In recent years, the need for cheap and fast processing of larger areas of thin organic films has become an increasingly important goal within a number of research fields that have emerged from the idea that solution processing of organic electronics has the potential for just that. In this context, roll-to-roll (R2R) vacuum free processing on flexible substrates has always been the final goal. Most of these technologies are now at the “verge of delivery” where science must be taken out of the protected environment of the laboratory, where it has proven successful for small-scale devices, in order to show that a transfer to large-scale production is actually possible. This might sound tedious, but the fact is that such a transfer is extremely challenging. First, all the preferred processing procedures in the laboratory and those suitable for R2R processing are generally not the same. Spin coating is, for example, among the most popular methods when it comes to small-scale processing of thin polymeric films in the laboratory, because it is easy to use and it is fairly cheap to acquire the necessary equipment, but the process does not comply well with larger areas and too much material is wasted for it to be compatible with high throughput production. Another example is vacuum deposition of electrodes, which because of the need for high vacuum is very time-consuming and a cost driver, which is why alternative techniques are necessary for larger production scale.
Second, there is a huge difference between preparing and aligning a small multilayer structure typically on a solid substrate like glass, and the precise coating and/or printing of large areas with the same accuracy on flexible substrates that moves with speeds on the order of 1–20 m/min (or preferably in the range of 60–300 m/min). This requires a high degree of technical skill, and the combination of scientific knowhow on how to tune the chemical and physical properties of the ink that is to be processed, with the technical knowhow of different processing procedures and the conditions they apply to. This simple challenge may very well become a bottleneck in the further development as such combined skills are not necessarily present in a research group.
This review is aimed at giving an overview of a series of R2R coating and printing techniques that can be used in the processing of organic thin films. We also review the devices that can be prepared using these techniques and give a broad overview of the recent progress within its applications in different research fields. Technical details on machine configurations and setups will not be covered as it is already well described elsewhere in literature.1–3
FILM PREPARATION METHODS
A large palette of different printing and coating techniques can be used for R2R processing on flexible substrates such as PET (polyethylene terephthalate), but depending on the ink and processing surface one particular choice is often critical for success or at least significantly better than the others. An essential difference between the coating and the printing techniques is that printings (with the exception of inkjet printing) are contact processes and enable a patterned deposition of the functional ink. Coating methods can either be contact-based or contact-free and are in general used for large-scale full-layer deposition of thin films. Patterning in coating is somehow limited, and is in reality restricted, to stripes. In the following section, we describe only the most prominent solution-based deposition technologies in the field of thin film electronics. A detailed description of the plethora of existing coating methods is beyond the scope of this review. Technical details to each film preparation method (coating/printing) can be found elsewhere in literature.3–8
Slot-Die Coating and Knife Coating
R2R coating is a one-dimensional continuous appliance of ink in a stripe with a defined width. In knife coating (often also referred to as blade coating), the ink is supplied in front of the knife which is placed very close to the substrate (see Fig. 1). As the substrate moves the fixed knife pushes the ink in front of it allowing only what corresponds to the gap to pass underneath the edge. The wet layer thickness can thus be regulated by adjusting the distance between the knife and the substrate and to some extent by the processing speed. The coating thickness can be estimated as being roughly half of the gap height. The technique is suitable for processing of wide areas without any pattern.
In slot-die coating, the ink is pumped through a coating head placed very close to the substrate, but without touching it. The constant supply of ink forms a standing meniscus between the moving substrate and the coating head. This creates a continuous coat of quite even thickness over a large area. In case the slot-die head is fitted with a meniscus guide it is possible to coat stripes in variable width depending on the shape of the meniscus guide (see Fig. 2). Because of the use of a pump it is possible to regulate the wet layer thickness with very high precision as it is defined by the pumping speed, the width of the meniscus, and the web speed. It is therefore classified as a premetered coating method.
Both knife and slot-die coating can be used for inks with large varieties in viscosity and solvents, and compared to many other coating and printing techniques they are also relatively forgiving with respect to wetting of the substrate as the ink is “poured” into the substrate compared to printing techniques where surface tension and surface energy play a more significant role in the transfer of the ink. Dewetting of the formed film will occur if the surface tension of the ink is larger than the surface energy of the substrate. The wetting of inks with a high surface tension (i.e., water-based inks and dispersions) can be achieved by pretreatment of the substrate with processes such as corona or plasma treatment that raises the surface energy of the substrate (this is in general valid for all coating and printing methods).
