Nanostructured Materials for Photovoltaic Cells

Transcription

1 Nanostructured Materials for Photovoltaic Cells Michael Beerman, Jasmine Gilchrist, and Harry Lewis Conventional solar cell devices were originally developed by government space programs and provide energy conversion efficiencies up to 30%, however they remain expensive due to stringent material purity requirements. Recent studies have shown that new hybrid polymer - nanoparticle solar cells have theoretical efficiencies that far exceed conventional systems. In this report we provide an overview of the field of photovoltaics, from conventional solid-state devices to current efforts with materials such as conjugated polymers and semiconducting nanoparticles. We will examine the theoretical limits of nanoparticle based systems and discuss the current challenges that researchers must surmount in order to achieve the high efficiencies required to be competitive in the consumer market. Photovoltaic (PV) systems are solid-state devices that directly convert incident light energy into electric current. In 1839, E. Becquerel first discovered that a current could be generated when two electrodes in an electrolytic solution were exposed to sunlight. However, photovoltaics were not actively pursued until the space race of the 1950 s and 60 s provided funding and incentive to produce efficient self-contained power generating systems. DM Chapin found that single crystalline silicon cells could produce current with an efficiency of 6% in The world s first solar cell powered satellite, Vanguard 1, was launched in It carried two radio transmitters, one powered by a battery, the other by photovoltaics. The battery powered unit ceased transmitting after 20 days, while the PV powered unit continued emitting radio signals until Vanguard 1 demonstrated the potential for improving the longevity of space craft missions, thus ushering in a decades long research program in the field of photovoltaics.[bailey] The US and Soviet space programs typically operated under very large budgets. Hence, expensive materials such as pure single crystalline silicon and GaAs could be utilized, with little research emphasis on reducing manufacturing costs. Terrestrial power generation, however, must compete with cheaper forms of energy generation, such as fossil fuels. Therefore, the cost of materials and fabrication must drop significantly before solar cells become competitive in today s energy market. Novel materials and fabrication methods, such as spin-coated electrically-conductive polymers and sol-gel synthesized semiconducting nanoparticles show promise to improve efficiency and reduce cost so that PV cells will become more pervasive in the near future. Conventional Devices Before discussing these novel materials and methods, it is important to understand conventional semiconducting photovoltaic systems. Conventional PV systems are composed of a cathode, p-doped and n-doped semiconductor regions, and an anode, as seen in Figure 1. When the device is in equilibrium, Figure 1a, the Fermi energy, E feq, is constant across all regions. When exposed to light, valence electrons are excited across the band gap - 1 -

2 into the conduction band, leaving a hole in the valence band. Promotion of electrons produces a non-equilibrium state, Figure 1b, in which the Fermi level, E fnoneq, shifts to a lower energy in the p-type region, and to a higher energy in the n-type region. The depletion zone is where the n- and p-doped regions meet. A diffusion gradient is present in the depletion region that acts to separate positively charged holes from negatively charged electrons: holes diffuse into the p-region while electrons diffuse into the n-region. A potential is produced due to this separation in charge, which generates a current that may be applied to an external load. Eventually the holes recombine with electrons at the surface of the cathode to complete the circuit.[brown] Figure 1: States of a solid-state semiconducting photovoltaic device. The gray regions represent filled electron bands. Vertical axis represents energy. a) equilibrium b) non-equilibrium Excitons are unstable particles with a relatively short half-life. They are formed when an electron is excited into the conduction band, but still remains quantum mechanically linked to the hole left behind in the valence band. The exciton must be de-coupled in order to generate free charge carriers, otherwise it will decay as the hole and electron recombine, releasing energy as a photon or a lattice vibration. Excitons that recombine do not contribute to current generation, and hence reduce the overall efficiency of the device. Typically the coupling Figure 1: States of a solid-state semiconducting photovoltaic device. The gray regions represent filled electron bands. Vertical axis represents energy. between electron-hole pairs for inorganic devices is negligible, however, the coupling