Materials Gateway
Resource Center
Login button
 Open/CloseSend Us Your Feedback

Symposium N: Next-Generation and Nano-Architectured Photovoltaics

Symposium N: Next-Generation and Nano-Architectured Photovoltaics Image


November 30 - December 4, 2008

Chairs
Valeria G. Stoleru
(On leave from The University of Delaware)
979-450-8607


         Andrew G. Norman
National Renewable Energy Laboratory
1617 Cole Blvd.
Golden, CO 80401
303-384-6483
N. J. Ekins-Daukes
Experimental Solid State Physics
Imperial College London
Blackett Laboratory
London, SW7 2BZ United Kingdom
44-207-594-7579
        






Symposium Support
Army Research Office


Proceedings to be published online
(see Proceedings Library at www.mrs.org/publications_library)
as volume 1121E
of the Materials Research Society
Symposium Proceedings Series.


* Invited paper

TUTORIAL


N

Approaches to High-Efficiency Photovoltaics
Sunday, November 30, 2008
1:30 PM - 5:00 PM
Room 201 (Hynes)

The theoretical limit of solar energy conversion is over 85%, yet the maximum efficiency of any solar cell in the laboratory is less than half this value; commercial solar cells are only one-fifth the value. The challenge for meeting future global energy demands is to develop solar cells that achieve efficiencies approaching the thermodynamic limit. Multiple-junction solar cells, called tandem cells, theoretically allow the approaching of the thermodynamic limit with an “infinite” stack of homojunctions. However, materials-related issues have limited these devices to three-junction design. Recently, new physical mechanisms have been proposed which allow higher efficiency for a given number of materials, and also offer other advantages such as reduced sensitivity to temperature or use of nanomaterials. The tutorial will present approaches to producing ultrahigh efficiency solar cells, and discuss the experimental and theoretical challenges in making these cells. These routes include multijunction solar cells consisting of integrated pn junctions, multiple transition cells (quantum well or intermediate band quantum dot cells), and multiple exciton generation. In addition, physical insights into operation and design rules for advanced concept cells will be presented.

Instructor:
Christiana Honsberg
University of Delaware



SESSION N1: Multiple Exciton Generation and Electron Transport
Monday Morning, December 1, 2008
Republic B (Sheraton)

8:30 AM *N1.1
Multiple Exciton Generation in Colloidal Semiconductor Nanocrystals for Enhanced Solar Energy Conversion. Randy J. Ellingson1, Matt Law1, Joseph Luther1, Justin Johnson1, Qing Song1, Barbara Hughes1, Wyatt Metzger3, Aaron Midgett1,2, Octavi Semonin1,2, Sean Sweetnam1, Matt C Beard1 and Arthur J Nozik1,2; 1Center for Chemical and Biosciences, National Renewable Energy Laboratory, Golden, Colorado; 2Department of Chemistry, University of Colorado, Boulder, Colorado; 3National Center for Photovoltaics, National Renewable Energy Laboratory, Golden, Colorado.

Generation of multiple excitons per absorbed photon for colloidal nanocrystals (NCs) of PbS, PbSe, PbTe, CdSe, InAs, and Si provides an avenue for increasing the conversion efficiency of solar cells by increasing the photocurrent resulting from sunlight’s higher-energy photons. Our research addresses the measurement of multiple exciton generation (MEG) in NC solutions and films, and the development of NC-based solar cells which efficiently harvest multiple electrons per absorbed photon. MEG measurement techniques and results will be described, and recent advances in the performance of PbSe NC-based devices fabricated using a room temperature layer-by-layer technique will be discussed.


9:00 AM N1.2
Charge Carrier Multiplication and Nature of Excited States in PbSe Quantum Dots. Laurens Siebbeles, Delft University of Technology, Delft, Netherlands.

The exploitation of carrier multiplication (CM) in QDs receives a lot of interest, since this offers prospects for the development of solar cells with a maximum power conversion efficiency as high as 44%. In the CM process the excess energy of charge carriers produced by photons with energy greater than the band gap is utilized to produce additional charge carriers and is not lost by thermalization of the photo-excited electron. Efficient CM has been reported for several semiconductor QDs: PbSe, PbS, PbTe, CdSe, InAs and Si. Recently, some of these reports were challenged by studies claiming that CM does not occur in CdSe, CdTe, and InAs QDs, thus raising legitimate doubts concerning the occurrence of CM in the remaining materials. In the work to be presented conclusive evidence is given for occurrence of CM in PbSe QDs using femtosecond transient photobleaching.[1] It is shown that a correct determination of CM efficiency requires spectral integration over the photobleach feature. The CM efficiency we obtain is significantly lower than has been reported previously. The nature of the electronic transition giving rise to the second peak in the optical absorption spectrum of PbSe QDs has been a subject of intense debate. Theoretical and experimental claims have been provided for the assignment of this feature as the 1Pe1Ph as well as the 1Sh,e1Pe,h transitions. We studied the nature of this absorption feature by femtosecond pump-probe spectroscopy and present evidence that the optical transition involves neither 1Se nor 1Sh states.[2] This suggests that it is the 1Ph1Pe transition that gives rises to the second peak in the absorption spectrum of PbSe QDs. References [1] M.T. Trinh, A.J. Houtepen, Juleon M. Schins, T. Hanrath, J. Piris, W. Knulst, A.P.L.M. Goossens and Laurens D.A. Siebbeles, Nano Lett., 8, 1713 (2008). [2] M.T. Trinh, A.J. Houtepen, J.M. Schins, J. Piris and L.D.A. Siebbeles, Nano Lett., ASAP Article; DOI: 10.1021/nl8010963.


9:15 AM *N1.3
Abstract Withdrawn



9:45 AM N1.4
Sensing Charge Through Arrays of PbSe Nanocrystals with a Narrow MOSFET. Tamar Mentzel1, Kenneth MacLean1, Scott Geyer2, Moungi Bawendi2 and Marc A Kastner1; 1Physics, MIT, Cambridge, Massachusetts; 2Chemistry, MIT, Cambridge, Massachusetts.

Understanding electron transport through localized states is important for realizing novel approaches to solar cells. The efficiency of existing solar cells is limited in part by transport through localized states and proposed novel solar cell materials, such as arrays of semiconductor nanocrystals (NCs), are based on transport through localized states. Measurement of current through localized states gives information only about the states through which tunneling is fastest as that is what limits the current. To understand the full dynamics of transport through localized states, we measure the charge in addition to the current through the localized states of an array of PbSe NCs. We fabricate a narrow Si MOSFET charge sensor (~100 nm x ~60 nm ) in close proximity (~1 micron away) to a three dimensional array of PbSe nanocrystals. We simultaneously measure the noise current of the MOSFET caused by nearby charge fluctuations, and the current through the NC array, as a function of temperature and electric field.


SESSION N2: Multiple Exciton Generation and Nanocrystal Thin Film Devices
Monday Morning, December 1, 2008
Republic B (Sheraton)

10:30 AM *N2.1
Understanding Quantum Dots For Solar Cells. Alex Zunger, National Renewable Energy Laboratory, Golden, Colorado.

Advanced electronic structure theory is capable of providing some important clues as to the applicability of quantum dots to both Direct Carrier Multiplication ( DCM,sometimes also called MEG) and to Intermediate Band Solar Cells ( IBSC) .Here I will review and summarize recent results about the basic electronic structure of both free-standing colloidal dots and epitaxial,semiconductor-embedded "SK" dots .I will then use these insights to discuss the basic mechanisms of DCM in colloidal dots and IBSC in epitaxial dots .This approach can be "inverted" -- given what we need for DCM and IBSC ,look for the materials that are close to satisfying this.Early results on the Inverse Design problem will be discussed.This work was performed in collaboration with A.Franceschetti;G.Bester,V.Popescu and funded by DOE Office of Science Basic Energy Science .


11:00 AM N2.2
Schottky-Quantum Dot Photovoltaics for Efficient Infrared Power Conversion. Keith W Johnston1, Andras Geza Pattantyus-Abraham1, Jason P Clifford1, Stefan H Myrskog1, Dean D MacNeil1,2, Larissa Levina1 and Edward H Sargent1; 1Electrical and Computer Engineering, University of Toronto, Toronto, Ontario, Canada; 2Chemistry, Universite de Montreal, Montreal, Quebec, Canada.

Third generation photovoltaics require efficiencies beyond the Schockley-Queisser single-junction limit at low costs. This can be achieved by fabrication of multiple junctions to efficiently convert visible and infrared wavelengths separately. We report here a solution-processed quantum dot device that achieves very promising efficiency in the infrared regime, which complements existing progress in the visible regime. Planar Schottky photovoltaic devices were prepared from solution-processed PbS nanocrystal quantum dot films with aluminum and indium tin oxide contacts. These devices exhibited up to 4.2% infrared power conversion efficiency at 975 nm, which is a three-fold improvement over previous results. Solar power conversion efficiency reached 1.8%. The simple, optimized architecture allows for direct implementation in multijunction photovoltaic device configurations. The mechanism for the photovoltaic effect in these devices will be discussed, and a simple model of device operation based on drift and diffusion will be presented.


11:15 AM N2.3
Using a Nanometer Scale MOSFET as a Charge Sensor. Kenneth MacLean1, Tamar S Mentzel1, Scott Geyer2, Moungi Bawendi2 and Marc A Kastner1; 1Physics, MIT, Cambridge, Massachusetts; 2Chemistry, MIT, Cambridge, Massachusetts.

Single electron charge sensing has been demonstrated as a powerful measurement technique, and has been primarily used to probe the quantum mechanical behavior of electrons at low temperatures. However, single electron charge sensing is not limited to the low temperature regime, so this technique may be used to investigate electronic behavior in a variety of technologically important materials, such as arrays of PbSe nanocrystals, which have been proposed as a novel photovoltaic material. We fabricate narrow Si MOSFET sensors designed to measure charge fluctuations in a material that is deposited near the channel of the device. We discuss our sensor design, sensitivity, and background charge fluctuations, and use it to study charge hopping in an array of PbSe nanocrystals.


11:30 AM N2.4
Fourfold Efficiency Improvement in PbS Quantum Dot Photovoltaic Devices via Trap State Passivation by Ethanethiol. Aaron Barkhouse, Andras G Pattantyus-Abraham and Edward H Sargent; Electrical and Computer Engineering, University of Toronto, Toronto, Ontario, Canada.

Ethanedithiol treatments on lead chalcogenide quantum dot films have been previously used to produce efficient photovoltaic devices through reduction in inter-dot spacing and increased crosslinking of dots. We report here that ethanethiol (EtSH) treatment, in addition to improving the hole mobility in PbS QD PV devices, also passivates PbS trap states in non-optimally exchanged nanocrystals. This is inferred from an improved PLQE in EtSH-treated dots and an increased built-in voltage (Vbi) in planar Schottky PV devices, consistent with a reduction in interface trap state density. The best EtSH-treated device had a monochromatic external quantum efficiency and power conversion efficiency of 22% and 2.56%, respectively, under 76 mW/cm2 illumination at 975nm. We show that the effects of EtSH on mobility, carrier lifetime, Vbi and depletion width are insufficient to account for the more than 4-fold increase in external quantum efficiency. We therefore believe that the increased efficiency comes from passivating traps states that favour non-radiative exciton recombination.


11:45 AM N2.5
Abstract Withdrawn


SESSION N3: Carbon Nanotubes: Carrier Multiplication and Photovoltaics
Monday Afternoon, December 1, 2008
Republic B (Sheraton)

1:30 PM *N3.1
Optical Properties of Single-Walled Carbon Nanotubes. Tony F. Heinz, Depts. of Physics and Electrical Engineering, Columbia University, New York, New York.

Semiconducting carbon nanotubes constitute a prototypical nanoscale system, one with fully defined crystallographic structure and highly one-dimensional character. Single walled-semiconducting nanotubes exhibit strong optical transitions that span the visible and near-infrared spectral range. The specific transition energies reflect the crystallographic structure of the nanotube, as well as the local environment. In this paper we will describe the nature of the optical transitions in carbon nanotubes and the significant role that excitonic interactions play in the excited states of these materials. The strong Coulomb interaction between charges not only yields tightly bound excitons, but also gives rise to exciton-exciton annihilation processes. These many-body effects suggest that semiconducting carbon nanotubes may also be candidates for materials that undergo the reserve process, i.e., multiple-exciton generation after absorption of a high-energy photon. Measurements of the optical properties of nanotubes have been performed both on ensemble samples and on individual nanotubes. Results of photocurrent spectroscopy and the implications of these measurements for the extraction of photogenerated electron-hole pairs will be presented.


2:00 PM N3.2
Study of Carrier Multiplication in Nanocrystal Quantum Dots and One-dimensional Carbon Nanotubes. Akihiro Ueda, Takeshi Tayagaki, Kazunari Matsuda and Yoshihiko Kanemitsu; Institute for Chemical Research, Kyoto University, Uji, Kyoto, Japan.

The strongly confined carriers in low-dimensional semiconductor nanostructures (nanocrystal quantum dots and carbon nanotubes) show unique optical properties compared with the case of bulk semiconductors. Strong Coulomb interactions between carriers cause various interesting phenomena such as carrier multiplication and quantized Auger recombination. The carrier multiplication, production of two or more electron-hole pairs with one high-energy photon, is one of the most attracting phenomena because of its physical and applicative interests [1]. Carrier multiplication phenomena have been observed in various semiconductor nanocrystals. However, the detailed mechanism of the carrier multiplication is unclear and several reports have pointed out the absence of carrier multiplication [2]. In this work, we studied temporal changes of the carrier density in CdSe nanocrystals, nanorods and single-walled carbon nanotubes by means of time resolved photoluminescence (PL) and femtosecond transient-absorption (TA) spectroscopy. The samples used in this work were CdSe/ZnS core/shell nanocrystals, CdSe nanorods, and single-walled carbon nanotubes. The PL dynamics of CdSe nanocrystals and nanorods were studied by a streak camera. The excitation laser energies were 3.1 and 6.2 eV. The PL dynamics under strong excitation showed faster decay compared with that under weak excitation. The origin of this fast decay is quantized Auger recombination and the lifetime of Auger recombination depended on the size of nanocrystals and nanorods. The PL dynamics under weak excitation at 6.2 eV was almost the same as that under weak excitation at 3.1 eV, which means the absence of carrier multiplication or very low efficiency of carrier multiplication under our excitation conditions. On the contrary, we clearly find carrier multiplication in carbon nanotubes under high-energy excitation [3]. The Auger recombination rate in carbon nanotubes is much faster than that in CdSe nanocrystals. From PL and TA studies on CdSe nanocrystals, nanorods, and carbon nanotubes, we discuss the efficiency and mechanism of carrier multiplication in zero- and one-dimensional semiconductor nanostructures. References [1] R. D. Schaller and V. I. Klimov, Phys. Rev. Lett. 92, 186601 (2004). [2] G. Nair and M. G. Bawendi, Phys. Rev. B 76, 081304(R) (2007). [3] A. Ueda, K. Matsuda, T. Tayagaki, and Y. Kanemitsu, Appl. Phys. Lett. 92, 233105 (2008).


2:15 PM N3.3
Double-Walled Nanotube-Silicon Heterojunction Solar Cells. Yi Jia1,2, Anyuan Cao2, Jinquan Wei1 and Dehai Wu1; 1Department of Mechanical Engineering, Tsinghua University, Beijing, China; 2Department of Mechanical Engineering, University of Hawaii at Manoa, Honolulu, Hawaii.

Carbon nanotubes (CNTs) show high charge mobility and excellent thermal stability, and their photovoltaic applications have been studied extensively, for example, CNTs were used as nanoscale fillers for polymeric solar cells to improve charge separation/transport, or as transparent electrode to improve charge collection. However, conjugated polymers used in constructing solar cells have very low charge mobility, a major factor that limits hole transport through the polymeric matrix to external electrodes and causes high charge recombination, and reported cell efficiencies remain at relatively low levels (<2%). The polymer-based hybrid solar cells also suffer from degradation in air despite that nanotubes are stable at much higher temperature. Several other recent reports involved high-conductivity nanotube films as transparent electrodes to replace indium tin oxide (ITO), however, the nanotubes were not involved into the photogeneration process and the cell efficiency is still limited by the polymeric materials forming the active layers. Here, we developed a novel method to configure double-walled carbon nanotube (DWNT)-silicon heterojunction solar cells. By simply coating thin nanotube films onto silicon wafers we have achieved power-conversion efficiencies of 5-7 % under one solar illumination at 100 milliwatts per square centimeter and negligible degradation of current density after hundreds of hour-exposure in air. The manufacturing process is simple and scalable, involving solution transfer of a porous, single-layer film of double-walled nanotubes onto silicon surface to form heterojunctions with silicon, and does not require separation of metallic and semiconducting nanotubes. We synthesized DWNTs by an improved chemical vapor deposition method. The as-grown DWNT films were treated in H2O2 and HCl solution to remove the impurities and spread to a single-layer film by solution processing. A piece of n-type silicon (n-Si) wafer with a window of pre-deposited insulating layer was immersed into water to pick up the DWNT film to form the heterojunction cell. The nanotubes serve multiple functions in our heterojunction cells, for example, as a heterojunction component for charge separation, as a highly-conductive percolated network for charge transport, and as a transparent electrode for light illumination and charge collection. Our method utilized the distinct properties of both crystalline silicon (e.g. diffusion length) and nanotubes (e.g. high mobility) to construct efficient solar cells and potentially could lead to lower-cost approaches based on nanomaterial-semiconductor heterojunction structures.


2:30 PM N3.4
Electrical and Optical Transport of GaAs/Carbon Nanotube Heterojunctions. Chen-Wei Liang and Siegmar Roth; Max-Planck-Institute, Stuttgart, Germany.

From the semiconductor industry, people have learned that semiconductor heterostructures are very applicable in various technologies, like solar cells or high-frequency communication. Inspired by the history, our work is dedicated to the understanding of the interface between semiconductor and CNTs in order to expand the range of the application of nanotube electronics. We have made heterojunctions consisting of nanotubes and an industrialized semiconductor--GaAs. The transport properties of different types of heterojunctions, p-doped and n-doped GaAs, have also been studied and compared. We found that the p-doped GaAs forms an ohmic contact with the nanotubes while the n-doped GaAs/nanotube heterojunction are rectifying. Given the rectifying effect of the n-doped GaAs/SWNT junctions, transport measurements under illumination were also carried out. The photovoltaic effect therefrom was observed with the light sources of either green laser or regular desk lamp.


SESSION N4: CdSe and CdTe nanocrystals: Synthesis, Properties, and Devices
Monday Afternoon, December 1, 2008
Republic B (Sheraton)

3:30 PM *N4.1
The Use of Multi-Size Arrays of Colloidal Quantum Dots to Study Energy and Electron Transport in QD Junctions. Emily Weiss1, Ryan Chiechi2, Scott Geyer3, Venda Porter3, Moungi Bawendi3 and George Whitesides2; 1Chemistry, Northwestern University, Evanston, Illinois; 2Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts; 3Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts.

This paper describes the electrical characteristics of junctions containing 3D arrays of colloidal CdSe quantum dots (QDs) of either a single size or multiple sizes. The electrodes were indium tin oxide (ITO) covered with a thin layer of poly(3,4-ethylenedioxyl- thiophene):poly(styrene sulfonate) (PEDOT:PSS), and eutectic Ga/In. The turn-on voltage of the junctions depended on the size of the QDs next to the PEDOT:PSS. We describe this dependence using a Marcus model to estimate the barrier for charge transfer induced by the energy gap at the QD/PEDOT:PSS interface. Size-selective photoexcitation of the arrays of multiple sizes of QDs helped to determine the location of the interface at which photoinduced charge separation occurred, whether the energy absorbed by the QDs was redistributed before charge separation, and the dependence of the photovoltage on the locations of various sizes of QDs within the junction.


4:00 PM N4.2
Temperature Dependent Characteristics of Thin Film Nanocrystal Solar Cells. Yvonne W. Rodriguez, Jeremy D Olson, Ingrid E Anderson, Glenn P Gray and Sue A Carter; University of California, Santa Cruz, Santa Cruz, California.

The current-voltage characteristics of thin film solar cells made from layers of CdTe and CdSe nanocrystals were measured as a function of temperature in the dark and under illumination. Data collection was done between 150 K and 380 K. Device structures were fabricated by spin casting the CdTe and CdSe layers onto ITO substrates. After a CdCl2 treatment, the devices were sintered and aluminum contacts were evaporated on. Strong temperature dependence was observed in both the open circuit voltage and the short circuit current density of these sintered devices. The short circuit current density increased as the temperature increased. This suggests transport dominated by thermally activated hopping in these sintered films. The open circuit voltage increased as the temperature decreased. This behavior has been seen in bulk inorganic solar cells. An analysis of the trap density, mobility and junction dynamics are made based on these measurements. The effects of sintering were further examined by measuring the photovoltaic properties of non-sintered cells. Though the open circuit voltage did increase with decreasing temperature as with the sintered films, little temperature dependence of the short circuit current density was observed below room temperature. Above room temperature, the short circuit current density abruptly begins a steady decline. In addition, the effect of plasma treatment and electrode material on the temperature dependent characteristics of these solar cells was studied. Devices were compared with CdSe-P3HT blends and other nanoparticle based thin film devices.


4:15 PM N4.3
Synthesis of New Organic/inorganic Heterostructures from CdSe Quantum Dots and Tetracyanoquinodimethane Derivatives. Anna Maria Laera1, Vincenzo Resta1, Emanuela Piscopiello1, Leander Tapfer1, Francesco Naso2, Francesco Babudri2, Gianluca Maria Farinola2 and Antonio Cardone3; 1Department of Advanced Physics Technology and New Materials (FIM), ENEA, Brindisi, Italy; 2Dipartimento di Chimica, Università degli Studi di Bari, Bari, Italy; 3ICCOM, CNR, Bari, Italy.

Quantum dot (QD) - based hybrid materials currently attract great research and technological interest due to their potential applications in next-generation photovoltaics. The extraction of photoexcited electrons from QDs with the subsequent charge transfer to a suitable matrix represents the key point to efficiently generate a photocurrent. Several ligands for the extraction and transfer of holes are known and applied, while few compounds for the electron extraction/transfer are known and investigated. Due to their peculiar electronic properties, the tetracyanoquinodimethane (TCNQ) derivatives are potential candidates as electron acceptors. These compounds form anion radicals at very positive potentials and form a charge-transfer complex with many donors. Here, we report on the synthesis of tetracyanoquinodimethane and anthraquinone derivatives functionalized with a phosphonic acid group to coordinate the surface of CdSe QDs. The tunable absorption of CdSe QDs covers most of the solar spectrum and, therefore, CdSe is a very suitable material for photovoltaic applications. Furthermore, the possibility to generate multiple excited electrons for a single photon, through a carrier multiplication process, would improve the maximum attainable efficiency of solar photon conversion considerably. We investigated the extraction of photoexcited electrons from CdSe QDs by means of TCNQ derivatives. CdSe QDs were prepared by colloidal synthesis by using trioctylphosphineoxide (TOPO) as solvent. The TOPO capping layer was replaced by a process of ligand exchange in order to obtain the desired new nanocrystal-surfactant complexes. The QDs, typically in range of d=2nm-5nm, were characterized by X-ray powder diffraction (XRPD) and high-resolution transmission electron microscopy (TEM). The successful ligand exchange was monitored by FTIR spectroscopy, that reveals the complete disappearance of the CH-aliphatic stretching of TOPO. The inorganic-organic heterostructure was further analyzed by photoluminescence (PL) and photoluminescence excitation (PLE) spectroscopy. Here, the complete quenching of fluorescence indicates that an electron transfer from the nanocrystals to the ligand may occur. The successful preparation of CdSe QDs functionalized by TCNQ-derivates and the hypothesized electron transfer indicates that this inorganic-organic structure may be a promising candidate to be used as building block in QDs-based hybrid photovoltaic cells.