Processing speeds for both knife and slot-die coating can be performed at speeds ranging from 0.1 to 200 m/min. Currently, manufacture of functional thin films (e.g., for organic solar cells) has been reported at speeds lower than 5 m/min. This is possibly linked to the need for a thermal treatment of the active layer post film formation. In laboratory-scale or pilot-scale equipment, the oven length is often limited and therefore processing speed is drying length dependent rather than ink coatability/printability dependent.
Opposed to the one-dimensional R2R coating, all the R2R printing techniques are two dimensional (2D) and thus allow for/require a defined pattern. In screen printing, the desired pattern is defined by the open area of an otherwise filled mesh. A squeegee, which moves relative to the mesh, then forces the ink through the open area and onto the substrate. The wet layer thickness is defined by the thickness as well as the open area of the mesh and generally relative thick wet layers can be achieved (10–500 μm). The technique is generally only useful for rather high-viscosity inks with thixotropic (shear-thinning) behavior, as inks with lower viscosities will simply run through the mesh.
There are two types of screen printing—flat bed and rotary screen printing—as illustrated in Figure 3. Flat bed screen printing is a stepwise process, where the screen is lowered on top of the substrate, the squeegee is then swiped over the screen resulting in the transfer of the ink, and finally the screen is lifted and the substrate is manually changed or moved forward after which the process can be repeated. Stepwise processes are generally not desirable in R2R processing as they are more time-consuming, but flat bed screen printing has been successfully adapted to R2R processing. The screens used are easily patterned in a variety of sizes and are fairly low in cost. This makes flat bed screen printing a powerful tool for small laboratory systems, either in a tabletop configuration or in a small- to medium-scale R2R configuration.
Fully continuous processing is best achieved through rotary screen printing, which uses the same principle as flat bed printing but in this case the web of the screen is folded into a tube and the squeegee and the ink are placed inside the tube. As the screen rotates with the same speed as the substrate the ink is continuously pushed through the open area of the screen by the stationary squeegee, making a full print upon every rotation. Much higher processing speeds can be achieved by use of rotary screen printing (>100 m/min) compared to flat bed (0–35 m/min), but the screens are quite expensive and they are much more difficult to clean because of the restricted access to the inside of the screen. The process furthermore requires a longer adjustment run-in (adjusting the print with previously processed layers) compared to flat bed. This gives a higher initial waste, but once the process is running the procedure is very reliable.
Gravure printing is a widely used printing technique commonly used in the printing of high-volume catalogs and magazines. The technique relies on surface tension transfer of ink from small engraved cavities in the gravure cylinder to the substrate. Good contact between the substrate and the gravure cylinder is ensured by use of a softer impression cylinder, and the final imprint is defined by the patterning of the cavities in the gravure cylinder. The cavities are continuously refilled by passing an ink bath and excess ink is removed by use of a doctor blade. A more advanced ink filling method is achieved by using a chambered doctor blade system, which can be advantageous for inks containing highly volatile solvents. The gravure cell volume and the pick-out ratio mainly define the transferred wet layer thickness on the substrate. Gravure printing applies well for printing of low-viscosity inks and is well suited for very high processing speeds (up to 15 m/s). Gravure offset printing, gravure coating, and reverse gravure coating are further methods based on an engraved gravure cylinder, whereas it is only patterned for the printing process.
In flexographic printing, the transfer of ink happens through direct contact of a soft printing plate cylinder, typically made of rubber or a photopolymer, onto which the desired motif stands out as a relief (like on a traditional stamp). The inking of the printing plate cylinder is provided via a ceramic anilox roller with engraved microcavities embedded into the exterior surface. The anilox cylinder is continuously supplied with ink by contact with a fountain roller that is partly immersed in an ink bath. Similar to gravure printing excess ink on the anilox is removed by a doctor blade ensuring good control of the wet layer thickness, which is defined by the volume of the cavities in the anilox cylinder (anilox volume) and the transfer rates from the printing plate cylinder to the substrate. The inking principle based on a chambered doctor blade system is also suitable in this case (similar to the gravure process).
