Linear Algebra and Vector Geometry 201-SN4-RE • CEGEP, Montreal

Corrected exercises: complex numbers, De Moivre's theorem and nth roots (201-SN4-RE)

This is the corrected exercise set on complex numbers for Linear Algebra and Vector Geometry, 201-SN4-RE (formerly 201-NYC), the CEGEP course of the Science program. Part A covers the basics: operations in algebraic form, conjugate, modulus and division, square roots by the algebraic method, conversion to trigonometric and exponential form, De Moivre's theorem and powers, and the nnth roots of a complex number. Part B moves to exam level: polynomial equations with real coefficients, a proof of the conjugate root theorem, a full problem on the six sixth roots of 6464, complex multiplication read as a rotation with the equilateral triangle criterion, the triple-angle formulas derived from De Moivre's theorem, and loci (perpendicular bisectors and circles) in the complex plane. Part C adds five classics: the fifth roots of unity and cos⁡72∘\cos 72^{\circ}, the factorization of z4+4z^{4}+4, a quadratic equation with complex coefficients, the Mandelbrot iteration and the addition of signals with phasors.

Exercises 0.1 and 0.2 open the set with nineteen direct calculations on complex numbers, like the drill lists of the textbooks. First ten calculations in algebraic form: sum and difference, product, square, powers of ii such as i103i^{103}, modulus, conjugate of a product, reciprocal and quotient through the conjugate, up to the complex solutions of 2z2−2z+5=02z^{2}-2z+5=0. Then nine conversions to exponential form: modulus and principal argument on the axes and in the quadrants, the algebraic form of 6ei5π66e^{i\frac{5\pi}{6}}, product and quotient in exponential form, powers by De Moivre's theorem such as (3−i)8(\sqrt{3}-i)^{8} and (1−i)−6(1-i)^{-6}, and a cube root of −27-27. Every value is checked as you type it.

This chapter closes the semester and its reputation is misleading: students find it easy as long as they stay in algebraic form, then get stuck as soon as arguments come in. The reason is almost always the same, and it has nothing to do with complex numbers: the unit circle is not under control. An argument placed in the wrong quadrant brings down the rest of the question. The traps are handled where they fall: the sign of −8i2-8i^{2} (exercise 1), the calculator arctangent that returns the angle of the opposite quadrant (exercise 2), the search for nnth roots stopped at the first one (exercise 4), the conjugate root property invoked for non-real coefficients (exercise 13) and the amplitude 3+4=73+4=7 instead of 55 (exercise 15).

The guiding principle of the chapter: complex numbers are added in algebraic form, and multiplied, divided and raised to a power in trigonometric form. Choosing the right form BEFORE starting the calculation is what separates a question settled in three lines from a two-page expansion. Nobody computes (1+i3)10(1+i\sqrt{3})^{10} with the binomial theorem.

17 corrected exercises • 170 points • 225 minutes

This chapter is part of 201-SN4-RE, every chapter with its set and sheet Before you start Revision sheet: the traps and the method for this chapter

Self-checking set Type your answers and check them question by question: the page tells you right or wrong before you open the solution. With an account, every answer right on the first try earns points, and your right answers stay filled in when you come back.

Course recap

  • • Algebraic form: z=a+biz=a+bi with a=Re(z)a=\text{Re}(z), b=Im(z)b=\text{Im}(z) and i2=−1i^{2}=-1. Conjugate: zˉ=a−bi\bar{z}=a-bi. Modulus: ∣z∣=a2+b2|z|=\sqrt{a^{2}+b^{2}}, and the key relation zzˉ=a2+b2=∣z∣2z\bar{z}=a^{2}+b^{2}=|z|^{2}, which is a REAL number.
  • • Division: multiply the numerator and the denominator by the conjugate of the denominator. z1z2=z1z2ˉz2z2ˉ=z1z2ˉ∣z2∣2\dfrac{z_{1}}{z_{2}}=\dfrac{z_{1}\bar{z_{2}}}{z_{2}\bar{z_{2}}}=\dfrac{z_{1}\bar{z_{2}}}{|z_{2}|^{2}}, which makes the denominator real.
  • • Trigonometric (polar) form: z=r(cos⁡θ+isin⁡θ)z=r(\cos\theta+i\sin\theta), with r=∣z∣r=|z| and θ=arg⁡(z)\theta=\arg(z). Exponential form: z=r eiθz=r\,e^{i\theta}. To find θ\theta, solve cos⁡θ=ar\cos\theta=\dfrac{a}{r} AND sin⁡θ=br\sin\theta=\dfrac{b}{r}: both equations together, never just one, because the pair of signs fixes the quadrant.
  • • Product and quotient in exponential form: r1eiθ1×r2eiθ2=r1r2 ei(θ1+θ2)r_{1}e^{i\theta_{1}}\times r_{2}e^{i\theta_{2}}=r_{1}r_{2}\,e^{i(\theta_{1}+\theta_{2})} and r1eiθ1r2eiθ2=r1r2ei(θ1−θ2)\dfrac{r_{1}e^{i\theta_{1}}}{r_{2}e^{i\theta_{2}}}=\dfrac{r_{1}}{r_{2}}e^{i(\theta_{1}-\theta_{2})}. The moduli multiply or divide, the arguments add or subtract.
  • • De Moivre's theorem: (r(cos⁡θ+isin⁡θ))n=rn(cos⁡(nθ)+isin⁡(nθ))\big(r(\cos\theta+i\sin\theta)\big)^{n}=r^{n}\big(\cos(n\theta)+i\sin(n\theta)\big), or in exponential form (reiθ)n=rneinθ\big(re^{i\theta}\big)^{n}=r^{n}e^{in\theta}.
  • • The nnth roots of z=reiθz=re^{i\theta}: there are exactly nn of them, given by zk=r1/n ei(θ+2kπn)z_{k}=r^{1/n}\,e^{i\left(\frac{\theta+2k\pi}{n}\right)} for k=0,1,…,n−1k=0, 1, \dots, n-1. They lie on the circle of radius r1/nr^{1/n}, evenly spaced by an angle 2πn\dfrac{2\pi}{n}, and therefore form a regular polygon with nn sides.
  • • Conjugate roots: if a polynomial has REAL coefficients and has the root z0z_{0}, then z0ˉ\bar{z_{0}} is also a root. Non-real roots always come in conjugate pairs.

Calculations: working in algebraic form

Exercise 0.1: Computing in algebraic form: ten direct calculations

For each line, carry out the calculation and write the result in the algebraic form a+bia+bi, then give what is asked: most often the real part aa and the imaginary part bb, which is a real number.

Compute as with polynomials in ii, then replace i2i^{2} by −1-1 and collect the real parts and the imaginary parts. To divide, multiply the numerator and the denominator by the conjugate of the denominator, which becomes real: (a+bi)(a−bi)=a2+b2(a+bi)(a-bi)=a^{2}+b^{2}. The modulus is ∣a+bi∣=a2+b2|a+bi|=\sqrt{a^{2}+b^{2}}, and a power of ii reduces to the remainder of the exponent on division by 44.

  • a) (5−3i)+(−1+7i)(5-3i)+(-1+7i), then (5−3i)−(−1+7i)(5-3i)-(-1+7i): give the imaginary part of each.
  • b) (2−5i)(3+4i)(2-5i)(3+4i).
  • c) (4−3i)2(4-3i)^{2}.
  • d) i58+i103i^{58}+i^{103}.
  • e) ∣−8+15i∣|-8+15i| and ∣6−2i∣|6-2i|.
  • f) (2+3i)(1−2i)‾\overline{(2+3i)(1-2i)}.
  • g) 11−3i\frac{1}{1-3i}.
  • h) 7+i1−2i\frac{7+i}{1-2i}.
  • i) (1+2i)2 (1−i)‾+5i1+2i(1+2i)^{2}\,\overline{(1-i)}+\frac{5i}{1+2i}.
  • j) The solutions of 2z2−2z+5=02z^{2}-2z+5=0 in C\mathbb{C}: give the real part and the imaginary part of the one whose imaginary part is positive.

Type your answers, the page tells you right or wrong 0/20

a)
b)
c)
d)
e)
f)
g)
h)
i)
j)
Show the solution

Answers

  • a) 4+4i4+4i and 6−10i6-10i; 44 and −10-10
  • b) 26−7i26-7i
  • c) 7−24i7-24i
  • d) −1−i-1-i
  • e) 1717; 210≈6.3252\sqrt{10}\approx 6.325
  • f) 8−i‾=8+i\overline{8-i}=8+i
  • g) 110+310i\frac{1}{10}+\frac{3}{10}i
  • h) 1+3i1+3i
  • i) −5+2i-5+2i
  • j) z=12±32iz=\frac{1}{2}\pm\frac{3}{2}i; 12\frac{1}{2} and 32\frac{3}{2}

The move that decides everything: replace i2i^{2} by −1-1 as soon as it appears, then collect the real part and the imaginary part. For a division, you never divide by a complex number: multiply the top and the bottom by the conjugate of the denominator, which becomes real.

a)

(5−3i)+(−1+7i)=4+4i(5-3i)+(-1+7i)=4+4i and (5−3i)−(−1+7i)=5−3i+1−7i=6−10i(5-3i)-(-1+7i)=5-3i+1-7i=6-10i: imaginary parts 44 and −10-10. Pitfall: distributing the minus sign over the −1-1 only, which gives 6+4i6+4i.

b)

(2−5i)(3+4i)=6+8i−15i−20i2=6+20−7i=26−7i(2-5i)(3+4i)=6+8i-15i-20i^{2}=6+20-7i=26-7i. Pitfall: reading −20i2-20i^{2} as −20-20, hence −14−7i-14-7i.

c)

(4−3i)2=16−24i+9i2=16−9−24i=7−24i(4-3i)^{2}=16-24i+9i^{2}=16-9-24i=7-24i. Pitfall: 16+9=2516+9=25, forgetting that i2=−1i^{2}=-1.

d)

Divide the exponent by 44: 58=4×14+258=4\times 14+2, so i58=i2=−1i^{58}=i^{2}=-1; 103=4×25+3103=4\times 25+3, so i103=i3=−ii^{103}=i^{3}=-i. The sum is −1−i-1-i.

e)

∣−8+15i∣=64+225=289=17|-8+15i|=\sqrt{64+225}=\sqrt{289}=17 and ∣6−2i∣=36+4=40=210≈6.325|6-2i|=\sqrt{36+4}=\sqrt{40}=2\sqrt{10}\approx 6.325. Pitfall: 36−4\sqrt{36-4}, squaring −2i-2i instead of −2-2; the modulus uses real numbers only.

f)

The product first: (2+3i)(1−2i)=2−4i+3i−6i2=8−i(2+3i)(1-2i)=2-4i+3i-6i^{2}=8-i, whose conjugate is 8+i8+i. The same result comes from 2+3i‾⋅1−2i‾=(2−3i)(1+2i)=2+4i−3i−6i2=8+i\overline{2+3i}\cdot\overline{1-2i}=(2-3i)(1+2i)=2+4i-3i-6i^{2}=8+i: the conjugate of a product is the product of the conjugates.

g)

11−3i=1+3i(1−3i)(1+3i)=1+3i1+9=110+310i\frac{1}{1-3i}=\frac{1+3i}{(1-3i)(1+3i)}=\frac{1+3i}{1+9}=\frac{1}{10}+\frac{3}{10}i. Pitfall: writing (1−3i)(1+3i)=1−9(1-3i)(1+3i)=1-9, when in fact −(3i)2=+9-(3i)^{2}=+9.

h)

7+i1−2i=(7+i)(1+2i)(1−2i)(1+2i)=7+14i+i+2i21+4=5+15i5=1+3i\frac{7+i}{1-2i}=\frac{(7+i)(1+2i)}{(1-2i)(1+2i)}=\frac{7+14i+i+2i^{2}}{1+4}=\frac{5+15i}{5}=1+3i.

i)

(1+2i)2=1+4i+4i2=−3+4i(1+2i)^{2}=1+4i+4i^{2}=-3+4i and 1−i‾=1+i\overline{1-i}=1+i, so (−3+4i)(1+i)=−3−3i+4i+4i2=−7+i(-3+4i)(1+i)=-3-3i+4i+4i^{2}=-7+i. Then 5i1+2i=5i(1−2i)1+4=i−2i2=2+i\frac{5i}{1+2i}=\frac{5i(1-2i)}{1+4}=i-2i^{2}=2+i. Total: −7+i+2+i=−5+2i-7+i+2+i=-5+2i.

j)

Δ=(−2)2−4×2×5=4−40=−36<0\Delta=(-2)^{2}-4\times 2\times 5=4-40=-36<0: two complex conjugate solutions, with Δ=6i\sqrt{\Delta}=6i. So z=2±6i4=12±32iz=\frac{2\pm 6i}{4}=\frac{1}{2}\pm\frac{3}{2}i; the one with positive imaginary part is 12+32i\frac{1}{2}+\frac{3}{2}i. Pitfall: dividing by 22 instead of 2a=42a=4.

