Today, this problem was given to me.

Let $F$ be a abelian free group of rank $r$. Show that it is isomorphic to an $r$-copy of $Z_{\infty}$.

I could do some messy job about it but so far I failed to solve it. The time for solving it in time is over, but please help me solve it. Thank you very much.

  • 1
    $\begingroup$ What exactly is Z_\infty ? $\endgroup$ – Ittay Weiss Dec 27 '12 at 7:00
  • $\begingroup$ @IttayWeiss: I had the same problem. Anyway, I assumed it is $\mathbb Z$ $\endgroup$ – Basil R Dec 27 '12 at 7:07
  • 1
    $\begingroup$ then now the question is, what is your definition for a free group of rank r? $\endgroup$ – Ittay Weiss Dec 27 '12 at 7:09
  • $\begingroup$ That is: there is a set $X$ of dimension r and $X$ can generate $F$. $\endgroup$ – Basil R Dec 27 '12 at 7:12
  • 3
    $\begingroup$ You cannot expect to solve a problem if you do not know the definitions of the words in the problem. $\endgroup$ – Derek Holt Dec 27 '12 at 10:04

I believe that you mean to say a free abelian group of rank $r$. The definition of a free abelian group $\mathfrak{F}$ of rank $r$ is a group with a generating set $\mathfrak{S}$ of size $r$ for which the only relation is that $[s,t]=1$ for each $s,t\in \mathfrak{S}$. Note that free abelian groups are not free groups when $r\geq 2$.

Once we have this definition digested, the path to victory is quite clear. Let $\epsilon_i$ denote the generator of the $i^{\rm th}$ $\mathbb{Z}$ in $\mathbb{Z}^{r}$. Number the generators in $\mathfrak{S}$ as $s_1,\ldots,s_r$. Define $\Phi:\mathfrak{F}\rightarrow \mathbb{Z}^r$ by $\Phi:s_i\rightarrow \epsilon_i$. Now, can you prove that $\Phi$ is an isomorphism?

We have the relations that $[s_i,s_j]=1$ for each $i,j$. Given an arbitrary word $w=s_{i_1}^{e_1}s_{i_2}^{e_2}\cdots s_{i_s}^{e_s}$ (with $i_1,\ldots,i_s$ not necessarily distinct) can you rewrite $w$ in the form $s_1^{f_1}\cdots s_r^{f_r}$? Where does $\Phi$ send that in $\mathbb{Z}^r$? Finally, can you prove that $\Phi$ is bijective using the fact there are no other relations on $\mathfrak{F}$?

  • $\begingroup$ "The only relation is" is too hand-wavy for my taste. I prefer the universal property. $\endgroup$ – Hagen von Eitzen Mar 10 '13 at 10:50
  • $\begingroup$ @HagenvonEitzen What's hand-wavy about it? The free abelian group of rank $r$ is the quotient of the free group on a set $\mathfrak{S}$ of $r$ generators by the normal subgroup generated by $\{[x,y]:x,y\in \mathfrak{S}\}$. $\endgroup$ – Alexander Gruber Mar 10 '13 at 10:55

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.