Skip to main content

Complex Numbers

Complex Numbers

Complex numbers extend the real number system by introducing a solution to the equation x2+1=0x^2 + 1 = 0. This seemingly abstract idea turns out to be indispensable in pure mathematics, engineering, and physics, providing powerful tools for analysing rotations, oscillations, and polynomial equations.

Board Coverage

BoardPaperNotes
AQAPaper 1Full coverage including transformations w=f(z)w = f(z)
EdexcelFP1/FP2De Moivre, roots of unity, loci in FP2
OCR (A)Paper 1De Moivre, exponential form, roots of unity
CIEP1Loci required; exponential form and roots of unity in depth

1. Review of A Level Complex Numbers

Definition. A complex number is a number of the form z=a+biz = a + bi where a,bRa, b \in \mathbb{R} and ii is defined by the property i2=1i^2 = -1. The set of all complex numbers is denoted C\mathbb{C}.

We call aa the real part (written Re(z)\operatorname{Re}(z)) and bb the imaginary part (written Im(z)\operatorname{Im}(z)). Two complex numbers are equal if and only if both their real and imaginary parts are equal.

1.1 The Argand Diagram

Definition. The Argand diagram is a representation of C\mathbb{C} as a plane, where the horizontal axis represents Re(z)\operatorname{Re}(z) and the vertical axis represents Im(z)\operatorname{Im}(z). The complex number z=a+biz = a + bi is plotted as the point (a,b)(a, b).

1.2 Modulus and Argument

Definition. For z=a+bi0z = a + bi \neq 0:

  • The modulus z|z| is defined as z=a2+b2|z| = \sqrt{a^2 + b^2}.
  • The argument arg(z)\arg(z) is the angle θ\theta (measured anticlockwise from the positive real axis) such that tanθ=ba\tan\theta = \dfrac{b}{a}, with π<θπ-\pi < \theta \leq \pi (the principal argument).

z=a+bi=z(cosθ+isinθ)=r(cosθ+isinθ)\boxed{z = a + bi = |z|(\cos\theta + i\sin\theta) = r(\cos\theta + i\sin\theta)}

where r=zr = |z| and θ=arg(z)\theta = \arg(z).

1.3 Arithmetic with Complex Numbers

For z1=a+biz_1 = a + bi and z2=c+diz_2 = c + di:

z1+z2=(a+c)+(b+d)iz1z2=(acbd)+(ad+bc)iz1z2=LBz1z2RB◆◆LBz22RB=(ac+bd)+(bcad)ic2+d2\begin{aligned} z_1 + z_2 &= (a + c) + (b + d)i \\ z_1 z_2 &= (ac - bd) + (ad + bc)i \\ \frac{z_1}{z_2} &= \frac◆LB◆z_1 \overline{z_2}◆RB◆◆LB◆|z_2|^2◆RB◆ = \frac{(ac + bd) + (bc - ad)i}{c^2 + d^2} \end{aligned}

where z2=cdi\overline{z_2} = c - di is the complex conjugate of z2z_2.

warning θ=arctan(b/a)\theta = \arctan(b/a) only gives the correct principal argument when a>0a > 0.
Worked Example: Modulus, argument, and polar form

Find the modulus, argument, and polar form of z=13iz = 1 - \sqrt{3}\,i.

z=LB12+(3)2RB=1+3=2|z| = \sqrt◆LB◆1^2 + (-\sqrt{3})^2◆RB◆ = \sqrt{1 + 3} = 2

Since (a,b)=(1,3)(a, b) = (1, -\sqrt{3}) lies in the fourth quadrant:

arg(z)=arctan ⁣(LB3RB◆◆LB1RB)=LBπRB◆◆LB3RB\arg(z) = \arctan\!\left(\frac◆LB◆-\sqrt{3}◆RB◆◆LB◆1◆RB◆\right) = -\frac◆LB◆\pi◆RB◆◆LB◆3◆RB◆

Polar form: z=2 ⁣(cos ⁣(LBπRB◆◆LB3RB)+isin ⁣(LBπRB◆◆LB3RB))z = 2\!\left(\cos\!\left(-\dfrac◆LB◆\pi◆RB◆◆LB◆3◆RB◆\right) + i\sin\!\left(-\dfrac◆LB◆\pi◆RB◆◆LB◆3◆RB◆\right)\right).


2. De Moivre's Theorem

Theorem (De Moivre). For any integer nn and any angle θ\theta:

(cosθ+isinθ)n=cos(nθ)+isin(nθ)\boxed{\left(\cos\theta + i\sin\theta\right)^n = \cos(n\theta) + i\sin(n\theta)}

Proof of De Moivre's Theorem (by induction for n0n \geq 0)

Base case (n=0n = 0): (cosθ+isinθ)0=1=cos0+isin0(\cos\theta + i\sin\theta)^0 = 1 = \cos 0 + i\sin 0. ✓

Inductive step. Assume (cosθ+isinθ)k=cos(kθ)+isin(kθ)(\cos\theta + i\sin\theta)^k = \cos(k\theta) + i\sin(k\theta) for some k0k \geq 0. Then:

(cosθ+isinθ)k+1=(cosθ+isinθ)k(cosθ+isinθ)=[cos(kθ)+isin(kθ)][cosθ+isinθ]=cos(kθ)cosθsin(kθ)sinθ+i[cos(kθ)sinθ+sin(kθ)cosθ]=cos((k+1)θ)+isin((k+1)θ)\begin{aligned} (\cos\theta + i\sin\theta)^{k+1} &= (\cos\theta + i\sin\theta)^k \cdot (\cos\theta + i\sin\theta) \\ &= [\cos(k\theta) + i\sin(k\theta)] \cdot [\cos\theta + i\sin\theta] \\ &= \cos(k\theta)\cos\theta - \sin(k\theta)\sin\theta + i\,[\cos(k\theta)\sin\theta + \sin(k\theta)\cos\theta] \\ &= \cos((k+1)\theta) + i\sin((k+1)\theta) \end{aligned}

using the compound angle identities. ✓

For negative integers, note that LB1RB◆◆LBcosθ+isinθRB=cosθisinθ=cos(θ)+isin(θ)\dfrac◆LB◆1◆RB◆◆LB◆\cos\theta + i\sin\theta◆RB◆ = \cos\theta - i\sin\theta = \cos(-\theta) + i\sin(-\theta), so the result follows. \square

Intuition. De Moivre's theorem says that raising a complex number on the unit circle to the nn-th power simply multiplies its angle by nn. This is because multiplication of complex numbers adds their arguments: r1eiθ1r2eiθ2=r1r2ei(θ1+θ2)r_1 e^{i\theta_1} \cdot r_2 e^{i\theta_2} = r_1 r_2 e^{i(\theta_1 + \theta_2)}.

2.1 Applications: Trigonometric Identities

De Moivre's theorem provides a systematic way to derive multiple-angle identities.

Example. Express cos3θ\cos 3\theta and sin3θ\sin 3\theta in terms of cosθ\cos\theta and sinθ\sin\theta.

By De Moivre: cos3θ+isin3θ=(cosθ+isinθ)3\cos 3\theta + i\sin 3\theta = (\cos\theta + i\sin\theta)^3.

Expanding the RHS using the binomial theorem:

(cosθ+isinθ)3=cos3θ+3icos2θsinθ3cosθsin2θisin3θ\begin{aligned} (\cos\theta + i\sin\theta)^3 &= \cos^3\theta + 3i\cos^2\theta\sin\theta - 3\cos\theta\sin^2\theta - i\sin^3\theta \end{aligned}

Equating real and imaginary parts:

cos3θ=4cos3θ3cosθ\boxed{\cos 3\theta = 4\cos^3\theta - 3\cos\theta}

sin3θ=3sinθ4sin3θ\boxed{\sin 3\theta = 3\sin\theta - 4\sin^3\theta}

2.2 Powers of Complex Numbers

To compute znz^n where z=r(cosθ+isinθ)z = r(\cos\theta + i\sin\theta):

zn=rn(cos(nθ)+isin(nθ))z^n = r^n\left(\cos(n\theta) + i\sin(n\theta)\right)

Worked Example: Computing a high power

Find (1+i)10(1 + i)^{10}.

First write in polar form: 1+i=2 ⁣(cosLBπRB◆◆LB4RB+isinLBπRB◆◆LB4RB)1 + i = \sqrt{2}\!\left(\cos\frac◆LB◆\pi◆RB◆◆LB◆4◆RB◆ + i\sin\frac◆LB◆\pi◆RB◆◆LB◆4◆RB◆\right).

(1+i)10=(2)10 ⁣(cosLB10πRB◆◆LB4RB+isinLB10πRB◆◆LB4RB)=32 ⁣(cosLB5πRB◆◆LB2RB+isinLB5πRB◆◆LB2RB)=32 ⁣(cosLBπRB◆◆LB2RB+isinLBπRB◆◆LB2RB)=32(0+i)=32i\begin{aligned} (1 + i)^{10} &= \left(\sqrt{2}\right)^{10}\!\left(\cos\frac◆LB◆10\pi◆RB◆◆LB◆4◆RB◆ + i\sin\frac◆LB◆10\pi◆RB◆◆LB◆4◆RB◆\right) \\ &= 32\!\left(\cos\frac◆LB◆5\pi◆RB◆◆LB◆2◆RB◆ + i\sin\frac◆LB◆5\pi◆RB◆◆LB◆2◆RB◆\right) \\ &= 32\!\left(\cos\frac◆LB◆\pi◆RB◆◆LB◆2◆RB◆ + i\sin\frac◆LB◆\pi◆RB◆◆LB◆2◆RB◆\right) \\ &= 32(0 + i) = 32i \end{aligned}

3. Roots of Unity

Definition. The nn-th roots of unity are the solutions to the equation zn=1z^n = 1 for nZ+n \in \mathbb{Z}^+.

By De Moivre's theorem, writing 1=cos0+isin0=cos(2kπ)+isin(2kπ)1 = \cos 0 + i\sin 0 = \cos(2k\pi) + i\sin(2k\pi) for any integer kk, the nn distinct solutions are:

zk=cos ⁣(LB2kπRB◆◆LBnRB)+isin ⁣(LB2kπRB◆◆LBnRB),k=0,1,2,,n1\boxed{z_k = \cos\!\left(\frac◆LB◆2k\pi◆RB◆◆LB◆n◆RB◆\right) + i\sin\!\left(\frac◆LB◆2k\pi◆RB◆◆LB◆n◆RB◆\right), \quad k = 0, 1, 2, \ldots, n-1}

3.1 Geometric Interpretation

The nn-th roots of unity lie on the unit circle z=1|z| = 1 in the Argand diagram, equally spaced at angles of LB2πRB◆◆LBnRB\dfrac◆LB◆2\pi◆RB◆◆LB◆n◆RB◆ radians apart. They form the vertices of a regular nn-gon inscribed in the unit circle, with one vertex at z=1z = 1.

