Aqueous Electrolytes Based Dye-sensitized Solar Cells using I^-/I₃^- Redox Couple to Achieve ≥ 4% Power Conversion Efficiency

Abstract

We report on the influence of water as an electrolyte on the photovoltaic performances. The photovoltaic performance was shown to be quite sensitive to the substituent on the donor group. An optimized efficiency of 4.41% in the presence of 100% water content using $I^-/I{_3}^-$ redox couple was obtained using the D21L6 organic dye.

Introduction

The current global warming and gradual depletion of fossil fuels urge the development of new renewable energy sources. Among the available technologies, dye-sensitized solar cells (DSSCs) are one of the most promising candi-dates due to their low cost and high conversion efficiency.1 Some polypyridyl ruthenium complexes,2 organic sensi-tizers,3 and porphyrins4 have achieved high power conver-sion efficiencies of 9-12％. These cells have been fabricated with I−/I3− or CoII/CoIII redox couple5 electrolytes using organic solvents.

However, organic based electrolytes degrade the durability and safety of the cell due to their volatility and flammability. Water might be a good candidate as the solvent medium because water-based DSSCs are less expensive and environ-mentally benign. The presence of water in the organic elec-trolyte is believed to be bad for efficient solar conversion efficiency due to formation of iodates,6 dye detachment,7 and band edge movement.8 Therefore, the prevention of water permeation on organic based DSSCs represents some challenges towards the encapsulation or barrier layers.9 Contrary to generally accepted view for water based elec-trolyte, a recent study published by O′ Regan and co-workers10 has shown that water content in the electrolyte is not linked to poor efficiency and/or durability in DSSCs. After this work, many studies have described an aqueous dye-sensitized solar cell based on the Fe(CN)64−/3−,11 I−/I3−,12 thiolate/disulfide,13 and Co2+/3+ redox couples14 using an organic sensitizer, showing high conversion efficiencies of 2.1-4.2％. These works encouraged us to revisit the aqueous based DSSCs in order to increase the conversion efficiency. In this study, we report aqueous based DSSCs using three organic sensitizers (D5L6,15  D21L6,16 and JK-310), in which the latter two dyes contain (hexyloxy)phenylamine and 4-(2-(2-methoxyethoxy)-ethoxy)phenylamine as the electron donor (Figure 1). We introduced a water soluble unit on an electron donor group to see the effect of wettability behavior and redox couple binding in water.

Figure 1.Structure of three sensitizers.

 

Experimental

General Methods. All reactions were carried out under argon atmosphere. Solvents were distilled from appropriate reagents. All reagents were purchased from Sigma-Aldrich and TCI. 1-Bromo-4-(2-(2-methoxyethoxy)ethoxy)benzene 117 was synthesized using a modified procedure of previous references. 1H and 13C NMR spectra were recorded with a Varian Mercury 300 spectrometer. MALDI-TOF Mass spectro-metry was performed using VoyagerTM DE-STR, 4700 Pro-teomics. Optimized structures were calculated by TD-DFT using the B3LYP functional and the 6-21G* basis set. The highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energies were deter-mined using minimized singlet geometries to approximate the ground state. UV-vis data was measured with a Perkin-Elmer Lambda 2S UV-visible spectrometer. Photolumine-scence spectra were recorded on a Perkin LS fluorescence Spectrometer. Cyclic voltammetry (CV) was performed using VersaSTAT 3 (Princeton Applied Research). Anhydrous acetonitrile was used as the solvent and 0.1 M of tetrabutyl ammonium hexafluorophosphate was used as the supporting electrolyte. TiO2 films stained with the sensitizers were used as working elec-trodes, a platinum wire as the counter electrode, and a Ag/Ag+ electrode as the reference electrode. The measurements were done using a scan rate of 50 mV/s and ferrocenium/ferrocene redox couple was used as an internal reference.

Electron Transport Measurements. The electron diffu-sion coefficient (De) and lifetimes (τe) in TiO2 photoelec-trode were measured by the stepped light-induced transient measurements of photocurrent and voltage (SLIM-PCV).18 The transients were induced by a stepwise change in the laser intensity. A diode laser (λ = 635 nm) as a light source was modulated using a function generator. The initial laser intensity was a constant 90 mW cm−2 and was attenuated up to approximately 10 mW cm−2 using a ND filter which was positioned at the front side of the fabricated samples (TiO2 film thickness = 8.4 µm; active area = 0.04 cm2). The photo-current and photovoltage transients were monitored using a digital oscilloscope through an amplifier. The De value was obtained by a time constant (τc) determined by fitting a decay of the photocurrent transient with exp(−t/τc) and the TiO2 film thickness (ω) using the equation, De = ω2/(2.77τc).18 The τe value was also determined by fitting a decay of photovoltage transient with exp(−t/τe).18 All experiments were carried out at room temperature.

