M. Waltl

A unified perspective of RTN and BTI

It has recently been suggested that random telegraph noise (RTN) and the bias temperature instability (BTI) are due to similar defects. Here we thoroughly analyze this hypothesis using nano-scale devices to show that (i) all defects that contribute to BTI recovery can also become spontaneously charged to produce an RTN event, (ii) most RTN defects also contribute to BTI recovery, (iii) the distribution of step-heights, capture and emission times is equally wide and similar for RTN and BTI, and (iv) both RTN and BTI defects are volatile, meaning that they can disappear and reappear. From these observations we conclude that RTN and the recoverable component of BTI are very likely due to the same defects. As a very important consequence, RTN and BTI must be analyzed and guardbanded against together. In particular, since conventional RTN analysis dominantly captures defects with the strongest contribution to the noise power, it misses the defects with large capture times. As we will show, however, it is exactly these defects with large capture times that may by chance become occupied at the same time after long times, thereby leading to very large NBTI-like threshold voltage fluctuations in an RTN setting. Conversely, conventional BTI analysis based on the expectation value of the stochastic trap behavior misses these RTN-like fluctuations when extrapolated down to operating voltages, potentially leading to wrong conclusions.

Advanced characterization of oxide traps: The dynamic time-dependent defect spectroscopy

An accurate understanding of oxide traps is essential for a number of reliability issues, including the bias temperature instability, hot carrier degradation, time-dependent dielectric breakdown, random telegraph and 1/f noise. Recent results have demonstrated that hole capture and emission into oxide traps in pMOS transistors are more complicated than the usually assumed Shockley-Read-Hall-like process. In particular, both charging and discharging proceed via a non-radiative multiphonon (NMP) mechanism involving metastable defect states. The existence of these metastable states can be demonstrated by extending the previously introduced time-dependent defect spectroscopy (TDDS) to a more general dynamic case by employing AC stresses and precisely timed discharge pulses during recovery. Application of AC stresses clearly reveals a frequency-dependence of the effective capture time, which confirms the existence of an intermediate metastable state. Application of pulses during recovery, on the other hand, allows extraction of the effective emission time also in depletion as well as in accumulation, thereby clearly revealing a metastable switching state. While all investigated traps show a frequency-dependent capture time constant, suggesting them to be of the same microscopic origin, we find two different kinds of emission behavior, namely fixed positive and switching traps. We finally demonstrate that our multi-state NMP model perfectly captures both cases.

NBTI in Nanoscale MOSFETs—The Ultimate Modeling Benchmark

After nearly half a century of research into the bias temperature instability, two classes of models have emerged as the strongest contenders. One class of models, the reaction-diffusion models, is built around the idea that hydrogen is released from the interface and that it is the diffusion of some form of hydrogen that controls both degradation and recovery. Although various variants of the reaction-diffusion idea have been published over the years, the most commonly used recent models are based on nondispersive reaction rates and nondispersive diffusion. The other class of models is based on the idea that degradation is controlled by first-order reactions with widely distributed (dispersive) reaction rates. We demonstrate that these two classes give fundamentally different predictions for the stochastic degradation and recovery of nanoscale devices, therefore providing the ultimate modeling benchmark. Using detailed experimental time-dependent defect spectroscopy data obtained on such nanoscale devices, we investigate the compatibility of these models with experiment. Our results show that the diffusion of hydrogen (or any other species) is unlikely to be the limiting aspect that determines degradation. On the other hand, the data are fully consistent with reaction-limited models. We finally argue that only the correct understanding of the physical mechanisms leading to the significant device-to-device variation observed in the degradation in nanoscale devices will enable accurate reliability projections and device optimization.

Long-Term Stability and Reliability of Black Phosphorus Field-Effect Transistors

Black phosphorus has been recently suggested as a very promising material for use in 2D field-effect transistors. However, due to its poor stability under ambient conditions, this material has not yet received as much attention as for instance MoS2. We show that the recently demonstrated Al2O3 encapsulation leads to highly stable devices. In particular, we report our long-term study on highly stable black phosphorus field-effect transistors, which show stable device characteristics for at least eight months. This high stability allows us to perform a detailed analysis of their reliability with respect to hysteresis as well as the arguably most important reliability issue in silicon technologies, the bias-temperature instability. We find that the hysteresis in these transistors depends strongly on the sweep rate and temperature. Moreover, the hysteresis dynamics in our devices are reproducible over a long time, which underlines their high reliability. Also, by using detailed physical models for oxide traps developed for Si technologies, we are able to capture the channel electrostatics of the black phosphorus FETs and determine the position of the defect energy band. Finally, we demonstrate that both hysteresis and bias-temperature instabilities are due to thermally activated charge trapping/detrapping by oxide traps and can be reduced if the device is covered by Teflon-AF.

