Synthesis of multi-donor dyes and influence of molecular design on dye-sensitized solar cells

Abstract

Multi-donor incorporated organic dyes were designed and synthesized and their structure–property relationship was investigated in dye-sensitized solar cells (DSCs). Five N,N-disubstituted aniline donor groups along with carbazole and thiophene secondary donors were combined into the (D)_n–π–A design with cyanoacrylic acid as the acceptor. The dyes showed broad absorption bands with absorption maxima in the range of 470–485 nm and optical band gap around 1.94–2.37 eV. Our results indicate that incorporation of bulky groups and bent-type architecture helps to improve the performance of DSCs.

---

Introduction

Low-cost and environment friendly technologies that can efficiently harness the solar energy would provide an attractive solution to the challenges for sustainable development.^1,2 Dye-sensitized solar cells (DSCs) are still not economically viable as the commercially available photosensitizers based on inorganic materials such as ruthenium complexes are expensive and require complicated preparation methods.^2–5 Ruthenium complexes such as N3,^6,7 N719 (ref. 8) and black dye⁹ are the most commonly used sensitizers in DSCs due to their intense and broad absorption of visible light.¹⁰ In contrast to inorganic ruthenium based sensitizers, metal-free organic dyes offer several advantages,^8,11–13 such as high molar absorption coefficients, molecular tailoring of the electronic band gap via structural modification,¹⁴ and useful for fabricating flexible DSC devices.^15–21 Structural design of organic dyes plays a crucial role in performance of DSCs for solar energy conversion. In place of the standard N719 sensitizer, the use of oligoacene or oligothiophene core dyes with coumarin donor groups and cyanoacrylic acid acceptors led to high efficiencies.^22–27 Several studies have been conducted with triphenylamine and carbazole group as donors in organic sensitizers due to their strong electron donating and hole transporting nature.^28–38 Multi-donor configuration (D_n–A) has shown significant improvements over the traditional D–A based molecules.^39–42

Owing to enormous efforts by researchers to design better organic dyes with high solubility, easy accessibility, high open-circuit voltage and photon-to-electron conversion efficiencies, detailed investigation of the structure–property relationship of organic dyes is the need of the hour towards the establishment of ‘molecular design’ concept for DSCs. Here, we explore the role of geometry (i.e. linear vs. bent), multiple donor groups attached to a single acceptor (push–push–pull) and changes in electronic properties of the molecule towards the performance of DSCs.

Structural differences are expected to exert significant influence on the packing of dye molecules on TiO₂ surface and affect the performance of DSCs. In addition, linear vs. curved or branched molecular structures of the dyes lead to reduced aggregation, followed by potential enhancement in electron transport.^43–45

Results and discussion

N,N-Dihexylaniline (DHA, in TC1) or phenylpyrrolidine (PPY, in TC2) is attached to the middle thiophene (electron rich π-spacer) group acts as donor and cyanoacrylic acid as the acceptor and anchoring group (Fig. 1). In TC3, TC4 and TC5 dyes, functionalization of the central carbazole moiety play a key role towards the final geometry of the molecule. In TC3 and TC4, the carbazole was linked to other groups via 3,6-linkage, whereas in TC5, a 2,7-linkage was used. N,N-Bis(4-methoxyphenyl)aniline (MTPA) and N-(2-ethylhexyl)carbazole attached to the thiophene ring act as donor groups in TC3 and TC5. In TC4, substituted phenylpyrrolidine and N-(2-ethylhexyl)carbazole act as donors and a cyanoacrylic acid as the anchor as well as acceptor group in all five dye molecules. It is conceivable that changes in geometry of the dye molecule would influence the packing on the electrode and electronic conjugation along the molecule, which will affect the performance of DSCs (Fig. 1).

Fig. 1 Molecular structure of synthesized dyes, TC1–TC5.

