ARTICLE pubs.acs.org/Langmuir
Lithium-Modulated Conduction Band Edge Shifts and Charge-Transfer Dynamics in Dye-Sensitized Solar Cells Based on a Dicyanamide Ionic Liquid Yu Bai,†,‡ Jing Zhang,† Yinghui Wang,† Min Zhang,*,† 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 ABSTRACT: Lithium ions are known for their potent function in modulating the energy alignment at the oxide semiconductor/dye/electrolyte interface in dye-sensitized solar cells (DSCs), offering the opportunity to control the associated multichannel charge-transfer dynamics. Herein, by optimizing the lithium iodide content in 1-ethyl-3-methylimidazolium dicyanamide-based ionic liquid electrolytes, we present a solvent-free DSC displaying an impressive 8.4% efficiency at 100 mW cm2 AM1.5G conditions. We further scrutinize the origins of evident impacts of lithium ions upon current density voltage characteristics as well as photocurrent action spectra of DSCs based thereon. It is found that, along with a gradual increase of the lithium content in ionic liquid electrolytes, a consecutive diminishment of the open-circuit photovoltage arises, primarily owing to a noticeable downward movement of the titania conduction band edge. The conduction band edge displacement away from vacuum also assists the formation of a more favorable energy offset at the titania/dye interface, and thereby leads to a faster electron injection rate and a higher exciton dissociation yield as implied by transient emission measurements. We also notice that the adverse influence of the titania conduction band edge downward shift arising from lithium addition upon photovoltage is partly compensated by a concomitant suppression of the triiodide involving interfacial charge recombination.
1. INTRODUCTION Dye-sensitized solar cells (DSCs) are the theme of much research owing to their large potential to cut the cost of solar electricity. The elaborate combination of a mesoporous titania film, a panchromatic ruthenium sensitizer, and an acetonitrile electrolyte has furnished a small-area DSC demonstrating the highest certified solar-to-electricity efficiency of 11.2% so far.1 Nevertheless, the unfeasible encapsulation of highly volatile and toxic acetonitrile for long-term stable DSC modules under a thermal stress has aroused great enthusiasm for solid-state devices based on hole conductors2 and polymer electrolytes,3 fostering a notarized record efficiency of ∼5.0%.4 In past years, the incorporation of room temperature ionic liquids into DSCs has also been intensively explored primarily on account of their negligible vapor pressures5,6 to address the volatility issue encountered by the archetypal cell. Until now, the only ionic liquid that could be utilized in conjunction with some iodide melts to fabricate an over 8% efficiency DSC is 1-ethyl-3-methylimidazolium tetracyanoborate (EMITCB),7,8 characteristic of a pretty high dynamic fluidity. However, a large-scale commercial availability of the anion of EMITCB presently remains an unsolved issue. r 2011 American Chemical Society
Actually, the easily accessible 1-ethyl-3-methylimidazolium dicyanamide (EMIDCA) first reported by MacFarlane et al.9 also features a comparable viscosity (∼21 cP at room temperature) with respect to EMITCB. Earlier studies on the exploration of EMIDCA in DSC electrolytes by several groups have just afforded cells having moderate efficiencies,1017 the best of which being 6.6% at AM1.5G full sunlight conditions.10 Very recently, we revisited the EMIDCA ionic liquid in DSCs, on the basis of the progress on high absorption coefficient ruthenium photosensitizers.1822 Even with the high-performance model dye C106 (Figure 1), the cell efficiency could only be improved to 7.3%.23 Joint electrical impedance and photoluminescence measurements further disclosed a low exciton dissociation yield in the EMIDCA-containing cell, owing to the dicyanamide-induced uplifting of the conduction band edge of a nanocrystalline titania film. Keeping in mind the potent function of lithium ions in modulation of the energetics of metal oxide semiconductors in contact with molecular solvent-based electrolytes,2428 we herein blend lithium Received: January 13, 2011 Revised: February 25, 2011 Published: March 25, 2011 4749
dx.doi.org/10.1021/la200156m | Langmuir 2011, 27, 4749–4755
Langmuir
Figure 1. Molecular structure of C106.
iodide with a dicyanamide-containing ionic liquid electrolyte via systematically altering the ratio of lithium to imidazolium ions. This optimization allows us to obtain a solvent-free DSC exhibiting an impressive 8.4% efficiency measured under irradiation of AM1.5G full sunlight. We further correlate important photovoltaic features such as incident photon-to-collected electron conversion efficiencies (IPCEs) and photocurrent densityvoltage (JV) characteristics, with the stepwise tuning of the titania conduction band edge by lithium ions and associated chargetransfer dynamics at the titania/dye/electrolyte interface.
