lable at ScienceDirect

Tetrahedron 72 (2016) 8387e8392

Contents lists avai

Tetrahedron

journal homepage: www.elsevier .com/locate/ tet

Design of high-performance dye-sensitized solar cells by variation ofthe dihedral angles of dyes

Kwang-Myeong Kim, Jong-In Hong*

Department of Chemistry, College of Natural Sciences, Seoul National University, Seoul 151-747, South Korea

a r t i c l e i n f o

Article history:Received 3 August 2016Received in revised form20 October 2016Accepted 23 October 2016Available online 4 November 2016

Keywords:Dye-sensitized solar cellDihedral angleAlkyl chain locationShort-circuit photocurrentOpen-circuit photovoltage

* Corresponding author.E-mail address: jihong@snu.ac.kr (J.-I. Hong).

http://dx.doi.org/10.1016/j.tet.2016.10.0570040-4020/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

Two new metal-free organic sensitizers (i.e., IAH and IDH) based on a D�p�A configuration weredesigned to elucidate the effect of the alkyl chain location at the thiophene p-bridge on their photo-voltaic performance. IAH, in which the alkyl chain is closer to the acceptor moiety, is more planar thanIDH, in which the alkyl chain is closer to the donor moiety. Therefore, IAH induces more effective p-conjugation and electron injection from the dye excited state to the TiO2 conduction band, leading to ahigher short-circuit photocurrent (Jsc). In contrast, IDH more effectively inhibits electron transfer fromthe conduction band of TiO2 to the dye excited state, resulting in a higher open-circuit photovoltage (Voc).Under standard AM 1.5 simulated sunlight, the IAH-based cells exhibited an overall conversion efficiencyof 6.90%; this value is higher than that of the IDH-sensitized cells, which reached 86% of the conversionefficiency of N719 cells.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Dye-sensitized solar cells (DSSCs) are intensively investigatedbecause of their cost-effectiveness and ability for application inflexible devices.1,2 In 2005, DSSCs employing ruthenium complexesas sensitizers achieved a power conversion efficiency (h) greaterthan 12% under a standard global air mass of 1.5.3 Meanwhile,metal-free organic dyes, which usually feature an electrondonorep bridgeeelectron acceptor (DepeA) configuration, arewidely used because of their low material costs, easy syntheticroutes, high molar extinction coefficients, and environmentalfriendliness, especially compared to ruthenium-based dyes.4e7

Over the past decade, many kinds of organic dyes with DepeAconfigurations have been explored for DSSCs.8e13 It is been wellknown that controlling the assembly morphology of sensitizers onthe TiO2 surface facilitates electron injection and suppresses chargerecombination.14,15 The introduction of alkyl chains to dye skele-tons based on thiophene or its derivatives has proven to be aneffective method for improving DSSC performance because theyprevent dye aggregation and reduce charge recombination be-tween the TiO2 semiconductor and electrolyte.16e20 However,depending on their location in the dye, alkyl chains may increase

the dihedral angle between the donor and p-bridge moieties; thiscould reduce the planarity of the molecular structure, impedeefficient intramolecular charge transfer, and influence electron in-jection to the TiO2 conduction band, which may reduce the short-circuit photocurrent (Jsc). Therefore, it is vital to determine theoptimal position of the alkyl chains to improve device performance.This study examines the effect of the alkyl chain location in twoindole-based DepeA dyes (i.e., IDH and IAH; Fig. 1) with n-hexylsubstitution at the p-bridge on DSSC performance. Compared toIDH, the JSC of IAH increased significantly, from 12.77 to16.16 mA cm�2; however, the open-circuit voltage (VOC) decreasedby 50 mV from 680 to 630 mV. The average efficiency of IAH-basedDSSCs is 6.90% with an I�/I3�-based electrolyte under AM 1.5simulated sunlight; in contrast, the average efficiency of IDH-basedDSSCs is only 6.27%. The structures of the dyes are shown in Fig. 1.

2. Results and discussion

2.1. Photophysical properties

The UVevis absorption spectra of IAH and IDH in chloroform atroom temperature are shown in Fig. 2a, and the corresponding dataare summarized in Table 1. Compared to that of IDH, the spectrumof IAH shows a red-shifted absorption maximum peak (lmax) andhigher corresponding molar absorption coefficients (ε); this wasattributed to the differences of the dihedral angles between the

Fig. 1. Structures of IDH and IAH.

