Gate tunable transport-gap to beat the Boltzmann limit: A graphene klein tunnel field effect transistor

With the impending end of Moore's Law and the abandonment of the ITRS roadmap, there is now a pressing need to explore new materials, architectures and possibly a new way of computation beyond Boolean logic. Of particular interest is the exploration of fundamental concepts that utilize novel physical mechanisms to beat the Boltzmann limit on the steepness of the gate transfer curve. Complimentary Metal Oxide Semiconductor (CMOS) technology is built around silicon based field-effect devices where the sub-threshold swing (k
_B
T ln 10/q ~ 60 mV/decade) is fundamentally limited by the tail of the Fermi-Dirac distribution of electrons in the contacts. To overcome this limit, new devices have emerged utilizing novel physical mechanisms such as - Tunnel FETs that abruptly open a tunneling channel in a p-i-n junction [1], negative capacitance based MOSFETs [2] that amplify voltage division across a regular oxide in series with a ferroelectric near transition, metal-to-insulator transition hyperFETs [3] that use opening of a Mott bandgap for voltage amplification at the source, NEMFETs that abruptly withdraw the channel from the drain end [4] and electrostrictive FET [5] that opens a physical gap with a piezoelectric field. All of these systems rely on depletion physics in addition to a gate enhancement of the transmission modes in the channel, except negative capacitance and hyperFETs where the enhancement happens externally at the voltage input. In this paper, we will argue how angular filtering in pristine graphene can produce a tunable transport-gap [6,7] which can in principle beat the Boltzmann limit over several decades while preserving its high mean-free path. We also discuss geometrical non-idealities as well as applications like RF devices that can still survive these non-idealities.