MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Inspired by the concept in the following post What are these compact sets called? We introduce the following concept:

Let $A$ be a unital $C^{*}$ algebra. We consider the unitary equivalent relation on the set of all projections in $A$. We denote by $n(A)$, the number of equivalents classes of projections in $A$. The number of connected components of $A$ is denoted by $ncc(A)$ and is defined by $ncc(A)=log_{2} n(A)$. For a non unital $C^{*}$ algebra $A$, we define $ncc(A)=ncc(M(A))$ where $M(A)$ is the multiplier algebra of $A$. This is a natural definition because for a locally compact Hausdorff space $X$ we have $ncc(C_{0}(X))=$ The number of connected components of $X$.

The concept in the above post is a motivation to ask:

Question 1: Assume that We have an extension $0\to A\to B\to C\to 0$ of $C^{*}$ algebras with $ncc(A)< \infty$ and $ncc(C)< \infty$. Does this implies that $ncc(B)$ is finite, too? How can $ncc(B)$ be controlled by $ncc(A)$ and $ncc(C)$?

Question 2: How can $ncc(A\otimes_{min} B)$ be controlled in term of $ncc(A)$ and $ncc(B)$? When $A$ is commutative, is it true to say that $ncc(A\otimes B) \leq ncc(A).ncc(B)$?

Note: The commutative case gives us enough motivation for such noncommutative questions.

share|cite|improve this question
As a sign of good faith, why don't you tell us what you've been able to work out when A is a noncommutative finite-dimensional Cstar algebra? – Yemon Choi May 27 '14 at 4:11
@YemonChoi For the matrix algebra n(A) is n+1.So when A,C are finite dimensional algebra, the first question is obvious, the second is not a so hard(for large value n, m) – Ali Taghavi Feb 6 at 10:55
Thanks. So according to your definition, $ncc(M_n({\bf C}))= \log_2(n+1)$? This seems slightly odd, and it's not clear to me why this definition is better than looking at equivalence classes of projections in stabilized versions of the algebra (i.e. $K_0)$. – Yemon Choi Feb 6 at 15:30
Also, why does an answer for the matrix algebra case tell you what is going on for the finite-dimensional case? It seems perfectly possible that if $A\subset M_n({\bf C})$ then two projections in $A$ could be unitarily equivalent in $M_n({\bf C})$ without being unitarily equivalent in $A$... – Yemon Choi Feb 6 at 15:32
@YemonChoi Thank s for your comments.yes it is possible. for example $A=M_{2}(\mathbb{C})\oplus M_{2}(\mathbb{C})\subset M_{4}(\mathbb{C})$. Let $e\in M_{2}(\mathbb{C})$ be the projection on $x$-axis. Then $e\oplus I_{2}$ and $I_{2} \oplus e$ are similar in the whole algebra but not similar in $A$. after all, I think the main question in the post is obvious for finite dim algebras since any such algebra is a direct sum of matrix algebras. – Ali Taghavi Feb 6 at 16:49

Your Answer


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

Browse other questions tagged or ask your own question.