Statistical analysis of the impact of charge traps in p-type MOSFETs via particle-based Monte Carlo device simulations

Rossetto, Alan Carlos Junior; Camargo, Vinícius V. A.; Both, Thiago Hanna; Vasileska, Dragica; Wirth, Gilson

doi:10.1007/s10825-020-01478-6

Cited by 10 publications

(9 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Black squares show the value obtained from the Monte Carlo simulations. Red dots show as predicted by (9). Very good agreement between analytical model and Monte Carlo simulations is seen.…”

Section: Monte Carlo Simulationssupporting

confidence: 60%

“…Please note that the moments of an exponentially distributed random variable X are given by <X n > = n!/λ n , with <X>=1/λ being the parameter of the exponential distribution [6]. This allows evaluation of equations (7) and (9) and comparison to the results from Monte Carlo simulations.…”

Section: Monte Carlo Simulationsmentioning

confidence: 99%

“…It is also important to address the question of how the VT jitter varies among devices that by design should be identical (have the same W x L). This question is addressed by looking at the variance of VTjitter 2 , which is , given by equation (9). Figure 5 shows how the VT jitter varies among transistors that by design should be identical.…”

Section: Monte Carlo Simulationsmentioning

confidence: 99%

“…The average amplitude to a trap is known to scale inversely with area, Ai~1/WL [1, 3, 7 and 8]. From (7) following relation between device area and average expected jitter value can then be written 𝑉 = 𝜎 ~1 𝑊𝐿 (10) For the variation of jitter among devices, from (9) following relation between device area and jitter variability among devices can be written 𝜎 ~1 (𝑊𝐿)…”

Section: A Model Derivationmentioning

confidence: 99%

See 3 more Smart Citations

Time Dependent Threshold Voltage Variability due to Random Telegraph Noise

Wirth

2020

2020 IEEE Latin America Electron Devices Conference (LAEDC)

View full text Add to dashboard Cite

Abstract-With the downscaling of MOSFETs to nanometer dimensions, transistor electrical parameter variability is produced by factors other than variations of physical dimensions and doping profiles, which are there since device fabrication and remain static over time. Besides these time-zero variability factors, factors that lead to performance variability from one instant in time to the other start playing a significant role. Random Telegraph Noise (RTN) is among these relevant time-dependent variability sources. In this work we extend the knowledge of the time-dependent random variability induced by RTN, by providing a statistical model for transistor threshold voltage jitter produced by RTN. The area scaling of is detailed and discussed, supporting designers in transistor sizing towards a more reliable design. Not only the jitter expected in a transistor is modeled, but also its variability among transistors that by design should be equal. Besides analytical modeling, Monte Carlo simulations are run. The simulations account for the charge carrier capture and emission events related to RTN, allowing the proper evaluation of the RTN related jitter. The Monte Carlo simulations validate the analytical model and illustrate the area scaling of jitter and its variability. IndexTerms-metal-oxide-semiconductor field-effect transistor (MOSFET) scaling, random telegraph noise (RTN), timing jitter, time-dependent variability.

show abstract

Section: Monte Carlo Simulationssupporting

confidence: 60%

Section: Monte Carlo Simulationsmentioning

confidence: 99%

Section: Monte Carlo Simulationsmentioning

confidence: 99%

Section: A Model Derivationmentioning

confidence: 99%

See 2 more Smart Citations

Time Dependent Threshold Voltage Variability due to Random Telegraph Noise

Wirth

2020

2020 IEEE Latin America Electron Devices Conference (LAEDC)

View full text Add to dashboard Cite

show abstract

“…Additionally, seemly identical device may exhibit a different number of traps (assumed to be Poisson distributed) and the properties of each trap are random (e.g. ∆I d is considered exponentially distributed [10]- [13] and τ log-uniformly distributed). Hence, each device has a different RTN behavior.…”

Section: Random Telegraph Noisementioning

confidence: 99%

Random Telegraph Noise Modeling for Circuit Analysis: RTN in Ring Oscillators

Silva

Both

Wirth

2022

IEEE J. Electron Devices Soc.

Self Cite

View full text Add to dashboard Cite

In highly scaled MOSFETs, random telegraph noise (RTN) can decrease the reliability and yield of circuits. RTN is produced by charge trapping, which in large devices results in 1/f noise. We derived analytical formulations for modeling the impact of RTN in the delay of inverters and in the jitter of ring oscillators. We show that the parameters of interest when characterizing RTN for circuit analysis are the distribution of current deviations and the density of traps in the space of area, energy and in log-space of time-constants. The model gives a direct relation between jitter variance in oscillators (or delay variance in inverters) and the power spectral density of RTN, which includes 1/f noise. The formulations can be written using time-or frequency-domain parameters.

show abstract