Skip to main content

Polynomials over the complex field in HSC Mathematics Extension 2: solving quadratics with complex coefficients, the conjugate root theorem, and complete factorisation

Syllabus dot point

“Solve quadratic equations with complex coefficients and factorise polynomials over the complex field, using the conjugate root theorem for real polynomials”

HSCMaths Extension 2Using Complex Numbers (MEX-N2)13 min read

Quick answer

The quadratic formula holds over C\mathbb{C}: z=−b±b2−4ac2az = \dfrac{-b \pm \sqrt{b^2 - 4ac}}{2a}, where the square root is a complex square root. For a polynomial with real coefficients, if α\alpha is a root then so is αˉ\bar{\alpha} (conjugate root theorem). The fundamental theorem of algebra guarantees a degree nn polynomial has exactly nn roots in C\mathbb{C} counted with multiplicity, so it factorises completely into nn linear factors over C\mathbb{C}.

Jump to a section
  1. What this dot point is asking
  2. Quadratics with complex coefficients
  3. The conjugate root theorem
  4. The fundamental theorem of algebra
  5. Sum and product of roots (Vieta)
  6. Strategy for complete factorisation
  7. How exam questions ask about complex polynomials
  8. Exam-style questions

What this dot point is asking

NESA wants you to work with polynomials over the complex numbers. You must solve quadratic equations whose coefficients may themselves be complex, apply the conjugate root theorem to real polynomials (complex roots occur in conjugate pairs), use the fundamental theorem of algebra to count roots, and factorise a polynomial completely into linear factors over C\mathbb{C}. The conjugate root theorem is the workhorse: it is what lets a single supplied complex root unlock a whole real polynomial.

Quadratics with complex coefficients

The quadratic formula

z=−b±b2−4ac2az = \frac{-b \pm \sqrt{b^2 - 4ac}}{2a}

applies to any quadratic az2+bz+c=0az^2 + bz + c = 0, even when a,b,ca, b, c are complex. The new feature is that b2−4ac\sqrt{b^2 - 4ac} is a complex square root. To find w\sqrt{w} for a complex w=p+qiw = p + qi, set w=u+vi\sqrt{w} = u + vi and solve u2−v2=pu^2 - v^2 = p and 2uv=q2uv = q (often together with the modulus equation u2+v2=∣w∣u^2 + v^2 = |w|, which makes the algebra fall out quickly). There are two square roots, differing only in sign, and the ±\pm in the formula accounts for both.

When the coefficients are real but the discriminant is negative, the formula still works directly: b2−4ac=−(4ac−b2)=i4ac−b2\sqrt{b^2 - 4ac} = \sqrt{-(4ac - b^2)} = i\sqrt{4ac - b^2}, and the two roots come out as a conjugate pair immediately.

The conjugate root theorem

When a polynomial P(z)P(z) has only real coefficients, its non-real roots come in conjugate pairs. That is, if α=a+bi\alpha = a + bi (with b≠0b \neq 0) satisfies P(α)=0P(\alpha) = 0, then P(αˉ)=0P(\bar{\alpha}) = 0 also. The reason is that conjugation respects addition and multiplication, so P(α)‾=P(αˉ)\overline{P(\alpha)} = P(\bar{\alpha}); since P(α)=0P(\alpha) = 0 and 0ˉ=0\bar{0} = 0, we get P(αˉ)=0P(\bar{\alpha}) = 0.

Geometrically, the roots of a real polynomial are symmetric about the real axis: every non-real root is mirrored by its conjugate. The figure builds this picture for the cubic z3−5z2+17z−13z^3 - 5z^2 + 17z - 13, whose roots are 11, 2+3i2 + 3i and 2−3i2 - 3i.

Stage 1, a single non-real root. Suppose a question tells you that 2+3i2 + 3i is a root of a polynomial with real coefficients. Plot it in the upper half-plane.

