Chapter 1
Introduction
1.1 Preface
Displays show everywhere in our daily life, such as, living room televisions, and computer monitors. People would like to pursue more realistic display images through the development of new technology. For the display revolution, the display images have changed from monochromatic to colorful, and the display thickness of display has become thinner because LCD (Liquid Crystal Display) replacing CRT (Cathode Ray Tube). In addition, high quality display is still developing for vivid images,. Now, for the next generation display, 3D images are surely to be presented in order to create images as true as real world.
3D images are much more natural and vivid than 2D images, and traditional displays cannot provide that. Therefore, there are more and more 3D display technologies and they are expected to provide depth information when displaying.
Fig. 1- 1 History of display technology
CRT Flat panel display HDTV Stereoscopic display Autostereoscopic display
1950 1975 1995 2000 2010
2
1.2 Principles of 3D image
Binocular vision allows humans to get the depth information that results from our two forward-facing eyes, called stereopsis [1][2]. There is a slightly different viewpoint perceived by each eye, and our brain will calculate this disparity as the sensation of depth [Fig. 1-2].
Fig. 1- 2 Horizontal disparity
Horizontal disparity
Left eye’s image Right eye’s image
Left and right images are displayed in different set of pixel on the display
3
With monocular vision [3], people can also perceive the depth of the real world by experiences and time learning. For example, the depth sensation of a picture can be felt by shading, interposition, linear perspective and so on, as shown in Fig 1-3.
Fig. 1- 3 Picture illustrating depth cues
Moreover, oculomotor depth can be categorized into vergence and accommodation [4] due to feedback from our eyes’ muscles. The vergence is defined as the angle of the viewing target to our eyes [Fig. 1-4]. The accommodation is the crystalline lens focus on an object to get a clear image, as shown in Fig. 1-5. Both of these are generally regarded to help judging the depth information.
Fig. 1- 4 Vergence angle
Interposition
Texture Gradient Aerial distortion
Shading
Linear perspective
40 ° 10 °
4
Fig. 1- 5 Accommodation
It is also necessary to imitate the binocular system allowing viewers’ left and right eye to each perceive its own image and form a 3D image in order to apply 3D images on flat panel displays.
Fig. 1- 6 Depth cues
Binocular cues
Oculomotor cues Monocular cues
Angular disparity Horizontal disparity
Interposition Linear perspective
Shading Texture gradient
Arial distortion
Texture gradient Arial distortion
Depth
cues
5
1.3 3D display technologies
Nowadays, 3D display technology has been briefly classified in two parts. One is stereoscopic which needs extra devices; the other is auto-stereoscopic which is without extra devices, as shown in Fig. 1-7. The details of two types are described following.
Fig. 1- 7 Classification of 3D display technology
1.3.1 Stereoscopic displays
Stereoscopic display requires viewers to wear a device such as glasses, or a helmet to see the 3D image [5][6]. The main purpose is to ensure images are seen correctly by the left and right eye respectively. For example, the shutter type, the display shows a left eye image in the first frame and the left glass will be transparent at the same time, while the right glass is blocked; on the contrary, in the second frame, the display shows a right image and the right glass will be transparent [Fig. 1-8].
Stereoscopic
6
The display is synchronized with the shutter glasses. One glass will be shuttered when the other is transparent. Therefore, our left and right eye receives different images and our brain then combines them into a 3D image. No matter what kind of stereoscopic display, the common issue is that the extra-devices are uncomfortable and inconvenient to wear.
Fig. 1- 8 Shutter type 3D display
1.3.2 Auto-stereoscopic displays
Auto-stereoscopic displays, for example, holographic type and multiplexed-2D type have the benefit of letting people see 3D images without extra devices. Although the holographic type can show a realistic 3D image, it is hard to fabricate on a large scale screen. Multiplexed-2D type has the advantage of its thin scale. Moreover, the multiplexed-2D type can easily be implemented with flat panel displays which are widely used today. As the result, many researches are interested in these aspects.
Following, we also simply divide multiplexed-2D type into time-multiplexed and spatial-multiplexed for describing clearly.
Right-eye image
Time Frame 1
Left-eye image Left-eye image Right-eye image
Frame 2 Frame 3 Frame 4
7
1.3.3 Multiplexed 2D type displays
The general concept of time-multiplexed and spatial-multiplexed is projecting different images to both eyes individually, as shown in Fig. 1-9. The time-multiplexed type with high resolution per view needs high speed display elements to project to left and right eyes sequentially [7][8][9]. The spatially multiplexed type provides left and right images at the same time nevertheless the lower resolution per view is caused.
Fig. 1- 9 Concept of the time and spatial multiplexed type
These time-multiplexed systems has many kind of configurations been proposed.
