10:15 AM - H1.3
High Open Circuit 1.1 eV Dilute Nitride III-V Quantum Well Solar Cell.
Aristotelis Fotkatzikis 1 2 , Andenet Alemu 1 , Lekhnath Bhusal 1 2 , Alex Freundlich 1 2 3
1 Photovoltaics and Nanostructures Laboratories, Center for Advanced Materials, University of Houston, Houston, Texas, United States, 2 Physics Department, University of Houston, Houston, Texas, United States, 3 Electrical and Computer Engineering Department , University of Houston, Houston, Texas, United States
Show AbstractIII-V dilute nitrides have attracted much interest over the past few years due to their captivating properties, which include a substantial reduction of the band gap and an increase of the electron effective mass upon incorporation of small amounts of N. These properties make this family of materials very attractive for a wide range of optoelectronic applications, including high-efficiency solar cells. The use of ~1.0 to 1.25 eV GaInNAs subcells, lattice matched to GaAs and Ge, has been suggested to enhance the efficiency of existing triple and quadruple junction solar cells. However unfavorable effects resulting from the addition of N in III-V semiconductors, such as poor minority carrier properties have thus far limited the success of this approach.(1)The use of III-V dilute nitride multi quantum wells (MQWs)inserted in the i-region of conventional III-V solar cells is a promising method to increase their efficiency, while alleviating many of the problems associated with bulk III-V based solar cells. Indeed, the insertion of multi quantum wells within the intrinsic region of conventional GaAs p-i-n solar cells has been predicted to yield practical efficiencies exceeding 35% at AM0. 2 III-V-N MQWs should extend the photon absorption range while alleviating minority carrier issues encountered in thick bulk-like GaInNAs. An added advantage of the use of III-V dilute nitride quantum wells is the stronger absorption coefficient, a result of the increase of the electron effective mass. In this work we discuss the development of 1.1 eV GaAsN/GaAs MQW solar cells grown by Chemical Beam Epitaxy (CBE). Device structures were fabricated on n-type GaAs (001) substrates by radio frequency (rf) nitrogen plasma-assisted CBE. 10 to 20 period MQWs of not intentionally doped strained GaAs1-xNx/GaAs quantum wells were inserted in the i-region of conventional GaAs solar cell structure at relatively low temperatures (440C
10:30 AM - **H1.4
Photogeneration and Carrier Collection in Quantum Confined p-i-n Solar Cells.
Alex Freundlich 1 2
1 Center for Advanced Materials, University of Houston, Houston, Texas, United States, 2 Physics and ECE, University of Houston, Houston, Texas, United States
Show AbstractWith staggering efficiency projections, quantum confined solar cells have been suggested and investigated to overcome the fundamental efficiency limitation of single junction solar cells. Thus far most practical realizations have relied on the introduction of a periodic array of low dimensional narrow bandgap semiconductor (wells, wires, dots) within the intrinsic (i) region of a wider bandgap p-i-n diode. While the approach has been shown to successfully extend the spectral sensitivity of the device, one major practical shortcoming resides in the difficulty of extracting photo-generated carriers from the quantum-confined region, which leads to important recombination losses and marginal (if any) efficiency enhancement. After a brief review of past and recent developments in quantum confined p-i-n diodes, the presentation will focus on the present understanding of recombination and escape mechanisms of photo-generated carriers in these devices and will discuss the importance of quantum confined region design parameters (well depth, hole and electron band discontinuities, carrier effective masses and i- region width,…) in maintaining an efficient collection of the photogenerated carriers. Finally the presentation will evaluate few emerging strategies (single confined-carrier devices, intermediate state assisted carrier escape) toward significantly enhancing practical efficiencies in quantum confined p-i-n solar cells.
11:00 AM - H1: Inter Band
BREAK
H2: Inorganic Quantum Dot and Nanowire Solar Cells
Session Chairs
Monday PM, November 26, 2007
Republic B (Sheraton)
11:30 AM - **H2.1
Photoluminescent Spectra of Nanostructured InAs Quantum Dots Enveloped in A GaAsSb Matrix.
Michael Levy 1 , Stephan Bremner 1 , Christiana Honsberg 1
1 Electrical and Computer Engineering, University of Delaware, Newark, Delaware, United States
Show AbstractMaterial properties of nanostructured InAs quantum dots enveloped in a GaAsSb matrix are presented here. This material system is of interest for it is identified as an absorbing medium for a multi-transition solar cell, in principle, capable of achieving greater than 50% efficient photovoltaic solar energy conversion. The structural properties of the samples are investigated with X-ray diffractometry and atomic force microscopy. The optical properties of the samples are investigated with photoluminescent spectroscopy and absorptance measurements. The matrices that were grown include GaAs0.88Sb0.12 matrices. Experimental studies reveal radiative transitions resulting from two distinct photo-induced transitions. These transitions result from interactions with photons with energies greater than and less than the bandgap of the barrier matrix.
12:00 PM - H2.2
Controlled Formation of Nanocrystal Superlattices in Conjugated Polymer Composites.
Alexandros Stavrinadis 1 , Richard Beal 1 , Jason Smith 1 , Hazel Assender 1 , Andrew Watt 1
1 Department of Materials, University of Oxford, Oxford United Kingdom
Show AbstractWe report the formation of nanocrystal superlattices during the post-synthesis processing of PbS nanocrystals/ MEH-PPV composites. Composites were synthesized by a single-step surfactant free route yielding well dispersed nanocrystals with broad size distribution (1-10nm) as characterized by transmission electron microscopy (TEM). When the composites were precipitated from solution by the rapid injection of methanol cubic and triangular colloidal particles, ~200nm in size were formed. Despite apparently monocrystalline electron diffraction patterns bright and dark field imaging shows the particles consist of individual nanocrystals which have self assembled with a high degree of crystallographic alignment. Absorption spectroscopy is used to show that nanocrystals retain their quantum confined properties when assembled. Using a higher alcohol to induce colloid precipitation such as 1-hexanol results in a modification of the superlattice structures observed and improved dispersion properties.
