Document not found! Please try again

An Energetic and Kinetic View on Cyclopentadithiophene Dye

Feb 3, 2011 - Ning Cai,. †,‡. Huijin Long,. †. Min Zhang,. †. Yinghui Wang,. † and Peng Wang*. ,†. †. State Key Laboratory of Polymer Ph...
0 downloads 0 Views 2MB Size
Download PDF
ARTICLE pubs.acs.org/JPCC

An Energetic and Kinetic View on Cyclopentadithiophene Dye-Sensitized Solar Cells: The Influence of Fluorine vs Ethyl Substituent Difei Zhou,†,‡ Ning Cai,†,‡ Huijin Long,† Min Zhang,† Yinghui Wang,† and Peng Wang*,† †

State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, China ‡ Graduate School, Chinese Academy of Sciences, Beijing 100039, China

bS Supporting Information ABSTRACT: A comprehensive understanding on the structure-property relationship of metal-free organic chromophores in dye-sensitized solar cells will form the basis of further design of advanced materials and realization of considerable enhancement of device performance, boosting the widespread utilization of solar energy at an affordable cost. In this paper we report profound influences of the fluorine versus ethyl substituent of the cyclopentadithiophene conjugated spacer of push-pull photosensitizers, upon the energetic and kinetic characteristics of dye-sensitized solar cells. Joint electrochemical and spectral measurements reveal that with respect to the fluorine-containing dye, the ethyl counterpart exhibits a 0.18 V more negative excited-state redox potential upon anchoring on titania, which overwhelms the 0.01 eV upward conduction band edge shift of the dye-coated titania film. The ethyl-correlated more favorable dye/titania interface energetics benefits a remarkably higher electron injection yield, which is verified with the transient emission and photocurrent action spectrum measurements. Furthermore, analysis on the electrical impedance data affords a notably smaller recombination reaction order for the cell made from the ethyl-substituting chromophore than that of the fluorine counterpart, accounting for a remarkably suppressed interfacial charge recombination with triiodide and thus explaining the observed higher open-circuit photovoltage.

1. INTRODUCTION One of the momentous global challenges is to transform inexhaustible sunlight into sustainable energy reserves, such as clean electricity and solar fuels.1 In order to attain the viable conversion of solar to electricity at an affordable cost, both scientists and engineers have been toiling to boost the performance of dyesensitized solar cells (DSCs), one of the cutting-edge photovoltaic technologies.2 The peculiar trait of this inorganic/organic hybrid photovoltaic cell rests with a nanoscale interpenetrating network, based on a mesoporous wide-band-gap oxide film coated with a visible and even infrared photon response chromophore, and further permeated with a hole-transport media. This delicate architecture can fully separate two critical physical processes of solar cells—light absorption and carrier transport, allowing the utilization of low-quality materials for efficient solarto-electricity conversion. A proper energy level alignment between the excited-state dye molecules and the electron-accepting species of oxide semiconductors would enable almost quantitative electron injection. Suppression of the interfacial charge recombination of photoinjected electrons with holes carried by electrolytes and/or oxidized dye molecules is intimately correlated with the r 2011 American Chemical Society

ultimate power output of a DSC. During the foregone two decades, consecutive material innovations and device engineering have been devoted to the efficiency enhancement of this photoelectrochemical device. Heretofore, a DSC made by the Sharp research group has pulled off a validated efficiency record of 11.1% by employing a panchromatic ruthenium photosensitizer.3 In view of the resource restriction of the noble-metal ruthenium, the exploration of metal-free push-pull organic chromophores for efficient dye-sensitized solar cells has attracted tremendous attention in recent years,4-20 primarily owing to the raw materials abundance. Aside from the electron donor and acceptor in a typical push-pull dye, the conjugated bridging segment is widely recognized of its significance for performance control of DSCs. In this regard, active investigations on thiophene-based organic photosensitizers have been witnessed since its initial incorporation in coumarin chromophores for dyesensitized solar cells.4 The structural versatility of thiophene Received: October 30, 2010 Revised: December 27, 2010 Published: February 3, 2011 3163

dx.doi.org/10.1021/jp110384n | J. Phys. Chem. C 2011, 115, 3163–3171

The Journal of Physical Chemistry C

ARTICLE

Figure 1. (A) Molecular structures of C223 and C224. (B) Plots of molar absorption coefficient as a function of wavelength for the chlorobenzene solutions. (C) Wavelength-dependent absorptions of dye-coated titania films immersed in the practical electrolyte for cell fabrication. (D) Photocurrent action spectra.

derivatives21 furnishes a flexible tailoring of dye molecules, which on the other side could bring on complicated substituentcorrelated photovoltaic behaviors of DSCs based thereon. Despite a continuous progress on organic-dye-based DSCs in past years, there is still an efficiency gap to fill when compared to the certificated ruthenium counterpart. This presents an ardent need for insightful understandings on the structure-property relationship of metal-free organic chromophores so as to expedite future development of promising sensitizers. Herein we investigate the profound impacts of two substituting groups with disparate electronic features, i.e. fluorine and ethyl, on the bridging carbon atom of the cyclopentadithiophene conjugated spacer, upon the photocurrent action spectra and current-voltage characteristics of organic push-pull dye-sensitized solar cells. Molecular structures of these two dyes are shown in Figure 1A. Our preliminary quantum calculations show that the employment of ethyl not only tunes both the dye HOMO and LUMO levels relative to those of the fluorine counterpart but also results in a distinguishable electron richness of the conjugated cyclopentadithiophene segment, as illustrated in Figure S1 and Table S1 in the Supporting Information. We expect that the noticeable substituent-correlated electronic features as well as the possibly different intermolecular interactions may effectively affect the critical energetic and kinetic characteristics of the titania/ dye/electrolyte interface, which will be scrutinized by means of electrochemical and photophysical measurements.