Synthesis

The synthetic pathways used for preparing the dyes are illustrated in Scheme 1. The first two dyes, TC1 and TC2 were obtained in a two-step reaction scheme, in which 2-bromo-5-formylthiophene was coupled to boronic ester of donor groups (e.g. DHA and PPY) using Suzuki coupling reaction, followed by Knoevenagel condensation with cyanoacetic acid. Similarly, TC3, TC4 and TC5 were also synthesized via a two-step sequence using Suzuki coupling reaction followed by Knoevenagel condensation. The compound, 3b or 3c were synthesized in a one-pot reaction, which consisted of two consecutive Suzuki coupling reactions. 3,6-Dibromocarbazole was reacted with 0.7 equiv. of 1b or 1c and stirred for 8 h, followed by the addition of 2.0 equiv. of 5-formyl-thiophene-2-boronic acid to the reaction mixture (Scheme 1). The same procedure was also used for the synthesis of compound 4. These reactions resulted in disubstituted side products from the corresponding boronic acids and bromocompounds; however, the sequential couplings of boronic acids led to target compound as the major product. In the next step, appropriate aldehyde was reacted with cyanoacetic acid in the presence of ammonium acetate and acetic acid (i.e. Knoevenagel condensation) to produce the target molecules. All synthesized compounds were characterized by ¹H and ¹³C NMR, MS-EI, FT-IR (for final dyes), MALDI-TOF and elemental analyses.

Scheme 1 Synthesis of multi-donor dyes (TC1–TC5).

Photophysical properties

The absorption and normalized emission spectra of all five organic dyes in chloroform solution are shown in Fig. 2. TC3, TC4 and TC5 showed two absorption maxima in the region of 310–485 nm. The absorption maximum in the shorter wavelength region is due to π–π* electronic transition of the conjugated molecule and the peak in the longer wavelength region is due to intramolecular charge transfer between the donor and acceptor moieties. Similar absorption maxima were also observed for TC1 (471 nm) and TC3 (470 nm) (Table 1).

Fig. 2 Absorption (A) and emission spectra (B) of TC1–5 in chloroform at room temperature.

Table 1 Summary of photo physical and electrochemical properties of TC1–5 dyes

Dye	λabs^a/nm (ε/M^−1 cm^−1)	λem^a/nm	Eg^b/eV	Eoxi^c/V	EHOMO^d/eV	ELUMO^e/eV
a Recorded in chloroform.b Optical band gap.c vs. ferrocene in 0.1 M Bu4NPF6 in THF, platinum disc as working electrode with a scan rate of 100 mV s^−1.d Calculated using the relationship EHOMO = −(E^onsetoxi + 4.8).e Calculated using the relationship ELUMO = −(optical band gap − EHOMO).
TC1	471 (22 894), 449 (22 101)	609	2.08	0.35	−5.00	−2.92
TC2	411 (24 529)	533	2.17	0.49	−5.15	−2.98
TC3	470 (33 901), 336 (42 655)	567	2.12	0.14, 0.64	−4.85	−2.73
TC4	457 (22 238), 311 (38 425)	555	2.19	0.27, 0.75	−4.97	−2.78
TC5	472 (10 233), 355 (8373)	589	2.05	0.13, 0.60	−4.84	−2.79

---

However, TC3 showed a higher molar absorptivity (ε) of 33 901 M⁻¹ cm⁻¹ as compared to TC1, ε = 22 894 M⁻¹ cm⁻¹. In TC3, the donor ability of the carbazole is significantly enhanced by the incorporation of trimethoxyphenyl amine group and 3,6-linakge on carbazole that increases the conjugation for the electron to migrate towards the cyanoacrylic acid moiety. The carbazole linked through 3- and 6-positions extends conjugation and improves electron transport through the organic sensitizer. Consequently, the enhanced donor ability of TC3 is beneficial to light harvesting and improving efficiency of DSCs.

A similar dye, 5-(4-(bis(4-methoxyphenylamino)styryl)thiophene-2-yl)-2-cyanoacrylic acid was synthesized and studied by Hagberg et al.,³⁶ displayed an absorption maximum of 462 nm (ε = 33 000 M⁻¹ cm⁻¹). In TC3, the ethylene group was substituted with (2-ethylhexyl)-9H-carbazole, resulting in a red shift of the absorption maximum to 470 nm (ε = 33 901 M⁻¹ cm⁻¹).

Comparison of electron donating ability of N,N-dihexylaniline (TC1) with phenylpyrrolidine (TC2) groups, significant red shift (60 nm) in absorption maximum was observed for TC1. Similarly, λ_max of TC4 was blue shifted by 13 nm as compared to TC3 with a lower absorption coefficient (ε = 22 238 M⁻¹ cm⁻¹).

