INDUSTRY PERSPECTIVE | TECHNOLOGY FOCUS

Organic materials Fantastic plastic

Polymer materials could bring down the cost of electricity production using photovoltaic technology to below $1 per watt for the first time, and enable mass-market, portable applications for photovoltaic technology.

Russell Gaudiana* and Transparent packaging † Transparent Christoph Brabec electrode Konarka Technologies, 100 Foot of John Street, Boott Mill South, 3rd Floor Lowell, Printed active 01852, USA material e-mail: *[email protected]; †[email protected] Primary uch of the early work on electrode photoactive materials for Light focused on M Substrate , which dominates Transparent packaging the commercial solar-energy field Electrons today. Several other materials, such as Transparent electrode amorphous silicon (a-Si), cadmium Active material External load telluride (CdTe) and copper indium (polymer blend) gallium selenide (CIGS), are also now in various stages of commercialization, and Primary electrode are known as thin-film technologies. The Substrate lower manufacturing costs and higher production throughput of these materials End view Angle view KONARKA potentially translate into lower electricity costs. Current thin-film technologies are expected to bring costs reasonably close to Figure 1 Polymer-based photovoltaic cells can be manufactured using standard printing processes. $1 per watt of electricity produced at peak solar power. There is, however, another technology that has the potential to and nanorods, metal oxides stained with dissolved in organic solvents or water, bring this cost down even further. Bulk dye molecules, as well as combinations are applied to a plastic sheet by means of heterojunction technology, using organic of these. a coating applicator. semiconductors and roll-to-roll coating Of all these technology platforms, Various printing and coating and printing techniques, could become organic photovoltaics is generating technologies have proven their the technology that makes considerable interest (Box 1). As compatibility with organic affordable to the general public. The the name implies, this technology semiconductor processing, among them technology uses abundantly available comprises carbon-based materials as gravure printing, flexo printing, screen non-toxic materials, is based on a donor and acceptor molecules. The printing, slot die coating and, most scalable production process with high most popular class of organic donor recently, ink-jet printing. The printing productivity, and requires low investment molecules are conjugated polymers, solvent is evaporated when heated to from the manufacturer. such as polythiophenes, polyfluorenes moderate temperatures, producing a A bulk heterojunction is a blend of or polycarbazoles. The material choice dried layer of the photoactive polymer. p- and n-type semiconductors, which for acceptors is much narrower — for The modules are encapsulated between forms molecular p–n diodes all over more than ten years substituted thin, flexible over-laminates, which the bulk layer. On light absorption, fullerenes have given by far the best protect the active layers from mechanical photo-induced charges are produced photovoltaic performance. abrasion and the environment. Capital by ultrafast charge transfer (within a The feature that differentiates this costs are very low, and the printing and few femtoseconds) between the two technology from all of the others is coating processes can be done at high semiconductor types. Various material its compatibility with high-speed and speed with no obvious limitation in the systems have been suggested for bulk low-temperature roll-to-roll processing. substrate width. The processing steps are heterojunction solar cells, including: The processes are typical of those used summarized in Fig. 1. organic semiconductors, inorganic and in the printing and coating industry in The combination of low-temperature organic semiconductor nanoparticles that solutions of the active materials, processing paired with high production nature photonics | VOL 2 | MAY 2008 | www.nature.com/naturephotonics 287 © 2008 Nature Publishing Group

INDUSTRY PERSPECTIVE | TECHNOLOGY FOCUS INDUSTRY PERSPECTIVE | TECHNOLOGY FOCUS

10 12 been made to the polymer structures 8 10 over the past few years (Fig. 2). These modifications include the incorporation 8 6 of comonomers that withdraw electrons 6 from the sea of electrons on the polymer 4 4 backbone, causing a large shift in the Current (mA) 2 Current (mA) absorption band towards the infrared. 2 Some of these modifications result in 0 0 absorption at wavelengths as long as 1,000 nm and more, enabling the cell –2 –2 –0.2 0.0 0.2 0.4 0.6 0.8 1.0 –0.2 –0.10.0 0.1 0.2 0.3 0.4 0.5 0.6 to absorb more than 50% of the total L radiation from the Sun. In addition, the

NRE Voltage (V) Voltage (V) structures are versatile enough, from a molecular architecture standpoint, Figure 2 These plots show record efficiencies achieved with organic photovoltaic technology as certified by the that the entire visible spectrum can be National Renewable Energy Lab (NREL). a, Results for a device with an active area of 1.024 cm2 and an efficiency covered as well. Modelling indicates of 5.21%. b, Results for a device with an active area of 0.685 cm2 and an efficiency of 5.24%. that some of these polymers, when combined with fullerene, will exhibit cell efficiencies between 7% and 10%. Even higher efficiency values are throughput suggests that attractive no complete life-cycle analysis has been expected for tandem or multiple junction energy payback times — the time it takes completed, but expectations are that the geometries, where solar cells of different to generate energy equivalent to that energy payback time can be as low as a bandgaps are stacked on top of each outlaid during fabrication — should be few weeks. other and interconnected in series. Each possible. With large-volume manufacture The efficiency of organic photovoltaic cell absorbs at a different wavelength, and reasonable efficiencies of 5% to technology is low when compared with reducing the amount of uncaptured 10%, these printed solar cells should silicon or compound semiconductor radiation that is lost as heat and enabling have the potential to go significantly technologies. However, many significant higher efficiencies. The materials for below $1 per watt of electricity. So far, modifications and improvements have tandem cells can come from a variety of