2. EXPERIMENTAL SECTION 2.1. Materials. Solvents such as acetonitrile, tert-butyl alcohol, and dimethyl sulfoxide were distilled before use. Chenodeoxycholic acid (cheno), 4-tert-butylpyridine (TBP), iodine (I2), and lithium iodide (LiI) were purchased from Fluka and used as received. 1,3-Dimethylimidazolium iodide (DMII), 1-ethyl-3-methylimidazolium iodide (EMII),29 and EMIDCA9 were prepared according to literature methods. The WER4-O scattering paste was received as a gift from Dyesol. Four electrolytes were applied for the cell fabrication: E1, DMII/EMII/EMIDCA/I2/ TBP (12/12/16/1.67/3.33, molar ratio); E2, DMII/EMII/EMIDCA/ I2/TBP/LiI (12/11.5/16/1.67/3.33/0.5); E3, DMII/EMII/EMIDCA/ I2/TBP/LiI (12/9.5/16/1.67/3.33/2.5); E4, DMII/EMII/EMIDCA/ I2/TBP/LiI (12/7/16/1.67/3.33/5). Note that, in this series of electrolytes, an invariable molar ratio of iodide to triiodide was kept to not alter the electrolyte equilibrium potential. The detailed synthesis of the C106 dye has been reported in our previous paper.21 2.2. Cell Fabrication. For the cells presented here, a bilayer titania film on the fluorine-doped tin oxide (FTO) conducting glass (Nippon sheet glass, solar, 4 mm thick) was used as the negative electrode, which constitutes a 10 μm thick transparent layer of 23 nm sized particles and an atop-deposited 5 μm thick second layer of scattering particles. A circular TiO2 electrode (∼0.283 cm2) was stained by incubating it for 16 h in a coadsorption solution of C106 (150 μM) and cheno (2 mM) in a mixed solvent of acetonitrile, tert-butyl alcohol, and dimethyl sulfoxide at a volume ratio of 9/9/2. After being rinsed with acetonitrile and dried by air flow, the dye-coated 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 by heating. The internal space was perfused with an ionic liquid electrolyte with the aid of a vacuum backfilling system. Ultimately, the hole in the positive electrode was closed hermetically with a Bynel sheet and a thin glass cover by heating. 2.3. Photovoltaic Characterization. A Keithley 2400 source meter and a Zolix Omni-λ300 monochromator equipped with a 500 W xenon lamp were used for photocurrent action spectrum measurements, with a wavelength sampling interval of 10 nm and a current sampling time of 2 s under full computer control. A Hamamatsu S1337-1010BQ silicon diode used for IPCE measurements was calibrated at the National
ARTICLE
Figure 2. JV characteristics measured under irradiation of 100 mW cm2 AM1.5G sunlight. The open symbols are experimental data, and the solid lines are fittings to the Shockley equivalent circuit. Institute of Metrology, China. A model LS1000-4S-AM1.5G-1000W solar simulator (Solar Light Co., Glenside, PA) in combination with a metal mesh was employed to give an irradiance of 100 mW cm2. The light intensity was tested with a PMA2144 pyranometer and a calibrated PMA 2100 dose control system. JV characteristics were obtained by applying a bias potential to a testing cell and measuring the photocurrent with a Keithley 2602 source meter under full computer control. The measurements were fully automated using Labview 8.0. A metal mask with an aperture area of 0.158 cm2 was covered on a testing cell during all measurements. An antireflection film (λ < 380 nm, ARKTOP, ASAHI Glass) was adhered to the DSC photoanode during IPCE and JV measurements. 2.4. Electrical Impedance Measurements. Electrical impedance experiments were carried out under illumination of a white LED with an IM6ex electrochemical workstation, with a frequency range from 50 mHz to 100 kHz and a potential modulation of 20 mV. A bias potential was applied to equal the open-circuit voltage at each irradiation intensity. The obtained impedance spectra were fitted with the Z-view software (v2.80, Scribner Associates Inc., Southern Pines, NC).
2.5. Absorption and Photoluminescence Measurements. The steady-state electronic absorption spectrum was recorded on a PerkinElmer Lambda 900 spectrometer. Transient absorption measurements were performed with an LP920 laser flash spectrometer pumped with a nanosecond tunable OPOLett-355II laser. The sample was kept at a 45° angle with respect to the excitation beam. The probe light passed through a bandpass filter (center wavelength 782 nm) was dispersed by a monochromator before being detected by a fast photomultiplier tube and recorded with a TDS 3012C digital signal oscilloscope. Photoluminescence decay traces were measured with a LifeSpec-II spectrometer, employing an EPL635 pulsed laser diode and a Hamamatsu H5773-04 photomultiplier. 2.6. Voltammetric Measurements. A CHI660C electrochemical workstation was used for square-wave voltammetry measurements. The measurement of the ground-state redox potential of C106 was accomplished by constructing a three-electrode electrochemical cell equipped with a platinum gauze as the counter electrode, a Ag/AgCl (saturated KCl) reference electrode, and a working electrode of a dyecoated titania film on FTO. In measuring the square-wave voltammogram of the ferrocene reference, we employed a working electrode of a 5 μm radius platinum ultramicroelectrode, a platinum foil auxiliary electrode, and a homemade reference electrode composed of a platinum wire dipped in an electrolyte-filling glass tube, the lower end of which is sealed with a porous ceramic frit.
