Current Opinion in Solid State and Materials Science 14 (2010) 123–130 Contents lists available at ScienceDirect Current Opinion in Solid State and Materials Science journal homepage: www.elsevier.com/locate/cossms Power from plastic Adam J. Moulé Chemical Engineering and Materials Science Department, University of California Davis, 1 Shields Ave., Davis, CA 95616, United States a r t i c l e i n f o Article history: Received 5 February 2010 Accepted 16 June 2010 Keywords: Polymer Fullerene Photovoltaic Solar cell a b s t r a c t Solar power is the most abundant renewable resource on our planet. In spite of this abundance, only 0.04% of the basic power used by humans comes directly from solar sources because harvesting solar energy using a photovoltaic (PV) panel costs more than burning fossil fuels. Solution-processable organic materials have recently been intensively studied for PV applications, not because they show a possibility for harvesting the sun’s power more efficiently, but because power generation from organic photovoltaic (OPV) materials will cost considerably less than other PV technologies. The cost/Watt savings comes from the possibility of using flexible substrates, using printable organic inks for the active layers, light-weight transport, low temperature and ambient pressure fabrication, and reduced materials costs. The ability to use solution processes for deposition is particularly exciting because products such as automobiles, freight containers, and building materials could be painted with photovoltaic coatings. This article gives an overview of the current state-of-the-art for OPV technology and discusses scientific issues that need to be addressed to facilitate scale-up of OPV. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Use of fossil fuels is raising the levels of greenhouse gas in the Earth’s atmosphere, which in turn threatens to cause climate change and unknown economic hardship. Conservative calculations show that in order to avoid the largest risks for climate change, humankind must drastically reduce the amount of primary energy that we use on a daily basis and simultaneously shift to energy sources that result in reduced carbon emission intensity [1]. It is well established that solar energy is by far the most abundant source of renewable energy. However, solar energy is difficult to harvest because it is a diffuse energy source that requires collection over large land areas and because the sun only shines during the day. Photovoltaics are a mature technology. The first silicon photovoltaic device was fabricated by Bell Labs in 1954. Since that time, photovoltaics have been greatly optimized and a power conversion efficiency (PCE) approaching the Schockley–Queisser limit for a single p/n junction of over 25% has been reported [2]. Photovoltaics only contribute 0.04% of the world’s total energy usage [3]. The reason is that, although solar power is the most abundant energy source, conversion of solar energy into electricity or solar fuels is far more expensive than burning fossil fuels [1]. Government incentives to install photovoltaic capacity has resulted in growth for the solar industry of over 30% per year for the last decade, but this growth curve is still too low to significantly reduce global greenhouse gas production, due to rising en- E-mail address: amoule@ucdavis.edu 1359-0286/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.cossms.2010.06.003 ergy demand. The DOE report on Basic Energy Needs for Solar Energy Utilization recommends development of a new class of photovoltaic collectors that are able to produce more Watts per dollar [3]. Fig. 1 shows a chart with power conversion efficiency vs cost in US$ per m2 for a PV unit. On the chart are marked three zones that represent technological and economic challenges for PV production. All of the PV devices that are currently commercially available fall into Zone I. Solution-processable organic photovoltaics (OPV) are currently being intensively investigated because this technology represents the best chance for developing a PV product in Zone II. The major advantages of using conjugated organic materials for PV units rather than inorganics are the low material and substrate costs and the ease of printing and fabrication. However, OPV devices are currently far less efficient and have shorter lifetimes than inorganic solar cells. The combination of low material and fabrication costs for OPV promises to reduce the unit cost by 10 [4]. This cost savings is based on the idea that mass production will decrease material costs and that fabrication can be performed in a reel-to-reel format onto flexible substrates. The resulting economic balance will yield a reduction in the costper-Watt for power produced from OPV devices when the device power conversion efficiency nears 10%. Zone III is shown more as an inspirational/political goal rather than an engineering challenge with a real chance of near term success. There are currently no viable technology paths that promise to both reduce module cost and also increase efficiency beyond the Schockley–Queisser limit. Far more basic and applied research is needed to develop OPV to become a viable large-scale electricity source. This article will discuss the state-of-the-art for organic photovoltaics and address 124 A.J. Moulé / Current Opinion in Solid State and Materials Science 14 (2010) 123–130 100 US$0.50/W US$0.10/W US$0.20/W Efficiency,% 80 Ultimate Thermodynamic limit at 1 sun 60 III US$1.00/W 40 SchockleyQueisser limit 20 0 0 100 US$3.50/W I II 200 300 400 500 Cost, US$/m 2 Fig. 1. PV source costs ($/Wp) as function of module efficiency and areal cost. For PV to provide the full level of carbon-free energy required for electricity and fuel–solar power cost needs to be 2 cents/kW h ($0.40/Wp [6]. Source: Green (2004). scientific and technological hurdles that must be overcome for scale-up to commercialization. A discussion of cost is found in Refs. [5,4]. 