in organic devices should be considered. An exciton produced further than about 10 nm from the depletion zone may not have enough time to migrate close enough to allow for separation. Therefore it is important to maximize contact area of the p-n junction. Photovoltaic cells have a theoretical limit to the magnitude of the open circuit voltage of each cell. Open circuit voltage is measured between the anode and cathode when the external load is removed from the circuit. The theoretical limit is determined by the band gap of the semiconductor. Higher voltages are only achieved by wiring many PV cells in arrays. Materials with wider band gaps will provide greater total output voltage for the same size array, and hence greater theoretical energy conversion efficiency for the same current output. In order to maximize efficiency, earth based solar cells must absorb photons of all wavelengths within the sun s emission spectrum that are transmitted through the earth s atmosphere. This leaves wavelengths between approximately 250 and 700 nm, which encompasses long-range ultra-violet, visible light and infrared. While solid-state photovoltaic cells have established longevity as an efficient renewable energy source in solar applications, they have not proliferated in the consumer market. This is due primarily to the brittle nature of - 2 -

3 single crystalline solids and their high production costs. The ideal solar cell would be flexible, cheap, easy to manufacture and incorporate into existing products, and adaptable to different lighting conditions. PV systems based on polymers with embedded nanoparticles may prove to have all of these qualities. Polymers Polymers are ideal candidates because they are highly versatile. They can be handled outside clean room conditions, unlike inorganic materials, allowing for relatively cheap large scale production. Polymers are flexible and moldable due to their amorphous structure. Polymers can easily be dissolved, mixed or separated, and applied to many different surfaces. Additionally, the physical, electrical, mechanical, chemical, and absorption spectrum characteristics can easily be tuned by altering the number and type of side groups on the polymer chain. The following paragraphs will discuss the history and nature of conjugated polymers, their electronic structure, the mechanism and effects of doping, and charge conduction and transport within conjugated polymers. Polymers that have always existed in nature in the form of bone, skin, wood and fiber are known as the first generation of polymeric materials. The second generation of polymers dates back to the 18 th century when chemists isolated organic compounds. In 1935, Wallace Carothers created the third generation, known as synthetic polymers. Conjugated polymers are considered the fourth generation of polymeric materials. Conjugated Polymers Polymers are comprised of molecules called monomers, which are repeated molecular units in the polymer. A conjugated molecule is defined by alternating double and single bonds. Single bonds consist of only one sigmabond (σ-bond), whereas double bonds contain both a σ-bond and a pi-bond (π-bond). In a conjugated system the σ- bond backbone is comprised of overlapping sp 2 hybrid orbitals in the x-plane, while the alternating π-bonds are comprised of overlapping p orbitals in the z-plane as shown in Figure 2. Figure 2: (top) Overlap seen in π and σ bonds Figure 3: (right) An energy level diagram shows the relative energies of single and bonded electrons in both ground and excited states. This example shows p-orbital electrons forming a π-bond. How do conjugated molecules conduct? The answer lies in the energy levels of molecular orbitals. Every orbital or energy state in an atom or molecule can have up to two electrons: one spin up and the other spin down. Electrons from atoms that are not bound together occupy atomic orbitals. When two atoms are bound together, the orbitals in the valence shell overlap this is called a molecular orbital (MO). Since the energy level for molecular orbitals is lower than either of the spins individually in atomic orbitals (AO), the atoms will move to the lowest energy state possible, forming what is known as a bonding orbital or ground state. The energy level for combining same spins is higher than both of the spins individually, the atoms require energy to move to this state, which is known as the anti-bonding orbital or excited state, as described in Figure 3. These anti-bonding orbitals allow electrons to be less localized and contribute to conduction. The type of bond (σ or π) determines the level of separation between the bonding and anti-bonding orbitals. The separation between bonding and anti-bonding orbitals in sigma bonds is so great that electrons sit only in the ground state. The amount of energy required to bring the bonding electrons to the anti-bonding level would break the σ-bond. Molecules comprised of only σ-bonds are insulators because the energy gap to reach the less localized anti-bonding orbitals is too great. In order to achieve semi-conducting properties, a weaker π-bond must be present - 3 -