In contrast to the traditional printing techniques where the ink is transferred from or through a permanent printing form to the substrate, inkjet printing is classified as a full digital nonimpact printing method with images generated on demand. Two-dimensional patterning is obtained by a pixilated pattern with a defined resolution. Each pixel will either receive a droplet of ink or not, as shown in Figure 4. The most abundantly used method, referred to as drop-on-demand (DOD), uses a piezoelectric material to shoot droplets from a nozzle which is situated ∼1 mm from the substrate. The DOD system requires many nozzles which in the early days caused limitation in web speeds and resolution, but today both high web speeds and high resolution can be achieved with the available commercial systems. The pattern shown in Figure 4 was printed on a R2R inkjet printer capable of a resolution of 600 DPI with web speeds ≤75 m/min. The major advantage of the inkjet printing technique compared to other 2D techniques is the possibility to change/adjust the printed pattern easily on a computer without the need to manufacture a physical printing form (i.e., it uses a digital master). Inkjet printing is a relatively new technology on an industrial scale and presents some limitation with respect to processing speeds and ink formulation. Especially the latter point has put constraints on the use of the technology for organic film formation.
Although spray coating cannot be counted as a printing method, it shows similarities to inkjet printing. In spray coating, a continuous spray of ink is generated in a nozzle instead of being digitally timed drops as in the case of DOD inkjet. In pneumatic-based systems, the ink gets atomized at the spray nozzle where pressurized air or gas (e.g., nitrogen) breaks up the liquid bulk into droplets.9 The liquid properties (surface tension, density, and viscosity), the gas flow properties, and the nozzle design influence the entire atomization process. The kinetic energy of the droplets helps them to spread upon impact on the substrate. Furthermore, the quality of the coated layer is based on several process parameters such as the distance of the spray nozzle to substrate, coating speed, and the number of sprayed layers. Other forms of spray generation are ultrasonication with a directed carrier gas10 or electrospraying.11 Pattern formation with spray coating is only possible with the help of shadow masks. The spray method has a high R2R compatibility, but the disadvantage of ink mist has to be considered, as it can cause contamination of the processing equipment. The ink loss when using a shadow mask and the low edge resolution is also a fundamental disadvantage. As a coating method for fully covering layers, it can be used in conjunction with laser ablation or other patterning methods postfilm formation.
CURRENT STATUS WITHIN DIFFERENT RESEARCH FIELDS
R2R coating and printing is an emerging processing method within the field of organic/polymer thin film device fabrication. “True” R2R equipment is very costly to acquire, and as a consequence reports of “true” R2R processing are limited to very few groups. To give a broader picture of how the above described processing methods have been used so far, we have chosen to include R2R-“compatible” processing techniques (sheet-to-sheet processing and the use of nonflexible substrates) in our overview of the current status of the field. It is important to stress though that there are huge differences between a process being claimed as R2R-compatible and true R2R processing, and the results obtained using the compatible processing will not be directly transferable to R2R! Table 1 gives a summary over various processing of layers in polymer-based devices including electrodes, blocking layers, and active layers and other layers in organic photovoltaic (OPV), organic thin film transistor (OTFT), polymer light-emitting diode (PLED), and electrochromic (EC) devices. The table also illustrates the fact that different coating or printing methods might be suited for processing of a specific layer, while proving less useful for the processing of others. Fixing the eye on a single technique for the processing of a whole device thus might prove less efficient than using several different techniques optimized for each specific layer. Figure 5 shows typical device architectures of OPVs, OTFTs, PLEDs, and EC devices.
Table 1. Overview of Reported Polymer Thin Film Devices Using R2R or R2R-Compatible Techniques
Large Area Organic Solar Cells
Organic solar cells are as indicated in Figure 5 multilayer structures of typically a transparent electrode, a hole or electron blocking layer, the active layer (comprising a bulk heterojunction of a polymeric absorber, which acts as a donor material and an acceptor material typically a fullerene), a hole or electron blocking layer (the opposite of the one placed on top of the transparent electrode), and finally a back electrode.
