We present the results of our experimental work concerning the sheet resistance, R,, temperature characteristics of resistance and relative power spectral density of l/f noise, S. versus the volume fraction ofwnductingmmponent , U , for RuOrglassmmposites.We eliminate the contributions of motact-film-resistive-~m interfaces to the measured characteristics. We find that the composites investigated can be mapped onto the threecomponent 3Drandomresistornetworl; (RRN) formedfrom wellconductingmetallic bonds, poorlyconductingmetal-insulator-meial (MIM) bondsand thosenotconducting. We provide the physical interpretation 01 the network's parameters: h,. h,, b , , i.e. the ratio of con. ductances of poorly and well conducting bonds, the ratio of their l/f noise relative power spectral densities and the fraction of metallic bonds in the set of all conducting bonds, respectively. The electrical transport characteristics of RuOrglass thick resistive films are interpreted with the help of bicritical behaviour of such a percolation network. We find qualitative agreement between S(R,) experimental data for a certain region of U and that from 3D RRN computer simulations using the Monte Carlo real space renormalization group algorithm.
The behavior of 1/f noise effective intensity in two-phase percolation systems and percolationlike systems with an exponentially wide distribution of bond resistances is reviewed. Monte Carlo simulations on random resistor networks are performed. For a two-phase system the numerical values of noise critical exponents ϭ1.54Ϯ0.025, Јϭ0.61Ϯ0.02, wϭ6.31Ϯ0.25, and wЈϭ6.9Ϯ0.25 are found in agreement with theoretical analysis performed with the help of a hierarchical model of a two-phase percolation system. For a system with an exponentially wide spectrum of bond resistances, i.e., a system in which bonds take on resistances rϭr 0 exp͑Ϫx͒, where ӷ1 and x is a random variable, it is assumed that in the individual resistors the noise generating mechanism obeys the form ͕␦r 2 ͖ϳr 2ϩ . In this case the effective noise intensity C e ϵS⍀, where S is the relative power spectral density of system resistance fluctuations and ⍀ is the system volume, is given by C e ϳ m exp͑Ϫx c ͒, where 1Ϫx c is the percolation threshold. The exponent m is ''double universal,'' i.e., it is independent of lattice geometry and of the microscopic noise generating mechanism. Numerical simulations performed for ϭ1 and 0 give approximately mХ2.3 and confirm this ''double universality'' of the exponent m. The connections between 1/f noise effective intensity and effective susceptibility in a two-phase weakly nonlinear percolation system are also established. ͓S1063-651X͑96͒04405-4͔ PACS number͑s͒: 64.60.Ak, 64.60.Ht * Electronic address: snar@carrier.kiev.ua
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.