Browsing by Author "Tseng, JC"
Now showing 1 - 3 of 3
- Results Per Page
- Sort Options
Item type:Item, Design optimization of ESD protection and latchup prevention for a serial I/O IC(2004-02-01) ;Huang, CY ;Chen, WF ;Chuan, SY ;Chiu, FCTseng, JCESD/latchup are often two contradicting variables during IC reliability development. Trade-off between the two must be properly adjusted to realize ESD/latchup robustness of IC products. A case study on SERIAL Input/Output (I/O) IC's is reported here to reveal this ESD/latchup optimization issue. SERIAL I/O IC features a special clamping property to wake up PC's during system standby situation. Along with high voltage operation, Input/Output (I/O) protection design of this IC becomes one of the most challenging tasks in the product reliability development. In the initial development phase, ignorance of latchup susceptibility resulted in severe Electrical Overstress (EOS) damage during latchup tests, and also gave a false over estimate of ESD protection threshold through parasitic latchup paths. The latchup origin is an output PMOS and floating-well ESD triggering NMOS beside the PMOS, and the main fatal link is this high-voltage (HV) NMOS connecting to a bi-directional SCR cell. This fatal link led to totally five latchup sites and three latchup paths clarified through careful and intensive FIB failure analysis, while this powerful SCR ESD device without appropriate triggering mechanism still could not provide sufficient product-level ESD hardness. Owing to there being no design window between ESD and latchup, the original several protection schemes were all abandoned. Using this bi-directional SCR ESD cell and proper triggering PNP bipolar transistors, a new I/O protection circuit could sustain at least ESD/HBM 4 kV and latchup triggering current 150 mA tests, thus accomplish the best optimization of ESD/latchup robustness. (C) 2003 Elsevier Ltd. All rights reserved.Item type:Item, Study on optimization of annular off-axis illumination using Taguchi method for 0.35 mu m dense line/space(1997-02-01) ;Loong, WA ;Tseng, JC ;Chen, TC ;Lung, CA交大名義發表The simulation and experiment study on optimizations of process factors under annular off-axis illumination (AOAI) using Taguchi method for design of experiment (DOE) for 0.35 mu m dense line/space in i-line was reported in this paper. The optimized simulation results for both resolution and DOF are as follows: inner ring sigma(i) = 0.35, outer ring sigma(o) = 0.70, NA = 0.5, negative bias on mask for line in dense line/space similar to 0.03 mu m (L/S = 0.320/0.380 mu m), positive bias on mask for isolated line similar to 0.03 mu m (iso. L = 0.38 mu m), overlapped DOF (dense + isolated) similar to 1.86 mu m, exposure dose similar to 278 mJ/cm(2). The experimental results are in good agreement with simulations. The optimized results by experiments with a NA of 0.57 and with AOAI are as follows: inner ring sigma(i) = 0.35, outer ring sigma(o)= 0.70, negative bias on mask for line in dense line/space similar to 0.015 mu m (L/S=0.335/0.365 mu m), positive bias on mask for isolated line similar to 0.03 mu m (iso. L = 0.38 mu m), overlapped DOF (dense + isolated) similar to 1.7 mu m, exposure time similar to 520 ms (exact dose uncertain). Without AOAI but with the same mask bias, the overlapped DOF is similar to 1.2 mu m by experiment. The overlapped DOF from mask patterns without any bias (L/S = 0.35/0.35 mu m; iso. L = 0.35 mu m) with 1/2 AOAI is similar to 0.93 mu m; and from conventional illumination without mask bias is up to similar to 0.72 mu m at most. The DOF, proximity effect and resolution for dense L/S are obviously improved from 1/2 AOAI exposure after optimization if compared with conventional exposure. The Taguchi method is also proved valuable in this study.Item type:Item, TiNx as a new embedded material for attenuated phase shift mask(1996-01-01) ;Loong, WA ;Chen, TC ;Tseng, JC ;交大名義發表應用化學系In this paper TiNx (x > 1.3) as a new material suitable for using as an embedded layer for an attenuated phase shift mask (APSM) is presented. TiNx thin film was formed by plasma sputtering under a gas mixture of Ar and N-2 (40:2 sccm). The related characteristics of TiNx at 365 nn (i-line) wavelength are as follows: n (refractive index) similar to 3.07; k (absorbance coefficient) similar to 0.531; R (reflectivity) 27 similar to 30%; rho (resistivity) similar to 52 mu Omega-cm (132 nm on quartz). For required phase shift degree theta=180 degrees, calculated thickness d of TiNx film is 88.2 nn, and transmittance T under 365 nm wavelength at this thickness is 14.5 % which is within the useful range for APSM. TiNx film also has good electrical conductivity, suitable for e-beam direct-write in patterning mask.