Incorporation of a fluorophenylene spacer into a highly efficient organic dye for solid-state dye-sensitized solar cells

Accepted Manuscript Title: Incorporation of a fluorophenylene spacer into a highly efficient organic dye for solid-state Dye-Sensitized Solar Cells Author: V. Leandri J. Zhang E. Mijangos G. Boschloo S. Ott PII: DOI: Reference:

S1010-6030(16)30133-2 http://dx.doi.org/doi:10.1016/j.jphotochem.2016.05.015 JPC 10234

To appear in:

Journal of Photochemistry and Photobiology A: Chemistry

Received date: Revised date: Accepted date:

22-2-2016 17-5-2016 19-5-2016

Please cite this article as: V.Leandri, J.Zhang, E.Mijangos, G.Boschloo, S.Ott, Incorporation of a fluorophenylene spacer into a highly efficient organic dye for solid-state Dye-Sensitized Solar Cells, Journal of Photochemistry and Photobiology A: Chemistry http://dx.doi.org/10.1016/j.jphotochem.2016.05.015 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Incorporation of a fluorophenylene spacer into a highly efficient organic dye for solid-state Dye-Sensitized Solar Cells

V. Leandri,a J. Zhang,b E. Mijangos,a G. Boschloo,*b S. Otta

a

Molecular Biomimetics, Department of Chemistry, Ångström Laboratory, Uppsala University (UU), SE-751 20 Uppsala, Sweden. E-mail: [email protected]

b

Physical Chemistry, Centre of Molecular Devices, Department of Chemistry, Ångström Laboratory, Uppsala University (UU), SE-751 20 Uppsala, Sweden. *Corresponding author. E-mail address: [email protected] (G. Boschloo).

Abstract A new efficient organic dye LEG4F, incorporating a fluorine-substituted phenylene unit in the spacer, has been synthesized and tested in dye-sensitized solar cells. Direct comparison with the parent dye LEG4 shows very similar performances in case of liquid electrolyte devices based on iodide/triiodide, yielding a power conversion efficiency of 6.8% under 1 sun and 8.0-8.2% under 0.5 sun illumination. However, LEG4F outperforms its fluorine-free analogue when the liquid electrolyte is replaced by the solid-state hole-transport material Spiro-OMeTAD, reaching 5.3% efficiency compared to 4.8% achieved by LEG4. We show that this improvement is due to the enhancement of the electron lifetime, which reduces recombination at the TiO2/dye/Spiro-OMeTAD interfaces. Keywords: Dye Sensitized Solar Cells, Organic Dyes, Fluorine Dyes.

Introduction Dye-Sensitized Solar Cells (DSSCs) are an attractive and promising photovoltaic technology for future commercialization, due to their high versatility and low-fabrication costs.1,2,3 Since the remarkable breakthrough by O´Regan and Grätzel in 1991,4 most of the research has been focused on liquidbased electrolyte devices in which a dye-sensitized TiO2 electrode and a counter-electrode are placed in contact through a liquid solution of the redox mediator. Due to the superior performances recorded with liquid electrolytes, highly efficient and competitive devices have been produced.567 Despite the outstanding performances of liquid DSSCs, a practical disadvantage of this system for future commercialization is the volatile nature of the solvents commonly employed. 5,6,7,8,9,10,11,12,13 In order to overcome this issue, alternative solid state DSSCs (ssDSSCs) based on hole-conducting small molecules or polymer, have been developed.14,15,16 However, the power-to-current efficiency of ssDSSCs is generally lower than that of their liquid counterpart due to incomplete pore-filling of the hole-conducting material into the mesoporous TiO2, remarkable sensitivity to moisture of the widely used hole-conductor 2,2´,7,7´-tetrakis(N,N-di-p-methoxyphenyl-amine)9,9´-spirobifluorene (SpiroOMeTAD), and the difficulty to control the amount of Spiro-OMeTAD+ used as dopant.17,18,19,20 In particular, the pore-filling problem has been addressed as a notable limitation due to an imperfect contact between the dye and the hole-transport material, which results in increased recombination.21,22 The introduction of a fluorine atom in organic and metalorganic materials for photovoltaic application has been widely studied in the last years.23,24,25,26,27 The main advantage of incorporating a fluorine atom in a suitable position is the stabilization of the oxidized form of the material, which leads to reduced recombination in the devices and overall higher performances.27,28,29 This property has been tested in ssDSSCs by Chou et al., leading to an enhanced efficiency for the fluorine substituted dye due to higher open-circuit voltage (Voc) and short-circuit current (Jsc).30 Organic dyes based on 4,4-dihexyl-4H-cyclopenta-[2,1-b:3,4-b’]dithiophene (CPDT) segment as conjugated spacer have shown remarkable efficiencies.31,32 Interestingly, in combination with the CPDT moiety, the introduction of additionally substituted phenyl rings on the triphenylamine donor resulted in organic dyes (Y123, LEG4) with superior performances,13,5 which are currently holding the record efficiency in ssDSSC.1819 In particular, Grätzel et al. included additional phenyl rings on the diphenylamine donor of the asymmetric porphyrins SM371 and SM315, breaking the existing record efficiency for liquidelectrolyte DSSCs.6 In this study, we designed and synthesized a new dye (LEG4F) by adding a phenyl ring with a fluorine substituent in the dye spacer of LEG4, with the aim of improving the dye performance in ssDSSCs.

