p-Type Dye-Sensitized Solar Cells Based on Delafossite CuGaO2Nanoplates with Saturation Photovoltages Exceeding 460 mVMingzhe Yu, Gayatri Natu, Zhiqiang Ji, and Yiying Wu*

Department of Chemistry & Biochemistry, The Ohio State University, 100 West 18th Avenue, Columbus, Ohio 43210, United States

*S Supporting Information

ABSTRACT: Exploring new p-type semiconductor nanoparticles alternative tothe commonly used NiO is crucial for p-type dye-sensitized solar cells (p-DSSCs)to achieve higher open-circuit voltages (Voc). Here we report the first applicationof delafossite CuGaO2 nanoplates for p-DSSCs with high photovoltages. Incontrast to the dark color of NiO, our CuGaO2 nanoplates are white. Therefore,the porous films made of these nanoplates barely compete with the dyesensitizers for visible light absorption. This presents an attractive advantage overthe NiO films commonly used in p-DSSCs. We have measured the dependenceof Voc on the illumination intensity to estimate the maximum obtainable Voc fromthe CuGaO2-based p-DSSCs. Excitingly, a saturation photovoltage of 464 mV hasbeen observed when a polypyridyl Co3+/2+(dtb-bpy) electrolyte was used. Under1 Sun AM 1.5 illumination, a Voc of 357 mV has been achieved. These are amongthe highest values that have been reported for p-DSSCs.

SECTION: Energy Conversion and Storage; Energy and Charge Transport

A p-type dye-sensitized solar cell (p-DSSC)1 is based on thecathodic sensitization of p-type semiconductors and thus

operates in a manner reverse to the conventional n-type DSSC.Recently, the research on p-DSSCs has attracted increasingattention because they can be integrated with n-DSSCs intotandem DSSCs, which hold a great promise for achieving highpower conversion efficiencies. For example, Lindquist et al.2

combined a NiO-based photocathode with a TiO2-basedphotoanode into a tandem DSSC with a Voc equal the sumof the Voc of the separate devices. Since then, much progresshas been achieved in the molecular engineering of sensitizers aswell as the developments of the redox mediators.3 The recentwork by Nattestad et al.4 in particular demonstrated that atandem DSSC can outperform either a p-DSSC or an n-DSSC.The results clearly show that tandem DSSCs are promising forrealizing substantially improved efficiencies.In principle, p-DSSCs should be able to work as efficiently as

n-DSSCs. However, in reality, the development of p-DSSCs hasbeen lagging behind that of n-DSSCs. The main reason isbecause no optimum wide bandgap p-type semiconductor isavailable that is equivalent to anatase TiO2 as in n-DSSCs. Todate, the predominant p-type semiconductor used in p-DSSCsis NiO. However, NiO is not optimal due to the followingreasons: (1) NiO absorbs a significant amount of visible light. A2.3-μm-thick film absorbs 30−40% of the incident photons overmost of the visible wavelength range.4 More transparent p-typesemiconductors are therefore desired. (2) The valence band(VB) edge of NiO is very close to the redox potential of thecommonly used triiodide/iodide (I3

−/I−) mediator, resulting inlow Voc. The VB edge of NiO is +0.54 V versus normalhydrogen electrode (NHE),1 while the redox potential of

triiodide/iodide is +0.35 V.5 The difference is only 190 mV.Therefore, with just a few exceptions,4,6−9 the reported Voc inmost prior works is in the range of 90−125 mV (see Table S1in the Supporting Information (SI)).1,10−17 (3) NiO has a lowhole mobility with an estimated hole diffusion coefficient ofonly 4 × 10−8 cm2/s,18 which may limit the diffusion length ofthe hole carriers. Therefore it is crucial to find alternative p-typesemiconductors. Some other p-type semiconductors have beeninvestigated, including CuO,13 CuSCN,19 and p-type dia-mond;20 however, the performances of these solar cells are notparticularly better than NiO p-DSSCs. The exploration ofcomplex oxides is a promising approach for identifying newsemiconductors for DSSCs due to their widely availablecompositions and tunable properties.21 The Cu(I) delafossiteternary oxides, CuMO2 (M = B, Al, Ga, In), are a group of p-type semiconductors with wide bandgap energies in the rangeof 3.4−4.0 eV and lower VB edges than NiO with a valence-band maximum (VBM) determined by the mixing of O 2porbitals and Cu 3d orbitals.22−25 Therefore, they should bemore transparent to visible light than NiO and be able toproduce higher photovoltages. The challenge is the difficulty insynthesizing the delafossite CuMO2 nanoparticles. For example,Nattestad et al. reported p-DSSCs based-on CuAlO2.