4:30 PM N4.4
Functionalized Semiconductor Nanocrystal Quantum Dots for Patterned, Multilayered Photovoltaic Devices. Sung Jin Kim1,2, Won Jin Kim2, Alexander N Cartwright1,2 and Paras N Prasad1,2,3; 1Electrical Engineering, University at Buffalo, State University of New York, Amherst, New York; 2Institute for Lasers, Photonics and Biophotonics, University at Buffalo, State University of New York, Amherst, New York; 3Chemistry, University at Buffalo, State University of New York, Buffalo, New York.

We report an approach for nanostructured photovoltaic devices using functionalized semiconductor nanocrystal quantum dots (NQDs) by incorporation of the functional ligand t-butoxycarbonyl (t-BOC) which has an acid-labile moiety. This change in the surface chemistry results in the ability to photo-pattern the NQDs where desired for a number of optoelectronic device geometries with sub-micron patterning capability and multi-layered nanostructure photovoltaic devices. We demonstrate an improved CdTe NQD photoconductive device using this patterning and a ligand modification process. In addition, we demonstrate a bi-layer solar cell using CdTe and CdSe NQDs. The photoconductor was fabricated using a metal semiconductor metal structure. The photoconductor with t-Boc functionalized NQDs showed at least 3 times better conductivity and photoresponse at an applied electric field of 300kV/cm when the ligand was shortened using UV excitation. The t-BOC functionalized NQDs enable all solution processing of multilayered and patterned structures without high temperature sintering or additional chemical treatment. By using this method, a bi-layered solar cell was fabricated. The preliminary results using an unoptimized structure show a 0.05% solar conversion efficiency and an open circuit voltage of 0.72V under AM1.5G 1 sun illumination.


SESSION N5: Intermediate Band Solar Cells
Tuesday Morning, December 2, 2008
Republic B (Sheraton)

8:30 AM *N5.1
Open Questions in the Implementation of the Intermediate Band (IB) Solar Cell. Antonio Luque and Antonio Marti; Instituto de Energia Solar, Universidad Politecnica de Madrid, Madrid, Spain.

The concept behind intermediate band (IB) solar cells lies in absorbing sub-band-gap photons by means of a band located within the band gap while electrons are extracted at a voltage limited by the band gap. While the concept is very attractive, there are several issues that have prevented so far from obtaining a cell with promising efficiency. Quantum dots (QD) can form an IB material from the QD confined states. However, a too small sub-band-gap current is obtained because of the low density of QDs (limited to about 1E17 per cm3) as consequence of their large size. Furthermore, not too many QD layers can be grown without spoiling by stress the solar cell performance. Wavelength selective light confinement might be a solution. In practical QD IB cells, the voltage is reduced by the quantum well associated to the wetting layer that is formed at the time that the QDs are grown and also by the QD levels that appear closely spaced near the valence (VB) and conduction (CB) bands. Alternatively to the QD approach, bulk IB materials are being developed. In these materials, the density of IB states may be much higher, of 1E21 per cm3 and higher. However deep levels in the semiconductor band gap are known to be the origin of non-radiative recombination. It could be argued that the introduction of levels within a semiconductor band gap, at such a high density, might reduce excessively its non-radiative lifetime. However, there are theoretical arguments and experimental data that support the conclusion that, when the Mott transition is reached, the non-radiative recombination disappears. IB solar cells have been formed in high mismatched alloys on the basis of the band anti-crossing mechanism. In this way, Zn-Mg-Te-O alloys have been prepared by O ion implantation followed by pulsed laser melting. Also Ga-As-P-N alloys by N ion implantation. In both cases an IB has been found by spectral photo-reflectance measurements. An IB material has also been prepared by solvo-thermal synthesis in V-In-S compounds and optical absorption has been found in agreement with preceding ab initio calculations. The preparation method is not compatible with cell manufacturing but it is expected that other preparation procedures will be realized able to be integrated in thin film cells fabrication. Deep levels are IB precursors and the knowledge of their position as deep levels can be transferred to a large variety of materials if they are known in another material. Based on this, Ti and Fe are found to be good candidates to form an IB material in copper gallium disulphide. These materials are compatible with the manufacturing process of CIS thin film cells. With the same grounds, other impurities can be predicted to form IB materials in components closer to those used in multi-junction solar cells.


9:00 AM N5.2
An Electronic Structure Study of Intermediate Band by Non-Magnetic Ion Defects in WO3. Muhammad N. Huda, Yanfa Yan and Mowafak M Al-Jassim; National Renewable Energy Lab, Golden, Colorado.

WO3 has a perovskite structure with one of the metal ion position vacant. The larger vacant space inside the crystal structure allows significant local relaxation in case of external doping. With an experimental band gap of around 2.6 eV this presents an interesting opportunity to produce intermediate band (IB) inside the forbidden gap to enhance photo-response. However, realizing IBs in metal oxides pose several problems. We have performed density functional theory (DFT) calculations to study these problems. We have found that non-magnetic ion defects did create IB over the valence band, however, these IBs depend highly on the defects concentration level and local magnetic effects. Moreover, localization of IB’s is one of the major concerns as this would define the performance of the semiconductor. We have found that spontaneous symmetry breaking associated with magnetic degrees of freedom could change the localization of these IBs. On the other hand very disperse doping would simply create a much localized unoccupied band in the gap, which would in turn act as a recombination center, hence degrading the conduction properties of the semiconductor. Over all, controlling these properties to obtain IB semiconductor are challenging issues. We would present our results to identify some of these challenges.


9:15 AM N5.3
Band Structure Calculation and Material Identification for Quantum Dot Intermediate Band Solar Cells. Som N. Dahal1, Stephen P Bremner2 and Christiana B Honsberg2; 1Physics and Astronomy, University of Delaware, Newark, Delaware; 2Electrical and Computer Engineering, University of Delaware, Newark, Delaware.

Intermediate band solar cells are proposed to overcome the single junction limit and can achieve theoretical efficiency as high as 63%. One of the approaches to realize intermediate band solar cells is the quantum dot heterostructures. The identification of proper material combinations (substrate/barrier/dot) for quantum dot intermediate band solar cells including realistic effects such as strain due to lattice mismatch and the effect of heavy and light holes and spin orbit interaction is essential. We present the band structure of quantum dot solar cells with the effect of strain and then band gaps associated with intermediate band solar cells (EIV, ECI and EG) are presented using k.p method.The efficiencies of the corresponding material systems are presented for AM 1.5 solar spectrum.


9:30 AM *N5.4
Applications of Highly Mismatched Alloys to Solar Power Conversion. Wladyslaw Walukiewicz, Materials Sciences Division, Lawerence Berkeley National Laboratory, Berkeley, California.

Highly mismatched alloys (HMAs) are compound semiconductor alloys composed of isovalent constituents with distinctly different electronegativities. In HMAs the hybridization of the extended states of the majority component with the localized states of the minority component results in band restructuring which is described by the band anticrossing (BAC) model. I will present results of our extensive experimental and theoretical studies on a large variety of HMAs. It will be shown that with proper choice of a minority component it is possible to independently control not only the band gap but also the locations of the conduction and the valence band edges of the alloys. This greatly expands the range of potential applications of semiconductor materials. In this talk I will focus on the features of HMAs that could be used for solar power conversion devices. For example, it has been shown that incorporation of dilute fractions of highly electronegative element into the anion site leads to formation of a new subband energetically separated from the original conduction band edge. By tuning the hybridization strength with the alloy fraction and the positions of the subbands with proper choice of the minority atoms, a multi-band system with desired positions and widths of the new subbands can be created. Such multiband materials satisfy the requirements for intermediate band solar cells (IBSCs) in which a more efficient utilization of the full solar spectrum is achieved using the intermediate bands as “stepping stones” facilitating optical transfer of electrons from the valence to the conduction band. Dilute group III-V nitrides or dilute group II-VI oxides are among the most extensively studied HMAs for IBSC applications. The materials can be synthesized using either ion implantation combined with pulsed laser melting or standard epitaxial growth techniques. Detail studies of electronic properties of ZnOxTe1-x with x ≤ 0.03 and GaNxAs1-x-yPy with 0.3 < y < 0.5 and x up to 0.02 demonstrate a formation of a narrow intermediate band below the conduction band. The location of the band can be adjusted to satisfy the requirements for the maximum efficiency IBSCs. Most recent studies led to a significant progress in understanding of structural, optical and electrical properties of these HMAs and photo responses from ZnOTe and GaNAsP systems has been clearly demonstrated. I will also discuss the N-rich InGaNAs alloys in which incorporation of dilute amounts of As shifts the valence band towards the oxygen redox potential of water making this material system suitable for solar water splitting. *In collaboration with Solar Energy Materials Research Group (http://emat-solar.lbl.gov/). Supported by the U.S. Department of Energy under Contract No. DE-AC02-05CH11231.


SESSION N6: Up/Down Conversion and Intermediate Band Solar Cells
Tuesday Morning, December 2, 2008
Republic B (Sheraton)

10:30 AM *N6.1
Up-Conversion in Multi-component Organic Systems: Enhancement of the Local Spectral Power Density of the Terrestrial Solar Irradiation? Stanislav Baluschev1 and Tzenka Miteva2; 1Max Planck Institute for Polymer Research, Mainz, Germany; 2Sony Deutschland GmbH, Materials Science Laboratory, Stuttgart, Germany.

Luminescence concentrators in which photon conversion processes such as down-shifting (DS), down-conversion (DC) and up-conversion (UC) are used, can convert via re-emission the solar spectrum to match the absorption properties of different photovoltaic (PV) devices. All these conversion process are expected to lead to increase in the local spectral power density of the Sun-irradiation which means, that more photons with suitable photon energy can be delivered to the PV device. The “classical” concepts focus on developing inorganic or organic PV devices better matching the polychromatic solar spectrum but often when increasing the absorption in e.g. the red part of the spectrum the absorption in blue and green decreases. Another trade-offs within these concepts could be the deterioration in the charge generation and transport in the PV devices. The main advantage of the photon conversion technology is that the concentrators can be considered and optimized as independent optical devices, having no influence on the electrical properties of the operating PV device or its architecture. Consequently, the same photon conversion devices can be combined with variety of PV devices. Until now, experimental demonstration of photon conversion processes excited with Sun-light (non-coherent excitation) and its application for increase of the performance of Si or organic solar cells is made only with luminescence concentrators based on DS. During the last 3 years we have been actively working on the utilisation of the process of UC for Solar spectrum concentration. We investigate and apply for spectrum concentration an original photon UC process based on energetically conjoined triplet-triplet annihilation (TTA) supported bimolecular (sensitizer-emitter) process. In contrast to the all previously described UC-techniques (such as simultaneous or sequential absorption of two or more photons, second and higher harmonic generation, UC-systems based on rare-earth doped phosphors) the fundamental advantage of the TTA based photon UC is its inherent independence on the coherence of the excitation light. Another principal advantage of this UC process is the very low required intensity (as low as 20 mWcm-2) and extremely low spectral power density (as low as 125µWnm-1) of the excitation source needed - so it can in reality be the Sun. We present the summarised results of our investigations and experiments and state the general requirements for the materials’ parameters of the couples sensitizer/emitter in order to obtain efficient energetically conjoined TTA−UC. We also present our working UC-devices for spectrum concentration via UC with up-converted fluorescence quantum yield of 3.2% realized at ultra-low excitation intensities (~20 mWcm-2). We have applied these devices to different solar cells and to our knowledge we are the first to demonstrate dye-sensitized solar cell with performance improved by Sun-light excited photon UC concentrators.


11:00 AM N6.2
Enhanced Upconversion by Plasmonic Field Concentration. Ewold Verhagen, L. (Kobus) Kuipers and Albert Polman; Center for Nanophotonics, FOM-Institute AMOLF, Amsterdam, Netherlands.

Plasmonic nanostructures, integrated with a solar cell, can lead to entirely new solar cell designs with lower materials costs or improved conversion efficiency. Here, we demonstrate how two novel plasmonic nanostructures, deposited onto a solar cell, can be used to upconvert infrared solar radiation to the visible, and inspire novel solar cell design that requires p-n junctions with only nanoscale dimension. Coaxial apertures, with overall diameters of ~400 nm, were made in a thin Ag film deposited onto SiO2 and Al2O3 substrates doped with optically active erbium ions. The apertures have localized plasmon resonances in the range 500-1800 nm, tunable by geometry, with typical bandwidths Δλ/λ=10. Finite-difference time domain simulations show field intensity enhancements in the nanoscale gaps of a factor 100. For arrays resonant at 1500 nm, optical measurements using a 1500 nm excitation source show that Er infrared-to-visible upconversion, integrated over the full array area, is enhanced more than 100 fold. While these experiments were done on samples made using electron-beam lithography, we will also demonstrate how soft-lithography techniques can be used to fabricate similar metal nanostructure over large areas characteristic for solar cells. In an alternative geometry, we couple infrared light into a hole array fabricated into a Ag film, that is coupled to a plasmonic nanotaper. Dispersion calculations, as well as near-field measurements of the plasmonic field symmetry, show large plasmon concentration toards the nanotaper, and a concomitant increased erbium upconversion. Calculations show that the field concentration in these nanotapers is - in principle - only limited by the sharpness of the taper tip. Calculations of geometries optimized for upconversion, taking into account Ohmic losses in the metal, will be presented. The metal hole array and nanotaper geometries can be readily integrated in existing solar cell metal contact designs, as will be discussed. Alternatively, they inspire entirely novel designs in which a nanoscale p-n junction is integrated with the plasmonic hotspot. Due to the small size, low-quality semiconductor material, with low minority carrier lifetime can be used, reducing cost. Several of these novel designs will be presented, and preliminary experimental data will be shown. We will elaborate on the pros and cons of implementing plasmonic nanostructures in solar cells in general, balancing losses and fabrication feasibility with the benefits of solar upconversion and reduced materials costs.


11:15 AM N6.3
Near-Infrared Quantum Cutting in SrF2: Pr, Yb. Andries Meijerink, Bryan van der Ende and Linda Aarts; Debye Institute, Utrecht, Netherlands.

Spectral conversion of the solar spectrum is a promising avenue to boost the energy efficiency of photovoltaic cells. Energy losses due to thermalization of hot electrons and holes can be minimized through near-infrared (NIR) quantum cutting, whereby one higher energy photon is converted into two NIR photons. Subsequent absorption of both NIR photons by the solar cell results in current doubling for the high energy part of the solar spectrum. Lanthanide ions are the prime candidates to realize efficient quantum cutting [1]. In this work, we present evidence for a close to 200 % quantum cutting efficiency of visible photons into NIR photons in SrF2:Pr3+, Yb3+ through resonant two step energy transfer. Conclusive information on the transfer efficiency is obtained from comparison of the diffuse reflection spectra and the excitation spectra. The intensity of 3H4→3PJ, 1I6 lines relative to the 3H4→1D2 lines is twice as large in the excitation spectrum of the Yb3+ emission than in the diffuse reflection (absorption) spectrum. This shows that absorption of a single photon in the 3PJ, 1I6 spectral region results in the emission of two NIR photons by Yb3+. Energy transfer occurs through two sequential resonant energy transfer steps between Pr3+ and Yb3+ with 1G4 as an intermediate level: Pr3+(3P0→1G4)-Yb3+(2F7/2→2F5/2) followed by Pr3+(1G4→3H4)-Yb3+(2F7/2→2F5/2). The energy transfer is effective at relatively low Yb3+ concentrations (5%), where concentration quenching of the Yb3+ emission is limited. Comparison of emission spectra, corrected for the instrumental response, for SrF2:Pr3+ (0.1%) and SrF2:Pr3+ (0.1%), Yb3+ (5%) reveal an actual conversion efficiency of 140%. Optimization of the synthesis conditions are expected to further improve this efficiency by reducing losses due to energy transfer to quenching sites. A next step towards the conversion of the short wavelength part of the solar spectrum will involve the inclusion of a sensitizer for the 3P0 level of Pr3+ which is able to absorb efficiently over a broad wavelength range (300-500 nm) and subsequent energy transfer to the 3P0 level of Pr3+. It has been calculated that (ideal) spectral downconversion can increase the Shockley-Queisser limit for p-n semiconductor solar cells up to 40% for an (optimum) bandgap of 1.05 eV, very close to the bandgap of c-Si [2]. The present results demonstrate the potential of spectral downconversion by lanthanide ions for reducing energy losses in photovoltaic devices and increasing the efficiency of c-Si solar cells to make them more cost-effective. [1] Wegh, R. T., Donker, H., Oskam, K. D. & Meijerink, A. Visible quantum cutting in LiGdF4:Eu3+ through downconversion. Science 283, 663-666 (1999). [2] Trupke, T., Green, M. A. & Würfel, P. Improving solar cell efficiencies by down-conversion of high-energy photons. J. Appl. Phys. 92, 1668-1674 (2002).


11:30 AM *N6.4
A Molecular Approach to the Intermediate Band Solar Cell. Timothy Schmidt1 and N. J Ekins-Daukes2; 1School of Chemistry, University of Sydney, Sydney, New South Wales, Australia; 2Blackett Laboratory, Imperial College, London, United Kingdom.

Traditional implementations of the intermediate band solar cell using inorganic materials suffer from short carrier lifetimes in the intermediate band. By harnessing the inter-system crossing processes which occur naturally in organometallic molecules, we store energy in molecular triplet states, forming a long-lived intermediate "band" after absorption of low energy photons. The carriers in the intermediate band undergo triplet-triplet annihilation, an Auger-like process, to produce carriers in the conduction "band". These carriers may recombine to produce up-converted light, or injected into a semiconducting substrate such as TiO2. Thermodynamic and kinetic modelling of a solar cell based on this mechanism show a limiting efficiency of about 40% under terrestrial insolation, optimized at a bandgap of about 1.75 eV and exceeding 35% for a range of bandgaps between 1.5 eV and 3.0 eV, allowing considerable flexibility in molecular engineering.


SESSION N7: Plasmonics in Photovoltaics
Tuesday Afternoon, December 2, 2008
Republic B (Sheraton)

1:30 PM *N7.1
Plasmonic Photovoltaics. Harry Atwater, Applied Physics, California Institute of Technology, Pasadena, California.

To date, little systematic thought has been given to the question of how plasmon excitation and light localization might be exploited to advantage in photovoltaics. Using insights derived from the other phenomena studied in the plasmonics field, we outline approaches to dramatically modify the light absorption and carrier collection characteristics of photovoltaic materials and devices. In particular, the ability of plasmonic structures to localize light at subwavelength dimensions is synergistic with use of ultrathin thin, quantum well, and quantum dot photovoltaic absorber materials. Conventionally, photovoltaic absorbers must be optically ‘thick’ to enable nearly complete light absorption and photocarrier current collection. They are usually semiconductors whose thickness is typically several times the optical absorption length. For silicon, this thickness is greater than 100 microns, and it is several microns for direct bandgap compound semiconductors, and high efficiency cells must have minority carrier diffusion lengths several times the material thickness. Thus solar cell design and material synthesis considerations are strongly dictated by this simple optical thickness requirement. Dramatically reducing the absorber layer thickness could significantly expand the range and quality of absorber materials that are suitable for photovoltaic devices by, e.g., enabling efficient photocarrier collection across short distances in low dimensional structures such as quantum dots or quantum wells, and also in polycrystalline thin semiconductor films with very low minority carrier diffusion lengths.


2:00 PM N7.2
Gold Nanoparticles for Enhancing Dye Absorption. Markus Hallermann1, Tapan K Sau1, Enrico Da Como1, Andrey L Rogach1, Jochen Feldmann1 and Omar Stern2; 1Photonics and Optoelectronics Group, Department of Physics and CeNS, Ludwig-Maximilians-Universität, Munich, Germany; 2GE Global Research Europe, Garching b. Munich, Germany.

Thin film solar cells are one of the most promising technologies for the development of third generation photovoltaic devices. While the thin film technology takes advantage from the easy implementation of such devices in many different applications, the low optical density limits the amount of solar energy harvested. Nanostructures able to enhance the absorption cross-section of the sensitizer are highly desirable in order to achieve higher efficiencies. Plasmonic nanostructures based on coupled metal nanoparticles can generate local fields of high intensity with spectral tunability across the visible spectral range, acting as nanoantennas [1]. A major challenge in the exploitation of this concept in the existing thin film solar cell technology is the spatial control of enhancement, avoiding loss mechanism such as charge carrier trapping or recombination. It is, therefore, important to design well localized enhancement effects involving the sensitizer. In this communication we present the combination of metal nanoparticles with the system dye/titanium dioxide (TiO2). The synthesis of the nanoparticles has been performed in solution ensuring a facile implementation in the low cost solution processing of dye/TiO2. We demonstrate the self-assembly of nanoparticles on TiO2 thin films coated with a monolayer of dye. By comparing the absorption spectra of the hybrid sample with the reference samples of dye/TiO2 and nanoparticle/TiO2, we show an enhancement of the absorption cross-section in the spectral regions corresponding to the nanoparticle plasmon resonance. [1] M. Ringler, A. Schwemer, M. Wunderlich, A. Nichtl, K. Kürzinger, T. A. Klar, J. Feldmann, Phys. Rev. Lett., 100, 203002 (2008)


2:15 PM N7.3
Thin-Film Silicon-Solar Cells and Metallic Tin Nanoparticles. Mads Moegelmose Kjeldsen, J. L Hansen, J. Chevallier and A. N Larsen; Department of physics and astronomy, Aarhus University, Aarhus, Denmark.

Crystalline silicon-thin film solar cells are intensively explored world wide due to the potential of reducing the material expense. However, light trapping becomes of particular importance when the active cell thickness is reduced. An attractive way of increasing light trapping is to utilize surface plasmons excited in metallic nanoparticles deposited on the top surface of the solar cell. By proper architecturing of nanoparticle materials, size and shape the light scattering is foreseen to be significantly increased and surface reflectivity can simultaneously be reduced. We will report on light absorption in silicon thin films with a dielectric top layer containing metallic tin-nanoparticles. Metallic tin, also known as β-Sn, is chosen over other metals for belonging to the same elemental group as silicon, therefore, tin as a contaminant in silicon does not introduce deep levels in the band gap, and moreover, the β-Sn bulk-plasmon energy is close to the silicon band-gap energy. A unique mesa-diode structure with the emitter at the bottom was developed and utilized for studying the photovoltaic characteristics of the nanoparticles structure. The mesa-diodes are produced by means of molecular-beam epitaxial growth of a p- layer on a n+ substrate followed by photolithography and etching. Subsequently, at the topmost, a layer of metallic tin-nanoparticles is produced by low temperature sintering of a thin silicon-dioxide film containing one layer of tin deposited by magnetron sputtering. The thickness of the tin layer is varied to study the impact of the size of the nanoparticles. Also, the doping concentration and thickness of the p- layer are varied to study the effects on the light absorption. First results demonstrate a photo-current enhancement caused by the presence of the tin nanoparticles, which from transmission electron microscopy (TEM) investigations are concluded to be spherical and in the metallic β-Sn phase. Combined results from TEM studies, measurement of photo currents, and absorption and reflection studies will be presented.


