Introduction
Dye sensitized solar cells (DSSCs) have attracted intense research interest due to their fabrication with low cost and high efficiencies of light-to-electrical conversion [1-5]. Although Ru-based complexes hold the efficiency recorded validated over 11% under full sunlight [6-9], their characteristics of high cost and relatively difficulty of purification limited their applications. On the other hand, organic dyes with easy structural tuning and generally high molar extinction coefficients have emerged as competitive alternatives to the Ru-based counterparts.
So far, many efficient organic dyes have been obtained, the highest efficiencies (η) of power conversion of DSSCs devices has achieved above 10% [10-14]. However, there were still some drawbacks to be overcome, such as the formation of aggregates on the surface of TiO2 and the relatively fast charge recombination of the injected electrons in the film of TiO2 with the redox species in the electrolyte. Thanks to enthusiastic efforts of chemists, some effective strategies have been proposed by extensive studies on molecular modification based on donor-π-acceptor configuration. Generally, the incorporation of the nonplanar structures into dye skeletons can hinder the π-π staking, and the introduction of long alkyl chains can block the redox species approaching the surface of TiO2, decreasing the concentrations of redox species at the vicinity of the TiO2, to increase the electron lifetime. Thus, our previous research has focused on the pyrrole-containing dyes [15,16], in which the triphenylamine (TPA) or indole was used as the electron donor, a cyanoacetic acid as the anchoring group. Another TPA group was introduced to the pyrrole ring through Carbon-Nitrogen single bond to suppress the possible aggregation of the dyes on the surface of TiO2, or a carbazolyl group with alkyl chains was employed to decrease the charge recombination. All of these dyes exhibited good performance, and the highest conversion efficiency was as high as 7.21% (91% of the standard cell from N719, 7.88%, measured under the same conditions). Apart from these modifications of the dye structures, the property of the pyrrole ring played a very important role: as one of the strongest electron-rich five-member heterocyclics, its abundant electron density should be beneficial to the charge transfer from the donor to the acceptor [17].
Recently, Liu group reported that the incorporation of the donor-π-acceptor dyes into the conjugated polymers could be an effective method to enhance the conversion efficiency for polymer DSSCs [18]. For example, the conjugated polymer containing TPA as the electron donating backbone, and an electron accepting side chain (cyanoacetic acid) with conjugated thiophene units as the linkers, gave good performance for the DSSC fabricated utilizing this material as the dye sensitizer. Thus, in view of the excellent properties of the pyrrole-based organic dyes, in this study, we designed a pyrrole-based conjugated oligomer, with phenyl-linked TPA as the isolation group, hexyloxyl and butyl moieties as the functional groups to align the oligomer and decrease the charge recombination. Here, the synthesis, structural characterization, electrochemical properties, and photovoltaic performance of pyrrole-based conjugated oligomer as organic dye for dye-sensitized solar cells were studied.
Experiment
Materials
Instrumentation
1H nuclear magnetic resonance (NMR) spectroscopy study was conducted with a Varian Mercury 300 spectrometer using tetramethylsilane (TMS; δ = 0 ppm) as internal standard. The Fourier transform infrared (FTIR) spectra were recorded on a Perkin Elmer-2 spectrometer in the region of 3000-400 cm-1 on NaCl pellets. Ultraviolet (UV)-visible spectra were obtained using a Schimadzu UV-2550 spectrometer. Gel permeation chromatography (GPC) analysis was performed on an Agilent 1100 series GPC system using a UV detector. Polystyrene standards were used as calibration standards for GPC. THF was used as an eluent and the flow rate was 1.0 mL/min. The electrochemical cyclic voltammetry was carried out on a CHI 600 voltammetric analyzer at room temperature in argon-purged anhydrous acetonitrile with tetrabutylammonium hexafluorophosphate (TBAPF6) as the supporting electrolyte at a scanning rate of 100 mV/s. A platinum disk and an Ag/Ag+ electrode were used as the working electrode and reference electrode, respectively. The ferrocene/ferrocenium redox couple was used for potential calibration.
Synthesis of oligomer P0
A mixture of compound 1 (1.00 g, 1.90 mmol), compound 2 (1.14 g, 1.90 mmol), sodium carbonate (2.01 g, 19.0 mmol), and tetrakis(triphenylphosphine)palladium (Pd(PPh3)4) (65.8 mg, 0.057 mmol) was carefully degassed and charged with argon. Subsequently, THF (60 mL) and deoxidized water (30 mL) were added by syringe. The reaction mixture was stirred at 80°C for 72 h. After cooled to room temperature, the organic layer was separated, THF was evaporated and methanol was added to precipitate the product, the oligomer P0 was further purified by reprecipitation with acetone/MeOH to afford yellow green solid P0 (0.78 g, 57.7%). Mw = 6700, Mw/Mn = 1.58 (GPC, polystyrene calibration). IR (thin film), υ (cm-1): 1657 (-CHO). 1H NMR (CDCl3) δ (ppm): 10.04 and 9.33 (-CHO), 7.78-6.74 (ArH and –CH= CH-), 4.00 (-O-CH2-), 3.81 (-N-CH2-), 1.71 (-CH2-), 1.28 (-CH2-), 0.87 (-CH3).
