Flexible Organic Light-Emitting Diodes Using Carbon-Based Transparent Electrodes

Lei Wu; Lihui Liu; Shufen Chen

doi:10.7536/PC200677

Progress in Chemistry ›› 2021, Vol. 33 ›› Issue (5) : 802-817 DOI: 10.7536/PC200677

research-article

Scan to read & Share

Flexible Organic Light-Emitting Diodes Using Carbon-Based Transparent Electrodes

Lei Wu ¹
, Lihui Liu ¹^,^*
, Shufen Chen ¹^,^*

Author information +

History +

PDF

Abstract

Due to the consuming upgrade and development of 5G technology, the future display technology focuses on ultra-high resolution, large-area, light weight, flexibility and low-cost. Organic light-emitting diodes(OLEDs) have attracted considerable attention as one of the most promising next-generation display technologies due to the advantages of self-emitting, ultra-thin thickness, low electric power consumption, large-area, and high flexibility. Flexible transparent conducting electrodes(TCEs) are primary to achieve the flexible, foldable and wearable OLEDs. However, novel TCEs are required to replace traditional indium tin oxide(ITO), due to its inherent brittle nature and steadily rising price. Carbon-based materials are the most promising alternate flexible TCEs because of their abundant resources and simple, cost-effective fabrication processes. Among the carbon-based materials, one-dimensional(1D) carbon nanotubes, 2D graphene and 3D interpenetrating network conducting polymers are drawing extensive interest owning to the excellent optical transparency, conductivity, flexibility and chemical functionalization characteristics, leading to remarkable achievements in optoelectronic devices as TCEs. In this review, carbon-based flexible TCEs, including carbon nanotubes, graphene and conducting polymers, are introduced, including the basic optoelectronic properties, preparation methods, and pattern technologies. Furthermore, a comprehensive overview of recent research progress for the flexible OLEDs using carbon-based TCEs are presented and summarized. Finally, we address the key challenges in current scale production and applications, and provide some potential proposals for future flexible OLEDs.

Contents

1 Introduction

2 Carbon-based flexible transparent electrodes

2.1 Carbon nanotube electrode

2.2 Graphene electrode

2.3 Conductive polymer electrode

2.4 Pattern methods

3 Research progress of FOLEDs using carbon-based electrodes

3.1 FOLED based on carbon nanotube electrode

3.2 FOLED based on graphene electrode

3.3 FOLED based on conductive polymer electrode

3.4 FOLED based on composite electrode

4 Conclusion and outlook

Graphical abstract

Keywords

flexible OLED / carbon nanotubes / graphene / conducting polymer / electrode patterning technology

Cite this article

Download citation ▾

Lei Wu, Lihui Liu, Shufen Chen. Flexible Organic Light-Emitting Diodes Using Carbon-Based Transparent Electrodes. Progress in Chemistry, 2021, 33(5): 802-817 DOI:10.7536/PC200677

登录浏览全文

4963

注册一个新账户忘记密码

1 Introduction

As early as 1963, Pope of New York University discovered the electroluminescent phenomenon of organic material single crystal anthracene, but its practical application was limited by the shortcomings of large thickness and high driving voltage, which did not attract people's attention at that time^[1]. Until 1987, Deng Qingyun, a Chinese-American scientist at Kodak, invented the organic light-emitting diode (OLED) with a sandwich structure^[2]. The OLED with this structure has the characteristic of low driving voltage, and the brightness reaches the 1000 cd·m^-2 at the voltage of 10 V, which arouses people's exploration of OLED. In 1992, Heeger et al. First fabricated flexible OLED devices using polyethylene terephthalate (PET) as a flexible transparent substrate^[3]. With the continuous development in recent decades, OLED display devices have reached the level of commercialization^[4]. OLED has a series of advantages, such as light weight, wide color gamut, wide viewing angle, high brightness and low power consumption, so it has been widely used in display, lighting and other fields^[5,6]. Currently, indium tin oxide (ITO) is commonly used as the bottom anode in such conventional OLED devices^[7]. The electrode has the advantages of high transmittance, low surface resistance, good conductivity, low surface roughness and work function matching, and is one of the best anode materials in traditional OLED devices^[8]. However, the mechanical endurance of ITO electrode material is poor, the texture is brittle, and it will break under a small tensile force^[9]. Moreover, indium and tin in the material are expensive rare metals, which limit the application of ITO in flexible optoelectronic devices.

Flexible OLED not only has many advantages of traditional OLED, but also has more portable, more transparent, foldable, stretchable, bendable and other advantages that traditional OLED does not have. With the continuous development of network information, terminal display is also constantly upgraded, and the development of flexible OLED is the trend of the times. Nowadays, flexible OLED has a wide range of applications in wearable display devices, electronic skin, foldable display devices, biosensing display, soft robots and so on^[10⇓~12]. The latest generation of foldable mobile phone screens, such as Samsung and Huawei, all use flexible OLED displays. At present, the methods to make OLED display devices flexible are mainly through the selection of substrate and electrode materials, the design of the structure and the optimization of the overall structure^[13⇓~15].

In many methods, the choice of flexible electrode material is the most critical. Because the mechanical properties of the electrode material will directly affect the flexibility of the device, at the same time, the conductivity of the electrode material will also affect the performance of flexible OLED devices. At present, flexible transparent electrode materials have a wide range of choices, mainly including metal oxides, metal nanomaterials and carbon-based materials, such as carbon nanotubes, graphene, conductive polymers, and composite electrode materials^[16]. Among them, metal nanomaterials have the advantages of good conductivity and high transmittance. However, the existence of nodes between the metal nanomaterials and the overlapping accumulation will make the surface roughness of the film higher, thus affecting the performance of the device^[17]. At the same time, the preparation process of this kind of electrode materials is complex, expensive and difficult to produce on a large scale, which will not be conducive to the practical application of industrialization. Carbon-based electrode materials have the advantages of high light transmittance, solution preparation, simple preparation process and compatibility with large area roll-to-roll preparation technology. In this paper, we will focus on the preparation and patterning methods of carbon-based electrode materials, as well as the research progress of flexible OLED based on carbon-based electrode. Finally, the status and challenges of flexible OLED are summarized, and the future development trend of flexible OLED is prospected.

2 Carbon-based flexible transparent electrode

In flexible OLED devices, flexible electrodes are generally deposited on the substrate in the form of thin films, which play a role in accessing external power and injecting carriers, and are an essential part of OLED devices. Therefore, the research of flexible electrode materials has been a hot direction in the field of flexible OLED. In general, there are several requirements for the selection of the electrode material. First, the surface roughness (RMS) of the electrode material film is required to be as small as possible, generally less than 10 nm. Because the RMS of the electrode material film will directly affect the morphology of the functional layer deposited on it, thus affecting the performance of OLED to a certain extent. Secondly, the conductivity of the electrode material is required to be as high as possible, while in OLED, the conductivity of the electrode is generally measured by the area resistance (R_s) or the conductivity (σ). High-performance OLEDs usually require the R_s of the electrode to be below the 100Ω·sq^-1, because the conductivity of the electrode material will greatly affect the performance of OLED devices. Then, the higher the light transmittance (T), the better, and generally, T is required to be 75% or more. Because the size of T will affect the light output of the OLED device, thereby affecting the brightness of the device. However, T and R_s are a pair of contradictory parameters. Since the larger the thickness of the electrode, the smaller the R_s, but the smaller T, a balance between T and R_s is required. Finally, the most important requirement is that the mechanical resistance of the flexible electrode material is good. Because with the change of mechanical stress, its R_s and T will not change significantly to ensure the stability of flexible OLED performance under bending, stretching, deformation and other conditions. The mechanical endurance of an electrode is generally measured by the change in conductivity (R_s or σ) of the electrode under bending deformation.

Commonly used carbon-based flexible transparent electrode materials can be divided into one-dimensional carbon nanotubes, two-dimensional graphene, conductive polymers and their composite electrode materials.

2.1 Carbon nanotube electrode

Since carbon nanotubes (CNTs) were synthesized by Iijima et al in 1991, CNTs have been widely used in various optoelectronic devices, supercapacitors, memory devices and other fields due to their excellent electrical, optical and mechanical properties^[18]^[19,20]. Especially in recent years, CNTs are often used as stretchable electrodes in various flexible and wearable devices due to their excellent mechanical properties^[21]. Chemical vapor deposition (CVD) is the most commonly used method to prepare CNTs^[22]. The method has the advantages of low cost and easy large-scale production, and CNTs with different tube diameters can be obtained by controlling the reaction temperature. A single CNT is like a fiber, and a CNT electrode material is generally composed of many fibrous CNTs in a vertical array, a horizontal array, or a random arrangement^[23]. Due to the strong van der Waals force between dense CNTs, CNTs are prone to aggregation, agglomeration, poor water solubility and dispersibility, thus affecting the quality of film formation^[24]. This can be modified by introducing hydroxyl or carboxyl groups on the surface of CNTs, thus improving the water solubility and dispersibility of CNTs, which is conducive to the spin-coating preparation of membrane electrodes^[25,26].

