A noise temperature analysis of the electrical degradation of thin nanostructured films

Thermal noise measurements at proper biasing conditions are shown to represent a powerful tool for the characterization of the homogeneity of thin nanostructured films and their adhesion to the substrate. By modeling a thinfilm as a two-dimensional random resistor network, we introduce a new type of excess-noise arising from local sources of Nyquist noise due to the presence of defective regions. The dishomogeneous Joule heating of the film is responsible for a thermal and electrical instability which is efficiently described by using a biased percolation model. The results of our simulations show that the Nyquist excess-noise temperature should provide a sensitive and non-destructive indicator of the packing density and of the quality of heat contact to the substrate of nanostructured films with grain size in the range 10 –500 nm.