For example, one uses a micro-optic structure on the principle of total internal reflection (TIR) that makes the light go through the desired direction, as shown in Fig.
1-10 [10]. Thus the 3D image can be observed by switching the two light sources sequentially.
Time-multiplexed Spatial-multiplexed
Time R L
Time R
L
Left-eye image
Right-eye image
8
Fig. 1- 10 3D display based on sequentially switching backlight with focusing foil There also are some configurations for spatial-multiplexed type such as parallax barrier [11][12] and lenticular lens array [13][14], and have been widely used in 3D flat panel display. These optical components are used to project the images to left and right eye-viewing windows at the desired viewing position. Both of them are briefly introduced in the following section.
Parallax barrier
Parallax barrier blocks the light by using strips of black mask. Its optical design adjusts the viewing angles and position by optimizing barrier’s geometry, as shown in Fig 1-11. The parallax barrier is perhaps the simplest way to do this work, however, the main disadvantage is that opaque areas will reduce the brightness of display.
Fig. 1- 11 Spatial multiplexed displays: parallax barrier θ
α
Lightguide LCD
Lamp 1 Lamp 2
9
Lenticular lens array
Lenticular lens array are typically cylindrical lenses arranged vertically with respect to a display. The cylindrical lenses direct the light from a pixel to the design angle in front of the display, as shown in Fig 1-12. For the transparent lens, the brightness can still be kept.
Fig. 1- 12 Spatial multiplexed displays: lenticular lens array
However, the image resolution decreases rapidly as more views were used in the horizontal direction for the basic type of spatial multiplexed multi-view display although it much easy to be implemented. Therefore, the company, Philips, proposed a slanted lenticular lens array [15][16] to share the resolution reduction between horizontal and vertical direction, as shown in Fig 1-13 to overcome this drawback,.
Because the resolution has been decreased in this method, and there is no need to show 3D images all the time. Thus, the 2D/3D switchable display is sure to carry out.
10
Fig. 1- 13 Slanted lenticular lens 3D display with multi-view by the Philips Company
1.3.4 2D/3D switching methods
The 2D/3D switchable displays provide more functions and options to users.
For example, users may watch an exciting video in 3D mode and read an interesting article switching to 2D mode. In these years, there are several techniques can switch the images between 2D and 3D, and briefly be divided into active parallax barriers [17], activated micro lens [18][19][20][21].
Active parallax barrier
The active parallax barrier is able to block the light as turn on voltage to form the barrier. Thus the 3D images can be obtained. N the other hand, , the parallax barrier become totally transparent and let all light pass through it to form 2D images when turning off the voltage. The same as mentioned before, the opaque areas still reduce the brightness of display in 3D mode.
11
Fig. 1- 14 2D/3D switching display with active parallax barrier
Active micro lens
Active micro lens has the similar function to the transparent lenticular lens.
Switch the lens by controlling the voltage can change the optical path to 2D and 3D modes. There are some devices can do this work, such as polarization activated micro-lens [22], active LC lenticular lens [23], and electric-field driven LC lens (ELC lens), are illustrated following.
1. Polarization activated micro-lens:
The polarization switching cell can change the polarization state likes the conventional LCD panel. And then the polarized light passes through the LC type GRIN lens, as shown in Fig 1-15. At the homogeneous state, the polarized light passes through the LC with the same refractive index of the glass. On the other hand, the changed-polarized light passes through the LC with different refractive index of the glass when applying a voltage on polarization switching cell. Thus, the inhomogeneous state will cause the refraction and will converge the light.
RIGHT
12
Fig. 1- 15 Polarization activated micro-lens 2. Active LC lenticular lens:
The polarized light has been converged by the different refractive index between LC and the glasses. The liquid crystal will change its direction and cause the LC refractive index to be same with the glass as applying a voltage on the LC lens.
However, the mismatching around the interface between concave structure and LC molecules is hard to achieve a complete lens off state.
Fig. 1- 16 Active LC lenticular lens
~ ~
Pixels Polarization
Switch LC type GRIN lens
(a) (b)
~ ~
Pixels LC Concave
lens
(a) (b)
13
3. Electric-field-driven LC lens (ELC lens):
The most attractive characteristic of ELC lens is its simple structure and easy fabrication. The ITO placed out of the top glass substrate would make the electric field distribution much smoother, but cause the high operating voltage. The bottom substrate was coated with ITO on the whole surface. As applying voltage on the top electrodes, the LC cell will become to a GRIN lens.
Fig. 1- 17 Electric-field-driven LC lens (ELC lens) 4. Multi-electrode driven liquid crystal lens (MeD-LC Lens)
The concept of MeD-LC lens is similar to ELC lens, as shown in Fig. 1-18. At the voltage off state, incident light passes through the MeD-LC lens cell without change of propagation direction. The electric field at the lens edge is stronger than at the lens center for applying certain voltage on different electrodes. Therefore, a non-uniform angle of LC director and the refractive index causes a phase difference and changes light direction.