CNT is used as a flexible electrode in optoelectronic devices, especially in flexible OLED, and its optical, electrical and mechanical properties are particularly important. Typically, a single CNT has an elastic modulus of up to 1 TPa and a tensile strength of up to 100 GPa, more than 10 times higher than industrial fibers. The T of the CNT film electrode is about 90%,R_s and about 100Ω·sq^-1[27]. These properties are affected by the concentration of CNT solution, the thickness of the film electrode and other factors. Baughman et al. Prepared an ultra-long CNT array sheet electrode with a width of 5 cm and a length of 1 m by CVD method. As shown in Figure 1A, the T in the visible light range is as high as 85%, and it has good mechanical properties. After 100% stretching, the conductivity does not change significantly, but the R_s is about 700Ω·sq^-1^[28]. Bao et al. prepared a stretchable CNT film electrode by spraying CNT solution on a flexible polydimethylsiloxane (PDMS) substrate^[29]. 2200 S·cm^-1(R_s=328Ω·sq^-1) and a light transmittance T = 79% was obtained. The conductivity of the obtained film electrode hardly changed after it was biaxially stretched 10,000 times. In order to obtain a CNT film electrode with better electrical conductivity and excellent mechanical properties, Lee et al. Prepared a CNT film electrode with a serpentine structure with a vertical array on the surface plated with nickel, as shown in Figure 1b^[23]. The method comprises the following steps of: firstly, depositing vertically arrayed CNTs on a SiO₂ substrate by a water vapor assisted CVD method; and then, depositing PDMS with a serpentine structure on the CNT film heated by the substrate by an inkjet printing mode, wherein the shape and the curing of the PDMS are controlled by the change of Joule heat of the substrate. After curing, the CNT thin film electrode is transferred to a flexible substrate by stripping, and then is plated with nickel to obtain a high-performance CNT thin film electrode, the R_s is as low as 6Ω·sq^-1, and the surface resistance R_s is still lower than the 100Ω·sq^-1 after 1000 times of cyclic stretching under the deformation degree of 100%. The PDMS transfer method can effectively avoid the influence of the CNT electrode processing process on the substrate. Pang et al. Used this PDMS template transfer method to prepare a CNT thin film electrode with a hexagonal grid, as shown in fig. 1C^[30].

The R_s of the obtained hexagonal grid CNT electrode hardly changes under 150% strain force. With the advent of various wearable electronic products, there are more and more studies on the integration of flexible stretchable electrodes on wearable textiles. Wang et al. integrated CNT electrodes on textiles by direct dip-coating^[31]. Zhao et al. integrated CNT electrodes on textiles by spray coating, as shown in fig. 1D^[32]. Due to the excellent mechanical properties of textile fibers, the reversible recovery deformation of CNT electrodes integrated on textile fabrics can reach 900%. After recovery, the conductivity does not change. This will greatly facilitate the recycling of wearable flexible electronic products. In a word, CNT has excellent mechanical properties and is one of the promising flexible electrode materials.

2.2 Graphene electrode

graphene is a two-dimensional carbon material formed by covalent bonding of sp², which has a hexagonal honeycomb structure and is widely used in flexible OLED. At present, the methods of preparing graphene include mechanical exfoliation, epitaxial growth, CVD, liquid phase exfoliation and oxidation-reduction^[34]^[35]^[36]^[37]^[38]. Among them, the mechanical exfoliation method is time-consuming and not conducive to large-scale preparation, which is generally only prepared in the laboratory. However, the epitaxial growth method requires a high temperature and high vacuum environment, the cost is high, and the transfer of the prepared graphene to the target substrate is difficult. The liquid-phase exfoliation method is relatively simple and low-cost, but it is easy to form agglomeration in the liquid phase due to the strong Van der Waals force between graphene sheets. The redox rule is to introduce functional groups such as hydroxyl groups into graphene by oxidation and then reduce them, which avoids the problem of agglomeration in solution^[39]. However, in the process of oxidation and reduction, it is easy to form lattice defects, which affect the conductivity of graphene. At present, high-quality graphene is mostly prepared by CVD method. Generally, the graphene prepared by CVD method needs to be transferred from the growth substrate to the target substrate in order to be applied to flexible devices. Therefore, the transfer of graphene is an important topic in the application of graphene. The transfer of graphene to the target substrate can be mainly divided into etching transfer and non-etching transfer^[40]. Etching transfer mainly uses metal as the growth substrate, and then transfers to the target substrate, and etches the metal by a certain way (such as acid, electrolysis, etc.). This method is not environmentally friendly and has high cost. While the non-etching transfer is mainly realized by introducing an intermediate supporting substrate. Generally, the support substrate is a high molecular polymer, such as polymethyl methacrylate (PMMA), PDMS, etc., and such an intermediate substrate needs to have a certain degree of adhesion. The condition of successful transfer is that the adhesion of graphene on the target substrate is greater than that of the intermediate substrate, and the adhesion of the intermediate substrate is greater than that of the growth substrate. Thereby realizing the transfer of the graphene by controlling the heating and pressurizing conditions. This method has low cost and is more environmentally friendly, but there will be polymer residues on the graphene. At present, the roll-to-roll method is usually used to transfer graphene on a large scale^[41]. The size of the single-layer graphene obtained by CVD combined with roll-to-roll transfer is up to 30 inches, the R_s of the film is as low as 125Ω·sq^-1, and the T in the visible light range is as high as 97.4%, but the graphene is easily contaminated in the transfer process. For example, there are polymer residues, amorphous carbon attachment and surface defects on the surface of the transferred graphene^[40]. This kind of contamination will greatly affect the performance of graphene electrodes. It is an important way to improve the performance of graphene electrodes by effectively cleaning the pollution caused by graphene in the transfer process. Liu et al. Recently proposed a method to clean the surface contamination of large area graphene^[33]. They effectively cleaned the contamination on the graphene surface by using the intermolecular force through the close contact between the roller wrapped by activated carbon with high specific surface area and the graphene surface, so as to obtain a large area of ultra-clean graphene, as shown in Fig. 1e and 1f. This ultra-pure graphene has a surface cleanliness of more than 99%, an ultra-high carrier mobility of up to 500 000 cm²·V^-1·s^-1, and a contact resistance as low as 118 Ω · μm.

Generally, the intrinsic graphene prepared by CVD method has low carrier concentration, high surface resistance and low work function, which leads to a high interfacial barrier between the graphene electrode and the organic material, making hole injection difficult. This greatly limits the application of graphene electrodes in high-performance flexible OLED devices. At present, there are two main methods to solve this kind of problem. One method is to use electrodes formed by stacking multiple layers of graphene by doping and modifying graphene, so as to effectively reduce the surface resistance and improve the work function. Wu et al. Doped single-layer graphene with boron, which reduced the R_s of intrinsic graphene from 350Ω·sq^-1 to 200Ω·sq^-1,T by only 0.2%. At the same time, the work function increases from 4. 7 eV to 5. 2 eV, which effectively improves the performance of the single-layer graphene electrode^[42]. Iijima et al. Wet doped graphene by HNO₃, and stacked the doped single-layer graphene layer by layer to form a four-layer graphene film electrode^[41]. The R_s of the graphene electrode is reduced from the 125Ω·sq^-1 to the 30Ω·sq^-1. In addition to the doping of small molecules into graphene, long-chain macromolecular acids have also been used to dope graphene in the past two years. Kwon et al. Doped graphene with perfluorosulfonic acid (PFSA), and the doped graphene electrode obtained ultra-high environmental stability^[43]. Compared with the intrinsic graphene electrode, the R_s is reduced by 56% and the work function is increased by 0. 8 eV. Therefore, doping is an efficient method to improve the performance of intrinsic graphene electrodes. In addition, another common method to solve the problem of high surface resistance and low work function of intrinsic graphene electrodes is to form composite electrodes through graphene and other materials (such as metal lattices, metal nanowires, conductive polymers, etc.). This method will be discussed in detail in the following sections. In a word, graphene electrode has a broad application prospect in flexible OLED devices.

2.3 Conductive polymer electrode

At present, another flexible electrode material in carbon-based electrode materials, namely conductive polymers, is considered to be the most promising material to replace rigid ITO electrodes. In particular, poly (3,4-ethylenedioxythiophene): polystyrenesulfonate (PEDOT: PSS) is one of the most popular flexible electrode materials. PEDOT: PSS has many excellent properties, such as high transmittance, low surface resistance, excellent mechanical properties, low surface roughness, good biocompatibility and solution processability, and can be patterned by inkjet printing. Because of these excellent properties, PEDOT: PSS has been widely used in flexible OLED, flexible solar cells, field effect transistors, supercapacitors, wearable flexible electronic devices, biosensors and so on^{[10,44⇓ ~46]}.

