| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 1546516 | Physica E: Low-dimensional Systems and Nanostructures | 2009 | 8 Pages |
We explore the characteristics of a new signal processing scheme based on the high-frequency electromechanical oscillations of a nanostructure formed by an oscillating metallic nanoparticle connected to the left and right electrodes by soft links. Because this system shows resonant behavior when the frequency of the applied electric potential is close to the characteristic natural frequency of the oscillating nanoparticle, a parallel arrangement of nanostructures with different frequencies can be excited selectively by an external time-dependent electrical signal with the appropriate resonant frequencies. The highly nonlinear system response makes it possible to devise a signal processing scheme, because those oscillators excited at resonance deliver a higher net current than those that are out of resonance. This property can be used to identify an input pattern codified in the external electrical signal. We discuss briefly the dynamics of the oscillating nanoparticle, paying special attention to the tunneling transitions to and from the electrodes at non-zero temperatures. We consider a device composed of N oscillators (one per bit of information) with different natural frequencies that are excited individually by the external electrical signal. As a proof-of-the-concept, we calculate the net charge delivered by each of the oscillators during a reference time and discuss theoretically the performance of the device in terms of the temperature, reference time, applied voltage, tunnel resistance asymmetry, noise, and damping parameter. In addition, we consider the case of an induced natural frequency by using a ligand whose conductance can be modulated externally. By imposing a periodic modulation of this conductance, a resonance of the charge delivered is observed, which can also be exploited to implement the signal processing device.