Experimental details All chemicals were purchased from Sigma-Aldrich and used as received unless noted otherwise. The organic dye LEG4F (structure in Fig. 1) was synthesized according to a synthetic procedure reported in the SI (Scheme S1). FTO-coated glass substrates were purchased from Pilkington. TEC15 and TEC8 were used for working electrodes and counter electrodes, respectively.

Optical characterization Equipment used for the evaluation of the molar extinction coefficient and UV-Vis absorption measurements: Varian Cary 50 UV/Vis Spectrometer. The absorption spectra of LEG4F and LEG4 on TiO2 were recorded on 4 m thick TiO2, electrodes fabricated according to the procedure reported in the liquid electrolyte DSSCs fabrication section. The electrodes were placed in a 0.1 mM solution of the dye in acetonitrile:t-butanol (1:1 volume ratio) for 3 hours and then rinsed with the same solvent mixture. Electrochemical characterization Electrochemical experiments were performed in a three-electrode electrochemical cell with a platinum mesh working electrode, glassy carbon counter electrode and Ag/AgNO3 (10mM/MeCN) as reference electrode. The samples were dissolved in 0.1 M t-Bu4NPF6 / DCM electrolyte solution in order to obtain dye concentration 10-3 M. The counter electrode was kept separate from the main solution by a salt bridge with glass frit tip. All electrochemical measurements were performed using an Autolab PGSTAT302 potentiostat/galvanostat with a GPES electrochemical interface. All experiments were carried out in a glovebox (MBraun) maintained at < 0.1 ppm O2 and H2O. The CVs of LEG4F and LEG4 on TiO2 were recorded on 4 m thick transparent TiO2 electrodes, fabricated and sensitized according to the procedure reported in the next section. Liquid electrolyte DSSCs fabrication TEC15 substrates were cleaned in an ultrasonic bath with detergent solution (RBS 25 from Fluka analytical), ethanol (VWR DBH Prolabo purity of 99.9%) and deionized water. The glass substrates were pre-treated in a 40 mM aqueous TiCl4 solution at 70°C for 90 min and then rinsed with water and ethanol. After drying in air the substrates were screen printed (active area 0.25 cm2) with a Dyesol 30 NR-D paste, 30 nm particle size (thickness around 8.0 m after sintering). The substrates were dried at 125 °C for 10 min before being screen-printed with a TiO2 scattering paste (Dyesol WER2-O). The samples were heated gradually at 180°C (10 min), 320°C (10 min), 390°C (10 min) and 500°C (60 min) in an oven (Nabertherm Controller P320) in air atmosphere. After sintering the samples were once again treated with 40 mM aqueous TiCl4 at 70°C, for 30 min. A final heating step (500°C for 60 min) was performed. Before immersing the electrodes in the dye bath the electrodes were cooled down to 90°C. Optimized substrates were screen-printed with a diluted (80% wt. paste, 20% wt. mixture of terpineol/ethyl cellulose) scattering paste (for a total thickness of 12-13 m), while non-optimized substrates were screen-printed with non-diluted scattering paste (for a total thickness of 11 m). Dye baths of LEG4 and LEG4F consisted of a 0.2 mM dye solution in acetonitrile:t-butanol (1:1 volume ratio). Different molar ratio of dye:chenodeoxycholic acid were added to the dye bath of LEG4F where reported. The films were left in a dye bath overnight (14 h) in the dark, rinsed with ethanol and assembled in a sandwich structure with the counter electrode. A 25 m thick thermoplastic Surlyn frame was employed (Meltonix 1170-25 from Solaronix). The electrolyte was introduced in the sealed devices through the predrilled hole by vacuum back-filling technique, sealed with a thermoplastic Surlyn cover and a glass coverslip. For counter-electrodes preparation, a predrilled one-hole TEC8 glass was cleaned following the same procedure reported for the working-electrodes, and heated in air at 400°C for 30 min to remove residual impurities. After cooling down to room temperature, 9 Lcm-2 of a 4.8 mM H2PtCl6 solution in ethanol was deposited