26

However, their synthesis of CuAlO2 was carried out by ahigh-temperature solid-state reaction followed by ball milling.The resulting CuAlO2 particles were large, and therefore theresultant porous CuAlO2 films exhibited limited surface area.

Received: March 23, 2012Accepted: April 10, 2012Published: April 10, 2012

Letter

pubs.acs.org/JPCL

© 2012 American Chemical Society 1074 dx.doi.org/10.1021/jz3003603 | J. Phys. Chem. Lett. 2012, 3, 1074−1078

In this work, we report delafossite CuGaO2 nanoplates andtheir first application for p-DSSCs that produce remarkablyhigh photovoltages. We chose CuGaO2 for the followingreasons: (1) its hydrothermal synthesis has been reported byPoeppelmeier27 and Jobic.28 Although these prior studies didnot produce nanoparticles, the hydrothermal synthesis ispromising to achieve smaller particle size than the CuAlO2

material synthesized from the solid-state synthesis. (2)CuGaO2’s band-structures have been calculated and exper-imentally investigated.29−32 It shows a direct forbidden bandgapof about 3.6−3.8 eV. An interesting phenomenon for thedelafossite CuMO2 is that the optically measured directbandgap increases from M = Al, Ga to In.31 Therefore,CuGaO2 should exhibit better transparency than CuAlO2. (3)The VB edge of CuGaO2 is about +0.6 V v.s. NHE, (i.e., +5.1eV below vacuum level),29 which is lower than that of NiO.33

CuGaO2-based p-DSSCs should thus be promising forproducing high photovoltages.Our detailed procedure of the hydrothermal synthesis is

shown in the SI, which was modified from a prior report.28

Consistent with its wide bandgap, our CuGaO2 product iswhite with a pale yellow tinge (Figure 1a). Therefore, theporous films made of these nanoplates barely compete with thedye sensitizers for light absorption. This is an attractiveadvantage over the dark-colored NiO films commonly used inp-DSSCs. The powder X-ray diffraction (XRD) showed that allobserved peak positions matched the diffraction pattern ofCuGaO2 with a delafossite structure from the InternationalCentre for Diffraction Data powder diffraction file (ICDD -PDF#41-0255) (Figure 1b). The scanning electron microscopy(SEM) and transmission electron microscopy (TEM) imagesshowed that the obtained particles had the nanoplatemorphology with an average size of about 200 nm and a

thickness of about 45 nm (Figure 1c,d). Each nanoplate issingle-crystalline as shown in the high-resolution TEM(HRTEM) (Figure 1e) and selected-area electron diffraction(SAED) (Figure 1f) images.CuGaO2 is known to be thermodynamically unstable below

600 °C, and the oxidation of Cu(I) in air would cause theformation of CuO and CuGa2O4.

34,35 This raised the concernof whether our CuGaO2 nanoplates could survive the thermalannealing process typically at 350−450 °C in the fabricationprocess of DSSCs, which is used to remove the organic residuesin the films and to enhance the connection between thenanoparticles. Therefore, XRD tests were performed to checkany possible phase changes after the samples were heated atdifferent temperatures in air. As shown in Figure 2, our

Figure 1. Characterization of our CuGaO2 nanoplates: (a) digital photo of the CuGaO2 (left) and NiO (right) particles dispersed in water; (b) XRDpattern, (c) SEM image, and (d) TEM image for the CuGaO2 nanoplates; (e) HRTEM image and (f) SAED pattern of a nanoplate along the [001]zone axis.

Figure 2. XRD pattern for CuGaO2 treated at different temperatures.

