1
$\begingroup$

Let $\mathcal A\subset B(H)$ be an unital $C^*$ algebra of operators on a Hilbert space $H$. Let's denote by $\mathcal P$ the set of projections in $\mathcal A$, that is $\mathcal P:=\{a\in \mathcal A \; | \; a=a^2=a^*\}$.

I'm wondering when $\mathcal P$ is closed under composition of operators. I was able to prove the following:

$\big(\mathcal P \text{ is closed under multplication}\big)\Leftrightarrow \big(\forall_{a,b\in \mathcal P}\; ab=ba\big)$

If $\mathcal A$ is a von Neumann algebra then projections spans set which is dense in $\mathcal A$ so in this case $\big(\mathcal P \text{ is closed under multiplication}\big)\Leftrightarrow \big(\mathcal A \text{ is abelian}\big)$. Can something similar be proven in general(without assuming $\mathcal A$ being a von Neumann algebra)?.

| cite | improve this question | |
$\endgroup$
1
$\begingroup$

No. If $\mathcal A=C_r(\mathbb F_2)$ (or any other projectionless unital C$^*$-algebra), then $$\mathcal P=\{0,I\}$$ is closed under multiplication.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Thank You for Your comment, Martin Argerami. So, there are noncommutative $C^*$ algebras where $\mathcal P$ is closed under multiplication. Is there any useful characterisation of this property? $\endgroup$ – Mogget Mar 30 '16 at 10:42
  • $\begingroup$ I have no answer for that, but I doubt so. It just happens that for some C $^*$-algebras projections are irrelevant. On the other hand many (most?) C $^*$-algebras contain copies of the matrix algebras, so $\mathcal P $ is not multiplicative. $\endgroup$ – Martin Argerami Mar 30 '16 at 11:48
  • $\begingroup$ Am I right that in this case center of $\mathcal A$ is non trivial? Does something change if we assume triviality of a center? $\endgroup$ – Mogget Apr 2 '16 at 10:22
  • $\begingroup$ The center of $C_r (\mathbb F_2) $ is trivial. $\endgroup$ – Martin Argerami Apr 2 '16 at 11:51
  • $\begingroup$ I might have not stated clearly what I meant. If we assume that the center is trivial, can we say something more about the question whether $\mathcal P$ is multiplicative? $\endgroup$ – Mogget Apr 2 '16 at 17:19

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.