2:30 PM N7.4
Coupling Au Nanoparticles Surface Plasmon Resonance with A-Si:H/c-Si Heterojunction Solar Cells for Enhanced Absorption and Efficiency. Giovanni Bruno, Maria M Giangregorio, Alberto Sacchetti, Pio Capezzuto and Maria Losurdo; PlasmaChemistry, IMIP-CNR, Bari, Italy.

A variety of approaches for increasing optical absorption in silicon solar cells is being investigated, with preference to light-absorption and trapping mechanisms that do not increase recombination losses. In this frame, exploitation of the localized absorption of metal nanoparticles (NPs) via surface plasmon resonance (SPR) has been proposed since the wavelength-selective photon absorption can be tailored by the NPs’ size, shape and local dielectric environment. Among the various solar cell technologies, over 85% of the current production is based on the well-established silicon wafer technology. In this frame, a-Si:H/c-Si heterojunction solar cells have been the subject of extensive research especially since Sanyo has demonstrated efficiencies >21% for heterojunction solar cells produced by plasma enhanced chemical vapor deposition (PECVD) of a-Si:H on crystalline silicon (c-Si). So far, there are no reports in literature about exploitation of SPR metal nanoparticles coupled directly with hydrogenated amorphous silicon (a-Si:H). Furthermore, although many studies and approaches reported in literature are mainly based on the use of Au NPs colloidal solutions requiring transferring and stabilization of the metal nanoparticles on the semiconductor/device platform of interest and additional steps of solvent evaporation, we are seeking and investigating dry processes for direct formation of metal NPs compatible with silicon-based photovoltaic technologies. In this contribution, we report on the integration of the a-Si:H PECVD with the Au sputtering process for depositing Au NPs directly on a-Si:H to exploit the metal nanoparticles surface plasmon resonance in enhancing light absorption. This approach allows Au nanoparticles to be added at the late stage of an integrated PV cell processing at a temperature of approximately 200°C by direct sputtering of Au NPs onto PECVD a-Si:H. The impact of Au nanoparticles density and size and of sputtering temperature on the heterojunction performance will be presented and discussed. The correlation among the electrical, morphological and optical properties of the Au-NPs/(n)a-Si:H/(p)c-Si heterojunctions is discussed, exploiting spectroscopic ellipsometry (UVISEL-Jobin Yvon), which is demonstrated to be useful for determining the a-Si:H film optical properties as well as the surface plasmon resonance of Au nanoparticles, whose diameter is determined by atomic force microscopy (AFM). The I-V characteristic of devices show that engineering Au nanoparticles’ diameter and density makes possible improvements in conversion efficiency. In particular, we demonstrate an improvement of 20% of the short-circuit current, Jsc, an increase of 3% of the fill factor, FF, an increase of 25% of the power output, Pmax, and, consequently, of the energy efficiency conversion for (n-type)a-Si:H/(p-type)c-Si heterojunctions with 20 nm Au nanoparticles (NPs), which are characterized by SPR at 572 nm.


2:45 PM N7.5
Light Trapping with Particle Plasmons. Kylie R. Catchpole1,2, Fiona Beck2, Ruud Schropp3 and Albert Polman1; 1Center for Nanophotonics, FOM Institute AMOLF, Amsterdam, Netherlands; 2Centre for Sustainable Energy Systems, Australian National University, Canberra, ACT, Netherlands; 3Debye Institute, University of Utrecht, Utrecht, Netherlands.

The excitation of surface plasmon resonances on metal nanoparticles is a promising way of increasing the efficiency of thin film solar cells. We develop a number of design principles for increasing the efficiency of solar cells (or light-emitting diodes) using metal nanoparticles. We show that potential path length enhancements of up to a factor of 28 can be achieved in the near-infrared for cylindrical and hemispherical Ag particles on Si, because the incident light leads to a dipole-type excitation very close to the interface with the substrate. (In fact, the path length enhancement for a horizontal electric point dipole is over twice the Lambertian value). In contrast, spherical particles have a low potential path length enhancement of 5-10, due to the low fraction of their volume close to the surface. Because of the exponential decay of high in-plane wavevector components of the scattered field, this results in poor coupling to the substrate. We also show that the scattering cross-section of a particle depends not only on the particle size and shape, but also its position relative to the substrate. For example, increasing the thickness of a Si3N4 spacer layer from 10nm to 30nm increases the scattering cross-section of a 100nm Ag particle on Si by a factor of 2. Finally, we show that Ag particles give much higher path length enhancements than Au particles. Optimizing plasmon mediated light trapping in thin film solar cells thus requires balancing scattering cross-section, fraction of light scattered into the substrate and the angular distribution of the scattered light. We apply the design principles to 100μm thick bifacial c-Si silicon solar cells and 350nm thick p-i-n a-Si:H solar cells deposited on glass substrates with Ag nanoparticles formed by evaporation followed by annealing, and show substantially increased near-infrared photocurrent. We also demonstrate that fine control over nanoparticle size and shape can be achieved with a process that is scalable to large areas using nanoimprinting, and we report results for silicon solar cells with Ag nanoparticles fabricated using nanoimprinting.


SESSION N8: Photon Management and Si Quantum Dot Solar Cells
Tuesday Afternoon, December 2, 2008
Republic B (Sheraton)

3:30 PM N8.1
3D Photonic Spectrally Selective and Diffractive Intermediate Filter for Micromorph Tandem Cell. Andreas Bielawny1, Johannes Upping1, Paul T Miclea1, Ralf B Wehrspohn1, Seung-Mo Lee2, Mato Knez2, Marius Peters3, Andreas Lambertz4 and Reinhard Carius4; 1Physics, Martin-Luther-University of Halle-Wittenberg, Halle, Germany; 2Max-Planck-Institute for Microstructure Physics, Halle, Germany; 3FMF (Freiburg Centre for Material Research), University of Freiburg, Freiburg, Germany; 4Institute of Energy Research, Forschungszentrum Jülich, Jülich, Germany.

In thin-film silicon tandem solar cells consisting of amorphous and microcrystalline silicon (a-Si/µc-Si) efficiency enhancements can be achieved by increasing the current density in the a-Si top cell to provide an optimized current matching at high current density. One option to achieve this is the increase of absorbance in the a-Si cell in its spectral region of low absorption by a selective reflector which reflects the spectral part of the spectrum that can be absorbed by the a-Si cell after multiple reflections. At present, optical interlayers of low refractive index are placed between the two cells making use of the high reflectivity at an interference maximum at a certain thickness of the layer. This effect depends on the angle of incidence and might be not ideally suited for strongly scattered light from the textured front TCO used in such tandem cells. However, for an ideal photon-management between top and bottom cell, a spectrally selective interlayer which is less dependent of the angle of incidence and provides further options for improved light trapping would be necessary. We have determined numerically the relative efficiency enhancement of an a-Si/µc-Si tandem solar cell using a conductive 3D-photonic crystal. The inverted opal is capable of providing a suitable optical band stop with high reflectance and the necessary long wavelength transmittance as well - leading to an increase of the efficiency of more than 10%. We have fabricated such structures by ZnO-replication of polymeric opals using atomic layer deposition on the rear side of a-Si solar cells. Completed with a back contact, this is the first step to integrate this novel technology into an a-Si/µc-Si tandem solar cell process. The spectral response of the cells will be presented and compared with reference cells containing varied interlayers.


3:45 PM N8.2
Very Effective Light Trapping in Grating Thin-Film Solar Cells. Mukul Agrawal and Peter Peumans; Electrical Engineering, Stanford University, Stanford, California.

Light-trapping is used in photovoltaic cells to increase the efficiency and lower the cost by reducing the amount of active material required to efficiently absorb sunlight. A well-known approach is to use textures that scatter incident rays into modes that are trapped in the absorbing layer. Since this geometric approach is ineffective for thin-film solar cells, an extension of this approach into the wave domain is needed. We show that the light trapping performance in the wave domain cannot surpass the limits obtained in the geometric limit (i.e. the Yablonovitch limit). Furthermore, we show that practical subwavelength structures can be designed with light-trapping performance that approaches this theoretical limit. Concrete light trapping structures are evaluated using a fast implementation of the rigorous coupled wave analysis method, which allows us to rigorously scan design space and determine optimal light trapping structures. Specifically, a three dimensional redistribution of the active material of a 250nm-thick hydrogenated amorphous silicon cell in the form of a two-dimensional in-plane photonic crystal provides a spectrally-averaged absorption enhancement that is several times higher than the state-of-the-art.


4:00 PM N8.3
An Array of Nanocones with Sharp Tips as Efficient Antireflection Layers for Solar Cells. Jia Zhu, Zongfu Yu, George Burkhard, Ching-Mei Hsu, Stephen Connor and Yi Cui; Stanford, Stanford, California.

Graded refractive-index layer has been actively pursued for broadband antireflection properties. However, current technologies are either limited to specific materials or requiring complicated process. Here an array of nanocones with sharp tips (~5nm) fabricated by a novel process is reported. The nanocones array can suppress the reflection of light at a wide rage of spectrum and angles of incidence. More importantly, the process is very simple and can be applied to different kinds of materials. Simulation results are also provided to explain the results. This technique could have many important applications in electro-optical devices, especially in Solar Cells and Light Emitting Diodes.


4:15 PM N8.4
Bandgap Engineering of Silicon Quantum Dot Nanostructures for High Efficient Silicon Solar Cell: The Tandem Approach. Georges Bremond and Bechir Rezgui; INL, INSA Lyon, Villeurbanne, France.

A tandem approach is proposed using Silicon nanostructures to increase the efficiency of so-called third generation photovoltaic solar cells.Si quantum dot nanostructures (or silicon nanocrystals)are synthesized by depositing silicon-rich nitride (SRN) layers using plasma-enhanced chemical vapour deposition (PECVD). We have shown the intrinsic formation of silicon nanocrystals (nc-Si) in non-stoechiometric amorphous hydrogenated silicon nitride (a-SiNx:H) layers using pure silane (SiH4) and ammonia (NH3) as reactants. The NH3 would provide more hydrogen in the silicon nitride film leading to an improvement of the crystallinity of Si quantum dots (QD) by favouring the disorder-to-order transition. Furthermore, hydrogen dissociated from the NH3 would passivate the surface of a Si QD more effectively.Transmission Electron Microscopy (TEM) was employed to explore the microstructure of the as-deposited Si-in-SiNx composite films. The chemical bonds of these films were examined by using Fourier Transform Infrared (FTIR) spectroscopy in the wavenumber range from 400 to 4000 cm-1 with a resolution of 4 cm-1. The photoluminescence (PL) property of silicon nanocrystals in silicon-rich nitride (SRN) layers are also investigated. The peak position of PL could be controlled by adjusting the flow rates of ammonia and silane . Two types of luminescent mechanisms, such as radiative defects in the film and the quantum confinement effect (QCE) in silicon nanocrystals, have been proposed to explain the origin of light emission from these structures. These two mechanisms are inherently coexisting in our samples and the photoluminescence spectrum depends on the contribution of each other. The optical absorption properties of the deposited films are obtained and analyzed from light transmittance measurements. Spectroscopique ellipsometry have been performed in order to analyse the refractive index and the extension coefficient. All these measurements were carried out at room temperature. These techniques have given good correlation in the extraction of the absorption coefficient induced by the Si nanocrystal in the visible /UV energy range. Measurements of photocurrent have shown a great increase of the induced currrent in the visible/UV energy range for an optimum of deposition conditions. These results will be discussed in order to reach a better knowledge of the physical properties of this third generation photovoltaic all silicon included material for the tandem solar cell application approach.


4:30 PM N8.5
Metamaterial Based on the Nanostructured Si for Multistage PV Conversion. Zbigniew T. Kuznicki and Patrick Meyrueis; Photonic Systems Laboratory, Illkirch, France.

Completing one-step PV conversion by additional new low-energy mechanisms is one of the most important challenges of modern photovoltaics. Silicon is a basic PV material which is not efficient enough to convert light into electricity in its bulk or thin film form because of its indirect bandgap. Progress in conversion efficiency requires breakthroughs. One way has been indicated by low-dimensional or nanostructured Si materials as, for example, nanoscale Si-layered systems combined with an active interface with its crystalline defects. We have demonstrated low-energy carrier multiplication experimentally under attenuated solar excitation in nanostructured Si. The investigation has been limited to comparative measurements of the short-circuit currents: reference and modified. Their ratio shows steps which allow estimating a characteristic energy E = 0.274 eV, previously determined by us from spectral response and modeling. The effect is particularly visible under weak incident beams. Another investigation of the modified surface activity, presented in, shows that the state of the semiconductor front surface/interface dominates first the optoelectronic properties of the MIND (multi-interface novel device) and, as a consequence, its PV performance. This behavior raises a series of paradoxes in comparison with well known Si photovoltaics. Most of these paradoxes can be explained on the basis of electronic and optical features as could be confirmed in simulations. The nonlinear behavior of the surface region versus incident light intensity shown by us previously results from the high free-carrier density. The incident flux acts through its intensity and spectral composition. The necessary modification consists of the introduction of a nanoscale Si-layered system into the surface zone of a conventional material (Si wafer). The nanoscale and multiinterface approaches modify the existing semiconductor, such as single-crystal Si, e.g. by making use of geometrical free-carrier confinement due to the CCL. This allows introducing useful energy levels and a ''new'' band structure, leading to a multilevel multistage PV conversion. A good “tuning” of device energy levels to the whole solar spectrum seems to be possible. In its simplest form it could be obtained by a superposition of only two groups of mechanisms having different energy thresholds: larger primary (conventional conversion stage) and smaller secondary (new conversion stages). The secondary mechanism combined with confinement peculiarities can be at the origin of multiple subpopulations distinguished from each other by their average energy. Certain non obvious features of the continuous nanoscale Si-layered system resulting from the buried amorphization can be noted.


4:45 PM N8.6
Nanostructured Silicon-based Layers for Photovoltaic Applications. Fabrice Gourbilleau1, Bechir Rezgui2, Abel Sibai2 and Georges Bremond2; 1CIMAP, ENSICAEN, Caen Cedex 04, France; 2INL, Insa Lyon/Université Lyon, Villeurbanne, France.

Since the 1990’s with the discovery of the quantum confinement properties in silicon when the Si grain size is reduced to the value of the Bohr radius (~5 nm), numerous studies have been carried out to produce materials having specific properties. Some of them concern the use of Si nanoclusters (Si-nc) such as nanomemories for microelectronic or as efficient sensitizer towards rare earth ions for optoelectronic applications. More recently, it has been proposed to take advantage of the quantum confinement effect in Si-nc for photovoltaic applications by managing the Si bandgap and consequently producing higher efficient tandem solar cells. Different approaches can be envisaged such as a composite Si-SiO2 or Si/SiO2 multilayer structure which requires a control of Si-nc size as well as a high Si-nc density in the silica host matrix for the achievement of solar cell device. In this context, we propose an original approach by means of reactive magnetron sputtering allowing the growth of either composite Silicon-rich silicon oxide (SRSO) layer or SRSO/SiO2 multilayers. The incorporation of a Si excess in the growing layer is obtained through the introduction of hydrogen in the Ar plasma for sputtering a pure silica target. In the case of the composite films, a mixture of Ar and H2 is used while for the multilayer approach the silica target is sputtered using either a mixture of Ar and hydrogen or pure argon gases for depositing SRSO or SiO2 sublayers, respectively. The control of the hydrogen quantity diluted in the argon gas allows to monitor the Si-nc density either in the SRSO composite or SRSO sub- layer. Moreover, the structure of the multilayer offers the possibility to manage the Si-nc size through the SRSO sublayer thickness. Microstructure of the deposited layers has been analyzed by means of X-ray diffraction, Infrared spectroscopy and High Resolution Electron Microscopy, while the optical properties have been studied by means of time resolved or not photoluminescent spectroscopy and absorption experiments. Thus, we have noticed an increase of the absorption coefficient by two orders of magnitude with the decreasing size of the Si-nc from 8 to 1.5 nm which is a promising result for the achievement of a Si-nc based tandem solar cell having a high efficiency.


SESSION N9: Poster Session I
Tuesday Evening, December 2, 2008
8:00 PM
Exhibition Hall D (Hynes)

N9.1
RF-Sputtered Ge-ITO Nanocomposite Thin Films for Photovoltaic Applications. Grace Shih, Tracie Bukowski, Barrett G Potter and Joseph H Simmons; Materials Science and Engineering, University of Arizona, Tucson, Arizona.

The use of inorganic quantum dot (QD) ensembles as an absorptive component in nanostructured photovoltaic junction designs is of current interest as an alternative to organic sensitizers. QD composition and quantum-size effects both can provide tuning of the QD optical absorption within the solar spectrum with the promise of improved energy conversion efficiency. For sensitization of thin film heterojunctions for photovoltaic energy conversion, the introduction of QD’s into junction structures represents an important challenge, requiring control of the mesoscale QD phase distribution along the film depth (e.g. within the depletion region of the heterojunction) and the QD-matrix interfacial electronic character (for efficient carrier extraction and transport). The current work introduces the use of a sequential, dual-source, RF-sputtering deposition technique to produce Ge-indium-tin-oxide (ITO) nanocomposite thin films with controlled QD phase distribution within the ITO matrix. Single and multilayer structures of Ge embedded in an ITO matrix have been formed and the technique has also been used to produce a QD layer in the vicinity of an ITO-Si heterojunction. Modification of local Ge volume fraction has been demonstrated through the control of substrate exposure times to each target. Transmission electron microscopy and optical absorption spectroscopy has confirmed the formation of Ge quantum dots upon post-deposition thermal annealing and photoconductivity has been observed in the Ge-ITO nanocomposite. Photovoltaic characterization of these structures will also be presented.


N9.2
Application of the Design of Experiments Method to the Study of the Intermediate-Band InAs/GaAs Quantum-dot Solar Cell. Alysha Grenko1, John J Walker2, Ibrahim Kimukin2, Brian Hoskins1 and Elias Towe1,2; 1Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania; 2Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania.

Conventional single-junction photovoltaic devices can only convert a narrow band of the solar spectrum into electric energy. Appropriately engineered semiconductor quantum-dots—inserted into the active regions of solar cells—could potentially capture a broader spectrum, thus improving device performance. Specifically, the quantum dots can be embedded in the i-region of a simple homojunction p-i-n device architecture. By controlling the size, shape, and composition of the quantum dots, their discrete energy levels can be made to form an “intermediate band” located within the energy gap of the homojunction semiconductor material. This intermediate band should lead to the absorption of low energy photons that would normally be lost to the photoelectric conversion process in a conventional single-junction device. This novel device architecture is expected to have a higher conversion efficiency than a single-junction cell because of the additional photocurrent that would be collected from the low energy photons. This paper studies the impact of various parameters on the characteristics of the intermediate-band solar cell device. The devices studied in this work are fabricated from material synthesized by the method of molecular beam epitaxy. The basic device architecture consists of 10 periods of InAs/GaAs quantum-dot layers inserted between the p-GaAs and n-GaAs layers of an otherwise conventional p-n junction structure. Several parameters have a major impact on the performance characteristics of the intermediate-band solar cell. These parameters include (i) the thickness of the InAs wetting layer from which the dots are formed; (ii) the substrate temperature at which a fraction of the spacer layer thickness between dots is grown; (iii) the thickness of the inter-dot spacer layer; and (iv) the magnitude of an electron sheet carrier concentration inserted between the inter-dot spacer layers during a growth interruption. To uncover the impact of each parameter on a performance characteristic, a structured approach is indispensable. In this work, we use a systematic statistical method, known as the design of experiments (DOE) method. The essence of the method is to uncover the impact of a factor (parameter) affecting the process and the output (characteristic) of the device. By varying the parameters listed above, and analyzing the resulting device characteristics, we have uncovered the factors that impact the spectral location and width of the intermediate band. Furthermore, we have identified the parameters that can be used to improve the photoresponse, including how to reduce non-radiative recombination losses in the quantum-dot region. This paper discusses the details of the application of the DOE method to the design of an intermediate-band solar cell structure, and highlights preliminary results obtained through use of the method.


N9.3
Fabrication of Optically Active Thin Films by Chemical Solution Deposition. Edita Garskaite, Mari-Ann Einarsrud and Tor Grande; IMT, NTNU, Trondheim, Norway.

Conversion of light in doped thin films is very attractive for increasing the efficiency of photovoltaic cells. Chemical solution deposition of thin films of such oxide materials is very attractive due to low costs and environmental issues. Here we present our resent data on preparation of optically active thin films prepared by aqueous sol-gel dip-coating route. Several different Er-doped optically transparent materials such as YAG, ZnO and In2O3 have been prepared. The influence of the precursor sol composition, deposition parameters, heat treatment of the films on the nanostructure and homogeneity of thin films on silicon substrates will be discussed. Finally, up-conversion process in thin oxide-based films has been demonstrated.


N9.4
Growth of Copper Indium Disulfide Nano-rods using Anodized Aluminum Nano-mask. Hong Lam1, Zhong Wei Zhang1,2, Chung Pui Chan1, Zhuo Chen1, Charles Surya1 and Chang Fei Zhu2,1; 1Department of Electronic and Information Engineering, The Hong Kong Polytechnic University, Hong Kong, Hong Kong, China; 2Department of Materials Science and Engineering, University of Science and Technology of China, Hefei, Anhui, China.

Nano-structured semiconductor material is considered advantageous in photovoltaic applications. In this work, Copper Indium Disulfide (CIS) nano-rods were electrodeposited by utilizing anodized aluminum oxide (AAO) nano-mask. The AAO nano-mask was obtained using a two-step anodization process. The anodization was carried out in a 0.05 to 0.3M oxalic acid solution at 1°C and a potential of 40 to 100V was applied. The alumina yielded in the first anodization step was dissolved by immersing the sample in a mixed acid solution with phosphoric acid and chromic oxide solution (6wt% H3PO4+1.8wt% H2CrO4) at 60°C. By controlling the time of the second step, we can control the ultimate thickness of the AAO film. Subsequently, the AAO foil was subjected to chemical etching (in 5wt.% H3PO4 for 30 to 50 minutes at 30°C) to enlarge the pores and remove the barrier layer at the bottom of AAO template. Such two-step process has shown to help improving the pore order in the AAO foil. A continuous layer of Cu and In film on the back side of the AAO layer was obtained by sequential deposition of elemental 5N Cu and In by e-beam evaporation. The total thickness of the stack was about 1μm. The stoichiometric ratio of Cu and In was maintained at 1:1. A 1μm thick layer of molybdenum was then deposited by e-beam on top of the Cu/In layers. To provide mechanical support, 100μm of copper was then electroplated onto the Mo layer. Depositions of Cu and In through the nanometer-scale windows in the AAO were performed by electrodeposition technique. We used a three electrode system with aquatic solutions containing copper and indium salts. The metal salts concentration was adjusted so that deposition rate of copper and indium ions is approximately 1:1. The electrodeposition process was conducted in constant current mode at 1mA and that deposition time ranged from 10 to 30 minutes. Columns of stoichiometric copper indium nano-rods were obtained and embedded inside AAO nano-holes. Finally, the AAO mask was eliminated using phosphoric acid and the nanowires were sulfurized in a quartz tube furnace with sulfur vapor at 450°C using argon as the carrier gas. X-ray diffraction characterization studies confirmed the presence of copper indium disulfide crystal structures in the material. Scanning electron microscope pictures of the columnar CIS nano-rods were taken. The structures were characterized for their optical and electrical properties.