12:15 PM - H2.3
Atomistic Theoretical Modeling of Auger Recombination in Si and Ge Nanocrystals.
Cem Sevik 1 , Ceyhun Bulutay 1
1 Physics and Institute of Materials Science and Nanotechnology, Bilkent University, Ankara Turkey
Show Abstract12:30 PM - H2.4
Photon Management for Solar Cells.
Andries Meijerink 1 , Peter Vergeer 1 , Linda Aarts 1 , Bryan van der Ende 1
1 Chemistry, Debye Institute, Utrecht Netherlands
Show AbstractThe largest energy loss in solar cells is related to the spectral mismatch between the absorption spectrum of the semiconductor material and the solar spectrum. For the most widely used c-Si solar cell, the main loss is related to the relaxation of high energy charge carriers created after absorption of relatively high energy photons (2-4 eV) to the 1 eV bandgap of c-Si. Alternatively, wider bandgap solar cells (like the Gratzel cell) experience a large energy loss due to the fact that the low energy part of the solar spectrum is not absorbed. In this contribution new strategies to reduce these losses will be presented based on photon management using the well-defined energy levels of lanthanides. The underlying mechanisms for the spectral conversion involve up- and downconversion processes where one higher energy photon is split into two lower energy photons (downconversion) or two lower energy photons are added up to one higher energy photon (upconversion). Recently we have shown that it is possible with downconversion to split high energy vacuum ultraviolet photons into two visible photons with a190% efficiency [1]. Here it will be shown that both resonant energy transfer and cooperative sensitization are promising for downconversion processes resulting in a conversion efficiency from visible to infrared close to 200%. For the couple Tb-Yb efficient downconversion from green to infrared is observed in Y1-xYbxPO4:Tb1%. Upon excitation in the 5D4 level of Tb3+ energy transfer to two neighboring Yb3+ ions occurs. The energy transfer mechanism is shown to be cooperative energy transfer via dipole-dipole interaction. In the fully concentrated material (x=1) the theoretical limit is 185% conversion efficiency. The actual quantum efficiency of the Yb3+ emission is significantly reduced by concentration quenching. Downconversion by resonant energy transfer is studied for a number of couples of lanthanide ions: Er-Yb, Ho-Yb, Pr-Yb and Tm-Yb. In a number of oxide and fluoride matrices efficient two-step energy transfer is observed resulting in the 1000 nm emission from Yb3+ after excitation in the 300-500 nm energy levels of the other lanthanide ion. Again, concentration quenching due to energy migration over the Yb-sublattice limits the output of the IR radiation by Yb3+. The quenching can be reduced by improving the synthesis procedure (less quenching centers) or by creating small nanoclusters of lanthanide ions (no energy migration).[1] R.T. Wegh, H. Donker, K.D. Oskam and A. Meijerink, Science 283 (1999) 593.
12:45 PM - H2.5
Large Area (> 1 cm^2) Arrays of Vertically-aligned Si Wires for Photovoltaic Device Applications.
Brendan Kayes 1 , Michael Filler 1 , Morgan Putnam 1 , Michael Kelzenberg 1 , Nathan Lewis 1 , Harry Atwater 1
1 , California Institute of Technology, Pasadena, California, United States
Show AbstractPhotovoltaic devices designed to achieve high cell efficiency in low quality materials must enable collection of low diffusion length charge carriers in optically thick absorber layers. An attractive approach to this challenge is a solar cell designed as a large array of vertically-aligned semiconducting wires with radial pn junctions that facilitate carrier collection over distances that are short relative to the optical thickness of the material [1]. Such designs require the achievement of large area arrays of vertically-oriented semiconductor wires. We demonstrate here the growth, by the vapor-liquid-solid (VLS) growth technique, of large arrays of silicon wires with diameters of 1.5 um and lengths of up to 75 um, with very low defect densities over areas greater than 1 cm^2, by exploiting photolithography to pattern arrays of selectively deposited Au catalyst at sites in hole arrays formed in a silicon oxide thin film.A key conclusion of our previous device design work is that the wires in the array should optimally have a radius approximately equal to the minority-carrier diffusion length in the material. This would provide an optimal tradeoff between increased current collection, and the loss of open-circuit voltage due to the increased junction area and surface area that is produced when the wire radius is decreased. Calculations based upon the equilibrium solubility of gold in silicon and the inferred mid-gap trap density, as well as near-field scanning optical microscope measurements of minority carrier diffusion length in actual wires grown with a Au VLS catalyst [2], indicate that the minority carrier diffusion length is 1 - 5 um for such Si wires. Vertically-oriented wire arrays were fabricated by patterning photoresist with holes 3 um in diameter and with a 7 um pitch, on Si (111) wafers with a 290 nm oxide coating. The oxide was then etched through the pattern. 500 nm of gold was evaporated, followed by liftoff, leaving gold only in the holes in the oxide. Silicon wires were then grown in a tube furnace at atmospheric pressure at 1000 oC, by the catalytic decomposition of SiCl4 at the surface of the gold particles, in a H2 ambient (via the VLS mechanism). Wires were vertically oriented from the substrate, and had diameters of 1.5 um and lengths up to 75 um. The pattern fidelity was maintained over areas greater than 1 cm^2. Transmission electron microscopy (TEM) indicated that these wires were single-crystalline and grew along the <111> direction. Comparison of wires grown on samples with and without an oxide layer revealed the importance of the oxide in maintaining pattern fidelity, by preventing catalyst diffusion and agglomeration that compete with wire growth in the initial stages of wire nucleation and growth. [1] B. M. Kayes, H. A. Atwater, and N. S. Lewis, J. Appl. Phys., 97, 114302 (2005) [2] M. D. Kelzenberg, unpublished results
H3: Multiexciton Generation
Session Chairs
Monday PM, November 26, 2007
Republic B (Sheraton)
2:30 PM - **H3.1
Multiple Exciton Generation in Semiconductor Quantum Dots: Applications to Third Generation Solar Photon Conversion.
