Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

If $\hat O$ is an operator and $c\in \mathbb C$

Then can I safely say that $[\exp(c\hat O)]^\dagger=\exp(c^*\hat O^\dagger)$?

My reasoning is by taking the transpose term by term in the power series definition of $\exp$ but I don't know if that is valid or not!



share|cite|improve this question
Yes ... you just have to verify, that $O \mapsto O^\dagger$ is continuous in the operator topology you are using. – martini Oct 1 '12 at 8:47
@martini: Thank you! I am wondering suppose it is given that $O \mapsto O^\dagger$ is continuous, can the transpose of all power series be defined by its term by term transpose? – George Oct 1 '12 at 8:50
@martini: Btw, I am not too sure what it means for $O \mapsto O^\dagger$ to be continuous in my operator topology... Would you mind briefly explaining or provide a link to some (relatively simple) references? (I haven't done operator topology.) – George Oct 1 '12 at 8:52
I'll write an answer, but you have to explain me one thing: How do you want the powerseries $\exp(O) = \sum_k \frac 1{k!}O^k$ to be defined? – martini Oct 1 '12 at 8:56
@martini: Thanks! I am not sure what you mean by how to define the power series...? – George Oct 1 '12 at 8:58
up vote 2 down vote accepted

So I suppose that $\hat O\colon X \to X$ is a continous operator on some complex Hilbert space $X$. Then one usually defines \[ \exp(\hat O) := \sum_{k=0}^\infty \frac 1{k!} \hat O^k = \lim_{K \to \infty} \sum_{k=0}^K \frac 1{k!} \hat O^k \] where the limit is meant with respect to the operator norm \[ \|T\| = \sup_{\|x\| = 1} \|Tx\| \] The above series converges in this norm as we have \[ \left\|\sum_{k=L+1}^K \frac 1{k!}\hat O^k\right\| \le \sum_{k=L+1}^K \frac 1{k!}\|\hat O\|^k \to 0, \quad K,L \to \infty\] So $\exp(\hat O)$ is a well defined operator $X \to X$.

Concernig it's transpose, if we try to compute it formally we have \begin{align*} \exp(\hat O)^\dagger &= \left(\sum_{k=0}^\infty \frac 1{k!} \hat O^k \right)^\dagger \\&= \left(\lim_{K\to \infty} \sum_{k=0}^K \frac 1{k!} \hat O^k \right)^\dagger \\&\stackrel?= \lim_{K\to \infty} \left(\sum_{k=0}^K \frac 1{k!} \hat O^k \right)^\dagger \\&= \lim_{K\to \infty} \sum_{k=0}^K \frac 1{k!} {\hat O^\dagger}^k \\&= \exp(\hat O^\dagger) \end{align*} The only step which doesn't follow from the linearity and antimultiplicativity of $(-)^\dagger$ is marked with $\stackrel?=$, we want to know if $(-)^\dagger$ exchanges with limits with respect to the operator topology, that is, if $(-)^\dagger\colon L(X) \to L(X)$ is continuous (where $L(X)$ denotes the set of linear continous functions $X \to X$). As taking the transpose is linear, it suffices to prove it is bounded. We have - denoting the scalar product of $X$ by $(-,-)$ - for $T \in L(X)$: \begin{align*} \|T^\dagger\| &= \sup_{\|x\| = 1} \|T^\dagger x\| \\&= \sup_{\|x\| = 1} \sup_{\|y\| = 1} \left|(T^\dagger x, y)\right| \\&= \sup_{\|x\| = 1} \sup_{\|y\| = 1} \left|(x, Ty)\right| \\&= \sup_{\|y\| = 1} \sup_{\|x\| = 1} \left|(Ty, x)\right| \\&= \sup_{\|y\| = 1} \|Ty\| \\&= \|T\|, \end{align*} so $T \mapsto T^\dagger$ is bounded (even an isometry) and hence, $\stackrel?=$ is verified, we do indeed have $\exp(\hat O)^\dagger = \exp(\hat O^\dagger)$.

The continuity of $(-)^\dagger$ implies of course, that for every convergent series of operators we can apply the transpose termwise.

share|cite|improve this answer
Thank you, martini, for the very detailed answer! :) – George Oct 1 '12 at 9:30

Your Answer


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

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