N9.5
Wet-Chemical Route to ZnO/CuInS2 Core-Shell Nanowire Array for Photovoltaic Applications. Jih-Jen Wu and Wan-Ting Jiang; Department of Chemical Engineering, National Cheng Kung University, Tainan, Taiwan.

Well-aligned ZnO/CuInS2 core-shell nanowire (NW) arrays have been synthesized on fluorine-doped tin oxide substrates using a simple wet-chemical route, i.e., aligned ZnO NW array first formed by an aqueous chemical bath deposition (CBD) and then conformal CuInS2 shells deposited on the ZnO NWs using the successive ionic layer absorption and reaction (SILAR) method. The thickness of the CuInS2 shell is variable by the cycle number of SILAR. The absorption edge of the ZnO/CuInS2 core-shell NWs is at ~700 nm after annealing in sulfur atmosphere. The ZnO/CuInS2 core-shell NW arrays with a NW length of 3 μm have been employed to be the n-type semiconductor and the absorber for fabricating the eta (extremely thin absorber) solar cells. Na2S liquid electrolyte and CuSCN were utilized as the p-type material in the eta solar cells. The results show that a current density and an efficiency of 3.4 mA/cm2 and 0.46%, respectively, are achieved in the ZnO/CuInS2/Na2S electrolyte eta solar cells, which is comparable to those of the N3-sensitized ZnO NW solar cells. Synthesis and characterization of the ZnO/CuInS2 core-shell NW arrays as well as the photovoltaic measurements of the ZnO/CuInS2 eta solar cells will be reported in the presentation in detail.


N9.6
Microstructure and Local Electronic Transport Properties of nc-Si:H Thin Films: Effect of Hydrogen Plasma Annealing. Pavel Dutta, Sanjoy Paul, David Galipeau and Venkat Bommisetty; EE, South Dakota State University, Brookings, South Dakota.

Improvement of the microstructural, optical and local transport properties of hydrogenated nanocrystalline silicon (nc-Si:H) is critical for applications in photovoltaics. Recent research showed that deposition conditions and post deposition plasma treatment strongly influence the surface morphology and electrical conductivity of thin films [1-3]. Present report describes the influence of hydrogen dilution, substrate temperature and post-deposition hydrogen plasma treatment on microstructural and local electronic properties of nc-Si:H thin films reactive sputter deposited on glass. The effect of post-deposition treatment with low power hydrogen plasma on surface morphology and local electronic properties will be discussed. XRD and Raman studies showed that the crystallinity in the films increases with hydrogen dilution (HD) and reaches maximum at 33%. Further increase in HD caused decrease in crystallinity. Films deposited at 200C with 33% HD produced films with highest crystallinity. Photoluminescence and FTIR results indicated that films deposited under these conditions contain largest density of Si-H bonds. Local electrical properties of nc-Si:H films were probed using current sensing-AFM (cs-AFM). Electrical conductivity of these films was low and produced low currents below the current threshold of preamplifier in cs-AFM. However, post-deposition hydrogen plasma treatment improved electrical conductivity significantly while decreasing surface RMS roughness. The effect of hydrogen plasma was most prominent during the initial few seconds of exposure to low power H2 plasma (10W, 10 sec) and there was a sharp increase in crystallinity and local electrical conductivity of nc-Si:H. Also, the improvements in crystallinity due to plasma treatment can be correlated to the hydrogen content of in the film. However, exposure to higher powered plasma and for prolonged duration showed marginal improvement in the structural and local transport properties. Results indicated that hydrogen plasma treatment can be an effective tool to engineer the structural and electrical properties of nc-Si:H 1. O.F.Vyvenko, O.Kruger and M.Kittler, Appl. Phys. Lett., 76, 6, 699 (2000). 2. K.Pangal, J.C.Sturm, S.Wagner and T.H.Buyuklimanli, J. Appl. Phys. 85, 3, 1900 (1999). 3. S. Sriraman,S. Agarwal, E.S.Aydil & D. Maroudas, Nature, Vol. 418, 62 (2002).


N9.7
Effect of Hydrogen Plasma Annealing on Optoelectronic Properties Nanocrystalline Silicon. Sanjoy Paul, P. Dutta, Dorin Cengher, David Galipeau and Venkat Bommisetty; EE, South Dakota State University, Brookings, South Dakota.

Nanocrystalline silicon (nc-Si) is an important photovoltaic material due to bandgap tunability, high optical absorption and low cost fabrication. Also, nc-Si can be deposited at low temperatures and on a variety of substrates making it potential candidate for tandem solar cell structures for efficient conversion of broadband solar spectrum. Improving optoelectronic properties requires detailed understanding of the interface structure between Si nanocrystals and surrounding amorphous tissue [1]. Deposition temperature, hydrogen dilution and plasma conditions are known to play an important role in determining nc-Si/a-Si interface [2,3]. This report presents optoelectronic properties of low temperature deposited nc-Si:H using RF sputtering and the use of post-deposition atomic hydrogen plasma treatment to improve crystallinity and optoelectronic properties. Nanocrystalline Si thin films of 100 nm thick were deposited on glass substrates using reactive sputter deposition and characterized using Raman, UV-vis spectroscopy, temperature dependent I-V, transient current measurement, transmission electron microscopy and atomic force microscopy. Hydrogen dilution impacts the Fermi level (EF) position and density of states (DOS) in the films significantly resulting change in optical and transport properties. Optical bandgap of nc-Si varied between 1.6 to 2.14 eV. Bulk defect density of all nc-Si thin films was estimated in the order of 1016 cm-3 from transient IV measurements. Electrical conductivity was about 10-5 S/cm. Arrhenius plots showed three distinct activation regions: intrinsic, extrinsic and partial ionization. Thermal activation energy calculated from high temperature region of the Arrhenius plot ranged between 0.73 and 0.88 eV. Post-deposition hydrogen plasma treatment improved crystallinity and electrical conductivity of nc-Si. Mild H2 plasma treatment (less than 1 W/cm2 plasma exposure for 10 s) at room temperature improved both structural and optoelectronic properties of nc-Si. While the effect of hydrogen plasma exposure was most significant during initial few seconds of exposure, higher plasma power and longer exposure periods showed incremental improvement in optoelectronic properties. Results suggest the potential use of hydrogen plasma during initial stages of nc-Si deposition to improve its structural and optoelectronic properties. [1] Seung Yeop Myong, Koeng Su Lim and Makoto Konagai, Appl. Phys. Lett., 88, 103120 (2006). [2] Yu. L. et. al., J. Appl. Phys., 94, 443 (2003) [3] S. Veprek, F. A. Sarott and Z. Iqbal, Phys. Rev. B36, 3344 (1987)


N9.8
Nano Si Structures for Solar Cell Application. Branko Pivac1, Pavo Dubcek1, Nikola Radic1, Hrvoje Zorc1, Sigrid Bernstorff2 and Branislav Vlahovic3; 1R. Boskovic Institute, Zagreb, Croatia; 2Sincrotrone Trieste, Trieste, Italy; 3Physics Department, North Carolina Central University, Durham, North Carolina.

One approach for silicon based next generation solar cells relies on the production of suitable Si nanostructured objects in wide bandgap material. Present research on Si nanosize structures is focused on the Si nanocrystals prepared by sputtering or physical vapor deposition of Si rich oxides and SiO2 multilayers on Si substrates. We present a study on amorphous SiO/SiO2 superlattice formation on Si substrate held at different elevated temperatures. Grazing-incidence small-angle X-ray scattering (GISAXS), X-ray reflectivity and photoluminescence were used to study such samples. From the 2D GISAXS pattern it is possible to determine the shape, size and inter-particle distance. Amorphous SiO/SiO2 superlattices were prepared by direct evaporation and/or magnetron sputtering of 2-5 nm thin films of SiO and SiO2 (10 layers each) from corresponding targets on silicon substrate. Rotation of the Si substrate during evaporation enables homogeneity of films over the whole substrate. After evaporation samples were annealed at 1100 C in different atmospheres. The analysis of the 2D GISAXS pattern has shown that Si nanocrystals are already present in the samples deposited at elevated temperatures. Using a Guinier approximation, the inter-nanocrystal distance and the thickness of the nanocrystals have been obtained. A long range ordering of nanocrystals deposited at elevated temperatures is observed.


N9.9
Abstract Withdrawn


N9.10
A New Method for Forming Surfactant-free PbSe Quantum Dot Films and Quantum Dot-polymer Composites for Excitonic Solar Cells. Gayan Dedigamuwa, Xiaomei Jiang, Jian Zhang, Pritish Mukherjee and Sarath Witanachchi; Department of Physics, University of South Florida, Tampa, Florida.

Multiple exciton generation in semiconductor quantum dots (QD) promises a new generation of solar devices that include flexible inorganic-organic hybrid structures. In these devices incorporation of PbSe or PbS quantum dots in a polymer matrix without a surfactant barrier at the QD-polymer interface is important for the dissociation of excitons and subsequent collection of carriers. We have developed a laser-assisted spray process to deposit surfactant-free crystalline PbSe nanoparticles on substrates. In the first step of the process, nanoparticles of PbSe were synthesized by the reaction between tributyl-phosphine selenium and lead oxide in the presence of Oleic acid and 1-octadecene. After washing the particles several times to remove the excess surfactant, they were dispersed in hexane. In the second step, toluene solution containing 8-10 nm size PbSe particles was atomized by a nebulizer to form an aerosol. SF6 was used as the carrier gas to inject the aerosol into the growth chamber through a nozzle. Heating of the SF6 gas by a CO2 laser beam at the nozzle caused the solvent and the surfactants to evaporate while PbSe nanoparticles were deposited on a substrate. Nanoparticles within the film grown by this method were uniformly distributed and in intimate contact with each other. TEM studies confirmed both the single crystal nature of each QD and the absence of surfactants. A second nebulizer was used to inject aerosols of P3HT polymer dissolved in toluene onto the substrate. SEM and TEM analysis of the co-deposited films showed a uniform distribution of PbSe particles within the polymer matrix. Conductivity of PbSe QD films and QD-polymer composite films have been measured in several electrode configurations. Comparison of these results with films formed with surfactant- coated QD will be presented.


N9.11
Solar Cell Efficiency Enhancement above the Shockley-Queisser Limit using Low Dimensional Absorbers. Ian Ballard, Keith Barnham and Nicholas Ekins-Daukes; EXSS Physics, Imperial College London, London, United Kingdom.

The detail-balance efficiency limit for photovoltaic solar energy conversion assumes isotropic, radiative emission as the sole means for carrier recombination. This is a good approximation for bulk semiconductors, but the valance bands in low dimensional structures can give rise to strongly directional emission, polarised in either the TE or TM direction. Typically, TE emission (perpendicular to the plane of the quantum well) couples to heavy-hole valance states, with some weaker coupling to light-hole states. TM emission couples exclusively to light-hole states. The degree of TE and TM emission can therefore be controlled by adjusting the energetic alignment of heavy and light hole bands via strain. In this way it is possible to engineer a material with intrinsic directional emission, either in TE or TM directions. In terms of detail balance efficiency limits, the highest efficiency occurs when the etendue of the incident radiation matches that of the escaping radiation. We show that this intrinsic directional emission gives rise to a new, marginally higher limiting efficiency for a single junction solar cell when low dimensional semiconductors are employed. Further we show how directional emission is useful in real, constrained devices that employ distributed Bragg reflectors to recycle photons.


N9.12
Abstract Withdrawn


N9.13
Effect of Back-Surface Reflectors on the Performance of Strain-Balanced Quantum Well Solar Cells. Jessica Adams1, Keith Barnham1, Ravin Ginige1, James Connolly1, Geoff Hill2, John Roberts2 and Tom Tibbits3; 1Department of Physics, Imperial College London, London, United Kingdom; 2EPSRC National Centre for III-V Technologies, Sheffield, United Kingdom; 3QuantaSol Ltd., Richmond upon Thames, United Kingdom.

The strain-balanced quantum well solar cell (SB-QWSC) is a GaAs p-i-n cell with multiple quantum well layers inserted into the i-region, confining carriers in the wells to discrete energy levels and extending the absorption edge to beyond that of the bulk. This results in a net efficiency increase over a conventional GaAs cell, as the loss in open-circuit voltage is over-compensated by the gain in photocurrent from the well region [1]. Recombination processes in a GaAs/InGaAs SB-QWSC have been found to be radiatively dominated at the high bias at which solar concentrator cells operate [2], and so cell efficiency can be improved by the incorporation of a reflector at the back of the n-region. For example, use of a distributed Bragg reflector (DBR) substrate can lead to an effect known as photon recycling [3], where a proportion of the radiative recombination is reflected back into the well region and reabsorbed, thus lowering the net radiative dark current. Cell efficiency is further increased as incident radiation not absorbed on the first pass is reflected back into the bulk material, increasing the short-circuit current. We present the characterisation of SB-QWSC devices grown by metal-organic vapour-phase epitaxy (MOVPE). Devices were compared from growth on doped and un-doped substrates. Different back-reflectors were investigated, including mirrored, polished, and saw-tooth etch. Furthermore, different device mountings were investigated, including gold epoxy, black paint, and an air gap between device and substrate. Dark current and quantum efficiency measurements were made on the anti-reflection coated test devices. Quantum efficiency measurements were fitted using in-house simulation software SOL, and parameters were taken from the fits to model device dark currents. A close match to experimental data was obtained. The software was then used to predict device efficiencies from the theoretical short-circuit current and open-circuit voltage. We will present details of experimental and modelling results found on these and on other devices. 1. K.W.J. Barnham et al., “Quantum Well Solar Cells and Quantum Dot Concentrators”, Chapter 16, Nanostructured Materials for Solar Energy Conversion, T.Soga (Ed.), Elsevier B.V., (2006), p.517. 2. J.P. Connolly, I.M. Ballard, K.W.J. Barnham, D.B. Bushnell, T.N.D. Tibbits, J.S. Roberts, “Efficiency Limits of Quantum Well Solar Cells”, Proc. 19th European Photovoltaic Solar Energy Conference, Paris, France, (2004), pp. 355-359. 3. D. C. Johnson et al., Appl. Phys. Letters, 90, 213505, (2007).


N9.14
Abstract Withdrawn


N9.15
Influence of Buffer Layer Patterning on Droplet Epitaxy and Photovoltaic Properties of InAs/GaAs Quantum Dots. Leon Webster1, Jia-Hung Wu2, Christopher Proctor4, Levi T Thompson3 and Rachel S Goldman2; 1Applied Physics, University of Michigan, Ann Arbor, Michigan; 2Materials Science and Engineering, University of Michigan, Ann Arbor, Michigan; 3Chemical Engineering, University of Michigan, Ann Arbor, Michigan; 4Physics, University of Michigan, Ann Arbor, Michigan.

Low-dimensional systems exhibiting quantum confinement have been predicted to lead to high efficiency intermediate-band solar cells (IBSCs). For example, highly ordered arrays of quantum dots (QDs) are expected to enable the formation of mini-bands of intermediate energy between the valence and conduction bands of the host matrix. In principle, this would allow two-photon carrier excitation processes, followed by resonant carrier transport between the QDs, thereby lowering cell series resistance and improving efficiency [1]. Although a number of groups have attempted to experimentally demonstrate high efficiency IBSCs based upon arrays of QDs, the performance to date has been limited, likely by the relatively large QD sizes and low QD densities resulting from the strain-driven Stranski-Krastanow (SK) QD growth process. An alternative approach to achievement of ultra-high QD densities is droplet epitaxy, which involves synthesis of In nanodroplets, followed by conversion to InAs QDs via exposure to an arsenic flux [2]. In this case, the QD size and density is determined primarily by the surface droplet size and density, rather than by the QD/matrix strain. Thus, surface patterning, prior to droplet epitaxy, provides a promising means for significant increases (decreases) in QD densities (sizes), while maintaining a specific spatial arrangement of QDs. For this purpose, we are exploring the influence of GaAs buffer layer patterning [3] on droplet epitaxy of InAs QDs. Our QDs consist of 2, 4, and 6 ML equivalent of In deposited on GaAs and subsequently exposed to As2 at 150°C, followed by annealing at 400°C to form InAs. Atomic force microscopy reveals an increase (decrease) in QD diameter (areal density) with increasing In exposure. Prior to QD formation, GaAs buffer layers were grown under several different conditions, involving a variety of growth rates, substrate temperatures, V/III ratios, and/or annealing sequences. For buffers grown at 580°C and As2/Ga = 13, 1-2 ML height step-terraces elongated along [-110] are apparent. Interestingly, QDs preferentially nucleate at the step-edges, similar to the case of SK QDs reported earlier [3]. We will discuss the effects of various buffer layer patterns on the QD sizes, densities, and spatial arrangements. Similar studies using focused-ion-beam and laser pre-patterned surfaces will be presented. Finally, the effects of buffer layer patterning on the photovoltaic properties of p-i-n structures incorporating droplet epitaxy QDs as the active region will be discussed. This work was supported in part by DOE and AFOSR. [1] A. Luque and A. Martí, Phys. Rev. Lett. 78, 5014 (1997) [2] S. Ohkouchi, Y. Nakamura, H. Nakamura, and K. Asakawa, Thin Solid Films, 464, 233 (2004). [3] W. Ye, S. Hanson, M. Reason, X. Weng, and R. S. Goldman, Journal of Vacuum Science & Technology B, 23, 1736 (2005).


SESSION N10: Epitaxial Quantum Dot and Quantum Well Solar Cells
Wednesday Morning, December 3, 2008
Republic B (Sheraton)

8:30 AM *N10.1
Efficiency Improvements in Triple-Junction Solar Cells using Quantum Materials. Simon Fafard, Cyrium Technologies Inc., Ottawa, Ontario, Canada.

High efficiency photovoltaic (PV) solar cells have been the topic of interesting research for several years [1], and recently the focus of major investments in the sector of renewable energy as the PV industry is growing at a phenomenonal rate. Triple-junction (3J) III-V cells grown on Ge substrates have reached a high level of perfection and maturity and have been commercialized using an active bottom Ge subcell, a lattice-matched (In)GaAs middle subcell, and a lattice-matched InGaP top cell [2]. The traditional market of these 3J cells has been for space power applications, but there are now over 30 companies developing concentrated PV (CPV) modules to pursue a competitive approach to economical solar energy. It is also well-known that the lattice-matched approach cannot provide the optimum combination of III-V semiconductors to optimize the efficiency of 3J CPV cells. Furthermore, high efficiency and low cost must be achieved simultaneously in order for the CPV market to take off. Advancements in the use of Metamorphic layers have lead to interesting devices with promising performance, but for the CPV market, trading cost and complexity to add marginal performance is not necessarily an attractive value proposition. For example the growth of 3J metamorphic cells requires the addition of epi-layers having significant thicknesses which would impact the cost in production. An attractive alternative that can be used to better optimize the design of multi-junction solar cells is the use of nanostructure-based materials to engineer semiconductor alloys with the desired effective bandgap while keeping the layers defect-free on the substrate's lattice constant. In this paper, I will discuss Cyrium's approach to improve the design and the efficiency of multi-junction solar cells and present the results obtained with triple-junction incorporating such quantum dot material in the middle subcell of devices optimized for CPV applications. For example, the bandgap combination of 0.7eV/1.2eV/1.7eV has one of the best thermodynamic efficiency limit for a 3J with a maximum potential of 59% [2]. We have demonstrated that the extension of the middle cell absorption to longer wavelengths can be achieved by adjusting the size, the shape, the composition of the quantum material and thus add over 10% of current in the middle cell by adding multiple layers of quantum dot material and thus better balance the device's current-matching without compromises on the cost and manufacturability. Different aspects and critical parameters of interest for cost-effective high-efficiency CPV cells will also be discussed. [1] J.M. Olson, S.R. Kurtz, A.E. Kibbler, and P. Faine, A 27.3% efficient Ga0.5In0.5P/GaAs tandem solar cell, Appl. Phys. Lett. 56, 623 (1990). [2] See for example the review and references in: F. Dimroth and S. Kurtz, High-Efficiency Multijunction Solar Cells, MRS BULLETIN 32, p.230, MARCH 2007.


9:00 AM N10.2
Quantum Dot Spectral Tuning of Multijunction III-V Solar Cells. Ryne Raffaelle, Seth Hubbard, Chris Bailey, Stephen Polly and Cory Cress; NanoPower Research Labs, Rochester Institute of Technology, Rochester, New York.

Enhancing the production of photocurrent in the middle junction of a InGaP/Ga(In)As/Ge triple-junction solar cells (TJSC) can directly lead to an increased overall solar conversion efficiency. This can potentially be accomplished by inserting a quantum dot (QD) superlattice (SL) into the middle junction cell. It has been predicted, using detailed balance modeling, that QD-SL enhanced TJSCs have an efficiency ceiling of 47% under a one-sun AM0 illumination spectrum. Regarding single junction devices, QD-SL enhanced GaAs cells have the added benefit of possible intermediate band effects, anisotropic absorption and enhanced radiation tolerance. Recently, the incorporation of InAs quantum dots QDs within a single junction GaAs solar has been shown to improve the short circuit current (Jsc) of single junction cells under simulated 1 sun air mass zero (AM0) illumination. In this demonstration, GaP strain compenstation layers were grown between each successive QD layer to balance the strain resulting from the lattice mismatch between the bulk GaAs and the InAs QDs. The incorporation of these layers were effective at mitigating the formation of defects and threading dislocations which reduce the minority carrier lifetime and resulted reduced Jsc and open circuit voltage (Voc) in devices grown without SC. The effect of GaP strain compensation layer thickness on the solar cell electrical and material properties was investigated theoretically and experimentally verified using high resolution X-ray diffraction (HRXRD). The optimal strain compensation layer thickness was then used to increase the QD-SL stacking from 5X to 10X and 20X. The material quality improvement in the SC QD-SL solar cells was manifested improved sub-GaAs bandgap spectral response and Jsc. The GaAs cell with optimized SC layer thickness was then utilized as the GaAs cell in a multijunction (MJ) device. The optoelectronic performance of these MJ solar cells were characterized in reference to baseline devices which did not contain QDs. Additionally, the effects of thermal cycling and alpha-particle irradiation were investigated to understand their applicability for space deployment. Thermal cycling consisted of repeated temperature cycling between 100 and 400 K with concomitant AM0 illuminated current-voltage (J-V) and quantum efficiency (QE) characterization. Similarly, the effects of alpha-particle irradiation on the J-V and QE response were measured with increased alpha-particle dose up to 10^12 alpha-particles/cm2. These results will be discussed in the context of improved photovoltaic performance with a particular focus on their potential utilization in space power systems.


9:15 AM *N10.3
Radiative Recombination in Strain-Balanced Quantum Well Concnetrator Cells in Single and Multi-Junction Configurations. Keith Barnham1, L. M Ballard1, B. C Browne1, J. C Connolly1, M. F Fuhrer1, R. Ginige1, G. Hill2, A. Ioannides1, D. C Johnson1, M. Mazzer1, J. S Roberts2 and T. D Tibbits3; 1Physics Department, Imperial College London, London, United Kingdom; 2EPSRC III-V National Centre for III-V Technologies, Sheffield, United Kingdom; 3QuantaSol Ltd., Richmond upon Thames, United Kingdom.

