Capture Time Constants Research Articles

Unveiling the Switching Mechanism of a TaO<italic>x</italic>/HfO2 Self-Selective Cell by Probing the Trap Profiles With RTN Measurements

Self-selectivecells (SSCs) with built-in nonlinearity provide a promising solution for overcoming the leakage current issue of 3-D vertical resistive random access memory arrays. However, deep understanding of the switching mechanism of the SSC device is still lacking, hindering the effective improvement of the device performance. In this letter, we investigated the switching mechanism of the TaO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</sub> /HfO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> SSC device by probing the trap profiles with RTN measurements. Both the vertical locations and energy levels (X <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</sub> , E <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</sub> ) of defects in the HRS and LRS could be calculated by the bias-dependence of the capture and emission time constants (τ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</sub> and τ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">c</sub> ). Using the comparison of the defect profiles before and after the resistive switching, an active region in the TaO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</sub> layer adjacent to the HfO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> layer could be clearly identified. Based on these results, a clear picture of resistive switching in the TaO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</sub> /HfO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> SSC was obtained.

A Prepulsing Technique for the Characterization of GaN Power Amplifiers With Dynamic Supply Under Controlled Thermal and Trapping States

The asymmetry between capture and release time constants associated with charge-trapping phenomena observed in the electrical characteristics of microwave gallium–nitride (GaN) field-effect transistors (FETs) introduces distortion in GaN-based power amplifiers (PA). The PAs that operate with supply modulation to increase efficiency are particularly affected by this phenomenon, since the GaN FET trap state exhibits a nonlinear dependence on the voltage applied to the device terminals. In this paper, a measurement approach and the setup are presented for a large-signal characterization of GaN-based PAs operated with dynamic bias supply: a new prepulsing technique is introduced, which enables the characterization of the PA in controlled charge-trapping and thermal states. The characteristics obtained with this technique are shown to give an accurate description of the PA performance in the actual application working conditions. The proposed approach is validated by using the measured data for the direct computation of predistortion functions for the linearization of a 9.7-GHz envelope tracking 10-W GaN monolithic microwave integrated circuit PA for amplitude-modulated pulsed radar transmitters. Additional research on the PA trap-induced performance degradation is also presented and can be explored to predict the PA performance for different parameters of the operative regime or for the formulation of the PA behavioral models.

(Invited) Generation-Recombination Noise in Advanced CMOS Devices

It is well-known that the noise power spectral density of scaled devices increases for smaller area. At the same time, the character of the noise spectrum changes from predominantly 1/f-like to Lorentzian, characterized by a plateau at low frequencies (f<fc) and a roll-off following 1/f2 [1]. Such a noise is related to trap-assisted generation-recombination events, with characteristic or corner frequency fc, which is related to the emission (te) and capture (tc) trap time constant of the carriers from and to the defect centre. In scaled CMOS transistors, two different behaviours of fc with respect to the gate voltage (VGS) can be distinguished: as shown in Fig.1 some traps exhibit little VGS dependence, while others show a more or less exponential increase with VGS [2]-[4]. In the first case, it is believed that the traps are in the depletion region of the semiconductor, while the second case corresponds to traps in the gate dielectric [1],[2] and is also known as Random Telegraph Noise (RTN). In this paper, it is shown how the Lorentzian parameters can be used to extract more information on the underlying traps. In the case of an oxide trap, it will be shown that the exponential slope at low or at high VGS corresponds with ae or ac, i.e., the exponential slope of the emission and capture time constant. This is illustrated by Figs 2 and 3. From these exponential coefficients, the trap depth in the oxide can be derived, as an alternative for the time constant ratio tc/te. In principle, one can rely then on the variation of fc with temperature T to extract the activation energy for carrier capture (high VGS) and/or emission (low VGS). In the case of semiconductor traps, the shift of fc with T, illustrated in Fig. 4 for a Si bulk nFinFET can be employed for the construction of an Arrhenius plot, like in Fig. 5. The slope yields the activation energy while a capture cross section can be derived from the intercept with the y-axis. This can also be applied to relaxed (r-) or strained (s) Ge pMOSFETs fabricated in different substrate types [6]. As shown in Fig. 6, the nature of the substrate has a strong impact on the corresponding surface trap density.