3.2 Sum and Product of Roots

Since the roots satisfy zn1=0z^n - 1 = 0, the sum of all nn-th roots of unity is zero:

k=0n1zk=0\sum_{k=0}^{n-1} z_k = 0

This follows from the coefficient of zn1z^{n-1} in zn1=0z^n - 1 = 0 being zero (by Vieta's formulas). Equivalently, the roots form a geometric series with ratio ω=e2πi/n\omega = e^{2\pi i/n} and first term 1, giving:

k=0n1ωk=LB1ωnRB◆◆LB1ωRB=LB11RB◆◆LB1ωRB=0\sum_{k=0}^{n-1} \omega^k = \frac◆LB◆1 - \omega^n◆RB◆◆LB◆1 - \omega◆RB◆ = \frac◆LB◆1 - 1◆RB◆◆LB◆1 - \omega◆RB◆ = 0

The product of all nn-th roots of unity is:

k=0n1zk=(1)n1\prod_{k=0}^{n-1} z_k = (-1)^{n-1}

Worked Example: Cube roots of unity

Find all cube roots of unity and verify that their sum is zero.

z3=1    zk=cos ⁣(LB2kπRB◆◆LB3RB)+isin ⁣(LB2kπRB◆◆LB3RB)z^3 = 1 \implies z_k = \cos\!\left(\dfrac◆LB◆2k\pi◆RB◆◆LB◆3◆RB◆\right) + i\sin\!\left(\dfrac◆LB◆2k\pi◆RB◆◆LB◆3◆RB◆\right) for k=0,1,2k = 0, 1, 2.

z0=cos0+isin0=1z1=cosLB2πRB◆◆LB3RB+isinLB2πRB◆◆LB3RB=12+LB3RB◆◆LB2RBiz2=cosLB4πRB◆◆LB3RB+isinLB4πRB◆◆LB3RB=12LB3RB◆◆LB2RBi\begin{aligned} z_0 &= \cos 0 + i\sin 0 = 1 \\ z_1 &= \cos\frac◆LB◆2\pi◆RB◆◆LB◆3◆RB◆ + i\sin\frac◆LB◆2\pi◆RB◆◆LB◆3◆RB◆ = -\frac{1}{2} + \frac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆\,i \\ z_2 &= \cos\frac◆LB◆4\pi◆RB◆◆LB◆3◆RB◆ + i\sin\frac◆LB◆4\pi◆RB◆◆LB◆3◆RB◆ = -\frac{1}{2} - \frac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆\,i \end{aligned}

Sum: 112+LB3RB◆◆LB2RBi12LB3RB◆◆LB2RBi=01 - \dfrac{1}{2} + \dfrac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆\,i - \dfrac{1}{2} - \dfrac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆\,i = 0. ✓

Product: 1(12+LB3RB◆◆LB2RBi)(12LB3RB◆◆LB2RBi)=1(14+34)=1=(1)311 \cdot \left(-\dfrac{1}{2} + \dfrac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆\,i\right)\left(-\dfrac{1}{2} - \dfrac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆\,i\right) = 1 \cdot \left(\dfrac{1}{4} + \dfrac{3}{4}\right) = 1 = (-1)^{3-1}. ✓


4. Exponential Form

Definition. The exponential form of a complex number z=r(cosθ+isinθ)z = r(\cos\theta + i\sin\theta) is:

z=reiθ\boxed{z = re^{i\theta}}

where eiθcosθ+isinθe^{i\theta} \equiv \cos\theta + i\sin\theta by Euler's formula.

Proof of Euler's Formula (from Maclaurin series)

The Maclaurin series for exe^x, cosx\cos x, and sinx\sin x are:

ex=n=0xnn!=1+x+x22!+x33!+x44!+e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!} = 1 + x + \frac{x^2}{2!} + \frac{x^3}{3!} + \frac{x^4}{4!} + \cdots

cosx=n=0(1)nx2n(2n)!=1x22!+x44!x66!+\cos x = \sum_{n=0}^{\infty} \frac{(-1)^n x^{2n}}{(2n)!} = 1 - \frac{x^2}{2!} + \frac{x^4}{4!} - \frac{x^6}{6!} + \cdots

sinx=n=0(1)nx2n+1(2n+1)!=xx33!+x55!x77!+\sin x = \sum_{n=0}^{\infty} \frac{(-1)^n x^{2n+1}}{(2n+1)!} = x - \frac{x^3}{3!} + \frac{x^5}{5!} - \frac{x^7}{7!} + \cdots

Substituting x=iθx = i\theta into the series for exe^x:

eiθ=1+iθ+LB(iθ)2RB◆◆LB2!RB+LB(iθ)3RB◆◆LB3!RB+LB(iθ)4RB◆◆LB4!RB+LB(iθ)5RB◆◆LB5!RB+=1+iθ+LBi2θ2RB◆◆LB2!RB+LBi3θ3RB◆◆LB3!RB+LBi4θ4RB◆◆LB4!RB+LBi5θ5RB◆◆LB5!RB+=1+iθLBθ2RB◆◆LB2!RBLBiθ3RB◆◆LB3!RB+LBθ4RB◆◆LB4!RB+LBiθ5RB◆◆LB5!RB\begin{aligned} e^{i\theta} &= 1 + i\theta + \frac◆LB◆(i\theta)^2◆RB◆◆LB◆2!◆RB◆ + \frac◆LB◆(i\theta)^3◆RB◆◆LB◆3!◆RB◆ + \frac◆LB◆(i\theta)^4◆RB◆◆LB◆4!◆RB◆ + \frac◆LB◆(i\theta)^5◆RB◆◆LB◆5!◆RB◆ + \cdots \\ &= 1 + i\theta + \frac◆LB◆i^2\theta^2◆RB◆◆LB◆2!◆RB◆ + \frac◆LB◆i^3\theta^3◆RB◆◆LB◆3!◆RB◆ + \frac◆LB◆i^4\theta^4◆RB◆◆LB◆4!◆RB◆ + \frac◆LB◆i^5\theta^5◆RB◆◆LB◆5!◆RB◆ + \cdots \\ &= 1 + i\theta - \frac◆LB◆\theta^2◆RB◆◆LB◆2!◆RB◆ - \frac◆LB◆i\theta^3◆RB◆◆LB◆3!◆RB◆ + \frac◆LB◆\theta^4◆RB◆◆LB◆4!◆RB◆ + \frac◆LB◆i\theta^5◆RB◆◆LB◆5!◆RB◆ - \cdots \end{aligned}

using i2=1i^2 = -1, i3=ii^3 = -i, i4=1i^4 = 1, i5=ii^5 = i, and so on. Grouping real and imaginary parts:

eiθ=LB(1LBθ2RB◆◆LB2!RB+LBθ4RB◆◆LB4!RB)RB=cosθ+iLB(θLBθ3RB◆◆LB3!RB+LBθ5RB◆◆LB5!RB)RB=sinθe^{i\theta} = \underbrace◆LB◆\left(1 - \frac◆LB◆\theta^2◆RB◆◆LB◆2!◆RB◆ + \frac◆LB◆\theta^4◆RB◆◆LB◆4!◆RB◆ - \cdots\right)◆RB◆_{=\,\cos\theta} + i\underbrace◆LB◆\left(\theta - \frac◆LB◆\theta^3◆RB◆◆LB◆3!◆RB◆ + \frac◆LB◆\theta^5◆RB◆◆LB◆5!◆RB◆ - \cdots\right)◆RB◆_{=\,\sin\theta}

Therefore eiθ=cosθ+isinθe^{i\theta} = \cos\theta + i\sin\theta. \square

4.1 Euler's Identity

Setting θ=π\theta = \pi:

eiπ+1=0\boxed{e^{i\pi} + 1 = 0}

This celebrated identity connects five fundamental constants: ee, ii, π\pi, 11, and 00.

Intuition. Euler's identity says that starting at the point 11 on the real axis and rotating by π\pi radians (half a turn) on the unit circle lands you at 1-1. The exponential eiθe^{i\theta} describes a point moving around the unit circle at a rate determined by θ\theta.

4.2 Exponential Rules for Complex Numbers

The standard laws of indices extend naturally:

z1z2=r1r2ei(θ1+θ2)(argumentsadd)z1z2=r1r2ei(θ1θ2)(argumentssubtract)zn=rneinθ(argumentmultiplies)\begin{aligned} z_1 z_2 &= r_1 r_2 \, e^{i(\theta_1 + \theta_2)} & &\mathrm{(arguments add)} \\ \frac{z_1}{z_2} &= \frac{r_1}{r_2} \, e^{i(\theta_1 - \theta_2)} & &\mathrm{(arguments subtract)} \\ z^n &= r^n e^{in\theta} & &\mathrm{(argument multiplies)} \end{aligned}
info with the cosθ+isinθ\cos\theta + i\sin\theta form. All boards require De Moivre's theorem. :::

5. Loci in the Argand Diagram

5.1 Circles: za=r|z - a| = r

Definition. The locus za=r|z - a| = r, where aCa \in \mathbb{C} and rR+r \in \mathbb{R}^+, is a circle with centre aa and radius rr in the Argand diagram.

za=r    LB(xα)2+(yβ)2RB=r    (xα)2+(yβ)2=r2|z - a| = r \iff \sqrt◆LB◆(x - \alpha)^2 + (y - \beta)^2◆RB◆ = r \iff (x - \alpha)^2 + (y - \beta)^2 = r^2

where a=α+βia = \alpha + \beta i and z=x+yiz = x + yi.

The region za<r|z - a| < r is the interior of the circle, and za>r|z - a| > r is the exterior.

5.2 Perpendicular Bisectors: za=zb|z - a| = |z - b|

This locus represents all points equidistant from aa and bb, which is the perpendicular bisector of the line segment joining aa and bb.

5.3 Half-Lines: arg(za)=α\arg(z - a) = \alpha

Definition. The locus arg(za)=α\arg(z - a) = \alpha is a half-line (ray) starting from the point aa (not including aa itself) making an angle α\alpha with the positive real direction.

