Complex numbers extend the real number system by introducing a solution to the equation
x2+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
Board
Paper
Notes
AQA
Paper 1
Full coverage including transformations w=f(z)
Edexcel
FP1/FP2
De Moivre, roots of unity, loci in FP2
OCR (A)
Paper 1
De Moivre, exponential form, roots of unity
CIE
P1
Loci 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+bi where a,b∈R
And i is defined by the property i2=−1. The set of all complex numbers is denoted
C.
We call a the real part (written Re(z)) and b the imaginary part (written
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 as a plane, where the
Horizontal axis represents Re(z) and the vertical axis represents
Im(z). The complex number z=a+bi is plotted as the point (a,b).
1.2 Modulus and Argument
Definition. For z=a+bi=0:
The modulus∣z∣ is defined as ∣z∣=a2+b2.
The argumentarg(z) is the angle θ (measured anticlockwise from the positive real
axis) such that tanθ=abWith −π<θ≤π (the principal
argument).
For negative integers, note that
L◆B◆1◆RB◆◆LB◆cosθ+isinθ◆RB◆=cosθ−isinθ=cos(−θ)+isin(−θ)
So the result follows. □
Intuition. De Moivre’s theorem says that raising a complex number on the unit circle to the
n-th power multiplies its angle by n. This is because multiplication of complex numbers Adds
their arguments: r1eiθ1⋅r2eiθ2=r1r2ei(θ1+θ2).
2.1 Applications: Trigonometric Identities
De Moivre’s theorem provides a systematic way to derive multiple-angle identities.
Example. Express cos3θ and sin3θ in terms of cosθ and sinθ.
The n-th roots of unity lie on the unit circle ∣z∣=1 in the Argand diagram, equally spaced at
Angles of L◆B◆2π◆RB◆◆LB◆n◆RB◆ radians apart. They form the vertices of a regular n-gon
Inscribed in the unit circle, with one vertex at z=1.
3.2 Sum and Product of Roots
Since the roots satisfy zn−1=0The sum of all n-th roots of unity is zero:
∑k=0n−1zk=0
This follows from the coefficient of zn−1 in zn−1=0 being zero (by Vieta’s formulas).
Equivalently, the roots form a geometric series with ratio ω=e2πi/n and first term 1,
Giving:
This celebrated identity connects five fundamental constants: e$$i$$\pi$$1And 0.
Intuition. Euler’s identity says that starting at the point 1 on the real axis and rotating by
π radians (half a turn) on the unit circle lands you at −1. The exponential eiθ
Describes a point moving around the unit circle at a rate determined by θ.
:::info info with the cosθ+isinθ form. All boards require De Moivre’s theorem.
:::
5. Loci in the Argand Diagram
5.1 Circles: ∣z−a∣=r
Definition. The locus ∣z−a∣=rWhere a∈C and r∈R+Is a Circle
with centre a and radius r in the Argand diagram.
∣z−a∣=r⟺◆LB◆(x−α)2+(y−β)2◆RB◆=r⟺(x−α)2+(y−β)2=r2
Where a=α+βi and z=x+yi.
The region ∣z−a∣<r is the interior of the circle, and ∣z−a∣>r is the exterior.
5.2 Perpendicular Bisectors: ∣z−a∣=∣z−b∣
This locus represents all points equidistant from a and bWhich is the perpendicular bisector Of
the line segment joining a and b.
5.3 Half-Lines: arg(z−a)=α
Definition. The locus arg(z−a)=α is a half-line (ray) starting from the point a
(not including a itself) making an angle α with the positive real direction.
The region α1<arg(z−a)<α2 is an angular sector (wedge) with vertex at a.
5.4 Combined Loci and Regions
Exam questions often require describing a region defined by combining loci, such as:
∣z−3∣≤2 and arg(z)≥L◆B◆π◆RB◆◆LB◆4◆RB◆: the part of a disc in a sector.
:::caution A half-line arg(z−a)=α does not include the point a. When shading
Regions, be careful about whether boundaries are included (solid line) or excluded (dashed line).
:::
:::tip 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 z satisfies ∣z−2i∣≤3 and
0≤arg(z)≤L◆B◆π◆RB◆◆LB◆2◆RB◆. Find the greatest possible value of ∣z∣ and the
Least possible value of ∣z∣.
The first condition: ∣z−2i∣≤3 is the closed disc of radius 3 centred at 2iI.e. At
(0,2).
The second condition: 0≤arg(z)≤L◆B◆π◆RB◆◆LB◆2◆RB◆ restricts z to the first
Quadrant (including axes).