on the glass substrate, followed by heating in air at 400°C for 30 min. Optimized devices were assembled using freshly prepared working electrodes and counter electrodes, while non-optimized devices were assembled using previously prepared working electrodes, stored in air for days/weeks, and re-heated to 90°C prior to dye-bath immersion. Solid State DSSCs fabrication Fluorine-doped SnO2 (TEC15) substrates were etched with Zn powder and HCl (4 M) to form the desired electrode pattern. The substrates were cleaned in an ultrasonic bath for half an hour with the following solvents: deionized water, acetone and ethanol. A compact layer of TiO2, intended to block the recombination current at the FTO support, was prepared on cleaned FTO substrate by spray pyrolysis. The solution used for the spray pyrolysis was composed of 0.2 M Ti-isopropoxide and 2 M acetylacetone in isopropanol. In order to generate the porous TiO2 layer, the colloidal TiO2 paste (the weight fraction of Dyesol DSL 18NR-T with terpineol is 46%) was spin-coated (2400 rpm for 30 seconds) on the compact layer surface in order to get 2 μm thick TiO2, then sintered by a gradient heating process: 180°C for 10 minutes, 320°C for 10 minutes, 400°C for 10 minutes, 450°C for half of an hour (Nabertherm Controller P320) in air atmosphere. After sintering, the films were treated with 40 mM TiCl4 aqueous solution at 70°C for 30 minutes. After rinsing with deionized water and ethanol, the films were sintered again by following the previously mentioned procedure. After cooling to 90°C, the films were immersed into the dye solutions (0.2 mM in acetonitrile:t-butanol, 1:1 volume ratio) for 18 hours. Subsequently, the films were rinsed in ethanol and dried by a N2 gas flow. The dye-sensitized TiO2 layer was covered with a spiro-OMeTAD HTM solution by spin-coating deposition (2400 rpm for 30 seconds). The solution is composed of 180 mg/mL HTM, 0.05 M LiN(CF3SO2)2 and 0.2 M 4-tert-butylpyridine (TBP) in chlorobenzene. Finally, a 200 nm thick Ag (Sigma-Aldrich; ≥ 99.99% trace metals basis) contact layer was deposited onto the dye-sensitized TiO2 layer by thermal evaporation, in a vacuum chamber (Leica EM MED020) with a base pressure of about 10-5 mbar, in order to complete the device. DSSCs characterization Current–voltage (I–V) measurements were carried out with a Keithley 2400 source/meter and a Newport solar simulator (model 91160); the light intensity was calibrated using a certified reference solar cell (Fraunhofer ISE), to an intensity of 1000 Wm-2. For the I–V measurements of liquid electrolyte-based devices, a black mask with an aperture of 0.6 cm x 0.6 cm, slightly larger of the active area (0.5 cm x 0.5 cm) was used.33 For ssDSSCs characterization, the size of the aperture employed was 0.2 cm2 (same as active area). The apparatus for incident photon to current conversion efficiency (IPCE) and electron lifetime measurements has been previously described.11 The reported efficiency is the maximum value obtained from a set of 4 devices. The average efficiency deviates from the maximum value for 0.1-0.2 units.

Results and discussion Optical and electrochemical characterization The molecular structures of the dyes LEG4F and LEG4 are depicted in Figure 1.

Figure 1: Molecular structures of LEG4F and LEG4 dyes

LEG4F has been synthesized according to a simple and highly efficient 5-step synthetic route reported in the SI (Scheme S1). LEG4F was optically characterized and its properties were compared to the parent dye LEG4 in order to get insight on the variations introduced by the additional building block in the spacer.

a)

b)

Figure 2: a) Normalized absorption spectrum of LEG4F (red) and LEG4 (black) in dichloromethane solution; b) Normalized absorption spectra of the dyes absorbed on the surface of TiO2.