We discuss how the spectral range of GaAs cells can be extended by GaAsP/InGaAs strain-balanced quantum wells without introducing dislocations. Such a cell has achieved above 27% single junction efficiency in the AM1.5 low-AOD spectrum at 500 suns concentration. At these concentrations the dark current demonstrates ideality factor n = 1 behaviour, and we have demonstrated that radiative recombination is the dominant loss mechanism. Novel features of the quantum well geometry make it possible to demonstrate dark-current reduction and efficiency enhancement by photon recycling. We will present new results indicating that radiative recombination is suppressed in deep well samples, and discuss the advantages of such deep wells in multi-junction structures.


SESSION N11: Hot Carrier and Epitaxial Quantum Dot Solar Cells
Wednesday Morning, December 3, 2008
Republic B (Sheraton)

10:30 AM *N11.1
Tapping the Thermal Energy: is a Hot Carrier Solar Cell Possible? Jean-Francois Guillemoles, IRDEP, CNRS, Chatou, France.

Even in an efficient solar cell, more than 30% of the incoming solar radiation is lost in the form of heat: the excess energy an average electron hole pair gets from the incoming photons is dissipated in less than a ps into lattice vibrations. Looking into the details of this dissipation process, it will be discussed what part of this heat is recoverable as electric power. An instance of that is given by hot carriers: electron-hole pairs that do not relax to the thermal distribution, even after several tenth of ps, leaving an opportunity for energy collection. The conditions, and specially the materials requirements, for an efficient operation of hot carrier solar cells will be explored theoretically, but also based on available experimental data.


11:00 AM *N11.2
Strain-Compensated Quantum Dot Superlattice for Use in High-Efficiency Solar Cells. Yoshitaka Okada, Research Center for Advanced Science & Technology, The University of Tokyo, Meguro-ku, Tokyo, Japan.

Low-dimensional quantum dots (QDs) that are incorporated in the active region of p-i-n junction solar cells (SCs) have attracted intense research as a possible means of exploiting the below bandgap photons to generate additional photocurrents beyond that corresponding to the band-to-band transition in the conventional single-junction SCs. In QDSCs, the quantum dots are required to be uniform in size and periodically distributed in all three dimensions, which would then result in the formation of an intermediate band or a miniband instead of a multiplicity of discrete quantized levels. With the optimized bandgap energies of QDs and host materials as well as the position and width of the intermediate band, high conversion efficiencies >50% are theoretically shown with QDSCs. The experimental demonstration of QDSCs requires placing a three-dimensional (3D) QD superlattice in the active region, and for this, the most popular fabrication technique of stacking QDs is to take advantage of spontaneous self-assembly or self-organization mechanism of coherent 3D islanding during growth known as Stranski-Krastanov (S-K) growth in lattice-mismatched epitaxy. However, the S-K growth in InAs/GaAs system usually results in degradation of QD structure and generation of misfit dislocations typically after 10 layers of stacking as a result of accumulation of internal strain beyond the critical thickness. We are currently investigating GaAs-based p-i-n QDSCs with up to 20 and more stacked layers of self-assembled InAs QDs grown by atomic hydrogen-assisted molecular beam epitaxy. Compensating for the compressive strain induced by InAs QDs with a spacer layer that produces an opposite tensile strain such as GaNAs dilute nitride seems to work very well to obtain improved size uniformity, a redshift of absorption edge, and to avoid generation of defects and dislocations. The total QD density of our current QDSC with 20 InAs QD stacks is on the order of 1012 cm-2, and the short-circuit current density is 21.1 mA/cm2. The filtered current above GaAs bandedge of 2.5mA/cm2 is four times higher than that for a strained QDSC without compensation with the identical cell structure.


11:30 AM N11.3
Behavior of Quantum Dot Enhanced Solar Cells under Solar Concentration. Seth M Hubbard, David Forbes, Ryne Raffaelle, Christopher Bailey, Stephen Polly and Cory Cress; NanoPower Research Laboratory, Department of Physics, Rochester Institute of Technology, Rochester, New York.

High efficiency III-V photovoltaics (PV) are gaining popularity in high concentration (100X and above) PV (HCPV) applications. This is due to the cost per Watt savings available by using ultra-high efficiency PV as the concentrator engine. The current state-of-the art HCPV cells (InGaP/GaAs/Ge multi-junction III-V devices) have a high degree of spectral sensitivity due to current matching requirements and materials constraints imposed by their epitaxial fabrication process. In particular, under the AM1.5d spectrum, the lattice-matched triple junction design leads to almost double the photocurrent production from the Ge bottom cell as from the top and middle cells. Due to the current matching requirement of the stacked design, much of this excess current is not extracted, thus leading to a lower overall efficiency. In order to harness this excess bottom cell current and improve overall efficiency, the middle cell bandgap must be lowered to ~1.2eV at 500X. One method to achieve this could involve insertion of low dimensional nanostructures such as quantum dot supperlattice (QD-SL) in the middle GaAs cell of a conventional triple junction concentrator cell. Bandgap engineering of the middle cell could lead to AM1.5d cell efficiency over 55% at 500X. In order to investigate the effect of quantum dot tuning of the middle cell, p-i-n GaAs concentrator solar cells both with and without a five-stack of InAs QDs inserted into the i-region were fabricated. Both the cells and InAs QDs were epitaxially grown on single crystal GaAs substrates using low-pressure organometallic vapor phase epitaxy. In order to mitigate the effects of InAs QD induced compressive strain, a GaP based strain compensation (SC) scheme was employed. The SC has previously been shown to result in improved one sun AM0 short circuit currents and reduction in defect density. In this work, the SC-QD solar cells have been evaluated at higher concentrations under simulated AM1.5d illumination. Concentration was provided by a large area pulsed solar simulator at NASA Glenn Research Center. The short circuit current of the QD cell maintained a liner relation to concentration up to 100X (1.6 A/cm2) . In addition, the short circuit current enhancement compared to the baseline (observed previously at 1X) was also maintained up to 100X concentration. Efficiency was observed to peak near 50X for both the baseline and SC-QD cells, likely due to parasitic series resistance. Additionally, as CPV are predicted to operate between 50-80 C°, the effects of elevated temperature operation on the SC-QD cell were measured and compared to the baseline. Finally, reports have indicated that light doping of the QD region may improve the current contributed from QD absorption at higher concentration. The optoelectronic results on SC-QD cells with Te delta-doping in the QD region have been grown and light IV, spectral response and concentration curves of these cells will be presented.


11:45 AM N11.4
Atomistic Tight Binding Study of Interband Light Transitions in Self-assembled InAs/GaAs Quantum Dots. Hoon Ryu1,3, Muhammad Usman1,3, Shaikh Ahmed2 and Gerhard Klimeck1,3; 1ECE, Purdue University, West Lafayette, Indiana; 2ECE, Southern Illinois University, Carbondale, Illinois; 3Network for Computational Nanotechnology, West Lafayette, Indiana.

III-V compound semiconductor (GaAs/InAs) quantum dots (QDs) have been extensively studied due to their potential applications in the optoelectronic devices such as Infrared (IR) photo-detectors and lasers at the long wavelengths (1.3-1.5 um). In the earlier times, QDs were expected to achieve a large optical gain at a small carrier density due to the strong quantum confinements arising from the 0-D density of states which prevents leakage of carriers. However, the experimental results demonstrated that some QDs exhibit very small gain, even leading to negligible inter-band ground state optical transitions [1]. While the previous theoretical studies using k.p method indicated that the built-in piezoelectric field and the geometry of QDs can reduce the gain significantly due to a lower overlap between the electron and hole wave functions [2,3], more sophisticated band models which incorporate the atomistic symmetry and interface anisotropy need to be employed for a correct understanding of these systems. Continuum methods such as effective mass or k.p neither guarantee the accuracy of quantized states, nor capture the non-degeneracy and optical polarization anisotropy for relatively large devices as they ignore the crystal symmetry and atomistic interface roughness [4]. The NanoElectronic MOdeling tool NEMO 3-D [5], which was built to handle systems of realistic size at the atomistic level, has shown its capability to simulate the strain in systems of up to 64 million atoms with the atomistic valence force field method. The electronic structure is typically computed with the semi-empirical sp3d5s* tight binding model that has demonstrated its correctness for the systems of up to 52 million atoms. The piezoelectric field arising from the off-diagonal components of the strain tensor is included in the Hamiltonian as an external potential by solving the Poisson equation on the zincblende lattice. This work uses NEMO 3-D to analyze the effect of the strain, piezoelectric field and QD geometry on the interband light transitions between the electron and hole energy levels in the InAs/GaAs pyramidal quantum dots. Experimental topologies are simulated and results are compared with the experiment and previous theoretical observations. We show that the strain and piezoelectric field reduces the magnitude of overlap integral by the elongation of wave-functions. This results in the weak optical matrix elements. Optical matrix element dependence on QD size is explained quantitatively. Our results clarify the physical picture and provide a qualitative and quantitative analysis to the experimentalists for the future design of laser devices based on InAs/GaAs QDs. [1] L. Harris et al., Appl. Phys. Lett. 75, 3512 (1999) [2] O. Stier et al., Phys. Rev. B 59, 5688 (1999) [3] A.D. Andreev et al., Appl. Phys. Lett. 87, 213106 (2005) [4] G. Bester et al., Phys. Rev. B 71, 045318 (2005) [5] G. Klimeck et al., IEEE Trans. Electron Devices 54, 2090 (2007)


SESSION N12: Luminescent Solar Concentrators
Wednesday Afternoon, December 3, 2008
Republic B (Sheraton)

1:30 PM *N12.1
Nanocrystal Doped Luminescent Solar Concentrators. Amanda J. Chatten1, Rahul Bose1, Daniel J Farrell1, Mauro Pravettoni1,4, Andreas Buechtemann2, Jana Quilitz2, James H Nelson3, Liberato Manna5 and Keith W. J Barnham1; 1Department of Physics, Imperial College London, London, United Kingdom; 2Fraunhofer-Institute for Applied Polymer Research, Potsdam, Germany; 3Department of Chemistry, University of California, Berkeley, Berkeley, California; 4Renewable Energy Unit, Joint Research Centre of the European Commission, Ispra, Italy; 5NNL-National Nanotechnology Laboratory of CNR-INFM and IIT Research Unit, Lecce, Italy.

The properties of nanocrystals can be exploited in the luminescent solar concentrator (LSC). LSCs concentrate light in addition to reducing spectral losses, and generally consist of transparent polymer sheets doped with luminescent species. Sunlight incident on the top surface is first absorbed by the luminescent species and is then re-radiated, ideally with high luminescence quantum efficiency (LQE), such that a fraction of the emitted light is trapped in the sheet and can be converted by a solar cell at the edge. Advantages over conventional geometric concentrators include that (i) the LSC reduces thermalization losses and heat dissipation in an attached solar cell by converting the incident spectrum to improve the match with the absorption spectrum of the cell, (ii) both direct and diffuse radiation can be collected, (iii) expensive solar tracking is not required and, (iv) LSCs are ideally suited to building integration via façades or active windows. In addition a stack of sheets can separate the light in order to maximize the power conversion efficiency. However, the development of this promising concentrator was initially limited by the performance of the organic dyes first utilized as the luminescent species. Particular problems were their narrow-band absorption spectra and poor stability under solar irradiation. We are currently evaluating the performance of nanocrystals (quantum dots and nanorods) as the luminescent species in the LSC. Nanocrystals have advantages over organic dyes in that (i) their absorption spectra are far broader, extending into the UV, (ii) their absorption properties may be tuned simply by the choice of nanocrystal size, and (iii) they are inherently more stable than organic dyes. In addition nanocrystals can provide high LQEs (LQE ~ 0.8 has been reported for both core-shell quantum dots and nanorods). Nanorods look particularly advantageous for this application as they provide a large effective red-shift thereby minimizing re-absorption losses. Furthermore they can be aligned and may also show directional emission. We have developed self-consistent 3D thermodynamic models for planar LSCs, modules and stacks which show excellent agreement with experiments on test devices. Although this approach is very fundamental (requiring a minimum of input data) and quick to run, the method is restricted to block shaped slabs containing a single homogeneously doped luminescent species. Therefore we have also developed phenomenological Monte-Carlo models based on the principles of ray-tracing. Comparison with measurements on small test slabs, modules and stacks show that both our modeling approaches can predict the room temperature red-shift in addition to the total flux escaping each surface, providing tools for optimization of the LSC. Results on test LSCs and predictions for large area devices will be presented in order to illustrate the advantages of employing nanocrystals as the luminescent species in the LSC.


2:00 PM *N12.2
High Efficiency Organic Solar Concentrators. Marc Baldo, Michael Currie, Jonathan Mapel, Timothy Heidel, Shalom Goffri and Carlijn Mulder; MIT, Cambridge, Massachusetts.

Organic solar concentrators are a class of luminescent solar concentrators that exploit advances in thin film organic semiconductor technology and low cost manufacturing processes to create efficient, large area optical concentrators for inorganic solar cells. We report single and tandem waveguide organic solar concentrators with quantum efficiencies exceeding 50% and projected power efficiencies up to 6.8%. Near field energy transfer, solid state solvation, and phosphorescence are employed within a thin film organic coating on glass to substantially reduce self absorption losses, enabling flux gains exceeding F = 10, meaning that a photovoltaic cell attached to the concentrator generates approximately 10 × the power of the photovoltaic cell without optical concentration. Flux gains of F > 10 in organic solar concentrators should enable the economical use of high performance photovoltaic cells in low cost systems.


2:30 PM N12.3
Reducing Re-absorption Losses in Luminescent Solar Concentrator with Nanorods. Rahul Bose1, Daniel J Farrell1, Amanda J Chatten1, Andreas Buechtemann2, Jana Quilitz2, Angela Fiore3, Liberato Manna3 and Keith W Barnham1; 1Department of Physics, Imperial College London, London, United Kingdom; 2Fraunhofer Institute for Applied Polymer Research, Potsdam, Germany; 3NNL-National Nanotechnology Laboratory of CNR-INFM and IIT Research Unit, Lecce, Italy.

Luminescent solar concentrators (LSCs) are flat plate concentrators that are well suited for building integrated photovoltaics. Sunlight is collected over the large front surface area of the LSC and emitted out of the thin edges at higher intensity, where photovoltaic (PV) cells can be attached. The LSC is inexpensive compared to the PV cells attached to it and is intended to reduce the cost of PV energy by making the cells more cost-effective. The variant of the LSC that appears most practical consists of a thin active layer on a transparent substrate, such as glass, of matching refractive index. Luminescent centres in the layer absorb the incident light and re-emit it. A portion of the luminescence is waveguided to the edges of the plate through total internal reflections, based on the refractive index difference at the interfaces to air, while some is lost out of escape cones. Concentration is achieved due to the large geometrical ratio between front and edge surfaces. Besides concentrating, the LSC also converts the absorbed spectrum into a relatively narrow emission peak, which can be matched to the PV cells absorption edge. The LSC is a static concentrator, in that it does not require solar tracking. It also collects both direct and diffuse irradiation and can be designed semi-transparent. These aspects make the LSC attractive for building integrated application. By using the LSC as a substitute for building components one could even offset some of the costs. One of the main challenges is the reduction of losses in the LSC in order to achieve higher concentrations. The predominant loss mechanisms are the non-radiative dissipation of photons absorbed by luminescent centres, which is governed by their luminescence quantum efficiency (LQE), and the loss of luminescence out of escape cones. These two kinds of losses occur each time luminescence is re-absorbed by the centres. Therefore, re-absorption poses a major problem. The conventional LSC uses dyes as luminescent centres. We are investigating inorganic nanocrystals as alternatives. These offer advantages, such as a broad absorption spectrum, spectral properties that can be tuned by the choice of size and photo-stability. Nanorods have additional features that could be utilised. A small overlap between their absorption and emission spectra means that re-absorptions and the related losses could be reduced. Moreover, they are expected to exhibit directional emission if they are aligned. We have developed a raytrace model that has been tested against experiments and other models and has proven useful for optimising the LSC. The model is based on geometrical optics and uses a Monte Carlo method. Small (5cm x 5cm) nanorod LSCs have been successfully fabricated with LQEs around 70%. We will present the characterisation of these nanorod LSCs and large area raytrace predictions.


SESSION N13: Nanowire Solar Cells
Wednesday Afternoon, December 3, 2008
Republic B (Sheraton)

3:15 PM *N13.1
Radial pn Junction Solar Cells, Past, Present, and Future. Brendan M. Kayes1, Harry A Atwater1 and Nathan S Lewis2; 1Applied Physics, California Institute of Technology, Pasadena, California; 2Chemistry, California Institute of Technology, Pasadena, California.

The key challenge facing the widespread adoption of photovoltaics (PVs) for terrestrial electricity production is reducing the cost per kilowatt hour of energy produced. Highly efficient PV devices currently require highly pure materials and expensive processing techniques, while low cost devices generally operate at relatively low efficiency. We proposed the radial pn junction, wire array geometry as a potential route to reducing materials costs while maintaining relatively high efficiencies, by allowing for efficient carrier collection despite low minority carrier diffusion lengths [1]. In the radial pn junction design the junction area is increased in order to decrease the path length any photogenerated minority carrier must travel to be less than its minority carrier diffusion length. Realizing this geometry in an array of semiconducting wires, by for example depositing a single-crystalline inorganic semiconducting absorber layer at high deposition rates from the gas phase by the vapor-liquid-solid (VLS) mechanism, allows for a “bottom up” approach to device fabrication, which can in principle dramatically reduce the materials costs associated with a cell. This talk will provide an overview of the field of radial pn junction solar cells as it stands today, and, in the light of previous results, in which directions we believe future opportunities might lie. First we will outline the history of, and motivation for, attempts to improve device efficiency by increasing pn junction area. Specifically, the Vertical Multijunction (VMJ) solar cell concept proposed by Wise [2], and the Parallel Multijunction (PMJ) cell of Green and Wenham [3] will be examined. We will discuss the similarities and differences between these designs and the radial pn junction wire array concept. Next we will outline and attempt to organize recent experimental realizations of the radial pn junction concept (although the first realization of this concept at the single wire level, to our knowledge, was in fact some time ago [4]). This will include discussion of the similarities and differences in the approaches taken and the results achieved, as well as of analogous systems and what we can learn from them. We will examine what is currently the “state of the art” and how this might be improved upon. Finally we will discuss recent innovations that point to the potential for low cost fabrication of wire-based devices. [1] B. M. Kayes, C. E. Richardson, N. S. Lewis, and H. A. Atwater, Conference Record of the Thirty-first IEEE Photovoltaic Specialists Conference 1, 55 (2005) [2] W. P. Rahilly, Conference Record of the Ninth IEEE Photovoltaic Specialists Conference 1, 44 (1972) [3] M. A. Green and S. R. Wenham, App. Phys. Lett. 65, 2907 (1994) [4] A. A. Shchetinin, A. I. Drozhzhin, N. K. Sedykh, and E. P. Novokreshchenova, Measurement Techniques 21, 502 (1978)


3:45 PM N13.2
Single and Tandem Axial p-i-n Nanowire Photovoltaic Devices. Thomas J Kempa1, Bozhi Tian1, Xiaolin Zheng2 and Charles M Lieber1,3; 1Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts; 2Mechanical Engineering, Stanford University, Stanford, California; 3School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts.

Nanowires represent a promising class of materials for exploring new concepts in solar energy conversion. Here we report the first experimental realization of axial modulation-doped p-i-n and tandem p-i-n+-p+-i-n silicon nanowire (SiNW) photovoltaic elements. SEM imaging of selectively etched nanowires demonstrates excellent synthetic control over doping and lengths of distinct regions in the diode structures. Current-voltage (I-V) characteristics reveal clear and reproducible diode characteristics for the p-i-n and p-n SiNW devices. Under simulated one-sun solar conditions (AM 1.5G), optimized p-i-n devices exhibited a Voc of 0.29 V, a maximum apparent Jsc of 3.53 mA/cm2, and efficiency of 0.52%. The response of Isc vs. Voc under varying illumination intensities shows that the diode quality factor is improved from n=1.78 (p-n) to n=1.28 (p-i-n i_length=4µm) by insertion of the i-type SiNW segment. The temperature dependence of Voc scales as -2.97 mV/K and extrapolates to the crystalline Si bandgap at 0 K, in excellent agreement with bulk properties. Finally, we demonstrate a novel tandem SiNW solar cell through series integration of two PV elements (p-i-n+-p+-i-n) and show that its Voc is on average 55% larger than the equivalent single p-i-n device. Fundamental studies of such well-defined nanowire photovoltaics will enable intrinsic limits of nanoscale elements to be defined.


4:00 PM N13.3
p-Si and p-InP Nanowire Array-Based Liquid-Junction Solar Cells. Rui Xiao1, Sarah Eichfeld2, Yoji Kobayashi1, Robin Woo3, Li Gao3, Robert Hicks3, Joan Redwing2, Theresa Mayer4 and Thomas Mallouk1; 1Department of Chemistry, The Pennsylvania State University, University Park, Pennsylvania; 2Materials Science and Engineering, The Pennsylvania State University, University Park, Pennsylvania; 3Department of Chemical and Biomolecular Engineering, University of California, Los Angeles, Los Angeles, California; 4Electrical Engineering, The Pennsylvania State University, University Park, Pennsylvania.

Nanowire arrays represent an interesting technological alternative to semiconductor thin film-based solar cells. Vertical nanowire arrays decouple the dimensions of light absorption and minority charge carrier collection, which occur in the axial and radial directions, respectively. High aspect ratio arrays should allow one to achieve relatively high quantum efficiencies across the spectrum, even with indirect gap semiconductors, and because of the short minority carrier collection length, they should be relatively tolerant of impurities and defects. We find by photoelectrochemistry and transmission electron microscopy that for single crystal p-InP nanowire arrays, the photovoltage is strongly dependent on the density of twin boundaries and on the nanowire diameter. For p-InP nanowires grown by metal-organic vapor phase epitaxy on Si(111), the degree of twinning can be tailored by controlling the substrate miscut angle and pre-annealing conditions. Photolithographic patterning and vapor liquid solid growth were combined to fabricate patterned single crystal Si wire arrays. The dependence of their photoelectrochemical properties on wire dimensions, inter-pore distance and doping density were studied.


4:15 PM N13.4
Characteristics of Radial p-n Junction Nanowire Solar Cells Grown by Vapor-liquid-solid Technique. Oki Gunawan and Supratik Guha; IBM Research, Yorktown Heights, New York.

Dense, semiconductor nanowire forest based solar cells are considered attractive because of the potential for higher light absorption characteristics as well as the ability to grow core-shell type radial p-n junctions. We report detailed photovoltaic and microstructural characterization of solar cells made of gold catalyzed silicon nanowires (NWs) with a core-shell radial p-n junction architecture, grown by the vapor-liquid-solid technique. The results of our studies show that nanowire based devices can result in enhanced efficiencies due to better light trapping characteristics. However, by varying the amount of Au catalyst used, we provide compelling evidence of the deleterious role of the residual Au impurities in acting as recombination centers in the wires. The diode characteristics indicate that the current in forward bias is dominated by generation recombination in the depletion layer and the estimated minority carrier lifetime from the electrical measurements is consistent with the expected concentration of gold impurities. Through measurement of the spectral response of the photovoltaic efficiency we show evidence for strong surface recombination effects in the nanowires that results in compromised efficiencies at short wavelengths. These results indicate that surface recombination due to Au impurities in NWs is a limiting factor for their performance. We finally show that longer and higher density NWs lead to better light trapping that increases the photocurrent, however the resulting higher surface recombination quenches the open circuit voltage. The results of this study shows that while there are clear benefits that arise from enhanced light absorption, nanowire based photovoltaic structures would benefit from crystal growth approaches that are free of gold.