The disc centre (0,2) with radius 3 extends from y=−1 to y=5 and from x=−3 to
x=3.
Greatest ∣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 y-axis at (0,5)Giving ∣z∣=5.
Least ∣z∣: We need the closest point in the region to the origin. The disc boundary is
(x2+(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)Which is the y-axis. The point (0,−1) is outside the first
Quadrant. Within the first quadrant, the closest point is where the circle meets the x-axis:
Setting y=0, x2+4=9⟹x=5. So ∣z∣=5.
6. Complex Transformations
6.1 The Mapping w=f(z)
A complex transformation is a function w=f(z) that maps points in the z-plane (Argand diagram
For z) to points in the w-plane (Argand diagram for w).
6.2 Linear Transformations: w=az+b
For w=az+b where a,b∈C and a=0:
Writing a=λeiα and b=μeiβ:
∣a∣=λ produces an enlargement (scale factor λ) about the origin.
arg(a)=α produces a rotation through angle α about the origin.
b produces a translation by the vector representing b.
The composition is: enlarge by ∣a∣Rotate by arg(a)Then translate by b.
6.3 Inversion: w=z1
The transformation w=z1 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+z1 and w=z+1z−1
These are common in exam questions. The general approach is:
Express z in terms of w: z=f−1(w).
Apply the given condition on z (e.g. ∣z∣=2) to find the locus of w.
Critical points. A critical point of a transformation w=f(z) is a point z0 where
f′(z0)=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 under the transformation w=z1.
This is a circle with centre (21,0) and radius 21 in the
w-plane.
:::info AQA places significant emphasis on complex transformations including w=f(z) mappings.
Edexcel and OCR cover this topic with less depth. CIE focuses more on loci than on transformations.
:::
Problem 4. The fifth roots of unity are ω0,ω1,ω2,ω3,ω4 where
ω=e2πi/5. Show that 1+ω+ω2+ω3+ω4=0 and deduce that
cosL◆B◆2π◆RB◆◆LB◆5◆RB◆+cosL◆B◆4π◆RB◆◆LB◆5◆RB◆=−21.
Hint
Sum the geometric series. Then separate real and imaginary parts.
Answer
The roots satisfy z5−1=0. The coefficient of z4 is 0, so by Vieta’s formulas,
∑k=04ωk=0.
Problem 5. Sketch on separate Argand diagrams the loci given by (a) ∣z−1−i∣=∣z−3+i∣
And (b) arg(z−2)=L◆B◆π◆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−(3−i)∣ is the perpendicular bisector of the segment joining
(1,1) and (3,−1). The midpoint is (2,0) and the slope of the segment is
3−1−1−1=−1So the perpendicular bisector has slope 1 and equation y=x−2.
(b)arg(z−2)=L◆B◆π◆RB◆◆LB◆3◆RB◆ is a half-line from (2,0) at angle
L◆B◆π◆RB◆◆LB◆3◆RB◆ to the positive real axis. Its equation is y=3(x−2) for
x>2.
Intersection: Setting x−2=3(x−2):
(x−2)(1−3)=0So x=2 (gives y=0But the half-line requires x>2) or
1=3Which is false.
There is no intersection. The half-line from (2,0) at angle π/3 has slope 3While The
perpendicular bisector has slope 1, and they only meet at the point (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 under the transformation
w=z−1z+1.
Hint
Express z in terms of w and substitute ∣z∣=2.
Answer
w=z−1z+1⟹wz−w=z+1⟹z(w−1)=w+1⟹z=w−1w+1.
Since ∣z∣=2:
w−1w+1=2⟹∣w+1∣=2∣w−1∣
Let w=u+vi:
(u+1)2+v2=2(u−1)2+v2
Squaring: (u+1)2+v2=4[(u−1)2+v2]
u2+2u+1+v2=4u2−8u+4+4v2
0=3u2−10u+3+3v2
3u2−10u+3v2+3=0
3(u2−310u)+3v2=−3
3(u−35)2−325+3v2=−3
3(u−35)2+3v2=316
(u−35)2+v2=916
This is a circle with centre (35,0) and radius 34 in the
w-plane.
Problem 7. Express L◆B◆(1+i)6◆RB◆◆LB◆(1−i3)4◆RB◆ in the form a+bi.
Hint
Write each term in exponential form and use the laws of indices.