The Journal of Physical Chemistry Letters Letter

dx.doi.org/10.1021/jz3003603 | J. Phys. Chem. Lett. 2012, 3, 1074−10781075

CuGaO2 nanoplates are stable when the temperature is at orbelow 350 °C. This agrees with Kumekawa et al.’s conclusionthat the CuGaO2 is thermo-kinetically stable below 360 °C.35

However, when the temperature is higher than 400 °C, CuOand CuGa2O4 diffraction peaks also appear, indicating that thefollowing decomposition reaction takes place: 4CuGaO2 +

O2⎯ →⎯⎯⎯⎯⎯⎯⎯≥ °400 C

2CuGa2O4 + 2CuO. The above result confirms thatour CuGaO2 nanoplates are compatible with the fabricationprocedure as long as the annealing temperature is below 350°C.Films made of our CuGaO2 nanoplates with thicknesses of 3

μm were sensitized by the organic P1 dye (Figure 4, inset),which contains triphenylamine as the donor and was firstreported by Qin et al.12 Two electrolytes were used: 0.1 M I2/1.0 M LiI/methoxypropionitrile (MPN) or 0.1 M Co3+/2+(dtb-bpy)/MeCN electrolyte. We have measured the dependence ofVoc on the illumination intensity in order to estimate themaximum Voc that can potentially be obtained from theCuGaO2-based p-DSSCs. Excitingly, a saturation photovoltageof 464 mV has been observed in our CuGaO2-based p-DSSCswith the cobalt (III/II) electrolyte. It has been shown that theDSSC’s current−voltage characteristics follow the expression ofa constant current source connected in parallel with a diode:36

=⎛⎝⎜⎜

⎞⎠⎟⎟V

nkTe

J

Jlnoc

photo

0

Here, n is the ideality factor of the solar cell, k is theBoltzmann constant, T is the absolute temperature, e is theelectric charge, Jphoto is the photocurrent density, and J0 is thesaturation current density. Since the Jphoto is proportional to theincident light intensity I0,

37 the Voc should be proportional tothe logarithm of the incident illumination intensity I0. Thislinear relationship would hold until the Voc reaches itssaturation value, which is determined by the VB edge of thesemiconductor and the redox potential of the electrolyte. Oncethe Voc gets saturated, the increasing I0 would not furtherincrease the Voc, and that saturated value is the maximum Vocwe can obtain from the DSSCs.Voc as a function of the illumination intensity is shown in

Figure 3 for the CuGaO2- and NiO-based p-DSSCs in the 0.1M I2/ 1.0 M LiI/MPN electrolyte or 0.1 M Co3+/2+(dtb-bpy)/MeCN electrolyte. With the I3

−/I− redox electrolyte, the

CuGaO2-based p-DSSCs produce a saturation Voc of 243 mV.By contrast, the saturation photovoltage of NiO cells is only132 mV. More promisingly, with the Co3+/2+(dtb-bpy)electrolyte, the saturation Voc of CuGaO2-DSSCs increasesfurther to 464 mV, which is an increase of more than 100%compared to the NiO-DSSCs under the same conditions.The saturation photovoltage is determined by the difference

between the VBM of CuGaO2 and the redox potential of theredox couple. The redox potential is +0.35 V vs NHE for I3

−/I−, and +0.39 V vs NHE for Co3+/2+(dtb-bpy)/MeCN.7 Thelarge increase in the saturation photovoltage cannot beexplained by the small difference in the redox potentials ofthe two redox couples. Therefore, the most likely reason is thechange of the CuGaO2/electrolyte interface. We speculate thatdifferent preferential ion adsorption occurs in differentelectrolytes. The different surface potential can induce theshift of the VBM relative to the electrolyte and thus the changein photovoltage. Zeta potential measurements are ongoing toconfirm the surface charge.Figure 4 shows the photocurrent−voltage curves of the

CuGaO2 cells under 1 Sun AM 1.5 illumination. At thiscondition, the Voc is 180 mV, which is significantly higher thanthe common NiO-based p-DSSCs in the similar LiI/I2electrolyte solutions. More remarkably, when the 0.1 MCo3+/2+(dtb-bpy)/MeCN electrolyte was used, a Voc of 357mV was obtained. This is among the highest values reported todate (see Table S1 in the SI).4,6−9 The current density of thecell using I3

