Take the 2-minute tour ×
Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics. It's 100% free, no registration required.

How can I tell if $A$ and $\exp(B)$ commute?

For $[A, B]$ it's simply $AB-BA$ and for $[\exp(A), \exp(B)]$ I think it'd be $\exp(A)\exp(B) - \exp(B)\exp(A) = \exp(A+B) - \exp(B+A) = 0$. Update: it's not generally true.

Is there a 'simple' way to find $[A, \exp(B)]$?

Or is this one of those problems where, if you encounter them at all, you are probably doing something wrong? The example I am encountering is $[\vec{S}, \exp(S_z)]$).

share|improve this question
Note that it is NOT true in general that for matrices $A$ and $B$, $e^Ae^B=e^{A+B}$. This is true provided $A$ and $B$ commute. I'm not sure if there is any simple formula for [A, e^B] for arbitrary matrices, but I doubt that there is. –  joshphysics Jan 23 '13 at 23:06
$[A,BC]=[A,B]C+B [A,C]$, specifically $[A,B^{n+1}]=[A,B B^n]=[A,B]B^n+B [A,B^n]$ and $f(B)=\sum a_n B^n$. –  NikolajK Jan 23 '13 at 23:25
@NickKidman: it'd be a good idea to state what you mean by $f(B)$ and $a_n$... –  Vibert Jan 23 '13 at 23:48
@Mark: Does $[S, \exp(Sz)]$ mean $[\vec{S}, \exp(S_z)]$? –  Qmechanic Jan 23 '13 at 23:49
It does yes. I guess we should consider [Sx, exp(Sz)], .., [Sz, exp(Sz)] seperately. But it's really more of a general question; spin is just the example that raised it. –  Mark Jan 23 '13 at 23:52

2 Answers 2

up vote 15 down vote accepted

If OP wants to evaluate $[A,e^B]$ in terms of $[A,B]$, there is a formula

$$\tag{1} [A,e^B] ~=~\int_0^1 \! ds~ e^{(1-s)B} [A,B] e^{sB}. $$

Proof of eq.(1): The identity (1) follows by setting $t=1$ in the following identity

$$\tag{2} e^{-tB} [A,e^{tB}] ~=~ \int_0^t\!ds~e^{-sB}[A,B]e^{sB} .$$

To prove equation (2), first note that (2) is trivially true for $t=0$. Secondly, note that a differentiation wrt. $t$ on both sides of (2) produces the same expression

$$\tag{3} e^{-tB}[A,B]e^{tB},$$

where we use the fact that


So the two sides of eq.(2) must be equal.

Remark: See also this related Phys.SE post. (It is related because $[A, \cdot]$ acts as a linear derivation.)

share|improve this answer
That's what I'm looking for, thanks! Any chance you can explain where this comes from, or point me to a reference? –  Mark Jan 24 '13 at 0:25
I updated the answer. –  Qmechanic Jan 24 '13 at 0:57


so in order to $A$ and $e^B$ to commute, $A$ should commute with $B$ and hence with any power of $B$. You can apply this to $[\vec{S},e^{S_{z}}]$


for $i=x,y,z$

share|improve this answer
Thanks! Is there an expression for $[A, \exp(B)]$ if $[A, B]$ isn't 0? –  Mark Jan 24 '13 at 0:03
The first equation Burzum wrote is such an expression... You can simplify or rewrite the infinite series once you know [A,B] –  zakk Jan 24 '13 at 0:05
Sorry, what I meant was an expression without infinite sum. E.g. with an expontentiation. Is it reducable to something like that? –  Mark Jan 24 '13 at 0:10

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.