Transparent electrode of organic light emitting diode (OLED) top emitting device

The original OLEDs were all bottom-emitting devices. The structure of the device from top to bottom is: opaque metal cathode / organic functional layer / transparent anode, light from the anode, hence the bottom emission, as shown in Figure 1 (a) Show.

Figure 1

(Color online) Bottom (a) and top (b) emission OLED

Figure 2

Electrical model (a) and optical model (b) for DMD electrode

In the active display, the OLED light-emitting device is controlled by a thin film transistor (TFT). Therefore, if the device emits light in the form of a bottom emission, the light is blocked by the TFT and the metal line on the substrate when passing through the substrate, thereby affecting the actual Luminous area. If the light exits from above the device, the circuit design of the substrate will not affect the light-emitting area of ​​the device. The operating voltage of the OLED is lower at the same brightness, and a longer service life can be obtained. Therefore, the top emitting device is the first choice for active display such as small screens such as mobile phones. The structure of the top emission device is: transparent or translucent cathode / organic functional layer / reflective anode [5], as shown in Figure 1 (b). In top-emitting devices, the choice of transparent electrodes is the most important, and a suitable transparent electrode will greatly improve the performance of the device.

Transparency and conductivity are two important parameters for evaluating transparent electrodes. The light transmission performance is determined by the film transmittance T, which can be measured by a spectrophotometer; the conductivity is commonly characterized by the square resistance Rs, which can be measured by the four-point resistance test method. For the transparent electrode, good light transmission performance and excellent electrical conductivity are often not satisfied at the same time, and comprehensive consideration is needed. The parameter for characterizing the photoelectric comprehensive performance is ΦH=T10/Rs[6], where Rs is the square resistance of the film, usually Need to reach the order of 10–2 to meet the application needs. The following is a brief introduction to the development of top-emitting transparent electrodes in OLEDs for the light transmission and conductivity of various types of electrodes.

1 transparent conductive oxide (TCO) electrode

1.1 Indium tin oxide (ITO)

Conductive metal oxide, the most commonly used is ITO, its work function is about 4.5 ~ 4.8eV [7], generally used as a conductive material for the anode, is a fairly stable, conductive and transparent material. Its resistivity is about 1×10–3 to 7×10–5 Ωcm, and the transmittance in the visible range is close to 90%. Therefore, the cathode of the first top-emitting OLED device is ITO [8].

Usually, ITO is deposited on a glass substrate by magnetron sputtering. During the film formation process, high-energy ions continuously hit the glass substrate, and finally form a crystalline conductive film with uniform density and excellent light transmittance [9]. However, when the organic functional layer film is pre-deposited on the substrate, the bombardment of the high-energy particles will seriously damage the organic layer, causing irreversible deterioration of the performance of the device. In order to solve this problem, a buffer layer is introduced between the organic layer/ITO. The buffer medium layer can be divided into two types: inorganic layer and organic layer.

(i) Inorganic barrier layer. In 1996, Gu et al. [8] first used 10 nm Mg:Ag (30:1) plus 40 nm ITO as the top-emitting cathode, and the transmittance was about 70% in the visible range, in 8-hydroxyquinoline aluminum. The transmittance of the luminescence peak of (Alq3) at 530 nm is 63%. The structure of the device is: ITO/TPD (20 nm) / Alq3 (40 nm) / Mg: Ag (10 nm) / ITO (40 nm) (TPD is N, N'-Bis (3-methylphenyl)-N, N'-bis (phenyl)benzidine), because it is a transmissive device, it can emit light from above and below. The intensity of light on each side is about 500?cd/m2 (10V working voltage), and the external quantum efficiency is 0.1%, which is lower than the same. The structure of the traditional bottom emitting device is about 0.25%. Mg and Ag are co-evaporated to the top of the organic layer, and the thickness is smaller than the skin depth of the light, which is used to enhance the injection of electrons while protecting the underlying organic layer. In order to avoid the damage of the organic layer caused by sputtering of ITO and the short circuit of the electrode, the sputtering power used is only 5W, and the deposition rate is only 0.05/s, so the sputtering of 40nm ITO is more than 2h, even low power. During sputtering, the device also has a large leakage current. During the sputtering process, Mg is oxidized, so that the resistance of the Mg:Ag/ITO interface is increased, and the ignition voltage is increased by 3V compared with the conventional bottom-emitting OLED device.