4:30 PM N13.5
Ni-catalyzed Si1-xGex Nanowire Arrays for Solar Cell Applications. Sang-Won Jee, Kwang-Tae Park, Hong-Seok Seo, Jin-Young Jung, Han-Don Um, Xiaopeng Li and Jung-Ho Lee; Chemical engineering, Hanyang university, Ansan, South Korea.

There has been a significant, resurgent interest in renewable energy systems. Nanowire (NW)-based solar cells are one of the promising candidates for next-generation photovoltaic devices. They offer many advantages such as the use of low-cost substrates and low-purity sources with defect-free crystalline morphology [1]. In the case of silicon nanowires (SiNWs), Au-catalyzed vapor-liquid-solid (VLS) growth has been a main focus to date owing to a superior single crystalline morphology obtained under low temperature (<430 °C) annealing. However, Au is not compatible to the conventional semiconductor processing since it is extremely difficult to remove the Au-Si alloy catalysts at the tip of SiNWs; moreover, Au has been known to act as a deep-level impurity inside the bandgap, causing serious leakage of device current. Here, we report for the first time the vertical growth of Ni-catalyzed Si1-xGex NWs for solar cell applications. Compared to Si, SiGe can absorb a wider range of the light spectrum [2] with increasing the carrier mobility. Figure 1 shows a conceptual schematic for the solar cell operation employing the Ni-catalyzed Si1-xGex NWs array which is vertically grown from the heavily boron doped Si(111) substrate. To measure the current-voltage (I-V) characteristics, a ~200-nm-thick Al layer was thermally evaporated on the wafer back-side. Si1-xGex NWs were prepared using the atmospheric VLS method using SiCl4 and GeCl4 source gases. After growth of SiGe NWs, n-type amorphous silicon thin film was deposited to conformally coat the NWs for the fabrication of radial p-n junction inside the nanowire bulk. Strong p-type behavior of undoped-grown SiGe NWs is originated from both the boron diffusion from the heavily doped substrate and the one-dimensional hole-gas feature due to the Ge-rich core [3]. The front side of the substrate with the NWs was contacted by pressing the NW arrays onto a transparent polymer covered with transparent conductive oxide (TCO). The I-V curves under illumination were measured using a solar simulator. [1] L. Tsakalakos et al, J. Nanophoton. 1, 013552 (2007) [2] G. Ganguly, T. Ikeda, T. Nishimiya, K. Saitoh, M. Kondo, and A. Matsuda, Appl. Phys. Lett. 69, 4224 (1996) [3] T. Matsui, M. Kondo, K. Ogata, T. Ozawa, and M. Isomura, Appl. Phys. Lett. 89, 142115 (2006)


SESSION N14: Poster Session II
Wednesday Evening, December 3, 2008
8:00 PM
Exhibition Hall D (Hynes)

N14.1
Branched ZnO Nanowires for Enhancing Energy Conversion Efficiency of Dye-Sensitized Solar Cells. Hsin-Ming Cheng1, Wei-Hao Chiu1, Chia-Hua Lee2, Song-Yeu Tsai2 and Wen-Feng Hsieh1; 1Department of Photonics and Institute of Electro-Optical Engineering, National Chiao Tung University, Hsinchu, Taiwan; 2Dye-Sensitized Solar Cells, Photovoltaics Technology Center, Industrial Technology Research Institute, Hsinchu, Taiwan.

The branched ZnO nanowires have been fabricated on FTO substrates using a solvothermal method for DSCs. To elucidate the nanomorphology, field emission scanning electron microscope (FESEM), field emission transmission electron microscope (FETEM), X-ray diffraction, and Raman spectroscopy are applied. The incident monochromatic photon to current conversion efficiency (IPCE) obtained for the branched ZnO nanowire DSCs was almost 1.5 times that of the bare ZnO nanowire. Furthermore, the short-circuit current density and the overall light conversion efficiency of the branched ZnO nanowire DSCs were almost twice higher than the bare ZnO nanowire ones. The improvement can be explained association with the enlargement of internal surface area within the photoelectrode without increasing interparticle hops. In addition, the sufficient dye-loading in branched ZnO nanowire DSCs was further evidenced from the lower series resistance Rs=(dV/ dI)I=0 and significantly enhanced IPCE spectra as compared with the bare ZnO nanowire ones. Accordingly, the concept of these one-dimensional branched nanostructures could simultaneously afford a direct conduction pathway and achieve higher dye adsorption to significantly enhance the overall energy conversion efficiency of the DSCs.


N14.2
Fabrication of Highly-Ordered TiO2 Nanotube Arrays and their Use in Dye-Sensitized Solar Cells. Tae-Sik Kang1,2, Barney E Taylor1,2, James R Deneault1,2, Hilmar Koerner1,2, Lawrence F Drummy1,3, Richard A Vaia1 and Michael F Durstock1; 1Air Force Research Laboratory, Wright Patterson AFB, Ohio; 2Universal Technology Corporation, Dayton, Ohio; 3UES, Inc., Dayton, Ohio.

In using nanocrystalline materials for solar cell applications, poor charge transport is one of the main problems due to a high degree of structural disorder. Vertically oriented TiO2 nanotube arrays provide an optimal material architecture for photoelectrochemical devices because of their large internal surface area, lower recombination losses, and vectorial charge transport along the nanotube axis. In this study, a process has been developed by which highly ordered TiO2 nanotube arrays were synthesized by an alumina nanotemplating method. The diameter and length of the TiO2 nanotube arrays can be controlled by varying the process parameters such as the aluminum anodization voltage and time. The TiO2 nanotube arrays were characterized using SEM, TEM, and XRD. The TiO2 nanotubes have the desired stoichiometric chemical composition and anatase crystal structure after firing at 500 oC for 30 min. Solar cells fabricated using a ruthenium dye-sensitized array of TiO2 nanotubes with various lengths and wall thickness were studied.


N14.3
Abstract Withdrawn


N14.4
Virus Templated Titania Multilayer Systems for Self-Assembled Solar Cells. Friederike Fleischhaker, Rebecca Ladewski, Forrest Liau, Shujun Chen, Naomi Jones, Angela Belcher and Paula Hammond; MIT, Cambridge, Massachusetts.

A number of challenges related to the development of new organic-inorganic photovoltaic systems remain, including the ability to enhance the materials interface and improve the control required in development of nanoscale materials. We can utilize alternating electrostatic assembly methods in combination with genetically engineered virus based templates to achieve nanostructured materials systems for dye-sensitized solar cells. Genetically-engineered bacteriophage have been designed to support the generation of a porous titania nanowire network that can be used with alternating assembly methods to obtain efficient solid-state dye-sensitized solar cells. X-ray diffraction measurements confirm that the in situ bacteriophage-to-titania conversion yields polycrystalline anatase nanowires. The resulting solid state dye-sensitized solar cells show an increase in power output (mW/cm2) compared to titania nanoparticle-based dye-sensitized solar cells under comparable assembly and testing conditions. Design, water-based assembly methods, and materials characterization of these systems will be discussed, as well as device characterization methods.


N14.5
Synthesis and Characterization of Non-compensated Chromium, Nitrogen Co-doped TiO2 Nanoparticles for Enhanced Solar Energy Utilization. Xiaofeng Qiu, Parans M Paranthaman, Hui Pan, Wenguang Zhu, Baohua Gu, Wei Wang, Gyula Eres and Zhenyu Zhang; Oak Ridge National Lab, Oak Ridge, Tennessee.

A conceptually new non-compensated co-doping of wide bandgap semiconductors has been proved as a promising strategy to fine-tune the electronic structures of semiconductor based photocatalysts. This research represents an important first step toward integrating advanced materials synthesis capabilities with fundamental understanding of materials requirements for high efficiency solar energy utilization, guided by theoretical modeling and simulations. Herein, we introduce a wet chemical approach to prepare non-compensated Cr-N co-doped TiO2 nanoparticles. Chromium (III) acetylacetonate, titanium (IV) isopropoxide, and ethylenediamine were used as starting precursors. XRD results prove that the co-doped TiO2 has an anatase structure and XPS results show that both Cr and N have been successfully incorporated into TiO2. UV-Vis absorption spectra of co-doped TiO2 exhibit significantly enhanced absorption in the visible-light region compared to either N-doped TiO2 or Cr-doped TiO2. These preliminary results show that the non-compensated co-doping may help stabilize the dopant elements within the host materials and enhance visible-light absorption of TiO2.


N14.6
Nanofibrous Titanate Coatings for Use in Flexible Dye Sensitized Solar Cells. Judith D. Sorge and Dunbar P Birnie; Materials Science and Eng., Rutgers University, Piscataway, New Jersey.

While dye sensitized solar cells (DSSC’s) have entered the commercial market, there are many variations in the production and properties of titania films that could be more efficient and/or cost effective than the current generation. High aspect ratio nanostructured titanium dioxide has been shown to enhance the efficiency of dye sensitized solar cells (DSSC’s); however, current growth methods, such as anodization of titanium metal or anodic alumina membrane templating, involve complex processes which might have expensive commercialization paths. In the current research, titanium metal is hydrothermally reacted in a basic medium, forming a fibrous coating of sodium titanate and titania which can then be placed directly into the solar cell as the photo anode. Depending upon the formulation utilized, the product morphology can be tuned to produce thick carpets with nanofiber strands. These strands have been characterized with transmission and scanning electron microscopy for composition and morphological analysis, as well as with x-ray diffraction. Cell efficiency comparisons are presented between cells made with a hydrothermally grown nanofiber coating and cells with more standard doctor bladed titania coatings. In spite of the excellent conduction pathways afforded by the crystalline nanostrands, the efficiency values for cells made with these nanofiber coatings are still rather modest. We present some model calculations that help explain the geometric limitations tied in to the growth patterns, surface area development, and space filling of the microstructures.


N14.7
Achievement of 4.7% Conversion Efficiency in ZnO dye- Sensitized Solar Cells Fabricated by Spray Deposition Using Hydrothermally Synthesized Nanoparticles. RangaRao Arnepalli and Viresh Dutta; Centre for Energy Studies, Indian Institute of Technology, Delhi, New Delhi, Delhi, India.

Abstract ZnO based Dye-sensitized solar cells (DSSCs) have been fabricated using hydrothermally synthesized ZnO nanoparticles by spray deposition. The effect of self-assembled nanostructures in ZnO photoelectrode, due to electric field during spray deposition, has been studied. Thickness of the photoelectrode is found to play a role on the cell performance, the cell with nanocrystalline film thickness of ~4.3 μm yielding an efficiency of ~2.8% for a cell area of ~3.2 cm2. On the other hand, the cell with ZnO nanostructures is found to yield an efficiency of ~4.7% (enhancement of ~60%) which is highest for the cell with area > 1 cm2 having a photoelectrode thickness of ~4.5 μm. Increased surface area due to the presence of ZnO nanostructures in the photoelectrode film helps in the adsorption of more number of dye molecules to the ZnO surface which contributes to the better cell performance. The improved dye- sensitized solar cell performance is also explained with the help of light scattering by the ZnO nanostructures through extending the distance traveled by light so as to increase the light-harvesting efficiency of photoelectrode film.


N14.8
Effects of Anodization Parameters on Titania Nanotube Arrays and the Performance of Dye-sensitized Solar Cells. Zhibin Xie, Stefan Adams, Daniel J Blackwood and John Wang; Materials Science & Engineering, National University of Singapore, Singapore, Singapore.

Dye sensitized solar cells (DSSCs) have attracted considerable attention as potential, cost-effective alternatives to silicon-based photovoltaic (PV) devices. The efficiency of DSSCs with a liquid electrolyte can reach as high as 11%. One of the factors limiting the performance of the DSSCs is the electron collection efficiency through the mesoporous titania layer. Titania nanotube arrays (TNAs) aligned perpendicular to electron collection electrodes could enhance electron transport and reduce recombination with redox electrolytes, leading to the higher charge collection efficiency. As a consequence, titania nanotube arrays replacing the conventional nanoparticle titania thin films is expected to improve the PV performance of DSSCs. Highly ordered, closely packed, and vertically oriented titania nanotube arrays with lengths exceeding 10μm were fabricated by anodizing titanium in an electrolyte composed of ammonium fluoride, ethylene glycol and deionized water. The microstructural morphology of the TNAs were characterized by scanning electron microscopy and X-ray diffraction and were adjusted by modifying the anodization voltage and time over a wide range. After anodization the nanotubes walls are amorphous, but they can be transformed into anatase of ca. 50 nm in crystallite size by annealing in air at 450°C for three hours. Anodization voltage and time greatly influence the photovoltaic performance of dye sensitized solar cells based on the TNAs by adjusting the available surface area. The resulting efficiency enhancement is dominated by the variation of Jsc. A promising efficiency of 4.16% (Jsc 7.68mA/cm2, Voc 0.803 and FF 67.4%) under AM 1.5 100mW/cm2 illumination was achieved.


N14.9
Novel Photovoltaics Based on Direct Interfacial Charge-Transfer from Dicyanomethylene Compounds to TiO2 . Takaya Kubo, Jun-ichi Fujisawa, Toshio Nagatani, Satoshi Uchida and Hiroshi Segawa; University of Tokyo, Tokyo, Japan.

Our society has been facing serious environmental problems closely related to global warming. Under these circumstances, renewable energy is becoming increasingly important. Among various renewable energy resources, solar energy has particular advantages since it is an abundant, limitless energy resource, and solar power generation emits virtually no pollutants and/or CO2. Although Si solar cells have already been put on the market and used worldwide, next-generation solar cells based on new concepts are currently attracting wide interest. This is because of their potential for converting solar energy to electricity at costs lower than the conventional Si solar cells. The next-generation solar cells include dye-sensitized solar cells and organic thin film solar cells, whose power conversion mechanisms are different from that of the Si solar cells wherein free carriers are produced at the pn junction. In dye-sensitized solar cells, sunlight is absorbed by dyes resulting in electron injection to the TiO2 conduction band from the excited dyes thereby generating electricity. On the other hand, the power generation in organic thin film solar cells takes place by exciton dissociation at the interface between electron donor and acceptor organic materials. Recently, we have found that some dicyanomethylene compounds (referred to as TCNX hereafter) bond to the surface oxide anion of TiO2 and form interfacial charge-transfer complexes, which give a broad absorption from the visible to the near infrared region caused by a direct interfacial charge-transfer transition (DICT) from the HOMO of TCNX to the TiO2 conduction band. We have also proposed novel photovoltaics based on DICT. In this presentation, we will discuss detailed properties of TCNX/TiO2 interfacial complexes and the performance of DICT-based solar cells. The present results give a new guiding principle for exploring novel photovoltaic devices. The TCNQ/TiO2 complex was synthesized by immersing a TiO2-substrate in a TCNQ-acetonitrile solution. Absorption of the TCNQ/ TiO2 photo-electrode caused by DICT gave a broad absorption band with an absorption edge at around 800 nm. DICT-based solar cells were fabricated with the TCNQ/TiO2 photo-electrode, a Pt-coated counter electrode, and a liquid electrolyte (2 mol/L LiI and 0.025 mol/L I2 in acetonitrile). I-V characteristics of the cells were measured under AM1.5G-100mW/cm2 illumination. Open-circuit voltage, short-circuit current density, and fill-factor were 0.36 V, 9.9 mA/cm2, and 0.61, respectively, which yielded a photo-to-electricity conversion efficiency of 2.2%. The IPCE spectrum of the cell was observed in the spectral region corresponding to the absorption of the photo-electrode. The IPCE reached a maximum of 75 % at 460 nm. Other TCNX/TiO2 photo-cathode electrodes formed with TCNE, TCNAQ, etc. were also confirmed to give rise to photo-to-electricity conversion in the visible to near infrared region.


N14.10
Multiwall Anatase TiO2 Nanotubes and Their Implications in Designing Efficient Photovoltaic Cells. Hyunchul Kim, Hyunjung SHin, Changdeuck Bae, Hyunjun Yoo and Youngjin Yoon; National Research Lab for Nanotubular Structures of Oxides, Center for Materials and Processes of Self-Assembly, and School of Advanced Materials Engineering, Kookmin University, Seoul, South Korea.

We suggest a new one-dimensional (1D) photoanode, multiwall anatase TiO2 nanotubes which were prepared by deposition of alternating TiO2/Al2O3 nanolaminates inside porous alumina membranes using atomic layer deposition (ALD: Shin et al, Adv. Mater. 2004, 16, 1197. and Bae et al, Chem. Mater. 2008, 20, 756.), followed by wet chemical etching of alumina used as a sacrificial material. All of the structural parameters including the diameter, length, wall thickness, interwall spacing, and number of wall layers of the multiwall TiO2 nanotubes were adjustable in a controlled fashion. Desired numbers of nanolaminate (3~7 nm TiO2 and 3~7 nm Al2O3 layers) were deposited, and upon the etching, the inside-wall’s shift by capillary action was detected. We suspect that this would be related with the etching and drying processes of the multiwall TiO2 nanotubes with the narrow nanocapillaries in their interior. Nonetheless, the inside of multiwall nanotubes with several micrometers in length was completely, selectively dissolved via a series of capillary insertion and subsequent diffusion-out of the dissolved species. The proposed structures were found to multiply the roughness factor (defined as the total surface area for unit area) of the anatase TiO2 nanotubes by simply increasing the number of wall layers, holding the areal density of the nanotube arrays. On the other hand, the specific surface area (defined as the total surface area per unit mass) was generally decreased with the increase of both the tube wall numbers and thickness. As for an optimum 1D structure, the thinner and the more for the wall layers of the multiwall TiO2 structures, the better in terms of the roughness factor. The present hierarchical structures should be useful as improved 1D charge collectors for application in dye-sensitized solar cells, organic-inorganic hybrid solar cells, Li-ion secondary battery, and other devices.


N14.11
Porous ZnO Matrix Decorated with Quantum Dots (CdS, PbS) for Solar Cell Applications. Paromita Kundu, Bratindranath Mukherjee and Ravishankar Narayanan; Materials Research Centre, Indian Institute of Science, Bangalore, India.

Attaching quantum dots (QD) to nanoporous oxide and controlling the size of QD is important for making solar cells. Controlling the surface area of the porous structure and ensuring uniform distribution of the QDs on the surface is critical for obtaining high efficiency cells. We present here the synthesis of QD sensitized wide band gap semiconductor material for photovoltaic device. The synthesis involves combustion method for producing porous structures followed by generation of nanoscale precursors on the surface to generate nucleation centre for the formation of sulfides. The size of the QD is controlled without using any surfactant. Electron dynamics in such couples has been studied indetail. Microstructural studies are done using high resolution electron microscopy (HRTEM).


N14.12
Growth of CdS Nanocrystals within Mesoporous TIO2 Films by using Cadmium Thiolate Derivatives as Unimolecular Precursors. Anna M Laera, Maria C Ferrara, Vincenzo Resta, Emanuela Piscopiello, Saverio Mazzarelli, Anna Mevoli and Leander Tapfer; Department of Advanced Physics Technology and New Materials, ENEA, Brindisi, Italy.

Photovoltaic (PV) cells based on quantum dot (QD) sensitized nanocrystalline titania (TiO2) in the last few years has focused research interest due to the advantages offered by QDs with respect to dye molecules. The size related tunability of their optical properties, the better heterojunction formation with hole conductors and the structural stability, have been recently joined by the perspective of higher conversion efficiencies due to inverse Auger effects (impact ionization) that is not possible in organic-dye sensitized solar cells (DSSC). Therefore, the development and optimization of new simple strategies to obtain QD-sensitized TiO2 will represent a crucial step for a breakthrough in the field of efficient PV materials. The most successful methodology for QD-sensitized TiO2 preparation is the absorption of pre-synthesized QDs onto TiO2 via chemical bath deposition (CBD). The pre-synthesized QDs absorption route involves only a thin surface layer of TiO2 and, therefore, the CBD based on an in-situ synthesis of QDs by using two precursor components in the mesoporous TiO2 matrix is the most frequently used. Semiconductor nanocrystals (NCs), such as CdS, CdSe, PbS, or InP QD, dispersed in titania were prepared and investigated. In the present work we propose an novel in-situ approach to synthesize CdS NCs in mesoporous TiO2 anatase films based on the use of an unimolecular precursor, containing both the metal and the non-metal part. This process ensures the stoichiometrical control of the synthesis and promotes the homogeneous multicomponent mixing. In previous works we demonstrated the suitability of cadmium thiolate as unimolecular precursors for the synthesis of nanometric metal sulphide even if they usually have a polymeric structure leading to low volatility and insolubility in typical organic solvents. These difficulties can be overcame by the preparation of new soluble adducts [Cd(SR)2]Ln, where L represents 1-methylimidazole, maximizing the absorption in the TiO2 matrix pores. The thiolate complex absorbed in TiO2 can be decomposed at 500°C in argon atmosphere in order to obtain the final CdS/TiO2 composite material. The presence of CdS NCs has been monitored by UV-Vis transmission spectroscopy carried out with non polarized light and at normal incidence with respect to the sample surface. The composites optical properties have been investigated by photoluminescence (PL) and PL excitation (PLE) spectroscopy. X-ray diffraction (XRD) measurements and transmission electron microscopy (TEM) have been performed in order to study the crystalline structure of the samples and to estimate the TiO2 and CdS NC size. The porosity of the TiO2 matrix and the composite material morphology were investigated by scanning electron microscopy (SEM). The proposed synthesis strategy allows to fabricate CdS/TiO2 composite material for PV applications, with an average size of d=10nm-100nm for TiO2 anatase crystallites and of d=1nm-5nm for CdS NCs.


N14.13
Self-Assembled Templates on Indium Tin Oxide Surfaces: Towards Nano-Structured Organic Photovoltaics. Travis L. Benanti1, Wook Jun Nam1, Zachary Gray1 and Stephen J Fonash1,2; 1Solarity, State College, Pennsylvania; 2Engineering Science and Mechanics, The Pennsylvania State University, University Park, Pennsylvania.

Efficient organic photovoltaic devices (OPVs) possess donor (D) and acceptor (A) phases structured on the nano-scale such that excitons reach a D-A interface within their finite lifetimes. Furthermore, the interpenetrating network of D and A phases must be bicontinuous in order to provide pathways for electrons and holes to reach the electrodes without becoming trapped. Current procedures for achieving this desired structure require the use of thermal annealing (e.g., P3HT:PCBM) or ultra high vacuum co-evaporation of D and A materials (e.g., CuPc:C60). Thermal annealing techniques, while simple in the laboratory, are challenging to optimize on a large scale. Ultra high vacuum evaporation of organic molecules is not yet a scalable, cost-effective process. A better way to control the structure of an OPV is to build the device directly on a nano-structured surface. If the surface has high aspect ratio features on the order of the exciton diffusion length, then conformal coating of the nano-structure fulfills the requirements for an interpenetrating, bicontinuous network -- even for layer by layer deposition of D and A. Indium tin oxide (ITO) is commonly employed as the top electrode in organic photovoltaic devices. So, a first step towards achieving the goal of nano-structured OPVs is to develop a cost-effective process for patterning ITO surfaces with nano-scale features. One viable method relies on the self assembling properties of block copolymers, such as poly(styrene-b-methyl methacrylate) (PS-b-PMMA). In this presentation, we describe our efforts: (i) to assemble PS-b-PMMA on ITO substrates; (ii) to develop alternative wet and dry etching techniques for removing the PMMA component. Commercially available ITO substrates are rough and multi-crystalline. The roughness of the surface tends to hinder self-assembly of thin (40-60 nm) PS-b-PMMA films. HBr plasma etching of ITO reduces its roughness and provides an ideal surface for block copolymer assembly. Controlling the orientation of PS-b-PMMA domains relative to the ITO substrate requires neutralization of the interactions between the polymer and the surface. We have found that effective neutralization of ITO -- with PS-r-PMMA brushes or with suitable silanes -- requires a high concentration of hydroxyl groups on the surface. This is effectively achieved by first soaking commercially available ITO slides in aqueous hydrogen peroxide and then irradiating them with an ozone-producing UV light source. Removal of the PMMA component from the self-assembled film is commonly done by irradiating the film with UV light and rinsing away the degraded PMMA blocks. A useful, dry alternative is to toughen the PS by exposing the film to RuO4 vapors. Then, oxygen plasma etching of the film effectively removes the PMMA, leaving behind the nano-porous PS template. The resulting templates extend over large areas of ITO and afford nano-structured surfaces upon which to build efficient OPVs.