−/I− is higher than that of cobalt electrolyte. Arecent study from Hamann group shows that the recombina-tion in cobalt(III/II) electrolyte is faster than I3

−/I− due to alower outers-sphere reorganization energy compared to theinner-sphere reorganization energy for the I3

−/I− redoxcouple.38 We think our results are probably due to the samereason.The major limitation of our current CuGaO2 cells is their low

current densities. Control experiments have been carried out toconfirm the photocurrent is from the dye-sensitized CGO (seeSI Figure S2). The photocurrent density of DSSCs isdetermined by the product of the light harvesting efficiency,the hole injection efficiency from the P1 dye to CuGaO2, andthe collection efficiency of the hole carriers. The abovesaturation Voc measurements show that the VB edge ofCuGaO2 is 243 mV more positive than the redox potential of

Figure 3. Dependence of Voc on light intensity: (a) CuGaO2-DSSC and NiO-DSSC with simple I3−/I− electrolyte; (b) CuGaO2-DSSC and NiO-

DSSC with Co3+/2+(dtb-bpy) electrolyte.

The Journal of Physical Chemistry Letters Letter

dx.doi.org/10.1021/jz3003603 | J. Phys. Chem. Lett. 2012, 3, 1074−10781076

I3−/I−. Considering the redox potential of I3

−/ I− is 0.35 V(versus the normal hydrogen electrode, NHE; same below),39

the VB edge of CuGaO2 is 0.59 V. The highest occupiedmolecular orbital (HOMO) position of P1 dye is known to beat 1.35 eV.12 Therefore, the hole injection from the HOMO ofthe dye into the VB of CuGaO2 is thermodynamically favorablewith a driving force of 0.76 eV. With this consideration,together with the desired donor−acceptor geometry of the P1dye, we think the hole injection efficiency should not be thefactor limiting the photocurrent density.In order to evaluate the light harvesting efficiency, we have

measured the dye adsorption isotherms. As shown in Figure S3in the SI, the maximum dye loading for a 1-μm thick film is66.77 nanomol/cm2 for NiO and 23.35 nanomol/cm2 forCuGaO2. This result shows that the low dye loading is clearly afactor that limits the current density of CuGaO2 cells. Weattribute the low dye loading to the relatively large particle sizeof the CuGaO2 nanoplates, which results in a small surface areafor the adsorption of dye molecules. Another possible limitationis the recombination between CuGaO2 and the dye,considering recombination has been identified to be a majorproblem in NiO-based p-DSSCs.7,33,40 The low dye loading andthe recombination issue will also limit the Voc of the CuGaO2-DSSCs. Further work on reducing the CuGaO2-particle sizeand understanding the device physics is ongoing.In conclusion, we report the first application of CuGaO2

nanoplates in p-DSSCs that produce high photovoltages. Thesenanoplates are thermally stable up to 350 °C, and therefore, arecompatible with the DSSC fabrication process. In contrast tothe brown color of NiO, these CuGaO2 nanoplates are off-white. Therefore, the porous films made of these nanoplatesbarely compete with the dye sensitizers for light absorption.This presents an attractive advantage over the NiO filmscommonly used in p-DSSCs. A Voc of 357 mV has beenachieved when a Co3+/2+(dtb-bpy) electrolyte was used as theredox shuttle under 1 Sun AM 1.5 illumination. Moreremarkably, a saturation photovoltage of 464 mV has beenachieved with the increasing illumination intensity. Our currentefforts are to further decrease the sizes of the nanoplates andtherefore to increase the dye loading.

■ ASSOCIATED CONTENT

*S Supporting InformationDetails of synthesis and characterization. A summary table ofthe major prior works on NiO-based p-DSSCs. The current−voltage curves of the CuGaO2 p-DSSCs with different CuGaO2film thicknesses. The current−voltage curves of the dye-sensitized FTO cells as “blank tests”. The dye loading isothermmeasurement information. This material is available free ofcharge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATION

Corresponding Author*Phone: (+1) 614-247-7810; Fax: (+1) 614-292-1685; E-mail:wu@chemistry.ohio-state.edu.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThe authors acknowledge the funding support from the U.S.Department of Energy (Award No. DE-FG02-07ER46427).