Check: multiply back. In h), (1+3i)(1−2i)(1+3i)(1-2i) must give 7+i7+i again; in g), (1−3i)(110+310i)=1(1-3i)\left(\frac{1}{10}+\frac{3}{10}i\right)=1; in j), the sum of the two solutions is 1=−ba1=-\frac{b}{a} and their product is 14+94=52=ca\frac{1}{4}+\frac{9}{4}=\frac{5}{2}=\frac{c}{a}.

Tick the exercises you have done or want to review: a free account, no password, keeps your ticks and your right answers from one visit to the next, tells you which chapter to tackle next and lets you ask for the exercise you are missing. Create your space, an email is enough.

Calculations: exponential form and De Moivre's theorem

Exercise 0.2: Converting to exponential form and computing powers: nine direct calculations

For each line, compute what is asked. Arguments are principal arguments, in (−π,π](-\pi, \pi], and are entered as multiples of π\pi: the argument −3π4-\frac{3\pi}{4} is typed −3/4-3/4.

The modulus first, r=a2+b2r=\sqrt{a^{2}+b^{2}}, then the argument from BOTH conditions cos⁡θ=ar\cos\theta=\frac{a}{r} and sin⁡θ=br\sin\theta=\frac{b}{r}, after locating the point in its quadrant. In exponential form, the moduli multiply and the arguments add; De Moivre's theorem gives (reiθ)n=rneinθ\left(re^{i\theta}\right)^{n}=r^{n}e^{in\theta}, for negative nn as well. Bring the final argument back into (−π,π](-\pi, \pi] before returning to algebraic form.

  • a) z=−7z=-7 and w=−3iw=-3i: give arg⁡(z)\arg(z) and arg⁡(w)\arg(w).
  • b) z=33+3iz=3\sqrt{3}+3i: give ∣z∣|z| and arg⁡(z)\arg(z).
  • c) z=−2−2 iz=-\sqrt{2}-\sqrt{2}\,i: give ∣z∣|z| and arg⁡(z)\arg(z).
  • d) z=6ei5π6z=6e^{i\frac{5\pi}{6}}: write zz in algebraic form.
  • e) 2eiπ12×5eiπ42e^{i\frac{\pi}{12}}\times 5e^{i\frac{\pi}{4}}: write the product in exponential form, then in algebraic form; give its modulus and its imaginary part.
  • f) 4ei2π32e−iπ2\frac{4e^{i\frac{2\pi}{3}}}{2e^{-i\frac{\pi}{2}}}: write the quotient in exponential form; give its modulus and its principal argument.
  • g) (3−i)8(\sqrt{3}-i)^{8}, by De Moivre's theorem, in algebraic form.
  • h) (1−i)−6(1-i)^{-6}: give its principal argument and its imaginary part.
  • i) The cube root of −27-27 obtained for k=0k=0 in the formula wk=r3 eiθ+2kπ3w_{k}=\sqrt[3]{r}\,e^{i\frac{\theta+2k\pi}{3}}, with θ∈(−π,π]\theta\in(-\pi, \pi]: give its real part and its imaginary part.

Type your answers, the page tells you right or wrong 0/18

a)
b)
c)
d)
e)
f)
g)
h)
i)
Show the solution

Answers

  • a) arg⁡(−7)=π\arg(-7)=\pi; arg⁡(−3i)=−π2\arg(-3i)=-\frac{\pi}{2}
  • b) z=6eiπ6z=6e^{i\frac{\pi}{6}}; 66 and π6\frac{\pi}{6}
  • c) z=2e−i3π4z=2e^{-i\frac{3\pi}{4}}; 22 and −3π4-\frac{3\pi}{4}
  • d) z=−33+3i≈−5.196+3iz=-3\sqrt{3}+3i\approx-5.196+3i
  • e) 10eiπ3=5+53 i10e^{i\frac{\pi}{3}}=5+5\sqrt{3}\,i; 1010 and 53≈8.6605\sqrt{3}\approx 8.660
  • f) 2e−i5π62e^{-i\frac{5\pi}{6}}; 22 and −5π6-\frac{5\pi}{6}
  • g) 256ei2π3=−128+1283 i256e^{i\frac{2\pi}{3}}=-128+128\sqrt{3}\,i; −128-128 and ≈221.703\approx 221.703
  • h) 18e−iπ2=−18i\frac{1}{8}e^{-i\frac{\pi}{2}}=-\frac{1}{8}i; −π2-\frac{\pi}{2} and −0.125-0.125
  • i) w0=3eiπ3=32+332iw_{0}=3e^{i\frac{\pi}{3}}=\frac{3}{2}+\frac{3\sqrt{3}}{2}i; 1.51.5 and ≈2.598\approx 2.598

The move that decides everything: place the point in the plane BEFORE looking for the argument. The quadrant fixes the sign and the value of the angle, which the arctangent alone cannot tell apart; after that, in exponential form, everything comes down to multiplying moduli and adding arguments.

a)

−7-7 lies on the negative real half-axis: ∣−7∣=7|-7|=7 and arg⁡(−7)=π\arg(-7)=\pi. −3i-3i lies on the negative imaginary half-axis: ∣−3i∣=3|-3i|=3 and arg⁡(−3i)=−π2\arg(-3i)=-\frac{\pi}{2}. Pitfall: giving arg⁡(−7)=0\arg(-7)=0 because the number is real; a negative real points to the left.

b)

∣z∣=27+9=36=6|z|=\sqrt{27+9}=\sqrt{36}=6, cos⁡θ=336=32\cos\theta=\frac{3\sqrt{3}}{6}=\frac{\sqrt{3}}{2} and sin⁡θ=36=12\sin\theta=\frac{3}{6}=\frac{1}{2}: first quadrant, θ=π6\theta=\frac{\pi}{6}. So z=6eiπ6z=6e^{i\frac{\pi}{6}}.

c)

∣z∣=2+2=2|z|=\sqrt{2+2}=2, cos⁡θ=−22\cos\theta=-\frac{\sqrt{2}}{2} and sin⁡θ=−22\sin\theta=-\frac{\sqrt{2}}{2}: third quadrant, θ=−3π4\theta=-\frac{3\pi}{4}, and z=2e−i3π4z=2e^{-i\frac{3\pi}{4}}. Pitfall: arctan⁡−2−2=arctan⁡1=π4\arctan\frac{-\sqrt{2}}{-\sqrt{2}}=\arctan 1=\frac{\pi}{4}, which is the argument of the OPPOSITE point.

d)

6ei5π6=6(cos⁡5π6+isin⁡5π6)=6(−32+12i)=−33+3i6e^{i\frac{5\pi}{6}}=6\left(\cos\frac{5\pi}{6}+i\sin\frac{5\pi}{6}\right)=6\left(-\frac{\sqrt{3}}{2}+\frac{1}{2}i\right)=-3\sqrt{3}+3i, that is a real part −33≈−5.196-3\sqrt{3}\approx-5.196 and an imaginary part 33. Pitfall: cos⁡5π6\cos\frac{5\pi}{6} is negative, the angle is in the second quadrant.

e)

The moduli multiply, the arguments add: 2×5=102\times 5=10 and π12+π4=π12+3π12=π3\frac{\pi}{12}+\frac{\pi}{4}=\frac{\pi}{12}+\frac{3\pi}{12}=\frac{\pi}{3}. So the product is 10eiπ3=5+53 i10e^{i\frac{\pi}{3}}=5+5\sqrt{3}\,i, with modulus 1010 and imaginary part 53≈8.6605\sqrt{3}\approx 8.660.

f)

The moduli divide, the arguments subtract: 42=2\frac{4}{2}=2 and 2π3−(−π2)=7π6\frac{2\pi}{3}-\left(-\frac{\pi}{2}\right)=\frac{7\pi}{6}. This angle exceeds π\pi: subtract 2π2\pi, which gives the principal argument 7π6−2π=−5π6\frac{7\pi}{6}-2\pi=-\frac{5\pi}{6}, and the quotient is 2e−i5π62e^{-i\frac{5\pi}{6}}. Pitfall: subtracting π2\frac{\pi}{2} instead of −π2-\frac{\pi}{2}, which gives π6\frac{\pi}{6}.

g)

3−i=2e−iπ6\sqrt{3}-i=2e^{-i\frac{\pi}{6}}, in the fourth quadrant. By De Moivre's theorem, (3−i)8=28e−i8π6=256e−i4π3=256ei2π3(\sqrt{3}-i)^{8}=2^{8}e^{-i\frac{8\pi}{6}}=256e^{-i\frac{4\pi}{3}}=256e^{i\frac{2\pi}{3}}, adding 2π2\pi. So (3−i)8=256(−12+32i)=−128+1283 i(\sqrt{3}-i)^{8}=256\left(-\frac{1}{2}+\frac{\sqrt{3}}{2}i\right)=-128+128\sqrt{3}\,i, with 1283≈221.703128\sqrt{3}\approx 221.703. Pitfall: forgetting to raise the modulus to the power 88, which gives −12+32i-\frac{1}{2}+\frac{\sqrt{3}}{2}i.

h)

1−i=2 e−iπ41-i=\sqrt{2}\,e^{-i\frac{\pi}{4}}, so (1−i)−6=(2)−6ei6π4=18ei3π2=18e−iπ2=−18i(1-i)^{-6}=(\sqrt{2})^{-6}e^{i\frac{6\pi}{4}}=\frac{1}{8}e^{i\frac{3\pi}{2}}=\frac{1}{8}e^{-i\frac{\pi}{2}}=-\frac{1}{8}i: principal argument −π2-\frac{\pi}{2}, imaginary part −18=−0.125-\frac{1}{8}=-0.125. Pitfall: the negative exponent ALSO changes the sign of the argument, −6×(−π4)=+3π2-6\times\left(-\frac{\pi}{4}\right)=+\frac{3\pi}{2}. Cross-check: (1−i)2=−2i(1-i)^{2}=-2i, so (1−i)6=(−2i)3=8i(1-i)^{6}=(-2i)^{3}=8i and 18i=−i8\frac{1}{8i}=-\frac{i}{8}.

i)

−27=27eiπ-27=27e^{i\pi}, so wk=3eiπ+2kπ3w_{k}=3e^{i\frac{\pi+2k\pi}{3}}. For k=0k=0: w0=3eiπ3=32+332iw_{0}=3e^{i\frac{\pi}{3}}=\frac{3}{2}+\frac{3\sqrt{3}}{2}i, that is a real part 1.51.5 and an imaginary part 332≈2.598\frac{3\sqrt{3}}{2}\approx 2.598. Pitfall: answering −3-3, the real root, which is w1w_{1}; the root for k=0k=0 is real only when the starting argument is zero.

