1

My group is $G=(\mathbb{Z}_3\times\mathbb{Z}_3)\rtimes\mathbb{Z}_3$ which is non abelian group of order $27.$

Now my problem is whether the group $1+J(FG)$ is abelian or non-abelian and what is its exponent? Here $F$ is any finite field of characteristic $3.$ I only know that $(1+J(FG))^{3^3}=1,$ by using below proposition given in the book "The Jacobson radical of group algebras" by G.Karpilovsky.

$\textbf{Proposition}$. Let $N$ be a normal subgroup of $G$ such that $G/N$ is $p$-solvable. If $|G/N|=np^a$ where $(p,n)=1$ then $$J(FG)^{p^a}\subseteq FG.J(FN)\subseteq J(FG)$$ In particular, if $G$ is $p$-solvable of order $np^a$ where $(p,n)=1,$ then $$J(FG)^{p^a}=0.$$

Please anyone try to help me . I will be very thankful. Thanks.

neelkanth
  • 6,048
  • 2
  • 30
  • 71

1 Answers1

1

It seems to me that since $J(FG)$ is maximal and contains the augmentation ideal, it contains $a-1, b-1$ for any elements $a,b\in G$. In particular, if you select $a,b$ such that $ab\neq ba$ this is true.

Then $a=a-1+1$ and $b=b-1+1$ are both in $1+J(FG)$ and they don't commute.

I'm not sure why anything harder is necessary...

rschwieb
  • 153,510
  • Thanks sir ....actually as you discussed in https://math.stackexchange.com/questions/2049645/computing-the-jacobson-radical-of-fg-with-charf-p-and-g-finite-with-a according to it J(FG ) is same as augmentation ideal as unique sylow subgroup is group itself so right side group being trivial and hence kernel is augmentation ideal. – neelkanth Nov 12 '18 at 16:32
  • my last question is about exponent of group $1+J(FG)$...thanks ... – neelkanth Nov 12 '18 at 16:36
  • Can i say its exponent must be $27$ as $3$ and $9$ order groups are always abelian.? – neelkanth Nov 12 '18 at 16:39
  • So $1+J(FG)$ is a non abelian group of exponent $27?$ – neelkanth Nov 12 '18 at 16:39
  • @neelkanth You asked if it was abelian or not, and I'm saying it doesn't appear to be. You do see that the map $x\to 1+x$ is one to one, right? And it is into hard to count the elements of the augmentation ideal. In this case it's a subspace of codimension $1$. Does that answer your question? – rschwieb Nov 12 '18 at 17:08
  • Yes i got it ...And thank you very much for your quick replay ...again thanks ... – neelkanth Nov 12 '18 at 17:16