In addition to the Mg:Ag-ITO transparent cathode, Burrows et al. [10] also studied a series of transparent cathodes of metal-ITO, such as Ca-ITO, LiF/Al-ITO. When the thickness of the metal layer is 10 nm, the transmittance of the Mg:Ag electrode and the Mg:Ag-ITO electrode is only about 50%, and the transmittance of the LiF/Al-ITO electrode is less than 20%. If it is a Ca-ITO electrode, The maximum transmission rate is over 80%. Further, the sputtering process using an Ar plasma, possible to reduce damage to the organic layer [11]. When a sputtered atom passes through an Ar plasma, high-energy atoms are scattered multiple times to reduce energy. Therefore, increasing the pressure of Ar (p) or the distance between the sputtering target and the substrate (L) reduces the Destruction of the organic layer. The thin layer of inorganic metal can provide good ohmic contact to the interface while providing protection to the organic layer, which facilitates the injection of carriers from the electrode to the organic transport layer. However, the thin metal layer will greatly limit the light transmittance of the electrode. When the thickness of the Mg:Ag alloy is 8 nm, the transmittance of the electrode is even less than 50%, which is a disadvantage of increasing the metal barrier layer.

Part of the transition metal oxide (TMO) can also be vapor deposited into a film to form a TMO-ITO electrode [12]. In 2008, Meyer et al. [12] studied the protective effect of WO3. Compared with the aforementioned metal barrier layer, oxide has the advantage of higher transmittance, which can effectively reduce the microcavity effect. At the same time, TMO has lift electrodes and The ability of carrier injection at the organic layer interface. In fact, the device reported by Meyer et al. is an ITO cathode/organic active layer/WO3-ITO anode inverted organic light emitting diode (IOLED). By changing the thickness of the WO3 layer (~60 nm), the device ITO/Bphen: Li (40 nm) / TPBi (5? nm) / TPBi: Ir (ppy) 3 (15 nm) / TCTA (40 nm) / WO3 (60 nm /ITO (60nm) (Bphen is bathothhenanthroline, TPBi is 1,3,5-tris(1-phenyl-1H-benzimidazol-2-yl)benzene, Ir(ppy)3 is tris(2-phenylpyridine)iridium, TCTA 4,4′,4′′-tris(carbazol-9-yl)-triphenylamine has extremely low leakage current (10–4 mA/cm2), and the transmission efficiency of the transmissive OLED exceeds 75%. 30lm/W, current efficiency is 38?cd/A.

(ii) Organic barrier layer. In 1998, Forrest et al. [13] used organic matter instead of inorganic metal as a barrier to increase the transmittance in the visible region. Three materials were selected, copper phthalocyanine (CuPc), phthalocyanine. Zinc phthalocyanine (ZnPc), a compound of ruthenium (3,4,9,10-perlyenetetracarboxylic dianhydride, PTCDA), found that ZnPc is similar to CuPc, and the energy barrier between ZnPc and CuPc and ITO is relatively large, thus reducing With the injection efficiency, the device's turn-on voltage rises from 4.2V (Mg: Ag as the cathode's top-emitting device) to 5.2V. Switching to PTCDA as a barrier, the effect will be even worse, the turn-on voltage is 20?V, and the quantum efficiency is only 1% of the device with ITO/CuPc as the cathode.

The reason why CuPc has better injection efficiency is because Cu-O bond is formed during the process of sputtering ITO, so many intermediate bands and surface states are introduced, and electron injection is easier; CuPc also acts as a protective organic layer. The effect, if the thickness of CuPc is reduced from 6? nm to 3nm, the leakage current of the device increases. In addition, the introduction of very thin Li (0.2 nm) at the interface between the electrode and the organic layer helps to increase electron injection by comparing ITO/CuPc/NPB/Alq3/CuPc/Li/ITO (NPB is N, N'-Bis -(1-naphthalenyl)-N,N'-bis-phenyl-(1,1'-biphenyl)-4,4'-diamine) and bottom-emitting ITO/CuPc/NPB/Alq3/Mg:Ag devices [ 14], found that their current-voltage curves are very similar, the current voltage is higher when the current density is above 10?mA/cm2. 2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline (BCP) can also be used as an organic protective layer [15]. In the device of the above structure, the external quantum efficiency is increased by 40% by using BCP instead of CuPc, and the electron injection and electron transport ability of BCP is better than that of Alq3 and CuPc, BCP/Li/ITO is used as the electrode, and the transmittance is in the visible light region. Nearly 90%, ηext=1.0%.