Pure poly (3,4-ethylenedioxythiophene) (PEDOT) is a hydrophobic polymer with a long chain and is difficult to dissolve in water, which hinders the solubility of PEDOT materials and greatly limits its application. Therefore, water-soluble polystyrenesulfonate (PSS) is usually added to the current commercial PEDOT: PSS solution to improve the solubility, but the insulating property of PSS limits the conductivity of PEDOT, so that the conductivity of PEDOT: PSS film without any treatment is only 1 S·cm^-1, which limits its application as an electrode in optoelectronic devices. It is necessary to improve the conductivity of PEDOT: PSS thin films by various chemical doping or post-treatment methods. For example, by doping various strong acids, weak acids, polyhydroxy compounds and ionic liquids, the σ of the PEDOT: PSS thin film can be greatly improved by three orders of magnitude from the original 1 S·cm^-1 to the 1000~4000 S·cm^-1^[47⇓~49]^{[50⇓⇓⇓~54]}^[55,56]. Acid doping treatment is the most common, but different acid treatments have different effects on improving the conductivity of PEDOT: PSS films. The most commonly used acids are hydrochloric acid, nitric acid, sulfonic acid, formic acid, concentrated sulfuric acid and so on, among which concentrated sulfuric acid has the best treatment effect. After being treated by concentrated sulfuric acid, the thickness of the film can reach more than 3000 S·cm^-1^[47]. However, concentrated sulfuric acid is more corrosive, which will cause T to decrease and affect the performance of the device in the subsequent process. Generally, the acid used for doping modification is strong acid with weak corrosiveness, such as sulfonic acid and phosphoric acid. These acid-modified polymers are in the 1000~2500 S·cm^-1[57]. Modified polyols, such as sorbitol and xylitol, are more likely to remain in the 1000~2000 S·cm^-1, and can improve both mechanical properties and wettability to the substrate^[50,52]. In the past two years, PEDOT: PSS doped with ionic liquids has attracted much attention. By doping PEDOT: PSS with ionic liquid, Bao et al. Made the σ of PEDOT: PSS film as high as 3100 S·cm^-1 and had excellent mechanical properties^[56]. After 1000 cycles at 100% strain, σ is still above the 3000 S·cm^-1. At present, there are two main reasons for the improvement of the conductivity of the modified PEDOT: PSS thin films: one is that the addition of various doping agents weakens the Van der Waals force between PEDOT and PSS chains, which leads to the separation of the two phases and the closer packing of PEDOT, thus improving the conductivity^[55]. Another explanation is that the additive protonates the PSS^- to form PSSH with a low boiling point, which then volatilizes during post-processing such as annealing. O that the content of the insulating PSS in the film is reduced, and the conductivity is improved^[47].

Another key for PEDOT: PSS to be used as a flexible electrode is to improve its mechanical properties, which can be achieved from two aspects: on the one hand, by adding certain modifying reagents, the flexibility of PEDOT: PSS can be changed essentially, so as to improve the tensile and bending resistance of the membrane electrode. Zhu et al. studied the improvement of mechanical properties of PEDOT: PSS film by TritonX-100, and found that the mechanical properties of PEDOT: PSS film could be significantly improved by adding appropriate TritonX-100^[58]. They attribute this improvement in mechanical properties to the addition of Triton X-100, which increases the free volume in the membrane and weakens the forces between polymer chains. As a result, the fluidity of the polymer chain is increased, which is manifested by an increase in elongation at break, resulting in an improvement in mechanical properties. Lee et al. improved the mechanical properties of PEDOT: PSS membranes by adding ionic liquid EMIM TCB^[55]. The results showed that the addition of EMIM-TCB decreased the tensile strength from 24. 9 MPa to 10. 1 MPa, increased the tensile strain limit from 5% to 28%, and decreased the elastic modulus from 550 MPa to 38 MPa. This improvement is also attributed to the addition of EMIM TCB, which increases the free volume of the PEDOT: PSS film, essentially acting as a plasticizer. Kim et al. modified PEDOT: PSS with polyurethane ionic liquid, which made the tensile property of PEDOT: PSS film as high as 600% and the Young's modulus as low as 7 MPa^[59]. On the other hand, by adding the polyhydroxy compound, hydrogen bonds are formed between the membrane electrode and the substrate to increase the adhesive force between the substrate and the PEDOT: PSS, thereby preventing the PEDOT: PSS from falling off from the substrate in the bending and stretching process and reducing the formation of cracks. For example, the mechanical properties of PEDOT: PSS films can be improved to some extent by adding D-sorbital, polypropylamine glucose hydrochloride (PAH-D), polypropylene, xylitol and so on^[60]^[61]^[62]^[52]. This improvement is mainly attributed to the enhanced adhesion between the PEDOT: PSS film and the substrate.

PEDOT: PSS is often used as an anode material in optoelectronic devices. The anode is used to inject holes into the HOMO orbital of the organic layer material, and the work function of the anode material is usually required to be as high as possible to reduce the hole injection barrier, so that the device performance can be effectively improved. However, the work function of PEDOT: PSS is about 4.9 ~ 5.1 eV. As an anode material, the work function of PEDOT: PSS is not ideal, which makes it one of the factors limiting the acquisition of high-performance devices. Therefore, it is very important to control the work function of PEDOT: PSS electrode. Lee et al. Regulated the work function and conductivity of PEDOT: PSS (PH 500) by PFSA and dimethyl sulfoxide (DMSO). The work function of PEDOT: PSS is adjusted to 5. 8 eV while ensuring high conductivity, which improves the performance of the device^[49]. Cao et al. Regulated the work function of PEDOT: PSS (AI 4083) by perfluoro ionomer (PFI) and n-butanol, improved the work function of PEDOT: PSS from the original 5. 3 eV to 5. 9 eV, and used it as a hole injection layer to fabricate blue OLED, and obtained 13.0 cd·A^-1 OLED^[63].

In order to reduce the manufacturing cost of flexible OLED, various research methods of all-solution preparation of OLED have been proposed. However, most of the cathode materials are prepared by vacuum evaporation, which limits the further development of all-solution devices. Recently, PEDOT: PSS has also been proposed as a cathode due to its excellent solution processability. For example, Lee et al. Used PEDOT: PSS as the anode and PDMS as the transfer solution to prepare flexible OLED^[64]. However, using PEDOT: PSS as the cathode, there is a big problem, that is, the high work function limits the injection of electrons into the organic layer, which greatly hinders the further improvement of device performance. The work function of a metal material generally used as a cathode is about 2.8 to 3.1 eV. The work function of PEDOT: PSS is about 4.9 eV. It is of great significance to explore the method of reducing the work function of PEDOT: PSS for the preparation of all-solution devices. Woon et al. mixed barium acetylacetonate (Ba(acac)₂) with PEDOT: PSS (AI 4083) in a certain proportion, and then spin-coated on ITO, and found that the work function was reduced to 3.6 eV^[65]. This is because the PSS in PEDOT: PSS and Ba form PSS-Ba, which forms a surface dipole of 1. 37 eV on the ITO surface, thus reducing the work function of PEDOT: PSS. Cai et al. Coated PEI and PEIE on the surface of PEDOT: PSS film to reduce the work function of PEDOT: PSS film^[66]. The results show that the work function of PEDOT: PSS can be reduced by 0. 3 ~ 0.8 eV by controlling the thickness of PEI coating. The reason for the decrease of the work function is also that the coating combines with PSS to form a dipole, which reduces the work function of PEDOT: PSS. In conclusion, PEDOT: PSS has great potential as a flexible electrode material. However, the conductivity of PEDOT: PSS needs to be further improved. In addition, it is necessary to find a method to control the work function of PEDOT: PSS while maintaining high conductivity. Further research in these directions will play an important role in improving device performance and reducing device manufacturing costs.

One-dimensional carbon nanotubes, two-dimensional graphene and polymer PEDOT: PSS have been used as transparent conductive electrodes in flexible optoelectronic devices. As shown in Table 1, we compared the performance of the three electrode materials with that of the conventional ITO electrode. Compared with ITO, the three kinds of carbon-based materials have excellent mechanical properties and are compatible with low-cost solution processing technology, but their electrical conductivity is poor under the condition of maintaining high light transmittance, so they need to be chemically modified or modified. In addition, carbon nanotube electrodes also face the problems of high contact resistance between tubes and high surface roughness, CVD graphene is easily contaminated during the transfer process, which affects its light transmittance and conductivity, and the acidity and hygroscopicity of PEDOT: PSS electrode itself lead to poor performance and stability of optoelectronic devices.

2.4 Patterning preparation method of carbon-based flexible transparent electrode

In flexible OLED devices, it is often necessary to pattern the light-emitting area of the OLED, as the light-emitting area determines the concept of the pixel in subsequent OLED device applications. At present, the patterning of the light-emitting area of flexible OLED is mostly determined by the pattern shape of the top electrode and the bottom electrode. The patterning of flexible electrodes is of great significance for the application of flexible OLED in the field of display. At present, due to the limitation of the work function and conductivity of the cathode in the flexible OLED device, the cathode material mostly adopts metals with low work function and good conductivity, such as Al, Ba and the like, and the anode mostly adopts ITO. The patterning of these electrodes mostly depends on various photolithography techniques and mask evaporation methods^[67⇓~69]^[14]. However, photolithography requires a series of tedious processes such as photoresist mask and etching. However, the mask evaporation method needs to be realized by a mask plate with a specific pattern, and the formed pattern has low precision and complex process, which is not conducive to the promotion of large-area low-cost OLED display products. In recent years, with the research and development of solution-machinable electrodes, various patterning methods of solution-machinable electrodes have been widely reported, such as screen printing, 3D printing, inkjet printing, nanoimprinting, directional wetting technology, and various mask coatings^[70,71]^[72]^[73⇓~75]^[76⇓~78]^[79]^[80⇓~82].