Both dyes exhibit a significant solvatochromism when dichloromethane (DCM) is replaced with acetonitrile:t-butanol (Fig 2, a). In DCM solution, LEG4F and LEG4 assume a dark purple color and the absorption spectra exhibit maximum wavelengths at 534 nm and 542 nm, respectively. However, when acetonitrile:t-butanol is employed as solvent, the solutions appear dark red and the maximum absorption wavelengths of LEG4F and LEG4 are shifted to 498 and 512, respectively. Table 1 summarizes the main optical parameters. Table 1: Optical parameters of dyes LEG4F and LEG4 dyes

Compound LEG4F LEG4 a

max (nm)a 534, 396, 336 542, 337 b

max (nm)b 498, 390, 331 512, 333

max (nm)c 474 482

fd 0.96 0.81

 (M-1cm-1) 50300b 49000f

c

Dichloromethane solution; Acetonitrile:t-butanol solution (1:1 volume ratio); Absorbed on the TiO2 surface d from acetonitrile:t-butanol solution (1:1 volume ratio) dye-bath; Oscillator strength measured in the range f 34 450-800 nm spectra in DCM (see SI for detailed procedure); Reported in literature.

In agreement with the data reported in literature,30,23 the introduction of an electron-withdrawing moiety in the -spacer causes a slight hypsochromic shift of the maximum absorption wavelength ascribed to the * charge transfer (CT) donor-acceptor transition. Nevertheless, the absorption profile of LEG4F appears more intense in the region between 300-500 nm and exhibits an additional optical transition located at 396 nm in DCM solution (Fig. 2, a). An additional blue-shift of the maximum absorption wavelengths of both dyes absorbed on the surface of TiO2, is observed (Fig. 2 b). This phenomena has been attributed to the deprotonation of the carboxylic acid moiety upon absorption on the TiO2 surface through bidentate mode.35,36 Finally, the insertion of the fluorinated phenyl ring in the -spacer of LEG4F, slightly increases the molar extinction coefficient from 49000 M-1cm-1 to 50300 M-1cm-1. The electrochemical properties of LEG4F were investigated by cyclic voltammetry (CV) (Figure 3). Due to the severe lack of solubility of the dye in acetonitrile, the electrochemical experiments in solution have been performed in DCM.

Figure 3: Cyclic voltammetry trace (Ag/AgNO3) of LEG4F in DCM solution (0.1 M TBAPF6 as supporting electrolyte).

Both compounds exhibit two distinctive reversible oxidation peaks in solution, the first related to the oxidation of the donor triphenylamine moiety, and the second probably related to the oxidation of the CPDT spacer. Table 2 summarizes the electrochemical data of LEG4F and LEG4. Table 2: Electrochemical properties of the dyes LEG4F and LEG4

Compound LEG4F LEG4

Eox (V vs NHE)a 0.95/1.26 0.87/1.26

LUMO (V vs NHE)b -1.10 -1.17

E0-0c 2.05 2.04

LUMO (V vs NHE)d -1.05 -1.12

a

E0-0e 2.19 2.15

Eox onTiO2 (V vs NHE)f 1.15/1.29 1.03/1.29g

b

Measured in DCM solution (supporting electrolyte 0.1 M TBAPF6); Calculated by subtracting E0-0 (from DCM) c from the oxidation potential of the dyes in DCM solution; Estimated from the Tauc plot of the absorption d spectra in DCM (SI, Figures S2 and S4); Calculated by subtracting E0-0 (from TiO2) the oxidation potential of the dyes on TiO2; eEstimated from the Tauc plot of the absorption spectra on TiO2 (SI, Fig. S3 and S5); fMeasured in g 0.1 M solution of TBAPF6 in acetonitrile as supporting electrolyte; Not reversible if oxidized to the second potential, dye decomposition observed after few scans.

The presence of the phenyl ring bearing a fluorine substituent, causes a shift of the first oxidation to more positive potential, which is in agreement with previous investigations.25,23 This shift slighly moves the LUMO level of LEG4F toward less negative potentials. Considering the conduction band potential of anatase TiO2 at -0.50 V vs NHE, and the potentials of the LUMO calculated from the dyes in DCM solution reported in Table 2, we calculate 0.60 eV (58 kJmol-1) and 0.67 eV (64 kJmol-1) injection driving force for LEG4F and LEG4, respectively. Although related to the dyes in solution, these values are a good approximation to the more precisely evaluated LUMO calculated from the redox potentials and UV-Vis of the dyes absorbed on TiO2, from which we estimate 0.55 eV (53 kJmol1 ) and 0.62 eV (60 kJmol-1) injection driving force for LEG4F and LEG4, respectively. Finally, it should be pointed out that the electrochemical data of the dyes in solution and absorbed on the surface of TiO2, show a smaller separation between the first and the second oxidation potential in the case of LEG4F. Interestingly, this separation is only 0.140 (13 kJmol-1) eV in the case of LEG4F absorbed on TiO2. Photovoltaic characterization Preliminary tests of LEG4F in liquid DSSCs were performed using a standard electrolyte composition based on the iodide/triiodide redox couple.37 Due to the lack of solubility of LEG4F in polar solvents, initial attempts were performed using DCM for the dye-bath (Table 3). Table 3: Effect of the dye-bath solvent on the photovoltaic performances

