The imaginary unit i, algebraic and trigonometric forms, the four operations and division by the conjugate, modulus and argument, De Moivre's theorem, and quadratic equations with a negative discriminant.
The equation x² = −1 has no real solution, so we define the imaginary unit i with i² = −1. Every complex number is written in algebraic form: z = a + bi, where a is the real part and b the imaginary part.
Real numbers fill the whole number line, and on that line the equation x² = −1 has no answer at all: the square of any real number is zero or positive. The way out is not to search harder for a real solution but to enlarge the number system, exactly as negative numbers once enlarged the counting numbers and fractions enlarged the integers. We add one new symbol whose only task is to supply the missing square root, and then we let ordinary algebra carry the rest of the work.
The enlargement is as small as it can be. Adjoin i, close the result under +, −, × and ÷, and you get the complex numbers. Nothing further is required, and the payoff is large: in ℂ every polynomial equation has a root, so the awkward case "no solution" disappears from algebra for good.
Cardano met √−15 in 1545 while solving cubic equations and dismissed such expressions as useless. Bombelli, around 1572, showed that they could be manipulated by the ordinary rules and still produce correct real answers in the end. Descartes used the word "imaginary" as a criticism, and the label stuck even after Gauss proposed "complex number" in 1831 and Argand and Wessel supplied the plane picture. The vocabulary is historical, not mathematical: i is no less concrete than −1, which Greek mathematicians also refused to accept as a number.
A complex number is an ordered pair of real numbers, written a + bi, where a, b ∈ ℝ and i is a symbol with the single property i² = −1. The set of all such numbers is ℂ = {a + bi : a, b ∈ ℝ}. Two complex numbers are equal exactly when both coordinates agree: a + bi = c + di if and only if a = c and b = d. That one rule is what makes "equate real parts, equate imaginary parts" a legal move in every problem below, and it is the reason a single complex equation carries two real equations inside it.
The operations are defined by treating a + bi as a polynomial in the symbol i and then replacing i² by −1 wherever it appears:
(a + bi) + (c + di) = (a + c) + (b + d)i
(a + bi)(c + di) = ac + adi + bci + bdi² = (ac − bd) + (ad + bc)i
The real numbers sit inside ℂ as the numbers with b = 0, and they keep their old arithmetic. So ℂ extends ℝ rather than replacing it, and every result about real numbers remains true inside the larger system.
| Symbol | Name | Meaning |
|---|---|---|
| i | imaginary unit | i² = −1 |
| z, w | complex numbers | z = a + bi, w = c + di |
| Re(z) | real part | a, a real number |
| Im(z) | imaginary part | b, a real number, not bi |
| z̄ | conjugate | a − bi |
| |z| | modulus | √(a² + b²), never negative |
| arg z | argument | the angle θ from the positive real axis |
| ℂ | the complex plane | all numbers a + bi |
One warning that saves marks: Im(z) is the real number b. Writing Im(3 + 4i) = 4i is a category error, and examiners deduct for it because the whole point of the notation is to separate the two coordinates.
Because i² = −1, the powers of i do not grow, they rotate: i¹ = i, i² = −1, i³ = −i, i⁴ = 1, and then the list repeats. The cycle has length four, so any power of i can be reduced by dividing the exponent by 4 and reading only the remainder. This single fact answers a large share of exam questions on the topic, and it also explains the geometry: multiplying by i is a quarter turn anticlockwise, so four turns bring you home.
The conjugate of z = a + bi is z̄ = a − bi: same real part, opposite imaginary part. Geometrically it is the mirror image of z in the real axis, which is why conjugating twice changes nothing. The useful properties all follow from the definition:
The modulus is |z| = √(a² + b²), the length of the arrow from the origin to the point (a, b). It is a distance, so it is never negative, and |z| = 0 happens only for z = 0. The properties worth memorising are:
The argument is the angle θ that the arrow makes with the positive real axis, measured anticlockwise and normally in radians. Together, r = |z| and θ locate the point completely, giving a = r cos θ and b = r sin θ. Two cautions. First, tan θ = b/a gives only a reference angle; the quadrant must be decided from the signs of a and b, because the tangent has period π while the point has period 2π. Second, θ is not unique: adding any multiple of 2π gives the same point. The principal value, written Arg z, is the one taken in (−π, π], and it is the value a calculator returns.
