Natalia Seoane

Advanced simulation of statistical variability and reliability in nano CMOS transistors

Increasing CMOS device variability has become one of the most acute problems facing the semiconductor manufacturing and design industries at, and beyond, the 45 nm technology generation. Most problematic of all is the statistical variability introduced by the discreteness of charge and granularity of matter in transistors with features already of molecular dimensions [i]. Two transistors next to each other on the chip with exactly the same geometries and strain distributions may have characteristics from each end of a wide statistical distribution. In conjunction with statistical variability [ii], negative bias temperature instability (NBTI) and/or hot carrier degradation can result in acute statistical reliability problems. It already profoundly affects SRAM design, and in logic circuits causes statistical timing problems and is increasingly leading to hard digital faults. In both cases, statistical variability restricts supply voltage scaling, adding to power dissipation problems [iii]. In this invited paper we describe recent advances in predictive physical simulation of statistical variability using drift diffusion (DD), Monte Carlo (MC) and quantum transport (QT) simulation techniques.

3-D Nonequilibrium Green's Function Simulation of Nonperturbative Scattering From Discrete Dopants in the Source and Drain of a Silicon Nanowire Transistor

As these As transistors are scaled to nanometer dimensions, the discreteness of the dopants becomes increasingly important. Transistors of 3 times 3 nm2 cross section contain, on average, approximately one dopant atom per nanometer of length, making any self-averaging impossible. The individual random dopants act as localized scatterers whose distribution, and therefore, impact on the electron transport, varies from device to device. This is complemented by electrostatic variation in the potential that controls the threshold voltage and the dominant current paths. The current density is greatly influenced by resonances associated with the attractive potential of the donors and screening effects. In this paper, for the first time, a full 3-D nonequilibrium Greens function (NEGF) simulation in the effective mass approximation has been used to study the influence of individual discrete donors in the source/drain on the I-V characteristics of a narrow n-channel Si nanowire transistor. We have compared devices with microscopically different configuration of dopants. The simulated variations in the I-V curves are analyzed with reference to the behavior of the transmission coefficients. We have highlighted the importance of resonance states when solving the NEGF and Poisson equations self-consistently.

A detailed coupled-mode-space non-equilibrium Green's function simulation study of source-to-drain tunnelling in gate-all-around Si nanowire metal oxide semiconductor field effect transistors

In this paper we present a 3D quantum transport simulation study of source-to-drain tunnelling in gate-all-around Si nanowire transistors by using the non-equilibrium Greens function approach. The impact of the channel length, device cross-section, and drain and gate applied biases on the source-to-drain tunnelling is examined in detail. The overall effect of tunnelling on the ID-VG characteristics is also investigated. Tunnelling in devices with channel lengths of 10u2009nm or less substantially enhances the off-current. This enhancement is more important at high drain biases and at larger cross-sections where the sub-threshold slope is substantially degraded. A less common effect is the increase in the on-current due to the tunnelling which contributes as much as 30% of the total on-current. This effect is almost independent of the cross-section, and it depends weakly on the studied channel lengths.

A detailed 3D-NEGF simulation study of tunnelling in n-Si nanowire MOSFETs

In this work we have computed the effect of tunnelling on the on- and off-currents for two different cross-sections (2.2×2.2 & 4.2×4.2 nm²) and for five different channel lengths (4, 6, 10, 12, 14 nm). The simulated Si NWT transistors have undoped channels, 0.8 nm SiO<inf>2</inf> oxide and 9 nm source/drain regions doped at 10²⁰cm⁻³. All the simulations in this work have been performed at room temperature. Two different drain voltages have been consider for which full Id-Vg characteristics were simulated. Due to the change in cross-section, thickness dependent effective masses from sp³d⁵ second-neighbour-basis tight-binding calculations [6] are used in our simulations.

Perturbative vs Non-Perturbative impurity scattering in a narrow Si Nanowire GAA transistor: A NEGF study

In this paper we study the effect of impurity scattering on the performance of a Si gate-all-around nanowire transistors. The non-equilibrium Green function formalism is used in order to describe the carrier transport. Impurity scattering is introduced using two different formalisms, one that considers the impurity potential as a small perturbation by introducing self energies and the other in which the impurity potential is described exactly by included the impurity potential through the Poisson equation. The non-perturbative method does not use a fitting parameter but the perturbative one uses a phenomenological constant that can be calibrated to match the result of the non-perturbative method. We confirm Ohms-law-type behaviour by using the perturbative approach for larger channel lengths.

Dopants and roughness induced resonances in thin Si nanowire transistors: A self-consistent NEGF-poisson study

Non-Equilibrium Green Function simulations of the effect of discrete ionised dopants and surface roughness in Silicon nanowire transistors show the strong presence of resonances in the transmission coefficients. These resonances or quasi-bound states are the main component in the screening of dopants and play an important role in the current flow. Resonances appear through a self-consistent calculation of the electron density and potential. In this work we study several examples that exhibit different types of resonances. We start with a single impurity case and gradually evolved to a complex case with several random impurities. Interface roughness in a very narrow nanowire could induce resonant cavities as is proven in this paper. The effect of these resonances in the current-voltage characteristic of the transistors is considered in detail.

3D NEGF simulation of ‘ab initio’ scattering from discrete dopants in the source and drain of a nanowire transistor

In this paper we study the fluctuations in the I-V characteristics due to coherent scattering off different configurations of discrete dopants in the source/drain of an extremely thin Si nanowire transistor. The self-consistent screening potential at the impurities is calculated using a full 3D NEGF formalism coupled to the 3D solution of the Poisson equation [3]. The fluctuations in the I-V characteristics can be understood by examining the energy dependence of the transmission coefficients. We show that the effective potential presented to electrons by a positively charged ionised donor does not necessarily have the simple form of an attractive screened Coulomb potential. Instead, the self-consistent electrostatics leads to an effective potential that resembles an inverted sombrero: the core is Coulomb like, but the radial dependence rises through a maximum before flattening to an asymptotically zero value. This maximum introduces quasi-bound (resonant tunnelling) states. For the first time, we demonstrate the effects of the sombrero potential on scattering and ultimately device current transmission. The whole screening occurs due to electrons congregating in the quasi-bound states of the sombrero potential

Current variations in pHEMTs introduced by channel composition fluctuations

We have studied the impact of the channel composition fluctuations in pseudomorphic high electron mobility transistors (PHEMTs) on the device characteristics. The simulations are carried out using a 3D parallel finite element device simulator based on the drift-diffusion approximation to the semiconductor transport. The variation of the drain current introduced by the channel composition fluctuations increase with the increase of the gate and the drain voltage.