2
$\begingroup$

It is well known that the cyclotomic polynomials $\Phi_n(x)$ are irreducible over the field of rationals $\mathbb{Q}$. I am curious about their reducibility over the real numbers $\mathbb R$.

We have $$\Phi_n(x)=\prod_{\substack{1\leqslant k\leqslant n \\\gcd(k,n)=1}}(x-\zeta_n^k)$$ where $\zeta_n:=e^{2\pi i/n}$.

So $\Phi_n$ is reducible when $2 k/n\in\mathbb Z$ for all $\gcd(k,n)=1$ which is clearly impossible.

So am I safe to conclude that $\Phi_n(x)$ are irreducible also over $\mathbb R$?

EDIT

As is mentioned by @Wojowu, every polynomial of degree $\geqslant 3$ is reducible in $\mathbb R$, so my question becomes... How to write down the irreducible factors of the cyclotomic polynomials over $\mathbb R$.

$\endgroup$
  • 5
    $\begingroup$ Every polynomial of degree greater than $2$ is reducible over $\mathbb R$. $\endgroup$ – Wojowu May 8 '19 at 20:58
  • $\begingroup$ @Wojowu, Thanks for the comment... $\endgroup$ – qinr May 8 '19 at 21:04
3
$\begingroup$

If $n \neq 1,2,3,4,6$ then $\phi(n) >2$ and $$Q(x)=(x-\zeta_n)(x-\zeta_n^{n-1})$$ is a divisor of your polynomial with real coefficients.

As for the edited question For each $\gcd(k,n)=1$ the polynomial $$Q(x ) = ( x-\zeta_n^k) (x-\zeta_n^{n-k}) $$ is irreducible over $\mathbb R$.

Extra Here is the proof that, for $n \neq 1,2$, and all $(k,n)=1, Q(x)$ is irreducible over $\mathbb R$.

First, it is easy to see that $\zeta_n^{n-k}=\overline{\zeta_n^{k}}$. You can see this either using the polar/trig form or by observing that $$\zeta_n^{n-k}=\frac{\zeta_n^{n}}{\zeta_n^{k}}=\frac{1}{\zeta_n^{k}}=\frac{\overline{\zeta_n^{k}}}{\zeta_n^{k}\cdot \overline{\zeta_n^{k}}}=\overline{\zeta_n^{k}}$$

From here it follows immediately that the coefficients are real.

If $Q(x)$ would be reducible, it would be a product of 2 linear terms, and hence its roots would be real. Therefore $$\zeta_n^{k} \in \mathbb R$$

But then $\zeta_n^{k}$ would be a real root to $x^n-1$, and hence $\zeta_n^{k} \in \pm 1$.

Since $(k,n)=1$ this is also a primitive $n$th root of unity, and hence $n \in \{1,2\}$, meaning the cyclotmic polynomial is linear.

$\endgroup$
  • $\begingroup$ Excellent answer. Any hint on how to prove that $Q(x)$ are irreducible? $\endgroup$ – qinr May 8 '19 at 21:15
  • $\begingroup$ Check that the coefficients of each $Q(x)$ are real, but the roots are not. $\endgroup$ – Sameer Kailasa May 8 '19 at 21:52
  • $\begingroup$ @qinr See the edit. $\endgroup$ – N. S. May 8 '19 at 22:25
  • $\begingroup$ @N.S. Thanks a lot! $\endgroup$ – qinr May 8 '19 at 22:42

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.