Arthur Nozik 1 2 , Matt Beard 1 , Randy Ellingson 1 , Joseph Luther 1 , Justin Johnson 1 , Qing Song 1 , Pingrong Yu 1 , Kathrine Gerth 1 , Matt Law 1
1 , NREL, Golden, Colorado, United States, 2 Department of Chemistry, University of Colorado, Boulder, Colorado, United States
Show AbstractIn order to utilize solar power for the production of electricity and fuel on a massive scale, it will be necessary to develop solar photon conversion systems that have an appropriate combination of high efficiency (delivered watts/m2) and low capital cost ($/m2). One potential, long-term approach to high efficiency is to utilize the unique properties of quantum dot nanostructures to control the relaxation dynamics of photogenerated carriers to produce either enhanced photocurrent through efficient photogenerated electron-hole pair (exciton) multiplication or enhanced photopotential through hot electron transport and transfer processes. To achieve these desirable effects it is necessary to understand and control the dynamics of hot electron and hole relaxation, cooling, charge transport, and interfacial charge transfer of the photogenerated carriers with femtosecond (fs) to ns time resolution. We have been studying these fundamental dynamics in various bulk and nanoscale semiconductors (quantum dots (QDs), and quantum wells) using fs transient, photoluminescence, and THz spectroscopy. We have observed very efficient and ultrafast multiple exciton generation (MEG) from absorbed single high energy photons in Group IV-VI and in Si QDs; the former system yields 3 excitons/photon at photon energies 3 times the QD bandgap. Such efficient MEG has the potential to greatly enhance the conversion efficiency of solar cells that incorporate QDs for both solar electricity and solar fuel (i.e. H2) production. Selected aspects of this work will be summarized and recent advances will be discussed, including work to measure MEG in the photocurrent. Various possible configurations for quantum dot solar cells that could produce ultrahigh conversion efficiencies for the production of electricity and solar fuels will be presented, along with progress in developing such new types of solar cells.
3:00 PM - H3.2
ZnO Nanorods Solar Cells Sensitized with PbS Quantum Dots.
Hung Liao 1 , Chin Lin 1 , San Chen 1
1 Department of Materials Science and Engineering, National Chiao Tung University, Hsinchu Taiwan
Show Abstract3:15 PM - H3.3
Space-separated Multiexciton Generation in Si Nanocrystals.
Dolf Timmerman 1 , Ignacio Izeddin 1 , Tom Gregorkiewicz 1
1 Van der Waals-Zeeman Institute, University of Amsterdam, Amsterdam Netherlands
Show AbstractIn the past, the phenomenon of multiple exciton generation has been reported for nanocrystals (NCs) of II-VI materials. In this case, formation of up to 7 excitons in a nanocrystal upon absorption of a single photon of sufficiently high energy has been shown. The threshold energy for this process is usually found to be equal or close to hν ≥ 3 EG, where EG denotes the HOMO-LUMO optical bandgap of the NC. In this contribution, we report on investigations of multiple exciton generation in Si nanocrystals. The study is performed on solid dispersions of small Si-NCs (diameter 2.7 – 3 nm) embedded in a SiO2 matrix. In the experiment, we follow efficiency of the exciton-related photoluminescence band at ~900 nm as a function of excitation wavelength. We then compare it with the separately measured absorbance. We conclude that when quantum energy of the incoming photons exceeds twice the bandgap of Si-NCs, two excitons are generated per one absorbed photon. Moreover, in contrast to the earlier reported multiple exciton generation in PbSe and CdSe, the two excitons are localized in two neighboring nanocrystals. The microscopic aspects and physical mechanisms of this process are discussed, and similarity with the earlier investigated energy transfer between Si-NCs and Er3+ ions [1] are discussed.[1] I. Izeddin et al., Phys. Rev. Lett. 97, 2007401:1-4 (2006).
3:30 PM - H3.4
Multiple Exciton Generation and Extraction from Hydrazine Treated PbSe Thin Film Device.
Sung Jin Kim 1 , Won Jin Kim 3 , Yudhisthira Sahoo 3 , Alexander Cartwright 1 3 , Paras Prasad 2 3
1 Department of Electrical Engineering, University at Buffalo, the State University of New York, Amherst, New York, United States, 3 Institution of Lasers, Photonics and Biophotonics, University at Buffalo, the State University of New York, Amherst, New York, United States, 2 Department of Chemistry, University at Buffalo, the State University of New York, Amherst, New York, United States
Show AbstractWe report Multiple Exciton Generation (MEG) and electrical extraction from a photoactive device using a thin film of PbSe nanoparticles with an excitonic absorption at 0.7 eV. The PbSe nanocrystal film was spin-cast onto an ITO coated glass substrate. This layer was subsequently dipped in a 1M solution of hydrazine in acetonitrile to remove the resident oleic acid ligand. The resulting hydrazine treated film was smooth and a mirror-like surface profile was observed. After the hydrazine treatment, an aluminum electrode was deposited on the top side of the device. To quantify the effects of MEG and extraction (Electrons collected/Photon absorbed), a wavelength tunable laser setup (Coherent Verdi 18, Mira 900, Rega 9000 and OPA9400) was used to scan the incident photon energy from 1.55eV to 3.1eV. The electrically measured external quantum efficiency was similar to the reported optically measured MEG results. Specifically, multiple electrons were extracted for high energy photons with energy above 2.8 times the PbSe nanocrystal excitonic absorption. At 4.35 times Eg(PbSe), 200% more electrons were extracted.
