The time dependent defect spectroscopy (TDDS) for the characterization of the bias temperature instability

Abstract

We introduce a new method to analyze the statistical properties of the defects responsible for the ubiquitous recovery behavior following negative bias temperature stress, which we term time dependent defect spectroscopy (TDDS). The TDDS relies on small-area metaloxide-semiconductor field effect transistors (MOSFETs) where recovery proceeds in discrete steps. Contrary to techniques for the analysis of random telegraph noise (RTN), which only allow to monitor the defect behavior in a rather narrow window, the TDDS can be used to study the capture and emission times of the defects over an extremely wide range. We demonstrate that the recoverable component of NBTI is due to thermally activated hole capture and emission in individual defects with a very wide distribution of time constants, consistent with nonradiative multiphonon theory previously applied to the analysis of RTN. The defects responsible for this process show a number of peculiar features similar to anomalous RTN previously observed in nMOS transistors. A quantitative model is suggested which can explain the bias as well as the temperature dependence of the characteristic time constants. Furthermore, it is shown how the new model naturally explains the various abnormalities observed.

On the frequency dependence of the bias temperature instability

“…We study the duty-factor and frequency dependence in the light of the recently proposed capture/emission time (CET) map model [7,[21][22][23] and demonstrate that the recoverable component of BTI can only to the first-order be captured by a two-state defect model. By using a three-state defect model, consistent with detailed studies on the microscopic defect properties [24,25], the experimentally observed frequency dependence can be reproduced for a wide range of technologies.…”

“…Note, however, that the effective time constants in the CET map do not directly correspond to physical defect parameters like thermodynamic energy-levels or relaxation energies but are functions of them [27]. In this sense we wish to highlight that the CET map model is not a new model meant to replace our previous modeling efforts such as [24,31]. Rather, the CET map model provides a convenient transformation of the experimental data to help understand the wide distribution of time constants.…”

“…The analytic CET map uses a recoverable (R) and a permanent (P) component, both of which are represented by bivariate normal distributions for the effective activation energies of the correlated capture and emission times. Given recent experimental evidence [21,24,[32][33][34], all time constants are assumed to be thermally activated,…”

Analysis of defect capture cross sections using non-radiative multiphonon-assisted trapping model

“…The use of this empirical parameter is needed to reproduce single defects experiment results where the capture rates are evaluated from variation of electrical parameters and indicate a smaller dependence than exponential with the field [23]. Additional studies based on time dependent defect spectroscopy (TDDS) in [24] show how this behavior can reach saturation at higher stress voltages. Huard [23] showed that a constant critical field of 3 MV/cm agrees well with the capture rates exponential dependence measured over different CMOS technologies.…”

“….Þ have been taken from measurement values based on scanning tunneling spectroscopy studies for single-defect Si dangling bonds [25]. The dynamics of other oxide traps (E c and E d [7,32]) are better described including metastable states [11,24]. However, EPR studies indicated that these defects relax almost immediately after the removal of the stress condition [9], and thus they have not been investigated in this study.…”