The one-dimensional SU
(n) Hubbard model is investigated using an analytical, perturbative renormalization-group treatment in the fermionic formulation of the model for
n>2. It is found that in the half-filled case, the umklapp processes couple the spin and charge degrees of freedom and generate a finite gap for arbitrary finite positive
U. This behavior is thus different from that at
1∕n filling, where there is one particle per site, in which case the dynamics of the two degrees of freedom can be separated and a critical
Uc is needed for the Mott transition.
We theoretically investigate the dynamics of insulator-metal transition induced by light-pulse excitation in one-dimensional (1D) and two-dimensional (2D) Mott insulators. We adopt the Pariser-Parr-Pople model, and pump-probe signal is obtained by numerically calculating the time development of the state excited by a light pulse. In the 1D case, when photoexcitation density is below about 10%, an extremely small portion of the energy eigenstates dominates the optical process, and the spin-charge separation holds in these dominant energy eigenstates. As a result, the Mott gap and the short-range antiferromagnetic (AF) spin order are preserved, and the spin relaxation does not occur. When the density is above the value, a metallic state without the Mott gap is photogenerated, and the magnitude of the short-range AF spin order is significantly reduced by photoexcitation. This is consistent with the experimentally observed photoinduced Mott transition. Furthermore, the spin relaxation occurs in the metallic state. This photoinduced Mott transition is a manifestation of the spin-charge coupling in the intensely photoexcited states. In the 2D case, the spin and the charge degrees of freedom are coupled even in the weakly photoexcited states, and the characteristic crossover associated with the excitation density is not observed.