3:45 PM - H3.5
Charge Transport in Arrays of Lead Selenide Nanocrystals.
Tamar Mentzel 1 , Scott Geyer 2 , Venda Porter 2 , Kenneth MacLean 1 , Moungi Bawendi 2 , Marc Kastner 1
1 Physics, MIT, Cambridge, Massachusetts, United States, 2 Chemistry, MIT, Cambridge, Massachusetts, United States
Show AbstractH4: Organic Solar Cells
Session Chairs
Monday PM, November 26, 2007
Republic B (Sheraton)
4:30 PM - **H4.1
Attainable Efficiencies in Organic Solar Cells.
Sean Shaheen 1 , Nikos Kopidakis 2 , Matthew Reese 2 , William Rance 3 , Jao van de Lagemaat 2 , David Ginley 2 , Garry Rumbles 2 , Brian Gregg 2 , Arthur Nozik 2
1 Dept. of Physics, University of Denver, Denver, Colorado, United States, 2 , National Renewable Energy Laboratory, Golden, Colorado, United States, 3 Physics, Colorado School of Mines, Golden, Colorado, United States
Show AbstractOrganic and nanostructured/hybrid solar cells rely on dissociation of an exciton at a donor-acceptor interface for charge carrier generation. In nearly all reported devices to date, there is a substantial (~ 1 eV) energy loss between the energy of the absorbed photon and the energy difference between the final separated electron and hole (which is considered to be the effective electronic band gap of the donor-acceptor couple). This talk begins by examining the role of the exciton binding energy, if any, in determining this energy loss and ultimately in determining the theoretically attainable efficiency. Practical issues relating to efficient carrier extraction are then addressed, with consideration of the competing mechanisms of carrier transport to the electrodes versus carrier recombination in the bulk. In particular, the size distribution and morphology of the donor and acceptor domains is shown to be critical in determining what fraction of the photogenerated carriers ultimately exit the device. Finally, incorporation of 3rd generation mechanisms, such as multiple junctions and frequency up-conversion, into organic device structures is discussed. Even if incorporation of these does not lead to efficiencies above the Shockley-Queisser limit, their incorporation may make sense economically, as is demonstrated by multijunction amorphous silicon devices. Here we briefly discuss practical issues as to their implementation and what sort of efficiency gains one might reasonably expect.
5:00 PM - **H4.2
``Solvent Annealing" Effect in Plastic Solar Cells.
Yang Yang 1 3 , Gang Li 1 3 , Yan Yao 1 , Hoichang Yang 2 , Vishal Shrotriya 3 , Guanwen Yang 1
1 Materials Sci. & Eng., UCLA, Los Angeles, California, United States, 3 , Solarmer Energy Inc., El Monte, California, United States, 2 Rensselaer Nanotechnology Center, Rensselaer Polytechnic Institute, Troy, New York, United States
Show Abstract5:30 PM - **H4.3
Additives Provide Kinetic Control of Morphology in Bulk Heterojunction Materials Processed via Spin Coating.
Gui Bazan 1 , Jeff Peet 1 , Jin Young Kim 1 , Daniel Moses 1 , Alan Heeger 1 , Thuc-Quyen Nguyen 2
1 Materials, University of California-Santa Barabara, Santa Barbara, California, United States, 2 Chemistry, University of California-Santa Barabara, Santa Barbara, California, United States
Show AbstractHigh charge separation efficiency combined with the reduced fabrication costs associated with solution processing (printing and coating) and the potential for implementation on flexible substrates make “plastic” solar cells a compelling option for tomorrow’s photovoltaics. The control the donor/acceptor morphology in bulk heterojunction materials as required for achieving high power conversion efficiency have is therefore of primary concern. We recently showed that by incorporating a few volume percent of alkanedithiols in the solution used to spin cast films comprising a low bandgap polymer and a fullerene derivative, the power conversion efficiency of photovoltaic cells (AM 1.5 conditions) is increased from 2.8% to 5.5% through altering the bulk heterojunction morphology. This approach provides an operationally simple and versatile tool available for the tailoring of the heterojunction solar cell morphology in systems where thermal annealing is not feasible and works even on a system in which long range polymer crystallinity is not observed. Increases in the hole mobility were measured by using FET devices. Modifiers other than alkanedithols work well, although the mechanism by which the heterojunction morphology is affected remains poorly understood. The effect of processing on charge mobility, photoresponse and other conjugated polymer structures will also be discussed.
Symposium Organizers
Mike McGehee Stanford University
David Ginger University of Washington
Christiana Honsberg University of Delaware
Jenny Nelson Imperial College London
Jiangeng Xue University of Florida
H5: Organic Bulk Heterojunctions
Session Chairs
Tuesday AM, November 27, 2007
Republic B (Sheraton)
9:30 AM - **H5.1
Correlation of Solar Cell Performance with Nanoscale Hole and Electron Transport in Bulk Heterojunction Conjugated Polymer/Fullerene Blends.