Characterization of charge trapping phenomena at III–N/dielectric interfaces

Charge trapping related phenomena are among the most serious reliability issues in GaN/AlGaN MIS-HEMTs technology. Today, many research efforts are undertaken to investigate and identify the defects responsible for device degradation. This work focuses on the trap sites located close to the interface with the dielectric, which are responsible for large voltage drifts in on-state conditions. We study the response of GaN/AlGaN/SiN systems to small and large signal excitation. Measurements performed with a lock-in amplifier enable us to deeply understand the dynamic behavior because of the improved time resolution and the versatility of the instrument. We investigate the frequency dispersion and the hysteresis of these devices and conclude that direct analysis of impedance characteristics is not sufficient to extract information about the interface trap response. We propose a methodology to study trapping phenomena based on transient measurement analysis, describing the approximations made and their effect on the accuracy of the result. Results on MIS test structures confirm the existence of a broad distribution of trap states. Capture time constants are found to be uniformly distributed in the experimental time window between 50μs and 100s.

Single vacancy defect spectroscopy on HfO2 using random telegraph noise signals from scanning tunneling microscopy

Random telegraph noise (RTN) measurements are typically carried out at the device level using standard probe station based electrical characterization setup, where the measured current represents a cumulative effect of the simultaneous response of electron capture/emission events at multiple oxygen vacancy defect (trap) sites. To better characterize the individual defects in the high-κ dielectric thin film, we propose and demonstrate here the measurement and analysis of RTN at the nanoscale using a room temperature scanning tunneling microscope setup, with an effective area of interaction of the probe tip that is as small as 10 nm in diameter. Two-level and multi-level RTN signals due to single and multiple defect locations (possibly dispersed in space and energy) are observed on 4 nm HfO2 thin films deposited on n-Si (100) substrate. The RTN signals are statistically analyzed using the Factorial Hidden Markov Model technique to decode the noise contribution of more than one defect (if any) and estimate the statistical parameters of each RTN signal (i.e., amplitude of fluctuation, capture and emission time constants). Observation of RTN at the nanoscale presents a new opportunity for studies on defect chemistry, single-defect kinetics and their stochastics in thin film dielectric materials. This method allows us to characterize the fast traps with time constants ranging in the millisecond to tens of seconds range.

Study of Hot-Carrier-Induced Traps in Nanoscale UTBB FD-SOI MOSFETs by Low-Frequency Noise Measurements

The hot carrier (HC)-induced traps in nanoscale fully depleted ultrathin body and buried oxide nMOSFETs are investigated by low-frequency noise (LFN) measurements in the frequency and time domains. The measured noise spectra are composed of 1/f and Lorentzian-type components. The Lorentzian noise is due to either generation–recombination noise or random telegraph noise (RTN). Based on the LFN results, the effect of the HC stress on fully depleted silicon-on-insulator MOSFETs is investigated after short- and long-time stress. The capture and emission time constants responsible for the RTN noise were calculated as the average duration time of the high/low drain current state, respectively. Analysis of RTN traps detected in fresh and HC-stressed devices indicates that the RTN amplitude is uncorrelated to the trap time constants, i.e., the impact of the trap depth from the interface is masked by that of the trap location over the channel. The overall results lead to an analytical expression for the RTN amplitude, enabling to predict the RTN changes from the subthreshold to the above-threshold region.

