Interface state generation during electrical stress in n-channel undoped hydrogenated polysilicon thin-film transistors

Bulgarian Academy of Sciences, Ulpia Serdica, Sofia-Capital, Bulgaria
Electronics Letters (Impact Factor: 1.07). 12/1998; 34(24):2356 - 2357. DOI: 10.1049/el:19981616
Source: IEEE Xplore

ABSTRACT Interface state generation effects due to hot carrier phenomena
are studied in n-channel undoped hydrogenated polysilicon thin-film
transistors under on-current bias-stress conditions. At the initial
stages of stressing, interface states are created as well as hot hole
trapping. As the stress process proceeds further, a saturation of the
density of generated interface states was found after which hot electron
injection into the gate oxide becomes the most important factor in
further device degradation

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We address the mechanisms responsible for the enhanced degradation in the polysilicon thin-film transistors under dynamic hot-carrier stress. Unlike the monotonic decrease of maximum transconductance (G/sub m max/) in static stress, G/sub m max/ in dynamic stress is initially increased due to the channel shortening effect by holes injected into the gate oxide near the drain region and then decreased due to tail states generation at the gate oxide/channel interface and grain boundaries. The threshold voltage variations are dominated by two degradation mechanisms: (1) breaking of weak bonds and (2) breaking of strong bonds to obey the power-time dependence law with a slope of 0.4. The degradation of the sub-threshold slope is attributed to intra-grain bulk states generation.
    IEEE Electron Device Letters 11/2001; DOI:10.1109/55.954916 · 3.02 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Proper electrical functionality of devices is dependent on the proper insulation between active/passive areas and metal interconnect. Although low-k materials are gradually replacing SiO2 to achieve lower parasitic capacitance, they are still SiO2-based. With the introduction of Cu as back-end metallization, the need for a barrier between insulator and metal arises, regardless of insulating materials. Ta has been established as a mainstream diffusion barrier in Cu interconnects. As device feature size shrinks, thickness of the Ta barrier also decreases. This raises a big concern on the effectiveness of thin, columnar Ta barriers in stopping the diffusion of Cu into SiO2, especially under simultaneous application of bias and thermal stresses. As the integrity of SiO2 is reliant on Ta to prevent Cu diffusion, bias-temperature stressing (BTS) would be a very suitable test to indirectly evaluate the Ta barrier through SiO2 integrity. In this study, the stability of a 5 nm Ta barrier is investigated under various bias and thermal stressing conditions through BTS analysis. Stressed samples are analyzed with high frequency C–V curves, and time-of-flight secondary ion mass spectrometry depth profiles. From the results, the 5 nm Ta barrier is proved to be stable in the Cu interconnect system. © 2004 American Vacuum Society.
    Journal of vacuum science & technology. B, Microelectronics and nanometer structures: processing, measurement, and phenomena: an official journal of the American Vacuum Society 07/2004; 22(4):1844-1850. DOI:10.1116/1.1776562 · 1.36 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Leakage current evolution during two different modes of electrical stressing in hydrogenated-undoped n-channel polysilicon thin film transistors (TFTs) is studied in this work. On-state bias stress (high drain bias and positive gate bias) and off-state bias stress (high drain bias and negative gate bias) were performed in order to study the degradation of the leakage current. It is found that during off-state bias stress the gate oxide is more severely damaged than the SiO2-polySi interface. In contrast, during on-state bias stress, two different degradation mechanisms were detected which are analyzed.
    Microelectronics Reliability 06/1999; 39(6):885-889. DOI:10.1016/S0026-2714(99)00118-3 · 1.21 Impact Factor