Every algebraic operation has a picture, and the picture is usually faster than the algebra. Addition is the parallelogram rule on arrows. Multiplication by w scales lengths by |w| and rotates by arg w, which is why powers spiral outward and roots sit on a circle. Conjugation is reflection in the real axis. The equation |z − z₀| = R is a circle of centre z₀ and radius R, and |z − z₀| = |z − z₁| is the perpendicular bisector of the segment joining the two centres. Recognising these four pictures converts many "solve for z" questions into one-line answers.
(a+bi) + (c+di) = (a+c) + (b+d)i — add like terms.
(a+bi)(c+di) = (ac − bd) + (ad + bc)i — expand, then replace i² by −1.
Division: multiply numerator and denominator by the conjugate of the denominator: (a+bi)/(c+di) = [(a+bi)(c−di)]/(c²+d²).
z = r(cos θ + i sin θ) with r = |z| and θ the argument. For powers: zⁿ = rⁿ(cos nθ + i sin nθ).
Hence i² = −1, i³ = −i, i⁴ = 1, and the powers repeat with period four.
Let z = a + bi and w = c + di, where a, b, c, d are real numbers. Working with symbols first, and substituting numbers second, prevents most sign errors. Addition and subtraction act on the two coordinates separately:
z + w = (a + c) + (b + d)i
z − w = (a − c) + (b − d)i
Multiplication is the distributive law applied twice, followed by i² = −1:
z · w = ac + adi + bci + bdi² = (ac − bd) + (ad + bc)i
The real part ac − bd collects the terms without i, and the imaginary part ad + bc collects the terms with exactly one i. A fast check on any multiplication is to test the result with a simple value, for example z = w = i, where the formula must give −1.
Division is not a new operation; it is multiplication by the reciprocal, and the reciprocal is found by rationalising. For w ≠ 0,
z / w = (a + bi)/(c + di) = [(a + bi)(c − di)] / [(c + di)(c − di)] = [(ac + bd) + (bc − ad)i] / (c² + d²)
The denominator collapses to the real number c² + d² = |w|², which is exactly why the method works: multiplying by the conjugate turns the denominator into a modulus squared, and a real denominator lets you split the answer into real and imaginary parts. Multiplying only the numerator by the conjugate changes the value of the fraction and is the most common error in the topic. As a final step, check that the result times w gives z back.
Since i² = −1, the powers repeat in a cycle of length four. Write the exponent as 4q + r with 0 ≤ r ≤ 3, then iⁿ = iʳ:
| Remainder r | iⁿ equals | Example |
|---|---|---|
| 0 | 1 | i¹⁰⁰ = 1, i²⁰²⁴ = 1 |
| 1 | i | i⁴⁵ = i |
| 2 | −1 | i²⁰²² = −1 |
| 3 | −i | i²⁰²³ = −i |
The same rule handles negative exponents, because i⁻ⁿ = 1/iⁿ. For example i⁻¹ = −i and i⁻² = −1.
With r = |z| and θ = arg z, the point (a, b) can be written in polar coordinates, which gives the trigonometric form
z = r(cos θ + i sin θ), where r = √(a² + b²), cos θ = a/r, sin θ = b/r.
The conversion back is a = r cos θ and b = r sin θ. Two numbers in this form are equal when their moduli are equal and their arguments differ by a multiple of 2π. The form is not unique, but the convention r ≥ 0 with −π < θ ≤ π is standard, and it is the one to use when the question says "the" argument.
Multiplication multiplies the moduli and adds the arguments; division divides the moduli and subtracts the arguments:
r₁(cos θ₁ + i sin θ₁) · r₂(cos θ₂ + i sin θ₂) = r₁r₂[cos(θ₁ + θ₂) + i sin(θ₁ + θ₂)]
r₁(cos θ₁ + i sin θ₁) / r₂(cos θ₂ + i sin θ₂) = (r₁/r₂)[cos(θ₁ − θ₂) + i sin(θ₁ − θ₂)]
In words: lengths multiply, angles add. This is the single most useful sentence in the topic, and it turns an ugly algebraic expansion into one line.
