Browsing by Author "Huang, C. -F."

Now showing 1 - 2 of 2

The Impact of TiN Barrier on the NBTI in an Advanced High-k Metal-gate p-channel MOSFET
(2017-01-01)
Huang, D. -C.
;
Hsieh, E. Ray
;
Gong, J.
;
Huang, C. -F.
;
Chung, Steve S.
This study investigates the bias temperature instability in high-k/metal-gate pMOSFETs with a TiN barrier layer sandwiched between the metal gate electrode and HfO2 dielectric and for reliability improvement of such devices. The experimental results clearly demonstrated that the diffusion mechanism of oxygen and nitrogen resulting from the post metallization treatment was the root cause of bias temperature instabilities in the p-channel MOSFETs. However, the device NBTI reliability is dependent on the nitrogen diffusion from the TiN layer, which degrades the SiO2/Si interfacial layer quality. Results show that by increasing the thickness of TiN barrier layer, the driving current will make worse the NBTI in p-MOSFET. Therefore, optimization needs to be considered for TiN as a barrier to improve the device reliabilities.
The Rouse-Mooney model for coherent quasielastic neutron scatterings of single chains well entangled in polymer melts
(2008-06-14)
Lin, Y. -H.
;
Huang, C. -F.
;
應用化學系
;
Department of Applied Chemistry
The dynamic structure factor (DSF) for single (labeled) chains well entangled in polymer melts has been developed based on the Rouse-Mooney picture; the DSF functions derived from the Langevin equations of the model in both discrete and continuous forms are given. It is shown that for all practical purposes, it is sufficient to use the continuous form to analyze experimental results in the "safe" q region (q being the magnitude of the scattering wave vector q) where the Rouse-segment-based theories are applicable. The DSF form reduces to the same limiting form as that of the free Rouse chain as q(2)a(2) or q(2)R(2)->infinity (a and R being the entanglement distance and the root mean square end-to-end distance, respectively), confirming what has been expected physically. The natural reduction to the limiting form allows the full range of DSF curves to be displayed in terms of the reduced Rouse variable q(2)(Z(d)t)(0.5) in a unified way. The displayed full range represents a framework or "map," with respect to which effects occurring in different regions of the DSF may be located and studied in a consistent manner. One effect is the significant or noticeable deviations of the theoretical DSF curves from the limiting curve in the region similar to 4>q(2)(Z(d)t)(0.5)>similar to 0.1 (a time region where t