Research and Development Trends in Flexible Display Technology

2.1 Structure of Flexible OLED displays

Although both LCDs and OLEDs have been raised as candidate devices for implementing flexible displays, OLED displays, which emit their own light and do not require a backlight, are thought to be preferable to LCDs that require hard backlights and are more difficult to make flexible. This section explains the structure of flexible OLED displays. Figure 2 (a) shows the basic pixel equivalent circuit used in OLED displays, which consists of two switching and driving thin-film transistors (TFT) for controlling the light emitted by the display along with a single storage capacitor. The voltage, which is the data written to the storage capacitor by the switching operation of the switching TFT, and the electric current flowing through and light emitted from the OLED are all controlled by the driving TFT. However, the actual structure is more complicated than this since a TFT circuit is required to compensate for variations between the TFT and OLED elements, as well as time variations in their characteristics.

Figure 2 (b) shows an example of the cross-sectional structure of flexible OLED displays. Here, we can see that the TFTs and OLEDs are formed on a thin, flexible plastic film substrate. Although curved displays can also be made using ultra-thin bendable glass substrates, plastic films have significant advantages in terms of reduced weight, improved flexibility, and higher impact endurance. However, among the issues with plastic films is the fact that, when compared to glass substrates, TFT and OLED fabrication process conditions are limited due to their low heat resistance, low chemical resistance, and high linear expansion coefficients^*2. Furthermore, since the sealant properties of plastic films against airborne oxygen and moisture are inferior to those of glass substrates, there is a large problem in terms of OLED service lifetime, since the device properties change due to oxygen and moisture exposure.

One potential solution to these issues is the use of polyimide, which is a plastic film with high heat resistance and a low linear coefficient of expansion and is currently being pursued for such applications by UBE Corporation. Polyimide, which is resistant to temperatures above 500 °C, can also be fully adapted to the TFT fabrication processes that are now used in glass substrate displays. However, there is a tendency for the optical transmittance to decrease as the heat resistance of the polyimide film increases, which limits its utility in display designs since a bottom-emission structure, in which the OLED emits light on the substrate side, cannot be used. Furthermore, while polyethylene terephthalate (PET) and polyethylene naphtholate (PEN) are low-cost plastic film materials that have high optical transmittance, as a result of their low heat resistance it would be difficult to incorporate these materials into TFT fabrication processes. Hence, it is clear that, in the future, it will be necessary to not only develop plastic materials with high heat resistance and high optical transmittance, but also to reduce the temperature of the display fabrication process.

2.2 Pixel Drive TFTs

A TFT is a type of field-effect transistor that is formed from thin films of semiconductor, insulator, and electrode (gate, source, and drain). In this section, we describe the TFTs needed to drive the pixels in flexible screens. TFTs can be broadly divided into two types, silicon- and oxide semiconductor-based, depending on the semiconductor material used in their fabrication. The former can be divided into amorphous silicon (a-Si) and low-temperature polycrystalline silicon (LTPS). Currently, a-Si TFTs are widely used in large-screen LCDs because they can be formed over large areas. However, since the carrier mobility, defined as the ease of motion of electrons (or holes) in the a-Si TFTs, is low (around 0.5 cm²/Vs) and the current they can carry is small, it is difficult to use these to drive OLED displays.

In contrast, LTPS-TFTs have a high carrier mobility of 50 to 100 cm²/Vs, and are employed in small flexible OLED displays, particularly in smartphones. Significantly, since their high carrier mobility means that a large current can be used even if the TFT area is small, they are also used as drive elements in high-resolution displays. Furthermore, LTPS-TFTs can also be formed in small areas not only in the pixel circuits, but also in the peripheral circuits used for driving the pixels, which provides the additional benefit of reducing the bezel (frame) size of the display. However, since the LTPS fabrication process requires laser irradiation to form polycrystallinie silicon, and this process does not support large area formation, it is difficult to apply it to large-screen displays.

Among other candidate oxide semiconductors, In-Ga-Zn-O (IGZO)²⁾, which was developed by a research group at the Tokyo Institute of Technology, is well-known. A notable feature of oxide TFTs is that their off-current is low compared to LTPS-TFTs, which makes it possible to suppress the data voltage drop caused by leakage current from the storage capacitor in the pixel equivalent circuit. This is also effective for decreasing the refresh rate of the storage capacitor, thus reducing power consumption. Furthermore, since oxide semiconductors can be deposited by sputtering^*3, which is a process that can be used to deposit films over large areas, and have a carrier mobility of around 10 cm²/Vs, which is sufficient for driving OLEDs, they are suitable for large-screen OLED displays. Because of these advantages, oxide semiconductors are used in various displays, from small to large, and are also attracting R&D attention in other areas.