Check: an exponential form is cross-checked by returning to algebraic form, reiθ=rcos⁡θ+irsin⁡θre^{i\theta}=r\cos\theta+ir\sin\theta, and a power or a root is checked by multiplying back: here w03=27eiπ=−27w_{0}^{3}=27e^{i\pi}=-27, and in g) the modulus ∣−128+1283 i∣=128×2=256=28|-128+128\sqrt{3}\,i|=128\times 2=256=2^{8}.

Part A: The basics (/50)

Exercise 1: Algebraic form, conjugate and division

Consider the complex numbers z1=3+2iz_{1}=3+2i and z2=1−4iz_{2}=1-4i.

-2-112345-5-4-3-2-1123z₁z₂
  • a) Compute z1+z2z_{1}+z_{2} and z1z2z_{1}z_{2} in algebraic form.
  • b) Give z1ˉ\bar{z_{1}} and z2ˉ\bar{z_{2}}, then compute ∣z1∣|z_{1}| and ∣z2∣|z_{2}|.
  • c) Write z1z2\dfrac{z_{1}}{z_{2}} in algebraic form.
  • d) Compute z1z1ˉz_{1}\bar{z_{1}} and comment on the result. Then find i2027i^{2027}.

Type your answers, the page tells you right or wrong 0/10

a)
b)
c)
d)
Show the solution

Answers

  • a) 4−2i4-2i; 11−10i11-10i
  • b) 3−2i3-2i, 1+4i1+4i; 13\sqrt{13}, 17\sqrt{17}
  • c) −517+1417i-\frac{5}{17}+\frac{14}{17}i
  • d) z1z1ˉ=13z_1\bar{z_1}=13; i2027=−ii^{2027}=-i
a)

z1+z2=(3+1)+(2−4)i=4−2iz_{1}+z_{2}=(3+1)+(2-4)i=4-2i. For the product, expand and then use i2=−1i^{2}=-1: z1z2=(3+2i)(1−4i)=3−12i+2i−8i2=3−10i+8=11−10iz_{1}z_{2}=(3+2i)(1-4i)=3-12i+2i-8i^{2}=3-10i+8=11-10i. The term −8i2-8i^{2} becomes +8+8, and that sign change is exactly what gets forgotten under pressure.

b)

z1ˉ=3−2i\bar{z_{1}}=3-2i and z2ˉ=1+4i\bar{z_{2}}=1+4i. The moduli: ∣z1∣=32+22=13|z_{1}|=\sqrt{3^{2}+2^{2}}=\sqrt{13} and ∣z2∣=12+(−4)2=17|z_{2}|=\sqrt{1^{2}+(-4)^{2}}=\sqrt{17}. The modulus uses the squares of the components, so the sign of the −4-4 disappears.

c)

Multiply the top and the bottom by the conjugate of the denominator: z1z2=(3+2i)(1+4i)(1−4i)(1+4i)\dfrac{z_{1}}{z_{2}}=\dfrac{(3+2i)(1+4i)}{(1-4i)(1+4i)}. Numerator: 3+12i+2i+8i2=3+14i−8=−5+14i3+12i+2i+8i^{2}=3+14i-8=-5+14i. Denominator: 12+42=171^{2}+4^{2}=17, a real number, which is the whole point of the operation. So z1z2=−5+14i17=−517+1417i\dfrac{z_{1}}{z_{2}}=\dfrac{-5+14i}{17}=-\dfrac{5}{17}+\dfrac{14}{17}i.

Check with the moduli: ∣z1z2∣=1317≈0.874\left|\dfrac{z_{1}}{z_{2}}\right|=\dfrac{\sqrt{13}}{\sqrt{17}}\approx 0.874, and directly 25+19617=22117≈0.874\dfrac{\sqrt{25+196}}{17}=\dfrac{\sqrt{221}}{17}\approx 0.874. Consistent.

d)

z1z1ˉ=(3+2i)(3−2i)=9−6i+6i−4i2=9+4=13z_{1}\bar{z_{1}}=(3+2i)(3-2i)=9-6i+6i-4i^{2}=9+4=13. The result is a REAL number, equal to ∣z1∣2=(13)2=13|z_{1}|^{2}=(\sqrt{13})^{2}=13. This is the property zzˉ=∣z∣2z\bar{z}=|z|^{2}, and it is what makes the division in part c) work.

For i2027i^{2027}: the powers of ii repeat with period 44, since i1=ii^{1}=i, i2=−1i^{2}=-1, i3=−ii^{3}=-i, i4=1i^{4}=1, and then the cycle starts again. So divide the exponent by 44: 2027=4×506+32027=4\times 506+3, the remainder is 33. Therefore i2027=i3=−ii^{2027}=i^{3}=-i. No calculator is needed, only the remainder on division by 44 matters.

-2-112345-5-4-3-2-1123z₁z₂z₁|z₁| = √13|z₂| = √17

Exercise 2: Modulus, argument and exponential form

Consider the complex number z=−2+2i3z=-2+2i\sqrt{3}.

-5-4-3-2-1123-112345z = −2 + 2i√3
  • a) Compute ∣z∣|z|, then find arg⁡(z)\arg(z) in radians.
  • b) Write zz in trigonometric form and in exponential form.
  • c) Give zˉ\bar{z} and −z-z in exponential form, without going back to the algebraic form.
  • d) Let w=1+iw=1+i. Write ww in exponential form, then compute zwzw in two ways, in exponential form and in algebraic form, and check that the moduli agree.

Type your answers, the page tells you right or wrong 0/6

a)
c)
d)
Show the solution

Answers

  • a) ∣z∣=4|z|=4, arg⁡z=2π3\arg z=\frac{2\pi}{3}
  • b) 4ei2π34e^{i\frac{2\pi}{3}}
  • c) zˉ=4e−i2π3\bar z=4e^{-i\frac{2\pi}{3}}, −z=4e−iπ3-z=4e^{-i\frac\pi3}
  • d) zw=42 ei11π12zw=4\sqrt2\,e^{i\frac{11\pi}{12}}
a)

∣z∣=(−2)2+(23)2=4+12=16=4|z|=\sqrt{(-2)^{2}+(2\sqrt{3})^{2}}=\sqrt{4+12}=\sqrt{16}=4. For the argument, impose BOTH conditions: cos⁡θ=−24=−12\cos\theta=\dfrac{-2}{4}=-\dfrac{1}{2} and sin⁡θ=234=32\sin\theta=\dfrac{2\sqrt{3}}{4}=\dfrac{\sqrt{3}}{2}. A negative cosine and a positive sine put the angle in the second quadrant, and the reference angle for 12\dfrac{1}{2} and 32\dfrac{\sqrt{3}}{2} is π3\dfrac{\pi}{3}. So θ=π−π3=2π3\theta=\pi-\dfrac{\pi}{3}=\dfrac{2\pi}{3}.

The classic pitfall: using the calculator for arctan⁡(23−2)=arctan⁡(−3)=−π3\arctan\left(\dfrac{2\sqrt{3}}{-2}\right)=\arctan(-\sqrt{3})=-\dfrac{\pi}{3}, and stopping there. That value is in the fourth quadrant, while zz is in the second. The arctangent function only ever returns angles between −π2-\dfrac{\pi}{2} and π2\dfrac{\pi}{2}: it cannot tell opposite quadrants apart. Always locate the point before concluding.

b)

Trigonometric form: z=4(cos⁡2π3+isin⁡2π3)z=4\left(\cos\dfrac{2\pi}{3}+i\sin\dfrac{2\pi}{3}\right). Exponential form: z=4ei2π3z=4e^{i\frac{2\pi}{3}}.

c)

The conjugate has the same modulus and the opposite argument: zˉ=4e−i2π3\bar{z}=4e^{-i\frac{2\pi}{3}}. Geometrically, it is the reflection in the real axis.

For −z-z, multiplying by −1-1 means multiplying by eiπe^{i\pi}, that is adding π\pi to the argument: −z=4ei(2π3+π)=4ei5π3-z=4e^{i\left(\frac{2\pi}{3}+\pi\right)}=4e^{i\frac{5\pi}{3}}, which is more conveniently written 4e−iπ34e^{-i\frac{\pi}{3}} after subtracting 2π2\pi. Geometrically, it is the point reflection through the origin, a rotation by a half-turn.

d)

∣w∣=1+1=2|w|=\sqrt{1+1}=\sqrt{2}, and the point (1,1)(1, 1) lies on the bisector of the first quadrant, so arg⁡(w)=π4\arg(w)=\dfrac{\pi}{4} and w=2 eiπ4w=\sqrt{2}\,e^{i\frac{\pi}{4}}.

In exponential form: zw=42 ei(2π3+π4)zw=4\sqrt{2}\,e^{i\left(\frac{2\pi}{3}+\frac{\pi}{4}\right)}. With a common denominator: 2π3+π4=8π12+3π12=11π12\dfrac{2\pi}{3}+\dfrac{\pi}{4}=\dfrac{8\pi}{12}+\dfrac{3\pi}{12}=\dfrac{11\pi}{12}. So zw=42 ei11π12zw=4\sqrt{2}\,e^{i\frac{11\pi}{12}}, in three lines.

In algebraic form: zw=(−2+2i3)(1+i)=−2−2i+2i3+2i23=−2−2i+23 i−23=(−2−23)+(23−2)izw=(-2+2i\sqrt{3})(1+i)=-2-2i+2i\sqrt{3}+2i^{2}\sqrt{3}=-2-2i+2\sqrt{3}\,i-2\sqrt{3}=(-2-2\sqrt{3})+(2\sqrt{3}-2)i.

Checking the moduli: ∣zw∣=(2+23)2+(23−2)2=(4+83+12)+(12−83+4)=32=42|zw|=\sqrt{(2+2\sqrt{3})^{2}+(2\sqrt{3}-2)^{2}}=\sqrt{(4+8\sqrt{3}+12)+(12-8\sqrt{3}+4)}=\sqrt{32}=4\sqrt{2}. The terms in 3\sqrt{3} cancel exactly, and we do get ∣z∣∣w∣=4×2|z||w|=4\times\sqrt{2}. Both forms describe the same number, but the exponential form gave the result without a single expansion.

-5-4-3-2-1123-112345z = −2 + 2i√32π/3|z| = 4

Exercise 3: De Moivre's theorem and powers

Consider the complex number z=1+i3z=1+i\sqrt{3}.

-2-1123-1123z = 1 + i√3
  • a) Write zz in exponential form.
  • b) Compute z6z^{6} using De Moivre's theorem.
  • c) Compute z10z^{10} and give the result in algebraic form.
  • d) Compute z−3z^{-3} in algebraic form, then check that z3×z−3=1z^{3}\times z^{-3}=1.

Type your answers, the page tells you right or wrong 0/6

a)
b)
c)
d)
Show the solution

Answers

  • a) z=2eiπ3z=2e^{i\frac\pi3}
  • b) z6=64z^6=64
  • c) z10=−512−5123 iz^{10}=-512-512\sqrt3\,i
  • d) z−3=−18z^{-3}=-\frac18
a)

∣z∣=12+(3)2=1+3=2|z|=\sqrt{1^{2}+(\sqrt{3})^{2}}=\sqrt{1+3}=2. Then cos⁡θ=12\cos\theta=\dfrac{1}{2} and sin⁡θ=32\sin\theta=\dfrac{\sqrt{3}}{2}, both positive, so first quadrant: θ=π3\theta=\dfrac{\pi}{3}. Thus z=2eiπ3z=2e^{i\frac{\pi}{3}}.

b)

By De Moivre's theorem: z6=26ei⋅6⋅π3=64 ei2πz^{6}=2^{6}e^{i\cdot 6\cdot\frac{\pi}{3}}=64\,e^{i2\pi}. But ei2π=cos⁡2π+isin⁡2π=1+0i=1e^{i2\pi}=\cos 2\pi+i\sin 2\pi=1+0i=1. So z6=64z^{6}=64, a positive real number. Compare with the work that expanding (1+i3)6(1+i\sqrt{3})^{6} with the binomial theorem would have taken: seven terms, with powers of 3\sqrt{3} and of ii to keep track of. The exponential form settles the question in one line.

c)

z10=210ei⋅10⋅π3=1024 ei10π3z^{10}=2^{10}e^{i\cdot 10\cdot\frac{\pi}{3}}=1024\,e^{i\frac{10\pi}{3}}. Bring the argument back into (−π,π](-\pi, \pi] by subtracting 4π=12π34\pi=\dfrac{12\pi}{3}: 10π3−12π3=−2π3\dfrac{10\pi}{3}-\dfrac{12\pi}{3}=-\dfrac{2\pi}{3}. So z10=1024 e−i2π3z^{10}=1024\,e^{-i\frac{2\pi}{3}}.