Dye LEG4F LEG4 LEG4F LEG4F LEG4F

Solventa DCM DCM ACN:valeronitrileb ACN:t-BuOHb ACN:t-BuOHb

Electrolyte A A A A B

Voc (mV) 640 535 685 715 720

Jsc (mAcm-2) 9.18 6.67 11.13 9.86 11.45

FF (%) 70 72 54 60 63

(%) 4.10 2.55 4.10 4.20 5.20

a

b

Solvent used for the dye-bath preparation; Solvents in 1:1 volume ratio; Electrolyte A: 0.5 M 1-Butyl-3Methylimidazolium Iodide, 0.1 M LiI, 0.05 M I2, 0.5 M 4-t-butylpyridine in acetonitrile (ACN); Electrolyte B: 1 M 1-Butyl-3-Methylimidazolium Iodide, 0.1 M LiI, 0.05 M I2, 0.5 M 4-t-butylpyridine in ACN.

The high solubility of LEG4F and LEG4 in dichloromethane shifts the equilibrium in favor of the dyes in solution, rather than on the surface of TiO2, causing low Voc and low efficiency. However, despite both dyes appear very soluble in DCM, LEG4F shows a higher Voc and significantly higher performance, which are the consequences of a superior dye-loading (SI, Fig. S1).38 This observation may be explained on the base of a higher affinity for TiO2, possibly due to an additional hydrogenbond interaction between the fluorine atom of the dye, and the proton of hydroxyl groups present on the surface of TiO2.25 In order to further improve the dye-loading and the efficiency, different solvents with gradually reduced solubility for the sensitizers have been tried. This investigation led to a significant increase of Voc and Jsc in case a relatively common acetonitrile/t-butanol mixture was used. The latter combination of solvents was thus employed for further investigations. Finally, higher concentration of the ionic liquid 1-butyl-3-methylimidazolium iodide in the electrolyte composition B, significantly increased the photocurrent (Jsc) yielding overall better efficiency. We speculate this effect may be the result of a better interaction of LEG4F with the electrolyte solution, due to a reduced resistance with the acetonitrile solvent. In fact, LEG4F is completely insoluble in acetonitrile. As a consequence, acetonitrile should not be able to efficiently solvate the dye molecules, limiting the electrolyte-dye interaction and causing higher resistance (lower FF). Since the addition of the fluorinated phenyl ring in the -spacer may induce dye aggregation, LEG4F was tested in combination with different amounts of chenodeoxycholic acid (CDCA). Table 4 reports the photovoltaic performances of DSSCs fabricated from LEG4F solutions with different ratio of chenodeoxycholic acid as deaggregating co-adsorbent. Table 4: Photovoltaic details of DSSCs based on LEG4F with different ratio of CDCA

Dye LEG4F

Dye:CDCA ratio no CDCA 1:10 1:30 1:50

Voc (mV) 720 710 685 655

Jsc (mAcm-2) 11.45 12.45 12.50 12.42

FF (%) 63 68 69 70

(%)a 5.20 6.00 5.90 5.70

a

Electrolyte composition B was used.

The significant photocurrent improvements that were recorded by adding CDCA to the sensitizing solutions, clearly show that the introduction of the fluorinated phenyl ring causes dye aggregation. Indeed, it is very well known that the length of the -spacer must be finely balanced accordingly to the position, the length, and the amount of alkyl chains present in the structure.31 Higher concentrations of CDCA diminish the amount of sensitizer absorbed on the surface of TiO2, which is an important aspect to consider in order to achieve higher performances. An inferior number of dye molecules absorbed on the surface causes a lower concentration of electrons injected into the conduction band of the semiconductor, which generates a more positive quasi-Fermi level, and a lower Voc. As expected, the value of the Voc is lower (655 mV) when a higher dye:CDCA ratio (1:50) is employed. Devices based on 1:10 and 1:30 of LEG4F:CDCA follow the trend of the Voc but show overall similar results. Indeed, the best performance is achieved by finely tuning the amount of CDCA employed in order to maximize the Voc, without significantly compromise the photocurrent due to

aggregation. Optimized working electrodes sensitized with LEG4F and LEG4 were used for the highest efficiencies, collected after one week from device assembly, in Table 5. Table 5: Photovoltaic parameters of optimized LEG4F and LEG4 DSSCs.