The disadvantage of the organic-ITO electrode is that the heat generated during the sputtering of ITO causes the organic matter to crystallize, which causes a change in the surface geometry, which causes the contact between the ITO electrode and the organic layer to deteriorate, and after the introduction of the organic barrier layer, Will bring a new barrier to the carrier, so that the exciton composite area moves to the cathode side, reducing the luminous efficiency.

In general, as a buffer layer, it is desirable to satisfy: (1) sufficient light transmittance; (2) certain conductivity; (3) formation of ohmic contact; (4) film formation process without destroying the organic layer; stability. Whether it is inorganic metal or organic as a barrier to block high-energy particles, it can have a good effect, reducing the leakage current of the device, but they solve the old problem and introduce new problems: the metal layer is not transparent, organic matter The introduction will bring a new barrier to carrier transport.

1.2 Other oxides

Although the performance of ITO is good, one of the main materials of ITO, In, is expensive due to low reserves; and In can diffuse from the electrode into the organic layer of the device, resulting in shortened lifetime of the device; in order to replace ITO, many other transparent Conductive oxides are prepared [16].

Similar to the composition scheme of ITO, these conductive oxides are either doped with another element in the oxide or mixed with two oxides. According to the band theory, the oxide generally has a wide band gap (greater than 3 eV). Therefore, it has a high transmittance (greater than 80%) in the visible light range, and the result of a large band gap is that the carrier concentration is low, so that the carrier concentration is increased by doping to cause transparent conductive oxidation. The material (TCO) film has both low resistivity and good light transmission. Generally, these oxides contain one or more of Zn, Gd, In, and Sn [17]. The optical and electrical properties of several commonly used conductive oxides including ITO are listed in Table 1.

Most TCO preparations require the introduction of sputtering, which faces the same problems as ITO, destroying the organic layer. In view of this, in 2006, Kim et al [27] using IZO Preparation Method Box Cathode Sputtering (BCS), to avoid sputtering plasma, prepared using the method of the TE OLED -6? V under the bias of the leakage current Very small, only 1 × 10–5 mA/cm2, but in order to get better conductivity, oxygen is introduced in the experiment, which brings new damage to the underlying organic layer.

In 2008, Meyer et al. [28] used AZO as both a cathode and an anode, and the uppermost AZO was formed by PLD. In the experiment, they adjusted the laser power to the AZO ablation threshold to reduce the kinetic energy of the particles as much as possible. Destruction of the organic layer, a transmissive OLED device having a transmittance of more than 73% is obtained, and the bottom-up structure is a glass substrate/AZO/BPhen: Cs2CO3/TPBi/TPBi: Ir(ppy)3/TCTA/WO3/AZO . The device has a current efficiency of 44 cd/A at a luminance of 100 cd/m2 and a power efficiency of 27 lm/W. Under the WO3 protective layer of 80 nm, the leakage current of the device is very low, only 3×10–5 mA/ Cm2.

In addition to hot evaporation of metals, organic matter, the solution prepared film can also be used as a protective layer. Sung et al. [29] reported that TiOx (30? nm) acts as an energy level matching layer between AZO and the light-emitting layer, which reduces the barrier of electron injection, and the titanium oxide layer prepared by solution method also effectively reduces sputtering. AZO damage to the luminescent layer. It is reported that the work function of the sputtered AZO electrode is reduced from 4.8 eV to 4.5 eV with the aid of the TiOx layer.

In 2003, Han et al. [30] used LiF/Al/ASO (Al-doped SiO) as the cathode of the top-emitting device. SiO and Al can be co-evaporated, thus avoiding sputtering. At the same time, this structure can also be used as a protective layer for sputtering ITO. The structure is a base/Al/ITO/TPD (60 nm)/Alq3 (40 nm)/LiF (0.5?nm)/Al (3 nm)/SiO:Al (30?nm) top emitting device at 20?V Under voltage, the brightness can reach 1600?cd/m2. They also studied the changes in electrode conductivity and transmittance with Al content. It was found that when the content of Al increases, the transmittance of the electrode decreases greatly, and the conductivity increases sharply, close to the conductivity of Al, in scanning electron microscopy. (SEM) It can be seen that when the Al content is increased to 54%, many isolated Al-integrated islands begin to appear in the film, with a size of about 7 nm. Further, using transmission electron microscopy (TEM) analysis, it was found that at 84% Al content, the film exhibited an fcc lattice structure of Al with a grain size of 15 nm and SiO was amorphous. With the appearance of fcc lattice, the film The density of the electronic states in the central and central regions increases, so the conductivity increases.