Synthesis of oligomer P1
The oligomer P0 (100 mg, 0.14 mmol) and cyanoacetic acid (16.2 mg, 0.19 mmol) were dissolved in THF (4 mL), and then 10 μL piperidine was added as the catalyst. The mixture was refluxed for 12 h. After cooled to room temperature, methanol was added to precipitate the product, the oligomer P1 was further purified by reprecipitation with acetone/MeOH to afford red solid P1 (70.5 mg, 60.7%). Mw = 3400, Mw/Mn = 1.55 (GPC, polystyrene calibration). IR (thin film), υ (cm-1): 2214 (-CN). 1H NMR (CDCl3) δ (ppm): 7.98–6.24 (ArH and –CH= CH-), 4.21-3.64 (-O-CH2- and-N-CH2-), 2.08 (-O-CH2-), 1.71 (-CH2-), 1.26 (-CH2-), 0.88 (-CH3).
Photoelectrode preparation
A screen-printed double layer of TiO2 particles was used as photoelectrodes. A 10 μm thick film of 13 nm size TiO2 particles (Ti-Nanoxide T/SP) was first printed on the fluorine-doped tin oxide (FTO) conducting glass and further coated by a 4 μm thick second layer of 400 nm light scattering anatase particles (Ti-Nanoxide 300). Sintering was carried out at 450°C for 30 min. Before immersed in the dye solution, these films were soaked in 0.2 M aqueous titanium(IV) chloride (TiCl4) solution overnight in a closed chamber, which has been proved to increase short-circuit photocurrent significantly. After being washed with deionized water and fully rinsed with ethanol, the films were heated again at 450°C for 30 min followed by cooling to 80°C and dipped into a 3 × 10-4 M solution of dyes in tetrahydrofuran (THF)/ethanol (2∶3) for 12 h at room temperature.
Counter platinum (Pt)-electrodes preparation
The Pt electrode was deposited on the cleaned FTO glass by coating with a drop of platinum (VI) tetrachloride dihydride (H2PtCl6) solution (0.02 M in 2-Propanol solution) with heat treatment at 400°C for 15 min. A hole (0.8mm diameter) was drilled in the FTO glass by a Drill-press as a counter electrode The perforated sheet was cleaned by ultrasound in an ethanol bath for 10 min.
DSSCs assemblage
The dye-covered TiO2 electrode and Pt-counter electrode were assembled into a sandwich-type cell and sealed with a hot-melt gasket of 25 μm thickness made of the ionomer Surlyn 1702 (Dupont). The size of the TiO2 electrodes used was 0.25 cm2 (i.e., 5 mm × 5 mm). A drop of the electrolyte was put on the hole at the back of the counter electrode. It was introduced into the cell via vacuum back filling. The hole in the counter electrode was sealed by a film of Surlyn 1702 and a cover glass (0.1 mm thickness) using a hot iron bar.
Photoelectrochemical measurements
Photovoltaic measurements were employed by an AM1.5 solar simulator equipped with a 300 W xenon lamp (Model No. 91160, Oriel). The power of the simulated light was calibrated to 100 mW/cm2 using a Newport Oriel PV reference cell system (Model 91150V). I (current)–V (voltage) curves were obtained by applying an external bias to the cell and measuring the generated photocurrent with a Keithley model 2400 digital source meter. The voltage step and delay time of photocurrent were 10 mV and 40 ms, respectively. Cell active area tested with a mask of 0.158 cm2.
Results and discussion
Synthesis and characterization
The synthetic route was shown in Scheme 1. The oligomer P0 was obtained by the palladium-catalyzed Suzuki coupling reaction between compounds 1 and 2. Through the post-function of the oligomer P0 by Knoevenagel condensation reaction, oligomer P1 was prepared with high yield, as determined by the infrared (IR) spectra (Fig. 1). In the IR spectra of oligomer P1, the strong peak of the aldehyde group in the oligomer P0 at about 1660 cm-1 diminished, and the characteristic peak of the cyano (CN) group appeared at 2214 cm-1. It should be pointed out that, by utilizing the Suzuki coupling reaction, different aryl moieties as the isolation group could be introduced into this system conveniently through single bonds. Thus, the size of steric hindrance for passivating the core component of the oligomer in donor-π-acceptor type could be tuned to optimize the properties of this material. Oligomer P1 was soluble in common solvents, such as THF, chloroform, dimethylfomamide (DMF)and dimethyl sulphoxide (DMSO). The molecular weight of oligomer P1 was determined by gel permeation chromatography (GPC) with THF as eluent, and polystyrene standards as calibration standards.