2. State-of-the-art The first OPV device was fabricated by Tang et al. at Kodak and had a power conversion efficiency (PCE) of 1.1%. This first device mimicked the structure of a traditional p–n junction. It was a bilayer stack of copper phthalocyanine as the electron donor and a perylene tetracarboxylic derivative for the electron acceptor [7]. This initial design was only able to deliver a low PCE because the light-induced excited states (excitons) are tightly bound in organic molecules and can only separate at donor/acceptor interfaces. Typical exciton diffusion lengths are 5–25 nm [8,9]. This small exciton diffusion length places a fundamental limit on the thickness that a material layer can have since light absorbed further from the interface will not result in charge separation and photocurrent. In the early 90s a new design for OPV active layers was developed called the bulk-heterojunction (BHJ). A BHJ consists of a mixture of donor and acceptor species in the same layer [10,11]. The layer is formed by solution casting the donor–acceptor mixture from a common solvent. As the solvent rapidly evaporates, phase separation occurs between the two components. The degree of phase separation is determined by the solubility of the components in the solvent, the peed with which the film dries, and the mutual solubility of components in each other [12,13]. The new BHJ design has the major advantage that the distance between the donor and acceptor in the layer is reduced and the probability of charge separation approaches unity. However, the mixed film has reduced order, leading to reduced charge mobility, greater trap density, and island domains that do not have a charge transport pathway to either electrode. Until 2007, nearly all increases in OPV device PCE ultimately came about as a result improved BHJ layer morphology. For example, Shaheen et al. found that using a solvent with improved mutual solubility for the two components could reduce the domain size of OC1C10–PPV/[6,6]phenyl-C61–butyric acid methyl ester (PCBM) mixtures [14]. Switching from the solvent toluene to chlorobenzene improved the PCE from 0.9% to 2.5%. Next it was discovered that the application of thermal treatment could be used to improve the hole mobility, increase crystallinity and increase efficiency in poly-3hexylthiophene (P3HT)/PCBM mixtures [15,16]. Simultaneously, it was found that the use of high-boiling-point solvents and long sol- vent-soaking times could optimize the morphology through self assembly [17,18]. Finally, it was found that the morphology could be optimized by using low-concentration additives [19] that could improve morphology either by selectively dissolving one the components [20,21] or by causing rapid crystallization of the polymer during drying [22,23]. All of the thermal, solvent, and additive techniques increase the PCE of a P3HT/PCBM device from 1% to >4%. An important research theme for the last 15 years has been the determination the mechanism for photocurrent production in a BHJ layer. Fig. 2a shows a diagram of a BHJ device based upon the commonly used mixture P3HT/PCBM. The mechanism for photocurrent production is depicted in Fig. 2b. Photons with energy above the optical band gap (Eg) are absorbed in the active layer by both the donor and acceptor materials, but since the donor polymer has a much larger absorption coefficient, the polymer typically absorbs the majority of the light. The absorbed photons form excitons, which for the purpose of this discussion are charge-separated excited states where both charges exist on the same species or polymer domain. In a BHJ, the exciton has a high probability of diffusing to a donor–acceptor interface and separating into a geminate charge pair. This geminate or bound pair, also known as a charge-transfer exciton, consists of a Coulombically bound hole on the donor and electron on the acceptor. The bound charge pair can be separated into free charges at room temperature when a sufficient electric field is applied. This built-in electric field is generated by sandwiching the BHJ layer between electrodes that have differing work functions. Finally, the free charges can hop from site-to-site under the influence of the built-in field until they reach their respective electrodes. Bound charge pairs can reform from free charges. Both bound pairs and free charges can recombine. Fig. 2b shows a generalized relaxation pathway. The kinetics of each process is strongly dependent on the morphology, donor/ acceptor mixing ratio, electric field, and specific donor and (a) LUMO offset loss Polym er LU eEg MO Fuller ene L h+ UMO Polym High Φ Electrode er HO Fuller MO Low Φ Electrode ene H OMO (b) D* δ+δ− Exciton A D+ A - + − E,T Geminate Pair D+ A- + ........ − Photocurrent Free Charges D A Recombination Fig. 2. (a) Energy level diagram for an organic bulk heterojunction PV device under short circuit conditions. (b) Cartoon of the charge-generation process: exciton formation, dissociation into geminate pairs, and separation of the geminate pairs into free charges under the influence of the electric field and temperature to produce photocurrent. Geminate pairs and free charges can recombine to form neutral species [25]. Source: Moule (2008). 125 A.J. Moulé / Current Opinion in Solid State and Materials Science 14 (2010) 123–130 acceptor species. Blom et al. has written a detailed review of these mechanisms [24]. Since 2007, a number of new record efficiencies have been reported that can be directly attributed to improved synthetic design of either the donor or acceptor. Referring to Fig. 2a, three successful synthetic strategies can be identified. First, the short circuit current density (Jsc) can be increased by lowering the Eg and maintaining the polymer HOMO–fullerene LUMO spacing. The second possibility is to increase the open circuit voltage Voc by maintaining the Eg but shifting the both the HOMO and LUMO of the polymer down in energy [26]. This shift will have the effect of lowering the energy loss labeled LUMO offset loss in the figure. The third possibility is to increase the Voc while maintaining the Eg by raising the LUMO level of the fullerene. Two articles have predicted that these combined adjustments can yield power efficiencies of over 10% [27,28]. A series of new co-polymer donors have been synthesized that use alternating electron rich and electron poor monomers. These so-called push–pull polymers have considerably smaller Eg than the homopolymers. Record efficiencies of 5.4% [20], 6.4% [29], and 7.9% [30] have been reported. All of the published results used various new push–pull polymers to achieve the high efficiencies. Refs. [31,32] detail the synthesis and design of new donor polymers for OPV. Fullerene-based acceptors have also recently been designed that increase the Voc of the OPV devices by adjusting up the LUMO level of the acceptor. A bis-aduct of PCBM has been found to increase the Voc by 0.15 V compared to the mono-aduct [33,34]. Another acceptor, 1-(3-hexoxycarbonyl)propyl-1-phenyl[6,6]-Lu3N@C81 (Lu3N@–PCBH) uses a C80 fullerene stabilized