Fabrication of Cells. For the preparation of DSSC, a washed FTO glass plate (Pilkington TEC Glass-TEC 8, Solar 2.3 mm thickness) was fabricated using a modified procedure of previous references.12(a) First, the FTO glass plate was subject to doctor blade printing of TiO2 paste (Solaronix, Ti-Nanoxide T/SP) and the second opaque layer containing 400 nm sized anatase particles (CCIC, PST-400C) was coated for the purpose of light scattering. The TiO2 electrodes were immersed into the dyes (D5L6, D21L6, and JK-310) solution (0.3 mM in acetonitrile:tert-butanol: ethanol = 2:1:1) and kept at room temperature for 16 hour. Counter electrodes were prepared by coating with a drop of H2PtCl6 soultion (2 mg Pt in 1 mL ethanol) on a FTO plate. A drop of electrolyte solution (2 M 1-propyl-3-methylimidazolium iodide (PMMI), 0.05 M iodine, 0.1 M guanidinium thiocyanate (GuSCN), 0.5 M tert-butylpyridine (TBP) in methoxypropionitrile (MPN) and water) was placed on the drilled hole in the counter electrode of the assembled cell and was driven into the cell via vacuum backfilling. Finally, the hole was sealed using additional surlyn and a cover glass (0.1 mm thickness).

4-(2-(2-Methoxyethoxy)ethoxy)-N-(4-(2-(2-methoxy-ethoxy)ethoxy)phenyl)-N-phenyl-aniline (2). A mixture of compound 1 (8.36 mmol, 2.30 g), aniline (3.80 mmol, 0.35 mL), Pd2(dba)3 (0.03 mol ％, 0.15 g), P(t-Bu)3 (0.03 mol ％, 0.04 mL), and t-BuONa (12.49 mmol, 1.20 g) in dry toluene was refluxed for 24 h. After cooling, the solvent was evaporated. And the reaction mixture was poured into water and extracted with methylene chloride. The organic layer was then dried over MgSO4 and the volatile organic solvent was removed using rotary evaporator. The pure product 2 was obtained by silica gel chromatography using a mixture of ethyl acetate and n-hexane (1:5) as an eluent (yield 90.8％). 1H NMR (300 MHz, Acetone-d6) δ 7.17 (t, J = 7.5 Hz, 2H), 7.01 (d, J = 9.0 Hz, 4H), 6.93 (d, J = 9.0 Hz, 4H), 6.87-6.85 (m, 3H), 4.10 (t, J = 4.5 Hz, 4H), 3.80 (t, J = 4.5 Hz, 4H), 3.64 (t, J = 4.5 Hz, 4H), 3.50 (t, J = 4.2 Hz, 4H), 3.29 (s, 6H). 13C NMR (75 MHz, Acetone-d6) δ 156.2, 149.7, 141.9, 130.5, 127.3, 121.4, 121.3, 116.2, 72.6, 71.2, 70.3, 68.5, 58.8. MS: m/z 481 [M+]. Anal. Calc. for C28H35NO6: C, 69.83; H, 7.33. Found: C, 69.44; H, 7.26.

4-Bromo-N,N-bis(4-(2-(2-methoxyethoxy)ethoxy)phen-yl)aniline (3). 4-(2-(2-Methoxyethoxy)ethoxy)-N-(4-(2-(2-methoxyethoxy)ethoxy)-phenyl)-N-phenyl-aniline 2 (5.21 mmol, 2.51 g) was dissolved in CH2Cl2 (50 mL). N-Bromo-succinimide (5.45 mmol, 0.97 g) was added in one portion at RT. The reaction was stirred at room temperature and was poured into 50 mL of water. The organic layer was separated and washed with 2 M NaOH, followed by an aqueous wash. The organic layer was dried over MgSO4 and the volatile organic solvent was removed using rotary evaporator. The pure product 3 was obtained by silica gel chromatography using a mixture of ethyl acetate and n-hexane (1:5) as an eluent (yield 96％). 1H NMR (300 MHz, Acetone-d6) δ 7.29 (d, J = 8.4 Hz, 2H), 7.05 (d, J = 8.4 Hz, 4H), 6.93 (d, J = 9.0 Hz, 4H), 6.76 (d, J = 8.4 Hz, 2H), 4.12 (t, J = 5.1 Hz, 4H), 3.80 (t, J = 4.8 Hz, 4H), 3.64 (t, J = 5.1 Hz, 4H), 3.50 (t, J = 5.1 Hz, 4H), 3.29 (s, 6H). 13C NMR (75 MHz, Acetone-d6) δ 156.5, 149.0, 141.0, 132.4, 127.6, 122.1, 116.3, 112.2, 72.5, 71.1, 70.2, 68.4, 58.8. MS: m/z 559 [M+]. Anal. Calc. for C28H34BrNO6: C, 60.00; H, 6.11. Found: C, 59.64; H, 6.06.