So:
21e5πi/6=21(cosL◆B◆5π◆RB◆◆LB◆6◆RB◆+isinL◆B◆5π◆RB◆◆LB◆6◆RB◆)=21(−L◆B◆3◆RB◆◆LB◆2◆RB◆+21i)=−L◆B◆3◆RB◆◆LB◆4◆RB◆+41i
Problem 8. The transformation T from the z-plane to the w-plane is given by w=z2. The
Region R in the z-plane is defined by 1≤∣z∣≤2 and
0≤arg(z)≤L◆B◆π◆RB◆◆LB◆4◆RB◆. Find and describe the image of R under T.
Hint
Under w=z2The modulus squares and the argument doubles.
Answer
If z=reiθThen w=r2e2iθ.
Modulus: 1≤r≤2⟹1≤r2≤4So 1≤∣w∣≤4.
Argument:
0≤θ≤L◆B◆π◆RB◆◆LB◆4◆RB◆⟹0≤2θ≤L◆B◆π◆RB◆◆LB◆2◆RB◆
so 0≤arg(w)≤L◆B◆π◆RB◆◆LB◆2◆RB◆.
The image is the region in the first quadrant of the w-plane between the circles ∣w∣=1 and
∣w∣=4Bounded by the rays arg(w)=0 and arg(w)=L◆B◆π◆RB◆◆LB◆2◆RB◆.
Problem 9. Solve the equation z3+z2+z+1=0 by recognising it as a geometric series,
And hence show that
cosL◆B◆2π◆RB◆◆LB◆4◆RB◆+cosL◆B◆4π◆RB◆◆LB◆4◆RB◆+cosL◆B◆6π◆RB◆◆LB◆4◆RB◆=−1.
Hint
Factor z3+z2+z+1=(z+1)(z2+1). The roots are the 4th roots of unity excluding z=1.
Answer
z3+z2+z+1=z−1z4−1=0⟹z4=1 with z=1.
The 4th roots of unity are 1,i,−1,−iSo the solutions are z=i,−1,−i.
Equivalently, the roots are ekπi/2 for k=1,2,3.
The sum of roots (by Vieta, coefficient of z2 divided by leading coefficient) is −1:
i+(−1)+(−i)=−1✓
Now: eiπ/2+eiπ+e3iπ/2=i+(−1)+(−i)=−1.
Separating real and imaginary parts:
cosL◆B◆π◆RB◆◆LB◆2◆RB◆+cosπ+cosL◆B◆3π◆RB◆◆LB◆2◆RB◆=−1I.e.
0−1+0=−1. ✓
Alternatively, the claim as stated uses
cosL◆B◆2π◆RB◆◆LB◆4◆RB◆+cosL◆B◆4π◆RB◆◆LB◆4◆RB◆+cosL◆B◆6π◆RB◆◆LB◆4◆RB◆=cosL◆B◆π◆RB◆◆LB◆2◆RB◆+cosπ+cosL◆B◆3π◆RB◆◆LB◆2◆RB◆=0+(−1)+0=−1.
✓ □
Problem 10. (a) Show that
L◆B◆1◆RB◆◆LB◆eiθ−1◆RB◆=−21−2icotL◆B◆θ◆RB◆◆LB◆2◆RB◆
For θ∈/2πZ.
(b) Hence, or otherwise, find k=1∑n−1L◆B◆1◆RB◆◆LB◆1−ωk◆RB◆
Where ω=e2πi/n.
Hint
For (a), multiply numerator and denominator by the conjugate e−iθ−1 and use half-angle
Identities. For (b), use the result from (a) with θ=2kπ/n.
Problem. Show that the product of all n-th roots of unity is (−1)n+1.
Solution. The n-th roots of unity are the roots of zn−1=0. By Vieta’s formulae, the
Product of all roots is (−1)n×(−1)=(−1)n+1.
Alternatively: the roots are 1,ω,ω2,…,ωn−1So the product is
ω0+1+2+⋯+(n−1)=ωn(n−1)/2=eπi(n−1)=(−1)n−1=(−1)n+1.
■
9. Common Pitfalls
| Pitfall | Correct Approach |
| ----------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------ | -------- | --- | --- | --- | --- | --- | --- | ---------------------------------------- | --- | ---- | --- | --- | --- | --- |
| Forgetting that arg(z) is measured from the positive real axis | arg(z) is the angle anticlockwise from the positive x-axis, range (−π,π] or [0,2π) |
| Confusing ∣z−w∣ with ∣z∣−∣w∣ | ∣z−w∣ is the distance between z and w; ∣z−w∣=∣z∣−∣w∣ |
| Missing roots when solving zn=w | There are always exactly n distinct roots; check your k values cover 0 to n−1 |
| Incorrectly applying de Moivre to non-integer powers | De Moivre’s theorem (cosθ+isinθ)n=cosnθ+isinnθ holds for integer n only |
10. Additional Exam-Style Questions
Question 8
Solve z4=8(1+i3)Giving roots in the form r(cosθ+isinθ).