by a Lu3N cluster and is solubilized by a similar adduct to PCBM. Lu3N@–PCBH yields an increase in the Voc of 0.2 V compared with PCBM [35,36]. No work has yet appeared that uses both the new donors and acceptors together. The increases in PCE over the last decade came about as a result of improved synthesis, morphology control, and a better understanding of the device physics. The synthesis achievements were due to increased understanding of how to synthesize low-bandgap polymers and how to control the energy levels of the polymers with respect to PCBM. However, morphology studies have been mostly performed on OC1C10–PPV/PCBM and P3HT/PCBM mixtures. The domain size, crystallinity, and curing method (temperature, solvent soak, or solvent additive) can now be chosen for these mixtures. A better understanding of optical and electrical loss mechanisms has also been gained for these mixtures. But, there is still quite a lot of academic research that needs to be performed in all three of these areas. Hopefully, the scientific community will have increased access to the new low-band-gap co-polymers for morphology and device physics studies in the near future. 3. From lab bench to rooftop In a research lab, OPV devices are fabricated in miniature to conserve materials and to allow researchers to test the greatest number of experimental possibilities. This experimental flexibility leads to the practical difficulty that experimental results and efficiency records are achieved under conditions that are not scalable to or practical for mass production of large area OPV devices. This section will address some of the issues related to scale-up. Fabrication of an OPV device involves four broad sequential steps. First, a conducting substrate with an optimized work function and surface energy is created. Second, a BHJ layer is coated onto the substrate. Next a second electrode is deposited onto the BHJ layer. Finally, the device is encapsulated in a material that protects the active layers from exposure to H2O, O2, and UV irradiation (Fig. 3). As discussed in the introduction, in order for OPV to be a viable competitor against other technologies, it is necessary to produce Regular Inverted -Mask and Etch Electrode -Clean Substrate -Prepare Surface Layer Prepare Conducting Substrate -Coat Active Layer -Remove Excess Solvent Coat Active Layer -Thermal Metal Evap OR -Coat metal NP Solution -Remove Excess Solvent -Sinter Deposit Coated Electrode -Apply/Coat Encapsulant -Remove Excess Solvent Apply Encapsulant -Mask Contacts -Clean Substrate -Prepare Surface Layer -Coat Active Layer -Remove Excess Solvent -Coat Transparent Electrode Material -Remove Excess Solvent -Apply/Coat Encapsulant -Remove Excess Solvent Fig. 3. A flow chart that shows the four sequential steps needed to fabricate an airstable OPV device for both regular and inverted fabrication. OPV at minimal cost. The fabrication will be discussed assuming a reel-to-reel fabrication. OPV devices can either be regular (i.e. holes move towards the substrate) or inverted (i.e. holes move away from the substrate). The direction of current flow is determined by the choice of electrode materials. One of the requirements for reel-to-reel coating is that the substrate be flexible so that it can be pulled through the reels. Regular devices can be fabricated onto transparent plastic foils that have been coated with a transparent conducting oxide (TCO) such as ITO, ZnO:Al or SnO2:F. The first step for fabrication has been shaded in Fig. 3 because highly conducting and transparent TCO materials are crystalline, not flexible, and will crack when drawn through bends with a low radius of curvature [37,38]. Cracks in the TCO degrade electrical performance of a device significantly. The shading indicates that this topic requires considerable innovation. An inverted device can either be coated onto a metal foil or onto a preformed metal part. Preparation of the substrate electrode requires similar steps for both device types. First the electrode must be either etched into the required shape or a resistive coating must be applied to prevent shorts where the electrodes will be externally contacted. Then the substrate material must be extensively cleaned and the electrode materials coated with a layer of material that sets the work function. Poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) is commonly used on top of ITO for a hole contact for regular devices. TiOx[39], ZnO [40], and Ca2CO3 [41] have all been investigated as electron contact interlayer materials. The second step is coating of the active layer mixture. This step is performed identically for both device types. The only difference that could arise in the film comes from possible differences in morphology. Since regular and inverted devices have differing electrodes the surface energies of the prepared substrates could be different. This difference in surface energy could lead to differing surface driven formation of morphology features. If the outer surface of the substrates have differing surface energies, the coating conditions must be adjusted to achieve similar layer thickness and dry morphology. For both device types, excess solvent must be driven from the BHJ film to ensure morphological stability [42]. Next comes the coated electrode. This step is marked in dark gray in Fig. 3 because this is the area that requires the most innovation for both device types. In the case of a regular device, a metal electrode must be deposited. A metal is always chosen for the back electrode material because it reflects transmitted light back through the active layer and thereby increases light absorption and photocurrent. Thermal evaporation of a metal is used in a research lab for electrode deposition. Thermal evaporation requires 126 A.J. Moulé / Current Opinion in Solid State and Materials Science 14 (2010) 123–130 small device areas and high vacuum conditions that are not compatible with reel-to-reel fabrication. Solution methods have been developed for printing a 50 lm Ag wire mesh from solution onto various substrates [43,37,?]