2
$\begingroup$

Let $k$ be a skew field. Assume that $A$ is finite $k$-algebra, i.e., ${\rm dim}_k A = [A:k] < \infty$.

Before asking I will enumerate two definitions :

Def : A $k$-algebra $A$ is $central$ if the center of $A$ is $k$.

Def : A $k$-algebra is $simple$ if only two sided-ideal is $0$ or $A$.

Question : Is there an example which is not simple but central ? Or vice versa ?

(1) Let $A=\{ X\in M_n({\bf R})|\ A$ is upper triangular and diagonal entries are same $\}$. Note that $A$ is not simple and not central. Consider ${\bf R}e_{1n}$ where $e_{1n}$ has only nontrivial entry at $(1,n)$.

(2) ${\bf H}$ is simple and central.

${\bf Reference}$ : When I study Bruer group, I found the following material : http://stacks.math.columbia.edu/download/brauer.pdf

$\endgroup$
  • 3
    $\begingroup$ Do you require $k$ to be non-commutative? Otherwise, an easy example of one of those is $\mathbb{C}$ as an $\mathbb{R}$-algebra. $\endgroup$ – Tobias Kildetoft Feb 17 '14 at 8:57
  • $\begingroup$ Yes. It need not to be non-commutative. ${\bf C}$ is simple but not central. Thank you. And maybe do you know an example of opposite case ? $\endgroup$ – HK Lee Feb 17 '14 at 8:59
  • $\begingroup$ ${\bf H}\oplus {\bf H}$ is not simple but central. $\endgroup$ – HK Lee Feb 17 '14 at 9:07
  • 1
    $\begingroup$ @HeeKwonLee Hm, well normally $\Bbb R$ in $\Bbb H\oplus \Bbb H$ would be identified with $\{(a,a)\mid a\in \Bbb R \}$, but the center of $\Bbb H\oplus \Bbb H$ is $\Bbb R\oplus \Bbb R$, so I don't think that fits the definition of "central." $\endgroup$ – rschwieb Feb 17 '14 at 13:10
  • 1
    $\begingroup$ @HeeKwonLee It's a little weird to ask for $k$ to be a skew field when algebras are conventionally defined over commutative rings, and then you are talking about $k$ being the center of an algebra and hence commutative anyway. It's not really a huge deal, but I just thought I'd throw the comment out. $\endgroup$ – rschwieb Feb 17 '14 at 13:28
2
$\begingroup$

You can check that the ring $T_n(k)$ of all upper triangular matrices over a field $k$ with $n>1$ is central but not simple.

To make this a complete solution for the posted question, I'd like to echo Tobais Kildetoft's excellent example already given in the comments: $\Bbb C$ is a simple noncentral $\Bbb R$-algebra.


Comment on the post (and one of its comments):

I'm not certain what you want a $k$-algebra over a noncommutative ring to be, but if you just want to talk about a ring $A$ that's a finite dimensional left vector space over $k$, we can do that. If $k$ isn't commutative, then no simple $k$-algebra (in this sense) will ever be central. So you could, for example take the $2\times 2$ matrix ring over $\Bbb H$ as a simple non-central $\Bbb H$-algebra (in this sense).

Since the center $Cen(R\oplus S)$ is $Cen(R)\oplus Cen(S)$, you won't ever be able to find an algebra whose center is a field by looking at the product of two $k$-algebras. For this reason, $\Bbb H\oplus\Bbb H$ is not central, as claimed in the comments.

$\endgroup$
  • $\begingroup$ Thank you for your introducing calculation way of center. $\endgroup$ – HK Lee Feb 18 '14 at 2:34

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.