Date of Award

Summer 2014

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Electrical and Computer Engineering

First Advisor

Kaushik Roy

Committee Chair

Kaushik Roy

Committee Member 1

Anand Raghunathan

Committee Member 2

Byunghoo Jung

Committee Member 3

Muhammad A. Alam


In contrast to conventional bulk-silicon technology, polysilicon (poly-Si) thin-film transistors (TFTs) can be implanted in flexible substrate and can have low process temperature. These attributes make poly-Si TFT technology more attractive for new applications, such as flexible displays, biosensors, and smart clothing. However, due to the random nature of grain boundaries (GBs) in poly-Si film and self-heating enhanced negative bias temperature instability (NBTI), the variability and reliability of poly-Si TFTs are the main obstacles that impede the application of poly-Si TFTs in high-performance circuits. The primary focus of this dissertation is to develop new design methodologies and modeling techniques for facilitating new applications of poly-Si TFT technology. In order to do that, a physical model is first presented to characterize the GB-induced transistor threshold voltage (V th)variations considering not only the number but also the position and orientation of each GB in 3-D space. The fast computation time of the proposed model makes it suitable for evaluation of GB-induced transistor Vthvariation in the early design phase. Furthermore, a self-consistent electro-thermal model that considers the effects of device geometry, substrate material, and stress conditions on NBTI is proposed. With the proposed modeling methodology, the significant impacts of device geometry, substrate, and supply voltage on NBTI in poly-Si TFTs are shown. From a circuit design perspective, a voltage programming pixel circuit is developed for active-matrix organic light emitting diode (AMOLED) displays for compensating the shift of Vth and mobility in driver TFTs as well as compensating the supply voltage degradation. In addition, a self-repair design methodology is proposed to compensate the GB-induced variations for liquid crystal displays (LCDs) and AMOLED displays. Based on the simulation results, the proposed circuit can decrease the required supply voltage by 20% without performance and yield degradation. In the final section of this dissertation, an optimization methodology for circuit-level reliability tests is explored. To effectively predict circuit lifetime, accelerated aging (i.e. elevated voltage and temperature) is commonly applied in circuit-level reliability tests, such as constant voltage stress (CVS) and ramp voltage stress (RVS) tests. However, due to the accelerated aging, shifting of dominant degradation mechanism might occur leading to the wrong lifetime prediction. To get around this issue, we proposed a technique to determine the proper stress range for accelerated aging tests.