = 3 + i
(b) z1 − z 2 = 1 − 7i
(c) z1 z 2 = 2 − 12i 2 − 3i + 8i = 14 + 5i
(d) z12 = 4 − 12i + 9i 2 = −5 − 12i
1 1 1 − 4i 1 − 4i 1 4
(e) = = = = − i
z 2 1 + 4i (1 + 4i )(1 − 4i ) 1 + 16 17 17
z 1 + 4i (1 + 4i )(2 + 3i ) (2 − 12) + (8 + 3)i 10 11
(f) 2 = = = =− + i
z1 2 − 3i (2 − 3i )(2 + 3i ) 4+9 13 13
(g) z12 − z 22 = ( z1 − z 2 )( z1 + z 2 ) = (1 − 7i )(3 + i ) = (3 + 7 ) + (− 21 + 1)i = 10 − 20i
( )
(h) z13 − z 23 = ( z1 − z 2 ) z12 + z1 z 2 + z 22 =
( ( ))
= (1 − 7i ) (− 5 − 12i ) + (14 + 5i ) + 1 + 8i + 16i 2 =
= (1 − 7i )(− 6 + i ) = (− 6 + 7 ) + (42 + 1)i = 1 + 43i.

Problem COM3_02.
z
z ∈ C such that is real. Show that z is imaginary.
z −i

z z −i+i i i ⋅i 1
Solution: is real ⇒ =1+ =1 + =1− is real.
z −i z −i z −i i(z − i) iz + 1
1 − iz + 1
∴ is real ⇒ is real. Hence iz is real ⇒ i (iz ) is imaginary. Thus z is
iz + 1 (iz + 1)(− iz + 1)
imaginary ⇒ z is imaginary.

Problem COM3_03.
a b
Complex numbers z1 = and z 2 = where a and b are real, are such that z1 + z 2 = 1 .
1+ i 1 + 2i
Find a and b.

Answer: a = 4 , b = −5 .

a a(1 − i ) a − ia a a
Solution: z1 = = = = − i,
(
1+ i 1+ i 1− i )( ) 1+1 2 2
b b(1 − 2i ) b − 2ib b 2b
z2 = = = = − i.
1 + 2i (1 + 2i )(1 − 2i ) 1 + 4 5 5
 a b a 2b 
Hence z1 + z 2 =  +  − i +  . But z1 + z 2 = 1 and a , b are real. Equating real and
2 5 2 5 
imaginary parts:
a b a 2b
+ = 1 and + = 0 . Therefore a = 4 , b = −5 .
2 5 2 5

Problem COM3_04.
The complex number z and its conjugate z satisfy the equation z z + 2 iz = 12 + 6 i . Find the
possible values of z.

Answer: 3 − i , 3 + 3i .

Solution: Let z = x + iy , where x , y are real. Then x 2 + y 2 + 2ix − 2 y = 12 + 6i . Equating real and
imaginary parts, x 2 + y 2 − 2 y = 12 and 2ix = 6i . Hence x = 3 and y 2 − 2 y − 3 = 0 . Therefore the
possible values of z are 3 − i and 3 + 3i .

Problem COM3_05.
Solve x 2 + 2 x + 2 = 0 .

Answer: − 1 ± i .

− 2 ± 2i
Solution: ∆ = −4 = 4i 2 ⇒ x = = −1 ± i .
2

Problem COM3_06.
Solve x 2 − 4 x + (1 − 4i ) = 0 .

Answer: x = −i or x = 4 + i .

Solution: Find ∆ : ∆ = 4 2 − 4(1 − 4i ) = 12 + 16i .

( )
Find square roots of ∆ : Let (a + ib )2 = 12 + 16i, a, b ∈ R . Then a 2 − b 2 + (2ab )i = 12 + 16i .
Equating real and imaginary parts, a 2 − b 2 = 12 and 2ab = 16 .
a2 −
a
64
2
( )( )
= 12 ⇒ a 4 − 12a 2 − 64 = 0 , a 2 − 16 a 2 + 4 = 0 , a real.

∴a = 4,b = 2 or a = −4,b = −2 . Hence ∆ has square roots 4 + 2i , − 4 − 2i . Use the quadratic

4 ± (4 + 2i )
formula: x 2 − 4 x + (1 − 4i ) = 0 has the solutions x = ,
2
∴x = −i or x = 4 + i .

Problem COM3_07.
Solve x 2 + ( 2 − i ) x − 2i = 0 .

Answer: i,− 2 .
Solution: Find ∆ : ∆ = (2 − i )2 + 8i = 3 + 4i . Find square roots of ∆ : Let
(a + ib )2 = 3 + 4i, a, b ∈ R . Then a 2 − b 2 = 3 and 2ab = 4 .
( )( )
∴a 4 − a 2 b 2 = 3a 2 and a 2 b 2 = 4 . Thus a 4 − 3a 2 − 4 = 0 ⇒ a 2 − 4 a 2 + 1 = 0 , a real, ∴a = 2,b = 1
or a = −2,b = −1 . Hence ∆ has the square roots 2 + i,− 2 − i .
Use the quadratic formula: x 2 + (2 − i )x − 2i = 0
− (2 − i ) ± (2 + i )
has solutions x = ∴x = i or x = −2 .
2

Problem COM3_08.
1 + i is a root of the equation x 2 + (a + 2i )x + (5 + ib ) = 0 , where a and b are real. Find the values
of a and b.