2. EXPERIMENTAL SECTION 2.1. Materials. Acetonitrile (AN) and chlorobenzene were distilled before use. All other solvents and reagents, unless otherwise stated, were of puriss quality and used as received. Phosphoryl trichloride, N-bromosuccinimide, Pd(OAc)2, 2-dicyclohexylphosphino-20 ,60 -dimethoxybiphenyl (SPhos), 2-cyanoacetic acid, piperidine, lithium iodide (LiI), iodine (I2), lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), 3α,7α-dihyroxy-5β-cholic acid (cheno), guanidinium thiocyanate (GNCS), and 4-tert-butylpyridine (TBP) were purchased

from Aldrich. 4,4-Difluoro-4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene (1a),22 4,4-diethyl-4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene (1b),23 4,4,5,5-tetramethyl-2-{4-[N,N-bis(4-hexyloxyphenyl)amino]phenyl}1,3,2-dioxaborolane,24 1,3-dimethylimidazolium iodide (DMII),25 1,3-dimethylimidazolium bis(trifluoromethanesulfonyl)imide (DMITFSI),25 and 1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide (EMITFSI)25 were synthesized according to literature methods. The scattering TiO2 paste (WER2-O) was received as a gift from Dyesol. 2.2. Dye Preparation. The synthetic route of C223 and C224 is presented in Scheme S1 of the Supporting Information. 4,4-Difluoro-4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene-2-carbaldehyde (2a). To a cold solution of 1a (0.140 g, 0.65 mmol) and N,N-dimethylformide (0.130 g, 1.82 mmol) in 1,2-dichloroethane (7 mL) at 0 °C was added phosphoryl trichloride (0.119 g, 0.77 mmol) under argon. The reaction solution was stirred at 80 °C for 20 h and then cooled to room temperature. Afterwards saturated sodium acetate aqueous solution (15 mL) was added. The mixture was further stirred at room temperature for 2 h. The crude product was extracted into dichloromethane, and the organic layer was washed with brine and water, and dried over anhydrous sodium sulfate. After removing solvent under reduced pressure, the residue was purified by column chromatography (ethyl acetate/petroleum ether 60—90 °C, 1/10, vol/vol) on silica gel to yield a yellow solid (0.155 g, 99% yield). 1H NMR (600 MHz, CDCl3) δ: 9.83 (s, 1H), 7.70 (s, 1H), 7.41 (d, J = 5.4 Hz, 1H), 7.18 (d, J = 4.8 Hz, 1H). 13C NMR (150 MHz, CDCl3) δ: 182.14, 149.77, 147.91 (t, J = 111.0 Hz), 145.43, 145.20 (t, J = 114.0 Hz), 138.99, 131.40, 129.16, 121.69, 116.48 (t, J = 984.0 Hz). MS (ESI) m/z calcd for (C10H4F2OS2): 241.97. Found: 242.98 ([M þ H]þ). Anal. Calcd for C10H4F2OS2: C, 49.58; H, 1.66. Found: C, 49.62; H, 1.63. 4,4-Diethyl-4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene-2-carbaldehyde (2b). To a cold solution of 1b (0.500 g, 2.13 mmol) and N,N-dimethylformide (0.430 g, 5.97 mmol) in 1,2-dichloroethane (10 mL) at 0 °C was added phosphoryl trichloride (0.390 g, 3164

dx.doi.org/10.1021/jp110384n |J. Phys. Chem. C 2011, 115, 3163–3171

The Journal of Physical Chemistry C 2.52 mmol) under argon. The reaction solution was stirred at the same temperature for 4 h. The same workup procedure as above (2a). Yellowish solid. Yield: 97%. 1H NMR (400 MHz, CDCl3) δ: 9.84 (s, 1H), 7.57 (s, 1H), 7.41 (d, J = 4.8 Hz, 1H), 6.99 (d, J = 4.8 Hz, 1H), 1.94 (m, 4H), 0.61 (t, J = 7.4 Hz, 6H). 13C NMR (150 MHz, CDCl3) δ: 182.57, 161.75, 157.42, 148.06, 143.22, 135.93, 130.12, 129.63, 121.79, 54.77, 30.06, 9.04. MS (ESI) m/z calcd for (C14H14OS2): 262.05. Found: 263.06 ([M þ H]þ). Anal Calcd for (C14H14OS2): C, 64.08; H, 5.38. Found: C, 64.04; H, 5.41. 6-Bromo-4,4-difluoro-4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene-2-carbaldehyde (3a). To a cold solution of 2a (0.141 g, 0.58 mmol) in tetrahydrofuran (20 mL) was added N-bromosuccinimide (0.114 g, 0.64 mmol) at 0 °C under argon. The reaction mixture was warmed to room temperature and stirred for 5 h, and then water (20 mL) was added. The crude product was extracted into dichloromethane, and the organic layer was dried over anhydrous sodium sulfate. After solvent was removed under reduced pressure, the residue was purified by column chromatography (ethyl acetate/petroleum ether 60-90 °C, 1/ 10, vol/vol) on silica gel to yield a yellow solid (0.180 g, 97% yield). 1H NMR (600 MHz, CDCl3) δ: 9.83 (s, 1H), 7.69 (s, 1H), 7.19 (s, 1H). 13C NMR (150 MHz, CDCl3) δ: 182.09, 149.05, 146.70 (t, J = 114.0 Hz), 145.75, 144.16 (t, J = 111.0 Hz), 139.00, 129.02, 124.52, 118.34, 116.28 (t, J = 987.0 Hz). MS (ESI) m/z calcd for (C10H3BrF2OS2): 319.88. Found: 320.86 ([M þ H]þ). Anal. Calcd for C10H3BrF2OS2: C, 37.40; H, 0.94. Found: C, 37.47; H, 0.83. 6-Bromo-4,4-diethyl-4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene-2-carbaldehyde (3b). The same procedure as above (3a). Yellowish solid. Yield: 96%. 1H NMR (600 MHz, CDCl3) δ: 9.84 (s, 1H), 7.56 (s, 1H), 7.02 (s, 1H), 1.92 (m, 4H), 0.61 (t, J = 7.2 Hz, 6H). 13C NMR (150 MHz, CDCl3) δ: 182.56, 160.40, 156.44, 147.10, 143.58, 136.30, 129.86, 124.90, 116.26, 55.61, 29.98, 9.01. MS (ESI) m/z calcd for (C14H13BrOS2): 339.96. Found: 340.98 ([M þ H]þ). Anal. Calcd for C14H13BrOS2: C, 49.27; H, 3.84. Found: C, 49.34; H, 3.79. 6-{4-[N,N-Bis(4-hexyloxyphenyl)amino]phenyl}-4,4-difluoro-4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene-2-carbaldehyde (4a). To a suspended solution of 3a (0.170 g, 0.53 mmol), Pd(OAc)2 (0.002 g, 0.01 mmol), SPhos (0.004 g, 0.01 mmol), and K3PO4 (0.561 g, 2.65 mmol) in dioxane/H2O (5/1, vol/ vol) (12 mL) was added 4,4,5,5-tetramethyl-2-{4-[N,N-bis(4hexyloxyphenyl)amino]phenyl}-1,3,2-dioxaborolane (0.363 g, 0.64 mmol). The reaction mixture was reacted at 40 °C for 2 h under argon and then water (50 mL) added. The crude compound was extracted into ethyl acetate, washed with brine and water, and dried over anhydrous sodium sulfate. After solvent was removed under reduced pressure, the residue was purified by column chromatography (ethyl acetate/petroleum ether 60-90 °C, 1/50, vol/vol) on silica gel to yield a red solid (0.344 g, 95% yield). 1H NMR (600 MHz, DMSO-d6) δ: 9.82 (s, 1H), 8.15 (s, 1H), 7.65 (s, 1H), 7.53 (d, J = 9.0 Hz, 2H), 7.07 (d, J = 9.0 Hz, 4H), 6.93 (d, J = 9.0 Hz, 4H), 6.74 (d, J = 9.0 Hz, 2H), 3.94 (t, J = 6.3 Hz, 4H), 1.69 (m, 4H), 1.41 (m, 4H), 1.31 (m, 8H), 0.88 (t, J = 6.8 Hz, 6H). 13C NMR (150 MHz, DMSO-d6) δ: 183.59, 155.72, 152.16, 150.16, 149.04, 147.67 (t, J = 111.0 Hz), 144.99, 142.90 (t, J = 111.0 Hz), 139.06, 135.34, 130.04, 127.25, 126.32, 123.92, 118.28, 116.75 (t, J = 972.0 Hz), 115.75, 115.52, 67.61, 30.96, 28.66, 25.16, 22.03, 13.85. MS (ESI) m/z calcd for (C40H41F2-