Dye LEG4Fa

Light Intensity 1 sun 0.85 sun 0.46 sun 0.114 sun 1 sun 0.85 sun 0.46 sun 0.114 sun

LEG4

a

Voc (mV) 825 815 800 760 790 785 770 725

Jsc (mAcm-2) 11.04 10.04 6.06 1.47 11.50 10.43 6.24 1.51

FF(%) 75 76 78 80 75 75 77 80

(%)b 6.80 7.30 8.20 7.80 6.80 7.20 8.00 7.70

b

Dye-bath 1:10=Dye:CDCA; Electrolyte composition B was used.

Despite the decrease of the photocurrent from the data initially recorded, great enhancements of the Voc and fill factor (FF), ensure superior power to current efficiency over time. A comparison with LEG4 dye shows equal performances of the dyes in combination with iodide/triiodide redox couple, although due to opposite trends in photocurrent and open-circuit voltage. Therefore, the data confirm that the inclusion of the fluorinated phenyl ring is able to reduce recombination (higher Voc) but, at the same time, limits the photocurrent. This latter aspect seems in agreement with the observations of Wei-Guang et al., who report a systematic trend of increased Voc and decreased Jsc by increasing the number of fluorine substituents on the ligands in a series of ruthenium complexes.26 In addition, the lower photocurrent can be additionally explained considering the lower driving force for electron injection into the conduction band of TiO2, due to the lower LUMO position of LEG4F (Table 2). LEG4F was additionally tested in solid-state DSSCs based on Spiro-OMeTAD as hole-transport material (Table 6). As previously seen in the iodide/triiodide based devices, LEG4F exhibits lower photocurrent but higher Voc. However, in the case of ssDSSCs (Figure 4a), this difference has a significant impact on the device efficiency (Table 6). As expected, due to imperfect contact between the dye and the hole-transport material, an increased Voc corresponds to an increased lifetime, which is able to reduce recombination and yield higher performance. Consequently, the longer lifetime and reduced recombination associated with LEG4F significantly reduce the shunt resistance at the materials interface, increasing the value of FF. Measurements of the electron lifetimes of the fabricated devices (Figure 4b) proved our rational design to be correct, confirming a further delay of the electron-recombination.

a)

b)

Figure 4: a) I-V characteristics of ssDSSCs based on LEG4F (red squares) and LEG4 (black squares); b) Electron lifetimes of the devices. Table 6: Detailed photovoltaic parameters of ssDSSCs under different light intensity

Dye LEG4F

LEG4

Light Intensity 1 sun 0.85 sun 0.46 sun 0.32 sun 0.114 sun 1 sun 0.85 sun 0.46 sun 0.32 sun 0.114 sun

Voc (mV) 870 860 840 830 800 830 820 800 790 750

Jsc (mAcm-2) 9.34 8.36 4.85 3.31 1.07 10.01 9.07 5.25 3.57 1.15

FF(%) 65 67 73 76 77 58 60 68 72 76

(%) 5.30 5.65 6.50 6.52 5.80 4.80 5.31 6.20 6.34 5.75

Analysis of the detailed photovoltaic parameters under different light intensity, additionally suggests lower charge-transport limitation in the case of LEG4F. Infact, despite the significantly different performance under 1 sun illumination, LEG4F and LEG4 exhibits very similar behavior under lower light intensities.

Figure 5: IPCE spectra of ssDSSC based on LEG4F (red) and LEG4 (black).

Finally, the lower photocurrent and incident-photon-to-current efficiency (IPCE, Figure 5) of LEG4F in ssDSSCs, can be ascribed to aggregation problems, previously observed in the devices based on iodide/triiodide. However, the well-established strategy of using CDCA as co-adsorbant,39,40,41,42 is not effective in the case of solid state DSSCs. We speculate that the imperfect contact at the dye/holetransport material (HTM) interface is the reason for this shortcoming. Indeed, the co-adsorbed CDCA partially replaces the dye, further reducing the possibility of a good dye-HTM connection.