Answer: a = −3 , b = −1 .

Solution: Substituting x = 1 + i , (1 + i )2 + (a + 2i )(1 + i ) + (5 + ib ) = 0 ,

∴(1 − 1) + 2i + (a − 2) + i(a + 2) + 5 + ib = 0 ,
∴(a + 3) + i(a + b + 4) = 0 , a, b ∈ R .
Equating real and imaginary parts: a + 3 = 0 and a + b + 4 = 0 .
Therefore a = −3 , b = −1 .

Problem COM3_09.
1 − 2i is one root of the equation x 2 + (1 + i )x + k = 0 . Find the other root and the value of k.

Answer: x = −2 + i , k = 5i .

Solution: Let z be the other root of the equation

x 2 + (1 + i )x + k = 0 . Then z + (1 − 2i ) = −(1 + i ) and z ⋅ (1 − 2i ) = k . Therefore
z = −(1 + i ) − (1 − 2i ) = −2 + i and k = (− 2 + i )(1 − 2i ) = (− 2 + 2) + i(4 + 1) = 5i . Hence k = 5i and
equation x 2 + (1 + i )x + k = 0 has roots x = 1 − 2i and x = −2 + i .

Problem COM3_10.
a and b are real numbers such that the sum of the squares of the roots of the equation
x + (a + ib )x + 3i = 0 is 8. Find all possible pairs of values a, b.
2

Answer: a = 3, b = 1 or a = −3, b = −1 .

Solution: Let z1 , z 2 are the roots of the equation x 2 + (a + ib )x + 3i = 0 . Then

z12 + (a + ib )z1 + 3i = 0 and z 22 + (a + ib )z 2 + 3i = 0 . But z12 + z 22 = 8 . Hence
8 + (a + ib )( z1 + z 2 ) + 6i = 0 . But z1 + z 2 = −(a + ib ) . Therefore 8 − (a + ib )2 + 6i = 0 ,
( )
∴(a + ib )2 = 8 + 6i , a, b ∈ R . Thus a 2 − b 2 + (2ab )i = 8 + 6i . Equating real and imaginary parts,

a 2 − b 2 = 8 and 2ab = 6 . a 2 −
a
9
2
( )( )
= 8 ⇒ a 4 − 8a 2 − 9 = 0 . a 2 − 9 a 2 + 1 = 0 , a real. ∴a = 3, b = 1

or a = −3, b = −1 .
Problem COM3_11.
Express (6 + 5i)(7 + 2i) in the form a + ib , where a and b are real, and write down
(6 − 5i)(7 − 2i) in a similar form. Hence find the prime factors of 32 2 + 47 2 .

Answer: 32 + 47i , 32 − 47i , 32 2 + 47 2 = 61 ⋅ 53 .

Solution: z = (6 + 5i )(7 + 2i) = (42 − 10) + i(12 + 35) = 32 + 47i . Since (6 − 5i)(7 − 2i) = z , then
2 2 2
(6 − 5i)(7 − 2i ) = 32 − 47i . It is clear z = 6 + 5i 7 + 2i . Therefore we obtain
( )( )
32 2 + 47 2 = 6 2 + 5 2 7 2 + 2 2 = 61 ⋅ 53 .

Problem COM3_12.
Find the modulus and principle argument of (a) − 2i ; (b) − 1 + 3i ; (c) i (i + 1) .

π 2π 3π
Answer: (a) 2, − ; (b) 2, ; (c) 2, .
2 3 4

Solution: In each case P( a, b) represents the complex number z = a + ib and θ is the principal
argument of z .

(a) z = - 2i
Im
z =2
p
0 arg z = -
θ Re 2

-2 P(0,-2)

(b) z = −1 + 3i
Im
P(− 1, 3 ) z = 1+ 3 = 2
3
π 2π
θ =π − ⇒ arg z =
3 3
θ

-1 0 Re
(ñ) z = i(i + 1) = −1 + i
Im
P(-1,1) z = 1+1 = 2
1
3π
θ θ = π − π4 ⇒ arg z =
4

-1 0 Re
Problem COM3_13.
 π π  π π  z z 
If z1 = 2 cos + i sin  and z 2 = 2  cos( − ) + i sin( − )  , find 1 and arg 1  .
 3 3  4 4  z2  z2 

7π
Answer: 2; .
12

π π
Solution: z1 = 2 and arg z1 = , z 2 = 2 and arg z 2 = − .
3 4
z1 z1 2 z  π  π  7π
= = = 2 , arg 1  = arg z1 − arg z 2 = −  −  = .
z2 z2 2  z2  3  4  12