■ REFERENCES(1) He, J.; Lindstrom, H.; Hagfeldt, A.; Lindquist, S.-E. Dye-Sensitized Nanostructured p-Type Nickel Oxide Film as a Photo-cathode for a Solar Cell. J. Phys. Chem. B 1999, 103 (42), 8940−8943.(2) He, J. J.; Lindstrom, H.; Hagfeldt, A.; Lindquist, S. E. Dye-Sensitized Nanostructured Tandem Cell-First Demonstrated Cell witha Dye-Sensitized Photocathode. Sol. Energy Mater. Sol. Cells 2000, 62(3), 265−273.(3) Odobel, F.; Le Pleux, L. Ø.; Pellegrin, Y.; Blart, E. NewPhotovoltaic Devices Based on the Sensitization of p-Type Semi-conductors: Challenges and Opportunities. Acc. Chem. Res. 2010, 43(8), 1063−1071.(4) Nattestad, A.; Mozer, A. J.; Fischer, M. K. R.; Cheng, Y. B.;Mishra, A.; Bauerle, P.; Bach, U. Highly Efficient Photocathodes forDye-Sensitized Tandem Solar Cells. Nat. Mater. 2010, 9 (1), 31−35.(5) Hagfeldt, A.; Graetzel, M. Light-Induced Redox Reactions inNanocrystalline Systems. Chem. Rev. (Washington, DC) 1995, 95 (1),49−68.(6) Gibson, E. A.; Smeigh, A. L.; Le Pleux, L.; Fortage, J.; Boschloo,G.; Blart, E.; Pellegrin, Y.; Odobel, F.; Hagfeldt, A.; Hammarstrom, L.A p-Type NiO-Based Dye-Sensitized Solar Cell with an Open-CircuitVoltage of 0.35 V. Angew. Chem., Int. Ed. 2009, 48 (24), 4402−4405.(7) Gibson, E. A.; Smeigh, A. L.; Le Pleux, L. C.; Hammarstrom, L.;Odobel, F.; Boschloo, G.; Hagfeldt, A. Cobalt Polypyridyl-BasedElectrolytes for p-Type Dye-Sensitized Solar Cells. J. Phys. Chem. C2011, 115 (19), 9772−9779.(8) Zhang, X. L.; Huang, F.; Nattestad, A.; Wang, K.; Fu, D.; Mishra,A.; Bauerle, P.; Bach, U.; Cheng, Y.-B. Enhanced Open-Circuit Voltageof p-Type DSC with Highly Crystalline NiO Nanoparticles. Chem.Commun. 2011, 47 (16), 4808−4810.(9) Kang, S. H.; Zhu, K.; Neale, N. R.; Frank, A. J. Hole Transport inSensitized CdS-NiO Nanoparticle Photocathodes. Chem. Commun.2011, 47 (37), 10419−10421.(10) Nakasa, A.; Usami, H.; Sumikura, S.; Hasegawa, S.; Koyama, T.;Suzuki, E. A High Voltage Dye-Sensitized Solar Cell using aNanoporous NiO Photocathode. Chem. Lett. 2005, 34 (4), 500−501.(11) Andrew, N.; Micheal, F.; Robert, K.; Yi-Bing, C.; Udo, B. Dye-Sensitized Nickel(II) Oxide Photocathodes for Tandem Solar CellApplications. Nanotechnology 2008, 19 (29), 295304.(12) Qin, P.; Zhu, H.; Edvinsson, T.; Boschloo, G.; Hagfeldt, A.; Sun,L. Design of an Organic Chromophore for p-Type Dye-SensitizedSolar Cells. J. Am. Chem. Soc. 2008, 130 (27), 8570−8571.(13) Sumikura, S.; Mori, S.; Shimizu, S.; Usami, H.; Suzuki, E.Photoelectrochemical Characteristics of Cells with Dyed and Undyed

Figure 4. Photocurrent−voltage curves for our CuGaO2-based p-DSSCs in an I3

−/I− electrolyte (blue) and in a Co(III/II) electrolyte(red). A “blank cell” without any dye-adsorption was also made(black) to confirm that the photocurrent was generated from the dye-sensitization.

