Resistance as a function of temperature and applied magnetic field for doped manganese perovskites is simulated on the basis of a random-resistor-network model. The network, consisting of ferromagnetic metallic particles with number density p and paramagnetic insulating particles with number density 1-p, is generated through the Monte Carlo method. Approximating p by the reduced magnetization (m) determined from the mean-field theory, we show that the simulation can yield the main features of the colossal magnetoresistance in doped manganese perovskites. Comparisons between simulated and experimental data are also presented for (La1-xYx)2/3Ca1/3MnO3 (x = 0.2). The excellent agreement between the simulations and experiments gives strong support to the present approach.

Investigations of structural and transport features of La 1−x Ba x MnO 3 with x = 1/3 and 2/3 are performed. The x = 2/3 sample is shown to be phase separated into a mixture of La 2/3 Ba 1/3 MnO 3 and BaMnO 3 with the same volume fractions, ∼50%. In this two-phase system, La 2/3 Ba 1/3 MnO 3 regions are connected in a percolative manner, so the electrical transport is dominated by flow along these percolative paths. Using the recently proposed random resistor network based on electronic phase separation between ferromagnetic metallic and paramagnetic insulating domains, we show that the model can yield results in quantitative agreement with the resistance versus temperature dependence measured for the x = 1/3 and 2/3 samples by using the metallic number density as a fitting parameter. This approach suggests a simple quantitative picture that can be used to explain the insulator-metal behaviour in La 1−x Ba x MnO 3 with higher x.

No abstract