ARTICLE

NO3S2): 685.25. Found: 686.51([M þ H]þ). Anal. Calcd for C40H41F2NO3S2: C, 70.04; H, 6.03; N, 2.04. Found: C, 70.10; H, 5.93; N, 2.09. 6-{4-[N,N-Bis(4-hexyloxyphenyl)amino]phenyl}-4,4-diethyl4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene-2-carbaldehyde (4b). The same procedure as above (4a). Orange oil. Yield: 96%. 1H NMR (600 MHz, DMSO-d6) δ: 9.81 (s, 1H), 7.97 (s, 1H), 7.51 (d, J = 9.0 Hz, 2H), 7.45 (s, 1H), 7.04 (d, J = 8.4 Hz, 4H), 6.92 (d, J = 9.0 Hz, 4H), 6.77 (d, J = 9.0 Hz, 2H), 3.94 (t, J = 6.3 Hz, 4H), 1.95 (m, 4H), 1.70 (m, 4H), 1.41 (m, 4H), 1.31 (m, 8H), 0.88 (t, J = 6.8 Hz, 6H), 0.55 (t, J = 7.5 Hz, 6H). 13C NMR (150 MHz, DMSO-d6) δ: 183.17, 163.05, 156.53, 155.46, 149.16, 148.35, 146.99, 142.53, 139.38, 132.62, 131.58, 126.86, 126.10, 125.32, 119.02, 116.86, 115.47, 67.59, 54.57, 30.95, 29.17, 28.66, 25.15, 22.02, 13.84, 8.88. MS (ESI) m/z calcd for (C44H51NO3S2): 705.33. Found: 706.67 ([M þ H]þ). Anal. Calcd for C44H51NO3S2: C, 74.85; H, 7.28; N, 1.98. Found: C, 74.92; H, 7.23; N, 2.01. 2-Cyano-3-{6-{4-[N,N-bis(4-hexyloxyphenyl)amino]phenyl}-4,4-difluoro-4H-cyclopenta[2,1-b:3,4-b0 ]dithiophene-2yl}acrylic Acid (5a, C223). To a stirred solution of 4a (0.220 g, 0.32 mmol) and cyanoacetic acid (0.082 g, 0.96 mmol) in chloroform (18 mL) was added piperidine (0.191 g, 2.25 mmol). The reaction mixture was refluxed under argon for 18 h and then acidified with 2 M hydrochloric acid aqueous solution (20 mL). The crude product was extracted into chloroform, washed with water, and dried over anhydrous sodium sulfate. After solvent was removed under reduced pressure, the residue was purified by flash chromatography with chloroform and methanol/chloroform (1/10, vol/vol) in turn as the eluent to yield a purple-black powder (0.217 g, 90% yield). 1H NMR (600 MHz, DMSO-d6 ) δ: 13.69 (s, 1H), 8.46 (s, 1H), 8.07 (s, 1H), 7.64 (s, 1H), 7.53 (d, J = 8.4 Hz, 2H), 7.06 (d, J = 8.4 Hz, 4H), 6.93 (d, J = 9.0 Hz, 4H), 6.74 (d, J = 8.4 Hz, 2H), 3.94 (t, J = 6.3 Hz, 4H), 1.70 (m, 4H), 1.41 (m, 4H), 1.31 (m, 8H), 0.88 (t, J = 6.8 Hz, 6H). 13 C NMR (150 MHz, DMSO-d6) δ: 163.65, 155.71, 152.66, 150.76, 149.02, 147.52 (t, J = 111.0 Hz), 146.47, 142.82 (t, J = 111.0 Hz), 139.03, 138.08, 135.87, 131.09, 127.23, 126.29, 123.95, 118.24, 116.95, 116.68 (t, J = 972.0 Hz), 115.69, 115.48, 97.01, 67.61, 30.98, 28.67, 25.18, 22.04, 13.85. HR-MS (MALDI): m/z calcd for (C 43H42F2 N2O4S2 ): 752.25541. Found: 752.25539. Anal. Calcd for C43H42 F2N2O4 S2: C, 68.59; H, 5.62; N, 3.72. Found: C, 68.64; H, 5.73; N, 3.67. 2-Cyano-3-{6-{4-[N,N-bis(4-hexyloxyphenyl)amino]phenyl}-4,4-diethyl-4H-cyclopenta[2,1-b:3,4-b 0 ]dithiophene-2yl}acrylic Acid (5b, C224). The same procedure as above (5a, C223). Purple powder. Yield: 84%. 1 H NMR (600 MHz, DMSO-d 6 ) δ: 13.36 (s, 1H), 8.42 (s, 1H), 7.93 (s, 1H), 7.53 (d, J = 9.0 Hz, 2H), 7.47 (s, 1H), 7.04 (d, J = 8.4 Hz, 4H), 6.92 (d, J = 9.0 Hz, 4H), 6.78 (d, J = 8.4 Hz, 2H), 3.94 (t, J = 6.3 Hz, 4H), 1.93 (m, 4H), 1.70 (m, 4H), 1.41 (m, 4H), 1.31 (m, 8H), 0.88 (t, J = 6.8 Hz, 6H), 0.56 (t, J = 7.2 Hz, 6H). 13 C NMR (150 MHz, DMSO-d 6 ) δ: 164.23, 163.79, 156.55, 155.53, 150.35, 149.22, 148.51, 147.04, 139.31, 135.51, 133.04, 132.86, 126.95, 126.20, 125.20, 118.90, 117.37, 116.88, 115.49, 93.34, 67.60, 54.35, 30.95, 29.23, 28.66, 25.16, 22.02, 13.86, 8.88. HR-MS (MALDI): m/z calcd for (C 47 H 52 N 2 O 4 S 2 ): 772.33685. Found: 772.33681. Anal. Calcd for C 47 H 52 N 2 O 4 S 2 : C, 73.02; H, 6.78; N, 3.62. Found: C, 73.09; H, 6.69; N, 3.70. 3165

