We know that the splitting field of $x^5 - 2 $ over $\mathbb Q$ is $\mathbb Q(2^{1/5}, \rho)$, where $\rho$ is a fifth root of unity.

Therefore, $\left[\mathbb Q(2^{1/5} , \rho) : \mathbb Q \right] = 20 $. Let $G$ be the galois group of $\mathbb Q(2^{1/5} , \rho)$. Then $|G| = 20$.

How to find the Galois group $G$?

Can we generalize to the Galois group of the splitting field of $x^p -2$ over $\mathbb Q$?

  • 2
    $\begingroup$ Very closely related. Could actually be a duplicate, but for hopefully understandable reasons I'm a bit reluctant to use my AA-powervote. The answer there is more about hints, but the asker of the duplicate elaborates nicely IMHO. Anyway, you get $C_p\rtimes C_{p-1}$ as the answer. $\endgroup$ – Jyrki Lahtonen Dec 18 '14 at 5:02

It's the Frobenius group $C_5\rtimes C_4$.

Hint: Start with $$\sigma:\left\{\begin{array}{l}2^{1/5}\mapsto \rho 2^{1/5}\\ \rho\mapsto \rho\end{array}\right.\hspace{15pt}\text{and}\hspace{15pt}\tau:\left\{\begin{array}{l}2^{1/5}\mapsto 2^{1/5}\\ \rho \mapsto \rho^2\end{array}\right.$$ To generalize, take a look at the order $\bmod p$ of the power in $\tau$, and play around with the conjugation $\tau^{-1}\sigma\tau$ until your group comes out to match the degree of your extension.


By Sylow's Theorem, we have that $P_5\trianglelefteq G$ is normal, which we expect since $\Bbb Q(\rho)/\Bbb Q$ is Galois. Since we know that $\text{Gal}(\Bbb Q(\rho)/\Bbb Q)\cong \Bbb Z/4$ we see that there is a surjection $G\to\Bbb Z/4$, hence we see that we have the short, exact sequence:

$$1\to \Bbb Z/5\stackrel{i}{\longrightarrow} G\stackrel{p}{\longrightarrow} \Bbb Z/4\to 1.$$

Now since the map $G\to \Bbb Z/4$ is surjective, $G$ has an element of order $4$, hence we have a section of the surjection, i.e. a map $\psi: \Bbb Z/4\to G$ such that $\psi\circ p= \text{id}_{P_2}$, and we conclude $G$ is a semidirect product, but since $P_2$ is not normal--$\Bbb Q \left(\sqrt[5] 2\right) /\Bbb Q$ is not Galois after all--we have that the product is not direct so:

$$G\cong \Bbb Z/4\ltimes \Bbb Z/5$$

We can generalize as well: With $\zeta_p$ a primitive $p^{th}$ root of $1$, we have $\text{Gal}\left(\Bbb Q(\zeta_p)/\Bbb Q\right)$ is always isomorphic to $\left(\Bbb Z/p\right)^*$ which is cyclic when $p$ is prime, the sequence will always split by the surjectivity of the map from $G\to\Bbb Z/(p-1)$ and you will always get the semi-direct product from the lack of being abelian.


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.