In TC5, the secondary donor and thiophene π-bridge were moved from 3,6-positions to less conjugated 2,7-positions of the carbazole group. This led to a linear molecular structure with a threefold decrease in the molar absorptivity of TC5, ε = 10 233 M⁻¹ cm⁻¹ compared to TC3. This implies that substitution at the 3,6-positions of the carbazole group is more favorable for electron migration as the substituents are located at para-positions to the electron rich nitrogen atom. TC3 showed lowest stokes shift of 97 nm and TC1 with highest value of 138 nm in the present series (Fig. 2B). The optical band gap was calculated using the intersecting point of the normalized absorption and emission curves of individual dyes (Table 1). The absorption spectra of dyes on TiO₂ (ESI, Fig. S2†) showed broad absorption bands which are in agreement with solution spectra and the loading is more or less uniform for all dyes.

Electrochemical properties

For efficient electron injection, the LUMO or excited state oxidation potential of the dye should be above the conduction band edge of TiO₂, and for efficient dye regeneration, the HOMO of the dye should lie below the energy level of iodine redox mediator (I⁻/I₃⁻) system.^31,40,46 To evaluate the thermodynamic basis of these electron transfer processes, cyclic voltammetry was used to identify the energy levels of the sensitizers. Cyclic voltammograms (CVs) were recorded at a constant scan rate of 100 mV s⁻¹ in THF solution (Fig. 3 and S1†). Dyes TC1 and TC2 showed the first oxidation potentials at 0.35 V and 0.49 V, respectively. The MTPA substituted dyes TC3 and TC5 showed significant two reversible oxidations and their HOMO energy levels were found at −4.85 eV. This first oxidation is assigned to MTPA moiety. In contrast, the PPY substituted dyes TC2, TC4 had the irreversible oxidation potentials (Fig. 3). However, TC3 and TC5 were found to be having similar HOMO energy levels as observed from UV-vis absorption spectra. As the dyes did not show the reduction potentials in the used voltage scan range with saturated calomel electrode, LUMO energy levels were calculated using the optical band gap and HOMO energy levels of individual dyes (Table 1).

Fig. 3 Cyclic voltammetry scans of dyes TC1, TC3 and TC5 at a scan rate of 100 mV s⁻¹ in THF solution. Potentials reported with respect to ferrocene.

Theoretical calculations

To further understand the photophysical and electrochemical properties, the Gaussian calculations were carried out on Gaussian 09 at the density functional theory (DFT) level with the B3LYP function using 6-31G* as a basis set. The HOMO/LUMO orbital distributions and the geometry optimized structures of carbazole-substituted and non-carbazole molecules were depicted in Fig. 4 and 5.

Fig. 4 Frontier orbital distribution of synthesized dye molecules were calculated by density functional theory at the B3LYP/6-31G(d) level.

Fig. 5 Geometry and energy optimized structure of dye molecules were calculated by density functional theory at the B3LYP/6-31G(d) level.

The calculated HOMO–LUMO levels and trends in energy gaps of the dye molecules were in good agreement with UV-vis absorption spectra of the dyes (Table 2). The frontier molecular orbital distributions of the dyes connected through the carbazole moiety were different from non-carbazole dyes. In case of TC1 and TC2, the HOMO levels were fully distributed on the whole molecule whereas the LUMO levels were localized in between thiophene and cyanoacrylic acid groups. The HOMO levels of dyes TC3, TC4 and TC5 were mostly localized on substituted amine moiety and partially on carbazole unit, which was observed in the electrochemical analysis. However, the LUMO levels were localized on thiophene, cyanoacrylic acid acceptor (Fig. 4).