Low-temperature polycrystalline-silicon and oxide (LTPO) technology reported by Apple, which combines LTPS-TFTs with oxide TFTs, has also been attracting attention recently³⁾. Since this technology makes it possible to utilize the high mobility advantages of LTPS-TFTs and the low off-current advantages of oxide TFTs, it can be used to make display bezels narrower and reduce power consumption by driving at a lower refresh rate. This LTPO technology is currently used primarily in smartwatches.

Separately, research is being pursued into forming oxide semiconductors by coating and printing processes with the aim of further increasing screen sizes and reducing the cost of flexible displays. If oxide semiconductors could be formed using a coating process that does not require the vacuum equipment that is needed for sputtering, it would make it possible to form large-area films regardless of equipment size limits, which would lead to reduced equipment and maintenance costs. Furthermore, since using a printing process makes it possible to form the semiconductor only on the required areas, it is expected to have reduced environmental impact.

In recent years, there has also been an abundance of research into stretchable TFT circuits with the aim of implementing stretchable displays. Two current methods for making TFT circuits stretchable involve forming all of the components that make up the circuit using stretchable materials, as reported by a research group at Stanford University⁴⁾, and a method of designating stretchable and non-stretchable areas on the substrate and then applying stretchable material to only the designated stretchable areas, as reported by a research group at the University of Tokyo⁵⁾, In the former case, the semiconductor, insulator, and electrode materials that make up the TFT all need to be made stretchable. Although there has been an abundance of research into stretchable TFT materials, it has been difficult to achieve the same carrier mobility and reliability as conventional TFTs due to material limitations.

In contrast, in the case of the latter method, where stretchable areas are specifically set, a method has been proposed where the areas that make up the TFTs are not stretchable, and only the areas between adjacent TFTs are made stretchable. By making the substrate itself stretchable and the fixed TFT areas non-stretchable, the TFTs can be made of conventional non-stretchable materials. However, problems may occur because stretchable wire materials have higher resistance than regular metal materials, so R&D is being pursued into fabrication processes, in addition to the development of materials such as carbon nanotubes and metal nanowires with the aim of reducing such resistance.

2.3 OLEDs

Ultra-thin, light-emitting OLEDs are effective light-emitting elements suitable for flexible displays. As shown in Fig. 3 (a), flexible OLEDs have a structure where an organic layer containing a light-emitting material (known as the light-emitting layer) is sandwiched between an anode and cathode. Next, holes injected from the anode, and electrons injected from the cathode recombine in the light-emitting layer to produce light. Materials with a work function (WF)^*4 of around 5.0 eV^*5, such as indium tin oxide, are widely used to produce anodes, and since there are numerous organic semiconductors that have around the same ionization energy (IE)^*6, holes can be easily injected from the anode.

In contrast, whereas the WF for the high conductivity materials used for the cathode, such as silver (Ag) and aluminum (Al), is around 4.0 eV, the electron affinity (EA)^*7 for many organic semiconductors is less than 3 eV, thus making it difficult to directly inject electrons. If electrons cannot be injected, it not only greatly increases the voltage needed for OLED light emission, it also throws off the balance of holes and electrons inside the OLED, which greatly reduces the light emission efficiency and drive stability. Hence, efficient electron injection from the cathode to the organic semiconductor is essential for realizing low-power, long-life OLEDs. Currently, alkaline metals such as lithium (Li) and cesium (Cs), which have a WF of around 2 to 3 eV, are typically used in the cathode to make electron injection easier. However, when a typical flexible substrate such as a plastic film is used, moisture in the air penetrates the substrate and degrades the alkali metal^{6) 7)}.

Against this background, there has recently been active research into technology for reducing the WF of the electrode surface by depositing a chemically stable organic semiconductor film on electrode surfaces such as Ag and Al (Table 1)^8)-13). For example, a pioneering research group at the Georgia Institute of Technology produced electrodes with a WF of around 3.1 eV using polyethylenimine⁸⁾, while a research group at the National University of Singapore reported creating electrodes with a WF of around 2.4 eV using multivalent anions⁹⁾. Separately, our institute and a research group at China’s Tsinghua University have proposed a method of reducing the WF using coordination bonds with phenanthroline derivatives and metal atoms [Fig. 3 (b), Table 1]. The coordination bond is accompanied by a small amount of negative charge transfer from the phenanthroline derivative side to the electrode side, and the WF can be controlled depending on the amount of charge transfer^{10) 11)}.

