I have few doubts regarding complex numbers. Help is as always , greatly appreciated. Nice solutions or hints are welcome.

**Q1)** If \(\alpha , \beta , \gamma\) are cube roots of \(p\) such that \(p<0\) , then for any \(x,y,z\) , the expression \(\dfrac{x\alpha+y\beta+z\gamma}{x\beta+y\gamma+z\alpha} = ?\)

**Q2)** If \(0<a,b<1\) such that the points \(z_1=a+i\) , \(z_2=1+bi\) and \(z_3=0\) form an equilateral triangle , find the value of \(a\) and \(b\).

**Q3)** Suppose \(z_1,z_2,z_3\) are vertices of an equilateral triangle inscribed in a circle given by \(|z|=2\). If \(z_1=1+i\sqrt{3}\) , then find \(z_2\) and \(z_3\).

**Q4)** If \(|Z|\leq 1 \ , \ |W| \leq 1\) then show that \(|Z-W|^2 \leq (|Z|-|W|)^2+(\text{arg}(Z) - \text{arg}(W))^2\)

**Q5)** Let \(\overline{b}z+b\overline{z}=c\) , \(b\neq 0\) , be a line in complex plane , where \(b\) is complex conjugate of \(b\). If a point \(z_1\) is reflection of point \(z_2\) in this line , show that \(c=\overline{z_1}b+z_2\overline{b}\).

## Comments

Sort by:

TopNewestQ1) Since \(\alpha\), \(\beta\), \(\gamma\) are the cube root so \(p\), it implies that they are the roots of the equation \(x^3 = p\). This implies that \(x= \sqrt[3]{p} \) \(\implies\) \(\{ \alpha, \beta, \gamma \} = \{ k , k\omega, k \omega ^2 \}\), where \(k\) is the real cube root of \(p\) and \(\omega\) is the primitive cube root of unity. So, WLOG let \(\alpha = k\), \(\beta = k \omega\) and \(\gamma = k \omega^2\).

Now consider the equation \(x \beta + y \gamma + z \alpha = kx \omega + ky \omega^2 + kz\). But since \(\omega ^3 = 1\).

\[\begin{align*} x \beta + y \gamma + z \alpha &= kx \omega + ky \omega^2 + kz \\ &= kx \omega + ky \omega^2 + kz \omega^3 \\ &= \omega (kx + ky \omega + kz \omega^2) \\ &= \omega (x \alpha + y \beta + z \gamma ) \end{align*} \]

So, we get that

\[\dfrac{x \alpha + y \beta + z \gamma}{x \beta + y \gamma + z \alpha} = \dfrac{1}{\omega} = \omega^2 \] – Surya Prakash · 1 year, 10 months ago

Log in to reply

Clarification:\(\omega\) can take any value of the two possible values: \(\dfrac{-1+i\sqrt 3}{2}\) and \(\dfrac{-1-i\sqrt 3}{2}\). – Sandeep Bhardwaj · 1 year, 10 months agoLog in to reply

– Abdur Rehman Zahid · 1 year, 10 months ago

Nice!!!Log in to reply

Q4:

Let \(Z = r_1 cis(\theta_1)\) and \(W = r_2 cis(\theta_2)\).

Let's consider: \(|Z - W|^2 = (r_1 cos(\theta_1) - r_2 cos(\theta_2))^2 + (r_1sin(\theta_1) - r_2sin(\theta_2))^2\) \(|Z - W|^2 = r_1^2 cos^2(\theta_1) + r_2^2 cos^2(\theta_2) - 2r_1 r_2 cos(\theta_1) cos(\theta_2) + r_1^2 sin^2(\theta_1) + r_2^2 sin(\theta_2) - 2r_1 r_2 sin(\theta_1) sin(\theta_2)\)

Doing a bit of algebra, and knowing some trig. identities, we obtain

\(|Z - W|^2 = r_1^2 + r_2^2 - 2r_1 r_2 cos(\theta_1 - \theta_2)\)

Subtracting and adding \(2r_1 r_2\) allows us to factorise as follows:

\(|Z - W|^2 = (r_1 - r_2)^2 + 2r_1 r_2 - 2r_1 r_2 cos(\theta_1 - \theta_2)\)

\(|Z - W|^2 = (r_1 - r_2)^2 + 2r_1 r_2(1 - cos(\theta_1 - \theta_2))\).