Optical properties
The UV-vis spectra of oligomer P1 were demonstrated in Fig. 2, while the maximum absorption wavelength (λmax) of the oligomer P1 in THF and on TiO2 surface shown in Table 1. There were two distinct absorptions around 340 nm and 460 nm in the UV-vis spectra of oligomer P1. The absorption band around 340 nm was assigned to a π-π* transition, while the absorption band with λmax around 460 nm corresponding to an intramolecular charge transfer (ICT) from the indole donor part to the acceptor end group. Compared to the UV-vis spectra of oligomer P1 tested in THF solution and in film, their λmaxs were almost the same, indicating that there were nearly no aggregation in the solid state of oligomer P1. When the oligomer P1 was adsorbed on TiO2 surface, the λmax did not show a big difference in comparison with that tested in THF solution either, suggesting that most of the dyes adsorbed on TiO2 surface were in the monomeric state. Its regular alignment should be ascribed to the rigid group, which was linked to the pyrrole moieties as the isolation group and its unique configuration.
Tab.1 Some characterization data of P1 |
| dye | Mwa) | Mw/Mna) | λmax/nmb) | ϵ at λmax/(M-1·cm-1)b) | λmax/nmc) | E0-0/eVd) | Eox/Ve) vs NHE | Ered/Ve) vs NHE | Egap/Vf) |
|---|---|---|---|---|---|---|---|---|---|
| LI-1 | 3400 | 1.5 | 462 | 21800 | 465 | 2.16 | 0.89 | -1.27 | 0.77 |
Notes: a) Determined by GPC in THF on the basis of a polystyrene calibration. b) Absorption spectrum of oligomer P1 measured in THF with the concentration of 3 × 10-5 M. c) Absorption spectrum of oligomer P1 adsorbed on the surface of TiO2. d) The bandgap, E0-0 was derived from the observed optical edge. e) The oxidation potential (Eox) referenced to calibrated Ag/Ag+ was converted to the NHE reference scale. Ered was calculated from Eox - E0-0. f) Egap was the energy gap between the Ered of oligomer P1 and the CB level of TiO2 (- 0.5 V vs NHE) |
Electrochemical properties
Cyclic voltametry was employed to investigate the redox behavior of oligomer P1 (Fig. 3). In the anodic scan, the onsets of oxidation of the oligomer P1 was found to occur at 0.69 V, which was converted to the normal hydrogen electrode (NHE) reference scale as 0.89 V, while the reduction potential vs NHE (Ered) could be calculated from Eox-Eo-o (Table 1). Both of the Eox and Ered corresponded to the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular obtital (LUMO), respectively. As shown in Fig. 4, both LUMO and HOMO levels of oligomer P1 agreed well with the requirements for an efficient dye sensitizer: first, the LUMO level was negative of the CB level (-0.5 V vs NHE) of TiO2, and the Egap was 0.77, indicating that the driving force was large enough for the injection; secondly, HOMO level of the oligomer P1 was more positive than the iodine redox potential (0.4 V), which ensured that the oxidized dye could be regenerated from the reduced species in the electrolyte.
Photovoltaic performance of DSSCs
Photovoltaic properties of the solar cell fabricated with the oligomer P1 was measured under simulated AM 1.5G irradiation. As shown in Fig. 5, the DSSC based on oligomer P1 gave a short-circuit photocurrent density (Jsc) of 1.22 mA/cm2, an open-circuit voltage (Voc) of 0.54 V, and a fill factor (FF) of 0.73, corresponding to an overall conversion efficiency (η) of 0.48%, without any of the additives typically used (e.g. 3a,4a-dihydroxy-5b-cholic acid) for enhancing conversion efficiencies. The relatively lower η value should be ascribed to the weak light harvesting capability of oligomer P1 on the surface of TiO2. As seen from its absorption spectrum the surface of TiO2, the alignment of the dyes was good, thus, the weak light harvesting capability should possibly be caused by the small amount of the dyes on TiO2 surface, which was usually affected by the size of the isolation group and the configuration of the oligomer. For the core structure of oligomer P1 in donor-π-acceptor type, phenyl-linked TPA as the isolation group perhaps was too large. Although they could well suppress the possible aggregation, its large size decreased the effective component on the surface of TiO2 for the light adsorption, then leading to the weak light harvesting capability. Therefore, in our further study, the size of the isolation group of the oilgomer should be adjusted to balance the two effects, the suppressing of the aggregations and the effective component on TiO2 surface for absorbing light.
Conclusions
In summary, we have demonstrated a new conjugated oligomer dye as an efficient organic sensitizer in DSSCs. After the incorporation of the donor-π-acceptor type dyes based on pyrrole as the core component and phenyl-linked TPA as the isolation group, the DSSC based on oligomer P1 gave a conversion efficiency of 0.48%. It was expected that higher conversion efficiencies of this kind of oligomers would be achieved by tuning the size of the isolation groups and the structure of the donor-π-acceptor type dyes.