The region α1<arg(za)<α2\alpha_1 < \arg(z - a) < \alpha_2 is an angular sector (wedge) with vertex at aa.

5.4 Combined Loci and Regions

Exam questions often require describing a region defined by combining loci, such as:

  • z32|z - 3| \leq 2 and arg(z)LBπRB◆◆LB4RB\arg(z) \geq \dfrac◆LB◆\pi◆RB◆◆LB◆4◆RB◆: the part of a disc in a sector.
A half-line arg(za)=α\arg(z - a) = \alpha does not include the point aa. When shading

regions, be careful about whether boundaries are included (solid line) or excluded (dashed line).

Always sketch loci problems. The algebraic description follows from the geometric picture.

Common exam technique: identify the boundary (circle, line, half-line), then determine which side of the boundary is included by testing a point. :::

Worked Example: Describing a locus algebraically

A complex number zz satisfies z2i3|z - 2i| \leq 3 and 0arg(z)LBπRB◆◆LB2RB0 \leq \arg(z) \leq \dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆. Find the greatest possible value of z|z| and the least possible value of z|z|.

The first condition: z2i3|z - 2i| \leq 3 is the closed disc of radius 3 centred at 2i2i, i.e. at (0,2)(0, 2).

The second condition: 0arg(z)LBπRB◆◆LB2RB0 \leq \arg(z) \leq \dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆ restricts zz to the first quadrant (including axes).

The disc centre (0,2)(0, 2) with radius 3 extends from y=1y = -1 to y=5y = 5 and from x=3x = -3 to x=3x = 3.

Greatest z|z|: The point in the region farthest from the origin is where the boundary of the disc intersects the first quadrant boundary furthest from the origin. The disc intersects the positive yy-axis at (0,5)(0, 5), giving z=5|z| = 5.

Least z|z|: We need the closest point in the region to the origin. The disc boundary is (x2+(y2)2)=9(x^2 + (y-2)^2) = 9. The closest point on this circle to the origin lies along the line from the origin through the centre (0,2)(0,2), which is the yy-axis. The point (0,1)(0, -1) is outside the first quadrant. Within the first quadrant, the closest point is where the circle meets the xx-axis: setting y=0y = 0, x2+4=9    x=5x^2 + 4 = 9 \implies x = \sqrt{5}. So z=5|z| = \sqrt{5}.


6. Complex Transformations

6.1 The Mapping w=f(z)w = f(z)

A complex transformation is a function w=f(z)w = f(z) that maps points in the zz-plane (Argand diagram for zz) to points in the ww-plane (Argand diagram for ww).

6.2 Linear Transformations: w=az+bw = az + b

For w=az+bw = az + b where a,bCa, b \in \mathbb{C} and a0a \neq 0:

Writing a=λeiαa = \lambda e^{i\alpha} and b=μeiβb = \mu e^{i\beta}:

  1. a=λ|a| = \lambda produces an enlargement (scale factor λ\lambda) about the origin.
  2. arg(a)=α\arg(a) = \alpha produces a rotation through angle α\alpha about the origin.
  3. bb produces a translation by the vector representing bb.

The composition is: enlarge by a|a|, rotate by arg(a)\arg(a), then translate by bb.

6.3 Inversion: w=1zw = \dfrac{1}{z}

The transformation w=1zw = \dfrac{1}{z} maps:

  • Circles not through the origin to circles.
  • Circles through the origin to straight lines not through the origin.
  • Straight lines through the origin to straight lines through the origin.
  • Straight lines not through the origin to circles through the origin.

6.4 Reciprocal: w=z+1zw = z + \dfrac{1}{z} and w=z1z+1w = \dfrac{z - 1}{z + 1}

These are common in exam questions. The general approach is:

  1. Express zz in terms of ww: z=f1(w)z = f^{-1}(w).
  2. Apply the given condition on zz (e.g. z=2|z| = 2) to find the locus of ww.

Critical points. A critical point of a transformation w=f(z)w = f(z) is a point z0z_0 where f(z0)=0f'(z_0) = 0. At a critical point, the mapping is not conformal (angles are not preserved).

Worked Example: Image of a line under inversion

Find the image of the line Re(z)=1\operatorname{Re}(z) = 1 under the transformation w=1zw = \dfrac{1}{z}.

Let z=x+yiz = x + yi with x=1x = 1, so z=1+yiz = 1 + yi and yRy \in \mathbb{R}.

w=11+yi=1yi1+y2=11+y2y1+y2iw = \frac{1}{1 + yi} = \frac{1 - yi}{1 + y^2} = \frac{1}{1 + y^2} - \frac{y}{1 + y^2}\,i

Let w=u+viw = u + vi. Then u=11+y2u = \dfrac{1}{1 + y^2} and v=y1+y2v = \dfrac{-y}{1 + y^2}.

Note that v=uyv = -uy, so y=vuy = -\dfrac{v}{u} (when u0u \neq 0).

Substituting: u=11+v2/u2=u2u2+v2u = \dfrac{1}{1 + v^2/u^2} = \dfrac{u^2}{u^2 + v^2}, giving u2+v2=uu^2 + v^2 = u, i.e.:

u2u+v2=0    (u12)2+v2=14u^2 - u + v^2 = 0 \implies \left(u - \frac{1}{2}\right)^2 + v^2 = \frac{1}{4}

This is a circle with centre (12,0)\left(\dfrac{1}{2}, 0\right) and radius 12\dfrac{1}{2} in the ww-plane.

AQA places significant emphasis on complex transformations including w=f(z)w = f(z) mappings.

Edexcel and OCR cover this topic with less depth. CIE focuses more on loci than on transformations.


7. Summary of Key Results

(cosθ+isinθ)n=cos(nθ)+isin(nθ)\boxed{(\cos\theta + i\sin\theta)^n = \cos(n\theta) + i\sin(n\theta)}

eiθ=cosθ+isinθ\boxed{e^{i\theta} = \cos\theta + i\sin\theta}

eiπ+1=0\boxed{e^{i\pi} + 1 = 0}

zk=e2kπi/n=cos ⁣(LB2kπRB◆◆LBnRB)+isin ⁣(LB2kπRB◆◆LBnRB),k=0,1,,n1\boxed{z_k = e^{2k\pi i/n} = \cos\!\left(\frac◆LB◆2k\pi◆RB◆◆LB◆n◆RB◆\right) + i\sin\!\left(\frac◆LB◆2k\pi◆RB◆◆LB◆n◆RB◆\right), \quad k = 0, 1, \ldots, n-1}

k=0n1zk=0\boxed{\sum_{k=0}^{n-1} z_k = 0}

za=r    circlecentrearadiusr\boxed{|z - a| = r \iff \mathrm{circle centre } a \mathrm{ radius } r}

arg(za)=α    halflinefromaatangleα\boxed{\arg(z - a) = \alpha \iff \mathrm{half-line from } a \mathrm{ at angle } \alpha}


Problems

Problem 1. Express z=3+iz = -\sqrt{3} + i in modulus-argument form and hence find z5z^5 in the form a+bia + bi.

Hint

Find z|z| and arg(z)\arg(z) first. Then apply De Moivre's theorem.

Answer

z=3+1=2|z| = \sqrt{3 + 1} = 2.

The point (3,1)(-\sqrt{3}, 1) is in the second quadrant. arg(z)=πarctan ⁣(LB1RB◆◆LB3RB)=πLBπRB◆◆LB6RB=LB5πRB◆◆LB6RB\arg(z) = \pi - \arctan\!\left(\dfrac◆LB◆1◆RB◆◆LB◆\sqrt{3}◆RB◆\right) = \pi - \dfrac◆LB◆\pi◆RB◆◆LB◆6◆RB◆ = \dfrac◆LB◆5\pi◆RB◆◆LB◆6◆RB◆.

z5=25 ⁣(cosLB25πRB◆◆LB6RB+isinLB25πRB◆◆LB6RB)=32 ⁣(cosLBπRB◆◆LB6RB+isinLBπRB◆◆LB6RB)=32 ⁣(LB3RB◆◆LB2RB+12i)=163+16iz^5 = 2^5\!\left(\cos\frac◆LB◆25\pi◆RB◆◆LB◆6◆RB◆ + i\sin\frac◆LB◆25\pi◆RB◆◆LB◆6◆RB◆\right) = 32\!\left(\cos\frac◆LB◆\pi◆RB◆◆LB◆6◆RB◆ + i\sin\frac◆LB◆\pi◆RB◆◆LB◆6◆RB◆\right) = 32\!\left(\frac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆ + \frac{1}{2}\,i\right) = 16\sqrt{3} + 16i


Problem 2. Use De Moivre's theorem to prove that cos4θ=8cos4θ8cos2θ+1\cos 4\theta = 8\cos^4\theta - 8\cos^2\theta + 1.

Hint

Expand (cosθ+isinθ)4(\cos\theta + i\sin\theta)^4 using the binomial theorem and equate real parts.

Answer

cos4θ+isin4θ=(cosθ+isinθ)4\cos 4\theta + i\sin 4\theta = (\cos\theta + i\sin\theta)^4.

Expanding: cos4θ+4icos3θsinθ6cos2θsin2θ4icosθsin3θ+sin4θ\cos^4\theta + 4i\cos^3\theta\sin\theta - 6\cos^2\theta\sin^2\theta - 4i\cos\theta\sin^3\theta + \sin^4\theta.

Real part: cos4θ=cos4θ6cos2θsin2θ+sin4θ\cos 4\theta = \cos^4\theta - 6\cos^2\theta\sin^2\theta + \sin^4\theta.

Using sin2θ=1cos2θ\sin^2\theta = 1 - \cos^2\theta:

cos4θ=cos4θ6cos2θ(1cos2θ)+(1cos2θ)2=cos4θ6cos2θ+6cos4θ+12cos2θ+cos4θ=8cos4θ8cos2θ+1\begin{aligned} \cos 4\theta &= \cos^4\theta - 6\cos^2\theta(1 - \cos^2\theta) + (1 - \cos^2\theta)^2 \\ &= \cos^4\theta - 6\cos^2\theta + 6\cos^4\theta + 1 - 2\cos^2\theta + \cos^4\theta \\ &= 8\cos^4\theta - 8\cos^2\theta + 1 \quad \square \end{aligned}

Problem 3. Find all solutions to z4=16iz^4 = 16i, expressing each in the form a+bia + bi.

Hint

Write 16i=16eiπ/216i = 16e^{i\pi/2} and use the roots formula.