4-(5'-(5,5-Dimethyl-1,3-dioxan-2-yl)-2,2'-bithiophen-5-yl)-N,N-bis(4-(2-(2-methoxyethoxy)ethoxy)phenyl)aniline (4). Under a nitrogen atmosphere, a mixture of compound 3 (2.00 mmol, 1.12 g), 4,4,5,5-tetramethyl-2-(5-(5,5-dimeth-yl-1,3-dioxan-2-yl)thiophen-2-yl)-1,3,2-dioxaborolane (0.10 mmol, 1.21 g), 2 M solution of K2CO3 (11.94 mmol, 1.65 g) in H2O (6.0 mL), Pd(PPh3)4 (2.00 mmol, 0.12 g) in dry THF was refluxed for 18 h. After cooling the solution, the organic layer was removed in vacuo. The crude product 4 was extracted with CH2Cl2, dried over MgSO4 and evaporated in vacuo leading to a dark yellow solid directly used in the next step. 1H NMR (300 MHz, Acetone-d6) δ 7.49 (d, J = 9.0 Hz, 2H), 7.26 (d, J = 3.9 Hz, 1H), 7.22 (d, J = 3.9 Hz, 1H), 7.13 (d, J = 3.9 Hz, 1H), 7.90 (d, J = 9.0 Hz, 4H), 7.05 (d, J = 3.9 Hz, 1H), 6.95 (d, J = 9.0 Hz, 4H), 6.87 (d, J = 8.7 Hz, 2H), 5.67 (s, 1H), 4.13 (t, J = 4.8 Hz, 4H), 3.81 (t, J = 4.5 Hz, 4H), 3.70 (s, 1H), 3.51 (t, J = 4.8 Hz, 4H), 3.31 (s, 6H), 1.22 (s, 6H).

5'-(4-(Bis(4-(2-(2-methoxyethoxy)ethoxy)phenyl)-amino)-phenyl)-2,2'-bithiophene-5-carbaldehyde (5). To the crude compound 4 (1.53 mmol, 1.16 g) in THF was added water. TFA (2.34 mL) was added to the solution. The mixture was stirred for 2 hrs. The solution was quenched with saturated aqueous sodium bicarbonate, extracted with CH2Cl2 and dried with MgSO4. The pure product 5 was obtained by silica gel chromatography using a mixture of ethyl acetate and n-hexane (1:1) as an eluent (yield 74.7％). 1H NMR (300 MHz, Acetone-d6) δ 9.91 (s, 1H), 7.91 (d, J = 4.2 Hz, 1H), 7.54-7.49 (m, 3H), 7.45 (d, J = 3.9 Hz, 1H), 7.35 (d, J = 3.9 Hz, 1H), 7.10 (d, J = 8.7 Hz, 4H), 6.96 (d, J = 9.0 Hz, 4H), 6.87 (d, J = 8.7 Hz, 2H), 4.12 (t, J = 4.5 Hz, 4H), 3.80 (t, J = 4.8 Hz, 4H), 3.64 (t, J = 4.2 Hz, 4H), 3.50 (t, J = 4.8 Hz, 4H), 3.29 (s, 6H). 13C NMR (75 MHz, Acetone-d6) δ 183.2, 156.5, 149.8, 146.9, 146.8, 142.1, 140.7, 138.8, 133.8, 128.2, 127.7, 126.9, 125.4, 124.7, 123.6, 119.9, 116.1, 72.3, 70.9, 70.0, 68.3, 58.5. MS: m/z 673 [M+]. Anal. Calc. for C37H39NO7S2: C, 65.95; H, 5.83. Found: C, 65.48; H, 5.74.