. The disadvantage of this technique is that the Ag mesh has reduced transparency compared to a TCO and the ‘‘windows” between the Ag wires also have to be filled with a transparent conductor to move charges to the wires and to planarize the Ag mesh. Low-temperature deposition of a transparent conducting electrode for an inverted device is even more difficult because oxides and conducting polymers have lower conductivity and are more complex materials than metals. For both device types, heat treatments must remain below 200 °C to prevent damage to the BHJ material. Finally, an encapsulant material must be applied that blocks O2, H2O, and UV radiation from the active layer material. The requirements of the encapsulant materials are that it be transparent, flexible, durable, impermeable, and low-cost. The deposition should also not destroy any of the layers already deposited and should not require curing temperatures above 200 °C. For this step, solution-based or lamination-based application of the encapsulant layer are preferable. Also for this step, the application should be identical for both device types. 3.1. Substrates The prerequisite for an OPV substrate material is that it be highly conducting. As discussed above, the substrate can either be a metal or a transparent conductor. Research lab OPV devices are typically fabricated on indium–tin oxide (ITO) coated glass substrates that are cut and etched to a predetermined size and shape. For both regular and inverted device types, transparent ITO coated polyethylene terephthalate (PET) substrates can also be used as a substrate material. The PET substrate has the significant advantage that it is flexible, and so is compatible with a reel-to-reel coating technique. However, ITO coated onto PET has been shown to develop cracks upon bending that leads to significant reduction in the conductivity of the substrate [37]. In general, TCO electrode materials that are suitable for electrodes have crystalline domains that are likely to be brittle and perform poorly after bending. For this reason, a number of groups are working on transparent electrode materials based on organic components. Single walled carbon nanotubes (CNT) have been studied for use as a transparent electrode material [45,46]. However, CNTs are synthesized as a mixture of semiconducting and metalic tubes. Schottky contacts form at junctions between the semiconducting and metalic tubes that cause significant reductions in conductivity, resulting in high sheet resistance [47]. A second approach is to replace the TCO with a graphene-based composite [48,49]. This approach leads to high sheet-resistance electrodes and is much too expensive with the current high price of graphene. A number of groups have used a doped conjugated polymer, such as highly doped PEDOT:PSS, to replace the TCO [50–52]. The conductivity of PEDOT:PSS is lower than that of most TCOs and so the resulting devices have reduced filling factor. Nevertheless, OPV devices with a PEDOT:PSS electrode based on P3HT/ bis-PCBM has been reported with a PCE of 3.5% [53]. Another approach to making an electrode transparent is to fabricate a nanoscale metal mesh with most of the area being transparent. OPV devices have been fabricated based on Ag nanowire meshes that show promising efficiency [43,44]. A recent article looked at the relationship between series resistance (Rs) that comes from the substrate itself and IV characteristics of OPV devices. They found that low Rs is critical for maintaining a high FF and PCE, especially as device area is increased [54]. A second and essential function of the electrode in OPV devices is charge selectivity. A metal can be chosen that has a work func- tion of that matches either the HOMO or LUMO of the BHJ mixture, but because metals have unfilled densities of states able the fermi level and because they form large dipoles on their surfaces when organic materials are added, metal electrodes have poor charge selectivity [55,56]. Thin (typically <100 nm) layers of various semiconductor materials are used to create charge selectivity. These layers are known as interlayers, hole/electron transport layers or hole/electron-only layers. For regular devices, PEDOT-PSS is spin coated to make the hole-collecting electrode selective [57]. It has also been shown that PEDOT:PSS also alters the wettability of the substrate to the polymer and thereby increases charge transport [58]. V2O5 [41,59], NiO [60], and other cross-linkable hole conducting polymers [61] have also been used as to create hole selective electrodes. These replacements for PEDOT:PSS have been reported to be more thermally stable and thereby, should lead to longer device lifetimes. However all record efficiencies have been posted using PEDOT:PSS. For inverted devices a solgel of TiO2 [39,62,63], ZnO [40,64–66], Cs2CO3 [41,67,59], or conjugated polyelectrolite[68] is coated onto the ITO surface to make an electron selective contact. Several recent review articles cover the fabrication and function of interlayers that are used for regular and inverted devices [69–71]. The thin electrode coatings listed above have very small crystalline domains and are therefore resistant to cracking upon mechanical stress. The small domain sizes also lead to much higher sheet resistances than ITO. So these coating materials are not suitable for complete transparent electrodes. In the introduction it was stressed that one major advantage of OPV technology is that it is compatible with flexible substrates. In fact, the coating of BHJ layers can conceivably be controlled better using a reel-to-reel coating technique rather than spin coating. This discussion of research into substrate materials reveals that all efficiency records have been recorded using ITO coated with PEDOT:PSS as the transparent electrode in spite of the fact that ITO has been widely predicted to be too expensive [72] and has been shown to perform very poorly in mechanical stress tests [37]. PEDOT:PSS is also not ideal for use in an OPV device because it is highly acidic [42,73,74] and thought to be a source of degradation in the active layer [25]. It is therefore extremely important that a strong and dedicated research emphasis be placed on the development and characterization of new transparent electrode materials that are compatible with OPV fabrication conditions. 