0
$\begingroup$

I am trying to find $[\mathbb Q(\omega):\mathbb Q]$, where $\omega=\cos(\frac {2\theta\pi}\phi)+i\sin(\frac {2\theta\pi}\phi),$ and $\theta,\phi\in\mathbb Z$.

I have tried putting $\omega$ into exponential form, so, $\omega=e^{\frac{2\theta i\pi}\phi}$.

Then considered $\omega^\phi$, so that we have $\omega^\phi=1$.

Thus $\omega$ is a root of the polynomial $x^\phi-1$.

Hence $\omega$ is algebraic over $\mathbb Q$.

I now want to use the fact that if this polynomial is irreducible in $\mathbb Q$, then $[\mathbb Q(\omega):\mathbb Q]=\phi$, since if it is irreducible then it will the the minimal polynomial of $\omega$ over $\mathbb Q$.

But I can't seem to prove that it is, not using Eisenstein's criterion anyway.

I have a feeling that cyclotomic polynomials may be useful, since they are all irreducible, but I can't quite see how to use this fact.

Any help would be appreciated, thanks.

$\endgroup$
  • $\begingroup$ $x^{\phi} - 1$ is not irreducible, it even has a rational root: $1$. $\endgroup$ – Paul K Feb 27 '17 at 15:11
2
$\begingroup$

You're right with your reasoning, as $\omega$ is indeed a root of a cyclotomatic polynomial. To notice this first reduce $\frac{\theta}{\phi}$ to $\frac{\theta_1}{\phi_1}$ such that $(\theta_1, \phi_1) = 1$ and $\phi_1>0$. Then we have that $\omega$ is a root of $\Phi_{\phi_1}$ as

$$\Phi_n = \prod_{1\le k \le n\text{ and }(n,k) = 1} (x - e^{2\pi \frac kn})$$

Obviously from this definition we have that the degree of the polynomial is the number of coprime integers to $\phi_1$ and less than $\phi_1$, which can be found using the Euler's Totient Function.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.