Conclusion In conclusion, a new organic dye LEG4F has been synthesized and its optical, electrochemical, and photovoltaic properties have been compared with those of the parent dye, LEG4. The results support the success of the design, showing reduced recombination in the photovoltaic devices where LEG4F is employed. However, this latter aspect seems to be negligible in the case of liquid DSSCs employing iodide/triiodide as redox shuttle, where recombination is minimal. On the other hand, the superior performance of LEG4F over LEG4 in solid-state DSSCs reflects the obstacles in the realization of devices with perfect contact at the dye/HTM interface. Therefore, in this scenario, an enhancement of electron lifetime corresponds to increased performance. Consequently, we propose LEG4F as a good candidate for the realization of solid state Dye-Sensitized Solar Cells with improved performance.

Acknowledgements Financial support from the Carl-Trygger Foundation (postdoctoral stipend V. L.), the Knut and Alice Wallenberg Foundation and the Swedish Research Council is gratefully acknowledged.

Notes and references 1

Y. Wu, W.-H. Zhu, S. M. Zakeeruddin and M. Grätzel, ACS Appl. Mater. Interfaces, 2015, 7, 9307−9318.

2

Y. Cui, L. Zhang, K. Lv, G. Zhou and Z.-S. Wang, J. Mater. Chem. A, 2015, 3, 4477–4483.

3

N. G. Park and K. Kim, Phys. Status Solidi Appl. Mater. Sci., 2008, 205, 1895–1904.

4

B. O’Regan and M. Grätzel, Nature, 1991, 353, 737–739.

5

A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin and M. Grätzel, Science, 2011, 334, 629–34.

6

S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. E. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazeeruddin and M. Grätzel, Nat. Chem., 2014, 6, 242–247.

7

A. Reynal, A. Forneli and E. Palomares, Energy Environ. Sci., 2010, 3, 805–812.

8

Y.-S. Yen, Y.-C. Hsu, J. T. Lin, C.-W. Chang, C.-P. Hsu and D.-J. Yin, J. Phys. Chem. C, 2008, 112, 12557–12567.

9

A. Abbotto, V. Leandri, N. Manfredi, F. De Angelis, M. Pastore, J.-H. Yum, M. K. Nazeeruddin and M. Grätzel, European J. Org. Chem., 2011, 6195–6205.

10

L.-Y. Lin, C.-H. Tsai, F. Lin, T.-W. Huang, S.-H. Chou, C.-C. Wu and K.-T. Wong, Tetrahedron, 2012, 68, 7509–7516.

11

S. M. Feldt, E. a Gibson, E. Gabrielsson, L. Sun, G. Boschloo and A. Hagfeldt, J. Am. Chem. Soc., 2010, 132, 16714–24.

12

V. Leandri, H. Ellis, E. Gabrielsson, L. Sun, G. Boschloo and A. Hagfeldt, Phys. Chem. Chem. Phys., 2014, 16, 19964–19971.

13

J.-H. Yum, E. Baranoff, F. Kessler, T. Moehl, S. Ahmad, T. Bessho, A. Marchioro, E. Ghadiri, J.-E. Moser, C. Yi, M. K. Nazeeruddin and M. Grätzel, Nat. Commun., 2012, 3, 631.

14

U. Bach, Y. Tachibana, J.-E. Moser, S. A. Haque, J. R. Durrant, M. Graetzel and D. R. Klug, J. Am. Chem. Soc., 1999, 121, 7445–7446.

15

L. Yang, J. Zhang, Y. Shen, B.-W. Park, D. Bi, L. Haggman, E. M. J. Johansson, G. Boschloo, A. Hagfeldt, N. Vlachopoulos, A. Snedden, L. Kloo, A. Jarboui, A. Chams, C. Perruchot and M. Jouini, J. Phys. Chem. Lett., 2013, 4, 4026–4031.

16

J. Zhang, L. Yang, Y. Shen, B. W. Park, Y. Hao, E. M. J. Johansson, G. Boschloo, L. Kloo, E. Gabrielsson, L. Sun, A. Jarboui, C. Perruchot, M. Jouini, N. Vlachopoulos and A. Hagfeldt, J. Phys. Chem. C, 2014, 118, 16591–16601.

17

W. H. Nguyen, C. D. Bailie, E. L. Unger and M. D. McGehee, J. Am. Chem. Soc., 2014, 136, 10996–1001.

18

J. Burschka, A. Dualeh, F. Kessler, E. Baranoff, N.-L. Cevey-Ha, C. Yi, M. K. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2011, 133, 18042–5.

19

B. Xu, E. Gabrielsson, M. Safdari, M. Cheng, Y. Hua, H. Tian, J. M. Gardner, L. Kloo and L. Sun, Adv. Energy Mater., 2015, 5, 1402340.

