4
$\begingroup$

Let $G$ be a finite simple group (nonabelian, for simplicity). Let $n(G,k)$ denote the number of elements of $G$ of order $k$. I'm interested in results which bound quantities related to $n(G,k)$.

For example, are there bounds for $\frac{n(G,k)}{|G|}$? For $\frac{n(G,k)}{k}$? For some other reasonable normalization of $n(G,k)$?

$\endgroup$
3
  • 1
    $\begingroup$ I think we have $n(G,2)\leq\frac34|G|$ and $n(G,3)\leq\frac78|G|$. $\endgroup$
    – MSMalekan
    Commented Dec 14, 2021 at 4:09
  • $\begingroup$ @GeoffRobinson Do we expect $n(G,k)$ to be largest for small $k$? (eg $k = 2$?). Is it possible to quantify how the bounds improve as $k$ increases? $\endgroup$ Commented Dec 14, 2021 at 18:39
  • $\begingroup$ Related question mathoverflow.net/q/337477/297 (but without the hypothesis that $G$ is simple, so the answers are pretty different). $\endgroup$ Commented Jun 2, 2022 at 14:42

2 Answers 2

8
$\begingroup$

For $w$ a word in a free group $F_d$, we can consider the word map $\bar w : G^d \to G$. Let $P_w(G)$ denote the proportion of $d$-tuples $x \in G^d$ such that $\bar w (x) = 1$. Your notation is related by

$$P_{x^k}(G) = \sum_{\ell \mid k} n(G, \ell) / |G|.$$

For any fixed nontrivial word $w$ we have $P_w(G) \to 0$ as $|G| \to \infty$ for nonabelian finite simple groups $G$. [1]

[1] Dixon, John D.; Pyber, László; Seress, Ákos; Shalev, Aner, Residual properties of free groups and probabilistic methods, J. Reine Angew. Math. 556, 159-172 (2003). ZBL1027.20013.

$\endgroup$
4
$\begingroup$

Here is a different, reasonably elementary, proof that for any $\varepsilon > 0,$ there are only finitely many non-Abelian simple groups $G$ with $n(G,2) > \varepsilon |G|.$

The number of involutions of such a $G$ is $\sum_{\chi \neq 1} \nu(\chi) \chi(1)$ where $\nu$ is the Frobenius-Schur indicator (which always takes value $0 , 1 $ or $-1$). This is certainly less than $\frac{|G|}{d},$ where $d$ is the minimum degree of a non-trivial complex irreducible character of $G$.

By Jordan's theorem on complex linear groups, for any positive integer $n$, there are only finitely many non-Abelian simple groups which have a non-trivial irreducible character of degree $n$, so the claimed result follows.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .