A note on the Baker-Campbell-Hausdorff series in terms of right-nested commutators (Q2657249)

Exponentials of non-commuting operators appear in many areas of physics and mathematics, ranging from quantum mechanics to the theory of Lie groups and Lie algebras to the numerical analysis of differential equations. (quotes from the present paper). For exponential operations \(e^{X_{1}}e^{X_{2}}\cdots e^{X_{n}}\), there is the problem of how to represent them by a single exponential operation \(e^{Z},(Z={\Phi(X_{1},X_{2},\cdots X_{n}))}\). The authors get compact expressions for the Baker-Campbell-Hausdorff series \(Z = log(e^{X} e^{Y} )\) in terms of right-nested commutators. The Baker-Campbell-Hausdorff theorem states that \(\Phi(X, Y )\) can be expressed as \[\Phi (X,Y)=X+Y+\sum\limits_{m\geq 2}\Phi _{m}(X,Y),\] \[ \Phi _{m}(X,Y)=\sum\limits_{p_{i},q_{i}}\frac{(-1)^{m-1}}{m}\frac{ [X^{p_{1}}Y^{q_{1}}\cdots X^{p_{m}}Y^{q_{m}}]}{ (\sum\nolimits_{i=1}^{m}(p_{i}+q_{i}))p_{1}!q_{1}!\cdots p_{m}!q_{m}!}, \quad (p_{i}+q_{i} >0),\] \([X^{p_{1}}Y^{q_{1}}\cdots X^{p_{m}}Y^{q_{m}}]\) denotes the right-nested commutator based on the word \(X^{p_{1}}Y^{q_{1}}\cdots X^{p_{m}}Y^{q_{m}}.\) Thus, for instance: \([XY^{2}X^{2}Y]=[XYYXXY]:=[X,[Y,[Y,[X,[X,Y]]]]].\) Let consider the general case. The Dynkin-Specht-Wever theorem states as the following: \(\Phi (X_{1},X_{2},\cdot \cdot \cdot ,X_{n})=\sum\limits_{m\geq 1}\Phi _{m}(X_{1},X_{2},\cdot \cdot \cdot ,X_{n}),\) where \(\Phi_{m}(X_{1},X_{2},\ldots ,X_{n})\) is a homogeneous polynomial of degree \(m\) in the non-commutative variables \(X_{1},\ldots,X_{n}\), \[ \Phi _{m}(X_{1},X_{2},\ldots,X_{n})=\sum\limits_{\substack{ i_{1}+\cdots +i_{n}=m \\ i_{j}\geq 0}}\frac{1}{i_{1}!\cdots i_{n}!}\varphi _{m}(\underbrace{X_{1},\ldots, X_{1}} _{i_{1}},\ldots,\underbrace{X_{n},\ldots, X_{n}} _{i_{n}}),\] \[ \begin{split} \varphi _{m}(X_{1},X_{2},\ldots,X_{m}) &=\frac{1}{m} \sum\limits_{\sigma \in S_{m}}(-1)^{d_{\sigma }}\frac{1}{\binom{m-1}{ d_{\sigma }}}X_{\sigma (1)}X_{\sigma (2)}\cdots X_{\sigma (m)} \\ &=\frac{1}{m^{2}} \sum\limits_{\sigma \in S_{m}}(-1)^{d_{\sigma }}\frac{1}{\binom{m-1}{ d_{\sigma }}}[X_{\sigma (1)},[X_{\sigma (2)}\cdots \lbrack X_{\sigma (m-1)},X_{\sigma (m)}]\cdots]], \end{split} \] \(S_{m}\) denotes the permutation group and \(d_{\sigma}\) is the number of descents in \(\sigma\). The authors state that they reduce the number of commutators by using as a starting point an explicit expression directly involving independent commutators and by deriving a complete set of identities arising among right-nested commutators as the following: \[\varphi _{m}(X_{1},X_{2},\ldots,X_{m})=\frac{1}{m} \sum\limits_{\sigma \in S_{m-1}}(-1)^{d_{\sigma }}\frac{1}{\binom{m-1}{ d_{\sigma }}}[X_{\sigma (1)},[X_{\sigma (2)}\cdots \lbrack X_{\sigma (m-1)},X_{m}]\cdots ]]. \] I recommend reading the present paper along with [\textit{M. Müger}, ``Notes on the theorem of Baker-Campbell-Hausdorff-Dynkin'', \url{https://www.math.ru.nl/~mueger/PDF/BCHD.pdf}].