Problem COM3_14.
Use the properties of modulus and argument of a complex number to deduce that (a)
1
z1 z 2 = z1 ⋅ z 2 ; (b)   =
1 1
; (c)   =
z 
1 (z )
 z  (z )  z2  z2
.
( )
Solution: Let r1 = z1 , r2 = z 2 and θ1 = arg z1 , θ 2 = arg z 2 . Then
z1 = r1 (cos θ1 + i sin θ1 ), z1 = r1 (cos(− θ1 ) + i sin (− θ1 )),
z 2 = r2 (cos θ 2 + i sin θ 2 ), z 2 = r2 (cos(− θ 2 ) + i sin (− θ 2 )).
(a) z1 z 2 = r1r2 (cos(θ1 + θ 2 ) + i sin (θ1 + θ 2 )) ⇒ z1 z 2 = r1r2 (cos (− (θ1 + θ 2 )) + i sin (− (θ1 + θ 2 ))).
( )
But z1 ⋅ z2 = r1r2 cos((−θ1) + ( −θ 2 )) + i sin(( −θ1) + ( −θ2 )) . Therefore z1 z 2 = z1 ⋅ z 2 .
(b) Let r = z and arg z = 0 . Then z = r (cos θ + i sin θ) , z = r (cos(− θ) + i sin (− θ)) and
1 1 1 1 1 1
= (cos (− θ ) + i sin (− θ)) . Thus   = (cos θ + i sin θ) and = (cos θ + i sin θ) . Hence
z r  z r (z ) r
1 1
 = .
 z  (z )
z1 r1 z  r1
(c) = (cos (θ1 − θ 2 ) + i sin (θ1 − θ 2 )) ⇒  1  = (cos (− (θ1 − θ 2 )) + i sin (− (θ1 − θ 2 ))) . But
z 2 r2  z2  r2

( z1 ) r1 ( (
cos − θ1 + θ 2 ) + i sin (− θ1 + θ 2 )) . Therefore,
 z1

( )
 z1
 =
=
(z 2 ) r2  z2 ( )
 z2
.

Problem COM3_15.
If z = 1 + i , find z10 and arg z 10 . ( )
π
Answer: 32 ; .
2

π
Solution: z = 1 + i = 2 ⋅  1 + i 1  = 2cis ,
 2 2 4
∴z = 2 and arg z =
π
4
10
( ) ( ) π 5π
. Then z10 = z 10 = 2 = 32 , arg z 10 = 10 arg z = 10 ⋅ = . But
4 2
5π 5π π
> π . The principal argument of z10 is − 2π = .
2 2 2

Problem COM3_16.
z = 1 + 3i . Find the smallest positive integer n for which z n is real and evaluate z n for this
value of n. Show that there is no integral value of n for which z n is imaginary.

Answer: 3, z 3 = −8 .

1
Solution: z = 2 +
3   π π π
( )
i = 2 cos + i sin  ⇒ z = 2, arg z = . If z n is real, then arg z n = kπ ,

2 2   3 3 3

( )
k is integral. But arg z n = n arg z . Therefore n ⋅
π
3
= kπ , k = 0,±1,±2, K ,
∴ n = 3k , k = 0,±1,±2, K
Hence the smallest positive integer n such that z n is real is 3.
( )
z 3 = 2 3 = 8 and arg z 3 = π , ∴ z 3 = −8 .

( )
If z n is imaginary, then arg z n =
π
2
( )
+ kπ , k is integral. But arg z n = n arg z . Therefore
π π 3
n ⋅ = + kπ, k = 0,±1, ±2, K , ∴ n = + 3k , k = 0,±1,±2, K
3 2 2
Hence there is no integral value of n for which z n is imaginary.

Problem COM3_17.
7−i  4  1 
Find the modulus of . Evaluate tan tan −1   − tan −1   . Hence find the principal
3 − 4i  3  7 
7−i
argument of in terms of π .
3 − 4i

π
Answer: 2 , 1, .
4

7−i
Solution: Let z1 = 7 − i , z 2 = 3 − 4i , and z = . Then z1 = 49 + 1 = 5 2 and
3 − 4i
1 4 z1
arg z1 = − tan −1   , z 2 = 9 + 16 = 5 and arg z 2 = − tan −1   , z = = 2 and
7 3 z2
4 1
arg z = arg z1 − arg z 2 = tan −1   − tan −1   . Use a well-known formula:
3 7
 4  1 4 1
tan  tan −1  − tan tan −1  −
 −1  4  −1  1    3  7 3 7 = 1 . Hence tan arg z = 1 . But
tan tan   − tan   = =
 3  7  1 + tan  tan −1 4  ⋅ tan  tan −1 1  1 + 4 ⋅ 1
   
 3  7 3 7
4
3
>
1
7
. Therefore −1  4 
3
−1  1 
arg z = tan   − tan   ∈ 0 , π
7  2
( ). Thus principal value of argument z is π4 .
7 −i  −1  4  −1  1   7 −i π
∴ Modulus of is 2 , tan  tan   − tan   = 1 , principal argument of is .
3 − 4i  3  7  3 − 4i 4

Problem COM3_18.
z
If z =r and arg z = θ , show that is real and give its value.
2
z +r2

1
Answer: .
2 r cos θ

2 z z 1 1 z
Solution: Noting r = zz , = = = . Hence is real. Since
z
2
+r 2
z
2
+ zz z+z 2 Re z z
2
+r2
z 1
Re z = r cos θ , = .
z
2
+r 2 2 r cos θ

Problem COM3_19.
Find the modulus and argument of each of the complex numbers z 1 = 2 i and z 2 = 1 + 3i. Mark
on an Argand diagram the points P, Q, R and S representing z1 , z 2 , z 1 + z 2 and z 1 − z 2
respectively. Deduce the exact values of arg ( z 1 + z 2 ) and arg ( z 1 − z 2 ) .

