We calculate quantum transport for metal-graphene nanoribbon heterojunctions within the atomistic self-consistent Schrodinger/Poisson scheme. Attention is paid on both the chemical aspects of the interface bonding as well the one-dimensional electros
tatics along the ribbon length. Band-bending and doping effects strongly influence the transport properties, giving rise to conductance asymmetries and a selective suppression of the subband formation. Junction electrostatics and p-type characteristics drive the conduction mechanism in the case of high work function Au, Pd and Pt electrodes, while contact resistance becomes dominant in the case of Al.
We present a systematic study of electron backscattering phenomena during conduction for graphene nanoribbons with single-vacancy scatterers and dimensions within the capabilities of modern lithographic techniques. Our analysis builds upon an textit{
ab initio} parameterized semiempirical model that breaks electron-hole symmetry and nonequilibrium Greens function methods for the calculation of the conductance distribution $g$. The underlying mechanism is based on wavefunction localizations and perturbations that in the case of the first $pi-pi{}^*$ plateau can give rise to impurity-like pseudogaps with both donor and acceptor characteristics. Confinement and geometry are crucial for the manifestation of such effects. Self-consistent quantum transport calculations characterize vacancies as local charging centers that can induce electrostatic inhomogeneities on the ribbon topology.
We report the results of an analysis, based on a straightforward quantum-mechanical model, of shot noise suppression in a structure containing cascaded tunneling barriers. Our results exhibit a behavior that is in sharp contrast with existing semicla
ssical models for this particular type of structure, which predict a limit of 1/3 for the Fano factor as the number of barriers is increased. The origin of this discrepancy is investigated and attributed to the presence of localization on the length scale of the mean free path, as a consequence of the strictly 1-dimensional nature of disorder, which does not create mode mixing, while no localization appears in common semiclassical models. We expect localization to be indeed present in practical situations with prevalent 1-D disorder, and the existing experimental evidence appears to be consistent with such a prediction.
In this paper we investigate Warm Electron Injection as a mechanism for NOR programming of double-gate SONOS memories through 2D full band Monte Carlo simulations. Warm electron injection is characterized by an applied VDS smaller than 3.15 V, so tha
t electrons cannot easily accumulate a kinetic energy larger than the height of the Si/SiO2 barrier. We perform a time-dependent simulation of the program operation where the local gate current density is computed with a continuum-based method and is adiabatically separated from the 2D full Monte Carlo simulation used for obtaining the electron distribution in the phase space. In this way we are able to compute the time evolution of the charge stored in the nitride and of the threshold voltages corresponding to forward and reverse bias. We show that warm electron injection is a viable option for NOR programming in order to reduce power supply, preserve reliability and CMOS logic level compatibility. In addition, it provides a well localized charge, offering interesting perspectives for multi-level and dual bit operation, even in devices with negligible short channel effects.
In this paper we investigate warm electron injection in a double gate SONOS memory by means of 2D full-band Monte Carlo simulations of the Boltzmann Transport Equation (BTE). Electrons are accelerated in the channel by a drain-to-source voltage VDS s
maller than 3 V, so that programming occurs via electrons tunneling through a potential barrier whose height has been effectively reduced by the accumulated kinetic energy. Particle energy distribution at the semiconductor/oxide interface is studied for different bias conditions and different positions along the channel. The gate current is calculated with a continuum-based post-processing method as a function of the particle distribution obtained from Monte Carlo. Simulation results show that the gate current increases by several orders of magnitude with increasing drain bias and warm electron injection can be an interesting option for programming when short channel effects prohibit the application of larger drain bias.
We present an atomistic three-dimensional simulation of graphene nanoribbon field effect transistors (GNR-FETs), based on the self-consistent solution of the 3D Poisson and Schroedinger equation with open boundary conditions within the non-equilibriu
m Greens Function formalism and a tight-binding hamiltonian. With respect to carbon nanotube FETs, GNR-FETs exhibit comparable performance, reduced sensitivity on the variability of channel chirality, and similar leakage problems due to band-to-band tunneling. Acceptable transistor performance requires effective nanoribbon width of 1-2 nm, that could be obtained with periodic etching patterns or stress patterns.
