Short proof of complex numbers

relimS
26-Jan-2025

Annoyingly, in a certain MO in a certain country (countries), to use com-

plex numbers would require proving even the most basic algebraic properties.
Here, I try to prove these basic properties as short as I can using knowledge
allowed in the VMO so that one can attempt to write down when the time
comes.

1
1 Basic definitions.
1. A complex number: is a number with 2 component a + bi, where a, b
are real numbers and i is a number such that i2 = −1.
For the rest of this document, i is defined to be a number such at
i2 = −1 and z is used to refer to a generic complex number like how x is
used for real numbers.

2. Real part (ℜ(z)) and Imaginary part (ℑ(z)): ℜ and ℑ are functions
that output real numbers that are respectively the Real and Imaginary
part of the input complex number.
Example: ℜ(a + bi) = a, ℑ(a + bi) = b, z = ℜ(z) + iℑ(z) with a and b are
real numbers.

3. Complex plane: is a plane formed by complex numbers. In terms of

Cartesian coordinate system, a complex number z is a point (ℜ(z), ℑ(z))
with the x-axis being the real axis and the y-axis being the imaginary
axis.
4. Negation: −z is defined as −z = −ℜ(z) − iℑ(z).

5. Conjugate: z is defined as z = ℜ(z) − iℑ(z).

6. Addition: z1 + z2 is defined as z1 + z2 = ℜ(z1 ) + ℜ(z2 ) + i(ℑ(z1 ) + ℑ(z2 )).
7. Subtraction: z1 − z2 is defined as z1 + (−z2 ).
8. Multiplication: z1 ·z2 or z1 z2 are defined as z1 z2 = (ℜ(z1 )+iℑ(z1 ))(ℜ(z2 )+
iℑ(z2 )) = ℜ(z1 )ℜ(z2 ) − ℑ(z1 )ℑ(z2 ) + i(ℜ(z1 )ℑ(z2 ) + ℑ(z1 )ℜ(z2 )).

9. Modulus: |z| is defined as the distance of z from 0. |z| is a real number

such that |z|2 = zz (by Pythagorean theorem).
10. Division: For non-zero z2 , z1 /z2 or
z1
z2
is defined as
z1 z1 z2 z1 z2 1
= = = z1 z2 .
z2 z2 z2 |z2 |2 |z2 |2

11. Argument: For non-zero z, arg(z) is defined as the directed angle from
the real axis to the vector from 0 to z.
12. iR: is a set of complex numbers such that their real components are zero.

2
2 Basic algebraic properties.
2.1 ℜ(z1 + z2 ) = ℜ(z1 ) + ℜ(z2 ), ℑ(z1 + z2 ) = ℑ(z1 ) + ℑ(z2 )
ℜ(z1 + z2 ) = ℜ(ℜ(z1 ) + ℜ(z2 ) + i(ℑ(z1 ) + ℑ(z2 ))) = ℜ(z1 ) + ℜ(z2 )
ℑ(z1 + z2 ) = ℑ(ℜ(z1 ) + ℜ(z2 ) + i(ℑ(z1 ) + ℑ(z2 ))) = ℑ(z1 ) + ℑ(z2 ) □

2.2 z1 + z2 = z2 + z1 , z1 z2 = z2 z1 , z1 (z2 + z3 ) = z1 z2 + z1 z3
z1 + z2 = ℜ(z1 ) + ℜ(zw ) + i(ℑ(z1 ) + ℑ(z2 )) is symmetric.
z1 z2 = ℜ(z1 )ℜ(z2 ) − ℑ(z1 )ℑ(z2 ) + i(ℜ(z1 )ℑ(z2 ) + ℑ(z1 )ℜ(z2 )) is symmetric.
z1 (z2 +z3 ) = ℜ(z1 )ℜ(z2 +z3 )−ℑ(z1 )ℑ(z2 +z3 )+i(ℜ(z1 )ℑ(z2 +z3 )+ℑ(z1 )ℜ(z2 +
z3 )) = ℜ(z1 )(ℜ(z2 ) + ℜ(z3 )) − ℑ(z1 )(ℑ(z2 ) + ℑ(z3 )) + i(ℜ(z1 )(ℑ(z2 ) + ℑ(z3 )) +
ℑ(z1 )(ℜ(z2 ) + ℜ(z3 ))) = z1 z3 + z1 z3 □

