Let w, z be complex numbers. Show that |wz|=|w||z|, and using the fact that x=|x|e^{arg(x)i}, show further that arg(wz)=arg(w)+arg(z) where |.| is the absolute value and arg(.) is the angle (in polar coordinates). Hence, find all solutions to x^n=1 .

Put w=a+bi, z=c+di where a, b, c, d are real, and i2=-1. Then |wz|2=|(a+bi)(c+di)|2=|ac-bd+(ad+cb)i|2=(ac)2+(bd)2-2bcd+(ad)2+(cb)2+2abcd=(ac)2+(bd)2+(ad)2+(cb)2=(a2+b2)(c2+d2)=|w|2|z|2. Taking square roots, we get |wz|=|w||z|, noting that these values can never be negative. For the argument, wz=|w|e^{arg(w)i}|z|e^{arg(z)i}=|w||z|e^{(arg(w)+arg(z))i} so that arg(wz)=arg(w)+arg(z). 

Now, suppose xn=1. Write x=|x|e^{arg(x)i}, 1=e^{0i}, then we rewrite the equation as (|x|e^{arg(x)i})n=e^{0i}. Since we are multiplying x to itself n times, we use the result from the previous part to get that xn=|x|ne^{n arg(x)i}. By comparing magnitute and arguments, we see that |x|=1 and arg(x)n=0, or equivalently, arg(x)n=2kpi for some natural number k, since e^{xi} is periodic in x with period 2pi. Notice that 0<=arg(x)<2pi, so that 0<=k<=n. So then arg(x)=2pi k/n for k=0, 1, ..., n, which gives us the n+1 different solutions x=e^{2ipik/n} for k=0, 1, ..., n. However, notice that when k=0, and when k=n gives us the same solution, so infact we only get n distinct solutions x=e^{2ipik/n} for k=0, 1, ..., n-1.

JK
Answered by Jihoon K. Maths tutor

7578 Views

See similar Maths A Level tutors

Related Maths A Level answers

All answers ▸

Integrate 4x^3 - 3x + 6


Simplify: (log(40) - log(20)) + log(3)


What are the advantages of using numerical integration (Trapezium rule, Simpson's rule and so on) over direct integration?


What is dot product and how to calculate it?


We're here to help

contact us iconContact ustelephone icon+44 (0) 203 773 6020
Facebook logoInstagram logoLinkedIn logo

MyTutor is part of the IXL family of brands:

© 2025 by IXL Learning