Physics – Condensed Matter
Scientific paper
1995-08-16
Phys. Rev. B 52 (95) 13707
Physics
Condensed Matter
4 pages, accepted for publication in Phys. Rev. B
Scientific paper
10.1103/PhysRevB.52.13707
We consider the $N$-component generalization of the four-slave-boson approach to the Hubbard model, where $1/N$ acts as the small parameter that controls the fluctuations about the saddle point, and address the problem of the appropriate choice of the bosonic hopping operator $z_{i\g}$. By suitably reorganizing the Fock space, we show that the square-root form for $z\is$ (originally introduced by Kotliar and Ruckenstein) reproduces the exact independent-fermion ($U=0$) results not only at the mean-field ($N=\infty$) level but also to {\evi all orders in the $1/N$ expansion}, provided one relaxes the usually adopted normal-ordered prescription for $z\is$. This ensures that $z\is$ needs not be modified at successive orders in the fluctuation expansion, and implies that all correlation functions are correctly recovered in the $U=0$ limit, a nontrivial result for the slave-boson approach. In addition, it provides a stringent requirement on the form of $z\is$, which might be also generalized to alternative slave-boson formalisms (like the spin-rotation-invariant formulation).
Arrigoni Enrico
Strinati Giancarlo Calvanese
No associations
LandOfFree
On an exact criterion for choosing the hopping operator in the four-slave-boson approach does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with On an exact criterion for choosing the hopping operator in the four-slave-boson approach, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and On an exact criterion for choosing the hopping operator in the four-slave-boson approach will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-75739