3.2. Coating In a lab setting, BHJ layers are typically formed using spin coating. Spin coating is a batch coating method that requires that each individual small-area substrate be loaded and coated separately. The spin-coating technique does not effectively scale-up to largearea coating. A more cost-effective alternative is to use flexible substrates that can be coated using a reel-to-reel technique in which the substrate is passed through a series of reels and the layers are sequentially coated using a compatible coating technique. Several groups have started working on reel-to-reel coating methods for OPVs. They have developed an inkjet printing (Fig. 4a) techniques for P3HT/PCBM BHJ layers on regular electrodes that yielded power conversion efficiencies of over 3% [75– 77]. The authors reported that the morphology formed from inkjet printing is distinctly different than from spin-coated or doctorbladed layers [78]. This result suggests that the current practice of studying the morphology of BHJ devices fabricated from spin coating may yield scientific data that is not technically relevant for the coating techniques most likely to be used. Shaheen et al. have used screen printing (Fig. 4b) to coat OC1C10–PPV/PCBM OPV devices [79]. Kim et al. have used a brush coating technique to coat P3HT/PCBM layers. They found that the PCE is highly dependent upon the substrate temperature during coating. A high A.J. Moulé / Current Opinion in Solid State and Materials Science 14 (2010) 123–130 127 Fig. 4. Schematic diagrams of (a) inkjet printing [76], (b) screen printing [79], (c) spray coating [81], and (d) slot die coater heads for single and multiple layers [42]. coating temperature led to phase separation on a larger length scale [80]. A likely coating technique to be used for industrial scale-up of BHJ printing onto flexible substrates is slot-die coating (Fig. 4d) [42,73,74]. Slot-die coating has the advantages of being amenable to reel-to-reel printing, it is a metered coating technique, which means that no ink is lost, and either single or multiple layers can be coated. Reel-to-reel coating methods are useful for coating onto flexible substrates, which should be an advantage of OPV technology since the active layer film is highly flexible. However, for many applications a flexible substrate is not suitable. For example if an OPV coating is to be used as a photovoltaic replacement for automobile paint, the BHJ layer would have to be coated onto the ridged and curved pieces of an auto body. For this application, a spray coating (Fig. 4c) technique is more appropriate. Several groups have spray coated OPV devices onto glass/ITO substrates and achieved PCEs of over 3% [82,83,81,84–87]. The main issues for spray-coating appear to be formation of sufficiently small droplets, wetting of the substrate, film drying rate, containing airborn solvent, and reducing the surface tension of the ink. 3.3. Coated electrodes For a functional OPV device, the active OPV layer is sandwiched between two electrodes and one of the electrodes must be transparent. These requirements mean that the second electrode must be deposited on top of the active OPV layer. Since the active layer is made from a polymer, the deposition technique must be compatible with low temperatures, to prevent damage, and/or hydrophilic solvents that will not redissolve the polymer layer. The need for a low-temperature deposition technique means that sputtering cannot be used on the polymer because the sputtered material will burn the organic upon deposition. However, if an oxide layer is deposited using a sol–gel, then a metal can be sputtered on top. In a research lab, the back electrode for a regular device is typically deposited by a metal evaporation source in high vacuum condi- tions. The metal must be evaporated at a rate 61 nm/s or the polymer can be burned [40]. Evaporation requires a vacuum source and is time consuming, which makes it a poor technique choice for high-speed and low-cost production of OPV devices. Ideally, a solution processing method should exist for the deposition of metal electrodes. The current best alternative to evaporation is to coat a suspension of metal nanoparticles (commercially available) and create a low conductivity porous electrode. Again because of the restriction that temperatures over 200 °C cannot be used to avoid damage to the OPV layer, the metal particle layer cannot be sintered to increase the connectivity between the particles. A more attractive solution for metal electrode deposition currently does not exist. One interesting advantage of the deposition of metal nanoparticles for electrodes is that the deposition technique is compatible with spray-coating conditions [88]. Deposition of transparent electrodes for inverted devices is more difficult. TCO materials such as ITO, ZnO, and SnO2:F can all be deposited using sputtering techniques, but as discussed above, these techniques can burn the underlying OPV material. In addition, sputtering requires a vacuum and cannot be performed continuously in a reel-to-reel deposition. Sol–gel methods exist to deposit some oxides, such as TiO2, but the resulting layers have much too high a sheet resistance to be used as the electrode material. In the section on substrates, we identified several alternatives for deposition of transparent electrodes, including CNTs, highly doped conjugated polymers, and Ag nanowire meshes. Deposition of the top electrode is more difficult because not only must the requirements of transparency and conductivity be met for the top electrode but the material must also be prepared in a form that can easily be coated from a solution in a reel-to-reel fabrication onto a low surface energy film. These very high requirements make the development of a coatable transparent electrode material the most difficult challenge for mass production of OPV devices. If such an electrode material can be developed, it will open the way for the development of truely transformational technologies such as photovoltaic car and building paints. 128 A.J. Moulé / Current Opinion in Solid State and Materials Science 14 (2010) 123–130 3.4. Encapsulating materials Acknowledgements Most research on OPV devices is performed in N2 gloveboxes. It is well known that OPV devices will degrade quickly on exposure to air. The mechanisms for degradation follow three known pathways. First, the metal electrode is typically a low work-function metal that oxidizes on exposure to O2 or H2O [89,90]. Second, if O2 is present in the active layer, it can be photo excited to singlet