11 π 5π
Answer: arg ( z 1 − z 2 ) = , arg ( z 1 + z 2 ) = .
12 12

Solution:

Im R

z1 + z2

2 P
3 Q

z1
S z2

z1 - z 2
0 1 Re

 π π π
z 1 = 2 i = 2  cos + i sin  , ∴ z1 = 2 and arg z 1 = .
 2 2 2
1 3
z 2 = 1 + 3i = 2  +i  = 2  cos π + i sin π  , ∴ z2 = 2 and arg z 2 = π .
2 2   3 3 3

OP = z1 , OQ = z2 . But z1 = z2 . Hence OP = OQ and OPRQ is a rhombus. Therefore
5π
∠POR = ∠QOR . Thus arg ( z 1 + z 2 ) = 1 (arg z 1 + arg z 2 ) = . Since diagonals OR and QP of the
2 12
π 11 π
rhombus OPRQ meet at right angle, arg ( z 1 − z 2 ) = arg ( z 1 + z 2 ) + = .
2 12
5π 11 π
∴arg ( z 1 + z 2 ) = , arg ( z 1 − z 2 ) = .
12 12

Problem COM3_20.
(a) Solve the equations x 2 + x + 1 = 0 and x 2 − 3 x + 1 = 0 . Plot on an Argand diagram the
points A and B representing the solutions of the first equation, and C and D representing the
solutions of the second, choosing A and C to lie above the real axis.
(b) Find the angles ∠AOB , ∠COD , ∠COA and ∠ACB .

1 3 3 1 2π π π π
Answer: (a) − ± i, ± i ; (b) , , , .
2 2 2 2 3 3 2 3

Solution: (a) Using the quadratic formula:

2 − 1 ± 3i 1 3
x + x + 1 = 0 ⇒ ∆ = −3 ⇒ x = =− ± i,
2 2 2
2 3 ±i 3 1
x − 3 x + 1 = 0 ⇒ ∆ = −1 ⇒ x = = ± i .
2 2 2

A(− 12 , 2
3
) Im

3 1
C( , )
2 2

Re

3 −1
D( 2
, 2)

B(− 12 , − 2
3
)

1 3 1 3 3 1
(b) Let x1 = − + i , x2 = − − i are the solutions of the first equation, and x3 = + i,
2 2 2 2 2 2
3 1
x4 = − i are the solutions of the second equation.
2 2
2π 2π π π
Then arg x1 = , arg x 2 = − , x1 = x 2 = 1, arg x 3 = , arg x 4 = − , x3 = x4 = 1 .
3 3 6 6
 2π π
Hence ∠AOB = 2 π − (arg x 1 − arg x 2 ) = , ∠COD = arg x 3 − arg x 4 = ,
3 3
π
∠COA = arg x1 − arg x 3 = . ∠ACB = ∠ACO + ∠BCO .
2
1
But ∠ACO = (π − ∠AOC ) , since AO = OC = 1 , and
2
1
∠BCO = (π − ∠BOC ) , since BO = OC = 1 .
2
1 1 π
Therefore ∠ACB = π − (∠AOC + ∠BOC ) = π − (2 π − ∠AOB ) = 1 ∠AOB = .
2 2 2 3
2π π π π
∴∠AOB = , ∠COD = , ∠COA = , ∠ACB = .
3 3 2 3

Problem COM3_21.
2
Let H and K be the points representing the roots of x + 2 px + q = 0 , where p and q are real and
2
p < q . Find the algebraic relation satisfied by p and q when ∠HOK is a right angle.

2
Answer: q −2p =0.

Solution:

A( −1, 2 ) Im

Re

B( −1, − 2 )

Using the quadratic formula:

2 − 2 p ± i2 q − p 2
x + 2 px + q = 0 ⇒ ∆ = 4 p 2 − 4 q ⇒ x = = − p ± q − p 2 i , since p 2 < q .
2
2 2
Let x3 = − p + i q − p and x4 = − p − i q − p . Since ∠HOK = 2 arg x3 , if p <0, or
π π 3π
∠HOK = 2 π − 2 arg x 3 , if p > 0 , ∠HOK = ⇒ arg x 3 = when p <0 or arg x 3 = when
2 4 4
 2
q− p
p >0. In each case must be equal to 1. Hence ∠HOK is a right angle when
p
2
q −2p =0.

Problem COM3_22.
On an Argand diagram the points A, B, C and D represent z1 , z 2 , z 3 and z 4 respectively.
Show that if z 1 − z 2 + z 3 − z 4 = 0 , then ABCD is a parallelogram, and if also
z 1 + iz 2 − z 3 − iz 4 = 0 , then ABCD is a square.