Answer

16i=16 ⁣(cosLBπRB◆◆LB2RB+isinLBπRB◆◆LB2RB)16i = 16\!\left(\cos\dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆ + i\sin\dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆\right).

zk=2 ⁣(cos ⁣(LBπ/2+2kπRB◆◆LB4RB)+isin ⁣(LBπ/2+2kπRB◆◆LB4RB)),k=0,1,2,3z_k = 2\!\left(\cos\!\left(\frac◆LB◆\pi/2 + 2k\pi◆RB◆◆LB◆4◆RB◆\right) + i\sin\!\left(\frac◆LB◆\pi/2 + 2k\pi◆RB◆◆LB◆4◆RB◆\right)\right), \quad k = 0, 1, 2, 3

k=0:z0=2 ⁣(cosLBπRB◆◆LB8RB+isinLBπRB◆◆LB8RB)=2 ⁣(LBLB2+2RB◆◆RB◆◆LB2RB+iLBLB22RB◆◆RB◆◆LB2RB)=LB2+2RB+iLB22RBk=1:z1=2 ⁣(cosLB5πRB◆◆LB8RB+isinLB5πRB◆◆LB8RB)=LB22RB+iLB2+2RBk=2:z2=2 ⁣(cosLB9πRB◆◆LB8RB+isinLB9πRB◆◆LB8RB)=LB2+2RBiLB22RBk=3:z3=2 ⁣(cosLB13πRB◆◆LB8RB+isinLB13πRB◆◆LB8RB)=LB22RBiLB2+2RB\begin{aligned} k = 0:\quad z_0 &= 2\!\left(\cos\frac◆LB◆\pi◆RB◆◆LB◆8◆RB◆ + i\sin\frac◆LB◆\pi◆RB◆◆LB◆8◆RB◆\right) = 2\!\left(\frac◆LB◆\sqrt◆LB◆2+\sqrt{2}◆RB◆◆RB◆◆LB◆2◆RB◆ + i\,\frac◆LB◆\sqrt◆LB◆2-\sqrt{2}◆RB◆◆RB◆◆LB◆2◆RB◆\right) = \sqrt◆LB◆2+\sqrt{2}◆RB◆ + i\sqrt◆LB◆2-\sqrt{2}◆RB◆ \\ k = 1:\quad z_1 &= 2\!\left(\cos\frac◆LB◆5\pi◆RB◆◆LB◆8◆RB◆ + i\sin\frac◆LB◆5\pi◆RB◆◆LB◆8◆RB◆\right) = -\sqrt◆LB◆2-\sqrt{2}◆RB◆ + i\sqrt◆LB◆2+\sqrt{2}◆RB◆ \\ k = 2:\quad z_2 &= 2\!\left(\cos\frac◆LB◆9\pi◆RB◆◆LB◆8◆RB◆ + i\sin\frac◆LB◆9\pi◆RB◆◆LB◆8◆RB◆\right) = -\sqrt◆LB◆2+\sqrt{2}◆RB◆ - i\sqrt◆LB◆2-\sqrt{2}◆RB◆ \\ k = 3:\quad z_3 &= 2\!\left(\cos\frac◆LB◆13\pi◆RB◆◆LB◆8◆RB◆ + i\sin\frac◆LB◆13\pi◆RB◆◆LB◆8◆RB◆\right) = \sqrt◆LB◆2-\sqrt{2}◆RB◆ - i\sqrt◆LB◆2+\sqrt{2}◆RB◆ \end{aligned}

Problem 4. The fifth roots of unity are ω0,ω1,ω2,ω3,ω4\omega^0, \omega^1, \omega^2, \omega^3, \omega^4 where ω=e2πi/5\omega = e^{2\pi i/5}. Show that 1+ω+ω2+ω3+ω4=01 + \omega + \omega^2 + \omega^3 + \omega^4 = 0 and deduce that cosLB2πRB◆◆LB5RB+cosLB4πRB◆◆LB5RB=12\cos\dfrac◆LB◆2\pi◆RB◆◆LB◆5◆RB◆ + \cos\dfrac◆LB◆4\pi◆RB◆◆LB◆5◆RB◆ = -\dfrac{1}{2}.

Hint

Sum the geometric series. Then separate real and imaginary parts.

Answer

The roots satisfy z51=0z^5 - 1 = 0. The coefficient of z4z^4 is 0, so by Vieta's formulas, k=04ωk=0\sum_{k=0}^{4} \omega^k = 0.

Alternatively: k=04ωk=LB1ω5RB◆◆LB1ωRB=LB11RB◆◆LB1ωRB=0\displaystyle\sum_{k=0}^{4}\omega^k = \frac◆LB◆1 - \omega^5◆RB◆◆LB◆1 - \omega◆RB◆ = \frac◆LB◆1 - 1◆RB◆◆LB◆1 - \omega◆RB◆ = 0.

Expanding using ωk=cosLB2kπRB◆◆LB5RB+isinLB2kπRB◆◆LB5RB\omega^k = \cos\frac◆LB◆2k\pi◆RB◆◆LB◆5◆RB◆ + i\sin\frac◆LB◆2k\pi◆RB◆◆LB◆5◆RB◆:

k=04ωk=LBk=04cosLB2kπRB◆◆LB5RB◆◆RBreal+iLBk=04sinLB2kπRB◆◆LB5RB◆◆RBimaginary=0\sum_{k=0}^{4}\omega^k = \underbrace◆LB◆\sum_{k=0}^{4}\cos\frac◆LB◆2k\pi◆RB◆◆LB◆5◆RB◆◆RB◆_{\mathrm{real}} + i\underbrace◆LB◆\sum_{k=0}^{4}\sin\frac◆LB◆2k\pi◆RB◆◆LB◆5◆RB◆◆RB◆_{\mathrm{imaginary}} = 0

The imaginary part is zero by symmetry (sinθ=sin(2πθ)\sin\theta = -\sin(2\pi - \theta)). The real part gives:

1+cosLB2πRB◆◆LB5RB+cosLB4πRB◆◆LB5RB+cosLB6πRB◆◆LB5RB+cosLB8πRB◆◆LB5RB=01 + \cos\frac◆LB◆2\pi◆RB◆◆LB◆5◆RB◆ + \cos\frac◆LB◆4\pi◆RB◆◆LB◆5◆RB◆ + \cos\frac◆LB◆6\pi◆RB◆◆LB◆5◆RB◆ + \cos\frac◆LB◆8\pi◆RB◆◆LB◆5◆RB◆ = 0

Since cosLB6πRB◆◆LB5RB=cosLB4πRB◆◆LB5RB\cos\frac◆LB◆6\pi◆RB◆◆LB◆5◆RB◆ = \cos\frac◆LB◆4\pi◆RB◆◆LB◆5◆RB◆ and cosLB8πRB◆◆LB5RB=cosLB2πRB◆◆LB5RB\cos\frac◆LB◆8\pi◆RB◆◆LB◆5◆RB◆ = \cos\frac◆LB◆2\pi◆RB◆◆LB◆5◆RB◆:

1+2cosLB2πRB◆◆LB5RB+2cosLB4πRB◆◆LB5RB=0    cosLB2πRB◆◆LB5RB+cosLB4πRB◆◆LB5RB=121 + 2\cos\frac◆LB◆2\pi◆RB◆◆LB◆5◆RB◆ + 2\cos\frac◆LB◆4\pi◆RB◆◆LB◆5◆RB◆ = 0 \implies \cos\frac◆LB◆2\pi◆RB◆◆LB◆5◆RB◆ + \cos\frac◆LB◆4\pi◆RB◆◆LB◆5◆RB◆ = -\frac{1}{2} \quad \square


Problem 5. Sketch on separate Argand diagrams the loci given by (a) z1i=z3+i|z - 1 - i| = |z - 3 + i|, and (b) arg(z2)=LBπRB◆◆LB3RB\arg(z - 2) = \dfrac◆LB◆\pi◆RB◆◆LB◆3◆RB◆. Find the complex number(s) satisfying both conditions simultaneously.

Hint

Part (a) is a perpendicular bisector. Part (b) is a half-line. Find their intersection.

Answer

(a) z(1+i)=z(3i)|z - (1 + i)| = |z - (3 - i)| is the perpendicular bisector of the segment joining (1,1)(1, 1) and (3,1)(3, -1). The midpoint is (2,0)(2, 0) and the slope of the segment is 1131=1\dfrac{-1 - 1}{3 - 1} = -1, so the perpendicular bisector has slope 11 and equation y=x2y = x - 2.

(b) arg(z2)=LBπRB◆◆LB3RB\arg(z - 2) = \dfrac◆LB◆\pi◆RB◆◆LB◆3◆RB◆ is a half-line from (2,0)(2, 0) at angle LBπRB◆◆LB3RB\dfrac◆LB◆\pi◆RB◆◆LB◆3◆RB◆ to the positive real axis. Its equation is y=3(x2)y = \sqrt{3}(x - 2) for x>2x > 2.

Intersection: Setting x2=3(x2)x - 2 = \sqrt{3}(x - 2):

(x2)(13)=0(x - 2)(1 - \sqrt{3}) = 0, so x=2x = 2 (gives y=0y = 0, but the half-line requires x>2x > 2) or 1=31 = \sqrt{3}, which is false.

There is no intersection. The half-line from (2,0)(2, 0) at angle π/3\pi/3 has slope 3\sqrt{3}, while the perpendicular bisector has slope 1, and they only meet at the point (2,0)(2, 0) which is excluded from the half-line.

Answer: No complex number satisfies both conditions simultaneously.


Problem 6. Find the image of the circle z=2|z| = 2 under the transformation w=z+1z1w = \dfrac{z + 1}{z - 1}.

Hint

Express zz in terms of ww and substitute z=2|z| = 2.

Answer

w=z+1z1    wzw=z+1    z(w1)=w+1    z=w+1w1w = \dfrac{z + 1}{z - 1} \implies wz - w = z + 1 \implies z(w - 1) = w + 1 \implies z = \dfrac{w + 1}{w - 1}.