A large palette of R2R techniques have been used in the preparation of organic solar cells, with slot-die coating at present being the most abundantly used for processing of hole/electron blocking layers and the active layer. On the other hand, screen printing is by far the most widely used for processing of the back electrode, and is a good example of how different processing procedures can be more suitable than others for the processing of a specific layer.
Krebs et al. introduced a new fabrication method for all R2R-processed OPVs on indium tin oxide (ITO)-coated PET substrates in 2009,18 with an architecture of PET|ITO|ZnO|active layer|PEDOT:PSS|silver, which has since been referred to as “ProcessOne.” The first three layers (ZnO, active layer, and PEDOT:PSS) were processed using slot-die coating, and finally a silver back electrode was slot-die coated or flat bed screen printed to finalize the stack before it was laminated. Since then, numerous productions of solar cells and modules based on this method have been published,20–26 and solar modules produced according to this have been used in interlaboratory studies,86, 87 round robins,88, 89 and in demonstrators of various kinds that have mostly involved sound, light, or in one case a laser (see Fig. 6).19, 23, 28, 90 The ProcessOne solar modules based on P3HT:PCBM are presently at a stage with a lifetime of 1 year (T80) when exposed to outdoor weather conditions.
Various examples of the use of R2R-processed low-bandgap (LBG) polymers have been reported21, 22, 24, 25 with efficiencies approaching what can be achieved with P3HT:PCBM. These are, however, still generally inferior in performance, which can be ascribed to the fact that it takes a large effort (and a lot of material) to fully explore a new polymer. This goes to show that direct transfer of small-scale technology cannot always be expected, as the use of several different LBG polymers in small-scale devices has proven to be more efficient than P3HT:PCBM, when prepared in a glove box and with evaporation of electrodes. In a more special case, Helgesen et al. used a R2R photonic flash lamp to remove the side chains of the used polymer by selective heating of the active layer to high temperatures without heat damaging the flexible plastic substrate. The reason for removing solubilizing side chains in this extra step was to improve device stability as the solubilizing side chains are known to induce instability to the polymer.24
Another of the more exotic examples of R2R-processed solar cells was presented by Hübler et al. who have prepared fully R2R-processed solar cells on paper using a combination of gravure and flexo printing.17 The processing starts with gravure printing of a glue onto a paper substrate. The glue patterned paper is then brought into contact with a zinc-coated transfer foil, transferring the zinc only onto the patterned glue, and the surface of the zinc is then oxidized creating a thin hole blocking layer. The active layer (P3HT:PCBM) is now coated on top of the ZnO by use of gravure printing, and the device is finalized by flexo printing a thin layer of PEDOT:PSS as a transparent back electrode. Small devices (0.09 cm2) showed maximum efficiencies of 1.31% at 600 W/m2. Figure 7 shows a picture of the final device.
One of the interesting things about the report from Hübler et al., besides the fact that they use paper as substrate, is that the architecture is ITO-free. ITO has for decades been the preferred transparent electrode material in many types of electronic devices, because of its high transparency combined with a low sheet resistance, but the scarcity of indium makes ITO quite expensive and the cost of indium accounts for the majority of the device cost. Much effort has been put into finding alternatives, and Galagan et al. showed that a silver collecting grid, screen printed on a PEN substrate and subsequently spin-coating of a high-conducting PEDOT:PSS solution on top, is a way of substituting ITO.32 Good efficiencies (1.93%, 4 cm2) were obtained using P3HT:PCBM for the active layer (spin-coated in glove box) and evaporative deposition of LiF/Al as the electrode, but the use of spin-coating, inert atmosphere, and low-pressure deposition makes the procedure difficult to transfer to a R2R setting.