(E)-3-(5'-(4-(Bis(4-(2-(2-methoxyethoxy)ethoxy)phen-yl)-amino)phenyl)-2,2'-bithiophen-5-yl)-2-cyanoacrylic acid (JK-310). The resulting carbaldehyde bithiophene (1.10 mmol, 0.74 g) and cyanoacetic acid (2.23 mmol, 0.19 g) were allowed to react in acetonitrile and chloroform in the presence of piperidine (2.23 mmol, 0.22 mL). The solution was refluxed for 18 h. After removal of solvent in vacuo, the crude product was extracted with methylene chloride and water. The red solid product JK-310 was obtained by silica gel chromatography using a mixture of methylene chloride and methanol (9:1) as an eluent (yield 75.7％). 1H NMR (300 MHz, DMSO-d6) δ 8.17 (s, 1H), 7.66 (d, J = 4.5 Hz, 1H), 7.46 (d, J = 8.4 Hz, 2H), 7.40 (d, J = 4.2 Hz, 2H), 7.31 (d, J = 4.2 Hz, 1H), 7.02, (d, J = 8.4 Hz, 4H), 6.92 (d, J = 8.7 Hz, 4H), 6.74 (d, J = 8.4 Hz, 2H), 4.06 (t, J = 4.5 Hz, 4H), 3.72 (t, J = 5.1 Hz, 4H), 3.58 (t, J = 5.1 Hz, 4H), 3.45 (t, J = 5.1 Hz, 4H), 3.24 (s, 6H) . 13C NMR (75 MHz, DMSO-d6) δ 164.2, 155.2, 148.3, 144.6, 142.1, 141.2, 139.5, 136.9, 135.0, 133.0, 127.7, 127.0, 126.3, 124.3, 119.7, 119.0, 116.3, 115.9,114.8, 107.6, 71.3, 69.7, 69.0, 67.3, 58.1. MS: m/z 740 [M+]. Anal. Calc. for C40H40N2O8S2: C, 64.85; H, 5.44. Found: C, 64.42; H, 5.36.

 

Results and Discussion

Scheme 1 illustrates the synthetic procedures of organic sensitizer JK-310. N-Phenylation of aniline with 1-bromo-4-(2-(2-methoxyethoxy)ethoxy)benzene was performed under Ullmann′s condition,19 followed by bromination to give 2. Suzuki coupling reaction20 of 2 with 4,4,5,5-tetramethyl-2-(5-(5,5-dimethyl-1,3-dioxan-2-yl)thiophene-2-yl)-1,3,2-dioxa-borolane led to 3. Subsequent cleavage of 1,3-dioxalane protecting group in aqueous acid afforded the aldehyde. The aldehyde, on reaction with cyanoacetic acid in the presence of a catalytic amount of piperidine in acetonitrile, produced the JK-310 sensitizer.

Scheme 1.Schematic diagram of the synthesis of JK-310.

Figure 2.Absorption and emission spectra of D5L6 (black solid line), D21L6 (blue solid line), and JK-310 (red solid line) in THF and absorption spectra of D5L6 (black dot line), D21L6 (blue dot line), and JK-310 (red dot line) adsorbed on TiO2 film.

The absorption and emission spectra of D5L6, D21L6 and JK-310 in THF are shown in Figure 2 and listed in Table 1, together with the UV-Vis absorption spectra of the corre-sponding sensitizers adsorbed on TiO2 film. The absorption spectrum of D5L6 displays visible band at 446 nm, which is due to the π-π* transition of the conjugate system. Under the same conditions, the D21L6 and JK-310 sensitizers that contain the hexyloxy or (2-methoxyethoxy)ethoxy unit on phenyl group cause a red shift to 465 and 470 nm relative to the D5L6, respectively. A red shift of D21L6 and JK-310 can be readily understood from the presence of an electron rich alkyloxy unit and molecular modeling studies of D5L6 and D21L6. The ground state structure of D5L6 possesses a 38° twist between the phenyl and thiophenyl unit. For D21L6, the dihedral angle is 25°. Therefore, a red shift of D21L6 relative to D5L6 derives from more delocalization over an entire conjugation system.

Table 1.aAbsorption spectra were measured in THF solution. bRedox potential of dyes on TiO2 were measured in CH3CN with 0.1 M (n-C4H9)4NPF6 with a scan rate of 50 mVs−1 (vs. NHE). cE0-0 was determined from inter-section of absorption and emission spectra in ethanol. dEs+/s* was calculated by Es+/s − E0-0.