Fig. 2. Absorption spectra of IDH and IAH (a) in chloroform solution (5.0 � 10�5 M)and (b) on TiO2 film (5 mm).

Fig. 3. Dihedral angles of IDH (a) and IAH (b).

K.-M. Kim, J.-I. Hong / Tetrahedron 72 (2016) 8387e83928388

donor and p-bridge moieties (see Fig. 3).The steric hindrance between an electron donor part and p-

bridge depends on the location of alkyl chains. The dihedral angleof IDH, where the alkyl chain is closer to the donor part, is 52�

which is much bigger than that of IAH (25�). The more twistedmolecular structure of IDH leads to a lower absorption maximumand molar absorption coefficients when compared to IAH.

Table 1The absorption spectral data of two dyes.

lmax (nm)a lmax (nm)b The

IAH 480 436 44IDH 465 423 42

a In chloroform solution (5.0 � 10�5 M).b On TiO2 film (5 mm).

The absorption peaks of the two dyes adsorbed on TiO2 exhibitsimilar blue-shifts (Fig. 2b) as those evident in the correspondingsolution spectra. This phenomenon was ascribed to deprotonationof the carboxylic acid groups and intermolecular aggregation.21

2.2. Electrochemical properties

The highest occupied molecular orbital (HOMO) and lowestunoccupied molecular orbital (LUMO) energy levels of dyes play acrucial role in electron injection to the conductive band (Ecb) of TiO2and dye regeneration during device operation.22 The electro-chemical properties of IAH and IDH were studied using CV in DCMsolutions containing 0.1M TBAPF6 as the supporting electrolyte andcalibrated with respect to ferrocene (Fig. 4). The LUMO levels of IDHand IAH were estimated to be �2.84 and �3.03 eV, respectively;these values are much higher than the energy level of the con-duction band of TiO2 (�4.26 eV), which ensures efficient chargeinjection from the dye excited state to the TiO2 conduction band.The HOMO levels of IDH and IAH are �5.01 and �5.19 eV, respec-tively, and are lower than the energy level of the I�/I3� redox couplein the electrolyte (�4.92 eV); thus, they are suitable to complete thecircuit of the device and regenerate the dyes.

To further elucidate the photophysical properties of the twodyes, DFT calculations were performed at the B3LYP/6-31G level;the results are shown in Fig. 5. The HOMOs are delocalizedthroughout the entire p-conjugated system, while the LUMOs arelocalized on the cyanoacrylic acid and thiophene groups. Fromthese results, we assume that intramolecular charge transfer be-tween the donor and acceptor would be favorable, allowing effi-cient electron injection from the dye to the conduction band of theTiO2 semiconductor.

2.3. Photovoltaic performance of the DSSCs

To ensure a valid comparison, all the dyes were used in thefabrication of DSSCs and tested under the same conditions. The J-Vcurves of the DSSCs are illustrated in Fig. 6. IAH with E1 (DMPII0.6 M, LiI 0.1 M, I2 0.03 M, TBP 0.5 M, and acetonitrile only) as thesensitizer produced a Jsc of 16.16 mA cm�2, Voc of 0.68 V, fill factor(FF) of 0.68, and performance up to 6.90%, which is comparable tothat of N719 (8.02%). Under the same conditions, the IDH-basedDSSC offered a Jsc of 12.77 mA cm�2, Voc of 0.72 V, FF of 0.72, andperformance up to 6.27%. The conversion efficiency (h) of IAH washigher than that of IDH by ~10% because of the larger Jsc, eventhough the Voc was smaller (see Table 2).

The broader range of light absorption for IAH compared to thatof IDH (Fig. 2b) may contribute to the higher Jsc, as confirmed by theIPCE spectra (Fig. 7). The IPCE curve of the IAH dye revealed a broadresponse from 340 to 740 nmwith a peak value of 66% at ~440 nm.Compared with IAH, IDH presented a slightly lower maximum IPCEvalue (64% at 440 nm), and the response band was much narrower.Therefore, the slightly lower Jsc of IDHwas reasonable. In the case ofN719, although it showed the broadest spectrum, the maximumvalues were relatively low, which led to a lower Jsc (15.12 mA cm�2)than that of IAH (16.16 mA cm�2).