dx.doi.org/10.1021/jp110384n |J. Phys. Chem. C 2011, 115, 3163–3171

The Journal of Physical Chemistry C

ARTICLE

Table 1. Spectral and Electrochemical Data of C223 and C224 dye

a λabs max (nm)

3 -1 a εabs cm-1) max (10 M

b λabs max,film (nm)

3 -1 b Rabs max,film (10 cm )

E0-0c (eV)

φDþ/Dd (V)

φD*/Dþe (V)

C223

555

33

490

5.2

2.01

-5.50

-3.49

C224

548

50

493

4.2

2.12

-5.43

-3.31

a abs The maximum absorption wavelength (λabs max) and maximum molar absorption coefficient (εmax) are derived from the electronic absorption abs measurements on the chlorobenzene solutions. b The maximum absorption wavelength (λabs max,film) and absorption coefficient (Rmax,film) are c measured with dye-coated titania films immersed in the practical electrolyte for cell fabrication. The zero-zero transition energy (E0-0) is estimated from the intersection of normalized absorption and emission spectra. d The ground-state (φDþ/D) redox potentials are reported vs vacuum. e The excited-state redox potentials (φD*/Dþ) are reported vs vacuum. φD*/Dþ is calculated by equation φD*/Dþ = φDþ/D þ E0-0/e without considering any entropy change during light excitation.

2.3. Electronic Absorption and Voltammetric Measurements. Electronic absorption spectra were recorded on a Per-

kin-Elmer Lambda 900 spectrometer. A CHI660C electrochemical workstation was used for cyclic voltammetric measurements in combination with a three-electrode electrochemical cell equipped with a platinum gauze counter electrode, a Ag/AgCl (sat. KCl) reference electrode, and a dye-coated titania film on fluorine-doped tin oxide (FTO) as working electrode. The redox potentials reported in this paper were calibrated with ferrocene as the internal reference (-5.14 V vs vacuum). 2.4. Cell Fabrication. A bilayer titania film was used as the negative electrode of DSCs. On a precleaned fluorine-doped tin oxide (FTO) conducting glass electrode (Nippon Sheet Glass, Solar, 4 mm thick), a 7-μm-thick transparent layer consisting of ∼23-nm-sized titania particles was first screen-printed and further coated by a 5-μm-thick second layer of scattering titania particles (WER2-O, Dyesol). The film thickness was monitored by a benchtop Ambios XP-1 stylus profilometer. The detailed preparation procedures of titania nanocrystals, paste for screenprinting, and nanostructured titania film were reported in a previous paper.26 A titania electrode (∼0.28 cm2) was stained by immersing it into a solution containing 150 μM C223 or C224 dye and 300 μM cheno coadsorbent in chlorobenzene for 5 h. After being washed with acetonitrile and dried by air flow, the dyecoated titania electrode was assembled with a thermally platinized FTO positive electrode possessing an electrolyte-perfusion hole, which was beforehand produced with a sand-blasting drill. The two electrodes were separated by a 30 μm thick Bynel (DuPont) hot-melt gasket and sealed up by heating. The internal space was perfused with an electrolyte via a vacuum backfilling system. Electrolyte composition: 1.0 M DMII, 20 mM I2, 50 mM LiI, 1.0 M TBP, and 0.1 M GNCS in acetonitrile. Ultimately, the hole on the positive electrode was closed hermetically with a Bynel sheet and a thin glass cover by heating. 2.5. Photocurrent Action Spectrum, Current-Voltage, and Electrical Impedance Measurements. These experiments were performed as described in our previous paper.27 2.6. Transient Absorption and Emission Measurements. Transient absorption measurements were performed with a LP920 laser flash spectrometer using a nanosecond tunable OPOLett-355II laser to supply a pump light. The sample was kept at a 45° angle with respect to the excitation beam. The probe light from a xenon arc lamp was passed through various optical elements, a testing sample, and a monochromator before being detected by a fast photomultiplier tube and recorded with a TDS 3012C digital signal analyzer. Emission spectrum and time-correlated single photon counting (TCSPC) measurements were carried out with a LifeSpec-II spectrometer employing an EPL485 pulsed laser diode and a Hamamatsu H5773-04 photomultiplier.

2.7. Computation. The ground-state geometries of C223Ti(OH)3 and C224Ti(OH)3 were fully optimized without symmetry constraints by the density functional theory (DFT) method,28 with hybrid Beck’s three-parameter functional and Lee-YangParr functional (B3LYP).29,30 The LANL2DZ basis set was used for the Ti atom and 6-31G (d, p) basis set for other atoms. All calculations reported here were performed with the Gaussian09 program package.31