As a mature process in the printing industry, screen printing has been gradually applied to the preparation of flexible electronic devices with the extensive research of flexible electronics. Screen printing has the characteristics of large-scale and large-size preparation, simple process and low cost. This two-dimensional printing process is mainly to deposit a layer of ink on a mask steel plate with a certain pattern, and then apply a certain pressure in the form of blade coating to deposit the ink on the substrate through the holes patterned by the mask plate, so as to form the required pattern on the substrate, as shown in Figure 2a. Due to a series of advantages of screen printing, it has also been used for the preparation of patterned electrodes recently. Lai et al. prepared a grid-like PEDOT: PSS electrode with high light transmittance and excellent mechanical properties by screen printing for the first time by adjusting the proportion of ink components and optimizing process parameters, as shown in Figure 2b^[70]. A flexible OLED was fabricated based on the latticed PEDOT: PSS electrode. In subsequent work, they still used screen printing to prepare honeycomb-shaped AgNW electrodes and applied them to patterned supercapacitors, as shown in Figure 2C^[71]. Although screen printing has many advantages, this process still requires a mask, and if different patterns need to be formed, a mask of different shape needs to be replaced. Moreover, the rheological properties and other mechanical properties of the ink have higher requirements, such as the viscosity, concentration and shear rate of the ink, which need to be controlled within a certain range. And the resolution of the pattern formed by this process is not high. These shortcomings have limited the further development of screen-printed patterned electrodes.

At present, inkjet printing and 3D printing are considered to be the most promising methods for large-scale patterned device fabrication. These two kinds of patterning preparation methods are similar, do not need a mask, and control the pattern shape, spacing, accuracy and so on by setting parameters through the computer terminal, which can realize large-scale, large-area, multi-shape and complex patterning electrode preparation. However, these two patterning methods have high requirements on the ink, such as the size of ink particles, the concentration of ink, rheological properties, surface tension, viscosity, contact angle with the substrate and other properties. If these properties are not well controlled, nozzle clogging, poor pattern accuracy, and even mutual accumulation of ink droplets may occur, resulting in failure to successfully form the desired pattern. The key to inkjet printing and 3D printing of flexible patterned electrodes is the regulation of ink properties and the development of various ink formulations. Zhao et al. successfully developed a conductive ink with excellent rheological properties and good conductivity for 3D printing by adjusting the ratio of solvent to PEDOT: PSS with water and DMSO as solvents, and successfully prepared patterned electrodes through 3D printing, as shown in Figure 2D, and applied them to various flexible biosensor devices^[72]. Hong et al. successfully realized the all-solution fabrication of OLED devices by inkjet printing of patterned Ag electrodes and PEDOT: PSS electrodes as the cathode and anode of OLED devices, respectively, such as the top electrode transferred in fig. 3A to avoid the influence of solvents on the functional layer^[73]. Zhang et al realized the transfer of patterned PEDOT: PSS electrodes to the skin by using hydrogel poly (lactic acid) (PVA) as an intermediate substrate and dodecylsulfonic acid (DBSA) to regulate the adhesion, as shown in Fig. 3B^[83]. In addition, Lee et al also realized the transfer of patterned PEDOT: PSS electrodes to different substrates such as glass, copper foil and PET by using PDMS as an intermediate substrate, as shown in Figure 3C^[84]. Therefore, the method of transferring patterned electrodes can effectively avoid the damage to the underlying functional layers, and is the most promising technology for preparing high-performance patterned devices in full solution.

The directional wetting technique utilizes the difference in hydrophobicity of the substrate to achieve patterning. Lee et al. formed hydrophilic and hydrophobic stripes on the substrate surface by directional wetting technology, that is, using superhydrophobic groups combined with UV/O₃ treatment, as shown in fig. 3D^[79]. When PEDOT: PSS aqueous solution is used to cover the substrate, PEDOT: PSS is selectively deposited on the hydrophilic stripe area due to the difference in hydrophilicity. Thereby forming a PEDOT: PSS stripe pattern on the substrate as shown in fig. 3e. However, such a patterning method using different hydrophilicities can only form a simple pattern, and the accuracy of the formed pattern is not high.

Nanoimprinting and mask coating are also commonly used methods to achieve patterning. Fanzio et al. Formed PEDOT: PSS stripe arrays with different resolutions and spacings by optimizing the spacing between the stripes of the working template and adjusting the imprinting temperature^[76]. Tormen et al. Successfully prepared various PEDOT: PSS array patterns by adjusting the imprinting temperature, pressure, humidity and using water vapor to assist nanoimprinting. The resolution of this electrode array can reach 100 nm, as shown in Figure 3F^[78]. The mask coating method is a method for preparing a patterned coating by using a specific pattern film layer as a mask plate, which does not require additional instruments and is simple to operate, but can only form a simple pattern with low resolution, and is suitable for applications requiring low pattern accuracy. Arias et al. realize that fabrication of a patterned OLED device by means of a mask coat of a dielectric layer material (PVDF), as shown in fig. 3G^[82]. At present, there are many methods to prepare patterned electrodes, among which printing and printing are more promising to realize the large-scale preparation of patterned flexible electronic devices. In the process of printing and printing patterns, optimizing ink formula, adjusting ink viscosity and improving ink rheological properties are the most critical. At the same time, the development of a universal ink with high conductivity and high performance is of great significance to the large-scale preparation of patterned flexible electronic devices.

3 Research progress of flexible OLED based on carbon-based electrode

3.1 OLED based on carbon nanotube electrode

As mentioned above, CNT tubes have excellent mechanical and electrical properties, and have been reported to be used in flexible OLED. However, due to the existence of high Schottky junction resistance and easy aggregation between CNTs, the photoelectric performance of CNT films decreases, which affects the performance of flexible OLED devices. In response to this problem, Cheng et al. Reported a method to prepare high-performance transparent electrodes by welding crossed single-walled carbon nanotubes together with graphitic carbon^[85]. The prepared transparent electrode exhibits a transmittance of 90% and an area resistance of 41Ω·sq^-1 under visible light of 550 nm. And then that surface resistance is further subject to HNO₃ treatment to reduce the surface resistance to 25Ω·sq^-1. The device using the CNT electrode exhibits a turn-on voltage (V_on) of 2.5 V, a current efficiency (CE) of 75 cd·A^-1, and excellent mechanical properties. This excellent performance is mainly due to the fact that graphitic carbon transforms the Schottky contact between metal and carbon nanotubes into ohmic contact, which significantly improves the conductivity of CNT electrodes. Pei et al. Also prepared red, green and blue OLED devices by using single-walled carbon nanotubes (SWCNT) as cathode and anode, respectively, using a vacuum-free rolling process^[86]. The V_on of the tested device is 3.8 V. At 10 V, the maximum brightness 1400 cd·m^-2 is achieved, and the maximum CE is 2.2 cd·A^-1. This efficiency value is comparable to that of an OLED using a SWCNT anode and an evaporated metal cathode. When bent at a bending radius of 2.2 mm, there was no significant decrease in performance. The key factors that limit the application of CNT electrodes in flexible OLED devices are the large surface roughness, high contact resistance and poor film formability of CNT films, so the research on the application of CNT as a flexible electrode alone in flexible devices is less than that of other carbon-based electrodes.

3.2 OLED based on graphene electrode

Compared with CNT electrodes, graphene is a two-dimensional material with low surface roughness and adjustable chemical properties, which has been widely used in flexible OLED devices. How to effectively improve the device performance is the key to the application and promotion of OLED. Ren et al. prepared a graphene oxide/graphene (GO/G) vertical heterostructure as the anode of OLED by oxidizing the top layer of the three-layer graphene film through O₃. The device structure and the pictures of the device during operation are shown in Figure 4A and Figure 4B^[87]. After 10 000 bending cycles, the R_s of the GO-G heterostructure electrode is almost unchanged. The luminous efficiency of the OLED based on the GO-G heterojunction electrode reaches the 82.0 cd·A^-1 and the 98.2 lm·W^-1 (PE), which are 14.8% and 15.0% higher than those of the OLED based on the ITO anode. Chen et al. Used PSS as a modified layer to cover the surface of graphene to form a dipole, which effectively improved the hole injection ability and stability of graphene, and reduced the leakage current of OLED devices^[91]. The CE of the blue and white OLED devices based on the modified graphene electrode is as high as 201.9 and 326.5 cd·A^-1, respectively. In addition to the interface modification of graphene, chemical doping modification is also an effective means. Lee et al. Doped four layers of graphene with PFSA, which greatly improved the work function of graphene electrode from the original 4.3 eV to 5.1 eV^[88]. Compared with the OLED(CE=82.7 cd·A^-1,PE=77.6 lm·W^-1) of the original four-layer graphene electrode, the PFSA-doped graphene electrode device shows higher luminous efficiency (CE=98.5 cd·A^-1,PE=95.6 lm·W^-1). The structure of the prepared OLED device and the current density-voltage curves (J-V curves) of the device before and after doping are shown in figs. 4C and 4D. The doping of PFSA can significantly improve the electroluminescent performance of OLED devices with graphene electrode. In addition, Wu et al. Also obtained high-performance OLED devices by n-type doping of graphene with boron, which is almost defect-free and has high mobility^[92]. The external quantum efficiency (EQE) of the original graphene OLED is only 19. 5%, while the EQE of the boron-doped graphene OLED is as high as 24. 6%.This excellent performance may be due to the high work function (5. 0 eV) of boron-doped graphene, which greatly improves the hole injection ability of the device. Therefore, at present, the performance of OLED devices based on graphene electrodes is mainly improved by modifying graphene.