Applying the multiplication rule n times gives De Moivre's theorem: for any integer n,
[r(cos θ + i sin θ)]ⁿ = rⁿ(cos nθ + i sin nθ)
The modulus is raised to the power and the argument is multiplied by the power, with the angle reduced modulo 2π at the end. The theorem also holds for negative n, where rⁿ means 1/r^|n|. It is the reason large powers such as (1 + i)²⁰ are computed in seconds in polar form and are hopeless by repeated expansion.
The same theorem run backwards gives the n distinct n-th roots of a complex number. For w = r(cos θ + i sin θ) and k = 0, 1, ..., n − 1,
z_k = r^(1/n) [cos((θ + 2πk)/n) + i sin((θ + 2πk)/n)]
The roots all have modulus r^(1/n), so they lie on a circle centred at the origin, spaced evenly by 2π/n. For w = 1 the roots are the n-th roots of unity, ω_k = cos(2πk/n) + i sin(2πk/n), with ω = ω₁ generating all of them as powers. Their sum is zero and their product is (−1)^(n−1); both facts are quick checks on a finished answer.
For ax² + bx + c = 0 with real coefficients and a ≠ 0, completing the square gives (x + b/2a)² = (b² − 4ac)/(4a²). If the discriminant Δ = b² − 4ac is negative, write Δ = −|Δ| and take the square root as i√|Δ|:
x = [−b ± i√(−Δ)] / (2a)
The two roots are conjugates of each other, they still satisfy Vieta's relations x₁ + x₂ = −b/a and x₁x₂ = c/a, and the parabola simply misses the x-axis. Real-coefficient polynomials always have complex roots in conjugate pairs, which is a useful way to recover a missing root from a given one.
Figure 1: the complex plane. z = a + bi is the point (a, b); the distance from the origin is the modulus |z| and the angle with the real axis is the argument θ.
Every complex number is a point and also an arrow from the origin. The horizontal coordinate is the real part, the vertical coordinate is the imaginary part, and the arrow's length is the modulus. Read a diagram in this order: locate the point, drop the two coordinates, then measure the angle from the positive real axis. That order prevents the usual confusion between the point (a, b) and the pair of numbers a and b.
Figure 2: a conjugate pair and the modulus circle. The point z = 3 + 2i and its mirror image z̄ = 3 − 2i are reflections in the real axis. Both lie on the same dashed circle, because conjugation preserves the modulus: |z| = |z̄| = √13. The two arrows have equal length and opposite angles θ and −θ, so their sum 6 is real and their product 13 is real as well.
| Algebraic statement | Picture | Consequence |
|---|---|---|
| z + w | parallelogram on the two arrows | |z + w| ≤ |z| + |w| |
| z − w | arrow from w to z | |z − w| is a distance |
| z · w | rotate by arg w, scale by |w| | arguments add |
| z̄ | reflection in the real axis | modulus unchanged, angle negated |
| i · z | quarter turn anticlockwise | i⁴ = 1 returns to start |
| |z − z₀| = R | circle, centre z₀ | locus of constant distance |
Figure 3 in words: powers spiral, roots divide the circle. Squaring doubles the angle and squares the radius, so repeated powers of a number with |z| > 1 spiral outward while a number with |z| < 1 spirals inward. The n-th roots of a number all share one radius and sit at equal angles, which is why the cube roots of 1 form an equilateral triangle on the unit circle and the fourth roots form a square. Sketching those few points before computing often reveals the answer.
(2 + 3i) + (4 − i) = 6 + 2i.
(1 + 2i)(3 − i) = 3 − i + 6i − 2i² = 3 + 5i + 2 = 5 + 5i.
(1 + i)/(1 − i) = (1+i)²/((1−i)(1+i)) = 2i/2 = i.
Solve x² + 2x + 5 = 0: discriminant 4 − 20 = −16, so x = (−2 ± 4i)/2 = −1 ± 2i.
Compute (3 + 4i)/(2 − i).