N14.14
Achieving Efficient Organic Photovoltaic Devices with Aligned Crystalline Organic Nanorods. Ying Zheng, Robel Bekele and Jiangeng Xue; Materials Science and Engineering, University of Florida, Gainesville, Florida.

Donor-acceptor (D-A) bulk heterojunctions with percolated nanostructures are highly desirable for efficient organic photovoltaic (OPV) devices due to their ability to achieve both good exciton diffusion (or charge generation) and charge collection. There have been many attempts in recent years to realize such nanostructures. One often discussed route was to create a nanostructured template composed of one material (D or A) and then infiltrate the gaps of that template with a second material, leading to an interdigitated nanostructure. However the success has been rather limited due to the challenges of achieving versatile control on the structure of the template with a low cost method and good infiltration of the second material into the nanostructured template. Here, we demonstrate a simple way to achieve interdigitated nanostructured D-A bulk heterojunction. First, aligned crystalline copper phthalocyanine (CuPc) nanorods were grown by tilting the substrate with respect to the molecular beam during the vacuum deposition, the so-called oblique angle deposition (OAD). The scanning electron microscope (SEM) images show nanopillars with 10-30 nm in diameter and 50 to 130 nm in height grown on top of indium tin oxide (ITO) substrates. Morphology of the nanostructured CuPc layers can be manipulated by varying the oblique deposition angle as well as the initial nucleation conditions. The bulk heterojunction was then completed by spin coating a [6,6]-phenyl-C61-butyric acid methyl ester (PCBM) layer onto the CuPc nanostructured film. Cross-section SEM images indicate that good infiltration of PCBM into the nanostructured CuPc was achieved by controlling the PCBM solution concentration and spin coating conditions. Finally, photovoltaic performance of the nanostructured CuPc/PCBM device is investigated. A more than two-fold increase of the power conversion efficiency (PCE) is achieved with the nanostructured CuPc/PCBM device compared with the planar CuPc/PCBM device. Further improvement of the PCE is realized by optimizing the annealing temperature of the device, which drives out residual solvent molecules. All these results demonstrate a new route to achieve high efficiency OPV devices for commercial application.


N14.15
Surface Modification of CdSe Quantum Dots with Regioregular Thiophene Oligomers for High Efficiency Hybrid Solar Cells. Hyo-Sung Choi and Tae-Ho Yoon; GIST, Gwangju, South Korea.

CdSe quantum dots were prepared and modified by ligand exchange method with regioregular thiophene oligomers to enhance their dispersion in P3HT and thus performance of CdSe/ P3HT hybrid solar cells. First, thiophene oligomers such as 2H4T(3,4’-dihexyl-2,2’,5’,2’’,5’’,2’’’-quaterthiophene) were synthesized by varying thiophene length and designed to have thiol functional group. Next, the thiophene oligomers were utilized to coat CdSe QDs via ligand exchange reaction, which were prepared by colloidal method and passivated by trioctylphosphineoxide(TOPO). Then, the modified CdSe QDs were blended with P3HT to evaluate their dispersion behavior and characterized by TEM, UV-vis and PL. In addition, bilayer solar cells were prepared and power conversion efficiency was evaluated as a function of thiophene length.


N14.16
Abstract Withdrawn


SESSION N15: Type II Band Offset Nanostructures and Related Materials
Thursday Morning, December 4, 2008
Republic B (Sheraton)

8:30 AM N15.1
Synthesis of Type II Nanocable Array Based on II-VI Semiconductors for Solar Energy Harvesting. Kai Wang1, Jiajun Chen1, Weilie Zhou1, Yong Zhang2, John Pern2, Yanfa Yan2 and Angelo Mascarenhas2; 1AMRI/Chemistry, Advanced Materials Research Institute/UNO, New Orleans, Louisiana; 2National Energy Renewable Laboratory, Golden, Colorado.

Nanocables based on II-VI semiconductors are predicted as potential candidate materials for solar energy harvesting. In this presentation, we present a successful synthesis of a large-area well aligned ZnO/ZnSe and ZnO/ZnS nanocable arrays directly on transparent conducting oxide (TCO) substrate by combining chemical vapor deposition and pulsed laser deposition. Their structural and optical properties are characterized by applying a set of comprehensive analysis techniques. Structural characterization reveals a sharp interface between the core and shell components. The shell with a thickness of 5-10 nm was observed epitaxially grown on the ZnO nanowire core, which was further confirmed by a line scan analysis of energy-dispersive X-ray spectroscopy. Compared to that of pure ZnO nanowires, photoluminescence measurements performed on the nanocables at room temperature demonstrated that the emission efficiency decreased and an absorption spectrum of the nanocable array showed a red tail, indicating a charge separation occurred in these heterostructures. This nanocable array synthesized on TCO substrate provides a novel, stable and efficient architecture for solar energy harvesting.


8:45 AM N15.2
Interplay Between Quantum Confinement, Strain, Compositional Disorder, and Exciton Dissociation in CdSe-CdTe Core-shell Nanowires. Thomas Sadowski and Rampi Ramprasad; University of Connecticut, Seymour, Connecticut.

The focus of this ab initio computational study is to provide an understanding of the impact of strain and component mixing on the stability, electronic structure, and tendency for electron-hole separation in core-shell CdSe-CdTe nanowires possessing a radial type II band offset. Recent studies suggest that enhanced electron-hole separation indeed occurs more easily in these materials than in single component systems [1]. In particular, first principles computational techniques have been applied to a series of wurtzite nanowires possessing regular hexagonal cross sections. These cross sections are roughly 30 Å in diameter and have nonpolar surface facets belonging to the {10 bar-1 0} family and contain atoms with one dangling bond. All nanorods were assumed to be infinitely long with a periodic length c and axis parallel to the wurtzite (0001) direction. The energies associated with the interface, strain, and mixing as well as the band gaps and locations of the electron and hole states are assessed as a function of varying number of CdSe pairs, strain, core-shell thicknesses, and interface roughness. The overlap of the electron and hole wavefunctions is also determined to quantify how these factors affect the extent of electron-hole separation for each topology. For this study, we considered six different cross sections obtained by varying the composition of CdSe throughout the structure with the impact of strain assessed by fixing the lattice parameter of CdSe along its axis at specific values and comparing with the results corresponding to the equilibrium value. At their equilibrium lattice parameter, homogeneous CdSe and CdTe nanowires were found to exhibit wide band gaps and significant overlaps of their electron and hole wavefunctions. Straining the CdTe nanowire at the CdSe lattice parameters resulted in a narrowing of the band gap and decrease in the overlap of the electron and hole states. Furthermore, a small valence band offset appeared between the CdTe atoms in the core and those able to relax on the surface. In the core-shell CdSe-CdTe nanowires, straining the system at the CdSe lattice parameters resulted in a larger band gap and more overlap than a similar structure maintained at its equilibrium lattice parameter, independent of whether the interface was abrupt or rough. Interestingly, the valence band offset between the CdSe and CdTe regions was found to be largely independent of the strain. Conversely, when CdSe pairs are uniformly mixed throughout CdTe, both the band gap and electron-hole wavefunction overlap decreased in the strained state compared to the equilibrium lattice parameter, as with the homogeneous topologies. 1. H. Zhong, Y. Zhou, C. Yang, Y. Li, J. Phys. Chem. C 111, 6538 (2007).


9:00 AM N15.3
Charge Separation Between Type II Aligned Closed Packed CdTe and CdSe Nanocrystals. Dieter Gross, Andrei Susha, Thomas Klar, Enrico Da Como, Andrey L. Rogach and Jochen Feldmann; Physics Department, Ludwig-Maximilians-Universitaet Muenchen, Munich, Germany.

We report on charge separation between type II aligned CdTe and CdSe nanocrystals [1], with the lowest excited energy state for electrons in CdSe and the lowest excited hole state in CdTe. This kind of energy level alignment facilitates the separation of photo-excited charge carriers in thin film solar cells. Two types of electrostatically bound nanocrystal structures have been studied: first, clusters of nanocrystals hold together by Ca(II) ions in aqueous solution, and second, thin film structures of nanocrystals created by layer-by-layer deposition in combination with polyelectrolytes. In both types of structures, short inter-particle distances of less than 1 nm have been achieved, whereby the isolating organic ligands on the nanocrystal surfaces and/or the polymer monolayers act as thin tunneling barriers between particles. We have observed an efficient quenching of photoluminescence and a reduced emission life time for CdTe nanocrystals in both types of type II hetero-structures. This result is explained by a spatial charge separation of the photo-excited electron-hole pairs due to tunneling of charge carriers. Type II heterostructures demonstrated here may find future applications in photovoltaics. [1] Gross, D.; Susha, A. S.; Klar, T. A.; Da Como, E.; Rogach, A. L.; Feldmann, J. NanoLett. 2008, 8, 1482-1485.


9:15 AM N15.4
Tin Monosulfide as Light-Absorbing Material in Thin Film Photovoltaic Applications. Ramprasad Chandrasekharan and Jeffrey R.S. Brownson; Energy and Mineral Engineering, The Pennsylvania State University, State College, Pennsylvania.

Recent spurt in research of tin monosulfide (SnS) as a promising light absorbing material for inorganic thin film photovoltaic applications can be attributed to its environmental friendliness and the cheap cost of raw materials tin and sulfur. With high absorption coefficient (α >10^4 cm-1) and favorable band gap (1.1-1.4 eV, p-type), SnS offers exciting opportunities for integration into nano-architectured devices such as eta (extremely thin absorber) - solar cells, operating at high efficiencies. In this research, SnS was electrochemically deposited on transparent conductive oxide (F-doped SnO2) - coated glass, and characterized for its photoelectrochemical properties to obtain band energy positions. For efficient conduction of photogenerated minority charge carriers away from the absorber, SnS was deposited on ZnO nanowires (wide band gap - 3.1 eV, n-type), and the interface was studied for its photovoltaic characteristics. Further investigations on the interface for its charge-transfer performance which is influenced by the presence of possible interfacial energy states, are required. Through characterization, the role of such states as recombination centers that reduce quantum yield in SnS-based photovoltaic devices can be better understood, providing research opportunities that can lead to the evolution of advanced solar cells that are commercially competitive and augment photovoltaic power conversion.


9:30 AM N15.5
Vertical Aligned Nanorod Assembly by Electrophoretic Deposition from Organic Solvents: Towards Aligned Nanorod Solar Cells. Kevin Michael Ryan and Shafaat Ahmed; Chemical and Environmental Sciences, University of Limerick, Limerick, Ireland.

The perpendicular alignment of II-VI nanorods in hybrid nanorod/polymer and all-nanorod solar-cells is sought to reduce charge recombination by providing directional charge pathways to the electrodes. Here we describe the organization of II-VI nanorods into perpendicular assemblies by a novel electrophoretic deposition approach. Monolayer assemblies are obtained by completely immersing parallel electrodes in a dilute solution of pyrolysis synthesized nanorods (5 nm × 30 nm). The positively charged nanorods deposit on the negative ITO electrode in ordered arrays which completely coat the electrode surface. The nanorods are hexagonally close packed with an interparticle separation of 2 nm occupied by interdigitated phosphonic acid surfactants. The nanorod supercrystals are characterized by TEM, HRTEM, HRSEM and I/V nanoprobe measurements. This approach allows deposition of CdSe, CdS and CdTe nanorods in very large areas optimal for perpendicularly aligned nanorod solar-cell application.


SESSION N16: Advances in Dye-Sensitized Solar Cells and Photocatalysis
Thursday Morning, December 4, 2008
Republic B (Sheraton)

10:15 AM N16.1
Using Long Range Forster Energy Transfer to Increase Light Absorption in Dye Sensitized Solar Cells. Brian E. Hardin1,2, Eric T Hoke1, Craig Peters1, Thomas Geiger3, Frank Nuesch3, Michael Graetzel2, Md. Khaja Nazeeruddin2 and Michael D McGehee1; 1Material Science and Engineering, Stanford University, Stanford, California; 2Laboratory for Photonics and Interfaces, Institute of Chemical Sciences and Engineering, Ecole Polytechnique Federale de Lausanne, Lausanne, VD, Switzerland; 3Laboratory of Functional Polymers, Empa, Swiss Federal Laboratories for Materials Testing and Research, Dubendorf, Switzerland.

The best solid-state dye-sensitized solar cells (ss-DSCs) have short circuit current of ~8 mA/cm2 which is roughly a factor of ~2.25 lower than what has been shown in DSCs with liquid electrolytes. With conventional sensitizing dyes that are usually based on ruthenium complex, 10-µm-thick films are needed to have enough dye to fully absorb the sunlight. Unfortunately, when the cell thickness exceeds 2.5 µm the internal quantum efficiency declines rapidly due to recombination and/or bad pore filling. The short circuit current density can be increased by placing chromophores inside the hole conductor that will absorb light and relay the energy via long range Forster energy transfer (FRET) to the dye attached at the surface. Using energy relay dyes has three important advantages. First, since the attached dye only has to absorb light over a smaller spectral region, it can be chosen to have a stronger and narrower absorption spectrum. Second, the conventional dye can be red shifted compared to the commonly used dyes since the energy relay dye can absorb higher energy photons. Finally, the energy relay dyes do not need to be attached to the TiO2 surface and can be placed in very large concentrations inside the hole transporter matrix. Thus the addition of energy relay dye into the films will make the overall absorption spectrum wider and stronger for the same film thickness. In a FRET enhanced DSC, energy transfer occurs from the relay dye to any of the adsorbed dye molecules on the titania surface. Quantum dots, luminescent dyes, and dendrimers were chosen as energy relay dye candidates because of their strong absorption, high photoluminescence efficiency (QD ~20-80%), and emission spectrums that overlaps with the absorption spectrum of the attached sensitizing dye. We have chosen a squaraine dye with a high molar extinction coefficient of ε=158,500/M-cm, which is a factor of 10 greater than that of the ruthenium dyes commonly used in DSCs. The energy relay dye systems have theoretical Forster radii which range from 4.5-6.2nm. With the CdSe/ZnS quantum dot system we have measured a Forster radius of 7.2nm. Energy transfer from a donor to an entire plane of acceptors is much faster and transfers energy more efficiently over larger distances (>15nm). Although the mesoporous titania surface is by no means flat, as a first approximation, the surface near the energy relay dye can be modeled locally as a plane. If dyes attached to the surface are tightly packed (0.5-1 dye molecule/nm2) on the TiO2 surface, the energy relay dye can transfer energy efficiently well beyond the FRET radius. With a moderate point-to-point FRET radius of 5 nm it is possible to transfer ~90% of the light absorbed by the energy relay dye in films with an average pore size of 20 nm. Modeling and device results will be presented.


10:30 AM N16.2
3D Dye-sensitized Solar Cells (Tandem cells) Prepared by Nano-interface Control- Improvement of Light Harvesting Properties and Electron Collection Properties-. Shuzi Hayase1, Fumi Inakazu1, Yusuke Noma1, Takeshi Kougo1, Shouhei Sakaguchi1, Yoshihiro Yamaguchi2 and Mitsuru Kono2; 1Kyushu Institute of Technology, Kitakyushu, Fukuoka, Japan; 2Nippon Steel Chemical Co., Ltd., Kitakyushu, Japan.

Improvements of light harvesting properties are crucial for increasing efficiencies of dye-sensitized solar cells (DSC). In order to cope with this, dye-bilayer structure of titania electrode are proposed. It was difficult to prepare the dye-bilayer structure of nano-titania layer by conventional dipping processes. Dye-bilayer structure was successfully prepared for the first time by controlling the reactivity between nano-surfaces and dye molecules under a pressurized CO2 condition. The reaction mechanism is discussed. Under the pressurized CO2 conditions, surfaces of nano-particles are passivated with CO2 molecules and the reaction rate between the nanoparticle surfaces and the dye molecules increased. Because of the activated nano-particle passivation, dye-staining occurred from the top of the titania layer to the bottom consecutively. The dye-bilayer consisting of Ru dyes having long wavelength and organic dyes having short wavelength covers wide ranges of wavelengths. The Jsc of the cell consisting of bi-layer structure was 21.8 mA/cm2, which was larger than that consisting of black dye only (20.4 mA/cm2). This strongly supported the effectiveness of the dye-bilayer structure. In order to improving electron-collection properties, TCO-less 3D DSC are prepared by using nano-technologies. The 3D DSC is characterized by porous thick Ti electrodes, which was prepared for the first time by nano-technologies. The 3D DSC shows higher photovoltaic performance than that of DSCs using transparent conductive layers (TCO). 10.4 % efficiency is reported. In addition, flexible-3D-DSCs prepared by roll-to-roll process are proposed. The 3D DSC consists of a Ti sheet / a core layer / a plastic sheet, where the core layer consists of a porous titania layer / a flexible porous metals/ a gel electrolytes core layer. Electrons are collected between the Ti sheet and the flexible porous metals. The DSCs can be prepared by roll-to-role processes relating to nano-structures and is expected to be low.


10:45 AM N16.3
Is Pore Filling a Problem in Solid-State Dye-Sensitized Solar Cells? I-Kang Ding1, Brian E Hardin1, Samuel J Rosenthal1, Michael Graetzel2 and Michael D McGehee1; 1Materials Science and Engineering, Stanford University, Stanford, California; 2Ecole polytechnique fédérale de Lausanne, Lausanne, Switzerland.

We report using X-ray Photoelectron Spectroscopy (XPS) depth profiling to investigate the infiltration of the organic hole transporting material, 2,2’,7,7’-tetrakis-(N,N-di-p-methoxyphenylamine)9,9’-spirobifluorene (Spiro-OMeTAD), inside nanoporous titania films that are used in solid-state dye-sensitized solar cells. XPS depth profiling technique is capable to measure the elemental composition on the surface as an argon sputtering gun gradually uncovers the interior of the film. Our results show that the Spiro-OMeTAD appear to infiltrate all of the way to the bottom of 4~6-micron-thick films made with 9 to 37 nm-diameter titania nanoparticles. This finding is in direct contradiction to the belief that Spiro-OMeTAD pore filling limits the optimum device thickness at 2 μm. We also found out that titania nanoparticle size and dye modification of the surface has no effect on pore-filling. If these titania nanoparticle films were completely filled with Spiro-OMeTAD, the carbon to oxygen ratio (C:O ratio) on the surface would be 1.84, assuming that the porosity of the TiO2 films is 60%, the density of Spiro-OMeTAD is 1.82 g/cm3 and the density of TiO2 is 3.9 g/cm3. The C:O ratio we experimentally measured is 0.3, which is much lower than the theoretical value. Therefore, we know that the pores are not completely filled, even though Spiro-OMeTAD does go all the way to the bottom of the films. We think the most likely explanation for this observation is that space is left behind when the solvent that carries the Spiro-OMeTAD into the pores evaporate. To test this hypothesis, we varied the concentration of the Spiro-OMeTAD solutions and performed depth profiling. As expected, we found that the C:O ratio increases with the Spiro-OMeTAD concentration in solution. The C:O ratio is constant with depth in each case. The C:O ratio is constant with depth in each case. The results indicate that approximately 15-20% of the pore volume is filled in conventional ss-DSCs. This fraction could be increased to 25% when we use a fully saturated solution of Spiro-OMeTAD. This finding suggests that the key to infiltrating more Spiro-OMeTAD is to avoid using a solvent.


11:00 AM N16.4
Infrared Dye and Quantum Dot Sensitized TiO2 Nanotube Arrays to harvest Red and Near-Infrared Photons. Karthik Shankar1,2, Sanjeev Sharma1, Oomman K Varghese1, Maggie Paulose1, John E Anthony4,5 and Craig A Grimes2,1,3; 1The Materials Research Institute, The Pennsylvania State University, University Park, Pennsylvania; 2Department of Electrical Engineering, The Pennsylvania State University, University Park, Pennsylvania; 3Department of Materials Science & Engineering, The Pennsylvania State University, University Park, Pennsylvania; 4Center for Applied Energy Research, University of Kentucky, Lexington, Kentucky; 5Department of Chemistry, University of Kentucky, Lexington, Kentucky.

The best dye sensitized solar cells (DSCs) currently convert incident sunlight into electricity at an efficiency greater than 11%[1]. This is rather close to the value of 10% efficiency reported by Gratzel et al[2] in 1993 indicating the difficulty encountered in the last 15 years in increasing the maximum conversion efficiency of DSCs. A principal reason is that currently used dyes harvest only a fraction of the available solar energy in the red and near-infrared region of the solar spectrum. Further improvements in efficiency for next-generation low cost solar cells require a greater harvesting of incident red and infrared photons. At the same time, candidate dyes that absorb in the infrared need to possess a smaller HOMO-LUMO separation as well as a suitable positioning of their HOMO orbital to allow efficient injection of photogenerated electrons into the host semiconductor (typically TiO2 or ZnO). We outline two approaches towards this end. The first approach consists of the use of a indolydine-derivatized cyclohexene sensitizer with a very high molar extinction coefficient of 190,000 L/(mol cm) at 780 nm in conjunction with TiO2 nanotube arrays to form solar cells that have a maximum incident photon-to electron conversion efficiency of 20% in the near-infrared, which is among the highest reported. A route to form mixed monolayers combining the infrared sensitizer with highly absorbing dyes in the visible region is also presented. The mixed monolayers harvest a very broad swathe of the AM 1.5G solar spectrum and provide an exciting route to improving the maximum efficiency of dye sensitized solar cells. The second approach consists of using a thin inorganic absorber layer of a lower band-gap semiconductor such as PbS. The inorganic layer may either consist of quantum dots anchored to the TiO2 surface or a chemically deposited layer with properties similar to that of the bulk. Using photoelectrodes consisting of PbS coated by chemical bath deposition on to TiO2 nanotube substrates, photocurrents of 8 mA/sq.cm under AM 1.5G illumination were obtained. 1. Gratzel, M. New nanocrystalline solar cells. Actualite Chimique, 57-60 (2007). 2. Nazeeruddin, M.K. et al. Conversion of Light to Electricity by Cis-X2bis(2,2'-Bipyridyl-4,4'-Dicarboxylate)Ruthenium(Ii) Charge-Transfer Sensitizers (X = Cl-, Br-, I-, Cn-, and Scn-) on Nanocrystalline TiO2 Electrodes. Journal of the American Chemical Society 115, 6382-6390 (1993).