The surface of graphene is easy to be contaminated during the transfer process, which will also seriously affect the performance of OLED devices. In the process of graphene transfer, the traditional intermediate substrate PMMA will form residues on the surface of graphene which are difficult to remove, resulting in poor performance of OLED. In order to solve this problem, Chen et al. inserted a layer of poly [ (9,9-dioctylfluorene-2,7-diyl) -co- (4,4 '- (N- (4-sec-butylphenyl) diphenylamine)] (TFB) between graphene and PMMA as a protective layer to avoid the residue of PMMA on graphene, and TFB also served as a hole injection layer^[93]. This approach effectively elevates the CE of the OLED from the 64.8 cd·A^-1 to the 74.5 cd·A^-1. Chen et al. Further found that the Joule heat generated in the current annealing can effectively remove the PMMA residue on the graphene surface, thereby eliminating the negative differential resistance and hysteresis, so that the CE of the device is improved from 30.9 to 41.6 cd·A^-1, and the CE of the device is 89.2 cd·A^-1 after using the outcoupling prism^[94]. Zhang et al. Also reported a method of using organic small molecule rosin as an intermediate support layer to achieve ultra-clean and residue-free transfer of graphene electrodes, thus achieving high performance of flexible OLED devices^[89]. The small molecule has good solubility and weak interaction force with the graphene, so that the intermediate substrate is easy to remove, and no residue is basically realized. The CE and PE of the graphene OLED prepared based on this transfer method can be as high as 89.7 cd·A^-1 and 102.6 lm·W^-1, and the maximum brightness can reach 10 000 cd·m^-2, as shown in Figure 4E and Figure 4F. The patterning of the graphene electrode requires the use of photoresist, and a separate support layer is required during the transfer process, which increases the possibility of contamination of the graphene electrode. In order to reduce contact contamination, Sun et al. Used photoresist as both a graphene transfer support layer and a photolithographic mask layer to prepare a graphene electrode with a stripe array, as shown in Figure 4G^[90]. The RMS of the electrode is the 444Ω·sq^-1 of the 0.675 nm,R_s. Based on this patterned graphene electrode, OLED devices with single-layer, double-layer and three-layer graphene electrodes were prepared respectively, and the device structure and J-V curves are shown in Figures 4h and 4i. The maximum luminances of single-layer, double-layer and three-layer graphene devices are 17 050, 63 700 and 75 370 cd·m^-2, respectively, which are greatly improved compared with the OLED devices with PMMA-transferred graphene electrodes. In a word, there are many reports on flexible OLED devices based on graphene electrode, and the improvement of device performance is mainly through improving the performance of graphene electrode, such as reducing the pollution of graphene surface, improving the conductivity of graphene, improving the work function of graphene and other effective means.

3.3 OLED based on PEDOT: PSS electrode

As mentioned above, PEDOT: PSS electrodes are widely used in flexible OLED devices due to their excellent mechanical properties, high transparency, solution processability and other properties. At present, the conductivity of PEDOT: PSS is the main factor that limits the further improvement of the performance of PEDOT: PSS OLED devices. Although many methods for improving the conductivity of PEDOT: PSS have been described, the σ of the PEDOT: PSS thin film is basically maintained at the 10²~10³S·cm^-1(R_s (about 100~1000Ω·sq^-1). At present, the highest σ of PEDOT: PSS film is obtained by concentrated sulfuric acid treatment, and σ can reach 3480 S·cm^-1(R_s (about 46.1Ω·sq^-1)^[96]. However, compared with the traditional ITO electrode (the R_s is less than the 10Ω·sq^-1), there is still a certain gap. Improving the conductivity of PEDOT: PSS thin films is still one of the main ways to improve the performance of devices. Although concentrated sulfuric acid can significantly improve the conductivity of PEDOT: PSS film electrode, the strong corrosion of sulfuric acid will damage the flexible substrate. In order to solve this problem, Lee et al. Used PDMS as an intermediate substrate to transfer PEDOT: PSS films treated by concentrated sulfuric acid on rigid substrates to flexible substrates by using different adhesion forces^[84]. Based on the transferred PEDOT: PSS as the top electrode, a flexible OLED was fabricated. The V_on of the device is 3 V and the maximum brightness is 5500 cd·m^-2. Although the conductivity of the PEDOT: PSS is lower than that of the ITO electrode, the PEDOT: PSS electrode has a higher work function and a lower hole injection barrier than the ITO electrode. Therefore, the OLED with PEDOT: PSS conductivity improved by modification can achieve the same performance as the OLED with ITO electrode. Elschner et al. prepared red, green and blue OLED devices by using DMSO modified PEDOT: PSS (PH 500) as the anode^[97]. It was found that the PE values of the red OLED with PEDOT: PSS anode and ITO anode at 100 cd·m^-2 were comparable, 15.9 and 17.3 lm·W^-1, respectively. For blue OLEDs, the PE(4.0 lm·W^-1) of PEDOT: PSS anodes is higher than the OLED(3.3 lm·W^-1) of ITO anodes. Kim et al. Used 2-ethoxyethanol to modify PEDOT: PSS film to improve conductivity^[98]. A white OLED was fabricated based on the modified PEDOT: PSS electrode, and the performance of the device was comparable to that of the ITO reference device. The modified PEDOT: PSS electrode device has a PE of 26.0 lm·W^-1 at a luminance of 1000 cd·m^-2, which is comparable to the reference device based on an ITO electrode (PE is 26.9 lm·W^-1). Kim et al. Doped PEDOT: PSS electrode with Pyronin-B to improve conductivity^[53]. At the same time, the small molecule compound hexaazatriphenylene hexanitrile (HAT-CN) is used as a hole injection layer to significantly improve the hole injection capability of the device. The flexible OLED device based on this structure has a V_on of 6. 8 V and a CE of 26.03 cd·A^-1, which is equivalent to the performance of ITO device (V_on=7.7 V,CE=27.95 cd·A^-1). Moreover, the brightness of the flexible OLED device based on PEDOT: PSS electrode remains 80% of the initial brightness after 900 cycles of bending with a radius of curvature of 5 mm.

In addition to improving the device performance by improving the conductivity, adjusting the work function of PEDOT: PSS is also one of the ways to improve the device performance. The increase of the work function of the anode is beneficial to lowering the hole injection barrier, so that holes can be injected more effectively. Lee et al. prepared a PEDOT: PSS electrode (AnoHIL) with high work function (5.8 eV) and good conductivity by using PFSA to regulate the work function of PEDOT: PSS and DMSO to improve the conductivity, which can also serve as a hole injection layer, as shown in Figure 5A^[49]. They fabricated a green OLED device based on this electrode. The device structure is shown in Figure 5B. The maximum CE is 128 cd·A^-1 and the maximum EQE is 33%, which is significantly higher than the OLED device with ITO electrode (CE_max=88 cd·A^-1,EQE_max=20%), which is the highest record of OLED based on PEDOT: PSS electrode at present. Therefore, by improving the conductivity and work function of the PEDOT: PSS electrode, the performance of the prepared OLED device can reach or even be superior to that of a device with an ITO electrode.

The method for improving that performance of the device by improve the conductivity and the work function of the PEDOT: PSS electrode is a modification method directly aiming at the electrode. In addition, the device performance can be improved by optimizing the structure of the OLED device. For example, the outcoupling efficiency of the device is increased by optimizing the substrate structure^[99⇓~101]. Shinar et al. fabricated a green OLED based on a PEDOT: PSS electrode by depositing a multilayer PEDOT: PSS film on a glass substrate as an anode^[100]. The maximum CE of the device is up to 40% of the 118 lm·W^-1,EQE up to the 127 cd·A^-1,PE. This high performance is due to the fact that the deposited multilayer PEDOT: PSS film plays a role in enhancing light extraction, weakening the microcavity effect, and suppressing the waveguide mode optical loss in the device. In addition, the device performance can be further improved by optimizing the electron injection layer, enhancing the electron injection ability, preparing the structure of the series OLED device, and improving the carrier injection balance^[102]^[103]^[104].

PEDOT: PSS is used as anode in more and more OLED devices because of its solution processability. However, most of the cathodes are metal electrodes deposited by vacuum evaporation. However, in the past two years, in order to reduce the manufacturing cost of OLED devices. It has been reported that OLED devices can be prepared by the way of full solution and no vacuum. At present, this kind of all-solution OLED devices are mostly realized by spin-coating or printing PEDOT: PSS as the anode and printing Ag paste as the cathode^[73,105,106]. Although the cathode is prepared by printing Ag paste, which avoids the use of vacuum evaporation, the properties of this Ag ink are difficult to control, and the price of Ag is also expensive. There are also OLEDs prepared with PEDOT: PSS as the anode, graphene as the cathode, or AgNWs as the cathode^[107]^[108]. In the last two years, there has even been work on the preparation of OLED with PEDOT: PSS as both anode and cathode in full solution, and the luminescence diagram and device structure of OLED with double PEDOT: PSS electrodes prepared in full solution are shown in Fig. 5C and 5d^[64,80,109]. However, compared with the vacuum evaporation device, there is a big gap in performance. At present, among the OLED devices prepared by all solutions, the yellow OLED device prepared by Colsmann et al. With PEDOT: PSS as the anode and PEDOT: PSS and AgNWs as the composite cathode by tape assisted spin coating has the best performance, and the maximum CE reaches 19.9 cd·A^-1^[80]. The first reason for the poor performance of these all-solution OLED devices is the mismatch of cathode work function, which leads to poor electron injection ability. Secondly, in the process of spin-coating, the mutual dissolution between layers destroys the morphology of the bottom film, which leads to poor contact between layers. In addition, the film prepared by spin coating is not as compact as the film formed by vacuum evaporation. These problems can seriously affect the performance of the device. This is also the key problem that needs to be solved urgently in order to realize the preparation of high performance OLED devices in full solution.