Simulating RF Performance of Proton Irradiated AlGaN/GaN High Electron Mobility Transistors (HEMT)s

AlGaN/GaN High Electron Mobility Transistors (HEMTS) subjected to proton irradiation of fluence up to 1014 cm-2 were simulated to observe the impact on small signal and RF characteristics. While a lot of work has covered dc degradation of HEMTs, very few studies exist which focus exclusively on RF degradation despite their high frequency applications. This work establishes a simulation framework capable of frequency domain device simulation up to 10 GHz. The role of defects generated by proton irradiation will be studied by carrying out low frequency transconductance simulation and current gain frequency sweeps to examine degradation of cut-off frequency. The primary objective will be to understand the role of partially ionized donor traps near the AlGaN/GaN interface and how their bias dependence and capture time constants may significantly impact cut-off frequency and current gain estimations at high frequencies.

Simulating RF Performance of Proton Irradiated AlGaN/GaN High Electron Mobility Transistors (HEMT)s

Due to their relative radiation hardness, High Electron Mobility Transistors (HEMTs) with an AlGaN/GaN heterostructure are of high interest for space applications. Many studies have been performed to quantify the level of degradation due to proton radiation, with the majority being focused on the effects on the DC or steady-state performance. Fewer studies have evaluated degradation of the radio frequency (RF) performance due to proton radiation. Chen et al. caution that degradation due to proton irradiation has a more measurable impact on the RF performance in comparison to the DC performance [1]. In particular, gate lag and increases in both channel resistance and device capacitance due to fast bulk and surface traps contribute more notably to RF degradation than to DC degradation [1]. Understanding the dynamics of these traps is required for modeling and simulation of GaN HEMTs for RF applications. In this paper, we aim to identify the necessary input parameters for future RF simulation of radiation effects on AlGaN/GaN HEMTs. In previous work, we have found that donor traps have a larger role than previously thought on the magnitude of DC performance degradation. We use experimental techniques to identify the donor trap lifetimes and energies, both of which are incorporated into a transient simulation of the HEMT and compared to results obtained experimentally. The experiments are long-term transient measurements of drain current after a step in gate voltage for various temperatures, resulting in the establishment of capture and release time constants associated with donor traps. (1) J. Chen, E. X. Zhang, C. X. Zhang, M. W. McCurdy, D. M. Fleetwood, R. D. Schrimpf, S. W. Kaun, E. C. H. Kyle, and J. S. Speck, IEEE Trans. Nucl. Sci, 61 (6), 2959, (2014).

Comprehensive studies on the accuracy of trap characterization by using advanced random telegraph noise simulator

Our developed noise simulator can represent the dynamic behaviors of electron and hole trapping and de-trapping via interactions with both the Si substrate and the poly-Si gate. Simulations reveal that the conventional analytical model using the ratio between the capture and emission time constants yields large errors in the estimates of trap site positions due to interactions with the Si substrate and poly-Si gate especially in thin gate insulator MOSFETs.

Investigation of Random Telegraph Noise in the Dark Current of Germanium Waveguide Photodetector

Dark current variations in the germanium waveguide photodetector were investigated. Contributing to the time dependent dark current variations, random telegraph noise (RTN) was observed and studied for the first time. The RTN at different reverse biases and temperatures were measured. By analyzing the capture and emission time constants, the single trap that was responsible for RTN was estimated to be around 18 nm from the interface between the N <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">++</sup> and the intrinsic region. The trap energy levels at different reverse biases were also extracted and the trapping and detrapping dynamics were explained. Through this procedure, the single trap that may lead to device failure can be detected and characterized without destroying the well-fabricated devices.