Hydrogen-related volatile defects as the possible cause for the recoverable component of NBTI

The recently suggested time-dependent defect spectroscopy (TDDS) has allowed us to study the recoverable component of NBTI at the single-defect level. To go beyond our previous efforts, we have performed a long-term TDDS study covering also the kilo-second time window. We found that even in this extended window NBTI recovery is due to a collection of first-order reactions. In particular, there is no trace of a diffusion-limited process as assumed in the reaction-diffusion model. Most intriguingly, the responsible traps show various degrees of volatility, that is, they can disappear and reappear. Our observations lend strong support to the idea that the recoverable component of NBTI is due to hydrogen-related defects which are active when a hydrogen atom is at the defect site and inactive when not.

The role of charge trapping in MoS2/SiO2 and MoS2/hBN field-effect transistors

The commonly observed hysteresis in the transfer characteristics of MoS2 transistors is typically associated with charge traps in the gate insulator. Since in Si technologies such traps can lead to severe reliability issues, we perform a combined study of both the hysteresis as well as the arguably most important reliability issue, the bias-temperature instability. We use single-layer MoS2 FETs with SiO2 and hBN insulators and demonstrate that both phenomena are indeed due to traps in the gate insulator with time constants distributed over wide timescales, where the faster ones lead to hysteresis and the slower ones to bias-temperature instabilities. Our data show that the use of hBN as a gate insulator considerably reduces the number of accessible slow traps and thus improves the reliability. However, the reliability of hBN insulators deteriorates with increasing temperature due to the thermally activated nature of charge trapping.

Gate-sided hydrogen release as the origin of "permanent" NBTI degradation: From single defects to lifetimes

The negative bias temperature instability (NBTI) in pMOS transistors is typically assumed to consist of a recoverable (R) and a so-called permanent (P) component. While R has been studied in great detail, the investigation of P is much more difficult due to the large time constants involved and the fact that P is almost always obscured by R. As such, it is not really clear how to measure P and whether it will in the end dominate device lifetime. We address these questions by introducing a pragmatic definition of P, which allows us to collect long-term data on both large and nanoscale devices. Our results suggest that (i) P is considerably smaller than R, (ii) that P is dominated by oxide rather than interface traps and therefore (iii) shows a very similar bias dependence as R, and finally (iv) that P is unlikely to dominate device lifetime. We argue that a hydrogen-release mechanism from the gate-side of the oxide, which has been suspected to cause reliability problems for a long time [1-6], is consistent with our data. Based on these results as well as our density-functional-theory (DFT) calculations we suggest a microscopic model to project the results to operating conditions.

The “permanent” component of NBTI revisited: Saturation, degradation-reversal, and annealing

While the defects constituting the recoverable component R of NBTI have been very well analyzed recently, the slower defects forming the more “permanent” component P are much less understood. Using a pragmatic definition for P, we study the evolution of P at elevated temperatures in the range 200°C to 350°C to accelerate these very slow processes. We demonstrate for the first time that P not only clearly saturates, with the saturation value depending on the gate bias, but also that the degradation at constant gate bias can also slowly reverse. Furthermore, at temperatures higher than about 300° C, a significant amount of additional defects is created, which are primarily uncharged around Vth but contribute strongly to P at higher VG. Our new data are consistent with our recently suggested hydrogen release model which will be studied in detail using newly acquired long-term data.

Microscopic oxide defects causing BTI, RTN, and SILC on high-k FinFETs

Reliability issues of MOSFETs such as bias temperature instability (BTI), random telegraph noise (RTN), and stress-induced leakage current (SILC), are linked to the trapping of charges in oxides. Even though the chemical structure of these oxide defects is still the subject of debate, detailed studies of these reliability phenomena have shown that their physical behavior can be successfully described by non-radiative multi-phonon (NMP) theory. In this work we characterize and study a pMOS high-k FinFET technology starting from degradation measurements up to the simulation of the energy barriers in the framework of NMP theory. This allows to investigate the aforementioned reliability issues all based on their common cause, the microscopic oxide defects.

Advanced data analysis algorithms for the time-dependent defect spectroscopy of NBTI

In order to identify the physical mechanisms behind the negative bias temperature instability (NBTI), the time-dependent defect spectroscopy (TDDS) has been recently proposed. The TDDS takes advantage of the fact that in nano-scaled devices only a handful of defects are present. As a consequence, degradation and recovery proceed in discrete steps, each of them corresponding to a charge capture or emission event. By repeatedly applying stress and recovery conditions, the TDDS analyzes the statistical properties of these discrete events. The measurement window of the TDDS is very large, but the occurrence of random telegraph noise (RTN) at certain biases/temperatures can limit its applicability. We have developed an advanced data analysis method which can also deal with data contaminated by RTN. The algorithm is based on the combination of a bootstrapping technique and cumulative sum charts. A benefit of the new method is the possibility to detect steps in a large class of different signals with a feasible amount of parameters. Moreover, de-/trapping parameters of the random telegraph noise (RTN) become accessible as well.