oxygen O2 which reacts with and bleaches the polymer. Third, upon exposure to long-term heating, the active layer can phase separate [91]. Both of the first two degradation mechanisms can be avoided if the O2 and H2O levels can be kept sufficiently low. However, the sensitivity of organic electronic materials to O2 is very high. O2 levels in the active layer must be kept four-ordersof-magnitude lower than can be attained using typical food packaging encapsulants [92]. Several groups have focused research efforts on creating solution-processable transparent encapsulant barriers that are based on a mixture of clay pellets and polymers [93]. Several groups have also published results that show OPV device lifetimes greater than several thousand hours, but unfortunately the identity of the encapsulant material is usually not disclosed [92,94,95]. The study of lifetime in OPV devices is particularly difficult because devices are very complicated, layered, and the defect states are below the limit of detection for most optical detection techniques. There is some evidence that defect states form at the metal electrode surface even in dark oxidant-free conditions [96]. In conclusion, it appears that encapsulant materials suitable for OPV applications have been developed though the identity of the materials is intellectual property. On the other hand there is not enough known about the basic mechanisms behind OPV device degradation. This author recommends much more basic research on the applied problem of lifetime enhancement in OPV devices with a particular emphasis on studying interfaces between layers and developing measurement techniques that are sensitive to interface states at burried interfaces. The writing of this article was funded by the California Solar Energy Collaborative and the United States Department of Energy under Grant No. DE-FG3608GO18018. Thanks to my group members David Huang, Scott Mauger, John Roehling, Lilian Chang, and Chris Rochester for editing. 4. Conclusions Organic photovoltaics is a facinating subject to study. It is necessary to be familiar with optical absorption processes, charge transport, polymer morphology, recombination kinetics, the relationship between polymer structure and electronic and optical properties, the interaction between metal and organic layers, and a host of other interesting basic science themes. The increase in the power conversion efficiency of organic photovoltaics from <2% to 7.9% in eight years is mostly due to intense and focused academic research on the basic science issues surrounding every aspect of this device type. In the past 2–3 years several companies have started to produce OPV materials and have reported record PCEs that cannot be matched by academic groups. This is, however, not the right time for reduced governmental support of academic research in this area. In fact much more research is needed. This review article has focused on the scale-up of OPV to a reel-to-reel solution-processed and mass-produced product. Almost the entire article stresses that research must still be done on flexible transparent electrodes, solution-processable electrodes (metal or transparent), developing reel-to-reel and spray coating methods, studying polymer morphology that results from the new coating methods, developing effective transparent O2 and H2O barrier materials, and learning to make all of these technologies work together. Largescale production and distribution of low-cost OPV modules can play an important role in greenhouse gas reduction and meeting global energy demands. These goals are urgent and need to be addressed by increased funding of basic and applied research in this area. References [1] Lewis NS. Powering the planet. MRS Bull 2007;32:808–20. [2] Green MA, Emery K, Hishikawa Y, Warta W. Solar cell efficiency tables (version 34). Prog Photovolt 2009;17(5):320–6. [3] Basic research needs for solar energy utilization. Technical Report, Department of Energy; 2005. [4] Brabec CJ, Hauch JA, Schilinsky P, Waldauf C. Production aspects of organic photovoltaics and their impact on the commercialization of devices. MRS Bull 2005;30(1):50–2. [5] Brabec CJ. Organic photovoltaics: technology and market. Sol Energy Mater Sol Cells 2004;83(2–3):273–92. [6] Green MA. Third generation photovoltaics: advance solar energy conversion. Springer-Verlag; 2003. [7] Tang CW. 2-Layer organic photovoltaic cell. Appl Phys Lett 1986;48(2):183–5. [8] Markov DE, Amsterdam E, Blom PWM, Sieval AB, Hummelen JC. Accurate measurement of the exciton diffusion length in a conjugated polymer using a heterostructure with a side-chain cross-linked fullerene layer. J Phys Chem A 2005;109(24):5266–74. [9] Barker JA, Ramsdale CM, Greenham NC. Modeling the current–voltage characteristics of bilayer polymer photovoltaic devices. Phys Rev B 2003;67(7):075205. [10] Yu G, Gao J, Hummelen JC, Wudl F, Heeger AJ. Polymer photovoltaic cells— enhanced efficiencies via a network of internal donor–acceptor heterojunctions. Science 1995;270(5243):1789–91. [11] Neuteboom EE, Meskers SCJ, van Hal PA, van Duren JKJ, Meijer EW, Janssen RAJ, et al. and photovoltaic properties of a new class of donor–acceptor materials. J Am Chem Soc 2003;125(28):8625–38. [12] Hoppe H, Sariciftci NS. Morphology of polymer/fullerene bulk heterojunction solar cells. J Mater Chem 2006;16(1):45–61. [13] Kim JY, Frisbie D. Correlation of phase behavior and charge transport in conjugated polymer/fullerene blends. J Phys Chem C 2008;112(45):17726–36. [14] Shaheen SE, Brabec CJ, Sariciftci NS, Padinger F, Fromherz T, Hummelen JC, et al. Appl Phys Lett 2001;78(6):841–3. [15] Padinger F, Rittberger RS, Sariciftci NS. Effects of postproduction treatment on plastic solar cells. Adv Funct Mater 2003;13(1):85–8. [16] Ma W, Yang C, Gong X, Lee K, Heeger AJ, stable Thermally. efficient polymer solar cells with nanoscale control of the interpenetrating network morphology. Adv Funct Mater 2005;15:1617–22. [17] Li G, Shrotriya V, Huang J, Yao Y, Moriarty T, Emery K, et al. High-efficiency solution processable polymer photovoltaic cells by self-organization of polymer blends. Nat Mater 2005;4:864–8. [18] Li G, Yao Y, Yang H, Shrotriya V, Yang G, Yang Y. Solvent annealing effect in polymer solar cells based on poly(3-hexylthiophene) and methanofullerenes. Adv Funct Mater 2007;17(10):1636–44. [19] Zhang FL, Jespersen KG, Bjorstrom C, Svensson M, Andersson MR, Sundstrom V, et al. Influence of solvent mixing on the morphology and performance of solar cells based on polyfluorene copolymer/fullerene blends. Adv Funct Mater 2006;16(5):667–74. [20] Peet J, Kim JY, Coates NE, Ma WL, Moses D, Heeger AJ, et al. Efficiency enhancement in low-bandgap polymer solar cells by processing with alkane dithiols. Nat Mater 2007;6(7):497–500. [21] Lee JK, Ma WL, Brabec CJ, Yuen J, Moon JS, Kim JY, et al. Processing additives for improved efficiency from bulk heterojunction solar cells. J Am Chem Soc 2008;130(11):3619–23. [22] Moulé AJ, Meerholz K. Controlling morphology in polymer–fullerene mixtures. Adv Mater 2008;20(2):240–5. [23] Moulé AJ, Meerholz K. Morphology