Table 2 Theoretical calculation results from density functional theory at the B3LYP/6-31G(d) level

Dye	HOMO (eV)	LUMO (eV)	Eg (eV)
TC1	−5.83	−2.68	3.15
TC2	−5.26	−2.41	2.85
TC3	−4.82	−2.56	2.26
TC4	−4.82	−2.47	2.35
TC5	−4.63	−2.45	2.18

---

Photovoltaic performances of DSCs

The photovoltaic performances of the organic dyes were evaluated by constructing DSCs using nanocrystalline TiO₂ and the iodide/triiodide redox mediator system. The J–V and IPCE curves (Fig. 6 and S3†) of TC1 to TC5 based devices and the corresponding results are shown in Table 3. The overall power conversion efficiency (η) was calculated from the short-circuit current (J_SC), open-circuit voltage (V_OC) and fill factor (FF). The open-circuit photovoltage and overall power conversion efficiency of the five dyes are in the order of TC5 > TC3 > TC1 > TC2 > TC4 and TC3 > TC2 > TC5 > TC4 > TC1, respectively. TC3 showed a higher short-circuit current (J_SC = 9.03 mA cm⁻²) and overall power conversion efficiency (η = 4.08%) than TC4 (η = 2.37%).

Fig. 6 J–V characteristics of solar cells of TC1–5 dyes under AM 1.5 illumination. The electrodes were immersed into a 0.25 mM solution of sensitizer in tetrahydrofuran solution.

This is due to the addition of a bulky MTPA unit, which increases the donor ability of the carbazole group and reduces dye aggregation at the TiO₂ interface. Among the D–π–A dyes, TC2 showed significantly higher performance as compared to TC1 and this could be due to the presence of linear alkyl chains in TC1 dye which facilitate more aggregation on TiO₂ surface.

From the results (Table 3), the D–D–π–A (push–push–pull type) configuration appears to give better photovoltaic performance. As seen from the absorption spectra, the increased propensity for the migration of the electrons towards anchoring group and TiO₂ surface is significantly enhanced with increase in number of donor groups and molar absorptivity, hence improving the light harvesting properties and short-circuit current.

Table 3 DSC performances of TC1–TC5 under AM 1.5 illumination

Dye	VOC (V)	JSC (mA cm^−2)	FF (%)	η (%)
TC1	0.704	4.45	63.7	1.99
TC2	0.651	7.35	67.2	3.22
TC3	0.748	9.03	61.1	4.08
TC4	0.625	5.45	69.5	2.37
TC5	0.793	4.80	68.4	2.60

---

Although the V_OC obtained for TC5 (793 mV) was higher than that of TC3, the overall power conversion efficiency produced was 2.60%, significantly lower than the 4.08% obtained for TC3. This is due to the lower short-circuit current in the TC5 based device (J_SC of 4.80 mA cm⁻²), which implies that substitution at the 3,6-positions of the carbazole moiety is suitable for achieving a better photovoltaic efficiency. On the other hand, the 2,7-positions of the carbazole are meta-directed with respect to the electron rich nitrogen atom, hence reducing its electron donating properties. It is conceivable that molecular structures of TC3 and TC5 are significantly larger than the others and such dyes will block the TiO₂ surface, thus reducing recombination and, consequently, increasing V_OC. Owing to the better coverage of the surface, the diffusion of electron acceptors present in the electrolyte towards the TiO₂ is hindered by the bulky dyes on the surface.^47,48 As shown in Fig. 2A, TC3 has the highest extinction coefficient for a broad range of wavelengths, which can explain its high J_SC owing to better light harvesting abilities. The peak extinction coefficient of TC2 is lower than that of TC3, which explains its lower current compared with TC3. Although TC2 has similar extinction coefficient spectrum as compared to TC1 and TC4, the observed high J_SC for TC2 is due to its more negative HOMO energy (Table 1), which should facilitate faster dye regeneration, thus improving J_SC.^49,50

From these preliminary findings, further improvements to the organic sensitizer could be done via proper design of multi-donor and bulky groups on the dye molecule. This would incorporate both strong donor properties, extended conjugation and low degree of aggregation.

Experimental

Materials and instruments

All chemicals and reagents were purchased from commercial suppliers (Sigma Aldrich, and Merck) and used without further purification. Solvents used for spectroscopic measurements were spectral grade quality. Compounds, 3,6-dibromo-9-(2-ethylhexyl)-9H-carbazole,⁵¹ 2,7-dibromo-9-(2-ethylhexyl)-9H-carbazole⁵² and N,N-bis(4-methoxyphenyl)aniline-4-boronic acid pinacol ester (1c)⁵³ were synthesized according to reported literature procedure. Synthesis of compounds 1a and 1b was provided in the ESI.† All reactions were monitored by thin-layer chromatography (TLC) on silica gel plates. Preparative separations were performed by column chromatography on silica gel grade 60 (0.040–0.063 mm) from Merck.