There were no significant differences in the dye loading

degree of blue shift (nm) ε, M�1 cm�1 E0-0

67,900 2.1657,980 2.17

Fig. 4. Cyclic voltammograms of IDH (a), IAH (b) and HOMO-LUMO energy levels ofdyes (c).

Fig. 5. Surface plots for the HOMOs and LUMOs of IAH and IDH.

Fig. 6. J-V curves of DSSCs based on IDH, IAH, and N719.

Table 2Performance of DSSCs based on IDH, IAH, and N719.

Jsc (mA/cm2) Voc (V) Fill

IDH 12.77 0.68 0.7IAH 16.16 0.63 0.6N719 15.12 0.76 0.7

Cell condition: 12 mm active layer, 10 mm scattering layer.E1: DMPII 0.6 M, LiI 0.1 M, I2 0.03 M, TBP 0.5 M, acetonitrile only.

Fig. 7. IPCE spectra of IDH, IAH, and N719 cells.

Fig. 8. Characteristics of IDH and IAH

K.-M. Kim, J.-I. Hong / Tetrahedron 72 (2016) 8387e8392 8389

(1.07 � 10�8 and 1.04 � 10�8 mol cm�2 for IDH and IAH, respec-tively) (Table 1). The only factor that could influence the DSSCperformance would be the dihedral angle between the donor andp-bridge moieties. Therefore, the smaller dihedral angle of IAHenables more efficient photoinduced electron transfer from the dyeto the TiO2 conduction band than the larger dihedral angle in IDH;this would lead to a higher Jsc value (Fig. 8). On the other hand, thelarger dihedral angle in IDH could impede back electron transferfrom TiO2 to the dye excited state (Fig. 8). In addition, it couldpromote the adsorption of IDH over a larger area of the TiO2 surface(Fig. 9); this would impede electron recombination between theTiO2 surface and electrolyte, thereby lifting the Fermi level of TiO2and increasing the Voc values (Fig. 8). To support our reasoning, weperformed electrochemical impedance analysis.

2.4. Electrochemical impedance analysis

EIS was performed in the dark to elucidate the origin of thehigher Voc values of IDH-based DSSCs. EIS is associated with thecharge recombination rate at the TiO2/electrolyte interface. The

Factor Efficiency (%) Dye loading (mol/cm2)

2 6.27 1.07 � 10�8

8 6.90 1.04 � 10�8

0 8.02

Fig. 9. Schematics of (a) IAH and (b) IDH adsorbed on TiO2 surface.

Fig. 10. Nyquist plots of IDH and IAH.

Scheme 1. Synthetic routes to IDH and IAH.

K.-M. Kim, J.-I. Hong / Tetrahedron 72 (2016) 8387e83928390

semicircle of IDH is bigger than that of IAH (Fig. 10); this resultsuggests that the twisted IDH structure could more effectivelysuppress electron recombination, leading to more electrons in theTiO2 conduction band and resulting in the higher Voc value.

3. Conclusion

We designed and synthesized two new D�p�A type metal-freeorganic sensitizers (i.e., IAH and IDH) with different alkyl chainlocations to investigate the effect of the dihedral angle between thedonor and thiophene p-bridge moieties on DSSC performance.Compared to IDH, which has the alkyl chain closer to the donormoiety, IAH, which has the alkyl chain closer to the acceptor moi-ety, has a much broader absorption spectrum on TiO2 film and thushigher IPCE values; this was attributed to its increased planarity. Asa result, IAH featured more effective p-conjugation, therebyinducing better electron injection from the dye excited state to theTiO2 conduction band, which resulted in a higher Jsc value. Conse-quently, the IAH-based cells exhibited an overall conversion effi-ciency of 6.90%, which reached 86% of that of N719 cells under thesame conditions. Our study revealed that the location of the alkylchain in D�p�A configuration organic sensitizers is an importantfactor to consider for tuning their photovoltaic parameters.

4. Experimental

4.1. Synthesis

3-hexyl-2-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)thio-phene and 4-hexyl-2-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)thiophene were purchased from TCI (see Scheme 1).

4.1.1. Synthesis of 7-bromo-4-p-tolyl-1,2,3,3a,4,8b-hexahydrocyclopenta[b] indole (1)

Compound 1 was synthesized according to a literatureprocedure.14

Yield: 65%. 1H NMR (300 MHz, CDCl3): d 7.17 (d, 6 Hz, 5H), 6.98(d, 9 Hz, 1H), 6.75 (d, 9 Hz, 1H), 4.76 (t, 6 Hz, 1H), 3.81 (d, 3 Hz, 1H),2.35 (s, 3H), 2.03 (d, 3 Hz, 2H), 1.89e1.84 (m, 2H), 1.82e1.79 (m, 2H),1.67e1.65 (m, 2H).