ThucQuyen Nguyen 1 , Mark Dante 1 , Jeffrey Peet 1
1 Chemistry and Biochemistry, University of California, Santa Barbara, Santa Barbara, California, United States
Show AbstractIn the past few years, conjugated polymer (CP) based photovoltaic devices have attracted a great deal of attention. In this work, we applied the various scanning probe techniques to characterize conjugated polymer/fullerene blend materials typically used to fabricate polymer solar cells. Atomic force microscopy (AFM) was used to characterize the film morphology whereas conducting AFM (C-AFM) was used to study the charge transport properties at the nanoscale. C-AFM in two different measurement modes was used to study the nanoscale charge transport of films containing poly(3-hexylthiophene)/[6,6]-phenyl C61-butyric acid methyl ester blend (P3HT:PCBM) and poly[2,6-(4,4-bis-(2-ethylhexyl)-4H-cyclopenta[2,1-b;3,4-b']dithiophene)-alt-4,7-(2,1,3 benzothiadiazole)](low bandgap polymer)/[6,6]-phenyl-C71butyric acid methyl ester (PCPDTBT:PC70BM). In C-AFM, the local current map at a constant bias and the surface topography are obtained simultaneously. Alternatively, local current-voltage curves can be measured from which the mobilities are calculated. Hole current images and mobilities were obtained using Pt-coated silicon probes whereas Mg-coated silicon probes were used to measure nanoscale electron mobilities. Variations in electrical properties and surface topography were examined. Nanoscale hole and electron mobilities increase upon annealing for P3HT:PCBM blends, consistent with increased internal order. A direct correlation between the C-AFM determined nanoscale properties and device efficiencies is observed.
10:00 AM - H5.2
Quantitative Nanoscale Composition and Efficiency Maps for Organic Solar Cells.
Benjamin Watts 1 , Daniel Queen 2 , A.L. David Kilcoyne 3 , Harald Ade 1 , Frances Hellman 2
1 Physics, North Carolina State University, Berkeley, California, United States, 2 ALS, Lawrence Berkeley National Laboratory, Berkeley, California, United States, 3 Physics, University of California, Berkeley, California, United States
Show AbstractSolar cells based on thin blend films of conjugated polymers and/or fullerene derivatives are a promising alternative to the currently available silicon-based solar cells. However, these systems tend to display complex segregation of the organic components during film formation, with the degree of segregation observed to correlate with device efficiency and shown to depend on fabrication parameters such as spin-casting spin-speed, solvent type and annealing conditions. However, the fundamental details of the relationship between the morphology of conjugated polymer blend films and their performance in photovoltaic devices is still not clear and techniques to quantitatively map these parameters with nanoscale resolution are required to enable further progress in this field.Recent work at the ALS in Berkeley, California, has utilized Scanning Transmission soft X-ray Microscopy (STXM) to produce quantitative, nanoscale composition maps of segregated conjugated polymer blend films [1], providing one half of the nanoscale composition-efficiency puzzle. In STXM, a thin sample is rastered across a focused soft X-ray beam at a number of photon energies to produce transmission images at resolutions down to 35 nm. By choosing photon energies corresponding to the molecular-structure dependent Near Edge X-ray Absorption Fine Structure (NEXAFS) resonances, chemical composition maps can be calculated via comparison of the transmission images to known NEXAFS spectra of the component chemical species. Because the NEXAFS resonances depend heavily on the types of bonding present, strong contrast can be achieved between materials with highly similar elemental and isotopic compositions, such as polymers.Now, we have developed a Soft X ray Beam Induced Current (SoXBIC) technique, in which a STXM instrument is extended to image a fully fabricated polymer solar cell device in a novel manner. While the transmitted light is measured and the device composition maps calculated as in normal STXM operation, the absorbed soft X-ray photons create charge carriers in the pixel-volume of the active polymer layer. Therefore, measuring the short-circuit current between the device electrodes, provides a second, current signal to complement the transmission (composition) signal for each pixel in the measured image. Hence, both the local composition and efficiency of a polymer solar cell device are simultaneously mapped at nanoscale resolution approaching the exciton diffusion length.We present details of polymer solar cell device fabrication on X-ray transparent substrates, the SoXBIC technique and new quantitative, nanoscale composition and efficiency maps of solar cell devices based on segregated PFB:F8BT blend films.This work was supported by the U. S. Department of Energy under Contract Nos. DE-FG02-98ER45737 and DE-AC02-05CH11231.[1] C.R. McNeill, B. Watts, L. Thomsen, H. Ade, N.C. Greenham, and P.C. Dastoor Macromolecules, 40, 3263–3270 (2007) DOI: 10.1021/ma070132d
10:15 AM - H5.3
Understanding P3HT Crystallite Orientation and Morphology in P3HT:PCBM Blends.
Alex Mayer 1 , Michael Toney 2 , Shawn Scully 1 , Michael McGehee 1
1 Materials Science and Engineering, Stanford University, Stanford, California, United States, 2 , Stanford Synchrotron Radiation Laboratory, Menlo Park, California, United States
Show AbstractOrganic bulk heterojunction solar cells based on poly(3-hexylthiophene) (P3HT) and phenyl-c61-butyric acid methyl ester (PCBM) have demonstrated a drastic increase in their power conversion efficiency as researchers gain control of the nanostructured morphology. In order to improve the efficiency further, a detailed understanding of the crystallographic orientation and morphology is required. Recently, researchers have found that decreasing the evaporation rate of the solvent in P3HT:PCBM solar cells greatly enhances power conversion efficiency of the device due to an increase in the hole mobility in the polymer phase. The mechanisms behind the mobility increase remain a puzzle. We employ 2D grazing incidence x-ray scattering (2D GIXS) and rocking curves of films spun from a slow and a fast drying solvent. We show for the first time that the orientation of the P3HT domains is highly dependent on the evaporation rate of the solvent and that the crystallographic orientation increases through the use of a slowly drying solvent. Interestingly, we find that the polymers preferentially orient with their insulating alkyl chains, normally thought to be the direction of slow transport, along the direction of charge transport. We rationalize this observation by noting that a percolating charge encounters many grain boundaries while traversing the cell that affect the charge to a varying degree. In a randomly oriented sample, the angle between adjacent grains can be higher than that found in an oriented sample, leading to a larger hopping barrier and thus a decreased mobility. Another puzzle that remains is that the P3HT photoluminescence is quenched in blends spun from both slow and fast drying solvents. But the internal quantum efficiency is ~100% for the quickly dried blends and <80% for the slowly dried blends even at reverse bias. This is a huge loss mechanism that needs to be explained if further optimization is possible. In order to explain this phenomenon, small angle x-ray scattering, surface x-ray scattering, and device modeling are employed.