NHK STRL also reported an electrode with a WF of around 3.1 eV that was created by utilizing the new concept of hydrogen bonding between organic materials near the cathode¹²⁾, and noted that it is possible to reduce the WF to 2.0 eV by using the polarization induced by hydrogen bonding with the light-emitting layer, in addition to coordination bonding between a strong organic base and Al [Fig. 3 (c), Table 1]¹³⁾. These results showed that a WF equivalent to Cs, which has the lowest WF among the alkali metals, can be obtained. It also solved the primary problem concerning Cs, which is its weakness against moisture, by utilizing a strong organic base, thus demonstrating that electrons can be injected into all kinds of OLEDs¹³⁾.

2.4 Quantum Dots

One important property demanded of displays is color reproducibility. The International Telecommunication Union Radiocommunication Sector (ITU-R) BT.2020 recommendation (BT.2020), which defines 4K/8K super Hi-Vision, specifies an extremely wide color gamut compared to the ITU-R BT.709 recommendation for HDTV. Reproducing a wide color gamut on a display requires red, green, and blue (RGB) light sources with high color purity. In the case of green, in particular, the requirements for satisfying BT.2020 are a full width at half maximum (FWHM) of around 30 nm or less for the light emission spectrum and no extraneous light emission components. Quantum dots (QDs) have attracted attention as a high-color-purity light-emitting material that meets these requirements. Quantum dots are semiconductor nanoparticles with sizes that range from several nanometers to several tens of nanometers, which exhibit a quantum size effect^*8 in which the energy state varies depending on the particle size¹⁴⁾. This makes it possible to tune the light emission wavelength, as shown in Fig. 4.

Furthermore, the FWHM for the light emission spectrum^*9 can be reduced and higher-color-purity emission can be obtained by limiting the variations in particle diameter. The two current methods for fabricating QDs are bulk semiconductor microfabrication and crystal growth. Among those fabricated by crystal growth, materials that are obtained by chemical synthesis in the liquid phase are called colloidal QDs, and will be described below.

Figure 5 shows a schematic of a QD. Here, it can be seen that the surface of a semiconductor nanoparticle, which is called the core, is covered by another semiconductor called the shell. Since the surface has a structure surrounded by hydrocarbon-based ligands, it can be dispersed in an organic solvent. Since a thin film can be fabricated by a coating process using a dispersion of QDs, this method can be applied to large-screen displays.

The three main methods shown in Fig. 6 have been investigated as examples of the application of QDs to displays. The first method uses QDs as a color conversion material in the backlight unit of an LCD (QD backlight LCD method), as shown in Fig. 6 (a). A white light source is obtained using a combination of blue LEDs as a blue light source with green and red light sources produced by the photoluminescence (PL)^*10 of QDs excited by the blue LEDs. In the second method, green and red pixels are obtained by arranging QDs that absorb blue light and convert it to green and red light over a blue microLED (μLED)^*11 or OLED, as shown in Fig. 6 (b). The third method uses EL obtained by injecting a current into blue, green, and red QDs, as shown in Fig. 6 (c). Since there is no efficiency loss due to wavelength conversion, the third method is expected to have the highest light utilization efficiency.

Next, we will explain the development of quantum dots. Initially, research was pursued into cadmium-based materials such as cadmium sulfide (CdS) and cadmium selenide (CdSe), which offered highly efficient, high-color-purity, light emission. A research group at Zhejiang University reported that high-color-purity light emission with a FWHM in the emission spectrum of 30 nm or less could be obtained by QD EL elements that use CdS or CdSe in the light-emission layer¹⁵⁾. However, these QDs are unpopular due to the toxicity of cadmium, and there are demands for switching to cadmium-free material for practical applications. Currently, although the color purity is not as good as cadmium materials, the efficiency is significantly improved. Indium phosphide (InP) and zinc selenide (ZnSe) have been studied as important cadmium-free material candidates, and Samsung Electronics has reported the production of red and blue EL elements^{16) 17)}. Our institute has also been pursuing research into cadmium-free QD Light-emitting Diode (QD-LED) for application to flexible displays, and has noted that the light emission efficiency can be improved by depositing a film containing a mixture of QDs and an appropriate electron transport material^{*12 18)}.

Chalcopyrite-based materials such as copper indium disulfide (CuInS₂) and silver indium disulfide (AgInS₂) have also attracted attention as candidate materials. Although it is thought to be difficult to obtain high-color-purity light emission from chalcopyrite-based materials because light emission is mainly associated with defects and has a wide spectral width, research groups at Osaka and Nagoya Universities recently reported obtaining narrow spectral width band edge emission^*13 from AgInS₂ QDs¹⁹⁾, and this is expected to become a new QD material. Our institute also created a prototype QD-LED using AgInS₂ and obtained green light emission with a narrow FWHM²⁰⁾.

