Channel Extension Effect in Poly-Silicon TFTs with LDD Structure

• Main Authors: Chih-Yu Yen, 顏志宇
• Other Authors: Hsiao-Wen Zan
• Format: Others
• Language: en_US
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/51879802224148647000

碩士 === 國立交通大學 === 顯示科技研究所 === 96 === Polycrystalline silicon thin film transistors (poly-Si TFTs) have been studied extensively for their application on system-on-panel (SOP) technology due to the high mobility. For actual applications, lightly-doped drain (LDD) structure is usually applied to poly-Si TFTs. When the TFTs turn on, the devices have channel extension effect to affect the electric characteristics. In this thesis, we will study on the influence factors on the LDD channel extension. And purpose a notion to design the optimal LDD which can get a balance between leakage current and LDD channel extension. In the beginning, we use Silvaco, the device simulation software, to simulate the influence of device structure. The result reveals that the vertical electric field between gate and drain determines the degree of channel extension. Second, to simulate the effect of temperature and get the relation which is that adding the temperature is helpful to reduce the channel extension effect. Then, change the density of states (DOS) of the thin film. The simulation focuses on the relation between DOS and different LDD doping concentrations. We find that the DOS and LDD doping concentration have specific effect on the channel extension. Base on previous results, we compare the thin film resistivity model with Silvaco simulation results. And get the conclusion which is the resistivity is other main reason causes the LDD channel extension. Finally, through above studies, we know the simplest method to reduce the extension length is to increase the doping concentration in LDD. However, this manner causes the ability of decreasing the leakage current at the same time. Hence, by the way of considering the extension length as the device is turning on and the parallel electric field as the device is turning off. We can get the optimal LDD doping concentration to get a balance between these two problems.