In algebraic form: cos⁡(−2π3)=−12\cos\left(-\dfrac{2\pi}{3}\right)=-\dfrac{1}{2} and sin⁡(−2π3)=−32\sin\left(-\dfrac{2\pi}{3}\right)=-\dfrac{\sqrt{3}}{2} (third quadrant, both negative). So z10=1024(−12−32i)=−512−5123 iz^{10}=1024\left(-\dfrac{1}{2}-\dfrac{\sqrt{3}}{2}i\right)=-512-512\sqrt{3}\,i.

d)

De Moivre's theorem also holds for negative exponents: z−3=2−3e−iπ=18×(−1)=−18z^{-3}=2^{-3}e^{-i\pi}=\frac{1}{8}\times(-1)=-\frac{1}{8}. Check: z3=8eiπ=−8z^{3}=8e^{i\pi}=-8, and (−8)×(−18)=1(-8)\times\left(-\frac{1}{8}\right)=1. Computing 1(1+i3)3\frac{1}{(1+i\sqrt{3})^{3}} directly would have taken an expansion and a division by the conjugate.

-9-8-7-6-5-4-3-2-1123-112345zz²z³ = −8per step: × 2, + 60°

Exercise 4: The nth roots of a complex number

We want to solve the equation z3=8iz^{3}=8i in C\mathbb{C}.

  • a) Write 8i8i in exponential form.
  • b) Find the three solutions of the equation in exponential form, then in algebraic form.
  • c) Check by direct calculation that 3+i\sqrt{3}+i is indeed a solution.
  • d) Plot the three solutions in the complex plane and describe the figure they form. What is their sum?

Type your answers, the page tells you right or wrong 0/6

a)
b)
c)
d)
Show the solution

Answers

  • a) 8i=8eiπ28i=8e^{i\frac\pi2}
  • b) 3+i\sqrt3+i, −3+i-\sqrt3+i, −2i-2i
  • c) (3+i)3=8i(\sqrt3+i)^3=8i
  • d) Equilateral triangle, zero sum
a)

The number 8i8i has real part 00 and imaginary part 88, so it lies on the positive imaginary half-axis. Its modulus is 88 and its argument π2\dfrac{\pi}{2}. So 8i=8eiπ28i=8e^{i\frac{\pi}{2}}.

b)

The cube roots have modulus 81/3=28^{1/3}=2 and arguments π2+2kπ3=π6+2kπ3\dfrac{\frac{\pi}{2}+2k\pi}{3}=\dfrac{\pi}{6}+\dfrac{2k\pi}{3} for k=0,1,2k=0, 1, 2.

For k=0k=0: z0=2eiπ6=2(32+12i)=3+iz_{0}=2e^{i\frac{\pi}{6}}=2\left(\dfrac{\sqrt{3}}{2}+\dfrac{1}{2}i\right)=\sqrt{3}+i.

For k=1k=1: the argument is π6+2π3=π6+4π6=5π6\dfrac{\pi}{6}+\dfrac{2\pi}{3}=\dfrac{\pi}{6}+\dfrac{4\pi}{6}=\dfrac{5\pi}{6}, so z1=2ei5π6=2(−32+12i)=−3+iz_{1}=2e^{i\frac{5\pi}{6}}=2\left(-\dfrac{\sqrt{3}}{2}+\dfrac{1}{2}i\right)=-\sqrt{3}+i.

For k=2k=2: the argument is π6+4π3=π6+8π6=9π6=3π2\dfrac{\pi}{6}+\dfrac{4\pi}{3}=\dfrac{\pi}{6}+\dfrac{8\pi}{6}=\dfrac{9\pi}{6}=\dfrac{3\pi}{2}, so z2=2ei3π2=2(0−i)=−2iz_{2}=2e^{i\frac{3\pi}{2}}=2(0-i)=-2i.

The mistake to avoid: stopping at the first root. An equation of degree 33 has exactly three solutions in C\mathbb{C}, and the parameter kk must run through 00, 11 and 22. Taking k=3k=3 would give z0z_{0} again, since it adds 2π2\pi to the argument.

c)

(3+i)2=3+23 i+i2=3+23 i−1=2+23 i(\sqrt{3}+i)^{2}=3+2\sqrt{3}\,i+i^{2}=3+2\sqrt{3}\,i-1=2+2\sqrt{3}\,i. Then (3+i)3=(2+23 i)(3+i)=23+2i+2⋅3 i+23 i2=23+2i+6i−23=8i(\sqrt{3}+i)^{3}=(2+2\sqrt{3}\,i)(\sqrt{3}+i)=2\sqrt{3}+2i+2\cdot 3\,i+2\sqrt{3}\,i^{2}=2\sqrt{3}+2i+6i-2\sqrt{3}=8i. The check is conclusive.

d)

The three roots all have the same modulus 22, so they lie on the circle of radius 22 centred at the origin. Their arguments are π6\dfrac{\pi}{6}, 5π6\dfrac{5\pi}{6} and 3π2\dfrac{3\pi}{2}, evenly spaced by 2π3\dfrac{2\pi}{3}, that is 120∘120^{\circ}. They are therefore the vertices of an EQUILATERAL TRIANGLE inscribed in that circle.

Their sum: (3+i)+(−3+i)+(−2i)=(3−3)+(1+1−2)i=0(\sqrt{3}+i)+(-\sqrt{3}+i)+(-2i)=(\sqrt{3}-\sqrt{3})+(1+1-2)i=0. The sum is zero. This is no coincidence: the nn nnth roots of a complex number are the vertices of a regular polygon centred at the origin, and the centroid of a regular polygon is its centre, so their sum is always zero as soon as n≥2n\geq 2. Another, algebraic reading: these roots are the solutions of z3−8i=0z^{3}-8i=0, a polynomial whose coefficient of z2z^{2} is zero, and by the relations between coefficients and roots, the sum of the roots equals that coefficient up to sign, so 00. This zero sum is an excellent way to check your roots at the end of a calculation.

-3-2-1123-3-2-1123√3 + i−√3 + i−2i

Exercise 5: Square roots of a complex number by the algebraic method

We look for the square roots of a complex number without going through the exponential form, by solving (x+iy)2=z(x+iy)^{2}=z directly. The method combines three real equations.

  • a) Find the square roots of z=3+4iz=3+4i: set (x+iy)2=3+4i(x+iy)^{2}=3+4i, write the system in xx and yy, adding the modulus equation x2+y2=∣z∣x^{2}+y^{2}=|z|, and solve it.
  • b) Find the square roots of z=−5+12iz=-5+12i in the same way.
  • c) Check each of your results by squaring, and explain why a non-zero complex number always has exactly two square roots, opposite to each other.
  • d) Application: solve the equation w2=3+4iw^{2}=3+4i in C\mathbb{C}, then explain how this method solves any quadratic equation with complex coefficients.

Type your answers, the page tells you right or wrong 0/6

a)
b)
c)
d)
Show the solution

Answers

  • a) ±(2+i)\pm(2+i)
  • b) ±(2+3i)\pm(2+3i)
  • c) Two opposite roots
  • d) w=±(2+i)w=\pm(2+i); complex discriminant
a)

Set (x+iy)2=x2−y2+2xyi=3+4i(x+iy)^{2}=x^{2}-y^{2}+2xyi=3+4i, which gives {x2−y2=32xy=4\begin{cases} x^{2}-y^{2}=3 \\ 2xy=4 \end{cases}. Add the modulus equation: ∣(x+iy)2∣=∣3+4i∣|(x+iy)^{2}|=|3+4i| gives x2+y2=32+42=5x^{2}+y^{2}=\sqrt{3^{2}+4^{2}}=5. Adding this to x2−y2=3x^{2}-y^{2}=3: 2x2=82x^{2}=8, so x2=4x^{2}=4 and x=±2x=\pm 2; subtracting: 2y2=22y^{2}=2, so y2=1y^{2}=1 and y=±1y=\pm 1. The condition 2xy=4>02xy=4>0 forces xx and yy to have the SAME sign. The square roots of 3+4i3+4i are therefore 2+i2+i and −2−i-2-i.

b)

Set (x+iy)2=−5+12i(x+iy)^{2}=-5+12i, so {x2−y2=−52xy=12\begin{cases} x^{2}-y^{2}=-5 \\ 2xy=12 \end{cases} and x2+y2=∣−5+12i∣=25+144=13x^{2}+y^{2}=|-5+12i|=\sqrt{25+144}=13. Adding this to x2−y2=−5x^{2}-y^{2}=-5: 2x2=82x^{2}=8, so x=±2x=\pm 2; subtracting: 2y2=182y^{2}=18, so y=±3y=\pm 3. The condition 2xy=12>02xy=12>0 forces the same sign. The square roots of −5+12i-5+12i are 2+3i2+3i and −2−3i-2-3i.

c)

Check: (2+i)2=4+4i+i2=4+4i−1=3+4i(2+i)^{2}=4+4i+i^{2}=4+4i-1=3+4i ✓, and (2+3i)2=4+12i+9i2=4+12i−9=−5+12i(2+3i)^{2}=4+12i+9i^{2}=4+12i-9=-5+12i ✓. The two roots of the same number are always opposite, because if r2=zr^{2}=z then (−r)2=r2=z(-r)^{2}=r^{2}=z as well, and a quadratic equation has no more than two solutions. A non-zero complex number therefore has exactly two square roots, symmetric about the origin. This is the difference with positive reals, where by convention only the positive root is kept; in C\mathbb{C}, neither of the two is preferred.

d)

The solutions of w2=3+4iw^{2}=3+4i are exactly the square roots found in a): w=2+iw=2+i or w=−2−iw=-2-i. The method is general: to solve aw2+bw+c=0aw^{2}+bw+c=0 with complex coefficients, compute the discriminant Δ=b2−4ac\Delta=b^{2}-4ac, which is a complex number, find a square root δ\delta of it by the method above, and the solutions are w=−b±δ2aw=\dfrac{-b\pm\delta}{2a}. The only new step compared with the real case is extracting a square root of a complex number, and that is exactly what this exercise teaches.

Part B: Exam level (/50)

Exercise 6: Polynomial equations and conjugate roots

This exercise is about solving polynomial equations in C\mathbb{C} and the property that governs their non-real roots.

  • a) Solve the equation z2−4z+13=0z^{2}-4z+13=0 in C\mathbb{C}.
  • b) Prove that if a polynomial with REAL coefficients has z0z_{0} as a root, then z0ˉ\bar{z_{0}} is also a root.
  • c) Let P(z)=z3−5z2+17z−13P(z)=z^{3}-5z^{2}+17z-13. Check that 11 is a root of PP, then factor PP and find all its roots in C\mathbb{C}.
  • d) Prove that ∣zw∣=∣z∣ ∣w∣|zw|=|z|\,|w| for all complex numbers zz and ww, then that z+w‾=zˉ+wˉ\overline{z+w}=\bar{z}+\bar{w}. Finally, explain with a counterexample why the result of b) would fail if the coefficients were not real.