Lee et al. [31] used Zn-doped In2O3 (IZO) as the transparent anode of the inverted device, HATCN (1,4,5,8,9,11-hexaaza triphenylene hexacarbonitrile) 50?nm as the hole injection layer, the whole The device has an average light transmittance of 81% in the visible range. From the experimental results, HATCN as an organic material can also provide protection for the underlying active layer, while the carrier injection capacity is improved, and the maximum current efficiency of the device reaches 67 cd/A.

Through high concentration of heavy doping, metal oxide can achieve high conductivity and high light transmittance at the same time. It is a good choice as a top-emitting transparent electrode, but the preparation process often needs to introduce a certain thermal effect (thermal annealing, high energy). Particles), which makes the preparation of TEOLED devices difficult.

2 ultra-thin composite metal electrode

The cathode material is generally metal. If the cathode is required to transmit light, the most direct method is to make the cathode thin, but too thin metal has many problems. When the mean free path of electrons in a metal becomes comparable to the film thickness, there is a large amount of additional scattering when electrons move near the surface, which causes a sharp increase in resistivity [32]; at the same time, too thin metal film stress Severe, easy to break and open circuit; In addition, the cathode metal generally has a low work function and is easily oxidized, resulting in a decrease in device life. Therefore, it is often changed to double-layer metal or metal alloy. The advantage of the double-layer metal electrode is that it can protect the active metal with an inactive metal. Both layers of metal can be formed by evaporation.

The bimetal electrode is difficult to ensure the balance between conductivity and light transmittance. Generally speaking, when the square resistance is 20 Ω/sq, the light transmittance can only reach about 70%, and further reducing the thickness of the metal film will cause The conductivity is drastically deteriorated. Therefore, using a metal material as a cathode, it can only be a translucent electrode, thus affecting the light extraction efficiency. Moreover, the micro-resonator [33] forming an opaque anode/organic layer/translucent cathode in the OLED forms a series of resonance modes, which also affects the device emission.

In 2001, Hung et al. [34] designed a multi-layer metal cathode structure for the first time. The ultra-thin LiF/Al double layer was used as the electron injection layer of the Ag electrode. The high conductivity Ag can reduce the square resistance. The index matching layer increases the light. The structure of the device is ITO/NPB (75 μm) / Alq3 (75 nm) / LiF (0.3 nm) / Al (0.6 nm) / Ag (20 nm) / Alq3 (52 nm), and the current efficiency is 2.75 cd / A. , about 90% of the bottom emitting device. Alq3 (52nm) acts as an index matching layer to improve the coupling light output of the device. Riel et al. [35] also reported that ZnSe is used as the index matching layer, and the luminous efficiency of the same device is improved by 1.7 times.

Ca, Mg, Ag, and Al have lower work functions and are more suitable for electron injection. In 2004, Pode et al. [36] studied several double-layered metal transparent electrodes, including Ca (10 nm) / Al (10 nm), Ca (10 nm) / Ag (10 nm), Mg (10 nm) / Ag (10 nm), And use Ag (10nm) and LiF (0.5nm) / Al (10nm) as a reference. The reflection coefficient of Al is very high, which is not conducive to light transmission. The transmittance of Ca/Al electrode is the worst, and the transmittance of Ca/Ag device is higher than that of other devices, and the ignition voltage is only 2.75V. The transmittance of the Ca/Ag electrode is also above 70% in the entire visible light range, and the other electrode transmittances fluctuate greatly in the visible range.

Many rare earth metals (RE) have very low work functions, mostly around 3.0 eV, and Yb can even reach 2.6 eV, which is lower than alkali metal Li. According to the law of interface energy level matching, it is not difficult to infer that the interface between the rare earth metal and the organic layer can form a good ohmic contact, and the electron injection performance is excellent. In addition, the rare earth metal has a low melting point and is easy to be vapor deposited. The researchers also conducted a series of studies on rare earth metal films as transparent cathodes. In 2002, Lai et al. [37] used Yb and Ag to obtain a transparent electrode (the ratio of Yb to Ag was 2.5:1), which was the same as that of Ag, LiF/Al, Mg/Ag electrodes. Under voltage, the device with Yb/Ag electrode has the best performance, and the current density and brightness are higher than the comparison device. This can be explained by the work function. The work function of Yb is only 2.6eV, Li is 2.9eV, Mg:Ag 3.7eV, Ag 4.2eV, so Yb's electron injection function is better. In 2006, Ran et al. [38] further studied rare earth metals/Au electrodes, including rare earth metals including Gd, Sm, Yb, Dy, Er, Ce, Tb. The authors selected Sm and Yb for further research because both elements are in the high-transmittance region and the melting point is the lowest. The structure of the top-emitting device is p–Si/NPB (60 nm)/Alq3 (60 nm). ) /RE (4nm) / Au (15? nm), the device with RE as the electrode has an emission rate of about 13% (ITO is 20% to 26%), which is much larger than the Al/Au electrode. Ma et al. [39] further studied the influence of the structural composition of the transparent electrode. The device structure was ITO/NPB (60 nm)/Alq3 (60 nm)/cathode, and the cathode was Yb (4 nm)/Au (15 nm), Yb:Au. (19nm), Yb: Ag (19nm), the co-steaming rate ratio is 2.5:1. The transmittances of several bimetal electrodes are listed in Table 2, where 530 nm is near the emission wavelength of Alq3, and 10 nm Ag film is used as a reference.