10:30 AM - H5.4
Structure and Diffusion in P3HT:PCBM Blends.
Benjamin Watts 2 , Lars Thomsen 1 , Warwick Belcher 1 , Harald Ade 2 , Paul Dastoor 1
2 Physics, North Carolina State University, Raleigh, North Carolina, United States, 1 Physics, University of Newcastle, Callaghan, New South Wales, Australia
Show AbstractNear-edge X-ray absorption fine structure spectroscopy (NEXAFS) and, its 2D mapping counterpart, scanning transmission X-ray microscopy (STXM) are extremely powerful; techniques for probing the structure, alignment and morphology of thin organic films on the nanometer scale [1]. Previous work by us has shown how these techniques can be applied to reveal insights into the morphology and property relationships for the typical semi-conducting polymer blends commonly used in organic electronic applications [2]. Most recently we have been able to use X-ray absorption techniques to probe the alignment and dynamic properties of polymer blends used in organic photovoltaic devices. Here we present our latest results on the structure and diffusion of P3HT:PCBM blends with a focus on their diffusion and phase segregation behaviour.References:1. Kilcoyne, A. L. D.; Tyliszczak, T.; Steele, W. F.; S., F.; Hitchcock, P.; Franck, K.; Anderson, E. H.; Harteneck, B. D.; Rightor, E. G.; Mitchell, G. E.; Hitchcock, A. P.; Yang, L.; Warwick, T.; Ade, H. J. Synchrotron Rad. 2003, 10, 125-136.2. McNeill, C. R.; Watts, B.; Thomsen, L.; Belcher, W. J.; Greenham, N. C.; Dastoor, P. C., Nano Lett. 2006, 6, 1202-1206.
10:45 AM - H5.5
Investigation of Loss Mechanisms in Polymer Blend Photovoltaics via Simultaneous Probing of Charge Separation and Charge Transport.
Astrid Gonzalez-Rabade 1 , Arne Morteani 1 , Richard Friend 1
1 Cavendish Laboratory, Cambridge University, Cambridge United Kingdom
Show AbstractHeterojunctions between organic semiconductors allow efficient operation of photovoltaic diodes. We investigate, via electromodulation spectroscopy, the photoluminescence and the photovoltaic behavior of electron- and hole-transporting polyfluorene derivatives at a broad range of dopant concentrations and at various electric fields. A strong inverse correlation between the exciplex emission and the photocurrent efficiency demonstrates that they both have the same origin and that exciplex formation is the main loss mechanism for charge generation. In addition, we find a quantitatively one-to-one correspondence between the external field induced reduction in exciplex formation (detected as quenching of exciplex luminescence) and the increase in current collected at the external circuit. Interstingly, eventhough the charge collection efficiency and the exciplex quenching are morphology and electric field-dependent, the ratio between these two quantities is constant (and equal to 1) for all morphologies. These findings show, not only that non-geminate recombination can be neglected for the charge collection efficiency, but also that the one-to-one relation is universal amongst substantially dissimilar morphologies, implying that morphology is not relevant for charge collection. Our results demonstrate that the losses during the geminate pair dissociation are morphology-dependent, while, strikingly, the losses during charge collection are not morphology-dependent and are therefore negligible.
11:00 AM - H5:Organic BHJ
BREAK
H6: Organic Solar Cell Device Physics I
Session Chairs
Tuesday PM, November 27, 2007
Republic B (Sheraton)
11:30 AM - **H6.1
Bimolecular Recombination in Polymer / Fullerene Solar Cells.
James Durrant 1
1 , Imperial College London, London United Kingdom
Show AbstractPolymer / fullerene blend films are attracting extensive interest for photovoltaic energy conversion. Several groups have now reported organic solar cells based on a blend of poly(3-hexylthiophene) (P3HT) and methanofullerene [6,6]-phenyl C61-butyric acid methyl ester (PCBM) with power conversion efficiencies of over 4%. The blending of the P3HT and PCBM species on the nanometer scale (the ‘bulk heterojunction’) is essential to enable efficient exciton dissociation. However a further consequence of this blending is that the charge carriers generated by exciton dissociation are not physically well separated, raising the possibility for significant interfacial bimolecular recombination of these carriers. Such recombination losses would compete with charge transport and collection at the device electrodes.In this study, we employ a range of methodologies to determine the charge carrier densities and bimolecular recombination rate constants for P3HT/PCBM films and solar cells, focusing on devices operating at open circuit, the conditions where bimolecular recombination losses are most likely to be important. Three different methodologies are employed to determine the charge carrier densities: photocurrent transients, charge extraction and transient absorbance. Three different methodologies are employed to determine recombination rate constants: transient photovoltage decays and transient absorption decays of both devices and thin films. Excellent agreement is found between all methodologies. The recombination rate constant is found to be strongly carrier density dependent, attributed to the effects of charge trapping in the film. From these measurements we determine the bimolecular recombination flux in devices under operation, and conclude that bimolecular recombination is indeed a key loss pathway limiting device performance.
12:00 PM - H6.2
Free Carrier Generation and Recombination in P3HT/PCBM Bulk Heterojunctions Studied by Time-Resolved Microwave Conductivity.