Time-Dependent 3-D Statistical KMC Simulation of Reliability in Nanoscale MOSFETs

This paper presents a thorough numerical investigation of statistical effects associated with charge trapping dynamics and their impact on the reliability projection in decananometer MOSFETs. By means of a novel 3-D kinetic Monte Carlo TCAD reliability simulation technology, we track the time-dependent variability associated with granular charge injection and trapping into the oxide traps. We consider the interactions between the statistical variability of the virgin transistors, introduced by the discreteness of charge and granularity of matter, and the stochastic nature of the trap distribution and the trapping process itself. As a result, the path to device failure (PtDF), defined as the stochastic succession of trapping events that bring the device parameters to a predefined failure criteria, is analyzed in detail. In particular, we show that the two stochastic components determining the PtDF, namely the traps' capture time constants and threshold voltage shifts, are uncorrelated. The charge injection variability is shown to play a dominant role in determining the statistical dispersion of the reliability behavior. Furthermore, we show that the short- and long-term reliability behaviors are uncorrelated. Finally, 3-D fringing and percolative effects are shown to play an important role in determining the statistical degradation of nanoscale MOSFETs.

Assessment of the Impact of Inelastic Tunneling on the Frequency-Depth Conversion from Low-Frequency Noise Spectra

An expression is derived for the conversion of the frequency axis of a low-frequency noise spectrum into a depth axis, based on the capture and emission time constant of a random telegraph signal. By connecting the corner frequency of the corresponding Lorentzian spectrum to the measured trapping time constant, the effect of multiphonon relaxation upon tunneling into an oxide trap can be included in a natural way. With this expression, one can derive in a straightforward manner some important trends with respect to the impact of several trap and material parameters on the tunneling depth, both for SiO2 and high- κ gate dielectrics.

On the Distribution of NBTI Time Constants on a Long, Temperature-Accelerated Time Scale

Recent investigations on individual defects contributing to negative bias temperature instability (NBTI) showed that the emission and capture time constants are thermally activated via an Arrhenius law. We apply this finding to conventional micrometer-sized devices where NBTI is the response of up to millions of defects. We rapidly switch the device temperature using an on-chip heating structure in order to accelerate NBTI stress and recovery and acquire experimental data on an up to 18 decades long time scale. On this extended time scale, we find that the distribution of NBTI defect time constants is log-normal with large mean and variance which follows directly from a normal distribution of energy barriers in Arrhenius law. As such, our work clearly identifies the role of temperature for NBTI and suggests a method for accurate life-time estimations.

The Effect of Adjacent Bit-Line Cell Interference on Random Telegraph Noise in nand Flash Memory Cell Strings

The effect of adjacent bit-line (BL) cell interference on BL current fluctuation (ΔIBL = high IBL - low IBL) due to random telegraph noise (RTN) in floating-gate nand Flash cell strings is characterized. It was found that the electron current density (Je) of a read cell can be appreciably different depending on the position in the channel width direction because of the interference from adjacent BL cells. The interference can be controlled by the state (program or erase) of the adjacent cells. We verified that ΔIBL due to RTN increases as a high-Je position is controlled to be close to a trap position in 32-nm nand Flash memory strings. Finally, it was also shown that the adjacent cell interference affects not only ΔIBL but also the ratio of capture and emission time constants [ln(τc/τe)].

Three-Dimensional Electrostatics- and Atomistic Doping-Induced Variability of RTN Time Constants in Nanoscale MOS Devices—Part II: Spectroscopic Implications

This paper investigates the impact of 3-D electrostatics and atomistic doping on the spectroscopic analysis of random telegraph noise (RTN) traps in nanoscale MOS devices. Using the numerical model and the template decananometer Flash cell presented in Part I of this paper, the gate bias dependence of the capture and emission time constants of oxide traps is shown to largely depend on the trap position over the channel, both in the subthreshold and in the on-state regime. This compromises the accuracy of any 1-D method for trap spectroscopy based on the time constants analysis and, due to the randomness in trap position and dopant placement in the substrate, calls into question the possibility for any accurate trap spectroscopy in nanoscale devices. Finally, the possibility to extract any information on the trap depth from the fluctuation amplitude of RTN waveforms is shown to be precluded by the large statistical spread of the amplitude itself, resulting in its negligible correlation with the trap position in the oxide and with the waveform time constants.