4 to decrease the separation between the ground and excited states. The π-bonds in double and triple bonds, require less energy for an electron to reach an excited state. The greatest delocalization is observed in alternating single and double bonds, also known as conjugation. The delocalization of electrons in polymers allows the electrons to move freely between the p z orbitals along the backbone of the polymer, giving the electrons mobility along the chain. Additionally, as more atoms are added to the chain, more π-bonding electrons are added to the system, which results in an increased number of energy states. The increase in the density of states decreases the band gap energy (E g ) between bonding and anti-bonding orbitals. The band gap, however, will level off at a certain chain length. Hence for a given polymer species, you cannot tune the band gap to absorb longer wavelengths, by simply increasing the conjugation length. In order to close the band gap further, the polymer must be doped. The length at which the change in band gap becomes negligible or independent of the number of monomer units is fuzzy and uncertain. However, some estimate the minimum effective length to be in the range of units, depending on the type of monomer used. Therefore, synthesis of a long chain is necessary to achieve the smallest band gap and thus, the greatest conductivity. The degree of delocalization along the polymer chain is difficult to measure, however, it is known that delocalization is proportional to the band gap, and that the band gap is inversely proportional to the peak in the absorption spectrum. In other words, the energy absorbed is the same amount required for the π electrons to be excited from the ground state to the excited state. The absorption spectrum of a polymer is affected by the degree of order within the polymer (including choice of side chains), choice of solvent, temperature, applied electric field, magnetic field, pressure, mechanical stress, and light [Kottan, Wallace]. Typical polymers that absorb photons within the critical range of nm appear in Figure 4. Figure 4: Some common conjugated polymers used in photo-voltaic devices. Doping Conjugated Polymers Without the help of dopants, the band gaps of these conjugated polylmers are not small enough to be used as semiconductors. For this reason, conjugated polymers did not come to the foreground of the scientific stage until polymeric doping was discovered. In 1974, Hideki Shirakawa, a chemist of the Tokyo Institute of Technology in Japan, created trans-polyacetylene (trans-pa) films. This conjugated polymer proved to be very interesting when, in 1977, Shirakawa, along with chemist Alan G. McDiarmid and physicist Alan J. Heeger, accidentally used too much iodine as a catalyst. They later discovered that the polymer s conductivity had increased 9 orders of magnitude! They had literally doped the polymer to transform the insulator into a p-type semiconductor. Doping is necessary to change the relative charge of the conjugated polymer. There are three main methods of doping: chemical, electrochemical, and transitory. Transitory methods do not involve the use of dopant counterions and include: photo-doping, charge-injection doping, and non-redox doping. Only chemical and electrochemical doping will be discussed here. In these processes, a polymer is doped by adding small and specific amounts of counterions to the polymer. The counterions exist to stabilize the doped state of the polymer. Since doping is a controlled process, scientists can obtain specific conductivity states ranging from insulating to semiconducting and sometimes even fully conducting. The process greatly effects electronic, electromagnetic, optical and structural properties of the polymer

5 P-type doping involves partial oxidation while n-type doping involves partial reduction of the delocalized π electrons of the polymer. In trans-polyactetylene, iodine is an effective oxidizing agent that creates a p-type polymer. In the same polymer, naphthalene is an effective reducing agent that creates an n-doped polymer. An example of iodine p-type doping is shown in Figure 9 [Kottan]. Figure 9: Iodine p-type doped trans-polyacetylene After doping, various characterization methods (i.e. x-ray diffraction, atomic force microscopy and scanning electron microscopy) are used to show structural changes in the polymer. Somet highly doped polymers show a large decrease in crystalline order. However, researchers have seen that an increase in order is desirable as it causes an increase in conduction and thus greater charge mobility. The effects molecular order can be seen in Figure 10 [Kottan]. So it would be more desirable, for