2.3 z1 + z2 = z1 + z2 , z1 z2 = z1 · z2 , (z1 /z2 ) = z1 /z2

z1 + z2 = ℜ(z1 ) + ℜ(z2 ) − ℑ(z1 ) + ℑ(z + 2) = z1 + z2
z1 z2 = ℜ(z1 )ℜ(z2 ) − ℑ(z1 )ℑ(z2 ) + i(ℜ(z1 )ℑ(z2 ) + ℜ(z2 )ℑ(z1 ))
z1 z2 = ℜ(z1 )ℜ(z2 ) − ℑ(z1 )ℑ(z2 ) − i(ℜ(z1 )ℑ(z2 ) + ℜ(z2 )ℑ(z1 ))
= z1 · z2 = ℜ(z1 )ℜ(z2 ) − (−ℑ(z1 ))(−ℑ(z2 )) + i(ℜ(z1 )(−ℑ(z2 )) + ℜ(z2 )(−ℑ(z1 )))

z1 1 1 z1
( ) = z1 z2 2
= z1 z2 =
z2 |z2 | z2 z2 z2
□

2.4 |z1 z2 | = |z1 ||z2 |, |z1 |/|z2 | = |z1 /z2 |

2
|z1 z2 | = z1 z1 z2 z2 = |z1 |2 |z2 |2

z1 2 z1 z1 |z1 |2
| | = ( )=
z2 z2 z2 |z2 |2

2.5 ℑ(z) = |z| sin(arg(z)), ℜ(z) = |z| cos(arg(z))

Use Cartesian coordinate system. □

2.6 arg(z1 z2 ) ≡ arg(z1 )+arg(z2 ) (mod 2π), arg(z1 /z2 ) ≡ arg(z1 )−

arg(z2 ) (mod 2π)
z1 z2 = ℜ(z1 )ℜ(z2 ) − ℑ(z1 )ℑ(z2 ) + i(ℜ(z1 )ℑ(z2 ) + ℜ(z2 )ℑ(z1 ))
Let us apply some trigonometric identities.
|z1 z2 | sin(arg(z1 z2 )) = ℑ(z1 z2 ) = ℜ(z1 )ℑ(z2 )+ℜ(z2 )ℑ(z1 ) = |z1 z2 |(cos(arg(z1 )) sin(arg(z2 ))+
sin(arg(z1 )) cos(arg(z2 ))) = |z1 z2 | sin(arg(z1 ) + arg(z2 )).
|z1 z2 | cos(arg(z1 z2 )) = ℜ(z1 z2 ) = ℜ(z1 )ℜ(z2 )−ℑ(z1 )ℑ(z2 ) = |z1 z2 |(cos(arg(z1 )) cos(arg(z2 ))−

3
sin(arg(z1 )) sin(arg(z2 ))) = |z1 z2 | cos(arg(z1 ) + arg(z2 )).
Hence, we have sin(arg(z1 z2 )) = sin(arg(z1 ) + arg(z2 )) and cos(arg(z1 z2 )) =
cos(arg(z1 ) + arg(z2 )).
arg(z1 /z2 ) + arg(z2 ) = arg(z1 /z2 · z2 ) = arg(z1 ). □

3 Basic propositions for complex bashing.

For the rest of the document, unless stated otherwise, lowercase letters,
except for i and z, are complex numbers representing points with the same
name in uppercase in the complex plane.
Example: a for point A.

3.1 Line AB and CD are parallel ⇔ (a − b)/(c − d) ∈ R.

arg((a − b)/(c − d)) ≡ arg(a − b) − arg(c − d) ≡ 0 (mod 2π) ⇔ arg(a − b) ≡
arg(c − d) (mod 2π).
arg(a − b) is the directed angle that a − b makes with the real axis. Therefore,
it is the directed angle that a makes with with a line parallel to the real axis
running through b so it is also the directed angle that line AB makes with the
real axis.
Similarly, arg(c − d) is the directed angle that line CD makes with the real axis.
Hence, Line AB is parallel to BC ⇔ arg(a − b) ≡ arg(c − d) (mod 2π). □

3.2 A, B, and C are collinear ⇔ (a − b)/(a − c) ∈ R.

A, B, and C are collinear if AB, AC are parallel. □

3.3 Line AB is perpendicular to CD ⇔ (a − b)/(c − d) ∈ iR.

Similar to the parallel proof, we have: arg(a − b) and arg(c − d) are respectively
directed angles that AB, CD make with the real axis. Hence, AB and CD are
perpendicular ⇔ arg(a − b) − arg(c − d) ≡ π/2 (mod 2π). □

3.4 A, B, C, and D are concyclic ⇔

a−b d−b
/ ∈R
a−c d−c
Similar to the parallel proof, we have arg(a − b), arg(a − c) are respectively the
directed angle that line AB, AC make with the real axis. Therefore, arg((a −
b)/(a − c)) is the directed angle that line AB makes with line AC.
Similarly, we have arg((d − b)/(d − c)) is the directed angle that line DB makes
with line DC.
Therefore, directed angles ∡BAC = ∡BDC
a−b d−b
arg( / ) = 0. □
a−c d−c

4
3.5 P ′ is the reflection of P with repsect to AB, H is the
image of P on AB:
(a − b)p + ab − ab (a − b)p + (a − b)p + ab − ab
p′ = ,h = .
a−b 2(a − b)

(a − b)p + ab − ab p + p′ (a − b)p + (a − b)p + ab − ab

Let p′ = ,h = = .
a−b 2 2(a − b)
The problem can be reduced to these following claims.