Solution
8(1+i3)=16eiπ/3.
zk=2exp(L◆B◆i(π/3+2kπ)◆RB◆◆LB◆4◆RB◆) for k=0,1,2,3.
The complex number z satisfies ∣z−1∣=∣z+1∣ and ∣z−3i∣=3. Find z.
Solution
∣z−1∣=∣z+1∣: perpendicular bisector of 1 and −1Giving Re(z)=0. So z=iy.
∣z−3i∣=3⟹∣iy−3i∣=3⟹∣y−3∣=3⟹y−3=±3.
y=6 or y=0. So z=6i or z=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+isinx connects exponential, trigonometric, and hyperbolic functions. See
Hyperbolic Functions.
11.3 Complex numbers and polar coordinates
Argand diagrams and polar form (r,θ) connect to polar coordinates. See
Polar Coordinates.
12. Key Results Summary
| Result | Formula |
| ----------------------------- | --------------------------------------------------------- | --- | -------------------------------- | --- | --- |
| Modulus | ∣z∣=a2+b2 for z=a+bi |
| Argument | arg(z)=arctan(b/a) (adjusting for quadrant) |
| Euler’s formula | eiθ=cosθ+isinθ |
| De Moivre | (cosθ+isinθ)n=cosnθ+isinnθ |
| n-th roots of unity | zk=e2πik/n, k=0,…,n−1 |
| Locus: circle | ∣z−a∣=r |
| Locus: perpendicular bisector | ∣z−a∣=∣z−b∣ |
| Locus: half-line | arg(z−a)=θ |
13. Further Exam-Style Questions
Question 11
Solve z3=−8iGiving roots in Cartesian form.
Solution
−8i=8e−iπ/2. Roots: zk=2exp(L◆B◆−iπ/2+2kπi◆RB◆◆LB◆3◆RB◆)
For k=0,1,2.
z0=2e−iπ/6=2(L◆B◆3◆RB◆◆LB◆2◆RB◆−2i)=3−i.
z1=2eiπ/2=2i.
z2=2e7iπ/6=2(−L◆B◆3◆RB◆◆LB◆2◆RB◆−2i)=−3−i.
z=3−i,2i,−3−i
Question 12
Prove that∣z1z2∣=∣z1∣∣z2∣ for any complex numbers z1,z2.
Solution
Let z1=r1eiθ1 and z2=r2eiθ2.
z1z2=r1r2ei(θ1+θ2).
∣z1z2∣=r1r2=∣z1∣∣z2∣. ■
14. Advanced Topics
14.1 The exponential form of complex numbers
Any non-zero complex number can be written as z=reiθ where r=∣z∣ and
θ=arg(z).
This form makes multiplication and division particularly simple:
z1z2=r1r2ei(θ1+θ2)
z1/z2=(r1/r2)ei(θ1−θ2)
14.2 Euler’s identity
Setting θ=π in Euler’s formula: eiπ+1=0.
This connects five fundamental constants: e$$i$$\pi$$1And 0.
14.3 Complex conjugate and roots of polynomials
If P(z) is a polynomial with real coefficients and z=a+bi is a root, then zˉ=a−bi
Is also a root. This is because P(z)=P(zˉ) for real-coefficient polynomials.
The complex numbers z and w satisfy |z| = 3$$|w| = 4And ∣z+w∣=5. Find ∣z−w∣.
Solution
∣z+w∣2=∣z∣2+∣w∣2+2Re(zwˉ)=9+16+2Re(zwˉ)=25.
Re(zwˉ)=0.
∣z−w∣2=∣z∣2+∣w∣2−2Re(zwˉ)=9+16−0=25.
∣z−w∣=5
18. Further Advanced Topics
18.1 Riemann surfaces and multi-valued functions
The complex logarithm, n-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∪{∞} 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)=L◆B◆1◆RB◆◆LB◆2πi◆RB◆∮Cz−af(z)dz
Residue theorem: ∮Cf(z)dz=2πi∑Res(f,ak)
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: p≡1(mod4)⟹p=a2+b2
Wilson’s theorem: (p−1)!≡−1(modp) for prime p
19. Further Exam-Style Questions
Question 19
Express cos5θ+isin5θ in terms of cosθ and sinθ using the binomial
Theorem.