Since z=2|z| = 2:

w+1w1=2    w+1=2w1\left|\frac{w + 1}{w - 1}\right| = 2 \implies |w + 1| = 2|w - 1|

Let w=u+viw = u + vi:

(u+1)2+v2=2(u1)2+v2\sqrt{(u+1)^2 + v^2} = 2\sqrt{(u-1)^2 + v^2}

Squaring: (u+1)2+v2=4[(u1)2+v2](u+1)^2 + v^2 = 4[(u-1)^2 + v^2]

u2+2u+1+v2=4u28u+4+4v2u^2 + 2u + 1 + v^2 = 4u^2 - 8u + 4 + 4v^2

0=3u210u+3+3v20 = 3u^2 - 10u + 3 + 3v^2

3u210u+3v2+3=03u^2 - 10u + 3v^2 + 3 = 0

3 ⁣(u2103u)+3v2=33\!\left(u^2 - \frac{10}{3}u\right) + 3v^2 = -3

3 ⁣(u53)2253+3v2=33\!\left(u - \frac{5}{3}\right)^2 - \frac{25}{3} + 3v^2 = -3

3 ⁣(u53)2+3v2=1633\!\left(u - \frac{5}{3}\right)^2 + 3v^2 = \frac{16}{3}

(u53)2+v2=169\left(u - \frac{5}{3}\right)^2 + v^2 = \frac{16}{9}

This is a circle with centre (53,0)\left(\dfrac{5}{3}, 0\right) and radius 43\dfrac{4}{3} in the ww-plane.


Problem 7. Express LB(1+i)6RB◆◆LB(1i3)4RB\dfrac◆LB◆(1+i)^6◆RB◆◆LB◆(1-i\sqrt{3})^4◆RB◆ in the form a+bia + bi.

Hint

Write each term in exponential form and use the laws of indices.

Answer

1+i=2eiπ/41 + i = \sqrt{2}\,e^{i\pi/4} and 1i3=2eiπ/31 - i\sqrt{3} = 2\,e^{-i\pi/3}.

LB(1+i)6RB◆◆LB(1i3)4RB=LB(2eiπ/4)6RB◆◆LB(2eiπ/3)4RB=LB8e3πi/2RB◆◆LB16e4πi/3RB=12ei(3π/2+4π/3)\frac◆LB◆(1+i)^6◆RB◆◆LB◆(1-i\sqrt{3})^4◆RB◆ = \frac◆LB◆(\sqrt{2}\,e^{i\pi/4})^6◆RB◆◆LB◆(2\,e^{-i\pi/3})^4◆RB◆ = \frac◆LB◆8e^{3\pi i/2}◆RB◆◆LB◆16\,e^{-4\pi i/3}◆RB◆ = \frac{1}{2}\,e^{i(3\pi/2 + 4\pi/3)}

3π/2+4π/3=LB9π+8πRB◆◆LB6RB=LB17πRB◆◆LB6RB=2π+LB5πRB◆◆LB6RB3\pi/2 + 4\pi/3 = \frac◆LB◆9\pi + 8\pi◆RB◆◆LB◆6◆RB◆ = \frac◆LB◆17\pi◆RB◆◆LB◆6◆RB◆ = 2\pi + \frac◆LB◆5\pi◆RB◆◆LB◆6◆RB◆

So: 12e5πi/6=12 ⁣(cosLB5πRB◆◆LB6RB+isinLB5πRB◆◆LB6RB)=12 ⁣(LB3RB◆◆LB2RB+12i)=LB3RB◆◆LB4RB+14i\dfrac{1}{2}\,e^{5\pi i/6} = \dfrac{1}{2}\!\left(\cos\dfrac◆LB◆5\pi◆RB◆◆LB◆6◆RB◆ + i\sin\dfrac◆LB◆5\pi◆RB◆◆LB◆6◆RB◆\right) = \dfrac{1}{2}\!\left(-\dfrac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆ + \dfrac{1}{2}\,i\right) = -\dfrac◆LB◆\sqrt{3}◆RB◆◆LB◆4◆RB◆ + \dfrac{1}{4}\,i


Problem 8. The transformation TT from the zz-plane to the ww-plane is given by w=z2w = z^2. The region RR in the zz-plane is defined by 1z21 \leq |z| \leq 2 and 0arg(z)LBπRB◆◆LB4RB0 \leq \arg(z) \leq \dfrac◆LB◆\pi◆RB◆◆LB◆4◆RB◆. Find and describe the image of RR under TT.

Hint

Under w=z2w = z^2, the modulus squares and the argument doubles.

Answer

If z=reiθz = re^{i\theta}, then w=r2e2iθw = r^2 e^{2i\theta}.

  • Modulus: 1r2    1r241 \leq r \leq 2 \implies 1 \leq r^2 \leq 4, so 1w41 \leq |w| \leq 4.
  • Argument: 0θLBπRB◆◆LB4RB    02θLBπRB◆◆LB2RB0 \leq \theta \leq \dfrac◆LB◆\pi◆RB◆◆LB◆4◆RB◆ \implies 0 \leq 2\theta \leq \dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆, so 0arg(w)LBπRB◆◆LB2RB0 \leq \arg(w) \leq \dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆.

The image is the region in the first quadrant of the ww-plane between the circles w=1|w| = 1 and w=4|w| = 4, bounded by the rays arg(w)=0\arg(w) = 0 and arg(w)=LBπRB◆◆LB2RB\arg(w) = \dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆.


Problem 9. Solve the equation z3+z2+z+1=0z^3 + z^2 + z + 1 = 0 by recognising it as a geometric series, and hence show that cosLB2πRB◆◆LB4RB+cosLB4πRB◆◆LB4RB+cosLB6πRB◆◆LB4RB=1\cos\dfrac◆LB◆2\pi◆RB◆◆LB◆4◆RB◆ + \cos\dfrac◆LB◆4\pi◆RB◆◆LB◆4◆RB◆ + \cos\dfrac◆LB◆6\pi◆RB◆◆LB◆4◆RB◆ = -1.

Hint

Factor z3+z2+z+1=(z+1)(z2+1)z^3 + z^2 + z + 1 = (z+1)(z^2+1). The roots are the 4th roots of unity excluding z=1z = 1.

Answer

z3+z2+z+1=z41z1=0    z4=1z^3 + z^2 + z + 1 = \dfrac{z^4 - 1}{z - 1} = 0 \implies z^4 = 1 with z1z \neq 1.

The 4th roots of unity are 1,i,1,i1, i, -1, -i, so the solutions are z=i,1,iz = i, -1, -i.

Equivalently, the roots are ekπi/2e^{k\pi i/2} for k=1,2,3k = 1, 2, 3.

The sum of roots (by Vieta, coefficient of z2z^2 divided by leading coefficient) is 1-1:

i+(1)+(i)=1i + (-1) + (-i) = -1 \quad \checkmark

Now: eiπ/2+eiπ+e3iπ/2=i+(1)+(i)=1e^{i\pi/2} + e^{i\pi} + e^{3i\pi/2} = i + (-1) + (-i) = -1.

Separating real and imaginary parts: cosLBπRB◆◆LB2RB+cosπ+cosLB3πRB◆◆LB2RB=1\cos\dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆ + \cos\pi + \cos\dfrac◆LB◆3\pi◆RB◆◆LB◆2◆RB◆ = -1, i.e. 01+0=10 - 1 + 0 = -1. ✓

Alternatively, the claim as stated uses cosLB2πRB◆◆LB4RB+cosLB4πRB◆◆LB4RB+cosLB6πRB◆◆LB4RB=cosLBπRB◆◆LB2RB+cosπ+cosLB3πRB◆◆LB2RB=0+(1)+0=1\cos\dfrac◆LB◆2\pi◆RB◆◆LB◆4◆RB◆ + \cos\dfrac◆LB◆4\pi◆RB◆◆LB◆4◆RB◆ + \cos\dfrac◆LB◆6\pi◆RB◆◆LB◆4◆RB◆ = \cos\dfrac◆LB◆\pi◆RB◆◆LB◆2◆RB◆ + \cos\pi + \cos\dfrac◆LB◆3\pi◆RB◆◆LB◆2◆RB◆ = 0 + (-1) + 0 = -1. ✓ \square


Problem 10. (a) Show that LB1RB◆◆LBeiθ1RB=12i2cotLBθRB◆◆LB2RB\dfrac◆LB◆1◆RB◆◆LB◆e^{i\theta} - 1◆RB◆ = -\dfrac{1}{2} - \dfrac{i}{2}\cot\dfrac◆LB◆\theta◆RB◆◆LB◆2◆RB◆ for θ2πZ\theta \notin 2\pi\mathbb{Z}.

(b) Hence, or otherwise, find k=1n1LB1RB◆◆LB1ωkRB\displaystyle\sum_{k=1}^{n-1}\frac◆LB◆1◆RB◆◆LB◆1 - \omega^k◆RB◆ where ω=e2πi/n\omega = e^{2\pi i/n}.

Hint

For (a), multiply numerator and denominator by the conjugate eiθ1e^{-i\theta} - 1 and use half-angle identities. For (b), use the result from (a) with θ=2kπ/n\theta = 2k\pi/n.

Answer

(a) LB1RB◆◆LBeiθ1RB=LBeiθ1RB◆◆LB(eiθ1)(eiθ1)RB=LBeiθ1RB◆◆LB2(eiθ+eiθ)RB=LBeiθ1RB◆◆LB22cosθRB\dfrac◆LB◆1◆RB◆◆LB◆e^{i\theta} - 1◆RB◆ = \dfrac◆LB◆e^{-i\theta} - 1◆RB◆◆LB◆(e^{i\theta} - 1)(e^{-i\theta} - 1)◆RB◆ = \dfrac◆LB◆e^{-i\theta} - 1◆RB◆◆LB◆2 - (e^{i\theta} + e^{-i\theta})◆RB◆ = \dfrac◆LB◆e^{-i\theta} - 1◆RB◆◆LB◆2 - 2\cos\theta◆RB◆.

Numerator: eiθ1=cosθ1isinθ=2sin2LBθRB◆◆LB2RB2isinLBθRB◆◆LB2RBcosLBθRB◆◆LB2RB=2sinLBθRB◆◆LB2RB ⁣(sinLBθRB◆◆LB2RB+icosLBθRB◆◆LB2RB)e^{-i\theta} - 1 = \cos\theta - 1 - i\sin\theta = -2\sin^2\dfrac◆LB◆\theta◆RB◆◆LB◆2◆RB◆ - 2i\sin\dfrac◆LB◆\theta◆RB◆◆LB◆2◆RB◆\cos\dfrac◆LB◆\theta◆RB◆◆LB◆2◆RB◆ = -2\sin\dfrac◆LB◆\theta◆RB◆◆LB◆2◆RB◆\!\left(\sin\dfrac◆LB◆\theta◆RB◆◆LB◆2◆RB◆ + i\cos\dfrac◆LB◆\theta◆RB◆◆LB◆2◆RB◆\right).