It can be seen from the table that the transmittance of the co-evaporated electrode is significantly higher than that of the two materials, because there is no reflective interface in the co-evaporated electrode. In the emission wavelength range of Alq3, the transmittance of Yb:Au (19nm), Yb:Ag (19nm) is close to 80%, and the transmittance of Yb(4nm)/Au(15nm) is 62%. Although co-evaporation can increase the transmittance of the device, the final brightness is much better than that of the separately vapor-deposited device. This is because the first Yb has a lower work function than the co-evaporated electrode, which can improve the electron injection efficiency.

Although the light transmittance of the bimetal transparent electrode is slightly insufficient (about 70%), it is simple and easy to take into account the ignition voltage (electron injection capability) and the film formation process. Currently, such a cathode is mostly used in commercial production.

3 dielectric / metal / dielectric (DMD) composite electrode

DMD multilayer film structure has been widely studied as a filter and heat mirror in the 1970s. In recent years, researchers have found that this structure is also an effective way to obtain transparent conductive films in visible light. A two-layer dielectric and an extremely thin metal layer (5-20 nm) can be considered as a parallel structure [40] (the electrical model and optical model of the DMD structure are shown in Figure 2), thus making the electrode extremely low. Resistivity RS; at the same time, the high refractive index dielectric layer can effectively extract more photons by using the interference effect. In addition, DMD has a certain ductility and can be used as an electrode material for flexible OLEDs .

In 1998, Bender et al. [41] first proposed the use of a multilayer film structure of ITO-metal-ITO as a transparent electrode. At that time, they used the form of CuAg alloy as the metal layer, resulting in 5.7 Ω/sq, 83 % transparent film. Several DMD electrodes that have been used in the fabrication of OLED devices are listed in Table 3.

The preparation of ITO requires sputtering, which should be avoided in the top emission device. Therefore, the researchers focused on some of the dielectric materials (such as MoO3, WO3 and ZnS) that can be prepared by thermal evaporation.

In 2009, Yun et al. [49] studied the inverted structure of Glass/Al(30nm)/pentacene(20nm)/Alq3(30nm)/NPB(50nm)/WO3(5nm)/Ag(15nm)/ZnS(40nm). The device, in which the WO3 layer can be regarded as a hole injection layer at the same time (reducing contact resistance), the device has a starting voltage of 4.7V and a maximum current efficiency of 9.5 cd/A. In 2010, Cho et al. [50] discussed in detail the effect of WO3/Ag/ZnS interlayer film thickness in OLED devices. The thinner WO3, the thicker the ZnS, the higher the current efficiency of the device, but the higher the refractive index of WO3. (2.3), too thin WO3 will limit the angle of light.

MoO3 is a low melting point oxide (795 ° C) that can be vapor deposited into a film. In 2011, Xie et al. [47] fabricated a top-emitting white OLED (Al(100nm)/MoO3(1.5nm)/m-MTDATA using a MoO3(40?nm)/Ag(17nm)/MoO3(40nm) structure electrode. 30 nm) / NPB (10 nm) / DPVBi (15 nm) / CBP (3 nm) / CBP: (F-BT) 2Ir (acac) (7 nm) / Bphen (30? nm) / LiF (1 nm) / Al (1 nm) / Ag(1nm)/MoO3(40 nm)/Ag(17nm)/MoO3(40nm) (where m-MTDATA is 4,4',4"-tris(3-methylphenyl-phenylamino)-tripheny-lamine, DPVBi is N,N'-bis-(1-naphthyl)-N,N'-diphenyl-1,1-biphenyl-4,4'-diamine/4,4'-bis(2,20-diphenylvinyl)-1,1 '-biphenyl, CBP is 4,4-N,N-dicarbazole-biphenyl, (F-BT)2Ir(acac) is bis(2-(2-fluorphenyl)-1,3-benzothiozolato-N,C2')iridium (acetylacetonate), the color rendering index (CRI) of the resulting device is 84 at a correlated color temperature of TC=3736?K, which is superior to the bottom emission device. By optimizing the thickness of the Ag layer, the electrode is in the visible range. The average light transmittance is above 84%, which is similar to the traditional indium tin oxide (ITO). It is prepared by thermal evaporation, which avoids the damage of the organic layer caused by thermal annealing. However, the current-voltage curve of the device From the above point of view, the electrode contact resistance is still large, and the charge injection is not sufficient.