11:15 AM N16.5
Bandgap Narrowing of Titanium Dioxides via Non-Compensated n-p Co-doping for Photocatalysis. Wenguang Zhu1,2, Xiaofeng Qiu2, Hui Pan2, Wei Wang2, Baohua Gu2, Mariappan Parans Paranthaman2, Gyula Eres2 and Zhenyu Zhang2,1; 1University of Tennessee, Knoxville, Tennessee; 2Oak Ridge National Laboratory, Oak Ridge, Tennessee.

Titanium dioxide TiO2 is a promising photocatalyst for solar hydrogen production from water, yet its photocatalytic efficiency is limited by its intrinsic wide-bandgap nature. In this talk, we present a conceptually new and intuitive approach, termed non-compensated n-p co-doping, to narrow the bandgap of TiO2. The validity of this approach has been demonstrated using first-principles calculations within density functional theory, showing that extra impurity bands are created in the gap region because of the non-compensated nature of the n-p co-doping, resulting in a narrowed bandgap around 2 eV. Moreover, the electrostatic attraction between the n and p dopants enhances their thermodynamic and kinetic solubility in the host semiconductors. *Research Supported by the DMSE program and grant number DE-FG02-05ER46209 of USDOE, grant number DMR-0606485 of USNSF, and LDRD program of ORNL.


11:30 AM N16.6
High Efficiency Photosplitting of H2O using TiO2/TiSi2 Heterostructure Complex Nanomaterials. Yongjing Lin, Sa Zhou, Xiaohua Liu and Dunwei Wang; Chemistry, Boston College, Chestnut hill, Massachusetts.

The ever-depleting reserves and the devastating environmental effects caused by burning fossil fuels - the dominating energy supply we rely on - has necessitated the development of new energy sources or carriers. Among energy forms that have been investigated, solar H2 from H2O splitting is particularly appealing as it utilizes the renewable solar energy and is environmentally friendly. Existing materials to absorb light and chemically split H2O are often suffering the dilemma of either poor catalytic reactivity or poor conductivity, either of which is detrimental to the targeted applications. We present in this talk our approach in producing heterostructure complex nanomaterials that are composed of multiple components, TiO2 as the shell and TiSi2 as the core. The shell is an excellent photocatalyst that also exhibits remarkable stabilities in aqueous solutions, and the core is a highly conductive material that has been reported to be a good photocatalyst recently, as well. The complex nature of the entire structure ensures high surface area (derived from the nanometer scales) and outstanding charge transport (all parts are connected through single-crystalline junctions). TiSi2 cores are synthesized by chemical vapor deposition and TiO2 shells are deposited by atomic layer deposition. H2O splitting efficiencies up to 16% in UV light (or 8% in Xenon light) have been reproducibly achieved. Preliminary experiments also show that significant improvement of visible light absorption can be obtained by chemical doping TiO2 during growth. Various parameters for efficiency optimizations will be discussed and rationalized. Our results shall inspire new nanomaterials by design for energy-related applications.


11:45 AM N16.7
Improved Performance of Quantum Well Solar Cells via Light Scattering from Dielectric and Metal Nanoparticles. Daniel Derkacs, Winnie V Chen, Paul K Yu and Edward T Yu; Electrical Engineering, University of California San Diego, La Jolla, California.

We have fabricated and characterized InP/InGaAsP quantum-well waveguide solar cells in which performance is improved via light scattering from deposited dielectric and metal nanoparticles. The integration of metal or dielectric nanoparticles above the quantum-well solar cell device is shown to couple normally incident light into lateral optical propagation paths, with optical confinement provided by the refractive index contrast between the multiple quantum-well region and surrounding material. To optimize collection of photogenerated carriers from the quantum wells and to minimize the reduction of Voc, moderate energy barriers and a sufficiently large electric field across the intrinsic region - on the order of 30 kV/cm or more - are typically required. However, the thin multiple-quantum-well region implied by this electric-field requirement can preclude efficient absorption of incident photons. Nanoparticle-induced scattering of incident photons into lateral propagation paths circumvents this conflict by providing photons with increased propagation distances even in thin layers, thereby decoupling photon and carrier transport paths and enabling, simultaneously, high efficiency in both photon absorption and photogenerated carrier collection. Nominally lattice-matched InP/InGaAsP multiple-quantum-well p-i-n solar cell structures grown by metal organic chemical vapor deposition with intrinsic region thicknesses of 250 nm were employed in our studies. Photovoltaic devices were fabricated using standard photolithography and metallization techniques, and dielectric or metal nanoparticles were deposited on device surfaces from colloidal solutions. With minimal optimization, quantum-well solar cell devices exhibited an increase of 7.4% in efficiency relative to InP homojunction reference devices. In addition, short-circuit current density increases of 12.9% or 7.3% and power conversion efficiency increases of 17% or 1% were observed upon incorporation of silica or Au nanoparticles, respectively. Measurements of photocurrent response spectra confirm the role of nanoparticle scattering in both increasing photon transmission into the semiconductor and enabling coupling of incident photons into waveguide modes within the semiconductor. A detailed theoretical analysis provides quantitative estimates of the efficiency of photon coupling into confined and leaky waveguide modes associated with the multiple-quantum-well structure, and has enabled design of device structures in which major improvements in this coupling efficiency are anticipated.


SESSION N17: Inorganic Absorber Sensitized Solar Cells
Thursday Afternoon, December 4, 2008
Republic B (Sheraton)

1:30 PM N17.1
Wet-chemical Synthesized Inorganic Extremely Thin Absorber (ETA) Solar Cells. Olivia Niitsoo, Miles G Page, Yafit Itzhaik, David Cahen and Gary Hodes; Department of Materials and Interfaces, Weizmann Institute of Science, Rehovot, Israel.

The quest for cheap, simple, reliable photovoltaic (PV) cells to help meet the need for clean energy is gaining momentum because of growing concern over the environmental and climatic impacts of continuing reliance on fossil fuels. One promising concept is the Extremely Thin Absorber (ETA) class of solar cells, in which an ultra-thin, light-absorbing layer is deposited on the large internal surface of high bandgap metal oxide semiconductor, and the volume of the cell is filled with an appropriate transparent hole conductor. The choice of transparent, high bandgap metal oxide materials for the ETA cell skeleton is mainly limited to TiO2 and ZnO (although other oxide materials are also being developed for this purpose), and the choice of stable wide bandgap inorganic hole conductors is even scarcer, CuSCN being the most successful one. However, there is a multitude of absorber materials. We investigate whether the ETA concept can be generalized in terms of inorganic absorbers, what are the factors that make some of them better than others in terms of solar cell performance, whether this performance is still controlled mainly by the bulk properties (i.e., light absorption, bulk recombination, energy levels) or if the thin film configuration reduces the relevance of recombination (due to the shorter distance that the carriers have to cross) and changes the band alignment (due to reduced dimension). We compare absorber materials (CdS, Cu2-xS, Sb2S3 and CdSe) that differ in optical and electronic properties, and in terms of the morphology of the material that deposits in the metal oxide matrix. We also address technological issues that intertwine with science: wet-chemical deposited (nano)crystalline absorbers do not necessarily form a perfect coating on the underlying metal oxide; therefore, there will be several paths for the photogenerated carriers to reach the collecting electrode, and chances of recombination increase. Recombination-reducing buffer layers between metal oxide and the absorber are recognized to be critical in solid state ETA cells. We have also used molecular monolayers as additional buffer layers to inorganic ones. We can summarize the present situation of ETA cells, based on published results and our own, by stressing that control over the interfaces in ETA cells is what is needed most to help overcome the technological problems that still exist. Such an effort is justified as it may open up a very large library of potential absorber materials that can then be exploited in low-cost metal oxide-based solar cells.


1:45 PM N17.2
3D Architectures of Semiconducting Nanogrids Sensitized with Quantum Dots for High Efficient Photoconversion. Xiangyang Kong, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, China.

Creating 3-D ordered nanostructures is of key interest in the design of novel nanoelectronic and nanophotonic devices. We demonstrated a novel technique to fabricate the grid-like nanostructures of semiconducting oxides by biotemplating approach, and assemble these grid-like nanostructures into 3-D open architectures by Langmuir-Blodgett method or ink-jet printing techniques. These semiconducting oxides grids sensitized with CdSe quantum dots can be integrated on a flexible substrate as 3-D open architectures for the photocatalytic electrode. We also report a new type of photovoltaic devices with composite structures, which is composed of two photoactive semiconductors electrodes, one using n-type TiO2 as a photo-anode, and another using p-type NiO as a photo-cathode. These electrodes formed as the stacked-grid arrays, not only provide a large surface area rather than mesoporous thin films, but also increase the light harvesting efficiency resulting from the light localization in the stacked-grid array. The stacked-grids arrays also play a role of channel behavior for the efficient electron diffusion and rapid transport to the conductive substrate, resulting in the increase of the photogenerated current. The promising DSSCs with composite structures show the significant photoconversion efficiency up to 9.8%. Keywords: grid-like nanostructures, 3-D open architectures, biotemplating approach, photoconversion.


2:00 PM N17.3
Photovoltaic Properties of CdTe Nanoparticles-Decorated Titania Nanotubes Arrays. Eugen Panaitescu, Dattatri Nagesha, Trifon Fitchorov, John S Morris, Srinivas Sridhar and Latika Menon; Physics, Northeastern Univ., Boston, Massachusetts.

Titanium oxide's wide bandgap allows for a low electron-hole recombination rate, but also limits the photons absorption spectrum to the ultraviolet range. Different methods such as dye sensitization or doping have expanded this spectrum into the visible. Another method is the conjugation of visible range active nanoparticles, and we are reporting on the successful production of hybrid structures involving titanium oxide nanotubes arrays (150 - 200 nm diameter) synthesized by anodization in fluorine based solution conjugated with CdTe quantum dots (less than 5nm diameter). Photoluminescence measurements revealed a peak around 620 nm specific to the CdTe nanoparticles, and proving the red shifting of the absorption spectrum. Further photovoltaic properties measurements of the CdTe nanoparticulate film coating titania nanotubes arrays compared with similar dye sensitized samples have been performed. Photoresponse parameters like short circuit current, open circuit voltage, maximum power and overall conversion efficiency have been measured under simulated solar radiation.


2:15 PM N17.4
Electron Injection from Colloidal PbS Quantum Dots into TiO2 and SnO2 Nanoparticles. Byung Ryool Hyun1, Adam C Bartnik1, Yu-Wu Zhong2, Liangfeng Sun1, Hector D Abruna2, Jason D Goodreau3, James R Matthews3, Thomas M Leslie3 and Nick F Borrelli3; 1Applied physics, Cornell University, Ithaca, New York; 2Chemistry and Chemical Biology, Cornell University, Ithaca, New York; 3Organic Technologies, Corning, Inc, Corning, New York.

Injection of photoexcited electrons from colloidal PbS quantum dots into TiO2 and SnO2 nanoparticles is investigated. The electron affinity and ionization potential of PbS quantum dots, inferred from cyclic voltammetry measurements, show strong size dependence due to quantum confinement. Based on the measured energy levels, photoexcited electrons should transfer efficiently from the quantum dots into TiO2 or SnO2 for quantum-dot diameters below ~4.3 nm and ~34 nm, respectively, because of the lower electron affinity of SnO2. Continuous wave fluorescence spectra and fluorescence transients of coupled PbS quantum dots are measured to verify the electron transfer. When coupled to TiO2, electron transfer is seen only in small PbS QDs; whereas in SnO2, electron transfer is still seen in the largest available QD diameter. The electron injection time from PbS QDs into TiO2 and SnO2 nanoparticles is obtained from transient absorption spectroscopy. The differences in electron transfer and recombination dynamics between TiO2 and SnO2 are highlighted. In addition, initial results obtained from Gratzel cells sensitized with PbS quantum dots are presented.


2:30 PM N17.5
Spectroscopic Characterization of Electron Injection and Charge Recombination in Tethered Quantum Dot-Metal Oxide Assemblies. David F Watson and Rachel Dibbell; Chemistry, University at Buffalo, Buffalo, New York.

Quantum dots may represent attractive alternatives to molecular sensitizers for applications in photocatalysis and solar energy conversion. Our laboratory has used surfactant-mediated self-assembly to prepare quantum dot-functionalized nanocrystalline TiO2 films. CdS and CdSe quantum dots are tethered to TiO2 surfaces through bifunctional organic linkers with terminal thiol and carboxylic acid groups. Equilibrium binding data reveal surface adduct formation constants (Kad) of 104 - 105 M-1 for the adsorption of CdSe quantum dots to thiolated TiO2 films. Quantum dot surface coverages are highest on mixed monolayers with both thiol- and methyl-terminated surfactants. Emission quenching and nanosecond transient absorption experiments have revealed that photoexcitation of surface-adsorbed quantum dots leads to efficient electron injection into the TiO2 substrate and long-lived charge-separated states. This presentation will focus on materials assembly and the spectroscopic characterization of electron injection and charge recombination processes. The influence of the molecular linker on the electron injection yield and the charge-recombination timescale will be highlighted. Specifically, we have found that the electron injection yield increases dramatically with decreasing linker length. Our findings lend insight into the factors governing the efficiency of bridge-mediated interparticle electron transfer processes.


SESSION N18: Inorganic/Organic Hybrid Solar Cells
Thursday Afternoon, December 4, 2008
Republic B (Sheraton)

3:15 PM N18.1
Investigating Charge Carrier Mobilities in Nanocrystal-Polymer Hybrid Photovoltaic Devices. Jian Xu, Fan Zhang and Shuai Gao; Engineering Science and Mechanics, Penn State University, University Park, Pennsylvania.

Hybrid plastic photovoltaic devices based on poly-(3-hexylthiophene) (P3HT) polymer blended with PbSe nanocrystal quantum dots (NQDs) have attracted wide attentions for their possible application in solar energy conversion [1-4]. In the host/guest composite, P3HT is known for its high hole-mobility and the PbSe NQDs dispersed in the polymer host act as electron acceptors. In addition, the size-tunable absorption of PbSe NQDs spans 1100 nm to 2500 nm [4], which sensitizes the polymer to the near-infrared (NIR) portion of the solar spectrum and enhances the harvest of infrared solar energy by the plastic photovoltaic cells. In this conference, we will present our study on the properties of charge separation and photo-carrier transport in P3HT/PbSe NQD composites by employing time of flight (TOF) method under selective excitation conditions. Both electron- and hole-photocurrent transients were measured under 532nm- and 1064nm-illumination, respectively. Charge carrier mobilities were calculated accordingly. In particular, the characterization of the carrier mobilities upon NIR excitation was employed to investigate the oleic acid-ligand effect on the charge-carrier-trapping kinetics at the P3HT/PbSe NQD interfaces. Our investigation confirms that the introduction of PbSe NQDs into P3HT thin films greatly balances the mobilities of holes and electrons. The hopping and tunneling transport of electrons in the PbSe nanocrystal network of the hybrid composite was verified by the experimental results, revealing the essential role of the infrared NQDs in the photovoltaic device under study.


3:30 PM N18.2
Improving PV Performance of Metal Oxide/Conjugated Polymer Based Hybrid Solar Cells using Interfacial Modifiers. Kethinni G Chittibabu, Dave Waller, Christoph Brabec and Russ Gaudiana; Konarka Technologies Inc, Lowell, Massachusetts.

High performance metal oxide/conjugated polymer based solar cells have been fabricated using interface modifiers at the metal oxide/conjugated polymer interface. The influence of interfacial modifiers on the short circuit current and open circuit volatage will be discussed.


3:45 PM N18.3
Hybrid Nanocomposites Based on ZnO Nanorods Functionalized with Supramolecular Porphyrin-Fullerene Complexes for Low Cost Solar Cells. Syed Mujtaba Shah1, Aiko Kira2, Hiroshi Imahori2, Frederic Fages1 and Jorg Ackermann1; 1Chemistry, Centre Interdisciplinaire de Nanoscience de Marseille (CiNaM), Marseille, France; 2Department of Molecular Engineering, Graduate School of Engineering, Kyoto University, Kyoto, Japan.

Hybrid composite materials based on surface functionalized semiconductor nanoparticles represent promising nanosystems for low cost photovoltaics. The use of anisotropic nanoparticles such as nanorods and nanowires opens new possibilities due to their unique electronic, optical and self-assembly properties. Here we report the synthesis of zinc oxide nanorods functionalized with supramolecular complexes of porphyrins and fullerenes. The formation of such donor-acceptor complexes on the surface of nanorods is aimed to increase the charge carrier injection from the porphyrin toward the ZnO. Absorption and fluorescence spectroscopy were used to investigate the influence of nanoparticle shape (nanorod or spheres), solvent and fullerene-porphyrin ratio on the complex formation and the photo-physics of the resulting hybrid nanoparticles. Furthermore we demonstrate that the co-grafting of porphyrin and fullerene introduce specific self-assembly properties to the hybrid nanostructures which leads to the formation of ordered aggregates with parallel oriented nanorods. Additionally the use of such hybrid nanorods in dye sensitized bulk heterojunction solar cells based on porphyrin-C61-ZnO/P3HT blends will be discussed.


4:00 PM N18.4
A Soft Lithography Route to Nanopatterned Photovoltaic Devices. Stuart Williams1, Meredith Hampton1, Vignesh Gowrishankar2, I-Kang Ding2, Joseph Templeton1, Edward Samulski1, Michael McGehee2 and Joseph DeSimone1,3; 1Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; 2Materials Science and Engineering, Stanford University, Stanford, California; 3Chemical Engineering, North Carolina State University, Raleigh, North Carolina.

We have fabricated bulk heterojunction photovoltaic (PV) cells using a perfluoropolyether (PFPE) elastomeric stamp to control the morphology of the donor-acceptor interface within devices. Devices were fabricated using the Pattern Replication In Non-wetting Templates (PRINT1) process to have nanoscale control over the bulk heterojunction device architecture. The low-surface energy, chemically resistant, variable modulus, fluoropolymer based molds used in PRINT provide a route to patterning, with nanometer resolution, general polymeric donor materials such as polythiophene and polyphenylenvinylene derivatives and ‘hard’ inorganic oxide structures typically used as acceptor materials in hybrid organic solar cells such as TiO2, ZnO, and CdSe. This “top-down” approach allows for patterning over large areas and for the functionalization of the donor/acceptor interface. Specifically, nanostructured anatase titania with post-like features ranging from 30-100 nm in diameter and 30-65 nm in height was fabricated to form the ordered bulk heterojunction of a titania-poly(3-hexylthiophene) (P3HT) PV-cell. The nanostructured devices showed a two-fold improvement in both short-circuit current (Jsc) and power conversion efficiency (ηeff) relative to reference bilayer cells. The titania was also functionalized with Ru(II) Z907 to increase the Jsc, open-circuit current (Voc), and ηeff. Additionally, we will discuss devices fabricated with other organic and inorganic materials in order to investigate the effect on cell performance of controlling the nanoscale architecture of the bulk heterojunction via patterning. 1. Rolland et al. J. Am. Chem. Soc. 2005, 127, 10096-10100


4:15 PM N18.5
Controlling the Distribution of CdSe Colloidal Quantum Dots in Conducting Polymer Nanocomposite Thin Films Using Matrix-Assisted Pulsed Laser Evaporation. Ryan Pate, Kevin R Lance and Adrienne D Stiff-Roberts; Duke University, Durham, North Carolina.

Bulk heterojunction conducting polymer/colloidal quantum dot (CQD) nanocomposites have been shown to significantly improve the performance of solar cells, compared to purely organic bi-layer devices, due to the enhancement of solar radiation absorption, exciton dissociation, and charge transport [1,2]. Unfortunately, the performance of these hybrid organic/inorganic nanocomposite solar cells continues to exhibit significantly lower efficiencies than their organic counterparts, i.e. conducting polymer/fullerene nanocomposites. This performance challenge is due in large part to the insulating surface ligands that encapsulate the CQDs and the uncontrolled distribution of CQDs that occurs inside the conducting polymer during solution processing [2,3]. Typical methods for depositing these hybrid structures include drop-casting and spin-casting; however, inadequate control of film thickness and uniformity, as well as an inability to control CQD distribution inside the polymer, makes consistent deposition results challenging to obtain. Matrix-assisted pulsed laser evaporation (MAPLE) is a novel vacuum deposition technique for the fabrication of polymer/CQD nanocomposites, and has been used to demonstrate the controlled deposition of many organic materials [4,5]. In this paper, we will use the MAPLE technique to demonstrate control over the distribution of CdSe CQDs in a polymer matrix, and we will compare the structural and electrical properties of these films as a function of CQD distribution. CdSe CQD/poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylenevinylene] (MEH-PPV) nanocomposite thin films will be deposited on glass substrates and carbon grids using MAPLE. The distribution of CdSe CQDs in the polymer film will be investigated as a function of MAPLE deposition parameters, including target composition, simultaneous vs. sequential nanocomposite deposition, and substrate-to-target distance. The influence of in-situ and post-deposition rapid thermal annealing will also be investigated. The MAPLE-deposited films will be compared to drop-cast films, and both will be characterized using transmission electron microscopy (TEM), Hall mobility measurements, and current-voltage measurements. From these data, we seek to identify an optimized CdSe CQD distribution in conducting polymer/CQD nanocomposite thin films, which has important implications to obtaining more efficient hybrid organic/inorganic photovoltaic devices. [1] Sun, B., et. al. (2005), J. App. Phys, 97, 014914. [2] Gunes, S., et al. (2007), Chem. Rev. 2007, 107, 1324. [3] Stiff-Roberts, A.D. (2005), et al., presented at 47th TMS Conf., Santa Barbara, CA. [4] Pate, R., et al. (2008), IEEE J. of Sel. Top. In Quan. Elec., 14, 4. [5] Toftmann, B., et. al. (2004), Thin Sol. Films 453/454, 117.


4:30 PM N18.6
Device Structure Modification: Towards A More Efficient Infrared Nanocrystal Solar Cell. Ting Zhu, Zhan'ao Tan, Shuai Gao, Fan Zhang and Jian Xu; Engineering Science and Mechanics, Penn State University, University Park, Pennsylvania.

In this abstract, we report the successful fabrication of a planar-mixed heterojunction (PM-HJ) like infrared solar cell based upon PbSe nanocrystal/P3HT nanocomposite. The previously reported PbSe nanocrystal/P3HT bulk heterojunction solar cell device by the authors enables a collection of solar spectrum to infrared range. To further improve the infrared photovoltaic response of the nanocrystal solar cell, an ultra-thin P3HT layer was added underneath the PbSe nanocrystal/P3HT active layer by spin casting method. The additional layer effectively suppresses the recombination current and improves the charge transportation. Thus it serves as an electron blocking and hole transport layer. The current structure enables us to combine the benefits of double-layer and bulk heterojunction structure. The underneath P3HT layer is kept intact because of the unique realignment property of alkyl side chain on the P3HT at the higher temperature. Under the illumination of 808 nm laser, 100 mW/cm2, the PM-HJ structured device showed a nearly triple improvement in photovoltaic response with a Voc of 0.38 V, a Jsc of 1.73 mA/cm2, a FF of 40%, and a PCE 0.26%, while the corresponding parameters of the control solar cell based on bulk-HJ structure is 0.25 V, 1.10 mA/cm2, 35%, and 0.10%, respectively. T. Zhu and Z. Tan contribute equally in this work.




MMR Ad

CIMTEC_2010

Asylum Research