3. RESULTS AND DISCUSSION We first measured the JV characteristics (Figure 2) of cells based on electrolytes E1E4 in conjunction with the C106 dye at an irradiation of 100 mW cm2 AM1.5G sunlight, and the 4750
dx.doi.org/10.1021/la200156m |Langmuir 2011, 27, 4749–4755
Langmuir
ARTICLE
Table 1. Detailed Photovoltaic Parameters Measured under an Irradiation of 100 mW cm2 AM1.5G Sunlighta electrolyte
Jsc (mA cm2)
Voc (mV)
FF
η (%)
E1
11.46
803
0.80
7.4
E2 E3
13.28 14.52
767 747
0.79 0.77
8.0 8.4
E4
15.04
715
0.75
8.1
a
The spectral distribution of our light resource simulates AM 1.5G solar emission with a mismatch factor of less than 5%. Jsc = short-circuit photocurrent density, Voc = open-circuit photovoltage, FF = fill factor, and η = total power conversion efficiency. The aperture area of a metal mask is 0.158 cm2. An antireflection film was adhered to the testing cell during measurement.
detailed photovoltaic parameters are collected in Table 1. A stepwise augmentation of the ratio of lithium to imidazolium ions in ionic liquid electrolytes E1E4 endows an increase of shortcircuit photocurrent density (Jsc) from 11.46 to 15.04 mA cm2. On the other hand, a remarkable 88 mV attenuation of opencircuit photovoltage (Voc) can be noted by changing the electrolyte from E1 to E4, along with a decreased fill factor (FF). Herein, owing to an electrolyte-dependent trade-off between photocurrent and photovoltage, the E3 cell affords an appreciable overall power conversion efficiency (η) of 8.4%, with Jsc, Voc, and FF being 14.52 mA cm2, 747 mV, and 0.77, respectively. Apparently, these preliminary photovoltaic evaluations show the adverse impact of lithium cations upon open-circuit photovoltage, which is consistent with previous studies.10,2628,30,31 On the basis of the recently proposed nonlinear recombination model in a DSC,33 the interfacial charge recombination rate (Un) can be expressed as Un ¼ k 0 n β
ð1Þ
where k0 is the apparent reaction rate constant and β is the reaction order of titania electrons. n denotes the free electron density and can be well described by the Boltzmann statistics assuming that Ec EF,n) . kBT as EF, n Ec n ¼ Nc exp ð2Þ kB T where Nc is the accessible density of states in the conduction band, kB the Boltzmann constant, and T the absolute temperature. EF,n and Ec represent the quasi Fermi level and conduction band edge of titania, respectively. Combining eqs 1 and 2 gives EF, n Ec β Un ¼ k0 Nc exp β ð3Þ kB T For a DSC at the open-circuit condition, the photocurrent generation rate (Φ) is fully compensated by the rate of interfacial charge recombination: Un ¼ Φ ¼
Jsc ed
ð4Þ
where d is the film thickness and e the elementary charge. Combining eqs 3 and 4 gives kB T Ec EF, redox 1 Jsc þ ln Voc ¼ ð5Þ e β dek0 Nc β kB T
Figure 3. (A) Electron transport resistance (Rt) and (B) interfacial charge recombination resistance (Rct) plotted against the potential bias (V). Calculation through R0 = d/[A(1 P)eμ0Nc] with μn = 1.0 cm2 V1 s1, Nc = 7.0 1020 cm3, a film thickness (d) of 10 μm, a cell area (A) of 0.283 cm2, and a porosity (P) of 0.64 affords an R0 of 8.77 105 Ω for the titania film utilized. (C) Jsc versus Voc obtained at different illuminations. The data points are experimentally obtained, and the solid lines are fittings in terms of proper functions. The blue, pink, green, and red dashed lines in panel C are drawn according to the Jsc values of E1E4 cells at the 100 mW cm2 AM1.5G conditions, respectively.