example, to use polymers with sidechains placed at regular intervals as opposed to polymers with randomly placed side-chains, as with poly(3- hexylthiophene) (P3HT). The downside to increased order tends to be a decrease Figure 10: Typical images from an SEM (scanning electron micrograph) where the random polymer (left) is less conductive than the more ordered polymer (right). in solubility, providing the polymers with better conduction and charge transport but rendering them more difficult to process and utilize. [Ryu] Polymeric Devices Since the their discovery, doped conjugated polymers have been utilized in diverse device applications including: electrophotographic imaging, optical sensors, transistors, biomaterials, conductive electrodes, corrosion protection agents, lasers, LEDs, and, of course, PV cells. In the PV cell industry, several types of polymeric devices exist including schottky, photoelectrochemical, and heterojunction or p-n junction devices. Device structure is absolutely vital to performance because polymeric properties are affected by the method of construction and by the presence or absence of other organic semi-conductors. The schottky device consists of one polymer layer sandwiched between two electrodes, where excitons are separated at the cathode. The photoelectrochemical device is similar to the Schottky, except that it contains a solid polymer electrolyte that transports electrons from the anode back to the polymer layer. Lastly, a heterojunction device consists of two or more polymer layers sandwiched between the anode and cathode. Since they are more efficient at exciton separation, heterojunction devices have more popularity and power conversion efficiency than both schottky and photochemical cells. For these reasons, only current advances in heterojunction devices will be discussed

6 Many researchers are currently utilizing a charge transfer (CT) state in blended polymers in order to maximize absorption of polymers in the near infrared regions (NIR). A CT state is a new low energy excited state that appears in a thin layer film made out of a blend of two polymers. In work done by Ruani et al, Zinc- Phthalocyanin (Zn-Pc) and buckministerfullerenes (i.e. buckey balls or C 60 ) are used to make several devices. One device consists of independent film layers, another is a co-sublimated (blended) layer, and the last device, as seen in Figure 5, has an intermediate co-sublimated layer of blended Zn-Pc and C 60 sandwiched between Zn-Pc and C 60. Each sublimation is carried out via molecular beam deposition under ultra high vacuum. This process yields a film thickness ranging from 30 to 200 nm for each layer, as measured by an AFM and the blended bi-molecular layer has a thickness of approximately 30nm. The blended layer produces a weak CT state that has a smaller band gap than either pure material at ~1.4eV as seen in Figure 6. The CT state is a result of the increased interaction between Zn-Pc and C 60 in the blended layer. In this layer, the materials undergo a phenomenon called spinodal decomposition, where the materials segregate into micro-domains. In each domain, one material is dominant and the other acts as a dopant, lowering the band gap and allowing energy from the NIR to be utilized [Ruani]. The work done by Ruani et al found a new means of maximizing contact area between p- and n- type materials. The co-sublimation technique creates a rough surface interface with embedded microdomains that has a lower band gap, which relaxes the coupling between the electron-hole pair. Also the surface area between micro-domains is larger than a conventional p-n junction. This increase in surface area creates more opportunities for excitons to separate and be collected before recombining, allowing for potentially higher power-conversion efficiencies. AFM provides a means of examining the bi-molecular interface to determine if the thickness or type of surface should be modified with different methods of processing (i.e. varying temperatures or solvents). Figure 5: A schematic of the Zn- Pc/C 60 heterojunction cell with a 30nm film of blended materials. Figure 6: Energy level diagram of the CT state with respect to ZnPc and C 60. Continuous lines represent the expected path of holes and electrons. Ruani s group is not alone in realizing the benefits of increased p-n junction surface area. An Austrian group, headed by Sean Shaheen, discovered an improvement in p-n material blending that delivers nearly three fold improvement in efficiency over existing polymer PV devices. Shaheen s device consists of a 100 nm layer of blended p- and n-type polymeric materials sandwiched between an ITO (indium tin oxide) anode and an Al (aluminum) cathode, as shown in Figure 7 [Shaheen]. PEDOT and LiF are present only to reduce the energy barriers for hole and electron collection at the electrodes. The active layer is a mix of MDMO-PPV ((poly)[2- methyl,5-(3,7-dimethyl-octyloxy)]-para-phenylene vinylene)) and the fullerene PCBM ([6,6]-phenyl C61 butyric acid methyl ester). MDMO-PPV is used as a p-type material and PCBM is used as the n- type material. Figure 7: Device structure of plastic solar cell with chemical structures of the compounds