Denominator: 22cosθ=4sin2LBθRB◆◆LB2RB2 - 2\cos\theta = 4\sin^2\dfrac◆LB◆\theta◆RB◆◆LB◆2◆RB◆.

LB1RB◆◆LBeiθ1RB=LB2sinLBθRB◆◆LB2RB ⁣(sinLBθRB◆◆LB2RB+icosLBθRB◆◆LB2RB)RB◆◆LB4sin2LBθRB◆◆LB2RB◆◆RB=LBsinLBθRB◆◆LB2RBicosLBθRB◆◆LB2RB◆◆RB◆◆LB2sinLBθRB◆◆LB2RB◆◆RB=12i2cotLBθRB◆◆LB2RB\frac◆LB◆1◆RB◆◆LB◆e^{i\theta} - 1◆RB◆ = \frac◆LB◆-2\sin\frac◆LB◆\theta◆RB◆◆LB◆2◆RB◆\!\left(\sin\frac◆LB◆\theta◆RB◆◆LB◆2◆RB◆ + i\cos\frac◆LB◆\theta◆RB◆◆LB◆2◆RB◆\right)◆RB◆◆LB◆4\sin^2\frac◆LB◆\theta◆RB◆◆LB◆2◆RB◆◆RB◆ = \frac◆LB◆-\sin\frac◆LB◆\theta◆RB◆◆LB◆2◆RB◆ - i\cos\frac◆LB◆\theta◆RB◆◆LB◆2◆RB◆◆RB◆◆LB◆2\sin\frac◆LB◆\theta◆RB◆◆LB◆2◆RB◆◆RB◆ = -\frac{1}{2} - \frac{i}{2}\cot\frac◆LB◆\theta◆RB◆◆LB◆2◆RB◆ \quad \square

(b) Using (a): LB1RB◆◆LB1ωkRB=LB1RB◆◆LBe2kπi/n1RB=12+i2cotLBkπRB◆◆LBnRB\dfrac◆LB◆1◆RB◆◆LB◆1 - \omega^k◆RB◆ = -\dfrac◆LB◆1◆RB◆◆LB◆e^{2k\pi i/n} - 1◆RB◆ = \dfrac{1}{2} + \dfrac{i}{2}\cot\dfrac◆LB◆k\pi◆RB◆◆LB◆n◆RB◆.

k=1n1LB1RB◆◆LB1ωkRB=k=1n1 ⁣(12+i2cotLBkπRB◆◆LBnRB)=n12+i2k=1n1cotLBkπRB◆◆LBnRB\sum_{k=1}^{n-1}\frac◆LB◆1◆RB◆◆LB◆1 - \omega^k◆RB◆ = \sum_{k=1}^{n-1}\!\left(\frac{1}{2} + \frac{i}{2}\cot\frac◆LB◆k\pi◆RB◆◆LB◆n◆RB◆\right) = \frac{n-1}{2} + \frac{i}{2}\sum_{k=1}^{n-1}\cot\frac◆LB◆k\pi◆RB◆◆LB◆n◆RB◆

The cotangent sum is zero by symmetry: cotLBkπRB◆◆LBnRB=cotLB(nk)πRB◆◆LBnRB\cot\dfrac◆LB◆k\pi◆RB◆◆LB◆n◆RB◆ = -\cot\dfrac◆LB◆(n-k)\pi◆RB◆◆LB◆n◆RB◆, so terms cancel in pairs.

Therefore: k=1n1LB1RB◆◆LB1ωkRB=n12\displaystyle\sum_{k=1}^{n-1}\frac◆LB◆1◆RB◆◆LB◆1 - \omega^k◆RB◆ = \frac{n - 1}{2}.

:::


8. Advanced Worked Examples

Example 8.1: De Moivre's theorem for cos5θ\cos 5\theta

Problem. Express cos5θ\cos 5\theta in terms of cosθ\cos\theta.

Solution. By de Moivre: (cosθ+isinθ)5=cos5θ+isin5θ(\cos\theta + i\sin\theta)^5 = \cos 5\theta + i\sin 5\theta.

Expanding by the binomial theorem:

(cosθ+isinθ)5=cos5θ+5icos4θsinθ10cos3θsin2θ10icos2θsin3θ+5cosθsin4θ+isin5θ(\cos\theta + i\sin\theta)^5 = \cos^5\theta + 5i\cos^4\theta\sin\theta - 10\cos^3\theta\sin^2\theta - 10i\cos^2\theta\sin^3\theta + 5\cos\theta\sin^4\theta + i\sin^5\theta

Equating real parts:

cos5θ=16cos5θ20cos3θ+5cosθ\boxed{\cos 5\theta = 16\cos^5\theta - 20\cos^3\theta + 5\cos\theta}

Example 8.2: Solving z6=64z^6 = -64

Problem. Solve z6=64z^6 = -64, giving answers in exponential form.

Solution. 64=64eiπ-64 = 64e^{i\pi}. The 6th roots are:

zk=641/6exp ⁣(LBi(π+2kπ)RB◆◆LB6RB)=2exp ⁣(LBi(2k+1)πRB◆◆LB6RB)z_k = 64^{1/6} \exp\!\left(\frac◆LB◆i(\pi + 2k\pi)◆RB◆◆LB◆6◆RB◆\right) = 2\exp\!\left(\frac◆LB◆i(2k+1)\pi◆RB◆◆LB◆6◆RB◆\right)

for k=0,1,2,3,4,5k = 0, 1, 2, 3, 4, 5.

z0=2eiπ/6z_0 = 2e^{i\pi/6}, z1=2eiπ/2z_1 = 2e^{i\pi/2}, z2=2e5iπ/6z_2 = 2e^{5i\pi/6}, z3=2e7iπ/6z_3 = 2e^{7i\pi/6}, z4=2e3iπ/2z_4 = 2e^{3i\pi/2}, z5=2e11iπ/6z_5 = 2e^{11i\pi/6}.

These lie on a circle of radius 2, at angles 30°,90°,150°,210°,270°,330°30°, 90°, 150°, 210°, 270°, 330°.

Example 8.3: Loci — perpendicular bisector

Problem. Find the Cartesian equation of the locus z34i=z+12i|z - 3 - 4i| = |z + 1 - 2i|.

Solution. Let z=x+iyz = x + iy.

z(3+4i)=z(1+2i)|z - (3+4i)| = |z - (-1+2i)|

(x3)2+(y4)2=(x+1)2+(y2)2\sqrt{(x-3)^2 + (y-4)^2} = \sqrt{(x+1)^2 + (y-2)^2}

Squaring: (x3)2+(y4)2=(x+1)2+(y2)2(x-3)^2 + (y-4)^2 = (x+1)^2 + (y-2)^2.

x26x+9+y28y+16=x2+2x+1+y24y+4x^2 - 6x + 9 + y^2 - 8y + 16 = x^2 + 2x + 1 + y^2 - 4y + 4

8x4y+20=0    2x+y=5-8x - 4y + 20 = 0 \implies \boxed{2x + y = 5}

This is the perpendicular bisector of the segment joining 3+4i3+4i and 1+2i-1+2i.

Example 8.4: Region defined by an inequality

Problem. Shade on an Argand diagram the region defined by z2i3|z - 2i| \leq 3 and 0arg(z)LBπRB◆◆LB4RB0 \leq \arg(z) \leq \dfrac◆LB◆\pi◆RB◆◆LB◆4◆RB◆.

Solution. z2i3|z - 2i| \leq 3 is the closed disc of radius 3 centred at 2i2i (i.e., (0,2)(0, 2)).

0arg(z)LBπRB◆◆LB4RB0 \leq \arg(z) \leq \dfrac◆LB◆\pi◆RB◆◆LB◆4◆RB◆ is the sector between the positive real axis and the line y=xy = x (for x0x \geq 0).

The required region is the intersection: a segment of the disc in the first quadrant between angles 00 and π/4\pi/4.

Example 8.5: Complex transformation — rotation and enlargement

Problem. The transformation TT maps the complex plane by w=(1+i)z+2iw = (1+i)z + 2i. Describe TT fully and find the image of the line Re(z)=1\mathrm{Re}(z) = 1.

Solution. w=(1+i)z+2i=2eiπ/4z+2iw = (1+i)z + 2i = \sqrt{2}\,e^{i\pi/4}\,z + 2i.

TT is an enlargement by scale factor 2\sqrt{2}, rotation by 45°45° anticlockwise about the origin, followed by a translation by 2i2i.

For Re(z)=1\mathrm{Re}(z) = 1: z=1+itz = 1 + it. w=(1+i)(1+it)+2i=1+it+it+2i=(1t)+i(3+t)w = (1+i)(1+it) + 2i = 1 + it + i - t + 2i = (1-t) + i(3+t).

Re(w)=1t\mathrm{Re}(w) = 1-t, Im(w)=3+t\mathrm{Im}(w) = 3+t. Eliminating tt: Im(w)=3+(1Re(w))=4Re(w)\mathrm{Im}(w) = 3 + (1 - \mathrm{Re}(w)) = 4 - \mathrm{Re}(w).

The image is the line u+v=4\boxed{u + v = 4} (where w=u+ivw = u + iv).

Example 8.6: Sum of roots of unity

Problem. Show that k=0n1ωk=0\displaystyle\sum_{k=0}^{n-1} \omega^k = 0 where ω=e2πi/n\omega = e^{2\pi i/n}.

Solution. This is a geometric series with ratio ω1\omega \neq 1:

k=0n1ωk=LB1ωnRB◆◆LB1ωRB=LB11RB◆◆LB1ωRB=0\sum_{k=0}^{n-1} \omega^k = \frac◆LB◆1 - \omega^n◆RB◆◆LB◆1 - \omega◆RB◆ = \frac◆LB◆1 - 1◆RB◆◆LB◆1 - \omega◆RB◆ = 0

\blacksquare

Example 8.7: Product of roots of unity

Problem. Show that the product of all nn-th roots of unity is (1)n+1(-1)^{n+1}.

Solution. The nn-th roots of unity are the roots of zn1=0z^n - 1 = 0. By Vieta's formulae, the product of all roots is (1)n×(1)=(1)n+1(-1)^n \times (-1) = (-1)^{n+1}.

