Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

I need one to test a theory. There are probably many, but I can't seem to think of a single one. My guess is examples are pretty big.

Is there a systematic way to find such examples? Are there databases one can go through to find these things?

share|improve this question
    
Are you thinking of finite groups? –  Yemon Choi Jan 27 '11 at 21:13
    
Finite. I'll change it. –  Makhalan Duff Jan 27 '11 at 21:15
    
Can't you take products of simple groups? –  José Figueroa-O'Farrill Jan 27 '11 at 21:19
    
A direct product of simple groups can be 2-generated. –  Mark Sapir Jan 27 '11 at 21:38
    
A direct product of enough copies of a given FSG is not 2-generated. –  Jonathan Kiehlmann Jan 27 '11 at 21:48
show 1 more comment

3 Answers 3

up vote 10 down vote accepted

The group $S_3\wr ({\mathbb Z}_2 \times {\mathbb Z}_2 \times {\mathbb Z}_2)$ where $S_3$ is the symmetric group with 6 elements, ${\mathbb Z}_2$ is the group with 2 elements, $\wr$ is the wreath product.

The fact that it does not have a center is proved by inspection. The fact that it needs at least 3 generators follows from the fact that ${\mathbb Z}_2^3$ is its quotient. There are lots of similar examples of course.

Update. In general , if you take any centerless group $G$ and any group $H$ that needs at least 3 generators, then the wreath product $G\wr H$ has both properties (is centerless and needs at least 3 generators). Another way to construct examples is (as Derek Holt comment below shows) to take any centerless finite group $G$ with nontrivial abelianization and take $G\times G\times G$.

share|improve this answer
2  
Isn't that written $S_3\wreath\mathbb Z_2$? –  Mariano Suárez-Alvarez Jan 27 '11 at 21:29
    
$A\wr B$ is the semidirect product of $A^B$ and $B$ where $B$ acts on the direct power $A^B$ by permuting components of vectors. There is a competing notation for the same thing $B\wr A$ used by some dynamics people. But the original (I think due Kaluzhnin and Krasner) is as above. Note that ${\mathbb Z} \wr S_3$ has a center, consisting of constant vectors in the base, also it is not that trivial to show that this wreath product needs 3 generators. –  Mark Sapir Jan 27 '11 at 21:35
    
Why not just $S_3^3$ ? –  Derek Holt Jan 27 '11 at 22:54
1  
@Mariano: Yes, $3!=6$. –  Mark Sapir Jan 27 '11 at 23:42
1  
@Mariano: Your computation is absolutely correct. –  Mark Sapir Jan 28 '11 at 0:50
show 3 more comments

The smallest example has order 18. It has a normal elementary abelian subgroup $N$ of order 9, and an element $t$ of order 2 such that $txt=x^{-1}$ for all $x \in N$.

share|improve this answer
1  
More generally, an extension of an elementary abelian 3-group $N$ of order $3^r$ by an element of order 2 that inverts everything in $N$ is the smallest example of a centreless group that needs $r+1$ generators. –  Derek Holt Jan 28 '11 at 9:15
add comment

I can't prove this, but the Rubik's cube group might work.

Update: so this doesn't work. Thanks for the comments.

share|improve this answer
1  
Why do you think so? In any case, gap tells me that the center is cyclic of order $2$ (I'm using the generators from gap-system.org/Doc/Examples/rubik.html) –  Mariano Suárez-Alvarez Jan 28 '11 at 8:41
1  
The non trivial central permutation is the one which flips the middle cube of each side. Cute. –  Mariano Suárez-Alvarez Jan 28 '11 at 8:45
    
A quick calculation in GAP or Magma also shows that this group is 2-generated. –  Derek Holt Jan 28 '11 at 9:13
1  
@Derek, how did you check? I tried GQuotients(FreeGroup(2), G : findall := false) with G is the cube group module its center (which is centerless) but it seems to take forever---I am pretty sure there are faster ways to do this? –  Mariano Suárez-Alvarez Jan 28 '11 at 9:17
2  
@Mariano: gap> G = Subgroup(G, [Random(G), Random(G)] ); true The probability of this returning true seems to be about 0.55. –  Derek Holt Jan 28 '11 at 12:07
show 1 more comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.