Abstract



next up previous contents
Next: Kurzfassung Up: PhD Thesis Predrag Habas Previous: PhD Thesis Predrag Habas

Abstract

 

In this thesis the following problems are considered:

The physics of the gate-depletion effect is analyzed. An analytical one-dimensional model of the MOS structure is presented, including heavy-doping effects in the gate. The impact of the active dopant concentration in the gate close to the oxide interface and the surface charge at the gate/oxide interface on the flat-band potential, the threshold voltage, the inversion-layer charge and the gate capacitance is modeled analytically without approximations and evaluated in a quantitative manner. A two-dimensional numerical model of the gate-depletion effect is developed and applied to study the degradation of the MOSFET performances with further scaling down of dimensions and supply voltages. By means of the analytical MOSFET theory and the numerical model, the differences in the degradation of current, mobility and inversion-layer charge in the linear and the saturation region are explained. Extensive comparison of the assumed physical gate-model with the own experimental quasi-static gate capacitance data and the data from the literature is carried out. Several physical phenomena in the gate which are responsible for the observed disagreements are discussed. This investigation may be of general interest for modeling of heavily doped space-charge regions.

The transient generation-recombination in MOSFETs is analyzed. The charge-pumping effect and the charge-pumping characteristics are studied both, analytically and numerically. The state of the art in the charge-pumping measurement techniques is given. The trap-dynamics equations are coupled with the two-dimensional transient semiconductor equations on a rigorous selfconsistent footing and implemented in an MOSFET numerical model. The model allows interface and bulk traps arbitrarily distributed in both, energy and position space. Algorithms are developed to achieve a stable and fast convergency of the iterative process, even in the presence of very high trap densities in devices. The time-discretization error in solving the trap equations is evaluated by analytical means and by comparison with the numerical model. An approximate analytical model of the complete characteristics charge-pumping current versus gate bottom level is developed. By using the exact numerical approach, the accuracy and limitations of the analytical model are studied. Different charge-pumping threshold voltages and charge-pumping flat-band potentials are defined and clarified. The accuracy of the expression for the charge-pumping current is evaluated for the case when the current is governed solely by the emission levels. An improved expression for the emission level is derived accounting for a finite width of the transition region of the non-steady-state occupancy function. The formula commonly used to extract the trap density distribution in the energy space is found to be inaccurate. In this study, the numerically calculated data are used instead of the experimental data. The two-dimensional transient model of the charge-pumping experiment is employed to fully clarify the geometric current component. Two mechanisms are found to be responsible for the geometric component. One of them is the transfer of the minority carriers which are emitted from the interface traps towards the bulk, but not towards the junctions as is usually assumed. The effect causes that the geometric component does not vanish at very long turn-off times. The impact of this novel effect on the capture mode in the measurements with the three-level gate waveform is pointed out.

By applying the fully numerical approach we have studied the changes in the charge-pumping characteristics after hot-carrier stress. The impact of the amount, nature and the location of the stress-generated traps on the characteristics is analyzed. Different stress-experiments are simulated, in order to judge what information the real experiments are able to provide. They model the hole trapping and the electron capturing on the trapped holes and on the neutral traps generated in the hole injection. The surface-potential perturbation and the shift in different gate-bias terminal characteristics which are induced by a localized surface charge are studied. Note that the potential perturbation is smaller for a localized charge sheet than for a uniform one. For the latter it is known to result in a simple shift on the voltage axis. An analytical model of the surface-potential perturbation is derived by solving the Laplace problem in the depleted MOS structure in the presence of a localized surface charge. A connection between the local band bending and the gate-bias shift is derived. The accuracy of the analytical model is studied by comparison with the numerical solution. The influence of the depletion-region width, the width of the charge sheet and the screening effect due to an eventual inversion at the interface on the local band bending and the gate-bias shift is analyzed both, qualitatively and quantitatively. The extraction of the spatial distribution of the trap density along the oxide/silicon interface by the charge-pumping technique is critically studied. Both cases, traps symmetrically distributed at the source and drain sides in virgin devices and a damaged region localized at the drain side in stressed devices are considered. An error in the application of some present techniques is found, which originates due to neglecting the changes in the emission levels for all traps in device during the course of experiment. The present methods are improved to account for this effect. An experimental method is proposed which does not suffer from this effect. In addition, a technique for the measurement of the spatial distribution of fixed oxide charge by the charge-pumping technique is proposed. Differences between the charge-pumping characteristics of LDD MOSFETs against conventional MOSFETs are analyzed by means of analytical considerations and the numerical model. In order to explain the tail in the rising edge of the characteristics of LDD devices, an analytical model of the gate-corner/LDD-region electrical-field fringing is derived by solving the Laplace problem in the oxide. The accuracy of the analytical solution to the fringing problem is investigated by comparison with the numerical model. The importance of the nonvanishing lateral self-induced field in the semiconductor near the gate edge is clarified. The degradation of -channel LDD MOSFETs under the electrical stress is studied. The spatial distributions of the stress-generated traps are extracted from the charge-pumping measurements at different stress moments. Assuming the extracted trap distributions as input, the charge-pumping characteristics are calculated by employing the numerical model. A good agreement between the calculated and the experimental characteristics confirms the accuracy of the obtained trap distributions.

A model of the band-to-band tunneling is developed and implemented in an MOSFET numerical simulator. The tunneling path is considered fully in two-dimensions. It is accounted for the linear variations of the electric field along the individual tunneling paths. The potential and the field distributions in the critical gate/drain overlap region are investigated. A model for the direct tunneling rate in a linearly variable field is proposed. It is derived by employing the Wentzel-Kramers-Brillouin-Jeffreys (WKBJ) approximation assuming the dispersion relation for the two -interacting bands.



next up previous contents
Next: Kurzfassung Up: PhD Thesis Predrag Habas Previous: PhD Thesis Predrag Habas



Martin Stiftinger
Sat Oct 15 22:05:10 MET 1994