7 Previously, toluene and other solvents were used to blend MDMO-PPV and PCBM before spin casting films, however, the group s latest efforts involve the use of chlorobenzene, a more effective solvent. Both chlorobenzene and toluene blended films were examined with an AFM to determine the surface morphology of each blend. The toluene cast film shows horizontal features measuring approximately 0.5 µm with vertical protrusions of up to 10 nm. The chlorobenzene cast film, on the other hand, shows horizontal features measuring only 0.1 µm with vertical protrusions up to 1nm. The surface morphology can be seen in Figure 8. Mechanical stiffness and adhesion properties measured during topographic imaging show that the valleys in the film have different concentrations of PCBM than the peaks. This indicates that reduced feature sizes correspond to better blending. Great differences in feature size are due to the greater solubility of PCBM in chlorobenzene than in the other solvents. The solvent also provides open confirmations for the polymer chains. Open confirmations increase interaction between the chains, causing a slight red shift in the absorbance to approximately nm. Furthermore, improved blending creates a smoother surface to give better contact with the LiF/Al cathode. While open-circuit voltages of the cells remain the same at 0.82V, the improved blending in the chlorobenzene cast device produced more than twice the current density than the toluene device. This increase in current density allows for a jump in the power conversion efficiency from 0.9% to 2.5%. The device also shows an impressive 9.5% power conversion efficiency at the peak absorbance wavelength of 488nm. Other means of increasing the absorption range and open circuit voltage must be found to improve efficiencies before the device can become a viable competitor on the energy market. Figure 8:AFM images and surface height profiles of a) toluene and b) chlorobenzene spin coated films Although polymers have potential as components for dynamic and efficient solar cells, they have several drawbacks. While the ease of chemical and electronic modification is a major advantage, it can be a minor inconvenience as the polymers must be encapsulated to maintain their desired properties. The main problem is that polymers have low carrier mobility due to their amorphous structure. This means big problems for exciton separation and major damage to competitive power conversion efficiencies. CdSe Hybrid Solar Cells In order to overcome the problems associated with low mobilities, most research today involves hybrid solar cells since they have shown promising results. Paul Alivisatos group has chosen to focus on a hybrid solar cell made of cadmium selenide and the semicrystalline polymer, poly(3-hexylthiophene), P3HT seen in Figure 11. Hybrid solar cells combine the excellent electronic properties of inorganic molecules with the reduced cost and flexibility of amorphous substrates [lbl.gov]. Hybrid solar cells are similar to conventional solar cells, but in this case, nanoscale CdSe acts as the electron receptor while P3HT acts as the hole receptor. P3HT was chosen due to the fact that it has the highest hole mobility in conducting plastics available today, while the inorganic semiconductor CdSe has a high electron mobility. High mobilities ensure fast charge transport and separation Figure 11: Conducting polymer poly(3- hexylthiophene) of the electron-hole pair, reducing current losses due to recombination and increasing the efficiency of the cell. Electrons and holes are exchanged at the interaction surface between CdSe and P3HT. Since an electron s position and direction can only be described by a probability, or wavefunction, the particular path that an electron - 7 -

8 will take is indeterminate. In an effort to direct the electron s path, the shape of the CdSe is long and narrow, like a wire, providing a directed, low-resistance path for the electron to follow. The nanorods may also be tuned to absorb different frequencies of light. The diameter of the rod determines the band gap, where smaller diameters absorb higher frequencies of light and larger diameters absorb lower frequencies. Alternatively, the length of the rods, as seen in Figure 12, is proportional to the efficiency of the cell. It was found that electron transport exhibited in the shorter nanoparticles is dominated by inefficient hopping, while band conduction is more prevalent in the longer particles. [Huynh, Dittmer] Figure 12: Transmission electron micrographs of CdSe particles with varying particle size: (a) 7x7 nm (b) 7x30 nm (c) 7x60 nm The solar cells