where EF,redox refers to the electrolyte Fermi level. In this expression, the Voc of a DSC displays an evident dependence on the energy alignment at the titania/electrolyte interface, the photocurrent generation rate, and the interfacial charge recombination kinetics. Thereby, we measured the impedance spectroscopies of our cells, from which the electron transport resistance Rt can be accurately modeled using a proper equivalent circuit based on the transmission line model.3437 Furthermore, the energy difference Ec EF,redox can be fitted through38 Ec EF, redox eV Rt ¼ R0 exp γ exp γ ð6Þ kB T kB T where R0 represents the electron transport resistance on the transport level and is considered invariable for a given film geometry and γ is an empirically introduced parameter correcting the discrepancy between the theoretical exponential factor and the practical one. As shown in Figure 3A, the Rt data as a function of the applied potential bias V could be well fitted to eq 6, yielding the Ec levels relative to EF,redox (Table 2). Note here that since the molar ratios between iodide and triiodide have been maintained constant in all electrolytes, the electrolyte Fermi level was taken as an invariant for our cells made with E1E4, according to the Nernst equation.39 Apparently, the enlargement of the ratio of lithium to imidazolium cations in electrolytes resulted in a consecutive downward displacement of the conduction band edge, adversely impacting the open-circuit photovoltage (eq 5). 4751
dx.doi.org/10.1021/la200156m |Langmuir 2011, 27, 4749–4755
Langmuir
ARTICLE
Table 2. Parameters Derived from Fittings on Rt and Rct Ec EF,redox
k0
U0k
electrolyte
(eV)
γ
(106 cm3(1β) s1)
β
(1022 cm3 s1)
E1
1.100
0.89
3.0
0.84
97
E2
1.056
0.92
5.0
0.80
24
E3
1.025
0.92
13.8
0.76
10
E4
0.983
0.95
22.0
0.73
4
To comprehend the charge recombination dynamics at the titania/electrolyte interface and assess its influence upon the open-circuit photovoltage, we further analyzed the charge recombination resistance Rct, which serves as an integrative representation of the interfacial charge-transfer characteristics. The apparent reaction rate constant k0 and reaction order β for titania electrons can be resolved with38 Ec EF, redox kB T βeV exp β Rct ¼ exp ð7Þ kB T kB T Ade2 k0 Nc β β where A is the active cell area. By deriving Rct through modeling the measured impedance spectroscopies, the values of k0 and β can be readily obtained through fitting the Rct data (Figure 3B) in terms of eq 7, as listed in Table 2. We note that through taking the reciprocal of the reaction order β as the ideality factor,33 the preceding JV curves can be excellently fitted according to a function illustrating the Shockley equivalent circuit,40 thus verifying the validity of the as-presented reaction orders. Herein, to evaluate the lithium impact on the interfacial charge recombination kinetics, we refer to an effective recombination rate constant U0k41 defined with U0k = k0Ncβ, which reconstructs eq 3 to be EF, redox Ec EF, n EF, redox Un ¼ U0k exp β exp β ð8Þ kB T kB T As presented in Table 2, U0k was noticeably diminished from 97 1022 to 4 1022 cm3 s1 by changing the electrolyte from E1 to E4, suggesting that the adsorption of lithium cations on titania tends to suppress the interception of photoinjected electrons by triiodide, being consistent with a previous observation.42 This could be related to the diminished driving force owing to a positive shift of the conduction band edge, and the detailed atomic picture needs to be addressed in the future. With the aid of eq 5, we remark that the gradually slowing charge recombination kinetics is inclined to pose a positive contribution to the opencircuit photovoltage. To figure out the influence of U0k upon Voc quantitatively, we further measured the JV characteristics under different illuminations and derived the JscVoc relationship43 as plotted in Figure 3C. The higher Jsc values of E2, E3, and E4 with respect to E1 tend to enhance Voc by 4, 6, and 7 mV, respectively. However, the Ec values of E2, E3, and E4 relative to E1 move downward by about 44, 75, and 117 mV (Table 2), introducing an adverse impact on Voc. Taking into account the Voc of the E2, E3, and E4 cells being 36, 56, and 88 mV lower than that of the E1 cell, we may derive that slower interfacial charge recombination rates of the E2, E3, and E4 cells are inclined to pose 4, 13, and 22 mV positive contributions to the open-circuit photovoltage. This result is in accordance with a gradually diminished U0k along with a stepwise augmentation of lithium ion content in our ionic liquid electrolytes. Thereby we conclude that the lithium ion induced
Figure 4. (A) Photocurrent action spectra of cells with electrolytes E1E4. (B) Photoluminescence traces of a C106-coated alumina film soaked in electrolyte E1 (trace a) and a C106-coated titania film immersed in the respective E1E4 (traces be). The instrument response function (IRF) has also been included. Smooth black lines are stretched exponential fittings to the raw data. The excitation wavelength is 639 nm, and the probe wavelength is 745 nm. The photoluminescence intensities have been corrected in terms of the absorbance at 639 nm. The arrow direction indicates the increase of the lithium cation concentration.