Solution:

Im
B

z2

z1
C
D
z4 z3

0 Re
→ →
If z 1 − z 2 + z 3 − z 4 = 0 , then z 1 − z 2 = z 4 − z 3 . But BA represents z 1 − z 2 , CD represents
→ → →
z 4 − z 3 . Therefore BA and CD are parallel. On the other hand, z 1 − z 4 = z 2 − z 3 . But DA
→ → →
represents z 1 − z 4 , CB represents z 2 − z 3 . Hence DA and CB are parallel. So we proved that
ABCD is a parallelogram.
 Im
B
A

z2

z1 C
D

z4
z3

0 Re
If z 1 + iz 2 − z 3 − iz 4 = 0 , then z 1 − z 3 = i ( z 4 − z 2 ) . Hence the diagonals CA and BD of the
parallelogram ABCD meet at right angle and CA = BD . Therefore ABCD is a square.

Problem COM3_23.
Use the vector representation of z1 and z2 on an Argand diagram to show that, if
π π
0 < arg z 2 < arg z 1 < , and arg ( z 1 − z 2 ) − arg ( z 1 + z 2 ) = , then z1 = z2 .
2 2

Im

C
A
z1 − z2
z1 + z2
z1
B
z2
0 Re

Solution:
→ → → →
Let OA , OB represent z1 , z 2 . Construct the parallelogram OACB . Then OC , BA represent
π → →
z1 + z 2 , z1 − z 2 respectively. Since arg ( z 1 − z 2 ) = arg ( z 1 + z 2 ) + , BA is obtained from OC by a
2
π
rotation anticlockwise about O through , followed by an enlargement in O . Therefore
2
diagonals OC and AB of the parallelogram OACB meet at right angle and OACB is a rhombus.
Hence OA = OB and z1 = z 2 .

Problem COM3_24.
 If z 1 = 24 + 7 i and z2 = 6 , find the greatest and least values of z1 + z 2 .

Answer: 19.

Solution: z1 + z 2 ≤ z 1 + z 2 = 25 + 6 = 31 and this greatest value of 31 is attained when z 2 = kz 1

for some positive real k . But z2 = 6 and z 2 = kz 1 ⇒ 6 = 25 k .
∴ z 1 + z 2 attained the greatest value of 31 when z 2 = 25 (24 + 7 i ) = 25 + 25 i .
6 144 42

z1 + z 2 ≥ z 1 + z 2 = 25 − 6 = 19 and this least value of 19 is attained when z 2 = − kz 1 for some

positive real k . But z2 = 6 and z 2 = − kz 1 ⇒ 6 = 25 k .

∴ z 1 + z 2 attained the least value of 19 when z 2 = − 25 (24 + 7 i ) = − 25 + 25 i .
6 144 42

Problem COM3_25.
On an Argand diagram the points A and B represent the numbers z1 and z 2 respectively. I is the
point (1; 0). D is the point such that triangle OID is similar to triangle OBA. Show that D
z1
represents .
z2
Solution:

Im
A

z1 B
D
z2 z3

0 I(1,0) Re

∆OID ≡ ∆OBA ,
OD OI z3 1 z1
∴ = ⇒ = ∠DOI = ∠AOB ⇒ arg z 3 = arg z 1 − arg z 2 . Hence z 3 = and
OA OB z1 z2 z2
arg z 3 = arg z 1 − arg z 2 .
z1 →
∴z 3 = and OD represents the quotient of z1 and z2 .
z2

Problem COM3_26.
Express − 1 + 3i in modulus/argument form. Use de Moivre’s theorem to show that
(− 1 + 3i ) + (− 1 − 3i )
n  2 nπ 
= 2 n +1 cos 
n
 , n is a positive integer. Evaluate this expression in each
 3 
of the cases n = 3 m , n = 3 m ± 1 , where m is a positive integer.
Answer: (
2 cos 2 π + i sin 2 π
3 3
); 2 n +1
, − 2n .

 1 3   2π 2π 
Solution: Let z = −1 + 3i . Then z = 2 − + i  = 2  cos + i sin  and
 2 2   3 3 
 
 2π 2π  n n 2nπ 2nπ 
− 1 − 3i = z = 2  cos − i sin  . Using de Moivre’s theorem z = 2  cos + i sin  . Now
 3 3   3 3 
n n n n
( )
z + ( z ) = z + z = 2 Re z = 2
n n +1
( )
 2nπ 
cos 
 3 
n n
 . Thus − 1 + 3i + − 1 − 3i = 2
n +1
(  2 nπ 
cos 
 3 
. ) ( )
( n
) ( n
If n = 3 m , − 1 + 3i + − 1 − 3i = 2 n +1 cos  )
 6mπ 
=2
n +1
.
 3 
( n
) ( n 
)
2π 
If n = 3 m ± 1 , − 1 + 3i + − 1 − 3i = 2 n +1 cos  2 m π ±  = −2 n .
 3 

Problem COM3_27.
Express z = 4 2 (1 + i ) in modulus/argument form. Hence find the two square roots of z and
mark their representations on an Argand diagram.

