Reading this book, I came across the following formula

$$e^A e^B = e^{A+B}e^{\frac{1}{2}[A,B]}$$

where $A$ and $B$ are two matrices and $[A,B] := AB-BA$.

I tried to find a proof, without success. It is impossible to use the binomial theorem since $A$ and $B$ do not commute. I've thought about developping the product in a power series, but I'm not sure the Cauchy product is allowed when $[A,B] \ne 0$.

I thought that $[A,B^k]$ could be useful so I searched the general expression

$$[A,B^n] = \sum_{i=0}^{n-1} B^i [A,B] B^{n-i-1}$$

Does anyone know a proof of the formula above? If so, is it possible to provide some hints? Or a full proof?


3 Answers 3


The equality is true when $A,B$ are quasi-commutative, that is, when $[A,B]$ commutes with both $A$ and $B$.


The author probably means an approximation rather than a true equality, as $(A+B)+\frac12[A,B]$ are only the first two terms in the BCH formula Zassenhaus formula.

  • $\begingroup$ How can one be sure that the terms after can be neglected ? $\endgroup$
    – ChocoPouce
    Commented Dec 2, 2013 at 17:24
  • $\begingroup$ @ChocoPouce The trailing terms are not always negligible. You may search the internet for convergence results. For example, see sec. 3 and fig. 1 of Casas et al., Efficient computation of the Zassenhaus formula. $\endgroup$
    – user1551
    Commented Dec 2, 2013 at 18:50

The book you referenced was "Quantum Physics in One Dimension". The operators in quantum mechanics have to obey particular commutation relations. For instance the operators $x$ and $p$ obey the canonical commutation relation,

$$ \left[ x, p \right] = i I. $$

This allows us to simplify the Zassenhaus formula. The operator $[x,p]$ commutes with every operator because it is proportional to the identity; if you consider the higher order terms in the Zassenhaus formula you will see they involve commutators which must be zero.

Meaning that,

$$\exp(x+p)= \exp(x)\exp(p)\exp(-[x,p]/2),$$

is an exact relationship for the operators $x$ and $p$.

Knowledge of specific commutation relations can allow you to simplify the Zassenhaus formula significantly.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .