Investigation of existing defects and defect generation in device-grade SiO2 by ballistic electron emission spectroscopy

Abstract

Degradation processes initiated by defect generation in device-grade SiO2 were studied by locally injecting hot electrons from a scanning tunneling microscope tip into Pd/SiO2/p-Si(100) metal–oxide semiconductor (MOS) structures. An analysis of the emerging collector current in the Si substrate, a technique known as ballistic electron emission microscopy, provides electron transport information, from which the oxide defect generation process was studied. The charging of the defects resulted in shifts of threshold energies for electron transport across the oxide. A novel sheet charge model was developed to assess the in-depth distribution and charge densities in the oxide from field-induced threshold shifts obtained from experiment. An as-fabricated MOS system with an oxide thickness of 71 Å was investigated and found to contain existing electron traps of charge densities in the range (0.7–2.8)×1013 e/cm2 that are distributed within a 30 Å region adjacent to the metal/oxide interface. Further stressing was performed at zero oxide bias with increasing tip voltages of up to −10 V. New electron traps characterized by charge densities of (1.9–3.6)×1013 e/cm2 and located within 40 Å of the SiO2/Si interface were generated when the kinetic energy of the electrons injected into the SiO2 conduction band exceeded 1.9 eV. This energy threshold is in very good agreement with the hydrogen-release energy that is frequently invoked to explain oxide degradation.

Hot-electron effects and oxide degradation in MOS structures studied with ballistic electron emission microscopy

“…The near coincidence of the curves at high oxide biases indicates that the charge-saturated region was temporarily neutralized, since upon returning to V ox ϭ 0 (open symbols), V th nearly assumed its original value. Consequently, only the minimum in the distributions is shown in Figure 7 as Curve C-the maximum lies ϳ1 eV higher [28]. For V ox Ͻ 0 (open symbols), the threshold increases at a rate greater than the rate one would expect by extrapolating the slope into the negative bias region.…”

“…An example is shown in Figure 5 for a 7.1-nm-thick SiO 2 layer. The net shift after saturation was about 1 eV, but could vary from sample to sample, as well as on a local scale of order 5-10 nm [28]. Under these conditions no dependence on the tip bias was noted, which led to the conjecture that the traps were due to defects that existed before the BEEM measurements were performed [28].…”

“…From [28], with permission. The tip bias V T was kept below Ϫ6 V to suppress the generation of new traps.…”

“…Curve A was obtained from sets of scans on widely separated virgin areas to prevent charge accumulations from affecting the thresholds [29]. The generation and filling of new traps (i.e., kinetic energies Ͼ2 V) was observed to be limited as well [28]. The neutralization is attributed to anode hole injection, a process that becomes efficient when the electron energy in the SiO 2 exceeds 5 eV [16], although charge leakage by tunneling, particularly near the anode, may also be a contributory process.…”

“…The charge distribution can be modeled uniquely by calculating the field dependence of the thresholds with a trial distribution and subsequently adjusting it until the details of the nonlinear threshold shifts of Figure 7 are duplicated [28]. A single ␦-or sheet charge exhibits a strictly linear field dependence of the threshold energy at low to moderate fields.…”