conduction band edge displacements have dominated the experimental Voc variations. The photocurrent action spectra of cells with E1E4 are presented in Figure 4A, where the IPCE is plotted as a function of the wavelength. Significantly enhanced photocurrent in the visible and infrared region can be noted along with the increment of the ratio of lithium to imidazolium, with the IPCE maxima consecutively increasing from 62% to 81%. To disclose the intrinsic origins of the lithium content correlated IPCE summits, we adopted the time-correlated single photon counting technique to estimate the yield of electron injection (ηinj) at the titania/ dye/electrolyte interface, which makes a critical determinant for the IPCE maximum.30,44,45 A control cell was first fabricated on the basis of a C106-coated alumina film soaked in the practical electrolyte E1, which presents a strong photoluminescence (trace a in Figure 4B) upon laser excitation at 639 nm. In virtue of the unfavorable energy alignment for exciton dissociation at the alumina/dye interface, we ascribe this photoluminescence decay to the irradiative and nonirradiative deactivation of excited-state C106 dye molecules. On the other hand, remarkably quenched photoluminescence (traces be in Figure 4B) is observed upon substituting alumina with titania, arising from expeditious electron injection at the titania/dye interface. Moreover, through applying a stretched exponential function I ¼ I0 exp½ðt=τÞR
ð9Þ
where I0 is the initial photoluminescence amplitude on alumina, R is the stretch parameter, and τ is the lifetime, the 4752
dx.doi.org/10.1021/la200156m |Langmuir 2011, 27, 4749–4755
Langmuir
ARTICLE
Table 3. Photoluminescence Average Lifetime (τh), Injection Rate Constant (kinj), Injection Efficiency (ηinj), and Reaction Free Energy (ΔG0) of Systems Based on Electrolytes E1E4 system
hτ (ns)
alumina/E1
29.3
kinj (108 s1)
ηinj (%)
ΔG0 (meV)
titania/E1
6.5
1.2
78
20
titania/E2
4.6
1.8
84
64
titania/E3
3.2
2.7
89
95
titania/E4
2.5
3.6
91
137
Figure 6. Temporal absorption profiles recorded at 782 nm for 13 μm thick, C106-coated titania films immersed in inert EMIDCA/TBP/ LiTFSI electrolytes (a, 40/3.33/0; b, 40/3.33/0.5; c, 40/3.33/2.5; d, 40/ 3.33/5, molar ratio) or electroactive electrolytes (e, E1; f, E2; g, E3; h, E4). The excitation wavelength is 688 nm, and the pulse fluence is 116 μJ cm2. Smooth lines are stretched exponential fittings to raw data obtained by averaging 800 laser shots.
Figure 5. (A) Square-wave voltammograms of a C106-coated titania film immersed in EMIDCA. The scan rate is 50 mV s1. (B) Normalized steady-state absorption spectrum of a C106-coated TiO2 film in contact with electrolyte E1. (C) Square-wave voltammogram of ferrocene in 1-ethyl-3-methylimidazolium dicyanamide with a platinum/E1 reference electrolyte.
photoluminescence average lifetime of the C106 dye anchored on a titania (τhtitania) or alumina (τhaluminia) film can be calculated with hτ = (τ/R)Γ(1/R), and Γ(x) is a Γ function. As shown in Table 3, increasing the lithium ion content in electrolyte reduces hτtitania from 6.5 to 2.5 ns. Through a combination of hτtitania and hτalumina (29.3 ns), the interfacial electron injection rate constant (kinj) can be derived by kinj = 1/τhtitania 1/τhalumina, displaying an evident acceleration feature (Table 3) with an increase of the lithium ion content. Furthermore, in consideration of the noninjection deactivation rate constant kd = 1/τhalumina, the yield of electron injection (ηinj) was estimated through ηinj = kinj/(kinj þ kd) to be 78%, 84%, 89%, and 91% for cells with the respective electrolytes E1E4, soundly explaining the aforementioned IPCE summits. We suspect that these electrolyte-correlated electron injection kinetics likely stem from the dissimilar energy offsets at the
titania/dye interface30 on account of the preceding discussion on the titania conduction band edge movements. As shown in Figure 5A, the ground-state redox potential φ0Sþ/S of the C106 dye was measured to be 5.44 V versus a vacuum. Note here that the φ0Sþ/S of C106 anchored on titania does not vary with the introduction of lithium ions into the EMIDCA ionic liquid. The excitation transition energy (E00) was estimated to be 1.67 eV from the onset of the electronic absorption spectrum (Figure 5B) of a C106-coated TiO2 film. We have also noticed that dissimilar to pushpull organic dyes,31 the C106 dye does not exhibit a significant E00 dependence on the lithium content in EMIDCAbased melts. Thereby, the excited-state potential φ0Sþ/S* of C106 in cells with electrolytes E1E4 can be derived to be 3.77 V versus a vacuum. Furthermore, by using a homemade threeelectrode electrochemical cell as described in the Experimental Section, we measured the square-wave voltammogram (Figure 5C) of ferrocene dissolved in EMIDCA, from which the EF,redox was deduced to be 4.89 eV versus a vacuum. Through combining the aforementioned Ec EF,redox, the reaction free energy ΔG0 for electron injection can be readily calculated with46 ΔG0 ¼ Ec eφ0Sþ =S