Type your answers, the page tells you right or wrong 0/5

a)
c)
d)
Show the solution

Answers

  • a) z=2±3iz=2\pm 3i
  • b) Real coefficients: P(z0ˉ)=P(z0)‾=0P(\bar{z_0})=\overline{P(z_0)}=0
  • c) 11, 2+3i2+3i, 2−3i2-3i
  • d) Q(z)=z−iQ(z)=z-i: −i-i is not a root
a)

The discriminant is Δ=(−4)2−4(1)(13)=16−52=−36\Delta=(-4)^{2}-4(1)(13)=16-52=-36. It is negative, so there is no real solution, but there are two complex conjugate solutions. Write −36=6i\sqrt{-36}=6i, so z=4±6i2=2±3iz=\dfrac{4\pm 6i}{2}=2\pm 3i. The solutions are z=2+3iz=2+3i and z=2−3iz=2-3i.

Check on the first one: (2+3i)2=4+12i+9i2=4+12i−9=−5+12i(2+3i)^{2}=4+12i+9i^{2}=4+12i-9=-5+12i, then −5+12i−4(2+3i)+13=−5+12i−8−12i+13=0-5+12i-4(2+3i)+13=-5+12i-8-12i+13=0. It works.

b)

Let P(z)=anzn+⋯+a1z+a0P(z)=a_{n}z^{n}+\dots+a_{1}z+a_{0} with all the aka_{k} real, and suppose P(z0)=0P(z_{0})=0. Take the conjugate of both sides: P(z0)‾=0ˉ=0\overline{P(z_{0})}=\bar{0}=0.

Now conjugation respects sums and products, that is z+w‾=zˉ+wˉ\overline{z+w}=\bar{z}+\bar{w} and zw‾=zˉ wˉ\overline{zw}=\bar{z}\,\bar{w}. So P(z0)‾=an‾ z0‾ n+⋯+a1‾ z0‾+a0‾\overline{P(z_{0})}=\overline{a_{n}}\,\overline{z_{0}}^{\,n}+\dots+\overline{a_{1}}\,\overline{z_{0}}+\overline{a_{0}}. This is where the crucial hypothesis comes in: since the coefficients are REAL, they equal their own conjugates, ak‾=ak\overline{a_{k}}=a_{k}. The expression therefore becomes anz0‾ n+⋯+a1z0‾+a0=P(z0‾)a_{n}\overline{z_{0}}^{\,n}+\dots+a_{1}\overline{z_{0}}+a_{0}=P(\overline{z_{0}}).

Putting it together: P(z0‾)=0P(\overline{z_{0}})=0, so z0‾\overline{z_{0}} is indeed a root. This is exactly what we see in a), where the two solutions are conjugates of each other.

c)

P(1)=1−5+17−13=0P(1)=1-5+17-13=0, so 11 is a root and PP is divisible by (z−1)(z-1). Polynomial division gives P(z)=(z−1)(z2−4z+13)P(z)=(z-1)(z^{2}-4z+13), which we check by expanding: z3−4z2+13z−z2+4z−13=z3−5z2+17z−13z^{3}-4z^{2}+13z-z^{2}+4z-13=z^{3}-5z^{2}+17z-13, as required.

The quadratic factor is exactly the one from part a), whose roots are 2±3i2\pm 3i. The three roots of PP are therefore 11, 2+3i2+3i and 2−3i2-3i.

We see the mechanism of b) at work: PP has degree 33 and real coefficients, so it has at least one real root (here 11), and its non-real roots necessarily come as a conjugate pair. A polynomial of odd degree with real coefficients can never have all its roots non-real, since they pair up two by two and an odd number cannot be split into pairs.

d)

For the modulus, use the exponential form. If z=r1eiθ1z=r_{1}e^{i\theta_{1}} and w=r2eiθ2w=r_{2}e^{i\theta_{2}}, then zw=r1r2ei(θ1+θ2)zw=r_{1}r_{2}e^{i(\theta_{1}+\theta_{2})}, whose modulus is r1r2=∣z∣ ∣w∣r_{1}r_{2}=|z|\,|w| since the moduli are positive. In algebraic form, with z=a+biz=a+bi and w=c+diw=c+di, one can also check directly that ∣zw∣2=(ac−bd)2+(ad+bc)2=a2c2+b2d2+a2d2+b2c2=(a2+b2)(c2+d2)=∣z∣2∣w∣2|zw|^{2}=(ac-bd)^{2}+(ad+bc)^{2}=a^{2}c^{2}+b^{2}d^{2}+a^{2}d^{2}+b^{2}c^{2}=(a^{2}+b^{2})(c^{2}+d^{2})=|z|^{2}|w|^{2}, the cross terms cancelling in pairs.

For the conjugate of the sum, with z=a+biz=a+bi and w=c+diw=c+di: z+w=(a+c)+(b+d)iz+w=(a+c)+(b+d)i, so z+w‾=(a+c)−(b+d)i\overline{z+w}=(a+c)-(b+d)i. On the other hand zˉ+wˉ=(a−bi)+(c−di)=(a+c)−(b+d)i\bar{z}+\bar{w}=(a-bi)+(c-di)=(a+c)-(b+d)i. The two expressions agree.

Finally, the counterexample. Consider Q(z)=z−iQ(z)=z-i, whose constant coefficient −i-i is NOT real. Its only root is z0=iz_{0}=i. But z0‾=−i\overline{z_{0}}=-i gives Q(−i)=−i−i=−2i≠0Q(-i)=-i-i=-2i\neq 0: the conjugate is not a root. The property of b) collapses as soon as a single coefficient stops being real, because the step ak‾=ak\overline{a_{k}}=a_{k} is precisely the one carrying the whole proof. That is why the statement insists on real coefficients, and exams often penalize applying the property blindly to a polynomial with complex coefficients.

Exercise 7: Problem: the six sixth roots of 64 and the regular hexagon

We want to solve the equation z6=64z^{6}=64 in C\mathbb{C}, then use the geometric structure of its solutions.

The equation looks harmless: many students answer z=2z=2 and move on. Yet degree 66 forces six solutions, four of which are not real.

  • a) Write 6464 in exponential form, then find the six solutions of the equation in exponential form.
  • b) Give the six solutions in algebraic form.
  • c) Describe the figure formed by the six solutions in the complex plane. Which ones are real?
  • d) Compute the sum of the six solutions, and justify the result in two ways, one geometric and one algebraic.
  • e) Compute the product of the six solutions by grouping the conjugate roots.
  • f) Show that multiplication by the solution z1=1+i3z_{1}=1+i\sqrt{3} corresponds, in the plane, to a rotation combined with a dilation. Give the 2×22\times 2 matrix of this transformation and check that it agrees with the modulus and the argument of z1z_{1}.

Type your answers, the page tells you right or wrong 0/8

a)
b)
c)
d)
e)
f)
Show the solution

Answers

  • a) zk=2eikπ3z_k=2e^{i\frac{k\pi}{3}}
  • b) 22, 1±i31\pm i\sqrt3, −1±i3-1\pm i\sqrt3, −2-2
  • c) Regular hexagon; 22 and −2-2 are real
  • d) Zero sum
  • e) Product −64-64
  • f) M=2R60∘M=2R_{60^\circ}, det⁡M=4\det M=4
a)

The number 6464 is a positive real, so its modulus is 6464 and its argument 00: 64=64ei064=64e^{i0}. The sixth roots have modulus 641/6=264^{1/6}=2, since 26=642^{6}=64, and arguments 0+2kπ6=kπ3\dfrac{0+2k\pi}{6}=\dfrac{k\pi}{3} for k=0,1,2,3,4,5k=0, 1, 2, 3, 4, 5.

The six solutions are therefore zk=2eikπ3z_{k}=2e^{i\frac{k\pi}{3}}, namely 2ei02e^{i0}, 2eiπ32e^{i\frac{\pi}{3}}, 2ei2π32e^{i\frac{2\pi}{3}}, 2eiπ2e^{i\pi}, 2ei4π32e^{i\frac{4\pi}{3}} and 2ei5π32e^{i\frac{5\pi}{3}}.

b)

Convert each one using the special values of the unit circle. z0=2(1+0i)=2z_{0}=2(1+0i)=2. z1=2(12+32i)=1+i3z_{1}=2\left(\dfrac{1}{2}+\dfrac{\sqrt{3}}{2}i\right)=1+i\sqrt{3}. z2=2(−12+32i)=−1+i3z_{2}=2\left(-\dfrac{1}{2}+\dfrac{\sqrt{3}}{2}i\right)=-1+i\sqrt{3}. z3=2(−1+0i)=−2z_{3}=2(-1+0i)=-2. z4=2(−12−32i)=−1−i3z_{4}=2\left(-\dfrac{1}{2}-\dfrac{\sqrt{3}}{2}i\right)=-1-i\sqrt{3}. z5=2(12−32i)=1−i3z_{5}=2\left(\dfrac{1}{2}-\dfrac{\sqrt{3}}{2}i\right)=1-i\sqrt{3}.

Check on z1z_{1}: its modulus is 22 and its argument π3\dfrac{\pi}{3}, so z16=26ei2π=64z_{1}^{6}=2^{6}e^{i2\pi}=64, as required.

c)

The six solutions all have the same modulus 22, so they lie on the circle of radius 22 centred at the origin, and their arguments are spaced by π3\dfrac{\pi}{3}, that is 60∘60^{\circ}. They are therefore the vertices of a REGULAR HEXAGON inscribed in that circle.

Only two of them are real: z0=2z_{0}=2 and z3=−2z_{3}=-2, those whose argument is 00 or π\pi. The other four come in conjugate pairs, z1z_{1} with z5z_{5} and z2z_{2} with z4z_{4}, which is expected since the polynomial z6−64z^{6}-64 has real coefficients. The naive answer z=2z=2 gave only one solution out of six.

d)

Sum: 2+(1+i3)+(−1+i3)+(−2)+(−1−i3)+(1−i3)2+(1+i\sqrt{3})+(-1+i\sqrt{3})+(-2)+(-1-i\sqrt{3})+(1-i\sqrt{3}). The real parts give 2+1−1−2−1+1=02+1-1-2-1+1=0 and the imaginary parts 3+3−3−3=0\sqrt{3}+\sqrt{3}-\sqrt{3}-\sqrt{3}=0. The sum is 00.

Geometric justification: the six points are the vertices of a regular hexagon centred at the origin. The sum of vectors arranged with point symmetry cancels out, because each vertex has its opposite in the figure (z0z_{0} with z3z_{3}, z1z_{1} with z4z_{4}, z2z_{2} with z5z_{5}). Adding three pairs of opposite vectors term by term gives the zero vector.

Algebraic justification: the six solutions are the roots of z6−64=0z^{6}-64=0. This polynomial reads z6+0z5+0z4+0z3+0z2+0z−64z^{6}+0z^{5}+0z^{4}+0z^{3}+0z^{2}+0z-64, and by the relations between coefficients and roots, the sum of the roots is minus the coefficient of z5z^{5} divided by the leading coefficient, that is −01=0-\dfrac{0}{1}=0.

e)

Group each root with its conjugate, which makes each partial product real. The two real roots: z0z3=(2)(−2)=−4z_{0}z_{3}=(2)(-2)=-4. The pair z1z5=(1+i3)(1−i3)=12+(3)2=1+3=4z_{1}z_{5}=(1+i\sqrt{3})(1-i\sqrt{3})=1^{2}+(\sqrt{3})^{2}=1+3=4, using zzˉ=∣z∣2z\bar{z}=|z|^{2}. The pair z2z4=(−1+i3)(−1−i3)=1+3=4z_{2}z_{4}=(-1+i\sqrt{3})(-1-i\sqrt{3})=1+3=4 in the same way.