Using the double-angle identity for cosine,

\(|Z - W|^2 = (r_1 - r_2)^2 + 4r_1 r_2(sin^2(\frac{\theta_1 - \theta_2}{2}))\)

Now, \(r_1, r_2 \leq 1\) and we allegedly know that \(|\theta| \geq |sin(\theta)|\) which can be easily proven.

Thus, we can say, \(|Z - W|^2 \leq (r_1 - r_2)^2 + 4(\frac{\theta_1 - \theta_2}{2})^2\)

\(|Z - W|^2 \leq (r_1 - r_2)^2 + (\theta_1 - \theta_2)^2\), by how we defined \(Z\) and \(W\), \(\theta_1\) and \(\theta_2\) are their principal arguments and \(r_1\) and \(r_2\) being their modulus respectively.

Therefore, \(|Z - W|^2 \leq (|Z| - |W|)^2 + (Arg(Z) - Arg(W))^2\).

As required. 🌝🌚

Please kindly point out any errors if there are any. – Vishnuram Leonardodavinci · 1 year, 10 months ago

Log in to reply

_– Nihar Mahajan · 1 year, 10 months agoLog in to reply

– Vishnuram Leonardodavinci · 1 year, 10 months ago

^^ Glad you liked it.Log in to reply

For Q3 use rotation of complex number. Let O(0,0) be the centre of circle then by rotating AO (A=\((1,\sqrt{3})\) by 120 degrees clockwise and anticlockwise obtain \(z_{2}\) and \(z_{3}\) – Tanishq Varshney · 1 year, 10 months ago

Log in to reply

@Vishnuram Leonardodavinci @Surya Prakash @Tanishq Varshney A Great Thanks for your time and help :)

Also I got Q5 ;) – Nihar Mahajan · 1 year, 10 months ago

Log in to reply

– Vishnuram Leonardodavinci · 1 year, 10 months ago

No worries. :D Glad to be of help.Log in to reply

– Surya Prakash · 1 year, 10 months ago

Ur welcome \(\Huge \ddot{\smile}\)Log in to reply

– Tanishq Varshney · 1 year, 10 months ago

Anytime ;)Log in to reply

Q3: Sorry man, I forgot how to include a diagram. Hopefully it was articulate enough to be of any use.

\(z_1 = 1 + i \sqrt{3}\) can be written as \(2e^{i \pi / 3}\).

Note, that the circle is about the origin, so we can get the principal argument using some circle properties (i.e., angle subtended at the circumference is half the angle subtended at the centre) as \(z_2\) and \(z_3\) as \(\pi\) and \(-\pi/3\).

Also, they all lie on the circumference of the circle \(\Rightarrow |z_2| = |z_3| = 2\)

Hence, \(z_2 = 2cis(\pi)\) and \(z_3 = 2cis(-\pi/3)\)

In Cartesian form, \(z_2 = -2\) and \(z_3 = 1 - i \sqrt{3}\) – Vishnuram Leonardodavinci · 1 year, 10 months ago

Log in to reply

I'll make it as concise as possible. :P

Q2: \(|z_1 - z_2| = |z_2 - z_3| = |z_3 - z_1|\) because it's an equilateral triangle.

\(|(a - 1) + (1-b)i| = |1 + bi|= |-a - i|\)

\(a^2 - 2a + 1 + 1 - 2b + b^2 = 1 + b^2 = a^2 + 1\)

Which gives us two systems of equations, namely, \(a^2 - 2a + 2 - 2b + b^2 = 1 + b^2\) and \(1 + b^2 = a^2 + 1\).

Simplifying them further yields, \(a^2 - 2a + 1 - 2b = 0\) and \(a^2 = b^2\).

Since \(0 < a,b < 1 \Rightarrow a = b\) from the second equation.

Noting the constraint and solving the system gives us the following solution: \(a = b = 2 - \sqrt{3}\) – Vishnuram Leonardodavinci · 1 year, 10 months ago

Log in to reply

Is it clear now brother? – Mohaiminul Adil · 1 year, 10 months ago

Log in to reply

@Surya Prakash @Chew-Seong Cheong @Akhil Bansal @Brian Charlesworth @Sudeep Salgia – Nihar Mahajan · 1 year, 10 months ago

Log in to reply