Gell-Mann and low formula for degenerate unperturbed states (Q625072)

A crucial ingredient in the case of many-body theory is the adiabatic switching. It allows to express the eigenstates of a Hamiltonian \(H_0+V\) in terms of eigenstates of \(H_0\). The main idea of this approach is to switch very slowly the interaction \(V\) to transform \(H_0+V\) using a typical form \(H_0+e^{-\varepsilon|t|}V\) where \(\varepsilon>0\) can eventually vanish. One may expect that the eigenstate \(H_0+V\) can be obtained as the limit of an eigenstate \(H_0\) evolved according to the time-dependent Hamiltonian \(H_0+e^{-\varepsilon|t|}V\) when \(\varepsilon\rightarrow0\), but this naive approach is not justified because the time-dependent eigenstate has no limit for \(\varepsilon\rightarrow0\). Partially this problem was solved by Gell-Mann and Low for cases when the initial state belongs to a non-degenerate eigenspace -- this is done by the Gell-Mann and Low wavefunction. But there is no generalization of this formula for a degenerated eigenvalue of \(H_0\). In this paper the authors are able to show that the switching can be performed provided the initial eigenstates are also eigenstates of \({\mathcal P}_0V{\mathcal P}_0|_{\varepsilon_0}\), the perturbation restricted to act on the degenerate eigenspace. The paper consists of three sections. In the second one we have the statement of the results starting with the approach based on the spectrum given by \(\widetilde{H}(\lambda)=H_0+\lambda H\) and the projectors \(\widetilde{P}_j(\lambda)\) associated with the \(N\) eigenvalues \(\widetilde{E}_j(\lambda)\) used as a part of the discrete spectrum \(\widetilde{H}(\lambda)\) for the case when the initial state is degenerate. The sequel \({\mathcal P}_0\) is defined as \({\mathcal P}_0=\sum_{j=1}^M\widetilde{P}_j(0)\). It allows to obtain the operator \({\mathcal P}_0V{\mathcal P}_0\), which is used for Theorem 7 that is a main result of the paper. In the third section we have the proof of the results which is based on the following steps: {\parindent=4mm \begin{itemize}\item[--] the usage of Kato geometric evolution; \item[--] an explicit description of initial subspaces in terms of the eigenvectors of \({\mathcal P}_0V{\mathcal P}_0|_{\varepsilon_0}\); \item[--] the limit of the full evolution of \(U_{\text{int}}^\varepsilon\) (as the unitary evolution generated by \(H(\varepsilon s)\)) related to the geometric evolution as \(\varepsilon\rightarrow0\) which is the adiabatic evolution. \end{itemize}}