Other examples where full R2R processing is actually achieved have recently been presented though.12–14 Larsen-Olsen et al. showed that the use of high-conductive PEDOT:PSS with or without a preprocessed silver collecting grid can be used as a substitute for ITO in inverted structure devices and modules with good results.12, 14 By appliance of a short pulse at high negative bias to the solar cell, it is possible to electrochemically “switch” (reduce) a thin layer of the top PEDOT causing a permanent conductivity change rendering it a rectifying junction. Further studies on the use of different types of front grids in the same process were carried out by Yu et al. showing that thermally embedded silver grids and flexo printed silver grids result in solar cells with similar efficiencies (slightly better for the embedded grid), whereas solar modules with inkjet printed silver front grid performed poorer, which was ascribed to the lower conductivity of the grid.13
Besides the previously mentioned example of use of paper as substrate by Hübler et al., the use of gravure printing for the preparation of solar cells has only been reported in very few cases.34–37 Kopola et al. reported the use of a desktop gravure printability tester (on PET, not R2R process) to process the hole transport layer (PEDOT:PSS) as well as the active layer (P3HT:PCBM) of single cells36 and of small modules.35 In both reported cases, the back electrode consisted of evaporated calcium and silver resulting in efficiencies of 2.8% for single cells (19 mm2) and 1.9% for small modules (five cells in series, 9.6 cm2) after optimization of the PEDOT:PSS ink with surfactants, wetting agents, and solvent mixtures.
Figure 8 shows an example of the necessity to optimize ink and processing conditions to obtain a smooth and homogeneous active layer and an example of a final flexible device. Voigt et al. recently reported the use of sheet-to-sheet gravure printing of inverted structured cells on PET after performing a systematic study of the wetting behavior of each layer.37 In this case, three of the layers (TiOx, P3HT:PCBM, and PEDOT:PSS) were processed by gravure, and the cells were finalized by evaporation of a back gold electrode (4.5 mm2, 0.6% PCE).
Also, flexo printing has only had limited application in solar cells and has so far not yet been used for processing of the active layer. The potential of flexographic printing as an extremely fast processing method has so far proven best for the preparation of front grid silver electrodes that were prepared at high speeds (25 m/min).13
Among the more specialized coating procedures used for R2R OPV fabrications can be mentioned double-slot-die coating, introduced in organic solar cells by Larsen-Olsen et al. in order to approach a further increase in the production throughput by simultaneous deposition of several layers of the solar cell stack.91 The double slot-die method illustrated in Figure 9 was used to coat an aqueous suspension of P3HT:PCBM nanoparticles and aqueous PEDOT:PSS at the same time with a processing speed of 1 m/min. Although the solar cells showed a poor performance (PCE of 0.03%) because of the complex bilayer formation process, it demonstrates the potential as a future processing method to lower the energy payback time of organic solar cells.
Another specialized processing method especially for research and development purposes is the differentially pumped slot-die coating.92 Hereby two components of the functional ink are mixed together with the ability to generate changing material ratios over the length of the running web. The fast screening method was used to determine the optimum donor–acceptor ratio and film thickness for organic solar cells. The specially designed slot-die head with very low dead volume requires very little amount of material, which makes it an ideal tool to screen new materials in a wide parameter space compared to spin coating. Furthermore, it directly shows the R2R processability. The process has later been used on several occasions to optimize donor/acceptor blends.21, 22, 25
Electrochromic devices are based on materials exibiting electrochromic behavior (materials that present two discrete optical appearances when in a reduced or oxidized electrochemical state). Testing of the electrochromic properties is usually performed by coating the polymer on a substrate with a transparent electrode (typically ITO), followed by immersing into an electrolyte solution together with an expendable counter electrode. The electrochromic properties, such as color change, change in transmittance (ΔT), and switching times, can then be examined through electrochemical oxidation and reduction of the polymer. When wanting to build a thin device the use of a simple counter electrode is generally not preferable because of the slow deterioration of this, but as a substitute one can make use of a complementary redox compound (a material that is reduced when the electrochrome is oxidized and vice versa) deposited on a second substrate also with a transparent electrode. The final device then consists of the two coated substrates with the electrolyte sandwiched in between as shown in Figure 5.