To evaluate the molecular energy levels, the electro-chemical properties of the three sensitizers were measured using cyclic voltammetry in acetonitrile. TiO2 films stained with sensitizer were used as working electrodes. The three sensitizers adsorbed on TiO2 film show a quasi-reversible behavior. The oxidation potentials of D5L6, D21L6, and JK-310, corresponding to HOMO level, were measured to be 1.23, 1.04, 1.01 V vs NHE, respectively, energetically favorable for iodide oxidation. The reduction potentials of the three sensitizers calculated from the oxidation potentials and the E0-0 determined from the intersection of absorption and emission spectra were calculated to be −1.11, −1.24, and −1.27 V vs NHE, ensuring an enough driving force for electron injection.

Figure 3.Isodensity surface plots of the HOMO, HOMO-1, LUMO, and LUMO+1 of JK-310.

Molecular orbital calculation has indicated that the HOMO of JK-310 is spread over the triphenylamino unit and that the LUMO is located on the cyanoacrylic unit and extends to the adjacent thiophene unit (Figure 3). Examination of the HOMO and LUMO of JK-310 indicates that HOMO-LUMO excitation moves the electron from the triphenylamino unit to the cyanoacrylic moiety. The change in local electron density distribution induced by photoexcitation results in efficient charge separation.

Table 2 demonstrates the device performance of three cells fabricated with the addition of up to 100％ water to the organic electrolyte. The electrolyte used in three cells con-sists of 2 M 1-propyl-3-methylimidazolium iodide (PMMI), 0.05 M iodine, 0.1 M guanidinium thiocyanate (GuSCN), and 0.5 M tert-butylpyridine (TBP) in methoxy propionitrile (MPN). In order to prevent phase separation in the aqueous electrolyte, 1％ surfactant Triton X-100 was added to the electrolytic solution.21 To see the effect of hydrophilic unit on the photovoltaic performances, we selected two known dyes (D5L615 and D21L616) and synthesized new organic dye (JK-310) having water-soluble substituent. As shown in Table 2, the open-circuit voltage (Voc) in the water-contain-ing electrolyte gradually increased with the incremental addition of water, but the short-circuit photocurrent (Jsc) sharply decreased under the same conditions. The decrease of photocurrent may be attributable to the dye detachment,7 formation of iodates,6 and decrease in electron lifetime.19 A similar phenomenon was observed in the previous work in an aqueous electrolyte.12

Table 2.aPerformances of DSSCs were measured with 0.18 cm2 working area. Electrolyte: 2 M 1-propyl-3-methylimidazolium iodide (PMMI), 0.05 M iodine, 0.1 M guanidinium thiocyanate (GuSCN), 0.5 M tert-butyl-pyridine (TBP) in methoxypropionitrile (MPN) and water.

Figure 4.J-V and IPCE spectra of (a) D5L6, (b) D21L6, and (c) JK-310.

The incident photon-to-current conversion efficiency (IPCE) of the devices based on the three dyes is presented in the inset of Figure 4. The IPCE of D21L6 cell exceeds 50％ in a broad spectrum range from 420 to 570 nm, reaching its maximum of 58％ at 480 nm in 100％ water based electro-lyte. The onset of IPCE spectrum employing D21L6 is 720 nm, about 40 nm red-shifted as compared to that using D5L6 due to an apparent red-shift in the absorption spec-trum. Under standard global air mass 1.5 solar condition, the D2L6 sensitized cell in 100％ water containing electrolyte gave a Jsc of 6.22 mAcm−2, a Voc of 0.75 V, and a ff of 0.73, corresponding to an η value of 3.40％. Under the same conditions, the D21L6 sensitized cell gave a Jsc of 7.46 mAcm−2, a Voc of 0.77 V, and a ff of 0.77, affording a η value of 4.41％, which is the highest one in aqueous DSSCs with I−/I3− electrolyte. A remarkable increase of efficiency in D21L6 with hydrophilic substituent relative to D5L6 may be due to the wettability increase between the adsorbed dye and electrolyte. The fill factor upon the incremental addition of water in both cases was slightly increased due probably to the fast catalytic characteristic, resulting in the reduced resistance. We also fabricated the cell using the sensitizer having hydrophilic unit (JK-310). Upon the incremental addition of water, the efficiency was dropped from 6.18％ in organic electrolyte to 3.77％ in 100％ water electrolyte. From these results, we have observed that the η value of the D21L6 based cell is higher than that of the D5L6 or JK-310 based cell due to a large photocurrent. The large photo-current in D21L6 relative to D5L6 or JK-310 may originate from a red-shifted absorption band or the minimum dye detachment from the TiO2 films in aqueous electrolyte, respectively. To check the effect of dye packing on three sensitizers, we measured the amount of dyes adsorbed on the TiO2 films by desorbing the dyes from the TiO2 surface with KOH. The amount of three dyes adsorbed on the TiO2 film were measured to be 8.04 × 10−6, 7.46 × 10−6, and 7.32 × 10−6 mmol cm−2 for D5L6, D21L6, and JK-310, respective-ly. To check the dye detachment from the TiO2 films in aque-ous electrolyte, we put three cells in 100％ water electrolyte for 1 h, and three cells were washed with acetonitrile. The amount of thee dyes remained on the TiO2 film were measured to be 4.86 × 10−6, 4.42 × 10−6, and 3.82 × 10−6 mmol cm−2 for D5L6, D21L6, and JK-310, respectively. Accordingly, the low adsorption of JK-310 relative to two dyes may be attributable to the facile dye detachment due to the hydrophilic nature, resulting in the reduction of photo-current.