ð10Þ
and the details are collected in Table 3. Evidently, a higher lithium content is intimately associated with a larger reaction free energy as a consequence of a lithium-induced displacement of the titania conduction band edge, soundly explaining the acceleration of the interfacial injection rate. The transient absorption measurements were further carried out to examine the electrolyte-correlated interfacial kinetic competition between oxidized dye interception by the titania electron 4753
dx.doi.org/10.1021/la200156m |Langmuir 2011, 27, 4749–4755
Langmuir and that by the electron-donating iodide (dye regeneration) in electrolytes. On the basis of our previous measurement on the transient absorption spectrum of the C106-coated titania film,32 a monochromatic light at 782 nm was selected to probe the dualchannel charge-transfer kinetics of oxidized dye molecules.4750 Excitation fluences were carefully controlled in our measurements to guarantee ∼2.4 1014 photons cm2 to be absorbed by the dye-coated titania film during every laser pulse. We also selected the pump wavelength according to a 0.2 optical density of the dye-coated titania film, so as to achieve a similar carrier distribution in the titania film. We first fit the raw data with a stretched exponential function (mΔOD µ exp[(t/τ)R]) and determine t1/2 in terms of the fitted curves. In the absence of electroactive iodide, the absorption decays (traces ad in Figure 6) produced t1/2 values in the range of 5.88.2 ms, indicating sluggish charge recombination between oxidized dye molecules and photoinjected electrons. Upon the employment of practical electrolytes E1E4, remarkably accelerated absorption decays (traces eh in Figure 6) were observed owing to the expeditious electron donation from iodide to oxidized dyes, affording comparable half-reaction times ranging from 6.3 to 7.7 μs. Albeit there is a slight dissimilarity in the respective charge recombination and dye regeneration kinetics, the kinetic branch ratios are over 750 for all electrolyte systems, apparently suggesting competition efficiencies close to unity. Hence, we conclude that the electrolyte-dependent dual-channel charge-transfer kinetics of oxidized sensitizers presented a negligible contribution to the IPCE maximum differences.
4. CONCLUSIONS In summary, through a systematic optimization of the ratio of lithium to imidazolium cations in a dicyanamide ionic liquid, we have fabricated a solvent-free dye-sensitized solar cell based on a highabsorptivity ruthenium dye C106, showing a solar-to-electricity conversion efficiency of 8.4% under AM1.5G full sunlight illumination. Furthermore, the roles of lithium cations in these electrolytes are investigated in detail with the aid of electrical impedance and transient emission spectroscopies. Numerical analysis of the charge transport resistance of the mesoporous titania film discloses that an increase of the lithium content in ionic liquid electrolytes confers a downward displacement of the conduction band edge upon the dye-coated titania film, dominating the observed open-circuit photovoltage diminishment, although the suppressed interception of photoinjected electrons by triiodide poses a positive contribution to Voc. Moreover, along with an increase of the lithium content in the electrolyte, a consecutive acceleration of the interfacial electron injection is probed via measuring dynamic emission traces. This is further rationalized with a positive dependence of the electron injection rate constant on the reaction free energy, which is closely related to the conduction band edge shifts. Our work will shed light on the future design of ionic liquid electrolytes for more efficient dye-sensitized solar cells. ’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected] (M.Z.);
[email protected] (P.W.).
’ ACKNOWLEDGMENT The National 973 Program (Grants 2007CB936702 and 2011CBA00702), the National Science Foundation of China
ARTICLE
(Grants 50973105 and 50773078), the Chinese Academy of Sciences (CAS) Knowledge Innovation Program (Grant KGCX2-YW-326), the Key Scientific and Technological Program of Jilin Province (Grant 10ZDGG012), and the CAS Hundred Talent Program are acknowledged for financial support. We are grateful to Dyesol for supplying the WER4-O scattering paste and to DuPont Packaging and Industrial Polymers for supplying the Bynel film.