π π
Answer: z = 8 cis , ±2 2 cis  .
4 8 

Solution:

Im
2 2 cis (π8 )

0 Re

−2 2 cis (π8 )
 1 1  π
z = 8 +i  = 8 cis ,
 2 2 4
π
∴ z = 8 and arg z = . By de Moivre’s theorem, one square root of z has modulus 2 2 and
4
π π
argument . Hence the two square roots of z are ± 2 2 cis  .
8 8 

Problem COM3_28.
Use de Moivre’s theorem to find in modulus/argument form the cube roots of − 2 − 2 i .

 π  11 π   5π 
Answer: 2 cis  − , 2 cis  − , 2 cis  .
 4  12   12 
  1 1    3π   3π  
Solution: z = −2 − 2 i = 2 2  − − i  = 8  cos  −  + i sin  −   .
 2 2    4   4 
3π
∴ z = 8 and arg z = − . By de Moivre’s theorem cube roots of z have modulus 2 and
4
π 2 πk
arguments − + , k = −1, 0, 1 . Hence the three roots of z are
4 3
 π  11 π   5π 
2 cis  − , 2 cis  − , 2 cis  .
 4  12   12 

Problem COM3_29.
Use de Moivre’s theorem to solve the equation z 5 = 1 . Show that the points representing the
five roots of this equation on an Argand diagram form the vertices of a regular pentagon of area
5 2π π
sin and perimeter 10 sin .
2 5 5

 2π   4π 
Answer: 1, cis  ±  , cis  ±  .
 5   5 

Solution:

Im

P1
z
P2 1
z
2 2π
5
2π
2π 5
5
P0
0 z 1 Re
2π 0

2π 5

5
z
3
P3 z
4

P4
5
z = 1 ⇒ z = 1 . Hence 5th roots of unity have modulus 1 and their representations
Pk (k = 0 ,1, 2 ,3,4 )
lie on the unit circle with the center in the origin. By de Moivre’s theorem one
root ( z 0 ) has argument zero, the others being equally spaced around the unit circle in the
 2π  2π 
Argand diagram by an angle . Hence the complex 5th roots of unity are 1, cis  ± ,
5  5 
 4π 
cis  ± .
 5 
2π π
Since ∠Pk OP k +1 = and OP k = OP k +1 = 1 , Pk Pk +1 = 2sin for any
5 5
k = 0 ,1, 2 ,3, 4 (P5 := P0 ) . Therefore the points Pk (k = 0 ,1, 2 ,3,4 ) form the vertices of a regular
5 2π π
pentagon of area sin (= 5 ⋅ (area of ∆P0 OP1 )) and perimeter 10 sin (= 5 ⋅ P0 P1 ) .
2 5 5

Problem COM3_30.
Show that the roots of z 6 + z 3 + 1 = 0 are among the roots of z
9
− 1 = 0 . Hence find the roots of
6 3
z + z + 1 = 0 in modulus/argument form.

 2π   4π   8π 
Answer: cis  ±  , cis  ±  , cis  ±  .
 9   9   9 

Solution: z
9
( )( )
− 1 = z 3 − 1 z 6 + z 3 + 1 . Therefore, if z 6 + z 3 + 1 = 0 , then z 9 − 1 = 0 . Hence the
9
roots of z 6 + z 3 + 1 = 0 are among the roots of z 9 − 1 = 0 . Let z = cos θ + i sin θ satisfy z =1.
Using de Moivre’s theorem, cos (9 θ) + i sin (9 θ) = 1 + 0 i
∴cos (9 θ) = 1 and sin (9 θ) = 0
∴9 θ = 2 πk , k integral.
∴θ = 29π k , k integral.
2π 2π
Taking θ = k , k = 0 ,1,K,8 gives 9 distinct numbers z with argument k .
9 9
If z+ z 3 + 1 = 0 , then z 9 = 1 but z 3 ≠ 1 . Hence the roots of z 6 + z 3 + 1 = 0 are
6

 2π   2π 
cos  k  + i sin  k  , k = 1,2 , 4 ,5,7 ,8 .
 9   9 
 2π   4π   8π 
∴z 6 + z 3 + 1 = 0 has the roots cis  ±  , cis  ±  , cis  ±  .
 9   9   9 

Problem COM3_31.
If z = cos θ + i sin θ , show that z
n
− z −n = 2 i sin n θ . Hence show that
1
sin 5 θ = (sin 5 θ − 5 sin 3θ + 10 sin θ) .
16

n
Solution: By de Moivre’s theorem and z = cos n θ + i sin n θ and
z
−n
= cos (− n θ) + i sin (− n θ) = cos n θ − i sin n θ . Then z n + z −n = 2 cos n θ and z n − z −n = 2 i sin n θ .