4.1.2. Synthesis of 7-(3-hexylthiophen-2-yl)-4-p-tolyl-1,2,3,3a,4,8b-hexahydro cyclopenta[b]indole (2)

A mixture of compound 1 (150 mg, 0.43 mmol), 3-hexyl-2-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)thiophene (156 mg,

K.-M. Kim, J.-I. Hong / Tetrahedron 72 (2016) 8387e8392 8391

0.52 mmol), and Pd(PPh3)4 (23 mg, 0.02 mmol) were dissolved in30 mL of THF followed by addition of 2 M aqueous K2CO3 solution(0.8 mL). The reaction mixture was refluxed under nitrogen at-mosphere for 12 h. After cooling to room temperature, the mixturewas extracted with chloroform, washed with water, and dried overanhydrous MgSO4. The crude product was purified by silica gelcolumn chromatography (1:2 hexane/CH2Cl2). Yield: 60%. 1H NMR(300MHz, CDCl3): d 7.35 (s,1H), 7.18 (s, 4H), 7.08 (d,1.8 Hz, 2H), 6.98(d, 6 Hz, 1H), 6.75 (d, 9 Hz, 1H), 4.76 (t, 6 Hz, 1H), 3.80 (t, 3 Hz, 1H),2.35 (s, 6H), 2.04e2.02 (m, 2H), 1.89e1.84 (m, 4H), 1.82e1.81 (m,2H), 1.79e1.55 (m, 3H), 1.45 (s, 3H), 1.27 (s, 1H). 13C NMR (75 MHz,CDCl3): d 129.824, 129.538, 129.474, 129.418, 128.595, 128.509,128.349, 128.121, 127.307, 125.814, 123.694, 122.297, 119.921,107.298, 69.157, 45.501, 33.899, 31.770, 31.181, 31.091, 29.252,28.887, 28.849, 28.714, 24.581, 22.734, 20.866, 14.212. HRMS (ESI):m/z observed 415.2305 (calculated for C28H-33NS [M]þ 415.2334).

4.1.3. Synthesis of 4-hexyl-5-(4-p-tolyl-1,2,3,3a,4,8b-hexahydrocyclopenta[b]indol-7-yl)thiophene-2-carbaldehyde (3)

Compound 2 (710 mg, 1.7 mmol) was dissolved in 20 mL of 1,2-dichloroethane and stirred under nitrogen gas. DMF (0.17 mL,2.2 mmol) and POCl3 (0.210 mL, 2.2 mmol) were added slowly tothis solution, and themixturewas refluxed for 24 h. After cooling toroom temperature, saturated aqueous NaOAc solution (10 mL) wasadded to themixture, whichwas then vigorously stirred for 1 h. Themixture was extracted with CH2Cl2, washed with water, and driedover anhydrous MgSO4. The crude product was purified by silicacolumn chromatography (1:2 hexane/CH2Cl2). Yield: 65%. 1H NMR(300 MHz, CDCl3): d 9.83 (s, 1H), 7.64 (s, 1H), 7.28e7.17 (m, 5H), 7.14(s, 1H), 6.92 (d, 6 Hz, 1H), 4.87 (t, 6 Hz, 1H), 3.87 (t, 3 Hz, 1H), 2.71 (t,615 Hz, 2H), 2.09 (s, 3H), 1.92 (s, 1H), 1.85e1.64 (m, 8H), 1.32e0.88(m, 8H). 13C NMR (75 MHz, CDCl3): d 187.412, 151.707, 144.203,140.5640, 136.194, 132.476, 131.148, 130.110, 128.928, 124.096,119.946, 118.956, 110.305, 74.312, 74.098, 46.834, 45.057, 35.98,35.289, 34.326, 34.042, 29.721, 24.217, 23.918, 21.039, 20.778.HRMS (ESI): m/z observed 443.2278 (calculated for C29H33NOS[M]þ 443.2283).