3. RESULTS AND DISCUSSION As Figure 1B presents, we first recorded the electronic absorption spectra of C223 and C224 dissolved in chlorobenzene so as to preliminarily evaluate the substituent influence on the light capture capacity, and the detailed parameters are collected in Table 1. The C223 dye featuring two fluorine substituents on the bridging carbon of cyclopentadithiophene has a maximum molar absorption coefficient (ε) of 33  103 M-1 cm-1 at 555 nm, which is remarkably augmented to 50  103 M-1 cm-1 at 548 nm upon replacing fluorine with ethyl for C224. This observation could be rationalized in terms of an evidently electron-richer character of the ethyl-tethered cyclopentadithiophene segment than the fluorine counterpart, as shown in Table S1 of the Supporting Information. Keeping in mind that either deprotonation concomitant with the carboxylate grafting of sensitizers on titania nanocrystals or solvent polarity can affect intramolecular charge transfer transition, we further measured the electronic absorption spectra of C223 and C224 coated on a 7 μm thick titania film and impregnated in the realistic electrolyte for cell fabrication. As presented in Figure 1C, an evident hypsochromic effect can be noted for both dyes, which could be ascribed to a weaker electron-withdrawing capability of the carboxylatetitanium assembly than the carboxylic acid. Moreover, we notice that the fluorine-containing C223 dye confers a nanocrystalline titania film with an even higher maximum absorption coefficient (R) than the ethyl counterpart, albeit its remarkably lower maximum molar absorption coefficient. By comparing the absorbance change of a dye solution (initially 150 μM in chlorobenzene) before and after dye up-taking with a titania film, we derived the surface coverages (Γ) of C223 and C224 anchored on titania nanocrystals, being 3.26  10-10 and 2.08  10-10 mol cm-2, respectively. Thereby, we can conclude that the increased dye loading of C223 relative to C224 significantly contributes to the observed superior light-harvesting capacity. Density functional theory (DFT) calculations on the geometries of C223Ti(OH)3 and C224Ti(OH)3 (Figure 2) show that the substituent alteration from fluorine to ethyl seems not to noticeably change the molecular footprint. Thereby, the ∼60% larger Γ of the fluorine-containing dye relative to the ethyl counterpart could be attributed to a much closer stacking of C223 on titania film. 3166

dx.doi.org/10.1021/jp110384n |J. Phys. Chem. C 2011, 115, 3163–3171

The Journal of Physical Chemistry C

ARTICLE

Figure 2. Optimized molecular geometries of (A) C223Ti(OH)3 and (B) C224Ti(OH)3.

We further measured cyclic voltammograms to probe the possible impact of dissimilar stacking modes upon the characteristics of diffusive lateral charge transfer32 within the grafted dye layer on titania. The ground-state redox potential (φDþ/D) can be determined by averaging the anodic and cathodic peak potentials of cyclic voltammograms shown in Figure 3A, being 0.36 and 0.29 V versus the ferrocene/ferrocenium reference for C223 and C224, respectively, anchored on titania. The 0.07 V more negative φDþ/D of C224 relative to C223 can be well understood on account of the electron-deficient character of fluorine with respect to ethyl. Furthermore, the peak current of anodic wave (Ipa) can be correlated with the scan rate (ν) through the following equation33 Ipa ¼ ð2:69  105 ÞAs cm n3 = 2 Dapp 1 = 2 ν1 = 2

ð1Þ

At a given film geometry with invariant electrode area (As), film thickness, and porosity, the dye molecule concentration (cm) is proportional to Γ. Herein, n is the transferred electron number and Dapp refers to the apparent hole diffusion coefficient. Through comparing the fitting slopes of data shown in Figure 3B, the Dapp in the molecule layer composed of the fluorine-containing dye is deducted to be ∼5 times higher than that of the ethyl counterpart. It seems that with respect to the case with C224, there is a stronger intermolecular electronic interaction of C223 molecules grafted on titania. We further carried out the incident photon-to-collected electron conversion efficiency (IPCE) measurement of a dye-coated titania film in conjunction with an acetonitrile electrolyte containing the iodide/triiodide redox couple. As shown in Figure 1D, compared to the fluorine counterpart, a perceptible blue shift of the photocurrent onset wavelength can be noted upon the substitution with the ethyl group, in good agreement with the preceding electronic absorption data (Figure 1C). However, albeit the remarkably lower absorptions of the C224 dye-coated titania film in the visible spectral region, the corresponding cell exhibits a remarkably higher IPCE summit of 93%, contrasting that of 66% upon the replacement of ethyl with fluorine.

Figure 3. (A) Cyclic voltammograms of dye-coated titania films. Scan rate: 60 mV s-1. Electrolyte: 1.0 M DMITFSI in acetonitrile. (B) Plots of anodic peak current (Ipa) as a function of square root of scan rate (ν1/2). The solid lines are fittings in terms of eq 1.

To comprehend the significant IPCE maximum variation, the time-correlated single photon counting (TCSPC) technique was deployed to scrutinize the yield of electron injection from the excited-state photosensitizer to titania, which could make a critical contribution to the IPCE summit.34-36 Using a nanoporous alumina film stained with C223 or C224, a control cell was first constructed in combination with the practical electrolyte and exhibited a strong emission upon laser excitation at 488 nm (dashed lines in Figure 4). Considering the thermodynamic retardance of electron injection at the dye/alumina interface, these photoluminescence decays can be attributed to the irradiative and nonirradiative deactivations of excited-state dye molecules (D*). However, remarkable photoluminescence quenching was observed upon the replacement of alumina with titania, implying the occurrence of electron injection at the energy-offset dye/ titania interface, which should be further proved by the succeeding transient absorption measurements. Through the calculation with integrals of areas under the emission traces on titania and those of alumina, the electron injection yields (ηinj) are roughly estimated to be 81% and 99% for cells made from the fluorineand ethyl-substituting dyes, respectively, echoing the notable IPCE height difference. It is suspected that the substituent-correlated molecular energy levels can profoundly influence the energy offset at the dye/ titania interface, leading to dissimilar electron injection yields. Thereby, we first measured the absorption and emission spectra of a dye-coated titania films (Figure S2, Supporting Information), from which the corresponding zero-zero transition energies (E0-0) can be estimated to be 2.01 and 2.12 eV for C223 and C224, respectively, by taking the intersections of the absorption and emission spectra. In further combination with the aforementioned ground-state redox potentials (C223, -5.50 V and C224, -5.43 V versus vacuum), we obtain the excited-state redox potentials (φD*/Dþ) of C223 and C224 grafted on titania, being -3.49 and -3.31 V versus vacuum. On the other side, the 3167

dx.doi.org/10.1021/jp110384n |J. Phys. Chem. C 2011, 115, 3163–3171

The Journal of Physical Chemistry C

Figure 4. Time-correlated emission decay traces. Excitation wavelength: 488 nm. The emission intensity (I) was corrected in terms of the film absorbance at 488 nm and further normalized with respect to the emission maximum of a dye-coated alumina film (Imax,alumina).

