centralizer
Let G be a group. The centralizer of an element a∈G is defined to be the set
C(a)={x∈G∣xa=ax} |
Observe that, by definition, e∈C(a), and that if x,y∈C(a), then xy-1a=xy-1a(yy-1)=xy-1yay-1=xay-1=axy-1, so that xy-1∈C(a). Thus C(a) is a subgroup of G. For a≠e, the subgroup is non-trivial, containing at least {e,a}.
To illustrate an application of this concept we prove the following lemma.
Lemma:
There exists a bijection between the right cosets of C(a) and the conjugates
of a.
Proof:
If x,y∈G are in the same right coset, then y=cx for some c∈C(a). Thus y-1ay=x-1c-1acx=x-1c-1cax=x-1ax.
Conversely, if y-1ay=x-1ax then xy-1a=axy-1 and xy-1∈C(a) giving x,y are in the same right coset.
Let [a] denote the conjugacy class of a. It follows that |[a]|=[G:C(a)] and |[a]|∣|G|.
We remark that a∈Z(G)⇔C(a)=G⇔|[a]|=1, where Z(G) denotes the center of G.
Now let G be a p-group, i.e. a finite group of order pn,
where p is a prime and n is a positive integer.
Let z=|Z(G)|.
Summing over elements in distinct conjugacy classes,
we have pn=∑|[a]|=z+∑a∉Z(G)|[a]|
since the center consists precisely of the conjugacy classes of
cardinality 1.
But |[a]|∣pn, so p∣z.
However, Z(G) is certainly non-empty, so we conclude that every
p-group has a non-trivial center.
The groups C(gag-1) and C(a), for any g, are isomorphic.
Title | centralizer |
---|---|
Canonical name | Centralizer |
Date of creation | 2013-03-22 12:35:01 |
Last modified on | 2013-03-22 12:35:01 |
Owner | drini (3) |
Last modified by | drini (3) |
Numerical id | 14 |
Author | drini (3) |
Entry type | Definition |
Classification | msc 20-00 |
Synonym | centraliser |
Related topic | Normalizer![]() |
Related topic | GroupCentre |
Related topic | ClassEquationTheorem |