’ REFERENCES (1) Chiba, Y.; Islam, A.; Watanabe, Y.; Komiya, R.; Koide, N.; Han, L. Jpn. J. Appl. Phys., Part 2 2006, 45, L638. (2) Snaith, H. J.; Schmidt-Mende, L. Adv. Mater. 2007, 19, 3187. (3) Stergiopoulos, T.; Arabatzis, I. M.; Katsaros, G.; Falaras, P. Nano Lett. 2002, 2, 1259. (4) Wang, M.; Liu, J.; Cevey-Ha, N.-L.; Moon, S.-J.; Liska, P.; Humphry-Baker, R.; Moser, J.-E.; Gr€atzel, C.; Wang, P.; Zakeeruddin, S. M.; Gr€atzel, M. Nano Today 2010, 5, 169. (5) Gorlov, M.; Kloo, L. Dalton Trans. 2008, 2655. (6) Zakeeruddin, S. M.; Gr€atzel, M. Adv. Funct. Mater. 2009, 19, 2187. (7) Bai, Y.; Cao, Y.; Zhang, J.; Wang, M.; Li, R.; Wang, P.; Zakeeruddin, S. M.; Gr€atzel, M. Nat. Mater. 2008, 7, 626. (8) Shi, D.; Pootrakulchote, N.; Li, R.; Guo, J.; Wang, Y.; Zakeeruddin, S. M.; Gr€atzel, M.; Wang, P. J. Phys. Chem. C 2008, 112, 17046. (9) MacFarlane, D. R.; Golding, J.; Forsyth, S.; Forsyth, M.; Deacon, G. B. Chem. Commun. 2001, 1430. (10) Wang, P.; Zakeeruddin, S. M.; Moser, J.-E.; Gr€atzel, M. J. Phys. Chem. B 2003, 107, 13280. (11) Kawano, R.; Matsui, H.; Matsuyama, C.; Sato, A.; Susan, M. A. B. H.; Tanabe, N.; Watanabe, M. J. Photochem. Photobiol., A 2004, 164, 87. (12) Dai, Q.; Menzies, D. B.; MacFarlane, D. R.; Batten, S. R.; Forsyth, S.; Spiccia, L.; Cheng, Y.-B.; Forsyth, M. C. R. Chim. 2006, 9, 617. (13) Chen, J.; Officer, D. L.; Pringle, J. M.; MacFarlane, D. R.; Too, C. O.; Wallace, G. G. Electrochem. Solid State Lett. 2005, 8, A528. (14) Wachter, P.; Zistler, M.; Schreiner, C.; Berginc, M.; Krasovec, U. O.; Gerhard, D.; Wasserscheid, P.; Hinsch, A.; Gores, H. J. J. Photochem. Photobiol., A 2008, 197, 25. (15) Singh, P. K.; Bhattacharya, B.; Nagarale, R. K.; Pandey, S. P.; Kim, K.-W.; Rhee, H.-W. Synth. Met. 2010, 160, 950. (16) Singh, P. K.; Jadhav, N. A.; Mishra, S. K.; Singh, U. P.; Bhattacharya, B. Ionics 2010, 16, 645. (17) G€oebel, R.; Hesemann, P.; Weber, J.; M€oller, E.; Friedrich, A.; Beuermann, S.; Taubert, A. Phys. Chem. Chem. Phys. 2009, 11, 3653. (18) Wang, P.; Zakeeruddin, S. M.; Humphry-Baker, R.; Moser, J. E.; Gr€atzel, M. Adv. Mater. 2003, 15, 2101. (19) Jiang, K.; Masaki, N.; Xia, J.; Noda, S.; Yanagida, S. Chem. Commun. 2006, 2460. (20) Gao, F.; Wang, Y.; Shi, D.; Zhang, J.; Wang, M.; Jing, X.; Humphry-Baker, R.; Wang, P.; Zakeeruddin, S. M.; Gr€atzel, M. J. Am. Chem. Soc. 2008, 130, 10720. (21) Cao, Y.; Bai, Y.; Yu, Q.; Cheng, Y.; Liu, S.; Shi, D.; Gao, F.; Wang, P. J. Phys. Chem. C 2009, 113, 6290. (22) Chen, C.-Y.; Wang, M.; Li, J.-Y.; Pootrakulchote, N.; Alibabaei, L.; Ngoc-le, C.-H.; Decoppet, J.-D.; Tsai, J.-H.; Gr€atzel, C.; Wu, C.-G.; Zakeeruddin, S. M.; Gr€atzel, M. ACS Nano 2009, 3, 3103. (23) Zhang, M.; Zhang, J.; Bai, Y.; Wang, Y.; Su, M.; Wang, P. Phys. Chem. Chem. Phys. 2011, 13, 3788. (24) Yates, D. E.; Healy, T. W. J. Chem. Soc., Faraday Trans. 1 1980, 76, 9. (25) Redmond, G.; Fitzmaurice, D. J. Phys. Chem. 1993, 97, 1426. (26) Kelly, C. A.; Farzad, F.; Thompson, D. W.; Stipkala, J. M.; Meyer, G. J. Langmuir 1999, 15, 7047. 4754