Figure 5. Plots of electron transport resistance (Rt) of titania film vs applied potential bias (V).

titania conduction band edge (Ec) can be derived through a theoretical analysis on the dependence of electron transport resistance (Rt) of the titania film on potential bias (V)37     Ec -EF,redox eV Rt ¼ R0 exp β exp -β ð2Þ kB T kB T where R0 represents the electron transport resistance in the conduction band and is taken as an invariant for a titania film with a given geometry. The potential bias V = (EF,n - EF,redox)/e, where EF,n and EF,redox represent Fermi levels of the titania film and redox electrolyte. kB is the Boltzmann constant and T the absolute temperature. β is introduced to account for the possible nonideal thermodynamical behavior of electrons in the titania conduction band.38 By modeling the impedance spectra in terms of a proper equivalent circuit,39,40 Rt can be obtained as a function of V, as shown in Figure 5. Fitting on the Rt data with eq 2 yields the (Ec - EF,redox) value to be 0.97 or 0.98 eV for a titania film coated with C223 or C224. Moreover, with the aid of a homemade three-electrode electrochemical system composed of a platinum foil as auxiliary electrode, a 5 μm radius platinum ultramicroelectrode as working electrode, and a reference electrode constructed with a platinum wire dipped in an electrolytefilling glass tube, the lower end of which was sealed by a porous ceramic frit, we measured the square-wave voltammogram (Figure S3, Supporting Information) of ferrocene dissolved in acetonitrile. By determining the formal potential of ferrocene to be 0.28 V versus EF,redox, the EF,redox of the electrolyte used for our

ARTICLE

Figure 6. Absorption change after applying a potential bias to a C223 or C224 dye-coated titania film immersed in EMITFSI. The applied potentials are reported vs the ferrocene/ferrocenium reference.

cell fabrication can be further derived to be -4.86 eV versus vacuum, affording the Ec values of -3.89 or -3.88 eV versus vacuum for a titania film coated with C223 or C224 dye molecules. These analyses allow us to calculate the energy offsets at the dye/titania interfaces for electron injection, being 0.40 and 0.57 eV for cells made from C223 and C224, respectively. Evidently, with respect to the fluorine counterpart, the ethylcontaining dye benefits the interfacial electron injection with a remarkable 0.17 eV larger thermodynamic driving force, significantly contributing to the higher electron injection yield. It is also worthy of noting that by comparing the photoluminescence decays of C223 and C224 anchored on an alumina film, a shorter photoluminescence lifetime can be perceived for the fluorinesubstituted dye, which also has an adverse effect on the electron injection yield. The laser flash photolysis technique was further adopted to picture the kinetic competition between the dye cations (Dþ) recombination with photoinjected electrons and their regeneration by iodide, so as to analyze the possible impact of the dualchannel charge-transfer kinetics of dye cations upon the yield of interfacial charge separation.41-43 As depicted in Figure 6, spectroelectrochemical measurements disclose that the C223 and C224 cations both feature a strong absorption in the infrared region, in comparison with their neutral counterparts. Thereby we chose a monochromatic light at 782 nm to probe the dualchannel charge-transfer kinetics of dye cations, on account of the signal sensitivity of our measuring system. Excitation fluences were carefully controlled in our kinetic experiments to ensure that ∼8.3  1013 photons cm-2 (∼ 0.8 photon per nanoparticle) were absorbed by the nanocrystalline titania film during every laser pulse. As Figure 7 presents, the absorption signals could be well fitted to a stretched exponential decay function (mΔOD µ exp[-(t/τ)R]) to afford the half-reaction times (t1/2). In the presence of an inert electrolyte containing the inactive bis(trifluoromethanesulfonyl)amide (TFSI) anion, the absorption decays (a and b) mainly reflect the charge recombination between dye cations and titania electrons, producing half-reaction times of 2.6 and 3.5 ms for C223 and C224, respectively. However, upon the incorporation of electroactive iodide ions, the absorption decays were significantly accelerated (c and d), yielding a half-reaction time of 8 μs for the cell with C223 and a noticeably increased t1/2 of 17 μs for the ethyl counterpart C224. Moreover, the kinetic branch ratios are over 200 for both 3168

dx.doi.org/10.1021/jp110384n |J. Phys. Chem. C 2011, 115, 3163–3171

The Journal of Physical Chemistry C

ARTICLE

essentially indicates an nc fluctuation, which can be described by the following continuity equation Dnc D2 nc edAs ¼ Jg þ D0 2 -Jr Dt Dx

ð3Þ

Eq 3 accounts for the collective contributions of photocurrent generation, charge transport, and recombination to the free electron density in a working cell, where t represents the time scale and x is the spatial coordinate. d is the titania film thickness, and D0 is the conduction band electron diffusion coefficient. Jg and Jr refer to the respective current of photogeneration and that of interfacial charge recombination with triiodide. At the steady-state and open-circuit conditions where the Voc is defined, eq 3 affords Jg = Jr, while Figure 7. Transient absorption traces at 782 nm of dye-coated titania films. Film thickness: 12 μm. Curves a and b were recorded with films immersed in an inert electrolyte composed of 1.0 M DMITFSI, 50 mM LiTFSI, and 1.0 M TBP in AN, while curves c and d in an electroactive electrolyte consisting of 1.0 M DMII, 50 mM LiI, and 1.0 M TBP in AN. Pulse fluence: (a) 63 μJ cm-2 at 687 nm; (b) 52 μJ cm-2 at 668 nm; (c) 64 μJ cm-2 at 677 nm; (d) 53 μJ cm-2 at 658 nm. Smooth lines are stretched exponential fittings over raw data obtained by averaging 800 laser shots.

Jr ¼ k0 ½I3-R nc γ edAs

ð4Þ

Herein k0 is the charge-transfer rate constant and is supposed to depend on both reaction activation energy (Ea) and temperature. [I3-] and nc refer to the concentrations of triiodide and free electrons, and R and γ represent the corresponding reaction order. With nc ≈ n0 exp[eV/kBT] and the background electron density in titania, n0 = Nc exp[(EF,redox - Ec)/kBT), we derive that     EF,redox -Ec eV Jr ¼ k0 ½I3-R Nc γ exp γ exp γ edAs ð5Þ kB T kB T Herein, we define a characteristic recombination rate   EF,redox -Ec - R γ Kr ¼ k0 ½I3  Nc exp γ kB T

ð6Þ

to depict the electron-exchange behavior at the titania/electrolyte interface. From eqs 5 and 6 we have   kB T -1 Jg ln Kr ð7Þ V¼ γe edAs

Figure 8. J-V characteristics measured at an irradiation of 100 mW cm-2 AM1.5G sunlight.