were made by co-dissolving 7 by 60-nm CdSe nanorods and P3HT in a pyridine and chloroform mixture. The 90% by weight CdSe solution was spin cast on an indium tin oxide glass substrate coated with a conducting polymer (see Figure 13). The efficiency of converting green light (515 nm wavelength) is 6.9% while the conversion of simulated sunlight was only 1.7%, according to preliminary results [Huynh, Peng]. In an effort to increase efficiency, it is important to increase Figure 13: PV Cell Device Structure the absorption of sunlight in the red part of the spectrum. While adjusting the diameter does allow the tuning of the bandgap somewhat, it is also important to keep a high aspect ratio (ratio of length to diameter) so that the path of the electrons is directed. Without a high aspect ratio, the electron transport will result in the inefficient hopping mentioned earlier. Another means of increasing the amount of absorption in the red part of the spectrum is by switching to other inorganic semiconductors like cadmium telluride (CdTe). Charge recombination also needs to be reduced to further increase efficiency. Increasing carrier mobilites reduces carrier densities, thereby lowering the likelihood of recombination. Enhancement of carrier mobility could be achieved by removing surface traps from the nanorod/polymer interface. This involves aligning the nanorods perpendicular to the substrate and further increasing their length. TiO 2 Hybrid Solar Cells Inorganic titanium dioxide, TiO 2, nanoparticles are n-type semiconductors that have a strong absorption peak in the ultraviolet, but are transparent to visible light. Additionally, TiO 2 nanostructures may be cheaply synthesized in a modest chemistry lab via sol gel processing.[sthathalos] Pharmaceutical companies utilize these particles as an ingredient in suntan lotion because of their UV light absorption properties and low cost. TiO 2 nanoparticles are also effective photo catalysts for treatment of air and water pollutants.[beydoun] Inorganic semiconducting nanoparticles show promise for use in photovoltaics because of improved light gathering properties as compared to conventional thin film PV cells. As discussed earlier, excitons must be generated near the interface between the n- and p- type materials. The total surface area of a collection of nanoparticles embedded in a polymer matrix is much larger then the contact area of the p-n junction in conventional devices. Additionally, the spherical geometry of particles allows for absorption of light with any angle of incidence, whereas conventional thin film systems have a maximum efficiency at incident angles close to the normal of the junction

9 Several research groups have shown interest in the optical properties and low cost of TiO 2 nanoparticles for use in photovoltaic systems. One such group at the Ecole Polytechnique University in Lausanne, Switzerland has achieved 10% efficiency with a new form of photovoltaic cell.[key] In their device, a porous network of TiO 2 nanoparticles is sandwiched between two electrodes. Figure 14 is a TEM image of a thin slice of the TiO 2 layer, showing the porous nature of the network. Since TiO 2 nanoparticles are transparent, as mentioned above, they do not absorb visible light efficiently. Therefore a photosensitive liquid dye, known as a sensitizer, is added to the network to improve absorption in the visible range. Additionally, an electrolyte is required to return the electrons from the cathode to the dye molecules. The reported 10% efficiency may be improved by broadening the spectrum over which the device can absorb light. Currently the system harvests only 45% of incident light because the absorption band is limited to high frequencies. By tapping into the lower frequency infrared portion of the spectrum, they predict efficiencies up to 20%. Figure 14: Transmission electron micrograph of anatase TiO 2 nanoparticles. The size distribution is about 5 to 20 nm. [Gratzel] Unfortunately this system has one major drawback: the electrolyte decomposes in sunlight! It may prove possible, however, to replace the liquid electrolyte with a charge transporting solid film. Another drawback is that the device may prove difficult to manufacture in large arrays. The researchers are confident that these setbacks are only temporary. If an appropriate solid film can be found along with a scalable manufacturing technique then future versions of this system may prove competitive in the energy market. Hybrid nanoparticle-polymer systems are another approach to photovoltaic devices. A team of scientists from the University of California, Santa Cruz and IBM Almaden Research Center are utilizing thin films of TiO 2 nanoparticles embedded in a polymer matrix.