20

N. Vlachopoulos, J. Zhang and A. Hagfeldt, Chim. Int. J. Chem., 2015, 69, 41–51.

21

H. J. Snaith, R. Humphry-Baker, P. Chen, I. Cesar, S. M. Zakeeruddin and M. Grätzel, Nanotechnology, 2008, 19, 424003.

22

P. Docampo, A. Hey, S. Guldin, R. Gunning, U. Steiner and H. J. Snaith, Adv. Funct. Mater.,

2012, 22, 5010–5019. 23

Y.-D. Lin and T. J. Chow, J. Photochem. Photobiol. A Chem., 2012, 230, 47–54.

24

Y. J. Chang and T. J. Chow, J. Mater. Chem., 2011, 21, 9523.

25

B.-S. Chen, D.-Y. Chen, C.-L. Chen, C.-W. Hsu, H.-C. Hsu, K.-L. Wu, S.-H. Liu, P.-T. Chou and Y. Chi, J. Mater. Chem., 2011, 21, 1937.

26

W.-K. Huang, H.-P. Wu, P.-L. Lin, Y.-P. Lee and E. W.-G. Diau, J. Phys. Chem. Lett., 2012, 3, 1830–1835.

27

A. C. Stuart, J. R. Tumbleston, H. Zhou, W. Li, S. Liu, H. Ade and W. You, J. Am. Chem. Soc., 2013, 135, 1806–1815.

28

H. Hsu, C. Cheng, W. Huang, Y. Lee and E. W. Diau, J. Phys. Chem. C, 2014, 118, 16904–16911.

29

J. W. Jung, J. W. Jo, C.-C. Chueh, F. Liu, W. H. Jo, T. P. Russell and A. K.-Y. Jen, Adv. Mater., 2015, 27, 3310–3317.

30

D.-Y. Chen, Y.-Y. Hsu, H.-C. Hsu, B.-S. Chen, Y.-T. Lee, H. Fu, M.-W. Chung, S.-H. Liu, H.-C. Chen, Y. Chi and P.-T. Chou, Chem. Commun. (Camb)., 2010, 46, 5256–5258.

31

M. Liang and J. Chen, Chem. Soc. Rev., 2013, 42, 3453–3488.

32

D. Joly, L. Pellejà, S. Narbey, F. Oswald, T. Meyer, Y. Kervella, P. Maldivi, J. N. Clifford, E. Palomares and R. Demadrille, Energy Environ. Sci., 2015, 8, 2010–2018.

33

S. Ito, M. K. Nazeeruddin, P. Liska, P. Comte, R. Charvet, P. Péchy, M. Jirousek, A. Kay, S. M. Zakeeruddin and M. Grätzel, Prog. Photovoltaics Res. Appl., 2006, 14, 589–601.

34

E. Gabrielsson, H. Ellis, S. Feldt, H. Tian, G. Boschloo, A. Hagfeldt and L. Sun, Adv. Energy Mater., 2013, 3, 1647–1656.

35

D. P. Hagberg, T. Edvinsson, T. Marinado, G. Boschloo, A. Hagfeldt and L. Sun, Chem. Commun., 2006, 2245–2247.

36

L. Ducasse, F. Castet, R. Méreau, S. Nénon, J. Idé, T. Toupance and C. Olivier, Chem. Phys. Lett., 2013, 556, 151–157.

37

D. Joly, L. Pellejà, S. Narbey, F. Oswald, J. Chiron, J. N. Clifford, E. Palomares and R. Demadrille, Sci. Rep., 2014, 4, 4033.

38

M. Pazoki, P. W. Lohse, N. Taghavinia, A. Hagfeldt and G. Boschloo, Phys. Chem. Chem. Phys., 2014, 16, 8503–8.

39

V. Leandri, R. Ruffo, V. Trifiletti and A. Abbotto, European J. Org. Chem., 2013, 2013, 6793– 6801.

40

J.-H. Yum, T. W. Holcombe, Y. Kim, K. Rakstys, T. Moehl, J. Teuscher, J. H. Delcamp, M. K. Nazeeruddin and M. Grätzel, Sci. Rep., 2013, 3, 2446.

41

M. Pastore and F. De Angelis, ACS Nano, 2010, 4, 556–62.

42

Y. Hao, X. Yang, J. Cong, X. Jiang, A. Hagfeldt and L. Sun, RSC Adv., 2012, 2, 6011.