The total product is (−4)(4)(4)=−64(-4)(4)(4)=-64. Check with the relations between coefficients and roots: the product of the roots of a polynomial of degree 66 is (−1)6a0a6=−641=−64(-1)^{6}\dfrac{a_{0}}{a_{6}}=\dfrac{-64}{1}=-64. It agrees. Grouping conjugates is the right strategy: each pair immediately gives the square of the modulus, and you never handle a non-real number.

f)

Multiplying a complex number x+iyx+iy by z1=1+i3z_{1}=1+i\sqrt{3} gives (x+iy)(1+i3)=x+ix3+iy+i2y3=(x−3y)+i(3x+y)(x+iy)(1+i\sqrt{3})=x+ix\sqrt{3}+iy+i^{2}y\sqrt{3}=(x-\sqrt{3}y)+i(\sqrt{3}x+y). In terms of the coordinate pair (x,y)(x, y), the transformation sends (x,y)(x, y) to (x−3y,3x+y)(x-\sqrt{3}y, \sqrt{3}x+y), which corresponds to the matrix M=(1−331)M=\begin{pmatrix} 1 & -\sqrt{3} \\ \sqrt{3} & 1 \end{pmatrix}.

Factor out the modulus: M=2(12−323212)=2(cos⁡π3−sin⁡π3sin⁡π3cos⁡π3)M=2\begin{pmatrix} \dfrac{1}{2} & -\dfrac{\sqrt{3}}{2} \\[4pt] \dfrac{\sqrt{3}}{2} & \dfrac{1}{2} \end{pmatrix}=2\begin{pmatrix} \cos\dfrac{\pi}{3} & -\sin\dfrac{\pi}{3} \\[4pt] \sin\dfrac{\pi}{3} & \cos\dfrac{\pi}{3} \end{pmatrix}. It is therefore the dilation by a factor of 22 combined with the rotation by π3\dfrac{\pi}{3}, that is 60∘60^{\circ}.

The agreement is perfect: ∣z1∣=2|z_{1}|=2 gives the dilation factor and arg⁡(z1)=π3\arg(z_{1})=\dfrac{\pi}{3} gives the angle of the rotation. Check with the determinant: det⁡(M)=(1)(1)−(−3)(3)=1+3=4=∣z1∣2\det(M)=(1)(1)-(-\sqrt{3})(\sqrt{3})=1+3=4=|z_{1}|^{2}, which is indeed the area factor expected from a dilation by a factor of 22.

A last reading of the whole problem: applying this transformation six times amounts to a rotation by 6×60∘=360∘6\times 60^{\circ}=360^{\circ}, that is the identity, combined with a dilation by a factor of 26=642^{6}=64. This is exactly the statement z16=64z_{1}^{6}=64, translated into the language of transformations. The hexagon of roots is nothing but the orbit of the point (2,0)(2, 0) under successive rotations by 60∘60^{\circ}.

-4-3-2-11234-3-2-112360°21 + i√3−1 + i√3−2−1 − i√31 − i√3

Exercise 8: Complex multiplication as a rotation and the equilateral triangle criterion

Multiplying by a complex number of modulus rr and argument θ\theta amounts to combining a rotation by the angle θ\theta with a dilation by a factor of rr. We use this geometric reading to move points of the plane.

-112345-1123ABC4 + icentre 1 + i
  • a) Find the image of the point z=2z=2 under the rotation about the origin by 60∘60^{\circ}, by multiplying by the appropriate complex number.
  • b) Find the complex number z′z' representing the image of the point z=4+iz=4+i under the rotation by 90∘90^{\circ} (counterclockwise) about the centre c=1+ic=1+i. Recall that the formula is z′=c+eiθ(z−c)z'=c+e^{i\theta}(z-c).
  • c) Consider the points AA, BB, CC represented by the complex numbers a=0a=0, b=2b=2 and c=1+3 ic=1+\sqrt{3}\,i. Show that triangle ABCABC is equilateral, first by the lengths.
  • d) Check the algebraic criterion for an equilateral triangle, a2+b2+c2=ab+bc+caa^{2}+b^{2}+c^{2}=ab+bc+ca, on this triangle, and explain the link with the rotation by 60∘60^{\circ}.

Type your answers, the page tells you right or wrong 0/7

a)
b)
c)
d)
Show the solution

Answers

  • a) 1+3 i1+\sqrt3\,i
  • b) 1+4i1+4i
  • c) Three sides of length 2
  • d) 2+23 i2+2\sqrt3\,i on both sides
a)

A rotation by 60∘60^{\circ} about the origin is multiplication by ei60∘=cos⁡60∘+isin⁡60∘=12+32ie^{i60^{\circ}}=\cos 60^{\circ}+i\sin 60^{\circ}=\dfrac{1}{2}+\dfrac{\sqrt{3}}{2}i. The image of z=2z=2 is therefore 2(12+32i)=1+3 i2\left(\dfrac{1}{2}+\dfrac{\sqrt{3}}{2}i\right)=1+\sqrt{3}\,i, that is the point (1,3)(1, \sqrt{3}). Check: its modulus is 1+3=2\sqrt{1+3}=2, the same as that of zz, which is expected since a rotation preserves distances to the origin.

b)

Apply z′=c+ei90∘(z−c)z'=c+e^{i90^{\circ}}(z-c) with ei90∘=ie^{i90^{\circ}}=i. First compute z−c=(4+i)−(1+i)=3z-c=(4+i)-(1+i)=3. Then i⋅3=3ii\cdot 3=3i. Finally z′=c+3i=(1+i)+3i=1+4iz'=c+3i=(1+i)+3i=1+4i. The image is the point 1+4i1+4i, that is (1,4)(1, 4). Translation by −c-c, rotation, then translation by +c+c: this is the universal pattern of a rotation about a centre other than the origin.

c)

Compute the three lengths as moduli of differences of the complex numbers. AB=∣b−a∣=∣2∣=2AB=|b-a|=|2|=2. AC=∣c−a∣=∣1+3 i∣=1+3=2AC=|c-a|=|1+\sqrt{3}\,i|=\sqrt{1+3}=2. BC=∣c−b∣=∣(1+3 i)−2∣=∣−1+3 i∣=1+3=2BC=|c-b|=|(1+\sqrt{3}\,i)-2|=|-1+\sqrt{3}\,i|=\sqrt{1+3}=2. The three sides all equal 22, so triangle ABCABC is equilateral.

d)

Check a2+b2+c2=ab+bc+caa^{2}+b^{2}+c^{2}=ab+bc+ca. On the left: 0+4+(1+3 i)2=4+(1+23 i−3)=4+(−2+23 i)=2+23 i0+4+(1+\sqrt{3}\,i)^{2}=4+(1+2\sqrt{3}\,i-3)=4+(-2+2\sqrt{3}\,i)=2+2\sqrt{3}\,i. On the right: ab+bc+ca=0+2(1+3 i)+0=2+23 iab+bc+ca=0+2(1+\sqrt{3}\,i)+0=2+2\sqrt{3}\,i. The two sides agree, the criterion holds. The link with the rotation: a triangle ABCABC is equilateral and positively oriented if and only if CC is the image of BB under the rotation by 60∘60^{\circ} about AA, which reads c−a=ei60∘(b−a)c-a=e^{i60^{\circ}}(b-a). Expanding this relation, with ei60∘e^{i60^{\circ}} a root of t2−t+1=0t^{2}-t+1=0, gives precisely the symmetric identity a2+b2+c2=ab+bc+caa^{2}+b^{2}+c^{2}=ab+bc+ca. The geometry of the equilateral triangle is thus encoded in a single equation between the complex numbers aa, bb and cc.

Exercise 9: De Moivre's theorem and the triple-angle formulas

De Moivre's theorem is not only for computing powers: once expanded, it produces the multiple-angle trigonometric formulas. Here we apply it to the triple angle.

  • a) Write (cos⁡θ+isin⁡θ)3(\cos\theta+i\sin\theta)^{3} in two ways: by De Moivre's theorem, and by expanding the cube of a sum.
  • b) By identifying the real and imaginary parts, deduce the formulas for cos⁡3θ\cos 3\theta and sin⁡3θ\sin 3\theta in terms of cos⁡θ\cos\theta and sin⁡θ\sin\theta.
  • c) Express cos⁡3θ\cos 3\theta in terms of cos⁡θ\cos\theta only, then sin⁡3θ\sin 3\theta in terms of sin⁡θ\sin\theta only.
  • d) Check the formula for cos⁡3θ\cos 3\theta at θ=30∘\theta=30^{\circ}, where cos⁡3θ=cos⁡90∘=0\cos 3\theta=\cos 90^{\circ}=0.

Type your answers, the page tells you right or wrong 0/3

b)
c)
d)
Show the solution

Answers

  • a) cos⁡3θ+isin⁡3θ=(c3−3cs2)+i(3c2s−s3)\cos 3\theta+i\sin 3\theta=(c^3-3cs^2)+i(3c^2s-s^3)
  • b) cos⁡3θ=c3−3cs2\cos 3\theta=c^3-3cs^2, sin⁡3θ=3c2s−s3\sin 3\theta=3c^2s-s^3
  • c) 4cos⁡3θ−3cos⁡θ4\cos^3\theta-3\cos\theta; 3sin⁡θ−4sin⁡3θ3\sin\theta-4\sin^3\theta
  • d) 00 for θ=30∘\theta=30^\circ
a)

By De Moivre's theorem: (cos⁡θ+isin⁡θ)3=cos⁡3θ+isin⁡3θ(\cos\theta+i\sin\theta)^{3}=\cos 3\theta+i\sin 3\theta. By expanding the cube, with c=cos⁡θc=\cos\theta and s=sin⁡θs=\sin\theta: (c+is)3=c3+3c2(is)+3c(is)2+(is)3=c3+3ic2s−3cs2−is3(c+is)^{3}=c^{3}+3c^{2}(is)+3c(is)^{2}+(is)^{3}=c^{3}+3ic^{2}s-3cs^{2}-is^{3}, using i2=−1i^{2}=-1 and i3=−ii^{3}=-i. Collecting: (c+is)3=(c3−3cs2)+i(3c2s−s3)(c+is)^{3}=(c^{3}-3cs^{2})+i(3c^{2}s-s^{3}).

b)

Identify the two expressions. Real part: cos⁡3θ=cos⁡3θ−3cos⁡θsin⁡2θ\cos 3\theta=\cos^{3}\theta-3\cos\theta\sin^{2}\theta. Imaginary part: sin⁡3θ=3cos⁡2θsin⁡θ−sin⁡3θ\sin 3\theta=3\cos^{2}\theta\sin\theta-\sin^{3}\theta. This is the fundamental mechanism: a single application of De Moivre's theorem delivers both triple-angle formulas at once, where classical trigonometry needs the addition formula twice.

c)

For cos⁡3θ\cos 3\theta, replace sin⁡2θ=1−cos⁡2θ\sin^{2}\theta=1-\cos^{2}\theta: cos⁡3θ=cos⁡3θ−3cos⁡θ(1−cos⁡2θ)=cos⁡3θ−3cos⁡θ+3cos⁡3θ=4cos⁡3θ−3cos⁡θ\cos 3\theta=\cos^{3}\theta-3\cos\theta(1-\cos^{2}\theta)=\cos^{3}\theta-3\cos\theta+3\cos^{3}\theta=4\cos^{3}\theta-3\cos\theta. For sin⁡3θ\sin 3\theta, replace cos⁡2θ=1−sin⁡2θ\cos^{2}\theta=1-\sin^{2}\theta: sin⁡3θ=3(1−sin⁡2θ)sin⁡θ−sin⁡3θ=3sin⁡θ−3sin⁡3θ−sin⁡3θ=3sin⁡θ−4sin⁡3θ\sin 3\theta=3(1-\sin^{2}\theta)\sin\theta-\sin^{3}\theta=3\sin\theta-3\sin^{3}\theta-\sin^{3}\theta=3\sin\theta-4\sin^{3}\theta. These two polynomials, in cos⁡θ\cos\theta and in sin⁡θ\sin\theta, are special cases of the Chebyshev polynomials.

d)

For θ=30∘\theta=30^{\circ}, cos⁡θ=32\cos\theta=\dfrac{\sqrt{3}}{2}. The formula gives cos⁡90∘=4(32)3−3⋅32=4⋅338−332=332−332=0\cos 90^{\circ}=4\left(\dfrac{\sqrt{3}}{2}\right)^{3}-3\cdot\dfrac{\sqrt{3}}{2}=4\cdot\dfrac{3\sqrt{3}}{8}-\dfrac{3\sqrt{3}}{2}=\dfrac{3\sqrt{3}}{2}-\dfrac{3\sqrt{3}}{2}=0. We do get cos⁡90∘=0\cos 90^{\circ}=0, which validates the formula on a known value.