3.4 OLED based on composite electrode

All kinds of carbon-based transparent electrodes can not meet the requirements of photoelectric devices for perfect electrodes, such as carbon nanotubes with large surface roughness and high contact resistance, intrinsic graphene with low conductivity and easy surface contamination, and the improvement of PEDOT: PSS electrode conductivity has reached a bottleneck. This limits the further improvement of OLED performance. In recent years, the performance of OLED devices based on various electrode materials is shown in Table 2. In addition to the modification of electrode materials to improve the performance, the formation of composite electrodes to improve the performance of devices is also an important way. By forming a composite electrode, the advantages of different electrode materials can be combined to optimize the performance of the electrode. Li et al. Prepared a hybrid electrode of single-layer graphene (SLG) and silver nanowires (AgNW)^[110]. AgNW can connect the grain boundaries of graphene and improve the transport of charge carriers. AgNW can reduce the R_s of SLG from 650Ω·sq^-1 to 27Ω·sq^-1 and still maintain the transmittance of 86.7%. A flexible OLED device with a maximum brightness of 15 000 cd·m^-2 was fabricated by using the composite electrode of graphene and AgNW. In addition to introducing AgNW into the composite electrode to enhance conductivity, other metals with higher conductivity may also be introduced. For example, Qin et al. Introduced copper to the graphene electrode. They used the introduced copper directly as a substrate for CVD growth of graphene. This not only plays a role in enhancing conductivity, but also can be used as a substrate to reduce the pollution problem in the process of graphene transfer^[111]. The maximum values of CE and PE of the devices based on copper/graphene composite anode are 6.1 cd·A^-1 and 7.6 lm·W^-1, respectively, which are significantly higher than those of the reference devices based on graphene anode, copper anode and ITO anode. Although this method of introducing high conductivity metal into the composite electrode can improve the conductivity of the carbon-based transparent electrode to a certain extent, it also has shortcomings, such as the decrease of transmittance and the increase of roughness. Therefore, some composite electrodes are only composited between modified carbon-based electrodes without introducing metal nanowires and nanogrids. For example, Song et al. prepared a composite electrode of graphene oxide (GO) and single-walled carbon nanotubes (SWCNT), and the surface resistance of the prepared composite electrode was 75Ω·sq^-1^[95]. A flexible OLED is fabricated based on the composite electrode, and the device structure is shown in fig. 5e. The device has a V_on of 2.0 V, a CE of 5.0 cd·A^-1,PE, and a 2.4 lm·W^-1,EQE of 1.9%, which is comparable to an OLED device with an ITO anode (V_on=2.0 V,CE=6.2 cd·A^-1,PE=2.6 lm·W^-1,EQE=2.3%). Yuan et al. Also prepared a PEDOT: PSS/CNT/PEDOT: PSS composite electrode similar to a sandwich structure, and the surface resistance of the composite electrode was 95.2Ω·sq^-1^[112]. The OLED device based on the composite electrode has a maximum brightness of 1598 cd·cm^-2 at 14 V and a maximum CE of only 1.5 cd·A^-1. The lower performance of the device may be due to the higher roughness, which results in poor contact between the functional layers. At present, there are ^[113~115] reports on improving the conductivity, mechanical properties, surface morphology, wettability, stability and other properties of single material electrodes by forming composite electrodes. However, the performance of composite electrodes formed by different materials is quite different, and the performance of devices is related to many factors. Therefore, not all composite electrodes can improve the performance of the device. In a word, composite electrodes can make the advantages and disadvantages of different electrode materials complementary, which is a method to improve the performance of devices.

4 Conclusion and prospect

In this paper, the application of carbon-based transparent electrode materials in flexible OLED is introduced, and the problems and solutions in the application are analyzed. The preparation, modification and patterning methods of several carbon-based transparent electrodes, such as carbon nanotubes, graphene and PEDOT: PSS, were introduced in detail. The development of this kind of carbon-based electrode material is of great significance to the further development and commercialization of flexible OLED. At present, flexible OLED devices based on carbon-based electrodes have been able to compete with traditional ITO electrode devices through electrode material modification, functional layer modification and device structure optimization.

At present, if flexible OLED needs to be commercialized, it still faces some other challenges besides device performance. The first is the stability and lifetime of the device. In flexible OLED, most of the functional layers are made of organic materials, especially carbon-based electrode materials, which are easily oxidized by air and water, thus greatly shortening the service life of the device. In traditional OLED devices, packaging technology is usually used to reduce the impact of external atmosphere on device life. The so-called packaging technology is to use a variety of waterproof, anti-oxygen, good sealing materials to seal the device. However, this kind of sealing material is basically an inorganic rigid material. If this kind of material is used to encapsulate the flexible OLED device, the flexibility of the OLED device will be greatly reduced, or even no longer flexible. Therefore, the development of flexible packaging materials is of great significance, or the oxidation resistance can be improved by modifying the functional layer materials of devices. Secondly, another important challenge is the cost of manufacturing. At present, except for the anode, most of the other functional layers of OLED devices with high efficiency are prepared by vacuum evaporation, which makes the cost of device preparation high. Similarly, this is also an important reason for the high price of rigid OLED displays that have been commercialized. In the previous paper, a process that can replace vacuum evaporation to prepare devices, that is, solution method to prepare functional films, was introduced. However, due to the problems of residual solvent, phase separation morphology and interlayer mutual solubility of organic thin films prepared by solution method, the performance of OLED prepared by this method still has a certain gap with that of OLED prepared by vacuum evaporation method. At present, inkjet printing is considered to be the most promising method to achieve low-cost and large-scale preparation of flexible optoelectronic devices. The key to inkjet printing is the development of various functional layer inks. The key to achieve efficient inkjet printing of flexible OLED devices lies in the development of functional layer materials and inks with orthogonal solvents to avoid the mutual dissolution of functional layers caused by solvents. At the same time, it is also very important to find and develop a kind of cathode material ink with matching work function and good conductivity for the realization of full printing OLED.

With the development of flexible electrode materials and functional layer materials with better performance and the optimization of preparation process, in the future, flexible OLED display devices that can be folded, stretched, bent and twisted will hopefully enter mass production and achieve the goal of commercialization. This kind of flexible OLED display device will be combined with various flexible sensing, flexible power supply and so on, pushing flexible electronic devices into a research boom. At the same time, these studies will greatly promote the further development of intelligent electronic devices, flexible biosensing displays, simulation robots, artificial organs and other fields.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Pope M, Kallmann H P, Magnante P. J. Chem. Phys., 1963, 38(8):2042.

[2]	Tang C W, VanSlyke S A. Appl. Phys. Lett., 1987, 51(12):913.

[3]	Gustafsson G, Cao Y, Treacy G M, Klavetter F, Colaneri N, Heeger A J. Nature, 1992, 357(6378):477.

[4]	Jou J H, Kumar S, Agrawal A, Li T H, Sahoo S. J. Mater. Chem. C, 2015, 3(13):2974.

[5]	Ma D G. Chin. J. Liq. Cryst. Displays, 2016, 31(3):229.

[6]	( 马东阁. 液晶与显示, 2016, 31(3):229.)

[7]	Li W Z. Technol. Innov. Appli., 2020,(08):135.

[8]	( 李伟章. 科技创新与应用. 2020,(8):135.).

[9]	Xu X J. Doctoral Dissertation of Institute of Chemistry Chinese Academy of Sciences, 2007.

[10]	( 徐新军. 中国科学院化学研究所博士论文, 2007.).

[11]	Giovanni L, Giovanni F C, Aurelio B, Nicolas Z M, Giuseppe C, Björn K, Wang Q, Norbert K, Stefan H, Emil J W L, Franco C. Nanoscale, 2012, 12(9):5444.

[12]	Han J W, Jung B, Kim D W, Lim K T, Jeong S Y, Kim Y H. Org. Electron., 2019, 73:13.

[13]	Wang B H, Facchetti A. Adv. Mater., 2019, 31(28):1901408.

[14]	Chen Z H, Fang R, Li W, Guan J G. Adv. Mater., 2019, 31(35):1900756.

[15]	Hao X M. Sens. World, 2019, 25(12):39.

[16]	( 郝晓明. 传感器世界, 2019, 25(12):39.)

[17]	Koo J H, Kim D C, Shim H J, Kim T H, Kim D H. Adv. Funct. Mater., 2018, 28(35):1801834.

[18]	Tao J S, Wang R P, Yu H, Chen L L, Fang D J, Tian Y, Xie J Y, Jia D M, Liu H, Wang J S, Tang F C, Song L, Li H B. ACS Appl. Mater. Interfaces, 2020, 12(8):9701.

[19]	Lim M S, Nam M, Choi S, Jeon Y, Son Y H, Lee S M, Choi K C. Nano Lett., 2020, 20(3):1526.

[20]	Kim K, Park Y G, Hyun B G, Choi M, Park J U. Adv. Mater., 2019, 31(20):1804690.

[21]	Wei B W, Wu X K, Lian L, Yang S, Dong D, Feng D X, He G F. Org. Electron., 2017, 43:182.

[22]	Iijima S, Ichihashi T. Nature, 1993, 363(6430):603.

[23]	Guo X Q, Bai Y Q, Bai Q Z, Wen N, Nie N. Carbon, 2018, (2):40.

