Excess thermal-noise in the electrical breakdown of random resistor networks

Abstract:

We discuss a new type of excess noise strongly sensitive to non-homogeneous Joule heating of random resistor network and associated with local sources of thermal noise. The evolution of the network towards an electrical breakdown of conductor-insulator type is then studied by using a biased percolation model and it is analysed in terms of an excess-noise temperature. Monte Carlo simulation results show a significant increase of the excess-noise temperature over the average temperature of the network. Remarkably the excess-noise temperature scales with the resistance with an exponent of about 3. The predictivity of the model can be tested on thin film resistors where the determination of the excess noise temperature should provide a valuable indicator of the defectiveness of the film.