2

$A,B$ are quantum mechanical operators.

  • $[A,B]\neq 0$ that is given.

  • $e^{A}=\sum_{n=1}^{\infty} \frac{A^n}{n!} $

Is the following correct?

$$[e^{A},e^{B}]=e^{A}e^{B}-e^{B}e^{A}=e^{A+B}-e^{B+A}=0 $$

Qmechanic
  • 201,751
MH Yip
  • 187

2 Answers2

4

An even simpler example: \begin{align} s_z=\left( \begin{array}{cc} 1 & 0 \\ 0 & -1 \\ \end{array} \right)\, ,& \qquad s_y=\left( \begin{array}{cc} 0 & i \\ -i & 0 \\ \end{array} \right)\\ e^{-i\alpha s_z}=\left( \begin{array}{cc} e^{-i \alpha } & 0 \\ 0 & e^{i \alpha } \\ \end{array} \right)&\qquad e^{-i\beta s_y}=\left( \begin{array}{cc} \cos (\beta ) & \sin (\beta ) \\ -\sin (\beta ) & \cos (\beta ) \\ \end{array} \right) \end{align} and clearly $[e^{-i\alpha s_z} ,e^{-i\beta s_y}]\ne 0$.

Note that, in connection to the comments on the answer of @user124864, $e^{-i\alpha s_z}e^{-i\beta s_y}\ne e^{-i\alpha s_z-i\beta s_y-\frac{1}{2}\alpha\beta[s_z,s_y]}$ either.

You can easily verify using the first few terms of the explicit expansion that, in general $$ e^A e^B= \sum_n \frac{A^n}{n!}\,\sum_m \frac{B^m}{m!} \ne e^{A+B}=\sum_{p}\frac{(A+B)^p}{p!}\, . $$ If anything: \begin{align} e^Ae^B &=\left(1+A+\frac{A^2}{2!}+\ldots \right) \left(1+B+\frac{B^2}{2!}+\ldots \right)\\ &= 1+ (A+B)+ \frac{1}{2!} (A^2+2AB+B^2)+\ldots \tag{1} \end{align} but $$ (A^2+2AB+B^2)\ne (A+B)^2= A^2+AB+BA+B^2 \tag{2} $$ unless $AB=BA$, i.e. unless $[A,B]=0$.

ZeroTheHero
  • 45,515
3

No this is not correct. The problem is that you have used $e^Ae^B = a^{A+B}$. This holds when $A$ and $B$ are numbers, so you might expect that it would also holds for matrices or even general operators. This is not in general the case, as the other answers show explicitly.

Inzinity
  • 822