control in solution-processed bulkheterojunction solar cell mixtures. Adv Funct Mater 2009;19:3028–36. [24] Blom PWM, Mihailetchi VD, Koster LJA, Markov DE. Device physics of polymer: fullerene bulk heterojunction solar cells. Adv Mater 2007;19(12):1551–66. [25] Moulé AJ, Meerholz K. Intensity-dependent photocurrent generation at the anode in bulk-heterojunction solar cells. Appl Phys B: Lasers Opt 2008;92(2):209–18. [26] Kooistra FB, Knol J, Kastenberg F, Popescu LM, Verhees WJH, Kroon JM, et al. Increasing the open circuit voltage of bulk-heterojunction solar cells by raising the LUMO level of the acceptor. Org Lett 2007;9(4):551–4. [27] Koster LJA, Mihailetchi VD, Blom PWM. Ultimate efficiency of polymer/ fullerene bulk heterojunction solar cells. Appl Phys Lett 2006;88(9):093511. [28] Scharber MC, Wuhlbacher D, Koppe M, Denk P, Waldauf C, Heeger AJ, et al. Design rules for donors in bulk-heterojunction solar cells—towards 10% energy-conversion efficiency. Adv Mater 2006;18(6):789–94. [29] http://www.konarka.com/index.php/site/pressreleasedetail/national_energy_ renewable_laboratory_nrel_certifies_konarkas_photovoltaic_s. National A.J. Moulé / Current Opinion in Solid State and Materials Science 14 (2010) 123–130 [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] Renewable Energy Laboratory (NREL) certifies Konarkas photovoltaic solar cells at 6.4. http://www.pvtech.org/news/_a/solarmer_breaks_organic_solar_pv_cell_ conversion_efficiency_record_hits_nre/. Solarmer breaks organic solar PV cell conversion efficiency record, hits NREL-certification, website; December 2009. Kroon R, Lenes M, Hummelen JC, Blom PWM, De Boer B. Small bandgap polymers for organic solar cells (polymer material development in the last 5 years). Polym Rev 2008;48(3):531–82. Thompson BC, Frechet JMJ. Organic photovoltaics–polymer–fullerene composite solar cells. Angew Chem Int Ed 2008;47(1):58–77. Lenes M, Wetzelaer G-JAH, Kooistra FB, Veenstra SC, Hummelen JC, Blom PWM. Fullerene bisadducts for enhanced open-circuit voltages and efficiencies in polymer solar cells. Adv Mater 2008;20(11):2116–9. Martijn L, Steve WS, Alex BS, David FK, Jan CH, Paul WMB. Electron trapping in higher adduct fullerene-based solar cells. Adv Funct Mater 2009;19(18): 3002–7. Ross RB, Cardona CM, Guldi DM, Sankaranarayanan SG, Reese MO, Kopidakis N, et al. Endohedral fullerenes for organic photovoltaic devices. Nat Mater 2009;8(3):208–12. Ross RB, Cardona CM, Swain FB, Guldi DM, Sankaranarayanan SG, Van Keuren E, et al. Tuning conversion efficiency in metallo endohedral fullerene-based organic photovoltaic devices. Adv Funct Mater 2009;19(14):2332–7. Krebs F. Polymer photovoltaics: a practical approach. SPIE Press; 2008. Paetzold R, Heuser D, Henseler K, Roeger S, Wittmann G, Winnacker A. Performance of flexible polymeric light-emitting doedes under bending conditions. Appl Phys Lett 2003;82(19). Waldauf C, Morana M, Denk P, Schilinsky P, Coakley K, Choulis SA, et al. Highly efficient inverted organic photovoltaics using solution based titanium oxide as electron selective contact. Appl Phys Lett 2006;89(23). White MS, Olson DC, Shaheen SE, Kopidakis N, Ginley DS. Inverted bulkheterojunction organic photovoltaic device using a solution-derived ZnO underlayer. Appl Phys Lett 2006;89(14). Li G, Chu CW, Shrotriya V, Huang J, Yang Y. Efficient inverted polymer solar cells. Appl Phys Lett 2006;88(25). Krebs FC. Fabrication and processing of polymer solar cells: a review of printing and coating techniques. Sol Energy Mater Sol Cells 2009;93(4):394–412. Tvingstedt K, Inganas O. Electrode grids for ITO-free organic photovoltaic devices. Adv Mater 2007;19(19):2893. Lee J-Y, Connor ST, Cui Y, Peumans P. Solution-processed metal nanowire mesh transparent electrodes. Nano Lett 2008;8(2):689–92. doi:10.1021/nl073296g. Wu ZC, Chen ZH, Du X, Logan JM, Sippel J, Nikolou M, et al. conductive carbon nanotube films. Science 2004;305(5688):1273–6. Rowell MW, Topinka MA, McGehee MD, Prall HJ, Dennler G, Sariciftci NS, et al. Organic solar cells with carbon nanotube network electrodes. Appl Phys Lett 2006;88(23). Topinka MA, Rowell MW, Goldhaber-Gordon D, McGehee MD, Hecht DS, Gruner G. Charge transport in interpenetrating networks of semiconducting and metallic carbon nanotubes. Nano Lett 2009;9(5):1866–71. Stankovich S, Dikin DA, Dommett GHB, Kohlhaas KM, Zimney EJ, Stach EA, et al. Graphene-based composite materials. Nature 2006;442(7100):282–6. Wu J, Becerril HA, Bao Z, Liu Z, Chen Y, Peumans P. Organic solar cells with solution-processed graphene transparent electrodes. Appl Phys Lett 2008;92(26):263302–3. Yinhua Z, Fengling Z, Kristofer T, Sophie B, Fenghong L, Wenjing T, et al. Investigation on polymer anode design for flexible polymer solar cells. Appl Phys Lett 2008;92(23):233308. Huang J, Wang X, Kim Y, deMello AJ, Bradley DDC, Demello JC. High efficiency flexible ITO-free polymer/fullerene photodiodes. Phys Chem Chem Phys 2006;8(33):3904–8. Hau SK, Yip H-L, Zou J, Jen AKY. Indium tin oxide-free semi-transparent inverted polymer solar cells using conducting polymer as both bottom and top electrodes. Org Electron 2009;10(7):1401–7. Do H, Reinhard M, Vogeler H, Puetz A, Klein MFG, Schabel W, et al. Thin Solid Films 2009;517(20):5900–2. Harding MJ, Poplavskyy D, Choong V-E, So F, Campbell AJ. Variations in hole injection due to fast and slow interfacial traps in polymer light-emitting diodes with interlayers. Adv Funct Mater 2010;20(1):119–30. Mihailetchi VD, Blom PWM, Hummelen JC, Rispens MT. Cathode dependence of the open-circuit voltage of polymer: fullerene bulk heterojunction solar cells. J Appl Phys 2003;94(10):6849–54. Braun S, Salaneck WR, Fahlman M. Energy-level alignment at organic/metal and organic/organic interfaces. Adv Mater 2009;21(14–15):1450–72. Frohne H, Shaheen SE, Brabec CJ, Muller DC, Sariciftci NS, Meerholz K. Influence of the anodic work function on the performance of organic solar cells. ChemPhysChem 2002;3(9):795–9. Mitchell WJ, Burn PL, Thomas RK, Fragneto G, Markham JPJ, Samuel IDW. Relating the physical structure and optical properties of conjugated polymers using neutron reflectivity in combination with photoluminescence spectroscopy. J Appl Phys 2004;95(5). Liao HH, Chen LM, Xu Z, Li G, Yang Y. Highly efficient inverted polymer solar cell by low temperature annealing of Cs2CO3 interlayer. Appl Phys Lett 2008;92(17). Irwin MD, Buchholz B, Hains AW, Chang RPH, Marks TJ, et al. p-type semiconducting nickel oxide as an efficiency-enhancing anode interfacial layer in polymer bulk-heterojunction solar cells. Proc Natl Acad Sci USA 2008;105(8):2783–7. 