The processing of polymeric ECs has been reported by use of spray coating,74, 76–80 inkjet printing,81, 82 screen printing,83, 84 and slot-die coating.74 Jensen et al. recently presented results showing that slot-die coating (or spray coating) of the electrochromic layers on flexible substrates can be achieved using a single roll coater with a roll having a diameter of 300 mm.74 By use of the EC commonly known as ECP-Magenta and a minimally color changing polymer (MCCP) as complementary redox compound, which were both slot-die coated on separate PET/ITO substrates, it was possible to make electrochromic devices >10 cm2, which could switch between magenta and a colorless state. Pictures of the coatings are shown in Figure 10. By connecting an electrochromic device (40 mm × 40 mm) directly with polymer solar cell modules or with batteries charged by polymer solar cells, they furthermore fabricated a demonstrator showing how different polymer thin film technologies can be combined to a final product. The small-scale approach was later upscaled to full R2R processing of ITO-free 18 cm × 18 cm electrochromic windows printed directly on barrier foil using flexographic printing of metal grids and slot-die coating of the electrochromic polymers (ECP-magenta and MCCP).74
With respect to spray coating especially the group of Reynolds has contributed to finding a series of polymers of different colors and optimizing the conditions for the use of this technique.76–80 As can be seen from Table 1, a multitude of different colored polymer are available, and recently Dyer et al. could announce that the color palette for spray-processable polymer electrochromics is complete.93
The use of inkjet printing of polymer electrochromes is highly useful when wanting to create patterned EC devices. This is well illustrated in a report by Shim et al., who showed that printing composite dispersions of polyaniline- or PEDOT-covered silica nanoparticles onto PET/ITO enabled preparation of electrochromic devices with good resolution (see Fig. 10).81
Screen printed flexible electrochromic devices were reported as early as in 1999 by Brotherston et al., who used a combination of PEDOT and V2O5 screen printed, respectively, onto separate ITO-coated Mylar substrates and finalized with an electrolyte sandwiched between the two substrates.84 The technique has not become widely used, as the only other example of the use of “screen printing” for electrochromic device preparation is the use of a precut barrier film and a screen printing squeegee for the processing of the electrolyte.83 As no actual screen was used in this latter example, the term screen printing should be taken with modification.
Thin Film Transistors
Although organic solar cells have so far been the most prominent technology to use R2R processing of functional polymer materials, printing is also one of the experimental techniques in a variety of fabrication methods for organic thin film transistors.94 In recent years, efforts have been made to R2R manufacture devices including organic thin film transistors to enable integrated circuitry such as inverters95 and ring oscillators.96 Still, most of the devices containing polymer materials are fabricated by sheet-to-sheet methods using R2R-compatible processes such as inkjet,63, 97 gravure,64, 65 screen printing,66 and spray coating.98 Kang et al., for example, utilize an optimized microgravure process to print silver patterns and poly(4-vinyl phenol) (PVP) dielectric layers.99 A transistor with record transition frequencies of >300 kHz was achieved with a spin-coated organic semiconductor poly(2,5-bis(3-alkythiophen-2-yl)thieno[3,2-b] thiophene) (pBTTT).
The transition from batch to R2R processing has partly been carried by Tobjörk et al. by using R2R reverse gravure coating for the deposition of a polymer semiconductor (P3HT) layer and a polymer dielectric PVP for the fabrication of low-voltage organic transistors with an on/off ratio of 100 and threshold voltage of 0.5 V.61 Silver source/drain electrodes and PEDOT:PSS polymer gate electrodes were sheet-to-sheet inkjet printed. Figure 11 (top) illustrates the processing methods used as well as the final device. The first integrated circuit fabricated completely by means of mass-printing technologies was reported by Huebler et al.62 The seven-stage ring oscillator contains organic field effect transistors in a top gate architecture with PEDOT:PSS source/electrodes prepared by offset lithographic printing. Gravure printing was used for the poly(9,9-dioctyl-fluorene-co-bithiophene) (F8T2) semiconductor layer and the first butadiene-styrene-copolymer low-k dielectric layer. The device was finalized with flexographic printing of a high-k BaTiO3 dielectric layer and silver gate electrodes. Although the frequency reached was only 3.9 Hz, it demonstrates the processability with mass-printing technologies with speeds in the order of 1 m/s. A picture of the seven-stage ring oscillator is show in Figure 11 (bottom).
A combined R2R flexo and gravure process was used to realize PEDOT:PSS source/drain electrodes with channel length down to 10 μm.100 A negative image of the electrodes was flexo printed from an amorphous perfluorinated poly(alkenyl vinyl ether). A secondary full layer gravure print of PEDOT:PSS leads to a self-formation of the electrodes. Full devices such as OFETs, inverters, and ring oscillators were fabricated with gravure printing of F8T2 semiconductor and dielectric material (butylene copolymer and PMMA). Flexo printing was utilized for the copper electrodes.