2 i sin θ = z − z −1 . Then 5
32 i sin
5
( ) 5
( ) = z − 5 z + 10 z − 10 z −
θ = z − z −1 . But z − z −1
5 5 3 1

(
+ 5 z −3 − z −5 = z 5 − z −5 − 5 z 3 − z −3 ) ( ) + 10 (z − z ) = 2i sin 5θ − 10 i sin 3θ + 20 i sin θ .
1 −1

1
Hence sin 5 θ = (sin 5 θ − 5 sin 3θ + 10 sin θ) .
16

Problem COM3_32.
 1, ω and ω 2 are the cube roots of unity. State the values of ω3 and 1 + ω + ω 2 . Hence show
that (1 + ω )2 12
( )( )(
= 1 and (1 − ω) 1 − ω 2 1 − ω 4 1 − ω5 1 − ω7 1 − ω8 = 27 . )( )( )
Answer: ω3 = 1 , ω2 + ω + 1 = 0 .

Solution: The cube roots of unity satisfy x

3
(
− 1 = 0 . But x 3 − 1 = ( x − 1) x 2 + x + 1 . Hence )
ω ≠ 1 ⇒ ω 2 + ω + 1 = 0 . Clearly, ω3 = 1 . Therefore 1 + ω ( 2 12
) = (− ω)12 = (ω )
3 4
=1.
Then ω 4 = ω3 ⋅ ω = ω ,
ω5 = ω3 ⋅ ω 2 = ω 2 ,
ω7 = ω6 ⋅ ω = ω ,
ω8 = ω6 ⋅ ω 2 = ω 2 .
Hence (1 − ω)(1 − ω2 )(1 − ω4 )(1 − ω5 )(1 − ω7 )(1 − ω8 ) =
((1 − ω)(1 − ω )) = (1 − ω − ω + ω ) = (2 − ω − ω )
2 3 2 3 3 2 3
=

(3 − (1 + ω + ω )) = 3 = 27 .
2 3 3

Problem COM3_33.
1, ω and ω 2 are the three cube roots of unity. Show that if the equations z
3
− 1 = 0 and
5
pz + qz + r = 0 have a common root,

then ( p + q + r )(p ω5 + q ω + r )(p ω10 + q ω2 + r = 0 . )

Solution: The cube roots of unity satisfy z 3 − 1 = 0 . Therefore, if z is a common root of the
equations z 3 − 1 = 0 and pz 5 + qz + r = 0 , then z is one of the cube roots. Thus
if z =1, then p + q + z = 0 ; if z = ω , then p ω5 + q ω + r = 0 ; if z =ω
2
, then 10
pω + qω2 + r = 0 .
Hence ( p + q + r )(p ω5 + q ω + r )(p ω10 + q ω2 + r = 0 . )
Problem COM3_34.
7 + 4i
Express z1 = in the form a + ib , where a and b are real. On an Argand diagram sketch the
3 − 2i
locus of the point representing the complex number z such that z − z1 = 5 . Find the greatest
value of z subject to this condition.

Answer: z1 = 1 + 2 i , 2 5.

7 + 4i ( 7 + 4 i )( 3 + 2 i ) ( 21 − 8 ) + (14 + 12 ) i
Solution: z1 = = = = 1 + 2i .
3 − 2i ( 3 − 2 i )( 3 + 2 i ) 9+4
 Im
Q(2,4)

Let A represent z1 = 1 + 2 i . Then

P
→
A(1,2) AP represents z − z1 and
z − z1 = 5 ⇒ AP = 5
. Hence P
lies on the circle center A (1, 2 ) and
radius 5 . So the locus of P has
0
Re equation ( x − 1) 2 + ( y − 2 ) 2 = 5 . Let
Q be the intersection of the line OA and the circle. Then the greatest value of z is OQ = 2 5 .

Problem COM3_35.
The point P represents the complex number z on an Argand diagram. Describe the locus of P
π
when arg ( z − 2 ) = arg ( z + 2 ) + .
2

2
Answer: y= 4−x .

→ →
Solution: Let A (2 , 0 ) , B (− 2 ,0 ) represent 2, − 2 respectively. Then AP and BP represent z −2
π →
and z +2 respectively. arg ( z − 2 ) = arg ( z + 2 ) + requires AP to be parallel to the vector obtained
2
→ π
by rotation of BP anticlockwise through an angle of .
2
Im

B(- 2, 0 ) 0 A( 2, 0 ) Re
arg( z + 2)
π arg( z - 2 )
2 π
2 z- 2
z+2

If P lies below the x-axis, AP must be parallel to a clockwise rotation of BP . This diagram
π
shows arg ( z − 2 ) = arg ( z + 2 ) − . Hence P must lie above the x-axis.
2
 Im
P

z+2 π
π 2 z −2
2
arg( z − 2 )
arg( z + 2)
B(−2, 0 ) 0 A( 2, 0 ) Re

π
Since alternate angles between parallel lines are equal, ∠BPA = as P traces its locus. Hence P
2
lies on the upper arc AB of a circle through A and B .

Im

x 2 + y 2 = 4, y > 0

π
2

B(- 2, 0 ) 0 A( 2, 0 ) Re

The center of this circle is the center of diameter AB . Hence the locus of P has equation
2 2 2
x + y = 4 , y > 0 , or y = 4 − x .

π
Let z = x + iy satisfies both z = z −2 and arg ( z − 2 ) = arg ( z + 2 ) + . Then x =1 and
2
y= 4 −1 = 3 . Hence z =1+i 3 .