Alternatively: the roots are 1,ω,ω2,,ωn11, \omega, \omega^2, \ldots, \omega^{n-1}, so the product is ω0+1+2++(n1)=ωn(n1)/2=eπi(n1)=(1)n1=(1)n+1\omega^{0+1+2+\cdots+(n-1)} = \omega^{n(n-1)/2} = e^{\pi i(n-1)} = (-1)^{n-1} = (-1)^{n+1}. \blacksquare


9. Common Pitfalls

| Pitfall | Correct Approach | | ----------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------ | -------- | --- | --- | --- | --- | --- | --- | --------------------------------------------------- | --- | ---- | --- | --- | --- | --- | | Forgetting that arg(z)\arg(z) is measured from the positive real axis | arg(z)\arg(z) is the angle anticlockwise from the positive xx-axis, range (π,π](-\pi, \pi] or [0,2π)[0, 2\pi) | | Confusing zw | z-w | with zw| z | - | w | | zw | z-w | is the distance between zz and ww; in general zwzw| z-w | \neq | z | - | w | | | Missing roots when solving zn=wz^n = w | There are always exactly nn distinct roots; check your kk values cover 00 to n1n-1 | | Incorrectly applying de Moivre to non-integer powers | De Moivre's theorem (cosθ+isinθ)n=cosnθ+isinnθ(\cos\theta + i\sin\theta)^n = \cos n\theta + i\sin n\theta holds for integer nn only |


10. Additional Exam-Style Questions

Question 8

Solve z4=8(1+i3)z^4 = 8(1 + i\sqrt{3}), giving roots in the form r(cosθ+isinθ)r(\cos\theta + i\sin\theta).

Solution

8(1+i3)=16eiπ/38(1+i\sqrt{3}) = 16e^{i\pi/3}.

zk=2exp ⁣(LBi(π/3+2kπ)RB◆◆LB4RB)z_k = 2\exp\!\left(\dfrac◆LB◆i(\pi/3 + 2k\pi)◆RB◆◆LB◆4◆RB◆\right) for k=0,1,2,3k = 0, 1, 2, 3.

z0=2(cos15°+isin15°)z_0 = 2(\cos 15° + i\sin 15°), z1=2(cos105°+isin105°)z_1 = 2(\cos 105° + i\sin 105°), z2=2(cos195°+isin195°)z_2 = 2(\cos 195° + i\sin 195°), z3=2(cos285°+isin285°)z_3 = 2(\cos 285° + i\sin 285°).

Question 9

Prove that cos4θ=38+12cos2θ+18cos4θ\cos^4\theta = \dfrac{3}{8} + \dfrac{1}{2}\cos 2\theta + \dfrac{1}{8}\cos 4\theta.

Solution

cos2θ=LB1+cos2θRB◆◆LB2RB\cos^2\theta = \dfrac◆LB◆1+\cos 2\theta◆RB◆◆LB◆2◆RB◆.

cos4θ=(LB1+cos2θRB◆◆LB2RB) ⁣2=LB1+2cos2θ+cos22θRB◆◆LB4RB\cos^4\theta = \left(\dfrac◆LB◆1+\cos 2\theta◆RB◆◆LB◆2◆RB◆\right)^{\!2} = \dfrac◆LB◆1 + 2\cos 2\theta + \cos^2 2\theta◆RB◆◆LB◆4◆RB◆.

cos22θ=LB1+cos4θRB◆◆LB2RB\cos^2 2\theta = \dfrac◆LB◆1+\cos 4\theta◆RB◆◆LB◆2◆RB◆.

cos4θ=14+LBcos2θRB◆◆LB2RB+LB1+cos4θRB◆◆LB8RB=38+LBcos2θRB◆◆LB2RB+LBcos4θRB◆◆LB8RB\cos^4\theta = \dfrac{1}{4} + \dfrac◆LB◆\cos 2\theta◆RB◆◆LB◆2◆RB◆ + \dfrac◆LB◆1+\cos 4\theta◆RB◆◆LB◆8◆RB◆ = \dfrac{3}{8} + \dfrac◆LB◆\cos 2\theta◆RB◆◆LB◆2◆RB◆ + \dfrac◆LB◆\cos 4\theta◆RB◆◆LB◆8◆RB◆. \blacksquare

Question 10

The complex number zz satisfies z1=z+1|z-1| = |z+1| and z3i=3|z-3i| = 3. Find zz.

Solution

z1=z+1|z-1| = |z+1|: perpendicular bisector of 11 and 1-1, giving Re(z)=0\mathrm{Re}(z) = 0. So z=iyz = iy.

z3i=3    iy3i=3    y3=3    y3=±3|z-3i| = 3 \implies |iy - 3i| = 3 \implies |y-3| = 3 \implies y - 3 = \pm 3.

y=6y = 6 or y=0y = 0. So z=6iz = 6i or z=0z = 0.


11. Connections to Other Topics

11.1 Complex numbers and matrices

Complex eigenvalues of 2×2 matrices correspond to rotation-scaling transformations. See Matrices.

11.2 Complex numbers and hyperbolic functions

eix=cosx+isinxe^{ix} = \cos x + i\sin x connects exponential, trigonometric, and hyperbolic functions. See Hyperbolic Functions.

11.3 Complex numbers and polar coordinates

Argand diagrams and polar form (r,θ)(r, \theta) connect to polar coordinates. See Polar Coordinates.


12. Key Results Summary

| Result | Formula | | ----------------------------- | --------------------------------------------------------- | --- | -------------------------------- | --- | --- | | Modulus | z=a2+b2 | z | = \sqrt{a^2+b^2} for z=a+biz = a+bi | | Argument | arg(z)=arctan(b/a)\arg(z) = \arctan(b/a) (adjusting for quadrant) | | Euler's formula | eiθ=cosθ+isinθe^{i\theta} = \cos\theta+i\sin\theta | | De Moivre | (cosθ+isinθ)n=cosnθ+isinnθ(\cos\theta+i\sin\theta)^n = \cos n\theta+i\sin n\theta | | nn-th roots of unity | zk=e2πik/nz_k = e^{2\pi ik/n}, k=0,,n1k = 0, \ldots, n-1 | | Locus: circle | za=r | z-a | =r | | Locus: perpendicular bisector | za=zb | z-a | = | z-b | | | Locus: half-line | arg(za)=θ\arg(z-a) = \theta |


13. Further Exam-Style Questions

Question 11

Solve z3=8iz^3 = -8i, giving roots in Cartesian form.

Solution

8i=8eiπ/2-8i = 8e^{-i\pi/2}. Roots: zk=2exp ⁣(LBiπ/2+2kπiRB◆◆LB3RB)z_k = 2\exp\!\left(\dfrac◆LB◆-i\pi/2 + 2k\pi i◆RB◆◆LB◆3◆RB◆\right) for k=0,1,2k=0,1,2.

z0=2eiπ/6=2 ⁣(LB3RB◆◆LB2RBi2)=3iz_0 = 2e^{-i\pi/6} = 2\!\left(\dfrac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆ - \dfrac{i}{2}\right) = \sqrt{3}-i.

z1=2eiπ/2=2iz_1 = 2e^{i\pi/2} = 2i.

z2=2e7iπ/6=2 ⁣(LB3RB◆◆LB2RBi2)=3iz_2 = 2e^{7i\pi/6} = 2\!\left(-\dfrac◆LB◆\sqrt{3}◆RB◆◆LB◆2◆RB◆ - \dfrac{i}{2}\right) = -\sqrt{3}-i.

z=3i,  2i,  3i\boxed{z = \sqrt{3}-i,\; 2i,\; -\sqrt{3}-i}

Question 12

Prove that z1z2=z1z2|z_1 z_2| = |z_1||z_2| for any complex numbers z1,z2z_1, z_2.

Solution

Let z1=r1eiθ1z_1 = r_1 e^{i\theta_1} and z2=r2eiθ2z_2 = r_2 e^{i\theta_2}.

z1z2=r1r2ei(θ1+θ2)z_1 z_2 = r_1 r_2 e^{i(\theta_1+\theta_2)}.

z1z2=r1r2=z1z2|z_1 z_2| = r_1 r_2 = |z_1||z_2|. \blacksquare


14. Advanced Topics

14.1 The exponential form of complex numbers

Any non-zero complex number can be written as z=reiθz = re^{i\theta} where r=zr = |z| and θ=arg(z)\theta = \arg(z).

This form makes multiplication and division particularly simple:

  • z1z2=r1r2ei(θ1+θ2)z_1 z_2 = r_1 r_2 e^{i(\theta_1+\theta_2)}
  • z1/z2=(r1/r2)ei(θ1θ2)z_1/z_2 = (r_1/r_2) e^{i(\theta_1-\theta_2)}

14.2 Euler's identity

Setting θ=π\theta = \pi in Euler's formula: eiπ+1=0e^{i\pi} + 1 = 0.

This connects five fundamental constants: ee, ii, π\pi, 11, and 00.

14.3 Complex conjugate and roots of polynomials

If P(z)P(z) is a polynomial with real coefficients and z=a+biz = a + bi is a root, then zˉ=abi\bar{z} = a - bi is also a root. This is because P(z)=P(zˉ)\overline{P(z)} = P(\bar{z}) for real-coefficient polynomials.

14.4 Regions in the Argand diagram

InequalityRegion
za<r\|z-a\| < rInterior of circle (open disc)
zar\|z-a\| \leq rClosed disc
α<arg(za)<β\alpha < \arg(z-a) < \betaSector (angular region)
Re(z)>k\mathrm{Re}(z) > kHalf-plane to the right of x=kx = k
Im(z)>k\mathrm{Im}(z) > kHalf-plane above y=ky = k

14.5 The exponential form of sin\sin and cos\cos

From eiθ=cosθ+isinθe^{i\theta} = \cos\theta + i\sin\theta and eiθ=cosθisinθe^{-i\theta} = \cos\theta - i\sin\theta:

cosθ=LBeiθ+eiθRB◆◆LB2RB,sinθ=LBeiθeiθRB◆◆LB2iRB\cos\theta = \frac◆LB◆e^{i\theta}+e^{-i\theta}◆RB◆◆LB◆2◆RB◆, \qquad \sin\theta = \frac◆LB◆e^{i\theta}-e^{-i\theta}◆RB◆◆LB◆2i◆RB◆

These are essential for deriving trigonometric identities and solving certain integrals.


15. Further Exam-Style Questions

Question 13

Express LB1+eiθRB◆◆LB1eiθRB\dfrac◆LB◆1+e^{i\theta}◆RB◆◆LB◆1-e^{i\theta}◆RB◆ in the form a+bia+bi.

