Simulation of quantum effects along the channel of ultrascaled Si-based MOSFETs

Abstract

Quantum transport simulations, including phase-breaking scattering, are used to observe the transition from classical to quantum transport in ultrascaled Si and SiGe heterostructure MOSFETs in order to gauge the potential effectiveness of semiclassical and pure phase-coherent quantum transport models as this transition is approached. It is shown that semiclassical models of transport along the length of the channel (as opposed to normal to the channel, where the importance of quantum mechanical effects has long been recognized) may remain reliable for channel lengths down to roughly 10 nm and perhaps beyond, and likely more reliable at this point than phase-coherent quantum transport simulations even when much of the transport is coherent/ballistic. As coherent transport effects within the channel eventually do become significant for ballistic carriers, the phase-breaking scattering rate, itself, also becomes a nonlocal function of the carrier's kinetic energy placing further demands on simulation. Simulations also reaffirm that for injection into the channel, the modeling of quantum transport effects such as tunneling, particularly in Si-SiGe heterostructure MOSFET's, will be important in much longer devices. However, even for this purpose it may not be possible to neglect the effects of inelastic scattering that can provide additional tunneling "paths.".