A prototype display using QD-LEDs as pixels was also created, and there is active research with reported examples of a 55-inch 4K display in which the pixels consisting of red, green, and blue quantum dots were painted separately by the inkjet method (BOE Technology)²¹⁾ and a color display in which the pixels consisting of red, green, and blue quantum dots were formed by fine patterning by photolithography (Sharp Corporation)²²⁾.

2.5 Applications Utilizing the Properties of Flexible Displays

Next, we will describe applications that utilize the properties of flexible displays²³⁾. Figure 7 shows a summary of flexible display shape examples. These can be divided into two fixed shape classes (Fig. 7 (a) to (c)) and a case where the shape is actively changed (Fig. 7 (d) and (e)).

Even when used in a flat state, as shown in Fig. 7 (a), if a plastic film that is thinner and lighter than a conventional glass substrate can be employed, it provides the significant advantages of a thinner, more lightweight display that is difficult to break even when subject to impact. These are important properties for mobile devices that demand reduced weight and improved impact resistance. Furthermore, for large applications such as televisions, such displays can be easily attached to a wall like a poster, as described earlier, which increases the amount of freedom when considering installation locations.

A curved display with a constant curvature across the entire screen, as shown in Fig. 7 (b), can prevent viewing distance differences and viewing angle differences between the center and edges of the screen when curved in a concave shape. These factors are expected to increase the image quality and sense of immersion. Furthermore, if curved in a convex shape, such devices can be expected to find applications such as wristwatch design improvements, wearable devices wrapped around a person’s arm, and signage when installed on cylindrical-shaped pillars.

Displays have also been proposed that provide a mechanism for switching between flat and curved shapes and even for freely changing the panel curvature radius depending on the viewing style. In particular, there are demands for applications used in vehicle-mounted displays that have a variety of curved areas, and the adoption of curved shape panels is expected to grow in the future. Locally curved shapes like those shown in Fig. 7 (c) are employed in smartphones. By folding the edges of the panel, the bezel can be made as narrow as possible when viewed from the front, and the area occupied by the screen can be increased. Furthermore, images can be displayed continuously not only on the front but also on the sides.

When the display is used in a fixed shape, as shown in Fig. 7 (a) to (c), it can be covered by a hard material such as glass even after formation on a plastic film, which makes it easy to obtain good sealing properties. In contrast, when used in variable shapes like those shown in Fig. 7 (d) and 7 (e), there is a need for better sealing and durability against repeated bending. Although this feature drastically increases the technical difficulties, it has the major advantage of improving device storability. Foldable type displays, which can be folded up like the one shown in Fig. 7 (d), have already been applied commercially to smartphones.

However, since it is difficult to avoid damage to the TFT, light-emitting elements, and plastic film if the display is completely folded up like origami, a mechanism is generally employed to limit the curvature of the bendable parts. Since this curvature has an effect on the overall device thickness when it is folded up, it is preferable to fold with a smaller curvature, and the required curvature radius will differ depending on whether or not the display folds at the front or at the back. Furthermore, since bending applies stress to the devices and wiring inside the display, research is ongoing into display structure designs, thinner displays, and stress dispersion methods.

The rollable format shown in Fig. 7 (e) is designed to be rolled up and stored when the display is not being viewed and spread out when it is being viewed. In the case of rollable displays, one problem is the strain that occurs due to the difference between the inner and outer diameters when the two substrates that make up the display are rolled up. This indicates that it is important to create thinner displays and design their structures to minimize such stress.

There is also active R&D into stretchable displays that use substrates that are not only flexible and bendable, but also stretchable. Stretchable displays can also be divided into the two cases of using a fixed shape (Fig. 8 (a)) and actively changing the shape (Fig. 8 (b)). If a highly stretchable material can be used as the substrate, it is expected that 3D shape displays that can be bent in multiple directions, like the spherical shape shown in Fig. 8 (a), could be realized. A method that uses a thermoplastic resin as the substrate for realizing 3D shapes has been proposed. In this case, after the display is fabricated, the resin is softened by heating the substrate, and the shape is formed by pressing it into a die in a vacuum.

However, to actively change the shape, as shown in Fig. 8 (b), the shape of not only the video but also the display itself must be freely changeable, and this is expected to produce a representation as if the object was actually there (free shape). Furthermore, this feature is expected to stimulate applications in wearable displays that can support the motion of muscles and joints due to their high stretchability, even when the devices are attached to the user’s skin. Although stretchable displays face numerous technical challenges, from the display structure to the development of the stretchable material, they offer even more appealing characteristics than flexible displays and will become the focus of future research trends.