[arango] They report an open circuit voltage of 0.7 V, however, the total quantum efficiency is only 4%. The anode of this device is an ITO patterned glass substrate, which has been surface activated with 3- aminopropyl. A viscous water solution of TiO 2 nanoparticles is then spread onto the ITO substrate, dried and annealed at 500 C to fuse the particles. The particles have a diameter of 80 nm, and self assemble into a regular array at the surface of the anode, with about 20 nm pores between particles, while further from the surface the particles are randomly dispersed. A thin layer of MEH-PPV polymer is then spin-cast over the TiO 2 layer. The polymer permeates into the pores, and the film transforms from opaque to transparent. A calcium cathode completes the circuit. Figure 15 is a schematic of the device. The open circuit voltage is dependent on the surface roughness of the ITO and the degree of selfassembly of the TiO 2 monolayer. Apparently charge collection is dominated by exciton diffusion at the TiO 2 interface, however, the authors struggled to provide an accurate model to explain charge transfer across the multi-component system. This device does not incorporate a cumbersome liquid electrolyte, hence it is stable in sunlight, cheap to manufacture and relatively easy to produce large arrays. Of course the efficiency must increase by a factor of about five before being readily accepted on the energy market. Figure 15: A hybrid polymer-nanoparticle PV cell. Conclusion The nanoparticles self-assemble at the ITO anode The field of photovoltaics has been steadily and fuse to form a porous network. progressing over the last half century. Research focusing on alternative materials promises breakthroughs in efficiency and new applications. Despite the obstacles, - 9 -

10 the development of polymer-based solar cells has advanced over the last few years, and commercialization will soon be within rach. The field would greatly benefit from a working model of charge transfer between the various components in these hybrid systems. Tribological characterization techniques, such as atomic force microscopy, could play an important role in determining the roughness properties at the complex interfaces, and hence provide insight for a functional charge transfer model. Armed with this model, polymer-nanoparticle hybrid device technology could leap forward and replace conventional solid-state solar cells in many areas. In addition, due to the limitations of conventional cells, previously untapped applications may emerge. For instance, future polymer based cells may be painted and cured on exterior surfaces of buildings and connected to the power grid. Lower costs and flexibility will usher in a new era of renewable energy, where alternative power generation is no longer hindered by high cost and cumbersome implementation methods. References 1. Bailey, S.G., Raffaelle, R., Emery, K., Space and Terrestrial Photovoltaics: Synergy and Diversity, Progress in Photovoltaics: Research and Applications, 10, , Brown, R.F., Solid State Physics, El Corral Publications, Sthathatos E., Tsiourvas D., Lianos P., Titanium dioxide films made from reverse micelles and their use for the photocatalytic degradation of adsorbed dyes, J. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 149, Issue: 1-3, April 15, 49-56, Beydoun D., Amal R., Low G., S Mcevoy, Role of nanoparticles in photocatalysis, Journal of Nanoparticle Research, 1, , Key, A., Gratzel, M., Low cost photovoltaic modules based on dye sensitized nanocrystalline titanium dioxide and carbon powder, Solar Energy Materials and Solar Cells, 44, no. 1, , Arango, A.C., Carter, S.A., Brock, P.J., Charge transfer in photovoltaics consisting of interpenetration networks of conjugated polymer and TiO 2 nanoparticles, Applied Physics Letters, 74, , Ryu, K.S., Chang, S.H., Kang, S-G., Oh, E.J., Yo, C.H., Physicochemical and Electrical Characterization of Polyaniline Induced by Crosslinking, Stretching, and Doping, Bull. Korean Chem. Soc., 20, No. 3, , Gratzel, M., Prospectives for Dye-sensitized Nanocrystalline Solar Cells, Progress in Photovoltaics: Research and Applications, 8, , Wallace, G.G., Dastoor, P.C., Officer, D.L., Too, C.O., Conjugated Polymers: New materials for photovoltaics, Chemical Innovation, 30, no. 1, 14-22, April Kottan Workshop Notes. June Ruani, G., Fontanini, M.M., Taliani, C., Weak Intrinsic Charge Transfer complexes: a new route for developing wide spectrum organic photovoltaic cells, J. Chem. Phys., 116, no. 4, , January 22, New Hybrid Solar Cells Combine Nanotech with Plastics. March 29, Articles/Archive/MSD-Alivisatos-solarcells.html 13. Huynh, W. U., Peng, X. G., Alivisatos, A.P. CdSe Nanocrystal Rods/Poly(3-hexylthiophene) Composite Photovoltaic Devices. Advanced Materials, 11, 923 (1999). 14. Huynh, W.U., Dittmer, J.J., Alivisatos, A.P. Hybrid Nanorod - Polymer Solar Cells. Science, 295, , March 29, Shaheen, S.E., Brabec, C.J., Sariciftci, N.S., Padinger, F., Fromherz, T., Hummelen, J.C. 2.5% efficient organic plastic solar cells. Applied Physics Letters, 78, no. 6, , February 5,