Claim 1: P P ′ is perpendicular to AB.

p − p′ (a − b)p − (a − b)p + ab − ab (a − b)p − (a − b)p + ab − ab a−b

= / =− .
p−p ′ a−b a−b a−b

p − p′ p − p′ p − p′
Hence, we have =− or ∈ iR. □
a−b a−b a−b

Claim 2: H is a point on AB.

h−a (a − b)p + (a − b)(p − 2a) + ab − ab (a − b)p + (a − b)(p − 2b) + ab − ab

= /
h−b 2(a − b) 2(a − b)

(a − b)p + (a − b)p − 2bb − ab − ab h−a

= = .□
(a − b)p + (a − b)p − 2aa − ab − ab h−b

3.6 Q is an intersection of line AB and line CD:

(ab − ab)(c − d) − (a − b)(cd − cd)
q= .
(a − b)(c − d) − (a − b)(c − d)

(ab − ab)(c − d) − (a − b)(cd − cd)

Let q = .
(a − b)(c − d) − (a − b)(c − d)
Since replacing {a, b} with {c, d} does not change q, it is suffices to only prove
A, B and Q are collinear.

q−a (ab − ab − a(a − b))(c − d) − (a − b)(cd − cd − a(c − d))

=
q−b (ab − ab − b(a − b))(c − d) − (a − b)(cd − cd − b(c − d))

−a(a − b)(c − d) − (a − b)(cd − cd − a(c − d))

=
−b(a − b)(c − d) − (a − b)(cd − cd − b(c − d))
cd − cd + a(c − d) − a(c − d) q−a
= = .□
cd − cd + b(c − d) − b(c − d) q−b

5
 a+b+c
3.7 G is the centroid of △ABC: g = .
3
Use Cartesian coordinate system or use Euler’s line. □

3.8 (Personally favourite) △ABC ∼

= △DEF ⇔
a−b d−e
= .
c−b f −e

|a − b| |c − b|

 = a−b d−e
|d − e| |f − e|

⇔ = .□
arg( a − b ) = arg( d − e )
 c−b f −e
c−b f −e

4 The unit circle.

A unit circle is a circle with 0 as the center (o = 0) and a radius of 1. The unit
circle holds tremendous significance when for complex bashing as it allows for
many expressions and formulas to be heavily simplified and shrunk down.
The equation of the unit circle: |z|2 = zz = 1.
A chord (on the unit circle) is a line segment from 2 points on the unit circle.

4.1 P ′ is the reflection of P with respect to chord AB,

H is the image of P on chord AB: p′ = a + b − abp,
p + a + b − abp
h= .
2
Let P ′ be the reflection of P with respect to chord AB, H be the image of P
on chord AB.
1
A and B are points on the unit circle so u = for u ∈ {a, b}.
u
b a
(a − b)p + ab − ab (a − b)p + −
′
p = = a b = −abp + a + b.
a−b 1 1
−
a b
p + p′ −abp + a + b + p
H is the image of P on chord AB so h = = . □
2 2

4.2 P is a point on (line through) chord AB ⇔ p+abp = a+b.

Let P ′ be the reflection of P with respect to chord AB.
p′ = a + b − abp
Therefore, P is a point on (line through) chord AB ⇔ p = p′ = a + b − abp. □

6
4.3 Equation of a line tangent to the unit circle at A:
z + a2 z = 2a.
For all points Z on the tangent to the unit circle at A, we have: ZA ⊥ OA.
a−z a−z
⇔ =− = az − 1. □
a a

4.4 P is the intersection of 2 lines tangent to the unit

2ab
circle at A and B: p = .
a+b
Let P be the intersection 2 lines tangent to the unit circle at A and B.

2a2 b − 2ab2 2ab

p + a2 p = 2a, p + b2 p = 2b ⇒ p = 2 2
= .□
a −b a+b

4.5 Q is an intersection of (line through) chord AB and

ab(c + d) − cd(a + b)
(line through) chord CD: q = .
ab − cd
ab(c + d) − cd(a + b)
q + abq = a + b, q + cdq = c + d ⇒ q = . □
ab − cd

4.6 H is the orthocenter of △ABC inscribed inside the

unit circle: h = a + b + c.
h−a b+c h−a
= =− .
b−c b−c b−c
Therefore, HA ⊥ BC. Similarly, HC ⊥ AB. □