Solution

LB1+eiθRB◆◆LB1eiθRB=LBeiθ/2(eiθ/2+eiθ/2)RB◆◆LBeiθ/2(eiθ/2eiθ/2)RB=LB2cos(θ/2)RB◆◆LB2isin(θ/2)RB=LBicos(θ/2)RB◆◆LBsin(θ/2)RB=icot(θ/2)\dfrac◆LB◆1+e^{i\theta}◆RB◆◆LB◆1-e^{i\theta}◆RB◆ = \dfrac◆LB◆e^{i\theta/2}(e^{-i\theta/2}+e^{i\theta/2})◆RB◆◆LB◆e^{i\theta/2}(e^{-i\theta/2}-e^{i\theta/2})◆RB◆ = \dfrac◆LB◆2\cos(\theta/2)◆RB◆◆LB◆-2i\sin(\theta/2)◆RB◆ = \dfrac◆LB◆i\cos(\theta/2)◆RB◆◆LB◆\sin(\theta/2)◆RB◆ = \boxed{i\cot(\theta/2)}

Question 14

Prove that the sum of the nn-th roots of unity is zero.

Solution

The nn-th roots of unity are 1,ω,ω2,,ωn11, \omega, \omega^2, \ldots, \omega^{n-1} where ω=e2πi/n\omega = e^{2\pi i/n}.

This is a geometric series: k=0n1ωk=LB1ωnRB◆◆LB1ωRB=LB11RB◆◆LB1ωRB=0\displaystyle\sum_{k=0}^{n-1} \omega^k = \frac◆LB◆1-\omega^n◆RB◆◆LB◆1-\omega◆RB◆ = \frac◆LB◆1-1◆RB◆◆LB◆1-\omega◆RB◆ = 0. \blacksquare

Question 15

Find all complex numbers zz such that zzˉ+z+zˉ=3z\bar{z} + z + \bar{z} = 3.

Solution

Let z=x+iyz = x+iy. Then zˉ=xiy\bar{z} = x-iy and zzˉ=x2+y2z\bar{z} = x^2+y^2.

x2+y2+2x=3    (x+1)2+y2=4x^2+y^2 + 2x = 3 \implies (x+1)^2 + y^2 = 4.

This is a circle with centre (1,0)(-1, 0) and radius 22. All complex numbers on this circle satisfy the equation.


16. Further Advanced Topics

16.1 The fundamental theorem of algebra

Every polynomial of degree n1n \geq 1 with complex coefficients has exactly nn roots (counting multiplicity) in the complex numbers.

This means: every polynomial can be factored as P(z)=a(zz1)(zz2)(zzn)P(z) = a(z - z_1)(z - z_2)\cdots(z - z_n).

16.2 Complex numbers as rotations and dilations

Multiplication by reiθre^{i\theta} represents:

  • Dilation by scale factor rr
  • Rotation by angle θ\theta anticlockwise

This provides a geometric interpretation of all complex arithmetic.

16.3 nn-th roots of any complex number

To solve zn=w=reiϕz^n = w = re^{i\phi}:

zk=r1/nexp ⁣(LBi(ϕ+2kπ)RB◆◆LBnRB)for k=0,1,,n1z_k = r^{1/n} \exp\!\left(\frac◆LB◆i(\phi + 2k\pi)◆RB◆◆LB◆n◆RB◆\right) \quad \text{for } k = 0, 1, \ldots, n-1

The roots lie on a circle of radius r1/nr^{1/n}, equally spaced.

16.4 The complex exponential function

ez=ex+iy=ex(cosy+isiny)e^z = e^{x+iy} = e^x(\cos y + i\sin y) for z=x+iyz = x + iy.

Key properties:

  • ez=ex|e^z| = e^x
  • arg(ez)=y\arg(e^z) = y (mod 2π2\pi)
  • ez1+z2=ez1ez2e^{z_1+z_2} = e^{z_1}e^{z_2}
  • ez0e^z \neq 0 for all zz
  • ez=1    z=2kπie^z = 1 \iff z = 2k\pi i for some integer kk

17. Further Exam-Style Questions

Question 16

Find all complex numbers zz such that z4=16z^4 = -16.

Solution

16=16eiπ-16 = 16e^{i\pi}. Roots: zk=2exp ⁣(LBi(π+2kπ)RB◆◆LB4RB)z_k = 2\exp\!\left(\dfrac◆LB◆i(\pi+2k\pi)◆RB◆◆LB◆4◆RB◆\right) for k=0,1,2,3k=0,1,2,3.

z0=2eiπ/4=2+i2z_0 = 2e^{i\pi/4} = \sqrt{2}+i\sqrt{2}, z1=2ei3π/4=2+i2z_1 = 2e^{i3\pi/4} = -\sqrt{2}+i\sqrt{2},

z2=2ei5π/4=2i2z_2 = 2e^{i5\pi/4} = -\sqrt{2}-i\sqrt{2}, z3=2ei7π/4=2i2z_3 = 2e^{i7\pi/4} = \sqrt{2}-i\sqrt{2}.

Question 17

Prove that for any complex number zz: z1z2=zˉ1zˉ2\overline{z_1 z_2} = \bar{z}_1 \cdot \bar{z}_2.

Solution

Let z1=a+biz_1 = a+bi and z2=c+diz_2 = c+di.

z1z2=(acbd)+(ad+bc)iz_1 z_2 = (ac-bd) + (ad+bc)i.

z1z2=(acbd)(ad+bc)i\overline{z_1 z_2} = (ac-bd) - (ad+bc)i.

zˉ1zˉ2=(abi)(cdi)=acadibci+bdi2=(acbd)(ad+bc)i\bar{z}_1 \cdot \bar{z}_2 = (a-bi)(c-di) = ac - adi - bci + bdi^2 = (ac-bd) - (ad+bc)i.

Equal. \blacksquare

Question 18

The complex numbers zz and ww satisfy z=3|z| = 3, w=4|w| = 4, and z+w=5|z+w| = 5. Find zw|z-w|.

Solution

z+w2=z2+w2+2Re(zwˉ)=9+16+2Re(zwˉ)=25|z+w|^2 = |z|^2 + |w|^2 + 2\mathrm{Re}(z\bar{w}) = 9+16+2\mathrm{Re}(z\bar{w}) = 25.

Re(zwˉ)=0\mathrm{Re}(z\bar{w}) = 0.

zw2=z2+w22Re(zwˉ)=9+160=25|z-w|^2 = |z|^2 + |w|^2 - 2\mathrm{Re}(z\bar{w}) = 9+16-0 = 25.

zw=5\boxed{|z-w| = 5}


18. Further Advanced Topics

18.1 Riemann surfaces and multi-valued functions

The complex logarithm, nn-th root, and inverse trigonometric functions are all multi-valued. Branch cuts are used to define single-valued branches (principal values).

18.2 The complex plane and stereographic projection

The extended complex plane C^=C{}\hat{\mathbb{C}} = \mathbb{C} \cup \{\infty\} is topologically a sphere (the Riemann sphere). Stereographic projection maps each point on the sphere (except the north pole) to a unique point in the complex plane.

18.3 Complex analysis connections

While complex analysis (Cauchy's theorem, residue calculus) is beyond A-Level, the fundamental concepts appear:

  • Cauchy's integral formula: f(a)=LB1RB◆◆LB2πiRBCf(z)zadzf(a) = \dfrac◆LB◆1◆RB◆◆LB◆2\pi i◆RB◆\displaystyle\oint_C \frac{f(z)}{z-a}\,dz
  • Residue theorem: Cf(z)dz=2πiRes(f,ak)\displaystyle\oint_C f(z)\,dz = 2\pi i \sum \text{Res}(f, a_k)

These are mentioned for context and further study.

18.4 De Moivre's theorem — number theory applications

De Moivre's theorem connects complex numbers to number theory:

  • Fermat's theorem on sums of two squares: p1(mod4)    p=a2+b2p \equiv 1 \pmod 4 \implies p = a^2 + b^2
  • Wilson's theorem: (p1)!1(modp)(p-1)! \equiv -1 \pmod p for prime pp

19. Further Exam-Style Questions

Question 19

Express cos5θ+isin5θ\cos 5\theta + i\sin 5\theta in terms of cosθ\cos\theta and sinθ\sin\theta using the binomial theorem.

Solution

(cosθ+isinθ)5=cos5θ+5icos4θsinθ10cos3θsin2θ10icos2θsin3θ+5cosθsin4θ+isin5θ(\cos\theta + i\sin\theta)^5 = \cos^5\theta + 5i\cos^4\theta\sin\theta - 10\cos^3\theta\sin^2\theta - 10i\cos^2\theta\sin^3\theta + 5\cos\theta\sin^4\theta + i\sin^5\theta.

Real part: cos5θ=cos5θ10cos3θsin2θ+5cosθsin4θ\cos 5\theta = \cos^5\theta - 10\cos^3\theta\sin^2\theta + 5\cos\theta\sin^4\theta.

Using sin2θ=1cos2θ\sin^2\theta = 1-\cos^2\theta:

=cos5θ10cos3θ(1cos2θ)+5cosθ(1cos2θ)2= \cos^5\theta - 10\cos^3\theta(1-\cos^2\theta) + 5\cos\theta(1-\cos^2\theta)^2

=cos5θ10cos3θ+10cos5θ+5cosθ10cos3θ+5cos5θ= \cos^5\theta - 10\cos^3\theta + 10\cos^5\theta + 5\cos\theta - 10\cos^3\theta + 5\cos^5\theta

=16cos5θ20cos3θ+5cosθ= \boxed{16\cos^5\theta - 20\cos^3\theta + 5\cos\theta}.

Question 20

Prove that z+w2+zw2=2(z2+w2)|z+w|^2 + |z-w|^2 = 2(|z|^2+|w|^2) for all complex numbers z,wz, w (the parallelogram law).

Solution

z+w2+zw2=(z+w)(zˉ+wˉ)+(zw)(zˉwˉ)|z+w|^2 + |z-w|^2 = (z+w)(\bar{z}+\bar{w}) + (z-w)(\bar{z}-\bar{w})

=z2+zwˉ+wzˉ+w2+z2zwˉwzˉ+w2=2z2+2w2= |z|^2+z\bar{w}+w\bar{z}+|w|^2 + |z|^2-z\bar{w}-w\bar{z}+|w|^2 = 2|z|^2 + 2|w|^2.

This is the parallelogram law: the sum of squares of the diagonals equals the sum of squares of all four sides. \blacksquare