Conjugate root pairs of a real polynomial, stage 1 For a polynomial with real coefficients, non-real roots occur in conjugate pairs symmetric about the real axis. Here the cubic has roots 1, 2 + 3i and 2 - 3i. Re Im 1 2 3 3i -3i 2 + 3i O Given a non-real root 2 + 3i of a real polynomial ...

Stage 2, the conjugate must appear. Because the coefficients are real, the conjugate 2−3i2 - 3i is forced to be a root too. It is the mirror image of 2+3i2 + 3i in the real axis, so the two roots sit symmetrically above and below.

Conjugate root pairs of a real polynomial, stage 2 For a polynomial with real coefficients, non-real roots occur in conjugate pairs symmetric about the real axis. Here the cubic has roots 1, 2 + 3i and 2 - 3i. Re Im 1 2 3 3i -3i reflect 2 + 3i 2 - 3i O ... its conjugate 2 - 3i must also be a root (mirror in Re axis).

Stage 3, the remaining root is real. The pair 2±3i2 \pm 3i contributes a real quadratic factor z2−4z+13z^2 - 4z + 13. A cubic has three roots, so the third must be real (a lone non-real root could not pair up). Here it is z=1z = 1, sitting on the real axis.

Conjugate root pairs of a real polynomial, stage 3 For a polynomial with real coefficients, non-real roots occur in conjugate pairs symmetric about the real axis. Here the cubic has roots 1, 2 + 3i and 2 - 3i. Re Im 1 2 3 3i -3i reflect 2 + 3i 2 - 3i 1 O The real cubic has roots 1, 2 + 3i, 2 - 3i: a real root + a pair.

A consequence of the pairing is that each conjugate pair α,αˉ\alpha, \bar{\alpha} contributes a real quadratic factor

(z−α)(z−αˉ)=z2−(α+αˉ)z+ααˉ=z2−2Re⁡(α) z+∣α∣2,(z - \alpha)(z - \bar{\alpha}) = z^2 - (\alpha + \bar{\alpha})z + \alpha\bar{\alpha} = z^2 - 2\operatorname{Re}(\alpha)\,z + |\alpha|^2,

which has real coefficients. This explains why every real polynomial factorises over the reals into linear and irreducible quadratic factors, and why a real polynomial of odd degree must have at least one real root.

The fundamental theorem of algebra

The fundamental theorem of algebra states that every polynomial of degree n≥1n \ge 1 with complex coefficients has at least one complex root. By repeatedly factoring out roots, a degree nn polynomial has exactly nn roots in C\mathbb{C} counted with multiplicity, and so factorises completely as

P(z)=a(z−z1)(z−z2)⋯(z−zn),P(z) = a(z - z_1)(z - z_2)\cdots(z - z_n),

where aa is the leading coefficient and z1,…,znz_1, \ldots, z_n are the roots. Over C\mathbb{C} there are no irreducible quadratics: everything splits into linear factors.

Sum and product of roots (Vieta)

Vieta's formulas relate the roots to the coefficients and are often the fastest route once a pair of roots is known. For a monic polynomial zn+cn−1zn−1+⋯+c0z^n + c_{n-1}z^{n-1} + \dots + c_0, the sum of the roots is −cn−1-c_{n-1} and the product of the roots is (−1)nc0(-1)^n c_0. As the 2023 quartic question shows, knowing one conjugate pair plus these two relations can pin down the remaining roots without any division.

Strategy for complete factorisation

Given a real polynomial, find one root (by inspection, the rational root test, or a supplied value). If it is non-real, immediately pair it with its conjugate to obtain a real quadratic factor, then divide. Continue until only linear factors and at most one quadratic remain, then solve the remaining quadratic by formula. If the question asks for factorisation over C\mathbb{C}, finish by splitting every real quadratic into its two linear conjugate factors.