In 2015, Banzai et al. [51] prepared a structure similar to the above, MoO3 (20 nm) / Ag (x nm) / MoO3 (20 nm), compared to the work of Xie et al., Banzai et al. The thickness is reduced to 20?nm, which greatly reduces the contact resistance of the electrode and improves the injection performance of the carrier. When the thickness of Ag increases to 10 nm, the square resistance is greatly reduced to 5.8 Ω/sq, which is mainly because the Ag layer forms a continuous film from the island shape. The electrode material exhibits light transmittance superior to that of the metal Ag film, and the light transmittance exceeds 70% at 550 nm. Although the light transmission is not as good as ITO, the brightness of the device can still be comparable to traditional devices due to the low barrier between the DMD and the organic layer.

The contact between the metal layer and the organic layer is the key to efficient carrier injection. Kim et al. [48] used surface modification to make the Ag layer flatter to improve light transmittance. The specific electrode structure was electron injection layer (EIL) (1 nm) / Ag (12 nm) / WO3 (40 nm), and the EIL layer was used. Cs2CO3, Rb2CO3, Rb2CrO4, CaCrO4, LiCoO2, LiMn2O4, Li2CO3, LiF were compared. The experimental data show that the transmittance of the electrode at 950% and 90.9% is 550 nm at the time of using LiCoO2 and LiMn2O4 as the EIL layer. At the same time, the square resistance of the electrode is also very low at 5.4 Ω/sq. A redox reaction occurs between the thin EIL layer atoms and the metal Ag layer, and the Ag atoms are oxidized to make the Ag layer flatter, thereby reducing light scattering and improving electron transport performance.

In summary, when a DMD structure is used as the top-emitting OLED transparent electrode, materials close to the organic layer side need to use extremely thin (<10 nm) materials such as WO3, which have functions similar to the injection layer; Conductivity, the thickness of the middle metal layer must at least be able to form a continuous film (>10nm); the dielectric layer of the outer layer needs to use a material with a high refractive index (refractive index matching layer), which is convenient for deriving more photons and lifting Quantum efficiency.

4 nanometer material electrode

In recent years, the development of nanomaterials has opened another door for transparent conductive materials. Generally speaking, nanomaterial electrodes exhibit a regular grid or irregular mesh structure through different manufacturing processes such as printing or solution processing [52, 53]. Compared with ITO electrodes, metal nanowires, carbon nanotubes, and graphene materials also have flexibility, which can be used as electrode materials for flexible OLEDs. The concepts of flexible electronics and flexible display technology have been proposed long before the discovery of organic light-emitting diodes, but suitable substrate materials and electrode materials have not been discovered.

4.1 Metal nanowires

The metal nanowire transparent electrode is not a complete film layer. Generally speaking, it is a conductive planar network formed by randomly connecting metal nanowires. The performance of metal nanowire transparent electrodes depends on the geometry of the nanowires (length to diameter ratio, specific surface area, diameter, etc.), the relationship between the lines and the distribution of the lines.

The most striking of these is the silver nanowire AgNW of the grid structure. AgNW with random network structure exhibits excellent electrical and optical properties: less than 20 Ω/sq square resistance, more than 85% visible light transmittance; simultaneous film formation is widely used in solution methods, compared to conventional vacuum Evaporation has the advantage of cost saving and large area production, so it is considered to have the potential to replace ITO. Table 4 lists the photoelectric properties of several common nanowire films. Due to the need to use the solution processing method, it is easy to destroy the underlying organic layer, which is a problem that metal nanowires must solve in the future for top-emitting OLED devices.

Direct application of AgNW to the substrate can result in undesirable surface undulations, resulting in a large number of short-circuit currents, resulting in reduced performance. In recent years, a large amount of research has been devoted to reducing surface roughness [54-58], and three methods of heating, pressurizing and introducing media have been developed.