[24]	( 郭晓琦, 白云起, 白青子, 温楠, 聂宁. 炭素, 2018, (2):40.).

[25]	Qu T Y, Sun Y, Chen M L, Liu Z B, Zhu Q B, Wang B W, Zhao T Y, Liu C, Tan J, Qiu S, Li Q W, Han Z, Wang W, Cheng H M, Sun D M. Adv. Mater., 2020, 32(9):1907288.

[26]	Li X L. Mod. Manuf. Technol. Equip., 2019, (4): 106,114.

[27]	( 李晓丽. 现代制造技术与装备, 2019, (4): 106,114.).

[28]	Su W G, Zhang C. Mod. Chem. Ind., 2019, 39(7):89.

[29]	( 苏暐光, 张策. 现代化工, 2019, 39(7):89.)

[30]	Hong S, Lee J, Do K, Lee M, Kim J H, Lee S, Kim D H. Adv. Funct. Mater., 2017, 27(48):1704353.

[31]	Yue C. Sci. Technol. Rev., 2012, 30(26):8.

[32]	( 岳臣. 科技导报, 2012, 30(26):8.)

[33]	Shen Q J, Liu X B, Jin W J. New Carbon Mater., 2013, 28(2):94.

[34]	( 申前进, 刘新斌, 晋卫军. 新型炭材料, 2013, 28(2):94.)

[35]	Zhang Y, Liu R N, Deng J H. Acad. J ZNU(Nat. Sci.), 2019, 39(2):21.

[36]	( 张燕, 刘瑞楠, 邓建华. 天津师范大学学报(自然科学版), 2019, 39(2):21.)

[37]	de Volder M F L, Tawfick S H, Baughman R H, Hart A J. Science, 2013, 339(6119):535.

[38]	Zhang M, Fang S, Zakhidov A A, Lee S B, Aliev A E, Williams C D, Atkinson K R, Baughman R H. Science, 2005, 309(5738):1215.

[39]	Lipomi D J, Vosgueritchian M, Tee B C K, Hellstrom S L, Lee J A, Fox C H, Bao Z N. Nat. Nanotechnol., 2011, 6(12):788.

[40]	Min H, Jang S, Kim D W, Kim J, Baik S, Chun S, Pang C. ACS Appl. Mater. Interfaces, 2020, 12(12):14425.

[41]	Sadi M S, Pan J J, Xu A C, Cheng D S, Cai G M, Wang X. Cellulose, 2019, 26(12):7569.

[42]	Gao Y, Guo F Y, Cao P, Liu J C, Li D M, Wu J, Wang N, Su Y W, Zhao Y. ACS Nano, 2020, 14(3):3442.

[43]	Sun L Z, Lin L, Wang Z H, Rui D R, Yu Z W, Zhang J C, Li Y, Liu X T, Jia K C, Wang K X, Zheng L M, Deng B, Ma T B, Kang N, Xu H Q, Novoselov K S, Peng H L, Liu Z F. Adv. Mater., 2019, 31(43):1970303.

[44]	Novoselov K S, Geim A K, Morozov S V, Jiang D, Zhang Y, Dubonos S V, Grigorieva I V, Firsov A A. Science, 2004, 306(5696):666.

[45]	Emtsev K V, Bostwick A, Horn K, Jobst J, Kellogg G L, Ley L, McChesney J L, Ohta T, Reshanov S A, Röhrl J, Rotenberg E, Schmid A K, Waldmann D, Weber H B, Seyller T. Nat. Mater., 2009, 8(3):203.

[46]	Li X, Cai W, An J, Kim S, Nah J, Yang D, Piner R, Velamakanni A, Jung I, Tutuc E, Banerjee S K, Colombo L, Ruoff R S. Science, 2009, 324(5932):1312.

[47]	Choucair M, Thordarson P, Stride J A. Nat. Nanotechnol., 2009, 4(1):30.

[48]	Zhao J P, Pei S F, Ren W C, Gao L B, Cheng H M. ACS Nano, 2010, 4(9):5245.

[49]	Ling B K, Que W X, Zhou K, Li S, Zhang T S. Adv. Struct. Mater., 2017, 83:177.

[50]	Ma L P, Ren W C, Cheng H M. Small Methods, 2019, 3(7):1900049.

[51]	Bae S K, Kim H, Lee Y, Xu X F, Park J S, Zheng Y, Balakrishnan J, Lei T, Ri Kim H, Song Y I, Kim Y J, Kim K S, Özyilmaz B, Ahn J H, Hong B H, Iijima S. Nat. Nanotechnol., 2010, 5(8):574.

[52]	Wu T L, Yeh C H, Hsiao W T, Huang P Y, Huang M J, Chiang Y H, Cheng C H, Liu R S, Chiu P W. ACS Appl. Mater. Interfaces, 2017, 9(17):14998.

[53]	Kwon S J, Han T H, Ko T Y, Li N N, Kim Y, Kim D J, Bae S H, Yang Y, Hong B H, Kim K S, Ryu S, Lee T W. Nat. Commun., 2018, 9:2037.

[54]	Song J, Lee H, Jeong E G, Choi K C, Yoo S. Adv. Mater., 2020, 32(35):1907539.

[55]	Khasim S, Pasha A, Badi N, Lakshmi M, Mishra Y K. RSC Adv., 2020, 10(18):10526.

[56]	Sirringhaus H, Kawase T, Friend R H, Shimoda T, Inbasekaran M, Wu W, Woo E P. Science, 2000, 290(5499):2123.

[57]	Bießmann L, Saxena N, Hohn N, Hossain M A, Veinot J G C, Müller-Buschbaum P. Adv. Electron. Mater., 2019, 5(2):1800654.

[58]	Kim T, Park S, Seo J, Lee C W, Kim J. Org. Electron., 2019, 74:77.

[59]	Jeong S H, Woo S H, Han T H, Park M H, Cho H, Kim Y H, Cho H, Kim H, Yoo S, Lee T W. NPG Asia Mater., 2017, 9(7):e411.

[60]	Nardes A M, Kemerink M, de Kok M M, Vinken E, Maturova K, Janssen R A J. Org. Electron., 2008, 9(5):727.

[61]	Oh J Y, Shin M, Lee J B, Ahn J H, Baik H K, Jeong U. ACS Appl. Mater. Interfaces, 2014, 6(9):6954.

[62]	Song W, Fanady B, Peng R X, Hong L, Wu L R, Zhang W X, Yan T T, Wu T, Chen S H, Ge Z Y. Adv. Energy Mater., 2020, 10(15):2000136.

[63]	Kim D H, Park N H, Kim T W. Nano Energy, 2019, 65:104027.

[64]	Zheng Y Q, Yu J L, Wang C, Yang F, Wei B, Zhang J H, Zeng C H, Yang Y. J. Phys. D: Appl. Phys., 2018, 51(22):225302.

[65]	Teo M Y, Kim N, Kee S, Kim B S, Kim G, Hong S, Jung S, Lee K. ACS Appl. Mater. Interfaces, 2017, 9(1):819.

[66]	Wang Y, Zhu C X, Raphael P, Bao Z N. Sci. Adv., 2017, 3(3):1602076.

[67]	Huseynova G, Hyun Kim Y, Lee J H, Lee J. J. Inf. Disp., 2020, 21(2):71.

[68]	Luo R B, Li H B, Du B, Zhou S S, Zhu Y X. Org. Electron., 2020, 76:105451.

[69]	Kim N, Lienemann S, Petsagkourakis I, Alemu Mengistie D, Kee S, Ederth T, Gueskine V, Leclère P, Lazzaroni R, Crispin X, Tybrandt K. Nat. Commun., 2020, 11(1):1.

[70]	Zhang H K, Zhang Y K, Yang G, Ren Z W, Yu W, Shen D, Lee C S, Zheng Z J, Li G. Sci. China Chem., 2019, 62(7):875.

[71]	Dong Q F, Zhou Y H, Pei J N, Liu Z Y, Li Y W, Yao S Y, Zhang J B, Tian W J. Org. Electron., 2010, 11(7):1327.

[72]	Kang D J, Cho H H, Lee I, Kim K H, Kim H J, Liao K, Kim T S, Kim B J. ACS Appl. Mater. Interfaces, 2015, 7(4):2668.

[73]	Zhong Z M, Ma Y W, Liu H L, Peng F, Ying L, Wang S R, Li X G, Peng J B, Cao Y. ACS Appl. Mater. Interfaces, 2020, 12(18):20750.

[74]	Kee S, Kim N, Park B, Kim B S, Hong S, Lee J, Jeong S, Kim A, Jang S, Lee K. Adv. Mater., 2018, 30(3):1703437.

[75]	Woon K L, Wong W S, Chanlek N, Nakajima H, Tunmee S, Lee V S, Ariffin A, Songsiriritthigul P. RSC Adv., 2020, 10(30):17673.

[76]	Cai W Z, Österberg T, Jafari M J, Musumeci C, Wang C F, Zuo G Z, Yin X L, Luo X H, Johansson J, Kemerink M, Ouyang L Q, Ederth T, Inganäs O. J. Mater. Chem. C, 2020, 8(1):328.

[77]	Chen Y, Zhang N, Li Y F, Bi Y G, Yue Y Y, Feng J, Sun H B. Adv. Opt. Mater., 2018, 6(13):1701348.

