Impact of the Surrounding Network on the Si-O Bond-Breakage Energetics
- PDF / 517,462 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 105 Downloads / 157 Views
1177-Z08-02
Impact of the Surrounding Network on the Si-O Bond-Breakage Energetics S.E. Tyaginov1,2, V. Sverdlov3, W. Gös1, Ph. Schwaha3, R. Heinzl3, F. Stimpfl3, T. Grasser1 Christian Doppler Laboratory for TCAD at the (3)Institute for Microelectronics, TU Wien, Gußhausstraße 25-29, A-1040 Vienna, Austria, (2) Ioffe Physico-Technical Institute, 26 Polytechnicheskaya Str., 194021 St.-Petersburg, Russia phone: +43-1-58801-36025; fax: +43-1-58801-36099; e-mail: [email protected] (1)
ABSTRACT We extend the McPherson Model for silicon-oxygen bond-breakage derived for a single SiO4 tetrahedron to capture the influence of the whole lattice. Several pair-wise potentials have been compared in the model including Mie-Grüneisen as well as diverse forms of TTAM/BKS. The contribution of the whole lattice substantially increases the activation energy for the Si-O bond rupture. The corresponding small transition rate of a non-distorted Si-O bond suggests that the interaction with the electric field alone can not be responsible for the bond-breakage and the contribution of other components such as energy delivered by particles and/or bond weakening is required. INTRODUCTION The energetics of the Si-O bond-breakage is one of the most crucial issues in the field of reliability of SiO2 films, especially in the context of Hot-Carrier-Injection (HCI) related degradation and the Time-Dependent-Dielectric-Breakdown (TDDB) [1-7]. There is still no consensus on whether the interaction of the dipole moment with the electric field or the energy delivered by particles is the driving force behind Si-O rupture in silica. As a consequence, several contradicting models based on one of the two main concepts exist: (i) the ThermoChemical Model [5,6] claiming that the applied field is responsible for bond-breakage, (ii) the Anode Hole Injection [2,7] and the Anode Hydrogen Release [8,9] models link Si-O breakage with the energy deposited by particles. All of these models have their shortcomings. Experimental data obtained by the group from IBM questioned the validity of AHI models [9]. The comparison between HCI-related degradation of p- and n-channel MOSFETs is presented in their work. They have shown that the hole component has to be at least comparable to the electron one in order to guarantee a considerable contribution to the defect density. However, in the AHI model a small anode hole is presumed and thus this approach is not suitable for describing the degradation mechanism in nMOSFETs. The shortcomings of two other models were well described in the review by Ghetti [10]. Thus, the AHR model cannot reproduce the isotope effect (hydrogen vs. deuterium annealing) on the oxide degradation. Moreover, in this model hydrogen is assumed to be present at Si/SiO2 interface and thus the substrate voltage dependence of stress-induced-leakage-current is not captured as well. On the other hand side, the Thermo-Chemical Model assumes that the electric field is the driving force for the breakdown, which is in contradiction to experimental evidence
Data Loading...