Figure 5 shows the electron diffusion coefficients and life-times of the DSSCs employing three dyes (D5L6, D21L6, and JK-310) and different electrolyte solvents (MPN and water) displayed as a function of the short-circuit current density and the open-circuit voltage, respectively. No signi-ficant differences among the De values were observed at the identical short-circuit current conditions by changing the dye and electrolyte solvent.22 Meanwhile, the clear changes in the τe values were seen and the cells employing D21L6 showed higher values of open-circuit voltage and electron lifetime compared to those of the cells with D5L6 and JK-310. Both the τe and Voc values were shown to be increased by employing water as the electrolyte solvent rather than MPN. The results of electron lifetimes are also well con-sistent with those of the Voc shown in Table 2 (Figure 4).

Figure 5.Variations in electron diffusion coefficients (a) and lifetimes (b) of the photovoltaic cells by changing the dye and electrolyte solvent.

The ac impedances of the cells were measured under illumination and dark conditions. Figure 6 shows the ac impedance spectra measured under illumination condition. Under the illumination of 100 mW cm−2 under open-circuit voltage conditions in 0％ and 100％ water containing elec-trolyte, the radius of the intermediate frequency semicircle in the Nyquist plot decrease in the order of D21L6 (9.02 𝛀) ＞ JK-310 (10.06 𝛀) ＞ D5L6 (10.28 𝛀) and D21L6 (17.81 𝛀) ＞ JK-310 (18.04 𝛀) ＞ D5L6 (18.62 𝛀) in organic based electrolyte under illumination and dark condition, respec-tively, indicating the improved electron generation and transport. This result is in accord with the trend of short circuit photocurrent shown in Table 1. In the dark under forward bias (−0.67 V), the radius of the intermediate frequency semicircle in 0％ and 100％ water containing electrolyte showed the increasing order of D21L6 (68.37 𝛀) ＞ JK-310 (53.87 𝛀) ＞ D5L6 (50.70 𝛀) and D21L6 (262.51 𝛀) ＞ JK-310 (262.42 𝛀) ＞ D5L6 (262.18 𝛀) in water based electrolyte under illumination and dark condition, respec-tively, in accord with the trends of the Voc and τe values. The small difference of RCT in dark condition is that the hydrophilicity of three dyes in water is similar, facilitating similar charge transfer at the electrode/electrolyte interface.

Figure 6.Electrochemical impedance spectra measured under the illumination (100 mW cm−2) and dark for the devices employing different dyes (■ D5L6, ● D21L6, ▲ JK-310).

 

Conclusion

In conclusion, we investigated the influence of water in the electrolyte on the photovoltaic performance. A solar-cell device based on the organic dye D21L6 in 100％ water electrolyte gave an overall conversion efficiency of 4.41％, which is the highest one in aqueous DSSCs with I−/I3− electrolyte. The power conversion efficiency of the devices based on the three sensitizers was shown to be sensitive to the substituent unit on donor group. In order to increase the conversion efficiency, the dye needs to be designed in consideration of the wettability and detachment of dye from the TiO2 film. Using an aqueous based electrolyte, a greener process for DSSCs fabrication can be designed for the realization of environmentally benign solar cells.