¹H (300 MHz) and ¹³C (75 MHz) NMR spectra were recorded on a Bruker spectrometer. IR spectra were measured on a Varian Bio-Rad Excalibur FT-IR spectrophotometer using KBr as matrix. Mass spectra (electron ionization) were obtained on a Finnigan TSQ7000 and MALDI-TOF on Bruker Autoflex III mass spectrometer. The UV-vis spectra were measured on a Shimadzu UV-160lPC spectrophotometer. Photoluminescence spectra were recorded with a Shimadzu RF-5301 PC spectrofluorophotometer in chloroform. Cyclic voltammetry was recorded with a computer controlled CHI electrochemical analyzer at a constant scan rate of 100 mV s⁻¹. Measurements were performed in tetrabutylammonium hexafluorophosphate (0.1 M) solution prepared in degassed tetrahydrofuran (THF). The electrochemical cell consists of three electrode systems with platinum disc as the working electrode, platinum rod as counter electrode and Standard Calomel Electrode (SCE) as a reference electrode. The potentials are calibrated using ferrocene as external standard. The onset of oxidation (E^onset_ox) was used to calculate the HOMO energy level using the relationship E_HOMO = −(4.8 + E^onset_ox).⁵⁴ Geometry optimizations were performed in Gaussian 09 (ref. 55) at the density functional theory (DFT) level with the B3LYP function and a 6-31G* basis set. The HOMO and LUMO surfaces were generated from the optimized geometries using GaussView 5.⁵⁶

DSCs were fabricated using dyes TC1 to TC5 and TiO₂ coated ITO electrodes with a triple layer structure consisting of transparent ITO layer (thickness 12 μm), TiO₂ nanoparticle (20 nm) layer and a scattering layer with larger TiO₂ nanoparticles (400 nm). The electrodes were immersed into a 0.25 mM solution of sensitizer in tetrahydrofuran, left overnight, removed from solution, rinsed in tetrahydrofuran and dried under the nitrogen gas flow immediately before assembling the cell. Anodes and platinized counter electrodes were sealed together in a sandwich configuration and were filled with an electrolyte [0.6 M propylmethylimidazolium iodide, 0.03 M I₂, 0.1 M guanidinium thiocyanate, and 0.5 M 4-tert-butylpyridine in a mixture of acetonitrile and valeronitrile (85 : 15 volume ratio)] by vacuum back filling.⁴⁰

