66 67 75 78 78 80 85 3.1 3.2 3.3 89 90 4.2 4.3 4.4.1 4.4.2 J =I-0, s =I-0 e 10 = 0 J =I-0, s = 0 e 1 0 =I-0 104 104 106 5 Conclusoes e sugestoes Apimdices B Justificativa do comportamento oscilatorio na curva da suscetibilidade magnetica da impureza 122 4.3 Diagrama de fluxo de energia para 0 Hamiltoniano de Falicov, Kimball e Ramirez de duas impurezas considerando J = -0.08, A = 10, 1 0 = a e s = 0.0096 para itera~oes (a) pares e (b) impares. . . . . . . . . . . . . . .. 107 4.4 Curvas da

more »

... tibilidade de carga da impureza do Hamiltoniano de Falicov, Kimball e Ramirez de duas impurezas considerando J = -0.08, A = 10. s = 0 e diferentes valores da intera<;ao RI\I\Y: (a) 10 = 0, (b) 1 0 = lac = -2.2T J \·, de duas impurezas considerando J = -0.08, A = 10, S = 0 e (a) 1 0 = 80T K (Efeito Kondo), (b) 10= -16TK (Intera<;ao RI\I\Y antiferromagnetica) ... 112 use SErNIC;:O DC: BIBLI~)TEC/.'l,. E )!'~FOR\/ AC;AO RI< I 1, equal to the ratio of two sucessive discrete energies. Once the conduction Hamiltonian is discretized, the model Hamiltonian reduces to a discrete series that can be diagonalized numerically. The computational cost of the diagonalization diminishes exponentially with 1/ In A, which makes it attractive to work with large A. unfortunately, the thermodynamical averages computed with Wilson's original version of the numerical renormalization group method and large A, computed as function of the temperature, display artificial oscilations with period In A and amplitude proportional to e-1r2 / lnA .