dx.doi.org/10.1021/la200156m |Langmuir 2011, 27, 4749–4755
Langmuir
ARTICLE
(27) Haque, S. A.; Tachibana, Y.; Willis, R. L.; Moser, J. E.; Gr€atzel, M.; Klug, R. D.; Durrant, J. R. J. Phys. Chem. B 2000, 104, 538. (28) Nakade, S.; Kanzaki, T.; Kubo, W.; Kitamura, T.; Wada, Y.; Yanagida, S. J. Phys. Chem. B 2005, 109, 3480. (29) Bonh^ote, P.; Dial, A. P.; Papageorgiou, N.; Kalyanasumdaram, K.; Gr€atzel, M. Inorg. Chem. 1996, 35, 1168. (30) Koops, S. E.; O’Regan, B. C.; Barnes, P. R. F.; Durrant, J. R. J. Am. Chem. Soc. 2009, 131, 4808. (31) Li, R.; Liu, D.; Zhou, D.; Shi, Y.; Wang, Y.; Wang, P. Energy Environ. Sci. 2010, 3, 1765. (32) Yu, Q.; Wang, Y.; Yi, Z.; Zu, N.; Zhang, J.; Zhang, M.; Wang, P. ACS Nano 2010, 4, 6032. (33) Bisquert, J.; Mora-Sero, I. J. Phys. Chem. Lett. 2010, 1, 450. (34) Bisquert, J. J. Phys. Chem. B 2002, 106, 325. (35) Fabregat-Santiago, F.; Garcia-Belmonte, G.; Bisquert, J.; Zaban, A.; Salvador, P. J. Phys. Chem. B 2002, 106, 334. (36) Bisquert, J.; Gr€atzel, M.; Wang, Q.; Fabregat-Santiago, F. J. Phys. Chem. B 2006, 110, 11284. (37) Fabregat-Santiago, F.; Bisquert, J.; Garcia-Belmonte, G.; Boschloo, G.; Hagfeldt, A. Sol. Energy Mater. Sol. Cells 2005, 87, 117. (38) Liu, J.; Li, R.; Si, X.; Zhou, D.; Shi, Y.; Wang, Y.; Jing, X.; Wang, P. Energy Environ. Sci. 2010, 3, 1924. (39) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; Wiley: Weinheim, Germany, 2001. (40) Potscavage, W. J., Jr.; Sharma, A.; Kippelen, B. Acc. Chem. Res. 2009, 42, 1758. (41) Barea, E. M.; Zafer, C.; Gultekin, B.; Aydin, B.; Koyuncu, S.; Icli, S.; Fabregat-Santiago, F.; Bisquert, J. J. Phys. Chem. C 2010, 114, 19840. (42) Kopidakis, N.; Benkstein., K. D.; van de Lagemaat, J.; Frank, A. J. J. Phys. Chem. B 2003, 107, 11307. (43) Kopidakis, N.; Neale, N. R.; Frank, A. J. J. Phys. Chem. B 2006, 110, 12485. (44) Koops, S. E.; Durrant, J. R. Inorg. Chim. Acta 2008, 361, 663. (45) Morandeira, A.; Lopez-Duarte, I.; Martínez-Díaz, M. V.; O’Regan, B. C.; Shuttle, C.; Haji-Zainulabidin, N. A.; Torres, T.; Palomares, E.; Durrant, J. R. J. Am. Chem. Soc. 2007, 129, 9250. (46) She, C.; Anderson, N. A.; Guo, J.; Liu, F.; Goh, W.-H.; Chen, D.-T.; Mohler, D. L.; Tian, Z.-Q.; Hupp, J. T.; Lian, T. Q. J. Phys. Chem. B 2005, 109, 19345. (47) Katoh, R.; Furube, A.; Barzykin, A. V.; Arakawa, H.; Tachiya, M. Coord. Chem. Rev. 2004, 248, 1195. (48) Nelson, J.; Haque, S. A.; Klug, D. R.; Durrant, J. R. Phys. Rev. B 2001, 63, 205321. (49) Clifford, J. N.; Palomares, E.; Nazeeruddin, M. K.; Gr€atzel, M.; Nelson, J.; Li, X.; Long, N. J.; Durrant, J. R. J. Am. Chem. Soc. 2004, 126, 5225. (50) Koops, S. E.; Barnes, P. R. F.; O’Regan, B. C.; Durrant, J. R. J. Phys. Chem. C 2010, 114, 8054.
4755
dx.doi.org/10.1021/la200156m |Langmuir 2011, 27, 4749–4755