4.1.4. Synthesis of (Z)-2-cyano-3-(4-hexyl-5-(4-p-tolyl-1,2,3,3a,4,8b-hexahydrocyclopenta[b]indol-7-yl)thiophen-2-yl)acrylic acid (IDH)

Amixture of compound 3 (100mg, 0.23 mmol), cyanoacetic acid(190 mg, 2.3 mmol), and ammonium acetate (20 mg, 0.25 mmol)were dissolved in a mixture of CH2Cl2 (30 mL) and acetonitrile(10 mL), which was then refluxed for 48 h. The solvent was evap-orated under vacuum and 1 N HCl aqueous solutionwas added. Themixture was then extracted with CH2Cl2, washed with water, anddried over anhydrous MgSO4. The crude product was purified bysilica column chromatography (100:10:1 CH2Cl2/MeOH/aceticacid). Yield: 70%. 1H NMR (300 MHz, DMSO-d6): d 8.07 (s, 1H), 7.63(s, 1H), 7.21 (s, 3H), 7.20 (s, 1H), 7.14 (d, 9 Hz, 1H), 6.89 (d, 9 Hz, 1H),4.90 (s, 1H), 3.86 (t, 3 Hz, 1H), 2.63 (t, 3 Hz, 3H), 2.29 (s, 4H),2.11e2.05 (m, 2H),1.91 (s, 2H),1.79e1.76 (m, 4H),1.57e1.42 (m, 4H),1.10 (s, 2H), 0.83 (s, 1H). 13C NMR (75 MHz, DMSO-d6): d 172.358,164.403, 148.551, 139.566, 138.815, 132.217, 131.799, 130.141,130.061,128.866,125.235,122.705,120.521,107.158, 58.948, 44.965,40.834, 40.556, 40.278, 40.000, 39.722, 39.444, 39.166, 35.380,33.475, 31.445, 28.945, 28.457, 24.406, 22.513, 21.406, 21.324,20.877, 14.288. HRMS (ESI): m/z observed 510.2349 (calculated forC32H34N2O2S [M]þ 510.2341).

4.1.5. Synthesis of 7-(4-hexylthiophen-2-yl)-4-p-tolyl-1,2,3,3a,4,8b-hexahydrocyclopenta[b]indole (4)

Compound 4 was synthesized similarly to compound 2.Yield: 60%. 1H NMR (300MHz, CDCl3): d 7.51 (d, 9 Hz, 2H), 7.16 (s,

1H), 7.13e7.09 (m, 5H), 6.98 (d, 9 Hz, 1H), 4.66 (t, 3 Hz, 1H), 3.91 (d,3 Hz, 1H), 2.37e2.36 (s, 3 Hz, 6H), 2.00 (t, 3 Hz, 3H), 1.86e1.80 (m,4H), 1.71e1.61 (m, 3H), 1.45 (s, 5H). 13C NMR (75 MHz, CDCl3):d 144.030, 129.905, 129.827, 129.778, 129.350, 128.357, 127.621,125.109, 124.864, 124.773, 124.720, 120.116, 119.772, 118.794,108.840, 69.356, 45.409, 35.054, 33.882, 31.816, 30.649, 30.480,29.852, 29.146, 24.567, 22.752, 20.884, 14.240. HRMS (ESI): m/zobserved 415.2308 (calculated for C28H33NS [M]þ 415.2334).

4.1.6. Synthesis of 3-hexyl-5-(4-p-tolyl-1,2,3,3a,4, 8b-hexahydrocyclopenta[b]indol-7-yl)thiophene-2-carbaldehyde (5)

Compound 5 was synthesized similarly to compound 3.Yield: 65%. 1H NMR (300 MHz, CDCl3): d 9.97 (s, 1H), 7.68 (d,

6 Hz, 2H), 7.46e7.36 (m, 4H), 7.24 (s, 1H), 7.08 (d, 9 Hz, 1H), 6.86 (d,9 Hz, 2H), 4.84 (t, 6 Hz, 1H), 3.86 (d, 6 Hz, 1H), 3.00e2.92 (m, 3H),2.37 (s, 1H), 1.93 (d, 3 Hz, 2H), 1.75 (d, 6 Hz, 3H), 1.73e1.68 (m, 7H),0.95e0.91 (m, 4H). 13C NMR (75 MHz, CDCl3): d 182.661, 139.077,138.854, 138.512, 135.429, 132.107, 129.872, 129.197, 128.777,128.618, 128.112, 125.377, 125.215, 124.461, 121.318, 120.581,107.040, 69.308, 45.280, 35.199, 33.663, 31.611, 30.748, 30.401,29.135, 28.891, 28.829, 28.545, 24.447, 22.6078, 20.845, 14.096.HRMS (ESI): m/z observed 443.2258 (calculated for C29H33NOS[M]þ 443.2283).