Full R2R gravure processing was used for the manufacturing of all-polymer field effect transistors with a yield of ∼75% out of a random selection of 50,000 produced transistors.60 The report highlights a special electrode layout to avoid longitudinal registration problems. The polymer materials used were PEDOT:PSS (source/drain, gate), PMMA and butylene copolymer (dielectric), and an amorphous poly(triphenylamine) (PTPA2) as semiconductor.
Polymer Light-Emitting Diodes
PLED is a class of organic light-emitting diodes (OLED) where the light-emitting layer is based on polymers opposed to small molecules, which are typically deposited by evaporation processes.101–104 OLEDs in general are widely explored and are already in use for display and lighting applications. The layer structure of PLEDs is almost identical to organic solar cells with a light-emitting layer instead of a light-absorbing layer as seen in Figure 5. Common conjugated polymer materials used as emitting layer are polyphenylene vinylenes such as poly[2-methoxy-5-(2′-ethylhexyloxy)-1,4-phenylenevinylene] (MEH-PPV).101 Another approach is the dispersion of small-molecule emitters in a nonconjugated polymer matrix, such as poly(methyl methacrylate) (PMMA) or polyvinylcarbazole (PVK).102 One advantage of using polymers is their solubility that enables the solution-based manufacturing similar to organic solar cells with a variety of potential printing and coating processes. Although the processing advantage is present, reports on R2R manufacturing are rare.
Various studies with R2R-compatible methods have been made and show the applicability of slot-die coating,67 screen printing,73, 105, 106 blade coating,68, 72, 107 gravure printing,69, 71, 108, 109 and inkjet printing,70, 110, 111 which has its advantages for pixel-based display applications. The following reports are highlighted for their closest potential in R2R up scaling. A sample device is shown in Figure 12.
Youn et al. fabricated a PLED based on the yellow-emitting phenyl-substituted poly(para-phenylene vinylene) (Super Yellow).68 All layers including PEDOT:PSS, ZnO, and an ionic solution containing tetra-n-butylammonium tetrafluroborate were blade-slit coated with a high layer uniformity on ITO-coated glass substrates. The luminous efficiency of the in-air processed samples reached 5.26 cd/A.
On another occasion standard multilayer blade coating has been used to coat several small-molecule materials and PVK host solutions to build multicolored light-emitting devices.72 The conducting polymer PEDOT:PSS was blade coated as well. The applied layer thicknesses were below 100 nm with uniformities within 10%. The report covers challenges in uniform layer formation, drying and interaction of solvents in the multilayer approach to prevent dissolution. Large-area devices with 6 cm2 active area and efficiencies up to 25 cd/A for green phosphorescent OLED materials were fabricated.
Gravure printing was used by Kopola et al. to prove the feasibility of large-scale fabrication of PLEDs for lighting applications.69 PEDOT:PSS and a blue-emitting polymer dissolved in o-xylene was gravure printed on ITO-coated glass with an active area of up to 30 cm2. Brightness levels of up to 1000 cd/m2 at 5.4 V were achieved. Ink modification and printing form optimization played an important role to achieve homogenous layers for a uniform light generation.
Finally, a fully slot-die-coated light-emitting device based on the light-emitting electrochemical cell (LEC) technology has been demonstrated by Sandström et al.67 For the first time, devices were fabricated fully under ambient conditions using roll coating on PET foil. This included air processing of the back electrode. The authors used the emissive conjugated polymer “Super Yellow” and an electrolyte consisting of potassium triflate (CF3SO3K) in poly(ethylene oxide) (PEO). Active areas of around 3 cm2 have been achieved with brightness levels of up to 150 cd/m2 at 10 V. The highest current efficiency was recorded with 0.6 cd/A at 50 cd/m2. The processing of the devices was shown to be very reliable because of a thick active layer and air-stable materials. Figure 13 shows pictures of the slot-die coating and the final device.
The achieved results for gravure, blade, and slot-die coating will enable an upcoming transition to full R2R processing of cost-friendly and vacuum-free manufacturing of large area light sources on flexible substrates. Inkjet printing methods will most likely be seen in display manufacturing.