Wei et al. [59] treated the spin-coated AgNW/PEDOT:PSS (Poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) double-layer film by hot pressing, and AgNW occurred under the action of thermal effect and stress. Fusion, link, so the RMS roughness is reduced by 40%, greatly reducing the generation of short-circuit current. The electrode material has a transmittance of 83% at a wavelength of 550 nm and a square resistance of 12 Ω/sq.

Although AgNW has outstanding performance as a flexible and transparent electrode, its excessive cost cannot be ignored, so materials such as copper nanowires (CuNW) have also been developed [53, 60-62].

4.2 Carbon nanotubes (CNT)

Carbon nanotube materials have unique physical and chemical properties and have developed rapidly in recent years. They have made good progress in laboratory research and are widely used in solar cell, touch panel , LCD , and OLED component structures. It is possible to replace the position of transparent conductive metal oxides in the application of optoelectronic devices.

Yu et al. [65] used single-walled carbon nanotubes (SWNTs) as the anode and cathode of the device, and the ignition voltage was only 3.8V, and the electrodes were bendable, and still had good performance in the curved state. The structure of the device is PET/SWNT/emissive polymer/SWNT/PET (PET is polyethylene terephthalate), the square resistance of the SWNT electrode is 500 Ω/sq, and the average transmittance in the wavelength range of 400 to 1100 nm is 85%. The average transmittance of the two-layer electrode is about 73%, and the transmittance of the entire blue-light device is also 70%. The blue light device has a brightness of 1400 cd/m2 at 10 V and a maximum efficiency of 2.2 cd/A. When the device is bent, the maximum brightness is 1260 cd/m2, and the maximum efficiency is 1.9 cd/A.

In the same year, Chien et al. [66] used a conventional LiF/Al structure as the cathode at the bottom, and transferred the spin-coated CNTs onto PEDOT:PSS by PDMS. As a top-emitting OLED anode, the whole device was Al/ from bottom to top. LiF/PVK: PBD: Ir(ppy)3/PEDOT-PSS/CNT, wherein PVK is poly(9-vinylcarbazole) and PBD is 2-(4-biphenylyl)-5-(4-tert-butylphenyl)-1. 3,4-oxadiazole, the maximum luminance is 3588?cd/m2, and the current efficiency is 1.24cd/A. The difficulty in preparing a CNT film by spin coating is that it is difficult to obtain a film that satisfies the requirements of light transmittance and conductivity. Experimental data show that when the light transmittance exceeds 80%, the square resistance is greater than 300 Ω/sq.

In the CNT film layer, there is a disordered structure in which randomly distributed nanowires are lapped, and the scattering of light by the disordered structure can be utilized to obtain uniform angular illuminance. Freitag et al. [67] found that OLED devices using CNTs as the top electrode can exhibit white light of the Lambertian type. The CNT layer used in this work has a light transmittance of 75%, and the thin layer of CNT deposited on the glass substrate has a low reflectance to visible light (<5%, which cannot be accurately obtained). The only drawback is that the square resistance is large, about 1500 Ω/sq, and the white OLED device is made of Al/MeO-TPD: F6-TCNNQ/NPB/NPB: Ir(MDQ)(acac)=/TCTA:Ir( Ppy)3/TCTA:MADN:TPBe/NET5/NET5:NDN1/CNT, where MeO-TPD is (N,N,N',N'-Tetrakis(4-methoxyphenyl)-benzidine, F6-TCNNQ is m2,2 -(perfluoronaphthalene-2,6-diylidene, Ir(MDQ)(acac) is Iridium(III)bis(2-methyldibenzo-[f,h]chinoxalin)(acetylacetonat), MADN is 2-Methyl-9,10-bis (naphthalen-2-yl)anthracene, TPBe is 2,5,8,11-Tetra-tert-butylperylene, brightness (474cd/m2) and power efficiency (0.2lm/W) are both low. Low reflectivity The Lambertian-type light emission is due to the strong scattering effect of CNTs on the photons emitted from the active layer, which provides a feasible solution to the problem of the viewing angle distribution of the top-emitting white OLED.

4.3 Graphene

Graphene is a hexagonal two-dimensional planar network structure formed by sp3 hybridized carbon atoms. It has excellent comprehensive performance: extremely high visible light transmittance, good electrical conductivity, high quality factor, suitable work function, Good mechanical stability and thermal stability, chemical stability.