The Journal of Physical Chemistry Letters Letter

dx.doi.org/10.1021/jz3003603 | J. Phys. Chem. Lett. 2012, 3, 1074−10781077

Nanoporous p-Type Semiconductor CuO Electrodes. J. Photochem.Photobiol., A 2008, 194 (2−3), 143−147.(14) Lepleux, L. C.; Chavillon, B.; Pellegrin, Y.; Blart, E.; Cario, L.;Jobic, S. P.; Odobel, F. Simple and Reproducible Procedure to PrepareSelf-Nanostructured NiO Films for the Fabrication of p-Type Dye-Sensitized Solar Cells. Inorg. Chem. 2009, 48 (17), 8245−8250.(15) Li, L.; Gibson, E. A.; Qin, P.; Boschloo, G.; Gorlov, M.;Hagfeldt, A.; Sun, L. Double-Layered NiO Photocathodes for p-TypeDSSCs with Record IPCE. Adv. Mater. 2010, 22 (15), 1759−1762.(16) Ji, Z.; Natu, G.; Huang, Z.; Wu, Y. Linker Effect in OrganicDonor−Acceptor Dyes for p-Type NiO Dye Sensitized Solar Cells.Energy Environ. Sci. 2011, 4 (8), 2818−2821.(17) Pellegrin, Y.; Le Pleux, L.; Blart, E.; Renaud, A.; Chavillon, B.;Szuwarski, N.; Boujtita, M.; Cario, L.; Jobic, S.; Jacquemin, D.; et al.Ruthenium Polypyridine Complexes as Sensitizers in NiO Based p-Type Dye-Sensitized Solar Cells: Effects of the Anchoring Groups. J.Photochem. Photobiol., A 2011, 219 (2−3), 235−242.(18) Mori, S.; Fukuda, S.; Sumikura, S.; Takeda, Y.; Tamaki, Y.;Suzuki, E.; Abe, T. Charge-Transfer Processes in Dye-Sensitized NiOSolar Cells. J. Phys. Chem. C 2008, 112 (41), 16134−16139.(19) Fernando, C. A. N.; Kitagawa, A.; Suzuki, M.; Takahashi, K.;Komura, T. Photoelectrochemical Properties of Rhodamine-C18Sensitized p-CuSCN Photoelectrochemical Cell (PEC). Sol. EnergyMater. Sol. Cells 1994, 33 (3), 301−315.(20) Nakabayashi, S.; Ohta, N.; Fujishima, A. Dye Sensitization ofSynthetic p-Type Diamond Electrode. Phys. Chem. Chem. Phys. 1999, 1(17), 3993−3997.(21) Tan, B.; Toman, E.; Li, Y. G.; Wu, Y. Y. Zinc Stannate(Zn2SnO4) Dye-Sensitized Solar Cells. J. Am. Chem. Soc. 2007, 129(14), 4162−4163.(22) Gillen, R.; Robertson, J. Band Structure Calculations of CuAlO2,CuGaO2, CuInO2 and CuCrO2 by Screened Exchange. Phys. Rev. B2011, 84 (3), 035125.(23) Shin, D.; Foord, J. S.; Payne, D. J.; Arnold, T.; Aston, D. J.;Egdell, R. G.; Godinho, K. G.; Scanlon, D. O.; Morgan, B. J.; Watson,G. W.; et al. Comparative Study of Bandwidths in Copper Delafossitesfrom X-ray Emission Spectroscopy. Phys. Rev. B 2009, 80 (23),233105.(24) Sheng, S.; Fang, G.; Li, C.; Xu, S.; Zhao, X. p-Type TransparentConducting Oxides. Phys. Status Solidi A 2006, 203 (8), 1891−1900.(25) Yanagi, H.; Kawazoe, H.; Kudo, A.; Yasukawa, M.; Hosono, H.Chemical Design and Thin Film Preparation of p-Type ConductiveTransparent Oxides. J. Electroceram. 2000, 4 (2), 407−414.(26) Nattestad, A. Dye-Sensitized CuAlO2 Photocathodes forTandem Solar Cell Applications. J. Photonics Energy 2011, 1 (1),011103.(27) Sheets, W. C.; Mugnier, E.; Barnabe, A.; Marks, T. J.;Poeppelmeier, K. R. Hydrothermal Synthesis of Delafossite-TypeOxides. Chem. Mater. 2006, 18 (1), 7−20.(28) Srinivasan, R.; Chavillon, B.; Doussier-Brochard, C.; Cario, L.;Paris, M.; Gautron, E.; Deniard, P.; Odobel, F.; Jobic, S. Tuning theSize and Color of the p-Type Wide Band Gap DelafossiteSemiconductor CuGaO2 with Ethylene Glycol Assisted HydrothermalSynthesis. J. Mater. Chem. 2008, 18 (46), 5647−5653.(29) Benko, F. A.; Koffyberg, F. P. Opto-electronic Properties ofCuAlO2. J. Phys. Chem. Solids 1984, 45 (1), 57−59.(30) Benko, F. A.; Koffyberg, F. P. The Optical Interband Transitionsof the Semiconductor CuGaO2. Phys. Status Solidi A 1986, 94 (1),231−234.(31) Nie, X.; Wei, S.-H.; Zhang, S. B. Bipolar Doping and Band-GapAnomalies in Delafossite Transparent Conductive Oxides. Phys. Rev.Lett. 2002, 88 (6), 066405.(32) Ueda, K.; Hase, T.; Yanagi, H.; Kawazoe, H.; Hosono, H.; Ohta,H.; Orita, M.; Hirano, M. Epitaxial Growth of Transparent p-TypeConducting CuGaO2 Thin Films on Sapphire (001) Substrates byPulsed Laser Deposition. J. Appl. Phys. 2001, 89 (3), 1790−1793.(33) Morandeira, A.; Boschloo, G.; Hagfeldt, A.; Hammarstrom, L.Photoinduced Ultrafast Dynamics of Comnarin 343 Sensitized p-