Write z = −1 + √3 i in trigonometric form and compute z³.
Compute (1 + i)⁸.
Find z with z² = 3 − 4i.
Solve z⁴ = 16.
Solve z² − 4z + 13 = 0.
Describe the set of points z with |z − 2| = 3, and then the set with |z − 2| = |z + 2i|.
Show that |z w| = |z| |w| for all complex numbers z and w.
| Form | Shape | Best for |
|---|---|---|
| Algebraic | a + bi | addition, subtraction, multiplication |
| Trigonometric | r(cos θ + i sin θ) | multiplication, division, powers |
| Exponential | r·e^(iθ) | compact powers and roots |
| Number | Re | Im | Conjugate | Modulus |
|---|---|---|---|---|
| 3 + 4i | 3 | 4 | 3 − 4i | 5 |
| −2 + i | −2 | 1 | −2 − i | √5 |
| 5 | 5 | 0 | 5 | 5 |
| −3i | 0 | −3 | 3i | 3 |
| 1 − i | 1 | −1 | 1 + i | √2 |
| Operation | Formula | Practical method | Quick example |
|---|---|---|---|
| Addition | (a + c) + (b + d)i | add like parts | (2 + 3i) + (4 − i) = 6 + 2i |
| Subtraction | (a − c) + (b − d)i | subtract like parts | (2 + 3i) − (4 − i) = −2 + 4i |
| Multiplication | (ac − bd) + (ad + bc)i | expand, then i² = −1 | (1 + 2i)(3 − i) = 5 + 5i |
| Division | [(ac + bd) + (bc − ad)i]/(c² + d²) | multiply by the conjugate | (1 + i)/(1 − i) = i |
| Power | rⁿ(cos nθ + i sin nθ) | De Moivre in polar form | (1 + i)² = 2i |
| n-th root | r^(1/n) at (θ + 2πk)/n | n roots, equally spaced | √16 has roots ±4 |
| Situation | Use | Reason |
|---|---|---|
| add, subtract | algebraic | coordinates combine directly |
| multiply two binomials | algebraic | expansion is short and exact |
| divide | algebraic with conjugate | the denominator becomes |w|² |
| high power | trigonometric | angle multiplies instead of expanding |
| roots | trigonometric | roots are equally spaced in angle |
| locus or geometry | drawing, then modulus | distances are visible |
| Identity | Statement | Typical use |
|---|---|---|
| Modulus squared | |z|² = z z̄ | turn distance into algebra |
| Real and imaginary parts | Re z = (z + z̄)/2, Im z = (z − z̄)/(2i) | extract coordinates from a formula |
| Reciprocal | 1/z = z̄/|z|² | invert without long division |
| Unit modulus | |z| = 1 implies z̄ = 1/z | simplify expressions on the unit circle |
| Triangle inequality | |z + w| ≤ |z| + |w| | bound a modulus without computing it |
In the exam: the most frequent asks are simplifying a power of i (divide the exponent by 4 and take the remainder), the modulus, and division by the conjugate.
Start every question by classifying it: is it a computation, a conversion, a power, or a locus? The classification decides the form to use, and choosing the form is most of the work. For a power, go polar immediately; for a division, go conjugate immediately; for a locus, draw the points before writing any algebra.
Budget the time in proportion to marks: a two-mark modulus or power-of-i item should take under a minute, a division three or four minutes including the check, and a De Moivre power with a large exponent about five. If a division is taking longer, the denominator is probably not the conjugate you think it is.
Always finish with a cheap verification. For a division, multiply the answer by the denominator. For a power, test the modulus: |zⁿ| must equal |z|ⁿ. For a quadratic, check the sum and product of the roots. For a locus, test one obvious point on the curve. These checks catch most sign errors in seconds.
Write the conjugate multiplication explicitly rather than in your head. Examiners award method marks for the line [(a + bi)(c − di)]/(c² + d²), and that line also protects you from the commonest error in the topic. Finally, if an answer contains a surd, leave it exact unless the question asks for a decimal: √13 is a better answer than 3.61, and it is the one that can be checked by squaring.
Compute i¹⁰⁰.