Key fact

The quadratic formula z=−b±b2−4ac2az = \dfrac{-b \pm \sqrt{b^2 - 4ac}}{2a} holds over C\mathbb{C} (the root is a complex square root). For a polynomial with real coefficients, non-real roots occur in conjugate pairs α,αˉ\alpha, \bar\alpha, each contributing the real quadratic factor z2−2Re⁡(α)z+∣α∣2z^2 - 2\operatorname{Re}(\alpha)z + |\alpha|^2. The fundamental theorem of algebra gives a degree nn polynomial exactly nn roots in C\mathbb{C}, so it factorises into nn linear factors.

How exam questions ask about complex polynomials

  • "Solve az2+bz+c=0az^2 + bz + c = 0 ... in Cartesian form": apply the quadratic formula; if the discriminant is negative (real coefficients) the roots are a conjugate pair p±qip \pm qi.
  • "Find the square roots of ww": set (u+iv)2=w(u + iv)^2 = w, equate real and imaginary parts, and solve (add the modulus equation to speed it up).
  • "α\alpha is a zero of P(z)P(z) with real coefficients. Explain why αˉ\bar\alpha is also a zero": cite the conjugate root theorem (conjugation preserves ++ and ×\times, so P(α)‾=P(αˉ)=0\overline{P(\alpha)} = P(\bar\alpha) = 0).
  • "Find the remaining zeros / factorise P(z)P(z)": pair the given root with its conjugate to get a real quadratic, then divide or use Vieta's sum and product.
  • "Factorise P(z)P(z) over C\mathbb{C} / over R\mathbb{R}": over C\mathbb{C} split into all linear factors; over R\mathbb{R} leave conjugate pairs combined as real quadratics.
  • "Hence solve …\ldots": a square-root or factor result from an earlier part is meant to be reused, often via completing the square.
Worked example

Factorise P(z)=z3−5z2+17z−13P(z) = z^3 - 5z^2 + 17z - 13 completely over C\mathbb{C}

Factorise P(z)=z3−5z2+17z−13P(z) = z^3 - 5z^2 + 17z - 13 completely over C\mathbb{C}, given that 2+3i2 + 3i is a root.

We are given one complex root of a real polynomial. The conjugate root theorem and polynomial division will give us all three roots and hence the complete factorisation.

Apply the conjugate root theorem: Because all coefficients of P(z)P(z) are real, non-real roots must occur in conjugate pairs. Since 2+3i2 + 3i is a root, 2−3i2 - 3i is also a root. We now have two of the three roots.

Form the real quadratic factor from the conjugate pair: Multiplying the two linear factors corresponding to the conjugate pair gives a quadratic with real coefficients. Using α+αˉ=4\alpha + \bar{\alpha} = 4 for the sum and ∣α∣2=22+32=13|\alpha|^2 = 2^2 + 3^2 = 13 for the product:

(z−(2+3i))(z−(2−3i))=z2−4z+13.(z - (2 + 3i))(z - (2 - 3i)) = z^2 - 4z + 13.

Divide P(z)P(z) by the quadratic factor: Dividing by the known factor reveals the remaining linear factor. Performing polynomial division:

z3−5z2+17z−13=(z2−4z+13)(z−1).z^3 - 5z^2 + 17z - 13 = (z^2 - 4z + 13)(z - 1).

The third root is therefore z=1z = 1. We can check the constant term: 13×(−1)=−1313 \times (-1) = -13, which matches PP.

Write the complete factorisation and verify: Combining all three linear factors gives the factorisation over C\mathbb{C}:

P(z)=(z−1)(z−(2+3i))(z−(2−3i)).P(z) = (z - 1)(z - (2 + 3i))(z - (2 - 3i)).

Verify by substituting z=1z = 1: P(1)=1−5+17−13=0P(1) = 1 - 5 + 17 - 13 = 0. Correct.

Final answer: P(z)=(z−1)(z−(2+3i))(z−(2−3i))P(z) = (z - 1)(z - (2 + 3i))(z - (2 - 3i)).

Worked example: a quadratic with complex coefficients

Solve z2−(3+i)z+(4+3i)=0z^2 - (3 + i)z + (4 + 3i) = 0

Solve z2−(3+i)z+(4+3i)=0z^2 - (3 + i)z + (4 + 3i) = 0, giving the roots in Cartesian form.