Type-Nanostructured NiO Films. J. Phys. Chem. B 2005, 109 (41),19403−19410.(34) Varadarajan, V.; Norton, D. P. CuGaO2 Thin Film SynthesisUsing Hydrogen-Assisted Pulsed Laser Deposition. Appl. Phys. A:Mater. Sci. Process. 2006, 85 (2), 117−120.(35) Kumekawa, Y.; Hirai, M.; Kobayashi, Y.; Endoh, S.; Oikawa, E.;Hashimoto, T. Evaluation of Thermodynamic and Kinetic Stability ofCuAlO2 and CuGaO2. J. Therm. Anal. Calorim. 2010, 99 (1), 57−63.(36) Sze, S. M. Physics of Semiconductor Devices, 2nd ed.; Wiley, Inc.:New York, 1981.(37) Huang, S. Y.; Schlichthorl, G.; Nozik, A. J.; Gratzel, M.; Frank,A. J. Charge Recombination in Dye-Sensitized Nanocrystalline TiO2Solar Cells. J. Phys. Chem. B 1997, 101 (14), 2576−2582.(38) Klahr, B. M.; Hamann, T. W. Performance Enhancement andLimitations of Cobalt Bipyridyl Redox Shuttles in Dye-Sensitized SolarCells. J. Phys. Chem. C 2009, 113 (31), 14040−14045.(39) Boschloo, G.; Hagfeldt, A. Characteristics of the Iodide/Triiodide Redox Mediator in Dye-Sensitized Solar Cells. Acc. Chem.Res. 2009, 42 (11), 1819−1826.(40) Smeigh, A. L.; Pleux, L. L.; Fortage, J.; Pellegrin, Y.; Blart, E.;Odobel, F.; Hammarstrom, L. Ultrafast Recombination for NiOSensitized with a Series of Perylene Imide Sensitizers ExhibitingMarcus Normal Behaviour. Chem. Commun. 2012, 48 (5), 678−680.

The Journal of Physical Chemistry Letters Letter

dx.doi.org/10.1021/jz3003603 | J. Phys. Chem. Lett. 2012, 3, 1074−10781078