Exercise 10: Problem: loci in the complex plane

The expression ∣z−a∣|z-a| reads as the distance from the point zz to the point aa in the complex plane. This single reading turns complex equations into classic geometric figures: lines, circles, perpendicular bisectors.

  • a) Find and describe the set of points zz such that ∣z−1∣=∣z+i∣|z-1|=|z+i|. Give its Cartesian equation.
  • b) Find the set of points zz such that ∣z−(2+i)∣=3|z-(2+i)|=3. State its nature, its centre and its radius, and say whether the point 4+2i4+2i lies on it, inside it or outside it.
  • c) Find the set of points zz such that ∣z−1∣=2 ∣z+1∣|z-1|=2\,|z+1|. Show that it is a circle and give its centre and its radius.
  • d) Describe geometrically the set of points z=x+iyz=x+iy such that Re(z)+Im(z)=1\text{Re}(z)+\text{Im}(z)=1, and explain why, unlike b) and c), this locus is a line.

Type your answers, the page tells you right or wrong 0/7

a)
b)
c)
d)
Show the solution

Answers

  • a) Line y=−xy=-x
  • b) Circle with centre (2,1)(2, 1), radius 3; point inside
  • c) Circle with centre (−53,0)\left(-\frac53, 0\right), radius 43\frac43
  • d) Line x+y=1x+y=1
a)

Setting z=x+iyz=x+iy: ∣z−1∣2=(x−1)2+y2|z-1|^{2}=(x-1)^{2}+y^{2} and ∣z+i∣2=x2+(y+1)2|z+i|^{2}=x^{2}+(y+1)^{2}. The equality ∣z−1∣=∣z+i∣|z-1|=|z+i| is equivalent to (x−1)2+y2=x2+(y+1)2(x-1)^{2}+y^{2}=x^{2}+(y+1)^{2}, that is x2−2x+1+y2=x2+y2+2y+1x^{2}-2x+1+y^{2}=x^{2}+y^{2}+2y+1, hence −2x=2y-2x=2y and y=−xy=-x. The set is the LINE with equation y=−xy=-x. Interpretation: ∣z−1∣=∣z+i∣|z-1|=|z+i| means that zz is equidistant from the points 11 and −i-i, so it is the perpendicular bisector of the segment joining these two points.

b)

∣z−(2+i)∣=3|z-(2+i)|=3 means that the distance from zz to the point 2+i2+i is always 33: it is the CIRCLE with centre (2,1)(2, 1) and radius 33. For the point 4+2i4+2i, compute its distance to the centre: ∣(4+2i)−(2+i)∣=∣2+i∣=4+1=5≈2.24|(4+2i)-(2+i)|=|2+i|=\sqrt{4+1}=\sqrt{5}\approx 2.24. Since 5<3\sqrt{5}<3, this point is INSIDE the disk, it is not on the circle.

c)

Setting z=x+iyz=x+iy and squaring: (x−1)2+y2=4((x+1)2+y2)(x-1)^{2}+y^{2}=4\big((x+1)^{2}+y^{2}\big). Expand: x2−2x+1+y2=4x2+8x+4+4y2x^{2}-2x+1+y^{2}=4x^{2}+8x+4+4y^{2}, that is 0=3x2+3y2+10x+30=3x^{2}+3y^{2}+10x+3, hence x2+y2+103x+1=0x^{2}+y^{2}+\dfrac{10}{3}x+1=0. Complete the square: (x+53)2+y2=259−1=169\left(x+\dfrac{5}{3}\right)^{2}+y^{2}=\dfrac{25}{9}-1=\dfrac{16}{9}. It is the CIRCLE with centre (−53,0)\left(-\dfrac{5}{3}, 0\right) and radius 43\dfrac{4}{3}. This is a circle of Apollonius: the locus of points whose distances to two fixed points have a constant ratio different from 11 is always a circle.

d)

The condition Re(z)+Im(z)=1\text{Re}(z)+\text{Im}(z)=1 reads directly x+y=1x+y=1: it is a LINE. The difference with b) and c) lies in the nature of the equation. A condition on the real and imaginary PARTS of zz is linear in xx and yy, so it gives a line. A condition of EQUAL MODULI, as in a), produces squares that cancel and also leaves a linear equation (a perpendicular bisector). But a condition of CONSTANT modulus, as in b), or of a RATIO of moduli different from 11, as in c), keeps the terms in x2+y2x^{2}+y^{2} and produces a circle. Recognizing the type of equation lets you predict the figure before calculating anything.

-4-3-2-1123456-3-2-1123451−i2 + i4 + 2ia)b)c)d)

Part C: The classics (/50)

Exercise 11: The fifth roots of unity and the cosine of 72°

Let ω=ei2π5\omega=e^{i\frac{2\pi}{5}}.

  • a) Give the solutions of z5=1z^{5}=1 and describe the figure they form.
  • b) Show that 1+ω+ω2+ω3+ω4=01+\omega+\omega^{2}+\omega^{3}+\omega^{4}=0.
  • c) Let t=ω+ω4t=\omega+\omega^{4}. Show that t=2cos⁡2π5t=2\cos\frac{2\pi}{5}, that ω2+ω3=t2−2\omega^{2}+\omega^{3}=t^{2}-2, then that t2+t−1=0t^{2}+t-1=0.
  • d) Deduce the exact value of cos⁡72∘\cos 72^{\circ}, then its value to three decimal places.
  • e) Compute the side length of the regular pentagon inscribed in the unit circle, to three decimal places.

Type your answers, the page tells you right or wrong 0/6

a)
b)
c)
d)
e)
Show the solution

Answers

  • a) Regular pentagon
  • b) Geometric sum equal to zero
  • c) t2+t−1=0t^2+t-1=0
  • d) cos⁡72∘=5−14≈0.309\cos 72^\circ=\frac{\sqrt5-1}{4}\approx 0.309
  • e) Side ≈1.176\approx 1.176
a)

zk=ei2kπ5z_{k}=e^{i\frac{2k\pi}{5}} for k=0,1,2,3,4k=0, 1, 2, 3, 4, that is 11, ω\omega, ω2\omega^{2}, ω3\omega^{3}, ω4\omega^{4}. Five points of the unit circle spaced by 72∘72^{\circ}: a REGULAR PENTAGON.

b)

It is a geometric sum with ratio ω≠1\omega\neq 1: ω5−1ω−1=1−1ω−1=0\frac{\omega^{5}-1}{\omega-1}=\frac{1-1}{\omega-1}=0. Geometrically, the vertices of a regular polygon centred at the origin add up to zero.

c)

ω4=e−i2π5=ωˉ\omega^{4}=e^{-i\frac{2\pi}{5}}=\bar{\omega}, so t=ω+ωˉ=2cos⁡2π5t=\omega+\bar{\omega}=2\cos\frac{2\pi}{5}. Then t2=ω2+2ω5+ω8=ω2+2+ω3t^{2}=\omega^{2}+2\omega^{5}+\omega^{8}=\omega^{2}+2+\omega^{3}, since ω5=1\omega^{5}=1 and ω8=ω3\omega^{8}=\omega^{3}: ω2+ω3=t2−2\omega^{2}+\omega^{3}=t^{2}-2. By b), 1+t+(t2−2)=01+t+(t^{2}-2)=0, that is t2+t−1=0t^{2}+t-1=0.

d)

t=−1±52t=\frac{-1\pm\sqrt{5}}{2}, and t=2cos⁡72∘>0t=2\cos 72^{\circ}>0 forces t=5−12t=\frac{\sqrt{5}-1}{2}. So cos⁡72∘=5−14≈0.309\cos 72^{\circ}=\frac{\sqrt{5}-1}{4}\approx 0.309. The negative root, −1−52\frac{-1-\sqrt{5}}{2}, equals ω2+ω3=2cos⁡144∘\omega^{2}+\omega^{3}=2\cos 144^{\circ}.

e)

The side is ∣ω−1∣|\omega-1|, and ∣ω−1∣2=(cos⁡72∘−1)2+sin⁡272∘=2−2cos⁡72∘=2−5−12=5−52≈1.382|\omega-1|^{2}=(\cos 72^{\circ}-1)^{2}+\sin^{2}72^{\circ}=2-2\cos 72^{\circ}=2-\frac{\sqrt{5}-1}{2}=\frac{5-\sqrt{5}}{2}\approx 1.382. Side ≈1.176\approx 1.176.

The thread: roots of unity turn a trigonometry question into a polynomial equation; their zero sum provides the key relation.

Exercise 12: The factorization of z⁴ + 4

We want to solve z4+4=0z^{4}+4=0 in C\mathbb{C}, then derive a real factorization from it.

  • a) Write −4-4 in exponential form and find the four solutions in algebraic form.
  • b) By grouping the conjugate roots, write z4+4z^{4}+4 as a product of two quadratic polynomials with real coefficients.
  • c) Check the factorization by noticing that z4+4=(z2+2)2−(2z)2z^{4}+4=(z^{2}+2)^{2}-(2z)^{2}.
  • d) Deduce that n4+4n^{4}+4 is never a prime number for an integer n≥2n\geq 2. Factor 34+43^{4}+4 and 54+45^{4}+4.
  • e) Compute the sum and the product of the four solutions, and check them against the coefficients of z4+4z^{4}+4.

Type your answers, the page tells you right or wrong 0/7

a)
b)
d)
e)
Show the solution

Answers

  • a) ±1±i\pm1\pm i
  • b) (z2−2z+2)(z2+2z+2)(z^2-2z+2)(z^2+2z+2)
  • c) Difference of two squares
  • d) 85=5×1785=5\times 17; 629=17×37629=17\times 37
  • e) Sum 0, product 4
a)

−4=4eiπ-4=4e^{i\pi}. The fourth roots have modulus 41/4=24^{1/4}=\sqrt{2} and arguments π4+kπ2\frac{\pi}{4}+\frac{k\pi}{2}: 2 eiπ4=1+i\sqrt{2}\,e^{i\frac{\pi}{4}}=1+i, then −1+i-1+i, −1−i-1-i and 1−i1-i. They form a square.

b)

(z−1−i)(z−1+i)=(z−1)2+1=z2−2z+2(z-1-i)(z-1+i)=(z-1)^{2}+1=z^{2}-2z+2 and (z+1−i)(z+1+i)=(z+1)2+1=z2+2z+2(z+1-i)(z+1+i)=(z+1)^{2}+1=z^{2}+2z+2. So z4+4=(z2−2z+2)(z2+2z+2)z^{4}+4=(z^{2}-2z+2)(z^{2}+2z+2).

c)

(z2+2)2−(2z)2=z4+4z2+4−4z2=z4+4(z^{2}+2)^{2}-(2z)^{2}=z^{4}+4z^{2}+4-4z^{2}=z^{4}+4, and a difference of two squares factors as (z2+2−2z)(z2+2+2z)(z^{2}+2-2z)(z^{2}+2+2z): this is indeed the factorization of b).

d)

n4+4=(n2−2n+2)(n2+2n+2)n^{4}+4=(n^{2}-2n+2)(n^{2}+2n+2), and for n≥2n\geq 2 the first factor equals (n−1)2+1≥2(n-1)^{2}+1\geq 2, while the second is larger still: the number has two factors greater than 11, so it is not prime. 34+4=85=5×173^{4}+4=85=5\times 17 and 54+4=629=17×375^{4}+4=629=17\times 37.

e)

Sum: (1+i)+(−1+i)+(−1−i)+(1−i)=0(1+i)+(-1+i)+(-1-i)+(1-i)=0, and the coefficient of z3z^{3} is zero. Product: (1+i)(1−i)×(−1+i)(−1−i)=2×2=4(1+i)(1-i)\times(-1+i)(-1-i)=2\times 2=4, equal to the constant term, since the degree is even.