photosensitizers, indicating that almost 100% of the dye cations can be intercepted by the electron-donating iodide anions. Thereby, we conclude that the difference in the dual-channel charge-transfer kinetic competition of dye cations does not make a concernful contribution to the observed IPCE height variation. We further recorded the photocurrent density-voltage (J-V) characteristics (Figure 8) of cells made from these two dyes to evaluate the substituent effect on photovoltaic parameters. The short-circuit photocurrent density (Jsc), open-circuit photovoltage (Voc), and fill factor (FF) of the C223 cell measured under an irradiation of 100 mW cm-2 AM1.5G sunlight are 10.86 mA cm-2, 709 mV, and 0.73, respectively, affording an overall power conversion efficiency (η) of 5.6%. Despite that the employment of C224 results in a slightly decreased FF of 0.70, it generates remarkably improved Jsc and Voc of 14.28 mA cm-2 and 793 mV, contributing to a significantly enhanced η of ∼8%. Obviously, the substituent alteration from fluorine to ethyl strongly affects the open-circuit photovoltage. For a typical DSC under illumination, the Voc is known as the energy offset between EF,n and EF,redox, the former of which is virtually a description of the excess carrier density (nc) in the titania film. Thereby, a change in Voc

which can be used to evaluate the relationship between photovoltage and kinetic competition between carrier generation and recombination. On the other side, through combining eqs 5-7 with the definition of the charge-transfer resistance, i.e., Rct = (∂Jr/ ∂V)-1,44 we can obtain   kB T eV Rct ¼ Kr-1 2 exp -γ ð8Þ e dAs γ kB T where Rct can be modeled from electrical impedance spectroscopies. Fitting the Rct data in Figure 9A can produce γ and Kr (Table 2). The C223 cell exhibits a Kr of 2.4  109 cm-3 s-1 in contrast to the 36-fold increased C224 counterpart, being 8.7  1010 cm-3 s-1. Moreover, with parameters listed in Table 2 and an assumed Nc of 7  1020 cm-3, we may also derive that the Ncγ exp[γ(EF,redox Ec)/kBT] term in eq 6 tends to enlarge the Kr of the C223 cell by over 1 order of magnitude relative to that of C224, being contrary to the observed actual Kr diminishment. Thereby, k0[I3-]R makes a predominant contribution to Kr, implying that there could be a noticeable molecule-related recombination reaction activation energy and/or the distribution profile and reaction order of triiodide, which should be specifically scrutinized in further studies. On the other side, in contrast to the high reaction order of 0.89 for the cell made from C223, the C224 cell presents a remarkably diminished γ of 0.68, leading to an evidently suppressed interfacial recombination (Figure 9B) at a given potential bias. Thereby, we can conclude from eq 7 that apart from the 3169

dx.doi.org/10.1021/jp110384n |J. Phys. Chem. C 2011, 115, 3163–3171

The Journal of Physical Chemistry C

ARTICLE

Figure 9. Plots of (A) interfacial charge-transfer resistance (Rct) and (B) simulated recombination current (Jr) as a function of potential bias (V). Solid lines in panel (A) are fittings according to eq 8.

Table 2. Parameters Obtained with Fittings on Electron Transport Resistance and Interfacial Charge-Transfer Resistance dye

Ec - EF,redox (eV)

γ

C223

0.97

0.89

C224

0.98

0.68

Kr (109 cm-3 s-1) 2.4 87

higher photocurrent, the notably smaller reaction order significantly contributes to the higher open-circuit photovoltage upon the employment of the ethyl-containing dye.

4. CONCLUSIONS To summarize, we have investigated the optoelectronic features of dye-sensitized solar cells based on two organic push-pull chromophores, featuring either the electron-deficient fluorine or electron-rich ethyl substituent on the cyclopentadithiophene conjugated linker. The substitution with the ethyl group negatively shifts the excitedstate redox potential of the dye molecule notably by 0.18 V relative to that of the fluorine counterpart. Since the substituent alteration has a relatively small impact on the titania conduction band edge, the ethyl-correlated thermodynamic tuning upon the dye molecule predominantly contributes to an evidently larger dye/ titania interfacial energy offset, soundly echoing the higher electron injection efficiency estimated from transient emission measurements and the elevated IPCE maximum. A remarkably decreased recombination reaction order is probed for the cell made from the ethyl-containing chromophore compared to the fluorine counterpart, which is found to be critically responsible for the enhanced open-circuit photovoltage, owing to a much slower charge recombination kinetics at the titania/electrolyte interface. Our work will shed light on future rational design of metal-free organic dyes for more efficient dye-sensitized solar cells. ’ ASSOCIATED CONTENT

bS

Supporting Information. Synthetic route, DFT calculation results, absorption/emission spectra, and voltammetric data. This information is available free of charge via the Internet at http:// pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT The National 973 Program (No. 2007CB936702 and No. 2011CBA00702), the National Science Foundation of China (No. 50973105 and No. 50773078), the Knowledge Innovation Program of CAS (No. KGCX2-YW-326), the Key Scientific and Technological Program of Jilin Province (No. 10ZDGG012), and the Hundred Talent Program of CAS are acknowledged for financial support. We are grateful to Dyesol for supplying the WER2-O scattering paste and to DuPont Packaging and Industrial Polymers for supplying the Bynel film. ’ REFERENCES (1) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15729. (2) Gr€atzel, M. Acc. Chem. Res. 2009, 42, 1788. (3) Chiba, Y.; Islam, A.; Watanabe, Y.; Komiya, R.; Koide, N.; Han, L. Jpn. J. Appl. Phys. 2006, 45, L638. (4) Hara, K.; Kurashige, M.; Dan-oh, Y.; Kasada, C.; Shinpo, A.; Suga, S.; Sayama, K.; Arakawa, H. New J. Chem. 2003, 27, 783. (5) Kitamura, T.; Ikeda, M.; Shigaki, K.; Inoue, T.; Anderson, N. A.; Ai, X.; Lian, T.; Yanagida, S. Chem. Mater. 2004, 16, 1806. (6) Horiuchi, T.; Miura, H.; Sumioka, K.; Uchida, S. J. Am. Chem. Soc. 2004, 126, 12218. (7) Thomas, K. R. J.; Lin, J. T.; Hsu, Y.-C.; Ho, K.-C. Chem. Commun. 2005, 4098. (8) Li, S.-L.; Jiang, K.-J.; Shao, K.-F.; Yang, L.-M. Chem. Commun. 2006, 2792. (9) Hagberg, D. P.; Edvinsson, T.; Marinado, T.; Boschloo, G.; Hagfeldt, A.; Sun, L. Chem. Commun. 2006, 2245. (10) Koumura, N.; Wang, Z.-S.; Mori, S.; Miyashita, M.; Suzuki, E.; Hara, K. J. Am. Chem. Soc. 2006, 128, 14256. (11) Kim, S.; Lee, J. K.; Kang, S. O.; Ko, J.; Yum, J.-H.; Fantacci, S.; De Angellis, F.; Di Censo, D.; Nazeeruddin, M. K.; Gr€atzel, M. J. Am. Chem. Soc. 2006, 128, 16701. (12) Edvinsson, T.; Li, C.; Pschirer, N.; Sch€oneboom, J.; Eickemeyer, F.; Sens, R.; Boschloo, G.; Herrmann, A.; M€ullen, K.; Hagfeldt, A. J. Phys. Chem. C 2007, 111, 15137. (13) Miyashita, M.; Sunahara, K.; Nishikawa, T.; Uemura, Y.; Koumura, N.; Hara, K.; Mori, A.; Abe, T.; Suzuki, E.; Mori, S. J. Am. Chem. Soc. 2008, 130, 17874. 3170