Nikos Kopidakis 1 , Andrew Ferguson 1 , Sean Shaheen 1 , Garry Rumbles 1
1 , National Renewable Energy Lab, Golden, Colorado, United States
Show AbstractPoly(3-hexylthiophene) (P3HT) is the prototypical polymer used as light absorber and exciton and hole transporter in efficient bulk-heterojunction organic photovoltaic devices. The study of photophysics of P3HT is facilitated by its rather unique property that photoexcitation produces free carriers without the presence of an acceptor, albeit with low yield. Extensive previous studied of TRMC in pristine P3HT has produced a wealth of experimental data and understanding of exciton and free carrier generation in this material. In the work presented here, we build upon these previous studies and extend them to include photoconductivity measurements in bulk heterojunctions of P3HT blended with varying amounts (1%-80% b.w.) of the acceptor fullerene derivative [6,6]-phenyl C61-butyric acid methyl ester (PCBM). We demonstrate how previous models for free carrier generation in pristine P3HT need to be modified to account for the presence of the acceptor and the corresponding high yield for free carrier generation by exciton dissociation at the donor-acceptor interface. We develop a kinetic scheme to describe free carrier generation and recombination that is consistent with our experimental data for both pristine P3HT and P3HT/PCBM blends. In the framework of this scheme, we discuss higher order processes that limit exciton generation in P3HT/PCBM bulk heterojunctions and show how these processes also limit free carrier generation. Implications of our findings on the operation and optimization of organic photovoltaic devices based on P3HT/PCBM bulk heterojunctions are discussed.
12:15 PM - H6.3
Monte Carlo Investigation of Bimolecular Recombination, Geminate Recombination and Space-charge Effects in Nanostructured Polymer Photovoltaic Devices.
Chris Groves 1 , R. Marsh 1 , Neil Greenham 1
1 Optoelectronics, Cambridge University, Cambridge, Cambridgeshire, United Kingdom
Show AbstractThe performance of organic photovoltaic devices (OPVs) is limited by a number of processes relating to the transport of charge carriers. Geminate pairs must first avoid mutual recombination, and thereafter avoid bimolecular recombination with other free carriers en-route to the collecting contacts. Furthermore, at high intensities, differences in carrier mobilities may lead to the accumulation of the least mobile carrier, giving space-charge limited photocurrents. Here we use a Monte Carlo model to examine these limiting factors and find their relative importance. The Monte Carlo model simulates the transport of polarons through a Cartesian geometry of electron- and hole-conducting sites, each with an associated Gaussian disorder. Electrostatic attractions between carriers and image charges at the electrodes are also included.In our simulations we keep one carrier mobility constant, and vary the other carrier mobility by two orders of magnitude in either direction. We find that the degree of geminate recombination at a given field is dependent largely upon the mobility of the most mobile carrier type, as might be expected. Bimolecular recombination increases with intensity and is of the order ~30% for some device configurations at the highest intensities considered. It is found that, over the operating voltage range of an OPV device, the degree of bimolecular recombination increases with increasing mobility. We ascribe this result to carriers diffusing further and so sampling more of the device, which in turn leads to increased opportunity for recombination. For the devices and conditions examined we do not find that the degree of geminate recombination falls with increasing intensity, suggesting the current is not space-charge limited.
12:30 PM - **H6.4
Dual Electron Donor/Electron Acceptor Character of a Conjugated Polymer in Efficient Photovoltaic Devices.
Chris McNeill 1 , Agnese Abrusci 1 , Jana Zaumseil 1 , Richard Wilson 2 , Mary McKiernan 2 , Jonathan Halls 2 , Neil Greenham 1 , Richard Friend 1
1 Cavendish Laboratory, University of Cambridge, Cambridge United Kingdom, 2 , Cambridge Display Technology Ltd., Cambridge United Kingdom
Show AbstractWe report efficient photovoltaic operation of a polymer blend consisting of poly(3-hexylthiophene) (P3HT) and the red polyfluorene poly((9,9-dioctylfluorene)-2,7-diyl-alt-[4,7-bis(3-hexylthien-5-yl)-2,1,3-benzothiadiazole]-2’,2’’-diyl) (F8TBT). Power conversion efficiencies of nearly 2% and external quantum efficiencies of over 25% are demonstrated, representing some of the highest values for all-polymer blends. These results are also significant as they demonstrate that F8TBT, previously used in efficient solar cells as an electron donor and hole transporter, can also be employed effectively as an electron acceptor and electron transporter. This dual nature of F8TBT is confirmed by the operation of F8TBT-based light-emitting ambipolar field-effect transistors. The dual electron donor/electron acceptor operation of F8TBT in photovoltaic devices, however, demonstrates that the ambipolar character of conjugated polymers is also valid in low charge density environments. Finally, the observation that efficient photovoltaic operation can result when P3HT is used in combination with F8TBT instead of the often-used (6,6)-phenyl C61-butyric acid methyl ester (PCBM), which has an electron mobility several orders of magnitude higher than F8TBT, questions the assumption that charge carrier mobility is limiting the development of organic photovoltaic devices. We present evidence that the separation of geminate electron-hole pairs is limiting the performance of this class of device rather than recombination resulting from poor charge transport.
H7: Organic Solar Cell Device Physics II
Session Chairs
Tuesday PM, November 27, 2007
Republic B (Sheraton)
2:30 PM - **H7.1
Charge Transport and Photogeneration in All-polymer Solar Cells.