A Simple Model for Capture and Emission Time Constants of Random Telegraph Signal Noise

A simple model for electron capture and emission process in oxide traps is introduced for analysis of random telegraph signal noise. Multiphonon emission capture theory is used for a “more fundamental” thermally activated capture cross-sectional model. Basically, Shockley-Read-Hall statistics is considered for time constant modeling. Particularly, the thermally activation model with the quantum effect is applied to emission time constant modeling. It is shown that the capture and emission process are controlled by characteristics of traps. Especially, the physical meaning of the lattice relaxation energy which is the most important model parameter and the trap characteristic is clarified.

(Invited) Random Telegraph Noise: From a Device Physicist's Dream to a Designer's Nightmare

An overview is given on Random Telegraph Noise (RTN) in MOS-based devices. First, the basic properties and physics are briefly outlined, emphasizing the stochastic nature of its main parameters: the capture and emission time constant, while its amplitude is fixed and specific for each trap. Different techniques exist to characterize RTN in MOS devices, either by time or frequency domain measurements. A distinction can also be made between dynamic equilibrium methods like noise spectroscopy or transient measurements, based on a Deep-Level Transient Spectroscopy approach. While single-device based RTN measurements are still of high value, for modern circuit applications, techniques are being developed allowing a fast assessment of RTN in a large number of transistors. The scaling of CMOS technologies to the 32 nm and below raises the issue of variability induced either by random dopant fluctuations or in general, by random charge fluctuations, both spatially and in time. As will be seen, the presence of traps responsible for RTN brings about an additional source of variability which may become dominant in future memories. Finally, it will be shown that RTN is not only a fundamental component of 1/f noise but the same oxide traps are also largely responsible for the Bias Temperature Instability (BTI) in scaled MOSFETs. This leads to a self-consistent model for BTI and a stochastic approach towards the reliability prediction of deep submicron CMOS technologies.

Time-resolved high-temperature detection with single charge resolution of holes tunneling into many-particle quantum dot states

We demonstrate detection of many-particle hole states in InAs/GaAs quantum dots with single charge resolution up to a temperature of 75 K. Capacitance-voltage measurements as well as time-resolved current measurements in an adjacent two-dimensional hole gas are used to determine the emission and capture time constants from 4 K up to 130 K. A transition from pure tunneling to thermally assisted tunneling is observed with increasing temperature. An equivalent circuit model gives access to the energy level splittings of the many-particle hole states and explains the broadening of the peaks at higher temperatures.

Extraction of trap location and energy from random telegraph noise in amorphous TiOx resistance random access memories

Random telegraph noise (RTN) has been studied in amorphous TiOx (α-TiOx) resistance switching random access memories (RRAMs). The RTN having two discrete current levels was observed only in the high-resistance state of the RRAMs. By investigating the bias dependence of capture and emission time constants, we extracted the vertical location of a trap responsible for the RTN in RRAM devices. The trap causing the RTN was found around 5.7 nm below the Ti (top electrode). The trap energy was less by 0.18 eV than the conduction band edge of the TiOx.

Time-dependent defect spectroscopy for characterization of border traps in metal-oxide-semiconductor transistors

We introduce the time-dependent defect spectroscopy (TDDS) for the analysis of a particular class of oxide defects known as ``border traps.'' These defects have a fundamental impact on the behavior of metal-oxide-semiconductor field-effect transistors and are commonly linked to the occurrence of random-telegraph noise, $1/f$ noise, and slow charging transients. The TDDS naturally extends the successful deep-level transient spectroscopy as it extracts both the capture and emission time constants. Analysis proceeds via the so-called spectral maps, which separate individual border traps by their characteristic times and their voltage step height. In contrast to standard random-telegraph noise analysis methods, where uncorrelated capture and emission events of only a few traps can already create convoluted noise patterns, the synchronization by the charging pulse yields the spectral maps, which allow for the analysis of a large number of defect occupancies in a single measurement. As a consequence, the TDDS allows us to monitor the defect parameters over exceptionally wide temperature and bias ranges.