~ ~
Pixels LC
(a) (b)
14
Fig. 1- 18 Principle of the MeD-LC lens
The principle of liquid crystal lens and driving method will introduce in the next sections.
1.4 Introduction to liquid crystal
Liquid crystal has an orientational order as the most important feature that makes anisotropic physical property such as an electric field, a magnetic field. For the rod-like molecules, the average directions of long molecule axes in nematic phase are along a common direction by the unit vector , as shown in Fig. 1-19.
Fig. 1- 19 Schematic diagram showing the orientation of rod-like molecules
~ ~
Pixels LC
(a) (b)
2D mode 3D mode
n
15
Because the permittivity in the direction parallel to is different that from the direction perpendicular to , the induced polarization depends on the orientation of the liquid crystal director with the respect to the applied field. The electric energy of liquid crystal per unit volume is approximately given by
(1-1) If the liquid crystal has a positive dielectric anisotropy ( ), the electric energy is minimized when liquid crystal director is parallel to the applied field ( ).
Conversely, if the liquid crystal has a negative dielectric anisotropy ( ), then the liquid crystal director tends to perpendicular to the applied field to minimized the electric energy [24].
Thus, the LC orientation will be changed by the electric energy, and the refractive index can be controlled by the applied voltage.
Fig. 1- 20 Geometry to calculate refractive index change with liquid crystal molecule orientation
In uniaxial crystals the index ellipsoid can be shown in Fig 1-20. The first mode, called the ordinary wave, has a refractive index no regardless of . In accordance
X Z
Xθ Zθ
nθ ne no
θ
16
with the ellipse shown in Fig 1-20, the second mode, called extraordinary wave, has a refractive index that varies from no when =0, to ne when =90. When the direction of liquid crystal is not rotated, the light will be affected by the refractive index ne. when the liquid crystal axis is rotate with an angle , the light will be affected by no
and ne.
This slice of the indicatrix can be expressed as Eq. 1-2.
(1-2) And
(1-3) Therefore, Eq.1-1 can be written as Eq.1-3.
(1-4)
Thus, the equivalence equation of rotate angle can be derived as Eq.1-6.
(1-6)
Since the nematic liquid crystal does not align uniformly across the cell in the thickness direction, a more useful form to solve the as Eq.1-7.
(1-7)
This formula indicates the relation between effective refractive index and angular orientation of the namatic liquid crystal.
17
1.5 Liquid crystal lens (LC lens)
The LC lens [25][26] has the same focus effect as the conventional optical lens.
Fig 1-21 shows how the conventional lens focuses the light. As a plane wave incident light passes through a lens, the wave fronts converge to a focal point due to the optical path difference (OPD). The OPD can be defined as (where n is refractive index and d is the path length).
Fig. 1- 21 Conventional lens
A graded–index (GRIN) material [27] has a refractive index that varies continuously with position, n(r). The OPD can be defined as . The relation between refractive index and position has been shown in Fig. 1-22. The refractive index increases from the outer to the center.
Fig. 1- 22 GRIN lens
f
Focal length
z
r
n
(r)18
As we assume that the focal point is larger than the incident position (f>r) and the refractive of the center is large than the edge ( n(0) > n(r) ). The light velocity is inverse proportional to the refractive index as formula shows. Therefore, the wave fronts will slow down in the dense region and speed up in the rare region and causing the light to focus.
Fig. 1- 23 The geometry of the focused GRIN Lens
A ray traverses the lens on the optical axis passes along the optical path length (OPL) as . For a ray traverses at a height , overlooking the slight bending of ray path, . Since a planer wave front must be bent into spherical wave fronts, the OPLs from one to the other, along any route must be equal as followed,
(1-8) And
(1-9) By substitution,
(1-10)
n 2
n 1
n (r) r
d
R F
19
Where is the refractive index difference between center and the edge ( ), if can be neglected, the focusing formula results in Eq.1-11.
(1-11) By substitution,
(1-12) Where can be written as a parabolic function with a constant ,
(1-13) According to the formula 1-13 derived above, the GRIN lens will focus at point f when the distribution of refractive index is closed to a parabolic function. The GRIN lens and conventional lens are both the fixed focal lens. Thus, the LC lens is proposed to be a tunable focus lens with variable focal length.
Focal length of a LC lens is defined in Eq. 1-11. The liquid crystal direction can be controlled by the applying voltage. When light passes through the LC layer, the continuous-changing refractive index causes the phase retardation to act as a lens.