Magda Mandoc 1 , Welmoed Veurman 1 , Jan Anton Koster 1 , Bert de Boer 1 , Paul Blom 1
1 Zernike Institute for Advanced Materials, University of Groningen, Groningen Netherlands
Show AbstractThe photogeneration mechanism in blends of poly[2-methoxy-5-(3', 7'-dimethyloctyloxy)-1,4-phenylene vinylene] (MDMO-PPV) and poly[oxa-1,4-phenylene-(1-cyano-1,2-vinylene)-(2-methoxy-5-(3',7'-dimethyloctyloxy)-1,4-phenylene)-1,2-(2-cyanovinylene)-1,4-phenylene] (PCNEPV) is investigated. The photocurrent in the MDMO-PPV:PCNEPV blends is strongly dependent on the applied voltage as a result of a low dissociation efficiency of the bound electron-hole pairs. The dissociation efficiency is limited by the low carrier mobilities, low dielectric constant and the strong intermixing of the polymers, leading to a low fill factor and a reduced photocurrent at operating conditions. Additionally, electrons trapped in the PCNEPV phase recombine with the mobile holes in MDMO-PPV phase at the interface between the two polymers, thereby affecting the open-circuit voltage and increasing the recombination losses. At an intensity of one sun, Langevin recombination of mobile carriers dominates over the trap-assisted recombination.
3:00 PM - H7.2
Defect Engineering in π-Conjugated Polymers.
Dong Wang 1 , Nikos Kopidakis 1 , Xin Ai 1 , Brian Gregg 1
1 , National Renewable National Lab, Golden, Colorado, United States
Show AbstractDefects, whether purposely added as dopants or not, usually control the electrical behavior of inorganic semiconductors. Defect engineering is thus a major field of research. In the newer field of organic semiconductors (OSCs), however, little is known about the number, influence or chemical identity of defect states. In OSCs with covalent disorder, deformations in a small fraction of π-bonds can have a profound influence on the electronic properties of the solid. We study the ostensibly pure π-conjugated polymers poly(3-hexylthiophene), P3HT, and poly(1-decyloxy-3-methoxy-2,5-phenylenevinylene), PPV. These commonly employed materials exhibit 9 - 12 orders of magnitude more charge carriers at equilibrium than they would if they were defect-free. Here we introduce chemical treatments, both of the polymers in solution and in solid films, that modify these electrical defects. The chemical reactions are chosen such that they can occur only at destabilized sites on the polymers. Treatment with nucleophiles such as hydride ion or methoxide ion leads to a decrease in film conductivity, an increase in both fluorescence intensity and lifetime, a greater exciton diffusion length, and Fermi level motion toward the middle of the bandgap. The addition of hydrogen or methanol across a strained double bond, converting unstable sp2 carbons into electroinactive sp3 carbons, is proposed as the mechanism. Treatment with electrophiles such as methyl cation, on the other hand, causes an increase in conductivity, Fermi level motion toward the valence band edge, and, again, an increase in exciton diffusion length. This is consistent with the addition of a methyl cation to the polymer backbone and concommittant elimination of an electron, leaving the counter anion unbound. The changes in optoelectronic properties of the π-conjugated polymers upon treatment are self-consistently explained by the chemical nature of the addition reactions.
3:15 PM - H7.3
Photoexcitations and Charge Generation Dynamics in P3HT/PCBM Blended Films Studied by Ultrafast Transient Absorption Spectroscopy.
Jorge Piris 1 , Thomas Eisenmayer 1 , Minh Trinh 1 , Juleon Schins 1 , Laurens Siebbeles 1
1 Optoelectronic Materials Section, Delft University of Technology, Delft Netherlands
Show AbstractWe studied the ultrafast processes following photo-excitation of thin films of the conjugated polymer poly(3-hexylthiophene) (P3HT) mixed with the electron acceptor [6,6]-phenyl-C61-butyric acid methyl ester (PCBM) by means of ultrafast transient absorption spectroscopy. We observed in the ps time domain exciton quenching and the subsequent formation of charge carriers as a function of the PCBM fraction in the films. We found that a relative large amount of charges are generated in the pure polymer. Addition of small quantities, ca. 2wt%, of the fullerene derivative reduces the fluorescence yield and lifetime by a factor 2. Nevertheless, it does not increase the initial yield for charge carrier generation significantly but results in longer-lived charge carriers. This is indication of the acceptor taking preferentially the place of the already existing charge separation centers in the pristine polymer. Increasing the PCBM content further increases the charge carrier yield but only up to double of that observed for the neat polymer at early times, reaching a maximum at ca. 20wt%. Higher acceptor concentrations result in reduced yield. In contrast, the most efficient solar cells contain ca. 50wt% of each component. This indicates that higher fullerene concentrations are necessary to transport the electrons than the optimum for charge generation.
3:30 PM - H7.4
Magnetic Studies of Singlet and Triplet Related Photovoltaic Processes in Organic Bulk Heterojunction Solar Cells.
Zhihua Xu 1 , Bin Hu 1
1 Materials Science and Engineering, University of Tennessee, Knoxville, Tennessee, United States
Show AbstractThe dynamic processes of photoexcitation generated excited states are critically important to the photovoltaic response in organic solar cells. Although the spin selection rule only allows the formation of singlet excitons under photoexcitation, both singlet and triplet excitons can coexist in organic materials due to the spin mixing from spin-orbital coupling or hyperfine interaction. When the different binding energies, lifetimes, and diffusion lengths are considered, the singlet and triplet excitons must have different contributions to the photovoltaic response. Therefore, understanding the singlet and triplet photovoltaic processes can influence the improvement of photovoltaic response in organic solar cells. In this report we will present our recent magnetic studies of singlet and triplet photovoltaic processes in organic bulk-heterojunction solar cells. Our experimental results of magnetic field-dependent photocurrent reveal that the singlet and triplet excitons undergo different photovoltaic processes, namely excito