探討HAN、OCB、TN和ITN模式液晶元件的預傾角之量測與TFT-TN面板受封裝製程後產生窗簾雲紋（Curtain Mura）之原因

Abstract

在液晶顯示器中液晶預傾角(Pretilt Angle)影響了許多液晶的光電特性如:液晶顯示器畫面品質、液晶的反應時間、液晶顯示器的壽命還有TN型液晶中畫素的開口率……等。在本論文中我們提出一新型、簡單且直接的方式量測不同配向模式的液晶樣品預傾角如:混合配向模(HAN Mode)、光學補償彎曲模(OCB Mode)、扭轉向列模(TN Mode)、反置扭轉向列模(ITN Mode)。實驗中我們利用塞曼效應雷射(Zeeman Laser)搭配晶體旋轉法測量液晶相位差值對於入射樣品角度的變化，接著以增廣式瓊斯矩陣(Extended Jones Matrix)計算相關的理論模型擬合實驗結果得到不同配向模式下的液晶預傾角。實驗中，我們也利用常規方式量測相同樣品製作流程中所製作之相對應的水平配向液晶(Homogenous Aligned)樣品。實驗結果顯示量測的預傾角角度和相對應的水平配向液晶樣品相當吻合，由此可證明此方法的可靠度及精準度。 在扭轉向列型液晶盒量測中，我們比較我們的量測方式以及Birecki和Kahn所提出的量測方式，結果顯示出我們的量測方式使用在較小液晶盒厚度(Cell Gap)的扭轉向列型液晶樣品中較為精準，此結果表示我們的方法更適用於目前量產線上的扭轉向列型液晶顯示器產品的量測。而在反置扭轉向列液晶樣品中，我們第一次發現液晶傾角隨著Z軸高度變化是以一非線性曲線存在而不是一常數值，因此實驗中我們所量測到的液晶預傾角為反置扭轉向列液晶樣品中平均液晶傾角。而在所有預傾角量測實驗中，量測變數只和預傾角有相當的敏感度和精準度，對於液晶盒厚度(Cell Gap)、扭轉角(Twist Angle)和波長(Wavelength)皆不敏感，因此我們提出的預傾角量測方式具有相當的可靠度與精確度。 本論文中，我們也探討TFT-TN面板受COG ACF封裝製程後產生窗簾雲紋（Curtain Mura）的原因，實驗結果顯示出封裝製程造成液晶盒厚度(Cell Gap)變化較微小因此不是產生漏光的主要原因，且預傾角(Pretilt angle)的變化在模擬顯示幾乎不會造成面板漏光，因此Curtain Mura的發生主要是在受封裝製程後應力造成扭轉角(Twist Angle)的變化導致TFT-TN面板產生漏光產生的不均勻的畫面品質。我們也提出將扭轉向列型面板參數設計在Gooch-Terry first minimum以消除應力引致扭轉角的變化造成的面板漏光。 此論文所提出的量測以及分析方式在液晶面板光學檢測上具有相當的應用性和潛力，此直接量測方式可應用於檢測液晶樣品預傾角隨著老化時間的變化、分析許多Mura現象產生的原因、檢測樣品經光配向後之液晶預傾角，以及不同液晶模態下預傾角對光電特性之影響等。

In TFT-LCDs, LC pretilt angle adjacent to cell boundaries is a very important parameter which affects image quality, response time, the LCD lifetime and, in the case of TN mode, also aperture ratios of display pixels in the panel. In this thesis, we provide new and direct methods for determining the pretilt angles of HAN, OCB, TN and ITN modes. For the measurements of pretilt angles, we applied a Zeeman-laser heterodyne-interference system to measure the phase retardation versus liquid-crystal-cell out-of-plane rotation angle (Ph.-I.A. curve). We also developed relevant theoretical models based on extended Jones matrix method and Oseen-Frank elastic continuum theory to calculate the Ph.-I.A. curves to fit the measured data for the derivations of pretilt angles. We compared the measured pretilt angles with the measured pretilt angles of accompanying tilted homogenous or homeotropic (ATH) cells. The results showed that the measured pretilt angles were in good agreements with those of ATH cells. For TN mode, the measured pretilt angles by our method were more accurate than published method by Birecki-Kahn on thin cells used in modern TFT-TN products. In ITN mode, for the first time, we discovered that the deformation of LC tilt angles within the cell formed a nonlinear profile and the measured pretilt angle was the average of the LC-direct tilt angles over that nonlinear profile. The proposed methods were only sensitive to pretilt angle with good accuracy and insensitive to cell gaps, twist angles as well as incident wavelength for measurements. In this thesis, we also investigated the occurrences of Curtain Mura in the TFT-TN panels after COG-ACF process used for module assembly. The results indicated that the Curtain-Mura phenomenon was dominant by the stress-induced change in the twist angle, less sensitive and insensitive to the corresponding induced changes in the cell gap and pretilt angle of the in-panel-LC medium, respectively. We also suggested a way to reduce the occurrence of Curtain Mura after COG-ACF process by designing the cell gap of TN panels to satisfy the condition of Gooch-Terry first minimum in transmission. In this case, the variations of transmission were insensitive to the stress-induced change of twist angles. Our methods have significant potential in monitoring the variations of pretilt angles within LCD panels and can also be applied to investigate the various Mura phenomena caused by non-uniform pretilt angles. Future work should be focused on the relationship between pretilt angles and electro-optical properties of these LC modes.