5
$\begingroup$

Is there a relation between the idempotent elements of a ring $R$ and those of $M_{n}(R)$ - the ring of $n \times n$ matrices over $R$?

$\endgroup$

2 Answers 2

8
$\begingroup$

Any idempotent $e$ of $R$ induces an idempotent $E=\mathop{diag}(e,\ldots,e)$ of $M_n(R)$. In fact, if $e_i$ are idempotents in $R$, then $E=\mathop{diag}(e_1,\ldots,e_n)$ is an idempotent of $M_n(R)$.

Conversely, if $R$ is nice enough, an idempotent $E$ in $M_n(R)$ can be diagonalized to $E=U^{-1} \cdot \mathop{diag}(e_1,\ldots,e_n) \cdot U$ for some $U \in M_n(R)$ and idempotents $e_i$ in $R$.

Of course, this relies crucially on $R$ being nice enough. One sufficient "nicety" condition is that $R$ is an AW*-algebra; see for example this paper.

$\endgroup$
0
$\begingroup$

Yes. You can see the relationship easily in the following way. Suppose the ring $R=R_1\times R_2$ so that there are two primitive idempotents $e_1$ and $e_2$ with $1=e_1 +e_2$. Next note that $M_n(R)\approxeq Hom_R(R^n,R^n)$. Then $M_n(R)\approxeq Hom_{R_1\times R_2}(\(R_1\times R_2\)^n,\(R_1\times R_2\)^n)\approxeq Hom_{R_1}(R_1^n,R_1^n)\times Hom_{R_2}(R_2^n,R_2^n) $ since $e_1 \centerdot Hom_{R_2}(R_2^n,R_2^n)=0$ and similarly for $e_2$. Thus idempotents are calculated relative to each factor in the ring decomposition.

$\endgroup$
6
  • 3
    $\begingroup$ I think you answered a question different from the one asked ;-) $\endgroup$ Nov 25, 2012 at 23:00
  • $\begingroup$ Dear Ray, Thank you very much of your answer. If we have idempotent of $R$ then it is easy to generated for $M_{n}(R)$, But i want to construct idempotent for $R$, if we have idempotent of $M_{n}(R)$. so the above proof is not useful for me, $\endgroup$
    – Ali
    Nov 26, 2012 at 4:19
  • 1
    $\begingroup$ Hi.Let $F$ be an infinite field, then there exist infinitely many distinct pair $(I,J)$ of minimal left ideals of $M_2(F)$ such that $M_2(F)=I \oplus J$. so this shows that in this case $F$ has only two trivial idempotent, But $M_2(F)$ has infinitely many nontrivial idempotent. you could find this point at exercise [11.b] page 443 of Hubgerford. So I think you should determine the property of your ring$R$ which entries of matrix ring come from it. $\endgroup$
    – Ali Reza
    Nov 26, 2012 at 5:59
  • $\begingroup$ Ok. But in local ring we have no problem. See Lemma 7 in the paper Journal of Algebra 301 (2006) 280–293, when is 2x2 matrix ring over a commutative local ring are strongly clean. $\endgroup$
    – Ali
    Nov 26, 2012 at 9:19
  • $\begingroup$ Dear friends. Let me ask my quedtion in another case. If $J$ is an ideal of $M_{n}(R)$ then there is an ideal $I$ of $R$ such that $J=M_{n}(I)$. Now, if $J$ is generated by a subset of idempotent of $M_{n}(R)$ say $S$, does $I$ is generated by a subet of idempotent of $R$ relatesd to $S$. $\endgroup$
    – Ali
    Nov 26, 2012 at 10:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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