Problem COM3_36.
Indicate on an Argand diagram the region which contains the point P representing z when
π π
z ≤ z −2 and − ≤ arg z ≤ .
4 4

Solution:
 Im
y=x

(1, 1)

A( 2, 0 )
0 Re

(1, − 1)
y = −x

x =1
π π
z = z −2 is the perpendicular bisector of OA . arg z = is the ray y =x, x >0 . arg z = − is
4 4
the ray y = −x , x > 0 .

Problem COM3_37.
z − 1 ≤ z − i and z − 2 − 2 i ≤ 1 . Sketch the region in the Argand diagram which contains the
point P representing z. If P describes the boundary of this region, find the value of z when
π
arg ( z − 1) = .
4

Answer: 2 + i , 3 + 2i .

Solution:

Im y=x

C
(2, 2)
P(3, 2)

B
1
P(2, 1)

A π
4

0 1 Re
Let A ,. B and Q represent 1, i, z respectively. If z −1 = z − i , then AQ = BQ and the locus of
1 1
Q is the perpendicular bisector of AB . Since AB has midpoint  ,  and gradient − 1 , the
2 2
1 1
locus of Q passes through  ,  with gradient 1 and has Cartesian equation y = x .
2 2
Let C represent 2 + 2 i . If z − 2 − 2 i ≤ 1 , then CQ = 1 and Q lies on or inside the circle
with center (2 , 2 ) and radius 1.
Let now z − 1 ≤ z − i and z − 2 − 2 i ≤ 1 . Then AQ ≤ BQ and CQ ≤ 1 . Hence Q lies on the
right-hand side of the perpendicular bisector of AB inside the circle centre C and radius 1, or Q
π
lies on the boundary of this region. If P describes the boundary of this region and arg ( z − 1) = ,
4
→ π
then CP = 1 and AP makes the angle with the positive x-axis. Thus we must solve
4
simultaneously two Cartesian equations (x − 2 )2 + ( y − 2 )2 = 1 and y = x − 1 . Substituting the
second equation into the first gives ( x − 2 )2 + (x − 3 )2 = 1 ⇒ 2 x 2 − 10 x + 12 = 0 ⇒ x = 2 , 3 ⇒ y = 1
(when x = 2 ), y = 2 (when x = 3 ). Therefore such P represents z = 2 +i and z = 3 + 2i .

Problem COM3_38.
z − 1 = 1 . Sketch the locus of the point P representing z on an Argand diagram. Hence deduce

that arg ( z − 1) = arg z 2 ( ).

Solution:

Im

z
z-1
θ 2θ B(2,0)

0 A(1,0) Re

→
Let A represent 1. Then AP represents z − 1 and AP = 1 . Hence P lies on the circle center
A (1, 0 ) and radius 1.
Let θ = arg z and B represent 2. Then ∠POB = θ and ∠PAB = arg (z − 1) . But ∠PAB = 2 ∠POB
and arg z ( ) = 2 arg z . Therefore arg (z − 1) = 2 θ = 2 arg z = arg (z ) as ∆ OAP
2 2
is isosceles with
AO = AP .

Problem COM3_39.
 1
The complex number z is given by z =t + , where t = r (cos θ + i sin θ) . Find the equation of the
t
π
locus of the point P which represents z on an Argand diagram when θ = and r varies.
4

2
Answer: x − y2 =2 .

1
Solution: Let P (x , y ) represent z = x + iy . Then x + iy = r (cos θ + i sin θ) + (cos θ − i sin θ ) =
r
 1  1
 r +  cos θ + i  r −  sin θ ,
 r  r
 1  1
∴x =  r +  cos θ and y =  r −  sin θ .
 r  r
 1 1  1 1
x = r +  and y =  r −  . Hence x 2 − y 2 = 2 .
 r  2  r  2
Im

− 2 2
0 Re

x 2 − y2 = 2

Problem COM3_40.
Indicate on an Argand diagram the region defined by the pair of inequalities and z ≤6
z − 5 ≤ 5 . Write down the range of values of arg z for such z. Find the values of z for which both
z = 6 and z − 5 = 5 .

π π 18 24
Answer: − < arg z < , z= ±i .
2 2 5 5

Solution:
 Im

x 2 + y 2 = 36 6
(x − 5)2 + y 2 = 25

0 5 Re

z =6 is the circle, center (0 ,0 ) and radius 6. z − 5 = 5 is the circle, center (5,0 ) and the radius 5.
Since y-axis is a tangent line to the circle z − 5 = 5 at point (0 ,0 ) , if z ≤ 6 and z − 5 ≤ 5 , then
π π
− < arg z < .
2 2
Let z = x + iy . Then z =6⇒x
2
+ y 2 = 36 , and z − 5 = 5 ⇒ ( x − 5 )2 + y 2 = 25 . Hence, if z such
2
that both z =6 and z −5 =5 , then both x + y 2 = 36 and x 2 + y 2 − 10 x + 25 = 25 . Therefore
10 x = 36 .
18
∴x = .
5
2
 18  24
∴y = ± 36 −   = .
5  5
18 24
Hence the values of z for which both z =6 and z −5 =5 are ±i .
5 5