Fig. 1- 24 LC cylindrical lens
Rubbing direction Electrode
(ITO) LC layer
Electrode Glass Glass
d
20
(a) (b) Fig. 1- 25 (a) Lens off (b) Lens on
1.6 LC lens with 2D/3D switching
The catalogue of 2D/3D LC lens switching methods as mentioned before is shown in Table. 1-1. Generally speaking, these LC lens have some advantages and also have some disadvantages. Active LC lenticular lens and polarization active micro-lens have an issue of complex fabrication. ELC lens and MeD-LC lens have much simpler fabrication; however, the long switching time between 2D and 3D is about 4~10 sec depending on the LC cell gap.
Glass ITO
LC layer
Glass ITO
V1
V0 V1>V0=0 volt
E-field
21
Table. 1- 1 2D/3D LC lens switching methods
1.7 Motivation and objective
People have been surrounded by displays more and more often in these days. 3D displays are much more realistic and may replace 2D displays as the next mainstream in the market. However, 3D displays still need to overcome some issues. The parallex barrier has lower brightness than lenticular lens. Both of parallex barrier and lenticular lens will loss more resolution as showing more views. Consequently, active optical 2D/3D switching devices are surely needed.
According to Table 1-1, the electrode type LC lens is a better choice for its simple fabrication. The High Resistance TFT Liquid Crystal Lens (HR-TFT LC lens) was combining high-resistance layer with TFTs structure. The high-resistance layer coating between the ITO electrodes created the continuous electric field. The TFTs is an on/off switch. The resistance value is corresponding to the applying gate voltage.
~ ~
~
~
22
HR-TFT LC lens has high resistance layer characteristic when the gate turns off. And it will be the TFTs letting current pass through the coating layer as the gate turns on.
Therefore, the HR-TFT LC lens has multifunction such as, 2D/3D switchable, 2D/3D rotatable, 2D/3D localized. In this thesis, the 2D/3D localized function was proposed.
Fig. 1- 26 Sketch of high-resistance TFT liquid crystal lens
1.8 Organization
This thesis is organized as following: the fabrication process of the HR-TFT LC lens will be introduced in detail in Chapter 2. In Chapter 3, the design and simulation result of the HR-TFT LC lens and discussion will be presented.
Additionally, this chapter also shows the driving method of the HR-TFT LC lens. The experimental result and summary will be shown in Chapter 4. Finally, the conclusions and future work of this thesis will be discussed inChapter 5.
Glass
LC layer Gate ITO
a-IGZO
ITO
Glass SiO2 Source/Drain
ITO
23
Chapter 2
Fabrication and Measurement System
2.1 Introduction
The fabrication of HR-TFT LC lens is described in this chapter. The thickness of substrate glass is 550 . Firstly, the glass substrate is cleaned by the standard process.
Secondly, the semiconductor process including spin coating, lithography, sputtering, PECVD and leave-off is utilized in order to obtain the desired pattern. Thirdly, high-resistance layer is coated by the sputter and annealed by the furnace. Fourthly, the alignment layer is made by the spin coating and rubbing techniques. Fifthly, the top and bottom glass is assembled by ball spacers 30 to control the cell gap, and then filed in LC (E7). Finally, wires for various voltage driving will connect to each electrode.
2.2 Fabrication processes
This section describes the cell fabrication process including spin coating, lithography, wet etching, sputtering, annealing, rubbing, assembling. The detail process steps were listed below and as shown in Fig. 2-1.
24
Fig. 2- 1 Fabrication steps
Glass cleaning
25
I. Glass cleaning
For the transparent characteristic, glass has been widely used as the substrate for the display. In the fabrication, the glass of 0.55 mm thick is chosen and coated with ITO thin film already. The detailed sequence is as follow:
Step 1 : Place the glasses on the Teflon carrier, and put into a container with acetone as shown in Fig. 2-2. Ultrasonic vibrate for 20 minutes to remove the organic contamination on the glasses.
Step 2 : Rinse the glasses by DI water for 1 minute.
Step 3 : Rubbing the glasses with detergent by hands.
Step 4 : Rinse the glasses by DI water for 1 minute.
Step 5 : Place the glasses on the Teflon carrier, and put into a container with DI water as shown in Fig. 2-2. Ultrasonic vibrate for 40 minutes to remove the remained particle and detergent on the glass.
Step 6 : Use N2 purge to dry the glasses, and place glasses into a glass container with a cover.
Step 7 : Put the glass container into an oven with 110℃ for 30 minutes.
Fig. 2- 2 Schematic picture of step1 and step 5
26
II. ITO pattering and insulator
The gate ITO pattering detailed sequence is as follow and also can refer to Fig. 2-4.
Step 1: Before the lithography process, glass substrate was cleaned by step I.
Step 1: Before the lithography process, glass substrate was cleaned by step I.