Statistical analysis of the impact of charge traps in p -type MOSFETs via particle-based Monte Carlo device simulations
- PDF / 1,809,330 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 40 Downloads / 139 Views
Statistical analysis of the impact of charge traps in p‑type MOSFETs via particle‑based Monte Carlo device simulations Alan C. J. Rossetto1 · Vinicius V. A. Camargo1 · Thiago H. Both2 · Dragica Vasileska3 · Gilson I. Wirth4
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In this paper, statistical analysis of the static impact of charge traps on the drain current of p-type metal–oxide–semiconductor field-effect transistors is presented. The study was carried out by employing a 3-D particle-based Monte Carlo device simulator, which is capable of accounting for the interplay between charge traps and the random dopant fluctuation effect. It was observed that the impact of a single charged trap on the transistor’s on-current is strongly dependent on the trap position along the channel length, on trap depth into the gate oxide, and on the trap position along the channel width. The current deviation estimated from statistical simulations is shown to be exponentially distributed, in agreement with experimental data from the literature. Results are also compared with uniform channel theory predictions. Keywords BTI · Monte Carlo · RTN · Numerical simulation · Oxide trap
1 Introduction The increasing demand for high-performance, high-density, and low-power circuits has pushed the complementary metal–oxide–semiconductor (CMOS) technology deep into nanometer dimensions. For deeply scaled devices, the discrete and stochastic nature of dopants and defects is responsible for a significant device parameter spread, thus becoming an important reliability issue. Understanding and properly modeling these deviations is critical not only for proper circuit design, but also for correct yield estimation. Two well-known reliability issues in modern CMOS devices are the random dopant fluctuation (RDF) and the trap activity. The first is related to the random number and random positioning of the dopant atoms in the channel * Alan C. J. Rossetto [email protected] 1
Centro de Desenvolvimento Tecnológico, Universidade Federal de Pelotas, Pelotas, RS 96010‑610, Brazil
2
Centro de Engenharias, Universidade Federal de Pelotas, Pelotas, RS 96010‑020, Brazil
3
School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287‑5706, USA
4
Programa de Pós‑Graduação em Microeletrônica, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS 91501‑970, Brazil
region of the transistors [1–3], which translate into seemingly identical devices to exhibit very different electrical characteristics [4–6]. The latter is known to be responsible for current degradation due to bias temperature instability (BTI) [7–9] and the random telegraph noise (RTN) [7, 10, 11]. These effects are not independent, though the impact of a charged trap is directly affected by RDF as for a nonhomogeneous doping configuration, the trap may induce RTN with a larger magnitude [12, 13]. Trap activity is also known to interact with other reliability issues such as hot-carrier injection (
Data Loading...