Steenberg et al. have latetly reported the preparation of 40-mm-thick poly[2,2′(m-phenylene)-5,5′bibenzimidazole] (PBI) films for fuel cells using both knife coating and slot-die coating.112 The membranes proved to have identical properties compared to traditionally cast membranes resulting in an increase in manufacturing speed by a factor of 100.
A last example that shows a quite exotic application of thin film processed organic polymers can be found in a recent publication by Hübler et al. on all-printed polymer loudspeakers.85 By utilizing flexographic and screen printing methods, they successfully applied PEDOT:PSS electrodes and the piezo electric polymer poly(vinylidene fluoride-trifluoroethylene) (P(VDF-TrFE)) on standard coated paper substrate. The loudspeakers with effective areas of 16 and 128 cm2 lead to sound pressure levels of 63.8 db at 5000 Hz.
SUMMARY AND OUTLOOK
Processing of organic thin films using R2R techniques clearly holds a promising future and for several technologies the realization of fast and low-cost manufacture lies within reach. There are issues though that still need to be overcome. Replacement of ITO as transparent electrode has commenced, but full integration of these new initiatives needs to be performed as they are still in their infant stage. Another key optimization parameter that needs improvement is the registration during R2R processing, which is used to align the different layers. Ultra-precise multilayer processing plays an important role for many types of devices such as organic TFTs, and better control of registration will be needed—especially for processing at very high speeds.
On the long term, a key feature for high throughput production will be that it does not harm the environment. A true large-scale production can only be realized by processing from water, and more research is needed on how this can be achieved. The use of different types of ink formulations (such as dissolved solutions, emulsions, and dispersed nanoparticles) as replacements to the typical use of orthogonal solvents for the processing of multilayer devices could be one suggestion on how to do this, but other solutions could prove more fruitful. One thing is clear though—if processing of large areas of organic thin films is to be realized through large-scale production, the typical toxic and chlorinated solvents that are generally used for processing of conjugated polymers must be replaced with something less harmful. In addition to solvent substitutions, processing procedures need to be more effective and new materials that can sustain processing in the ambient need to be developed. In-line processing (where several layers are processed right after each other during the same run) and even lower drying temperatures are ways of improving effectiveness. For the materials it is more difficult to define a specific strategy, but ideally these should be developed in tandem with the solvent substitutions, thus approaching the two issues simultaneously.
This work has been supported partially by the EUIndian framework of the “Largecells” project that received funding from the European Commission's Seventh Framework Programme (FP7/2007–2013; grant no. 261936); partial financial support was also received from the European Commission as part of the Framework 7 ICT 2009 collaborative project ROTROT (grant no. 288565).
Roar R. Søndergaard received his PhD from Risø DTU—National Laboratory for Sustainable Energy and has a background in organic synthesis. Through synthesis and device preparation he has previously worked within areas of molecular wires, water processed organic solar cells, and improved stability of polymers for roll-to-roll (R2R)-processed organic solar cells. He is currently working as a researcher at DTU Energy Conversion where he is focusing on large-scale manufacture and optimization of R2R-processed organic solar cells through different processing methods.
Markus Hösel received his Diploma degree (Dipl.-Ing.) in Microtechnology/Mechatronics from the Chemnitz University of Technology (CUT) in 2007. After working as a construction engineer at a printing machine manufacturer and as a research associate at CUT, he is currently pursuing his PhD at DTU Energy Conversion under the mentorship of Prof. Frederik C. Krebs. His research interest is focused on roll-to-roll large-scale fabrication methods for organic photovoltaics.
Frederik C. Krebs received his PhD from the Technical University of Denmark (2000) and has since then worked in the field of functional organic materials for application as solar cells, fuel cells, electrochromics, light-emitting devices, and photocatalysis. A focus area has been polymer solar cells manufactured by roll-to-roll methods for solar energy conversion. He is currently a professor at DTU Energy Conversion within areas of research that include new materials with low bandgap and novel processing capability, large area processing and manufacture of polymer solar cells, all aspects of roll-to-roll printing-coating-processing and testing, life cycle analysis, stability and lifetime testing, degradation mechanism studies, outside testing and demonstration.