4.1.7. Synthesis of (Z)-2-cyano-3-(3-hexyl-5-(4-p-tolyl-1,2,3,3a,4,8b-hexahydrocyclopenta[b]indol-7-yl)thiophen-2-yl)acrylic acid (IAH)

IAH was synthesized similarly to IDH.Yield: 70%. 1H NMR (300 MHz, DMSO-d6): d8.16 (s, 1H), 7.57 (s,

1H), 7.41 (s, 1H), 7.21 (d, 9 Hz, 1H), 7.14 (s, 3H), 7.01 (d, 6 Hz, 1H),6.84 (d, 9 Hz, 1H), 4.93 (t, 6 Hz, 1H), 3.85 (t, 6 Hz, 1H), 3.17 (s, 1H),2.73 (t, 6 Hz, 3H), 2.31e2.29 (m, 4H), 2.09 (t, 6 Hz, 2H), 1.91 (s, 3H),1.82e1.78 (m, 3H), 1.73 (s, 4H), 1.24 (s, 2H). 13C NMR (75 MHz,DMSO-d6): d 172.476, 149.051, 139.455, 136.278, 132.031, 130.453,130.318, 129.551, 127.567, 126.516, 124.334, 122.940, 122.833,122.045, 120.772, 119.020, 107.341, 58.965, 44.965, 35.380, 33.475,31.445, 28.946, 28.457, 24.406, 22.513, 21.406, 21.324,20.877,14.288. HRMS (ESI): m/z observed 510.2347 (calculated forC32H34N2O2S [M]þ: 510.2341).

4.2. Preparation of solar cells

The anode films for the DSSCs were fabricated under standardconditions. Fluorine-doped tin oxide (FTO) glass substrates werewashed by consecutive sonication in dilutedMucasol liquid cleaner,water, and isopropyl alcohol (IPA) for 20 min each to remove dirtand any remnant contaminants.

A layer of 4 mm TiO2 paste (20 nm) was printed onto the FTOconducting glass, which was then heated at 150 �C for 10 min. Thisprocess was repeated twice to achieve a 12 mm-thick film. Theresulting surface was finally coated with a scattering layer (10 mm)of 200 nm TiO2 paste. The electrodes were gradually sintered to460 �C for 15 min to generate a three-dimensional TiO2 nano-particle network.

Subsequently, the films were soaked in a 10 mM TiCl4 aqueoussolution for 30 min at 70 �C, washed with deionized water, andfurther annealed at 450 �C for 10 min. After cooling to ~50 �C, theelectrodes were immersed into a 0.5 mM dye bath in chloroben-zene for 6 h under dark conditions at 55 �C. After the immersionprocess, the electrodes were rinsed with ethanol and chloroben-zene to remove any non-adsorbed dye and then dried in air.

Platinum counter electrodes were prepared by coating each FTOplate with a drop of H2PtCl6 solution and heating them to 450 �C.Two holes (0.10 and 0.06 mm in diameter) were drilled into thecounter electrodes for injection of the electrolyte. The dye-

K.-M. Kim, J.-I. Hong / Tetrahedron 72 (2016) 8387e83928392

adsorbed TiO2 electrode and Pt counter electrode were attachedusing a hot-melt polymer film at about 100 �C. The liquid electro-lyte consisting of 0.6 M 1,2-dimethyl-n-propylimidazolium iodide(DMPII), 0.1 M LiI, 0.03 M I2, and 0.5 M tert-butylpyridine (TBP) inacetonitrile was injected into the cell through the drilled holes.Finally, the holes were sealed to prevent leakage of the electrolyte.