[78]	Jung W B, Jang S, Cho S Y, Jeon H J, Jung H T. Adv. Mater., 2020, 32(35):1907101.

[79]	Cui N, Tang Q X, Ren H, Zhao X L, Tong Y H, Liu Y C. J. Mater. Chem. C, 2019, 7(18):5385.

[80]	Zhou L, Yu M J, Chen X L, Nie S H, Lai W Y, Su W M, Cui Z, Huang W. Adv. Funct. Mater., 2018, 28(11):1870072.

[81]	Li D D, Liu X, Chen X, Lai W Y, Huang W. Adv. Mater. Technol., 2019, 4(8):1900196.

[82]	Yuk H, Lu B Y, Lin S, Qu K, Xu J K, Luo J H, Zhao X H. Nat. Commun., 2020, 11:1604.

[83]	Park J, Yoon H, Kim G, Lee B, Lee S, Jeong S, Kim T, Seo J, Chung S, Hong Y. Adv. Funct. Mater., 2019, 29(34):1902412.

[84]	Yoon D, Kang M Y, Kim J, Kang K T. Appl. Sci., 2018, 8(2):203.

[85]	Ding Z C, Xing R B, Fu Q, Ma D G, Han Y C. Org. Electron., 2011, 12(4):703.

[86]	Fanzio P, Cagliani A, Peterffy K G, Sasso L. Microelectron. Eng., 2017, 176:15.

[87]	Wang S J, Dou X Y, Chen L, Fang Y, Wang A Q, Shen H B, Du Z L. Nanoscale, 2018, 10(24):11651.

[88]	Radivo A, Sovernigo E, Caputo M, Zilio S D, Endale T, Pozzato A, Goldoni A, Tormen M. RSC Adv., 2014, 4(64):34014.

[89]	Kang S J, Lee B Y, Lee S H, Lee S D. Sci. Rep., 2019, 9:13066.

[90]	Zhang M, Höfle S, Czolk J, Mertens A, Colsmann A. Nanoscale, 2015, 7(47):20009.

[91]	Kim G, Shin D, Lee J, Park J. Org. Electron., 2019, 66:116.

[92]	Han D, Khan Y, Gopalan K, Pierre A, Arias A C. Adv. Funct. Mater., 2018, 28(37):1802986.

[93]	Zhang S M, Ling H N, Chen Y H, Cui Q Y, Ni J H, Wang X C, Hartel M C, Meng X, Lee K, Lee J, Sun W J, Lin H S, Emaminejad S, Ahadian S, Ashammakhi N, Dokmeci M R, Khademhosseini A. Adv. Funct. Mater., 2020, 30(6):1906016.

[94]	Kim N, Kang H, Lee J H, Kee S, Lee S H, Lee K. Adv. Mater., 2015, 27(14):2317.

[95]	Jiang S, Hou P X, Chen M L, Wang B W, Sun D M, Tang D M, Jin Q, Guo Q X, Zhang D D, Du J H, Tai K P, Tan J, Kauppinen E I, Liu C, Cheng H M. Sci. Adv., 2018, 4(5):9264.

[96]	Yu Z B, Liu Z T, Wang M L, Sun M L, Lei G T, Pei Q B. J. Photon. Energy, 2011, 1(1):11003.

[97]	Jia S, Sun H D, Du J H, Zhang Z K, Zhang D D, Ma L P, Chen J S, Ma D G, Cheng H M, Ren W C. Nanoscale, 2016, 8(20):10714.

[98]	Kwon S J, Han T H, Ko T Y, Li N N, Kim Y, Kim D J, Bae S H, Yang Y, Hong B H, Kim K S, Ryu S, Lee T W. Nat. Commun., 2018, 9:2037.

[99]	Zhang Z K, Du J H, Zhang D D, Sun H D, Yin L C, Ma L P, Chen J S, Ma D G, Cheng H M, Ren W C. Nat. Commun., 2017, 8:14560.

[100]

Chen Y, Zhang N, Li Y F, Bi Y G, Yue Y Y, Feng J, Sun H B. Adv. Opt. Mater., 2018, 6(13):1701348.

[101]

Chen S F, Zhang Q, Shang W J, Liu L H, Yu H T, Zhang S, Deng L L, Wang M, Wang M H, Li X, Mi B X, Huang W. Sci. Rep., 2018, 8:8155.

[102]

Wu T L, Yeh C H, Hsiao W T, Huang P Y, Huang M J, Chiang Y H, Cheng C H, Liu R S, Chiu P W. ACS Appl. Mater. Interfaces, 2017, 9(17):14998.

[103]

Liu L H, Shang W J, Han C, Zhang Q, Yao Y, Ma X Q, Wang M H, Yu H T, Duan Y, Sun J, Chen S F, Huang W. ACS Appl. Mater. Interfaces, 2018, 10(8):7289.

[104]

Zhang Q, Chen S F, Zhang S, Shang W J, Liu L H, Wang M H, Yu H T, Deng L L, Qi G Q, Wang L Y, Han S Y, Hu B, Kang Q, Liu Y J, Yi M D, Ma Y W, Yang W J, Feng J, Liu X G, Sun H B, Huang W. J. Mater. Chem. C, 2018, 6(8):1926.

[105]

Lee B R, Kim J S, Nam Y S, Jeong H J, Jeong S Y, Lee G W, Han J T, Song M H. J. Mater. Chem., 2012, 22(40):21481.

[106]

Kim N, Kee S, Lee S H, Lee B H, Kahng Y H, Jo Y R, Kim B J, Lee K. Adv. Mater., 2014, 26(14):2268.

[107]

Fehse K, Walzer K, Leo K, Lövenich W, Elschner A. Adv. Mater., 2007, 19(3):441.

[108]

Seo Y K, Joo C W, Lee J, Han J W, Cho N S, Lim K T, Yu S, Kang M H, Yun C H, Choi B H, Kim Y H. Org. Electron., 2017, 42:348.

[109]

Hippola C, Kaudal R, Manna E, Xiao T, Peer A, Biswas R, Slafer W D, Trovato T, Shinar J, Shinar R. Adv. Opt. Mater., 2018, 6(4):1701244.

[110]

Cai M, Ye Z, Xiao T, Liu R, Chen Y, Mayer R W, Biswas R, Ho K M, Shinar R, Shinar J. Adv. Mater., 2012, 24(31):4337.

[111]

Ang P Y, Will P A, Lenk S, Fischer A, Reineke S. Sci. Rep., 2019, 9:18601.

[112]

Höfle S, Schienle A, Bruns M, Lemmer U, Colsmann A. Adv. Mater., 2014, 26(17):2750.

[113]

Chiba T, Pu Y J, Kido J. Adv. Mater., 2015, 27(32):4681.

[114]

Liu S H, Yu H W, Zhang Q Y, Qin F S, Zhang X, Zhang L T, Xie W F. J. Mater. Chem. C, 2019, 7(18):5426.

[115]

Zhen H Y, Chen Y X, Zhang C L, Zhou Z G, Li K, Zhao Y, Mei X F, Ling Q D. J. Mater. Chem. C, 2017, 5(26):6400.

[116]

Zimmermann J, Schlisske S, Held M, Tisserant J N, Porcarelli L, Sanchez-Sanchez A, Mecerreyes D, Hernandez-Sosa G. Adv. Mater. Technol., 2019, 4(3):1800641.

[117]

Matyba P, Yamaguchi H, Eda G, Chhowalla M, Edman L, Robinson N D. ACS Nano, 2010, 4(2):637.

[118]

Liang J J, Li L, Niu X F, Yu Z B, Pei Q B. Nat. Photonics, 2013, 7(10):817.

[119]

Shu Z, Kemper F, Beckert E, Eberhardt R, Tünnermann A. Mater. Today: Proc., 2017, 4(4):5039.

[120]

Li H Y, Liu Y F, Su A Y, Wang J T, Duan Y. Sci. Rep., 2019, 9:17998.

[121]

Meng H, Luo J X, Wang W, Shi Z J, Niu Q L, Dai L, Qin G G. Adv. Funct. Mater., 2013, 23(26):3324.

[122]

Gu Z Z, Tian Y, Geng H Z, Rhen D S, Ethiraj A S, Zhang X C, Jing L C, Wang T, Xu Z H, Yuan X T. Appl. Nanosci., 2019, 9(8):1971.

[123]

Gan D L, Huang Z Q, Wang X, Jiang L L, Wang C M, Zhu M Y, Ren F Z, Fang L M, Wang K F, Xie C M, Lu X. Adv. Funct. Mater., 2020, 30(5):1907678.

[124]

Nair N M, Pakkathillam J K, Kumar K, Arunachalam K, Ray D, Swaminathan P. ACS Appl. Electron. Mater., 2020, 2(4):1000.

[125]

Zhang Q, Lu Y, Liu Z W, Yu H T, Duan Y, Liu L H, Chen S F, Huang W. Org. Electron., 2019, 72:30.

Funding

National Key Research and Development Program of China(2017YFB0404501)

National Foundation for Science and Technology Development of China(973 project,2015CB932203)

National Major Fundamental Research Program of China(91833306)

National Natural Science Foundation of China(61705111)

National Natural Science Foundation of China(61704091)

Science Fund for Distinguished Young Scholars of Jiangsu Province of China(BK20160039)

Synergetic Innovation Center for Organic Electronics and Information Displays, the Startup Foundation of Nanjing University of Posts and Telecommunications(NY217010)