Show that $G/\Phi(G)$ is a non-abelian simple group, where $\Phi(G)$ denotes the Frattini subgroup of $G$

So $G/\Phi(G)$ can't be abelian since if it were then is would be solvable and since $\Phi(G)$ is a solvable normal subgroup of $G$, it would imply that $G$ is solvable.

For the next part, I think I was able to prove that $G$ is simple which would mean $G/\Phi(G)$ is simple by the correspondence theorem, but my intuition is telling me that showing $G$ is simple is a little to over reaching so I think I made a mistake.

For the sake of contradiction suppose $G$ has a proper non trivial normal subgroup. Let $N$ be a minimal proper normal subgroup and let $P$ be a Sylow subgroup of $N$. So by the Fratini Argument $G = N_G(P)N$. Since $N$ is minimal normal $N_G(P)$ must be a proper subgroup. But $N_G(P)N/N \cong N_G(P)/N_G(P)\cap N$ which is solvable since $N_G(P)$ is solvable and the quotient group of a solvable group is solvable.

  • $\begingroup$ You have just proved that $P=N$. $\endgroup$
    – Derek Holt
    Jul 21, 2020 at 18:23
  • 2
    $\begingroup$ You will not be able to prove that $G$ is simple, because there are counterexamples. For example, $\mathrm{SL}_2(5)$ has order $120$, is not soluble (it has a quotient $A_5$) and has no subgroup $A_5$ as it has a single element of order $2$). $\endgroup$ Jul 21, 2020 at 18:28

1 Answer 1


Let $N$ be any proper normal subgroup of $G$. If $N \not\le \Phi(G)$, then there is a maximal subgroup $M$ of $G$ with $N \not \le M$. So then $G = NM$ by maximality of $M$, and then $N$ and $M$ are both solvable and hence so is $G$.

So $N \le \Phi(G)$ and $G/\Phi(G)$ must be simple.


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.