The transmittance of single-layer graphene is 97.7% [68], and the absorption in the visible range is 2.3%. The reflectivity of graphene is very low, the reflectivity of single-layer graphene in the visible range is less than 0.1%, and the reflectivity at 10 layers is about 2%. Multilayer graphene is optically equivalent to the superposition of non-contact single-layer graphene. The transmittance and absorption are linear with the number of layers. The 4-layer CVD graphene still has a transmittance of about 90%. Transmittance of commercial ITO, FTO and AZO.

The carrier mobility of graphene is very high (>20000 cm2/Vs), even if the carrier concentration is low relative to ITO (~2'1011cm-2), it still retains high conductivity; ultra-thin single layer The thickness and lower carrier density increase the light transmission properties of the material (about 98%). At the same time, graphene also has excellent mechanical properties: tensile strength 130? GPa, Young's modulus 1 TPa, in the application of flexible devices, compared with the traditional TCO film has more advantages.

In summary, graphene is an ideal transparent conductive material, and its work in organic solar cells and OLED devices has been reported in recent years [69-71]. The initial preparation methods of graphene are mainly mechanical stripping method and epitaxial growth method, but these two methods are not suitable for large-area low-cost preparation of devices. Wu et al. [72] prepared a graphene film by spin coating, and then used it instead of ITO as an anode to prepare an OLED device. Its performance is close to that of ITO as an anode. However, in order to reduce the square resistance of graphene, it needs to be annealed at a temperature of 1100 ° C after spin coating. Most of the organic layers have a lower glass transition temperature, and the transparent cathode needs to be formed on the organic layer, and annealing will destroy the organic Layer, so graphene can now only be used for transparent anodes. If you can improve the production of graphene, then graphene will also be a good choice for transparent cathodes.

5 Summary

Transparent conductive materials play an important role in the field of information and energy technology. In the past few decades, in order to realize the advantages of OLED, such as transparency and flexibility, researchers have optimized various types of transparent electrodes.

At present, the transparent electrodes applied to the top-emitting OLED can be roughly classified into four types, one is a transparent conductive oxide electrode, including the most common transparent ITO for bottom emission, and the transparent conductive oxide film is in the visible light range. Has a high transmittance, can meet the needs of transparent electrodes, but the transparent conductive oxide has a relatively high work function and poor electron injection capability, so the transparent conductive oxide is used as a transparent cathode, and the device's ignition voltage and operating voltage are Both are higher.还有就是成膜工艺上， 由于其高熔点， 一般说来采用对有机层破坏较小的真空蒸镀很难获得透明导电氧化物薄膜．

另一类透明电极是薄层Ca、Mg、Sm等低功函再加上高功函数保护的复合金属， 相对于其他各类透明电极，这类材料与有机层间接触电阻很小， 适合得到低启亮电压的器件． 但是金属对可见光有很强的吸收， 即使很薄其透过率也比较低， 并且低功函的金属容易被氧化， 需要用Ag、Au来保护， 电极的透光率很难达到较高（80％）的水平． 同时， 金属膜较高的反射率会增强OLED器件中的微腔效应， 使出光的颜色和强度随视角改变而产生严重的漂移．

第三类是DMD结构的透明电极， 一般选导电性高且反射率较低的Ag作为中间层， DMD结构可以同时满足导电性与透光性， 同时还兼具柔性， 适用于柔性基底， 最大的优势在于不需要借助高温， 但Ag等金属原子在工作过程中容易结块， 导致薄膜的连续性被破坏， 有可能影响到器件的长期使用． 目前看来， 如果控制好DMD与活性层间的接触电阻， 它将会是未来顶发射透明电极最好的选择．

最后一类是纳米材料电极， 为了获得足够的导电性， 纳米线电极在成膜之后需要进行加热、加压等后处理，这就给有机层带来破坏． 目前多采用层压法等转移的方法将制备好的纳米材料薄膜转移至有机层上方， 这样受限与材料间较差的黏附力， 有机层与电极间很难形成充分的接触， 给电荷的注入带来不利的影响， 因此寻找合适的转移方法至关重要．

总体来说， 作为顶发射OLED的透明电极的材料， 需要从透光性、导电性和接触电阻3个方面考虑， 目前常见的4类电极性能列于表5．

近年来， 导电聚合物薄膜作底发射OLED透明阳极的研究相继发表［73～75］， 量子效率也已经可以和ITO器件相媲美， 但聚合物薄膜基于溶液法的成膜工艺会在顶发射器件中引入溶剂分子， 严重影响器件的使用寿命，更加适合的工艺还有待开发。