Roberto Grassi

Multilayer Black Phosphorus as a Versatile Mid-Infrared Electro-optic Material.

We investigate the electro-optic properties of black phosphorus (BP) thin films for optical modulation in the mid-infrared frequencies. Our calculation indicates that an applied out-of-plane electric field may lead to red-, blue-, or bidirectional shift in BPs absorption edge. This is due to the interplay between the field-induced quantum-confined Franz-Keldysh effect and the Pauli-blocked Burstein-Moss shift. The relative contribution of the two electro-absorption mechanisms depends on doping range, operating wavelength, and BP film thickness. For proof-of concept, simple modulator configuration with BP overlaid over a silicon nanowire is studied. Simulation results show that operating BP in the quantum-confined Franz-Keldysh regime can improve the maximal attainable absorption as well as power efficiency compared to its graphene counterpart.

Optimization of n- and p-type TFETs Integrated on the Same

Design of a suitable technology platform is carried out in this paper for co-integration of simultaneously optimized n- and p-type tunnel field-effect transistors (TFETs). InAs/AlxGa1-xSb heterostructures are considered, and a 3-D full-quantum simulation approach is adopted to investigate the combined effect of Al mole fraction x and transverse quantization on band lineups at the heterojunction. Design optimization leads to a TFET pair with similar dimensions and feasible aspect ratios realized on the same InAs/Al0.05Ga0.95Sb platform. These devices exhibit average subthreshold slopes below 60 mV/dec and relatively high ON-currents of 270 (n-TFET) and 120 μA/μm (p-TFET) at a low-supply voltage VDD=0.4 V. Combined ON- and OFF-state performance of the proposed technology platform is expected to be compatible with low operating power applications, while potential candidates for low standby power scenarios are obtained by reducing TFET cross sections from 10 to 7 nm.

{\rm InAs}/{\rm Al}_{x}{\rm Ga}_{1-x}{\rm Sb}

Recent experiments show that a substantial energy gap in graphene can be induced via patterned hydrogenation on an iridium substrate. Here, we show that the energy gap is roughly proportional to N(H)(1/2)/N(C) when disorder is accounted for, where N(H) and N(C) denote concentrations of hydrogen and carbon atoms, respectively. The dispersion relation, obtained through calculation of the momentum-energy resolved density of states, is shown to agree with previous angle-resolved photoemission spectroscopy results. Simulations of electronic transport in finite size samples also reveal a similar transport gap, up to 1 eV within experimentally achievable N(H)(1/2)/N(C) values.

Technology Platform

For the first time, a simulation study is reported of a device formed by stacking an n+-Si layer (emitter), a monolayer graphene sheet (base), and a second n-Si layer (collector), operating as a graphene-base heterojunction transistor. The device differs from the recently proposed hot-electron graphene-base transistor (GBT), where graphene is sandwiched between the two dielectric layers, in the current flow being regulated mainly by thermionic emission over the potential-energy barrier, rather than by tunneling through the emitter-contact Schottky barrier. The simulations are based on a 1-D quantum transport model with the effective mass approximation and nonparabolic corrections. In addition to being much easier to fabricate compared with the GBT, the device is shown to be able to provide 104 ON/OFF current ratio, current densities well in excess of 0.1 A/μm2 and cutoff frequencies well above 1 THz, together with an intrinsic dc small-signal voltage gain larger than 10. Even though the simulation model is somewhat idealized, since ballistic transport is assumed and Si-graphene interfaces are ideal, our results show that this device is a serious competitor for high-frequency RF applications.

Scaling of the energy gap in pattern-hydrogenated graphene.

In this paper, we clarify the physical mechanism for the phenomenon of negative output differential resistance (NDR) in short-channel graphene FETs through nonequilibrium Greens function simulations and a simpler semianalytical ballistic model that captures the essential physics. This NDR phenomenon is due to a transport mode bottleneck effect induced by the graphene Dirac point in the different device regions, including the contacts. NDR is found to occur only when the gate biasing produces an n-p-n or p-n-p polarity configuration along the channel, for both positive and negative drain-source voltage sweep. In addition, we also explore the impact on the NDR effect of contact-induced energy broadening in the source and drain regions and a finite contact resistance.