Box 1 The structure of an organic photovoltaic cell

All photovoltaic cells have several the near infrared. As the active-layer common features: they must have two p-type carrier: coating is a fraction of a micrometre electrodes and a layer of photoactive electron donators thick (100–200 nm), the polymers must material that absorbs light (a photon) have a very high absorptivity as well. + and generates current (an electron). When choosing which polymers to use Positive charge + + The key to the success of organic e– as the donor and acceptor molecules, e– photovoltaics is its two-component Negative charge e– material scientists must look at the active layer, which on coating and different energy states of the molecular drying forms a very unique morphology electrons in the donor and the acceptor n-type carriers: (shown schematically in Fig. B1) molecule. All charge carriers need electron acceptors Primary electrode referred to as a bulk heterojunction. to be transported across the bulk to The main feature of this heterojunction the electrodes before recombination morphology is the intertwining of takes place. At typical carrier lifetimes phases of each of the components, which of a few microseconds, charge- spontaneously occurs when the solvent Figure B1 The mechanism of charge transfer and carrier mobilities of 10–3 cm2 V–1 s–1 is evaporated. In current designs of transport in a bulk heterojunction structure. or higher are required for loss-free polymer photovoltaic cells, one of the carrier collection. components is a polymer that has three These selection criteria significantly functions: to absorb light; to inject an narrow down the polymeric structures electron into the second component; External quantum efficiencies of up to that have potential for efficient and to carry the resultant hole to one 80% have already been demonstrated, photovoltaic energy conversion. Finding of the electrodes. The other component though today’s power-conversion suitable polymers for this application is a fullerene derivative. Its function is efficiency is only in the regime of 5–6%. is challenging because a compromise to accept an electron and carry it to the The current generated by the cell must always be reached between the other electrode. The primary advantage is related to the absorption spectrum choice of bandgap (which dominates of the bulk heterojunction is the very of the polymer, which is determined the short-circuit current) and the position high surface area that is formed between by its molecular structure. The of the electronic levels (which dominates the two phases, which directly affects structural design of these molecules the open-circuit voltage). The product of the efficiency of charge transfer between is adjusted to absorb broadly across the short-circuit current and the open- the polymer and the fullerene phases. the solar spectrum from the blue to circuit voltage dominates the efficiency.

288 nature photonics | VOL 2 | MAY 2008 | www.nature.com/naturephotonics © 2008 Nature Publishing Group

INDUSTRY PERSPECTIVE | TECHNOLOGY FOCUS INDUSTRY PERSPECTIVE | TECHNOLOGY FOCUS

conjugated polymers, always combining Potential applications are in a narrow-bandgap semiconductor with electricity-generating awnings that can a wide-bandgap one. Efficiencies over be rolled up for storage or windows in 6% have already been reported for a office buildings and greenhouses, where tandem architecture based on polymer two transparent electrodes are used so and fullerene composites (Kim, J. Y. et al. that the module is semitransparent. Science 317, 222–225; 2007). The biggest market for organic solar Although still low compared with panels is large-area rooftop applications silicon and compound semiconductor in residential and commercial technologies, these efficiencies buildings. This will require efficiencies represent a big step forward in polymer between 7% and 10% and lifetimes of photovoltaic technology. Polymer 7–10 years. Konarka is now running systems also have many advantages over extensive lifetime investigations, and conventional photovoltaic technology. the company’s cells regularly pass more For example, organic photovoltaic than 1,000 hours under accelerated modules are lightweight, have a high degradation. To verify the lifetime power-to-weight ratio (more than of 7–10 years, the degradation over 100 mW g–1 at 5% efficiency) and they around 3–5 years must be measured are mechanically flexible (Fig. 3). This and then extrapolated. So far, Konarka makes them particularly useful for has been measuring the lifetime

portable applications, which represent the KONARKA of its cells for more than one year first target markets for the technology. (extrapolated out to three years) with no Potential uses include battery chargers for degradation observed. mobile phones, laptops, radios, flashlights, Figure 3 Polymer photovoltaic technology has a high This shows the enormous progress toys, and almost any handheld device power-to-weight ratio, which allows thin modules that that has been made over the past few that uses a battery. The modules can be have the added advantage of being lightweight and years in designing environmentally stable adhered to the outside face of a briefcase flexible, making it ideal for portable applications. organic materials and interfaces. Until or a piece of clothing, or they can be recently it was thought that organic incorporated into the housing of a device. solar-cell lifetimes were restricted to They can also be rolled up or folded for 1–2 years, but today it is clear that the storage in a pocket when not in use. The Organic photovoltaic modules will technology could reach lifetimes of first products serving this market should probably also be used in architectural 5–10 years, making it a strong contender be available later in 2008. or building-integrated applications. for consumer solar-cell applications.

nature photonics | VOL 2 | MAY 2008 | www.nature.com/naturephotonics 289 © 2008 Nature Publishing Group