Identify the coefficients and the discriminant: Here a=1a = 1, b=−(3+i)b = -(3 + i), c=4+3ic = 4 + 3i. The discriminant is b2−4acb^2 - 4ac. Compute b2=(3+i)2=9+6i+i2=8+6ib^2 = (3 + i)^2 = 9 + 6i + i^2 = 8 + 6i, and 4ac=4(4+3i)=16+12i4ac = 4(4 + 3i) = 16 + 12i, so

b2−4ac=(8+6i)−(16+12i)=−8−6i.b^2 - 4ac = (8 + 6i) - (16 + 12i) = -8 - 6i.

Find a complex square root of the discriminant: Set −8−6i=u+iv\sqrt{-8 - 6i} = u + iv, so u2−v2=−8u^2 - v^2 = -8 and 2uv=−62uv = -6. The modulus equation u2+v2=∣−8−6i∣=64+36=10u^2 + v^2 = |-8 - 6i| = \sqrt{64 + 36} = 10 helps: adding to u2−v2=−8u^2 - v^2 = -8 gives 2u2=22u^2 = 2, so u2=1u^2 = 1 and u=±1u = \pm 1; then v2=9v^2 = 9, v=±3v = \pm 3. Since 2uv=−6<02uv = -6 < 0, uu and vv have opposite signs, so −8−6i=±(1−3i)\sqrt{-8 - 6i} = \pm(1 - 3i).

Apply the quadratic formula: With −b=3+i-b = 3 + i and the square root ±(1−3i)\pm(1 - 3i):

z=(3+i)±(1−3i)2.z = \frac{(3 + i) \pm (1 - 3i)}{2}.

Compute and state both roots: Take the two signs in turn:

z=(3+i)+(1−3i)2=4−2i2=2−i,z=(3+i)−(1−3i)2=2+4i2=1+2i.z = \frac{(3 + i) + (1 - 3i)}{2} = \frac{4 - 2i}{2} = 2 - i, \qquad z = \frac{(3 + i) - (1 - 3i)}{2} = \frac{2 + 4i}{2} = 1 + 2i.

The roots are not conjugates of each other, which is expected because the coefficients are not all real.

Final answer: z=2−iz = 2 - i or z=1+2iz = 1 + 2i.

Common traps
Applying the conjugate root theorem to complex coefficients
Roots come in conjugate pairs only when all coefficients are real. A polynomial with non-real coefficients need not have conjugate-paired roots (as the worked example shows).
Forgetting αˉ\bar{\alpha} when forming the quadratic factor
Using only α\alpha gives a complex linear factor and loses the clean real quadratic, complicating the division.
Sign error in z2−2Re⁡(α)z+∣α∣2z^2 - 2\operatorname{Re}(\alpha)z + |\alpha|^2
The middle coefficient is minus twice the real part, and the constant is the squared modulus, not the modulus.
Leaving a quadratic unfactored over C\mathbb{C}
Over the complex field every quadratic factors into two linear terms; the question asks for complete factorisation, so finish the job.
Choosing the wrong sign pairing in a square root
When 2uv2uv is negative, uu and vv must have opposite signs; do not just take both positive.
Exam technique

When a complex root is handed to you and the coefficients are real, write down its conjugate immediately, that is often the first mark. Build the real quadratic factor as z2−2Re⁡(α)z+∣α∣2z^2 - 2\operatorname{Re}(\alpha)z + |\alpha|^2 and either divide or use Vieta's sum and product of roots to get the rest, whichever the marks suggest. State clearly whether you are factorising over R\mathbb{R} (keep real quadratics) or over C\mathbb{C} (split into linear factors). For a quadratic with complex coefficients, do not expect conjugate roots; carry the complex square root through the formula carefully and verify a root by substitution at the end.

Exam-style questions

Questions in the style of NESA exam questions on this dot point, each with a worked answer. They are written by ExamExplained unless tagged "Past paper"; the year shows the paper a question is modelled on.