Graphene-Base Heterojunction Transistor: An Attractive Device for Terahertz Operation

A simulation study aimed at investigating the main features in dc and small-signal operating conditions of the hot-electron graphene base transistor (GBT) for analog terahertz operation is presented. Intrinsic silicon is used as reference material. The numerical model is based on a self-consistent Schrödinger-Poisson solution, using a 1-D transport approximation and accounting for multiple-valley and nonparabolicity band effects. Some limitations in the extension of the saturation region and in the output conductance related to the finite quantum capacitance of graphene and to space charge effects are discussed. A small-signal model is developed that catches the essential physics behind the voltage gain and the cutoff frequency, which shows that the graphene quantum capacitance severely limits the former but not the latter. According to simulations carried out within the ballistic transport approximation, a 20-nm-long GBT can achieve at the same time a voltage gain larger than 10 and a cutoff frequency largely above 1 THz within a reasonably wide bias range.

Contact-Induced Negative Differential Resistance in Short-Channel Graphene FETs

A mode space (MS) tight binding approach for the simulation of armchair graphene nanoribbon FETs is discussed. It makes use of slab-dependent modes and a novel criterion for mode selection, going beyond the one based on the lowest energy subbands. For ideal ribbons, we show that by splitting the modes into decoupled groups, the new method provides results almost identical to the real space (RS) with a speedup of more than one order of magnitude. Even in the presence of edge roughness, which tends to couple the modes, the MS approach still offers a sizable computational advantage with respect to the RS, while retaining a good accuracy.

Graphene Base Transistors: A Simulation Study of DC and Small-Signal Operation

We show that a ballistic quantum transport model based on the effective mass approximation can fairly well describe the I-V characteristics of armchair carbon nanoribbon FETs, in all bias conditions, including the regimes dominated by direct or band-to-band-tunneling, provided first-order nonparabolic corrections are included in the simulation. This is achieved by means of an energy (position) dependent effective mass. The analysis is supported by comparisons with a full atomistic tight-binding model.

Mode Space Approach for Tight Binding Transport Simulation in Graphene Nanoribbon FETs

Through self-consistent quantum transport simulations, we evaluate the RF performance of monolayer graphene field-effect transistors in the bias region of negative output differential resistance. We show that, compared with the region of quasi-saturation, a voltage gain larger than 10 can be obtained, at the cost of a decrease in the maximum oscillation frequency of about a factor of 1.5-3 and the need for a careful circuit stabilization.

Tight-binding and effective mass modeling of armchair carbon nanoribbon FETs

The many unique properties of graphene, such as the tunable optical, electrical, and plasmonic response make it ideally suited for applications such as biosensing. As with other surface-based biosensors, however, the performance is limited by the diffusive transport of target molecules to the surface. Here we show that atomically sharp edges of monolayer graphene can generate singular electrical field gradients for trapping biomolecules via dielectrophoresis. Graphene-edge dielectrophoresis pushes the physical limit of gradient-force-based trapping by creating atomically sharp tweezers. We have fabricated locally backgated devices with an 8-nm-thick HfO2 dielectric layer and chemical-vapor-deposited graphene to generate 10× higher gradient forces as compared to metal electrodes. We further demonstrate near-100% position-controlled particle trapping at voltages as low as 0.45 V with nanodiamonds, nanobeads, and DNA from bulk solution within seconds. This trapping scheme can be seamlessly integrated with sensors utilizing graphene as well as other two-dimensional materials.The capability of positioning target molecules onto the edges of patterned graphene nanostructures is highly desirable. Here, the authors demonstrate that the atomically sharp edges of graphene can be used as dielectrophoretic tweezers for gradient-force-based trapping applications.