Solution: 100 ÷ 4 leaves remainder 0, so i¹⁰⁰ = 1.
What is the modulus of 3 − 4i?
Solution: √(9 + 16) = 5.
Simplify (2 + i)(2 − i).
Solution: 4 − i² = 4 + 1 = 5.
Write 1/i in algebraic form.
Solution: 1/i = −i.
Solve x² + 4 = 0.
Solution: x = ±2i.
Write √3 + i in trigonometric form.
Solution: r = √(3 + 1) = 2. The point is in the first quadrant, so θ = arctan(1/√3) = π/6. Therefore √3 + i = 2(cos π/6 + i sin π/6). Check: 2 cos π/6 = √3 and 2 sin π/6 = 1.
Compute (1 − i)⁶.
Solution: |1 − i| = √2 and arg(1 − i) = −π/4. By De Moivre, the modulus becomes (√2)⁶ = 8 and the angle becomes −6π/4 = −3π/2. Since cos(−3π/2) = 0 and sin(−3π/2) = 1, the value is 8i. Check by cubing twice: (1 − i)² = −2i, so (1 − i)³ = −2i(1 − i) = −2 − 2i, and squaring that gives 4 + 8i − 4 = 8i.
Find the modulus of (2 + 3i)(1 − i) without expanding.
Solution: moduli multiply, so the answer is |2 + 3i| · |1 − i| = √13 · √2 = √26. Expanding as a check: (2 + 3i)(1 − i) = 2 − 2i + 3i − 3i² = 5 + i, and |5 + i| = √26.
Solve z² − 4z + 13 = 0 and verify with Vieta.
Solution: Δ = 16 − 52 = −36, so z = (4 ± 6i)/2 = 2 ± 3i. Sum: (2 + 3i) + (2 − 3i) = 4 = −b/a. Product: (2 + 3i)(2 − 3i) = 4 + 9 = 13 = c/a.
Simplify i + i² + i³ + i⁴.
Solution: substitute the cycle: i + (−1) + (−i) + 1 = 0. In general, any four consecutive powers of i sum to zero, which is why sums of this type collapse.
For z = 2 − i, compute z z̄ and 1/z.
Solution: z z̄ = 2² + 1² = 5. Then 1/z = z̄/|z|² = (2 + i)/5 = 2/5 + (1/5)i. Check: (2 − i)(2 + i)/5 = 5/5 = 1.
Find the cube roots of 8.
Solution: write 8 = 8(cos 0 + i sin 0). The roots have modulus 8^(1/3) = 2 and angles 0, 2π/3, 4π/3. They are 2, 2(cos 2π/3 + i sin 2π/3) = −1 + √3 i, and −1 − √3 i. Check the sum: 2 − 1 + √3 i − 1 − √3 i = 0, as required.
Find the argument of −2i, and of −1 − i.
Solution: −2i lies on the negative imaginary axis, so its principal argument is −π/2. For −1 − i the point is in the third quadrant with reference angle π/4, so arg = −3π/4 (equivalently 5π/4).
A complex number has |z| = 5 and Re z = 3. Find its possible imaginary parts and both numbers.
Solution: 9 + b² = 25 gives b² = 16, so b = ±4. The numbers are 3 + 4i and 3 − 4i; they are conjugates, as the symmetry of the circle requires.
Simplify (1 + i)/(2 − i) and check the result.
Solution: multiply by 2 + i: numerator (1 + i)(2 + i) = 2 + i + 2i + i² = 1 + 3i; denominator 4 + 1 = 5. So the value is (1 + 3i)/5 = 1/5 + (3/5)i. Check by multiplying by (2 − i): (1 + 3i)(2 − i)/5 = (2 − i + 6i − 3i²)/5 = (5 + 5i)/5 = 1 + i.
Show that the four fourth roots of 1 sum to zero, and name them.
Solution: the roots are 1, i, −1, −i, obtained from angles 0, π/2, π, 3π/2. Their sum is 1 + i − 1 − i = 0, and their product is 1 · i · (−1) · (−i) = −1, which matches (−1)^(n−1) with n = 4.
Free · three languages · works offline