129 [61] Hains AW, Liu ABF, Martinson J, Irwin MD, Marks TJ. Anode interfacial tuning via electron-blocking/hole transport layers and indium tin oxide surface treatment in bulk-heterojunction organic photovoltaic cells. J Mater Chem 2010;20:595–606. [62] Yoshikawa O, Fujieda T, Uehara K, Yoshikawa S. High performance polythiophene/fullerene bulk-heterojunction solar cell with a TiOx hole blocking layer. Appl Phys Lett 2007;90:163517. [63] Bouclé J, Ravirajan P, Nelson J. Hybrid polymer–metal oxide thin films for photovoltaic applications. J Mater Chem 2007;17:3141–53. [64] Olson DC, Lee Y-J, White MS, Kopidakis N, Shaheen SE, Ginley DS, et al. Effect of ZnO processing on the photovoltage of ZnO/poly(3-hexylthiophene) solar cells. J Phys Chem C 2008;112(26):9544–7. [65] Yip HL, Hau SK, Baek NS, Ma H, Jen AKY. Polymer solar cells that use selfassembled-monolayer-modified ZnO/metals as cathodes. Adv Mater 2008;20(12):2376. –+ [times cited: 2]. [66] Yip HL, Hau SK, Baek NS, Jen AKY. Self-assembled monolayer modified ZnO/ metal bilayer cathodes for polymer/fullerene bulk-heterojunction solar cells. Appl Phys Lett 2008;92(19) [times cited: 3]. [67] Huang YYJ, Xu Z. Low-work-function surface formed by solution-processed and thermally deposited nanoscale layers of cesium carbonate. Adv Funct Mater 2007;17(12):1966–73. doi:10.1002/adfm.200700051. [68] He C, Zhong C, Yang R, Yang R, Huang F, Bazan GC, et al. Origin of the enhanced open-circuit voltage in polymer solar cells via interfacial modification using conjugated polyelectrolytes. J Mater Chem 2010;20:2617–22. [69] Chen LM, Xu Z, Hong Z, Yang Y. Interface investigation and engineering— achieving high performance polymer photovoltaic devices. J Mater Chem 2010;20:2575–98. [70] Steim R, Brabec CJ. Interface materials for organic solar cells. J Mater Chem 2010;20:2499–512. [71] Ma H, Yip H-L, Huang F, Jen AK-Y. Interface engineering for organic electronics. Adv Funct Mater 2010;20:1371–88. [72] Yang F, Forrest SR. Organic solar cells using transparent SnO2-f anodes. Adv Mater 2006;18(15):2018. [73] Krebs FC. Polymer solar cell modules prepared using roll-to-roll methods: knife-over-edge coating, slot-die coating and screen printing. Sol Energy Mater Sol Cells 2009;93(4):465–75. [74] Krebs FC, Jorgensen M, Norrman K, Hagemann O, Alstrup J, Nielsen TD, et al. A complete process for production of flexible large area polymer solar cells entirely using screen printing—first public demonstration. Sol Energy Mater Sol Cells 2009;93(4):422–41. [75] Xia YJ, Friend RH. Polymer bilayer structure via inkjet printing. Appl Phys Lett 2006;88(16):163508. [76] Hoth CN, Choulis SA, Schilinsky P, Brabec CJ. High photovoltaic performance of inkjet printed polymer: fullerene blends. Adv Mater 2007;19(22):3973–8. [77] Hoth CN, Schilinsky P, Choulis SA, Brabec CJ. Printing highly efficient organic solar cells. Nano Lett 2008;8(9):2806–13. [78] Schilinsky P, Waldauf C, Brabec C. Performance analysis of printed bulk heterojunction solar cells. Adv Funct Mater 2006;16:1669–72. [79] Shaheen SE, Radspinner R, Peyghambarian N, Jabbour GE. Fabrication of bulk heterojunction plastic solar cells by screen printing. Appl Phys Lett 2001;79(18):2996–8. [80] Kim S-S, Na S-I, Jo J, Tae G, Kim D-Y. Efficient polymer solar cells fabricated by simple brush painting. Adv Mater 2007;19:4410–5. [81] Steirer KX, Reese MO, Rupert BL, Kopidakis N, Olson DC, Collins RT, et al. Ultrasonic spray deposition for production of organic solar cells. Sol Energy Mater Sol Cells 2009;93(4):447–53. [82] Treossi E, Liscio A, Feng XL, Palermo V, Mullen K, Samori P. Large-area bicomponent processing of organic semiconductors by spray deposition and spin coating with orthogonal solvents. Appl Phys A—Mater Sci Process 2009;95(1):15–20. [83] Hoth CN, Steim R, Schilinsky P, Choulis SA, Tedde SF, Hayden O, et al. Topographical and morphological aspects of spray coated organic photovoltaics. Org Electron 2009;10(4):587–93. [84] Girotto C, Rand BP, Genoe J, Heremans P. Exploring spray coating as a deposition technique for the fabrication of solution-processed solar cells. Sol Energy Mater Sol Cells 2009;93(4):454–8. [85] Vak D, Kim S-S, Jo J, Oh S-H, Na S-I, Kim J, et al. Fabrication of organic bulk heterojunction solar cells by a spray deposition method for low-cost power generation. Appl Phys Lett 2007;91(8):081102. [86] Green R, Morfa A, Ferguson AJ, Kopidakis N, Rumbles G, Shaheen SE. Performance of bulk heterojunction photovoltaic devices prepared by airbrush spray deposition. Appl Phys Lett 2008;92(3):033301–3. [87] Ishikawa T, Nakamura M, Fujita K, Tsutsui T. Preparation of organic bulk heterojunction photovoltaic cells by evaporative spray deposition from ultradilute solution. Appl Phys Lett 2004;84(13):2424–6. [88] Hau SK, Yip H-L, Leong K, Jen AKY. Spraycoating of silver nanoparticle electrodes for inverted polymer solar cells. Org Electron 2009;10(4):719–23. [89] Kim JB, Kim CS, Kim YS, Loo YL. Oxidation of silver electrodes induces transition from conventional to inverted photovoltaic characteristics in polymer solar cells. Appl Phys Lett 2009;95(18). [90] Reese MO, White MS, Rumbles G, Ginley DS, Shaheen SE. Optimal negative electrodes for poly(3-hexylthiophene): [6,6]-phenyl c61-butyric acid methyl ester bulk heterojunction photovoltaic devices. Appl Phys Lett 2008;92(5). [91] Chirvase D, Parisi J, Hummelen JC, Dyakonov V. Influence of nanomorphology on the photovoltaic action of polymer–fullerene composites. Nanotechnology 2004;15(9):1317–23. 130 A.J. Moulé / Current Opinion in Solid State and Materials Science 14 (2010) 123–130 [92] Dennler G, Lungenschmied C, Neugebauer H, Sariciftci NS, Latreche M, Czeremuszkin G, et al. A new encapsulation solution for flexible organic solar cells. Thin Solid Films 2006;511:349–53. [93] Ebina T, Mizukami F. Flexible transparent clay films with heat-resistant and high gas-barrier properties. Adv Mater 2007;19(18):2450–3. [94] Houch JA, Schilinsky P, Choulis SA, Rajoelson S, Brabec CJ. The impact of water vapor transmission rate on the lifetime of flexible poly solar cells. Appl Phys Lett 2008;93:103306. [95] Lungenschmied C, Dennler G, Neugebauer H, Sariciftci SN, Glatthaar M, Meyer T, et al. organic solar cells. Sol Energy Mater Sol Cells 2007;91(5):379– 84. [96] Reese MO, Morfa AJ, White MS, Kopidakis N, Shaheen SE, Rumbles G, et al. Pathways for the degradation of organic photovoltaic p3ht:Pcbm based devices. Sol Energy Mater Sol Cells 2008;92(7):746–52.