dx.doi.org/10.1021/jp110384n |J. Phys. Chem. C 2011, 115, 3163–3171

The Journal of Physical Chemistry C (14) Forneli, A.; Planells, M.; Sarmentero, M. A.; Martinez-Ferrero, E.; O'Regan, B. C.; Ballester, P.; Palomares, E. J. Mater. Chem. 2008, 18, 1652. (15) Wang, Z.-S.; Cui, Y.; Hara, K.; Dan-oh, Y.; Kasada, C.; Shinpo, A. Adv. Mater. 2007, 19, 1138. (16) Thomas, K. R. J.; Hsu, Y.-C.; Lin, J. T.; Lee, K.-M.; Ho, K.-C.; Lai, C.-H.; Cheng, Y.-M.; Chou, P.-T. Chem. Mater. 2008, 20, 1830. (17) Wang, Z.-S.; Koumura, N.; Cui, Y.; Takahashi, M.; Sekiguchi, H.; Mori, A.; Kubo, T.; Furube, A.; Hara, K. Chem. Mater. 2008, 20, 3993. (18) Zhou, G.; Pschirer, N.; Sch€oneboom, J. C.; Eickemeyer, F.; Baumgarten, M.; M€ullen, K. Chem. Mater. 2008, 20, 1808. (19) Kumaresan, D.; Thummel, R. P.; Bura, T.; Ulrich, G.; Ziessel, R. Chem.—Eur. J. 2009, 15, 6335. (20) Nishida, J.-i.; Masuko, T.; Cui, Y.; Hara, K.; Shibuya, H.; Ihara, M.; Hosoyama, T.; Goto, R.; Mori, S.; Yamashita, Y. J. Phys. Chem. C 2010, 114, 17920. (21) Perepichka, I. F.; Perepichka, D. F. Handbook of ThiopheneBased Materials: Applications in Organic Electronics and Photonics; Wiley: Weinheim, 2009. (22) Ie, Y.; Nitani, M.; Ishikawa, M.; Nakayama, K.-i.; Tada, H.; Kaneda, T.; Aso, Y. Org. Lett. 2007, 9, 2115. (23) Yan, P; Xie, A; Wei, M; Loew, L. M. J. Org. Chem. 2008, 73, 6587. (24) Li, R.; Liu, J.; Cai, N.; Zhang, M.; Wang, P. J. Phys. Chem. B 2010, 114, 4461. (25) Bonh^ote, P.; Dias, A.-P.; Papageorgiou, N.; Kalyanasumdaram, K.; Gr€atzel, M. Inorg. Chem. 1996, 35, 1168. (26) Wang, P.; Zakeeruddin, S. M.; Comte, P.; Charvet, R.; Humphry-Baker, R.; Gr€atzel, M. J. Phys. Chem. B 2003, 107, 14336. (27) Liu, J.; Li, R.; Si, X.; Zhou, D.; Shi, Y.; Wang, Y.; Jing, X.; Wang, P. Energy Environ. Sci. 2010, 3, 1924. (28) Runge, E.; Gross, E. K. U. Phys. Rev. Lett. 1984, 52, 997. (29) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (30) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785. (31) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009. (32) Hutchison, G. R.; Ratner, M. A.; Marks, T. J. J. Am. Chem. Soc. 2005, 127, 16866. (33) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; Wiley: Weinheim, 2001. (34) Hara, K.; Wang, Z.-S.; Sato, T.; Furube, A.; Katoh, R.; Sugihara, H.; Dan-oh, Y.; Kasada, C.; Shinpo, A.; Suga, S. J. Phys. Chem. B 2005, 109, 15476. (35) Tachibana, Y.; Moser, J. E.; Gr€atzel, M.; Klug, D. R.; Durrant, J. R. J. Phys. Chem. 1996, 100, 20056. (36) Koops, S. E.; O'Regan, B. C.; Barnes, P. R. F.; Durrant, J. R. J. Am. Chem. Soc. 2009, 131, 4808. (37) Fabregat-Santiago, F.; Bisquert, J.; Garcia-Belmonte, G.; Boschloo, G.; Hagfeldt, A. Sol. Energy Mater. Sol. Cells 2005, 87, 117. (38) Jennings, J. R.; Ghicov, A.; Peter, L. M.; Schmuki, P.; Walker, A. B. J. Am. Chem. Soc. 2008, 130, 13364. (39) Bisquert, J. J. Phys. Chem. B 2002, 106, 325. (40) Fabregat-Santiago, F.; Garcia-Belmonte, G.; Bisquert, J.; Zaban, A.; Salvador, P. J. Phys. Chem. B 2002, 106, 334.

ARTICLE

(41) Haque, S. A.; Tachibana, Y.; Willis, R. L.; Moser, J. E.; Gr€atzel, M.; Klug, D. R.; Durrant, J. R. J. Phys. Chem. B 2000, 104, 538. (42) Pelet, S.; Moser, J.-E.; Gr€atzel, M. J. Phys. Chem. B 2000, 104, 1791. (43) Ardo, S.; Meyer, G. J. Chem. Soc. Rev. 2009, 38, 115. (44) Bisquert, J.; Mora-Ser o, I. J. Phys. Chem. Lett. 2010, 1, 450.

3171

dx.doi.org/10.1021/jp110384n |J. Phys. Chem. C 2011, 115, 3163–3171