Modeling Information Processing Using Nonidentical Coulomb Blockade Nanostructures

In recent years, molecular-protected metallic nanoparticles (NPs) have attracted a great deal of attention. Because of their reduced size, they behave like tiny capacitors so that there is an energy penalty when adding an electron to the NP which suppresses the electric current at a potential lower than a threshold value. This phenomenon is known as Coulomb blockade (CB) and allows the transport of electrons to be modulated through an external gate provided that the energy penalty is higher than the thermal energy. Together with the possibility of tailoring their properties, molecular protected NPs are potential candidates as future components of high density, low consumption electronics. However, they face a number of problems before they can be considered as a technological viable option. To be used at room temperatures, NPs radii need to be in the nanometer range, and then fabrication processes lead to significant variability in the NPs physical properties. We use here two systems, a XOR gate and a R-SET model which mimics some characteristics of neurons, to show strategies that may be used to cope with the variability problem so that a robust information processing can be achieved despite using nominally different components.