The thread: the complex roots of a real polynomial group into conjugate pairs that give real quadratic factors.

Exercise 13: A quadratic equation with complex coefficients

We solve the equation z2−(3+i)z+4+3i=0z^{2}-(3+i)z+4+3i=0 in C\mathbb{C}.

  • a) Compute the discriminant Δ\Delta.
  • b) Find the two square roots of Δ\Delta by the algebraic method.
  • c) Deduce the two solutions of the equation.
  • d) Check them using the sum and the product of the solutions.
  • e) Are the two solutions conjugate? Why does this not contradict the conjugate root property?

Type your answers, the page tells you right or wrong 0/8

a)
b)
c)
d)
e)
Show the solution

Answers

  • a) Δ=−8−6i\Delta=-8-6i
  • b) ±(1−3i)\pm(1-3i)
  • c) 2−i2-i and 1+2i1+2i
  • d) Sum 3+i3+i, product 4+3i4+3i
  • e) No: the coefficients are not real
a)

Δ=(3+i)2−4(4+3i)=9+6i−1−16−12i=−8−6i\Delta=(3+i)^{2}-4(4+3i)=9+6i-1-16-12i=-8-6i.

b)

Look for δ=x+iy\delta=x+iy with x2−y2=−8x^{2}-y^{2}=-8, 2xy=−62xy=-6 and x2+y2=∣Δ∣=64+36=10x^{2}+y^{2}=|\Delta|=\sqrt{64+36}=10. Hence x2=1x^{2}=1 and y2=9y^{2}=9, and xy<0xy<0 forces opposite signs: δ=1−3i\delta=1-3i or δ=−1+3i\delta=-1+3i.

c)

z=(3+i)±(1−3i)2z=\frac{(3+i)\pm(1-3i)}{2}: z1=4−2i2=2−iz_{1}=\frac{4-2i}{2}=2-i and z2=2+4i2=1+2iz_{2}=\frac{2+4i}{2}=1+2i.

d)

Sum: (2−i)+(1+2i)=3+i(2-i)+(1+2i)=3+i, the opposite of the coefficient of zz. Product: (2−i)(1+2i)=2+4i−i−2i2=4+3i(2-i)(1+2i)=2+4i-i-2i^{2}=4+3i, the constant term. Both relations hold.

e)

No: the conjugate of 2−i2-i is 2+i2+i, which is not a solution. There is no contradiction: the conjugate root property only holds for a polynomial with REAL coefficients, and here two coefficients are not real.

The thread: the quadratic formula remains valid in C\mathbb{C}; the only new step is extracting a complex square root.

Exercise 14: Problem: the iterations of the Mandelbrot set

For a complex number cc, define the sequence z0=0z_{0}=0 and zn+1=zn2+cz_{n+1}=z_{n}^{2}+c. The number cc belongs to the Mandelbrot set if this sequence stays bounded. Take as given that as soon as a term satisfies ∣zn∣>2|z_{n}|>2, the sequence is unbounded.

  • a) Compute the terms z1z_{1} to z4z_{4} for c=1c=1. Does the number 11 belong to the set?
  • b) Same question for c=−1c=-1.
  • c) Compute z1z_{1} to z4z_{4} for c=ic=i and conclude.
  • d) For c=−2c=-2, compute the first terms. For c=0.5c=0.5, find the first index nn such that ∣zn∣>2|z_{n}|>2.
  • e) For real cc strictly greater than 14\frac{1}{4}, show that zn+1−zn≥c−14z_{n+1}-z_{n}\geq c-\frac{1}{4}, and conclude.

Type your answers, the page tells you right or wrong 0/9

a)
b)
c)
d)
e)
Show the solution

Answers

  • a) 1,2,5,261, 2, 5, 26: no
  • b) −1,0,−1,0-1, 0, -1, 0: yes
  • c) ii, −1+i-1+i, −i-i, −1+i-1+i: yes
  • d) c=−2c=-2: bounded; c=0.5c=0.5: n=5n=5
  • e) Increase of at least c−14c-\frac14: divergence
a)

z1=1z_{1}=1, z2=2z_{2}=2, z3=5z_{3}=5, z4=26z_{4}=26. From z3z_{3} on, the modulus exceeds 22: the sequence is unbounded, 11 does not belong to the set.

b)

z1=−1z_{1}=-1, z2=0z_{2}=0, z3=−1z_{3}=-1, z4=0z_{4}=0: the sequence alternates between −1-1 and 00, it is bounded, and −1-1 belongs to the set.

c)

z1=iz_{1}=i; z2=i2+i=−1+iz_{2}=i^{2}+i=-1+i; z3=(−1+i)2+i=(1−2i−1)+i=−iz_{3}=(-1+i)^{2}+i=(1-2i-1)+i=-i; z4=(−i)2+i=−1+iz_{4}=(-i)^{2}+i=-1+i. The sequence repeats −1+i-1+i, −i-i: it is bounded, with modulus at most 2\sqrt{2}, and ii belongs to the set.

d)

For c=−2c=-2: z1=−2z_{1}=-2, z2=2z_{2}=2, then zn=2z_{n}=2 for every n≥2n\geq 2. The sequence is bounded, and −2-2 is the leftmost point of the set. For c=0.5c=0.5: z1=0.5z_{1}=0.5, z2=0.75z_{2}=0.75, z3=1.0625z_{3}=1.0625, z4≈1.629z_{4}\approx 1.629, z5≈3.153z_{5}\approx 3.153: the first index is n=5n=5.

e)

For real cc, every term is real, and zn+1−zn=zn2−zn+c=(zn−12)2+c−14≥c−14z_{n+1}-z_{n}=z_{n}^{2}-z_{n}+c=\left(z_{n}-\frac{1}{2}\right)^{2}+c-\frac{1}{4}\geq c-\frac{1}{4}. The sequence therefore increases by at least c−14>0c-\frac{1}{4}>0 at each step: zn≥n(c−14)z_{n}\geq n\left(c-\frac{1}{4}\right) tends to infinity. On the real axis, the set does not extend beyond 14\frac{1}{4} on the right.

The thread: iterating z↦z2+cz\mapsto z^{2}+c combines the product and the sum of complex numbers; the modulus decides divergence.

Exercise 15: Problem: adding sinusoidal signals with phasors

A signal acos⁡(ωt+φ)a\cos(\omega t+\varphi), with a>0a>0, is represented by the complex number a eiφa\,e^{i\varphi}, its PHASOR: the signal is the real part of a eiφeiωta\,e^{i\varphi}e^{i\omega t}. Two signals with the same angular frequency are added by adding their phasors.

  • a) Give the phasors of s1(t)=3cos⁡(ωt)s_{1}(t)=3\cos(\omega t) and of s2(t)=4cos⁡(ωt+π2)s_{2}(t)=4\cos\left(\omega t+\frac{\pi}{2}\right).
  • b) Compute the phasor of s1+s2s_{1}+s_{2} in exponential form: amplitude and phase, to the hundredth of a degree.
  • c) Write s1+s2s_{1}+s_{2} in the form Acos⁡(ωt+Φ)A\cos(\omega t+\Phi) and check the value at t=0t=0.
  • d) Three signals 2cos⁡(ωt)2\cos(\omega t), 2cos⁡(ωt+2π3)2\cos\left(\omega t+\frac{2\pi}{3}\right) and 2cos⁡(ωt+4π3)2\cos\left(\omega t+\frac{4\pi}{3}\right) are added. What is the amplitude of the sum?
  • e) Differentiating acos⁡(ωt+φ)a\cos(\omega t+\varphi) amounts to multiplying its phasor by iωi\omega. For ω=2 rad/s\omega=2\ \mathrm{rad/s}, give the phasor and the amplitude of the derivative of s1s_{1}.

Type your answers, the page tells you right or wrong 0/6

a)
b)
c)
d)
e)
Show the solution

Answers

  • a) 33 and 4i4i
  • b) 5ei 0.9275e^{i\,0.927}, that is 53.13∘53.13^\circ
  • c) 5cos⁡(ωt+0.927)5\cos(\omega t+0.927); 3 at t=0t=0
  • d) Zero sum
  • e) Phasor 6i6i, amplitude 6
a)

s1s_{1}: 3ei0=33e^{i0}=3. s2s_{2}: 4eiπ2=4i4e^{i\frac{\pi}{2}}=4i.

b)

3+4i3+4i, with modulus 55 and argument arctan⁡43≈0.927 rad\arctan\frac{4}{3}\approx 0.927\ \mathrm{rad}, that is 53.13∘53.13^{\circ}: 5ei 0.9275e^{i\,0.927}.

c)

s1(t)+s2(t)=5cos⁡(ωt+0.927)s_{1}(t)+s_{2}(t)=5\cos(\omega t+0.927). At t=0t=0: 3+4cos⁡π2=33+4\cos\frac{\pi}{2}=3, and 5cos⁡(0.927)=5×0.6=35\cos(0.927)=5\times 0.6=3. The amplitude of the sum, 55, is not 3+4=73+4=7: the two signals are shifted by a quarter of a period.

d)

The three phasors 22, 2ei2π32e^{i\frac{2\pi}{3}} and 2ei4π32e^{i\frac{4\pi}{3}} are the vertices of an equilateral triangle centred at the origin: their sum is zero, and the sum of the signals is identically zero. This is the principle of three-phase current, whose three phases balance out.

e)

The phasor of s1′s_{1}' is 2i×3=6i=6eiπ22i\times 3=6i=6e^{i\frac{\pi}{2}}: s1′(t)=6cos⁡(2t+π2)=−6sin⁡(2t)s_{1}'(t)=6\cos\left(2t+\frac{\pi}{2}\right)=-6\sin(2t), with amplitude 66. This is indeed the derivative of 3cos⁡(2t)3\cos(2t).

The thread: phasors replace the trigonometry of signals with the same frequency by the addition and multiplication of complex numbers.

-11234567-11234534i3 + 4i: amplitude 553.13°
Something missing from this set? A type of exercise your teacher gives, a problem you are stuck on, a mistake you spotted: tell me, I read every request.

Sending goes through your free space: that is how I can answer you and tell you when the exercise is added. An email is enough, no password. Create your space. Rather not create an account? Write to me directly.

What is it about?
I read every request myself, nobody else sees it. No need to write your name or your school.
When your request is dealt with, you hear about it by email and in your space.

By sending, you agree that I read your request to answer it. It is kept with your account, erased with it, and never published: privacy policy.

Previous chapter Matrix models: Markov, Leontief, Leslie, 201-SN4-RE Next chapter Applications in science, 201-SN4-RE

© Ahmed Squalli Houssaini. Exercise set published at www.letuteurscientifique.ca/en/exercices/cegep-sn4-complex-numbers. Free for personal and classroom use; republishing it elsewhere requires written permission (legal notice).

See also

Looking for a linear algebra tutor in Montreal?

Get in touch for a first session. We work on 201-SN4-RE at the real level of CEGEP tests, from the algebraic form to nth roots and polynomial equations, in English or in French, in Montreal or online.

Site by Studio Squalli