4.3. General information

All solvents and reagents were purchased from TCI and Aldrichand used without further purification. All reactions were per-formed under argon or nitrogen atmosphere and assessed using DCKieselgel 60 F254 thin layer chromatography. 1H and 13C NMRspectra were recorded on an Advance 300 MHz Bruker spectrom-eter in CDCl3 and DMSO-d6, respectively. High resolution mass(HRMS) data were taken by Q-TOF 5600 Series (AB SCIEX), with ESIpositive mode. FT-IR data were recorded using a Nicolet 6700(Thermo Scientific). Ultravioletevisible (UVevis) spectra of thesedyes in solution were measured using a Beckman DU 650 spec-trophotometer. The theoretical energy levels and minimizedstructures were obtained by density functional theory (DFT) cal-culations at the B3LYP/6-31G* level using the Gaussian 09 package.Electrochemical redox potentials were measured by cyclic vol-tammetry (CV) using a CH instrument 660 ElectrochemicalAnalyzer (CH Instruments, Inc., Texas) with 0.1 M tetrabuty-lammonium hexafluorophosphate (TBAPF6) in a CH2Cl2 (DCM) so-lution at room temperature. CV data were recorded using a three-electrode cell with a glassy carbon (GC) working electrode, Ptwire counter electrode, and Ag/Agþ reference electrode on anelectrochemistry workstation at a scan rate of 0.1 V s�1. TBAPF6(0.1 M in dry acetonitrile) was used as the supporting electrolyte.The ferrocene/ferrocenium (Fc/Fcþ) redox couple served as thereference potential of the reference electrode to correct the elec-trochemical redox potentials of the dyes with respect to NHE. Thecurrent-voltage (J-V) characteristics of the DSSCs were measuredusing a 1000 W xenon light source, which was calibrated as aportable solar simulator (PEC-L01) using a KG5-filtered Si referencesolar cell. Photon-to-current conversion efficiency (IPCE) spectra ofthe DSSCs were measured using an IPCE measuring system (PV

Measurements) from 400 to 800 nm under simulated AM 1.5 Glight. Electrochemical impedance spectra (EIS) were recorded usinga solar simulator (Netport) with a potentiostat/galvanostat (Auto-lab, PGSTAT 302N) under dark conditions.

Acknowledgments

This research was supported by the National Research Founda-tion (No. 2013R1A1A2074468) funded by the MSIP. We thank Dr.Hongse Oh for his experimental help and advice.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.tet.2016.10.057.

References

1. O'Regan, Gr€atzel M. Nature. 1991;353:737.2. Gr€atzel M. Nature. 2001;414:338.3. Nazeeruddin MK, Angelis FD, Fantacci S, et al. J Am Chem Soc. 2005;127:16835.4. Gao P, Tsao HN, Gr€atzel M, Nazeeruddin MK. Org Lett. 2012;14:4330.5. Liang M, Chen J. Chem Soc Rev. 2013;42:3453.6. Kim B-G, Zhen C-G, Jeong EJ, Kieffer J, Kim J. Adv Funct Mater. 2012;22:1606.7. Wu Y, Zhu WH, Zakeeruddin SM, Gr€atzel M. ACS Appl Mater Interfaces. 2015;7:

9307.8. Chen B-S, Chen D-Y, Chen D-Y, et al. J Mater Chem. 2011;21:1937.9. Ren X, Jiang S, Cha M, Zhou G, Wang Z-S. Chem Mater. 2012;24:3493.

10. Hua Y, Chang S, Huang D, et al. Chem Mater. 2013;25:2146.11. Cao D, Peng J, Hong Y, Fang X, Wang L, Meier H. Org Lett. 2011;13:1610.12. Koumura N, Wang Z-S, Mori S, Miyashita M, Suzukiand E, Hara K. J Am Chem

Soc. 2006;128:14256.13. Horiuchi T, Miura H, Sumioka K, Uchida S. J Am Chem Soc. 2004;126:12218.14. Pei K, Wu YZ, Islam A, et al. ACS Appl Mater Interfaces. 2013;5:4986.15. Pei K, Wu Y, Islam A, et al. J Phys Chem C. 2014;118:16552.16. Feng Q, Zhang Q, Lu X, Wang H, Zhou G, Wang Z-S. ACS Appl Mater Interfaces.

2013;5:8982.17. Shahid M, Thomas M-W, Labram J, et al. Chem Sci. 2012;3:181.18. Yum J-H, Thomas WH, Kim YJ, et al. Chem Commun. 2012;48:10727.19. Jiang S, Lu X, Zhou G, Wang Z-S. Chem Commun. 2013;49:3899.20. Kakiagem K, Aoyama Y, Yano T, et al. Chem Commun. 2014;50:6379.21. Kitamura T, Ikeda M, Shigaki K, et al. Chem Mater. 2004;16:1806.22. Hagfeldt A, Gr€atzel M. Chem Rev. 1995;95:49.