2023 HSC-style2 marks
Solve the quadratic equation z^2 - 3z + 4 = 0, where z is a complex number. Give your answers in Cartesian form.
Show worked answer →

Apply the quadratic formula with a = 1, b = -3, c = 4.

z = (3 +/- sqrt(9 - 16))/2 = (3 +/- sqrt(-7))/2.

Since sqrt(-7) = sqrt(7) i, this gives

z = 3/2 +/- (sqrt(7)/2) i.

So the two solutions in Cartesian form are z = 3/2 + (sqrt(7)/2) i and z = 3/2 - (sqrt(7)/2) i (a conjugate pair, as expected for a real quadratic with negative discriminant).

Mark notes: 1 mark for a correct use of the quadratic formula reaching sqrt(-7), 1 mark for both solutions in Cartesian form.

2023 HSC-style3 marks
The complex number 2 + i is a zero of the polynomial P(z) = z^4 - 3z^3 + c z^2 + d z - 30 where c and d are real numbers. (i) Explain why 2 - i is also a zero of P(z). (ii) Find the remaining zeros of P(z).
Show worked answer →

Part (i). Because P(z) has real coefficients, its non-real zeros occur in complex conjugate pairs (the conjugate root theorem). Since 2 + i is a zero, its conjugate 2 - i is also a zero.

Part (ii). Let the other two zeros be a and b. P(z) is monic of degree 4, so by Vieta's formulas:

Sum of zeros = -(-3)/1 = 3: (2 + i) + (2 - i) + a + b = 3, so 4 + a + b = 3, giving a + b = -1.

Product of zeros = -30/1 = -30: (2 + i)(2 - i) a b = -30. Since (2 + i)(2 - i) = 4 + 1 = 5, we get 5 a b = -30, so a b = -6.

The remaining zeros satisfy t^2 + t - 6 = 0, that is (t + 3)(t - 2) = 0.

So the remaining zeros are z = 2 and z = -3.

Mark notes: 1 mark for the conjugate root theorem explanation, then for part (ii) 1 mark for setting up sum and product of roots and 1 mark for solving to get z = 2 and z = -3.

2021 HSC-style4 marks
(i) Find the two square roots of -i, giving the answers in the form x + iy, where x and y are real numbers. (ii) Hence, or otherwise, solve z^2 + 2z + 1 + i = 0 giving your solutions in the form a + ib where a and b are real numbers.
Show worked answer →

Part (i). Let (x + iy)^2 = -i, so x^2 - y^2 + 2xy i = 0 - 1 i. Equating parts: x^2 - y^2 = 0 and 2xy = -1. From x^2 = y^2 with opposite signs needed (since 2xy < 0), take y = -x; then 2x(-x) = -1 gives x^2 = 1/2, so x = 1/sqrt(2), y = -1/sqrt(2) or x = -1/sqrt(2), y = 1/sqrt(2).

The two square roots are (1/sqrt(2))(1 - i) and -(1/sqrt(2))(1 - i), that is sqrt(2)/2 - (sqrt(2)/2) i and -sqrt(2)/2 + (sqrt(2)/2) i.

Part (ii). Complete the square: z^2 + 2z + 1 + i = (z + 1)^2 + i = 0, so (z + 1)^2 = -i.

Then z + 1 = +/- sqrt(-i) = +/- (sqrt(2)/2)(1 - i) from part (i).

z = -1 + (sqrt(2)/2)(1 - i) = -1 + sqrt(2)/2 - (sqrt(2)/2) i,
or z = -1 - (sqrt(2)/2)(1 - i) = -1 - sqrt(2)/2 + (sqrt(2)/2) i.

Mark notes: parts (i) and (ii) were 2 marks each. 1 mark for setting up the simultaneous equations and 1 mark for both square roots; then 1 mark for reducing the equation to (z + 1)^2 = -i and 1 mark for both solutions.

Practise this

Sources & how we know this

ExamExplained