Synthesis and characterization

General synthetic precursors for Suzuki coupling reaction. Compounds 2a, 2b, 3b, 3c and 4 were synthesized from the appropriate bromo compound and boronic ester using Suzuki coupling reaction. Method described for 2a was also used for the synthesis of other compounds. In a typical procedure, compound 1a (0.387 g, 1 mmol) and 2-bromo-5-formyl thiophene (0.191 g, 1 mmol) were dissolved in THF (15 mL) under inert atmosphere. To the reaction mixture, 2 M K₂CO₃ (10 mL) and Pd(PPh₃)₄ (0.188 g, 0.16 mmol) were added and stirred at 80 °C for 12 hours. The reaction progress was monitored by TLC. The reaction mixture was extracted with dichloromethane, organic layer was washed with brine, dried over anhydrous Na₂SO₄ and solvent was removed under reduced pressure. The crude compound was purified using silica gel column chromatography with DCM–hexane (1 : 1) as eluent.
Synthesis of 2-[4-(N,N-dihexylamino)phenyl]-5-formyl-thiophene (2a). A bright yellow liquid (0.339 g, yield 91.1%). MS (EI) m/z: 371.3 (M⁺), 371.6 (calcd). ¹H NMR (300 MHz, CDCl₃, ppm): δ = 9.79 (–CHO), 7.64–7.65 (d, 1H, J = 3.93 Hz), 7.51–7.53 (d, 2H, J = 8.88 Hz), 7.19–7.20 (d, 1H, J = 4.11 Hz), 6.61–6.64 (d, 2H, J = 8.88 Hz), 3.27–3.32 (t, 4H), 1.59 (s, 4H), 1.33 (s, 12H), 0.92 (s, 6H). ¹³C NMR (75.4 MHz, CDCl₃, ppm): δ = 182.27 (–CHO), 156.32, 14.06, 139.65, 138.21, 127.66, 121.06, 119.68, 111.50, 53.47, 51.03, 31.71, 29.72, 27.23, 26.79, 22.70. Elem. anal. calcd for C₂₃H₃₃NOS: C, 74.34%; H, 8.95%; N, 3.77%; S, 8.63; found: C, 74.22%; H, 8.71%; N, 3.69%; S, 8.74%.
Synthesis of 2-[4-(pyrrolidin-1-yl)phenyl]-5-formyl thiophene (2b). Compound 1b (0.546 g, 2 mmol) and 2-bromo-5-formyl thiophene (0.31 g, 1.62 mmol) were reacted under Suzuki coupling conditions to yield a yellow solid (2b, 0.40 g, 95.7%). MS (EI) m/z: 257.1 (M⁺), 257.4 (calcd). ¹H NMR (300 MHz, CDCl₃, ppm): δ = 9.81 (s, 1H, –CHO), 7.67–7.68 (d, J = 3.39 Hz, 1H), 7.54–7.57 (d, J = 8.7 Hz, 2H), 7.22–7.23 (d, J = 4.11 Hz, 1H), 6.55–6.58 (d, J = 8.88 Hz, 2H), 3.32–3.34 (t, 4H), 2.02–2.06 (quintet, 4H). ¹³C NMR (75.4 MHz, CDCl₃, ppm): δ = 182.38 (–CHO), 156.55, 148.70, 139.79, 138.14, 127.62, 121.14, 120.12, 111.83, 47.60, 25.47. Elem. anal. calcd for C₁₅H₁₅NOS: C, 70.01%; H, 5.87%; N, 5.44%; S, 12.46; found: C, 70.15%; H, 5.89%; N, 5.29%; S, 12.68%.
Synthesis of 5-[6-(4-pyrrolidin-1-yl-phenyl)-9-(2-ethylhexyl)-9H-carbazol-3-yl]thiophene-2-carbaldehyde (3b). Compound 3b was synthesized by following a similar synthetic procedure used for 2a. 3,6-Dibromo-9-(2-ethyl-hexyl)-9H-carbazole (0.846 g, 2 mmol) and 1b (0.546 g, 2 mmol) were reacted under Suzuki reaction conditions for 8 hours. After which, 5-formyl-thiophene-2-boronic acid (0.35 g, 2.24 mmol) and Pd(PPh₃)₄ (0.047 g, 0.04 mmol) were added, stirred at 80 °C for another 12 hours, poured into water (100 mL) and the mixture was extracted with DCM (2 × 100 mL). The collected organic fractions were dried over anhydrous Na₂SO₄, filtered, concentrated under reduced pressure and the crude product was purified using column chromatography with 5% ethyl acetate in hexane as eluent to yield a bright yellow solid (3b, 0.033 g, 12%). MS (EI) m/z: 534.5 (M⁺), 534.8 (calcd). ¹H NMR (300 MHz, CDCl₃, ppm): δ = 9.89 (s, CHO, 1H), 8.43–8.44 (d, 1H), 8.28 (d, 1H), 7.75–7.79 (m, 2H), 7.69–7.73 (dd, 1H), 7.60–7.63 (d, J = 8.7 Hz, 2H), 7.46–7.47 (d, 1H), 7.41–7.44 (d, 1H), 7.39–7.42 (d, 1H), 6.69–6.72 (d, J = 8.55 Hz, 2H), 4.17–4.19 (d, 2H), 3.37 (t, 4H), 2.05 (t, 4H), 1.29–1.42 (m, 9H), 0.85–0.96 (m, 6H). ¹³C NMR (75.4 MHz, CDCl₃, ppm): δ = 182.59 (–CHO), 156.53, 147.02, 141.93, 140.20, 137.89, 133.64, 128.84, 127.92, 125.40, 124.30, 123.89, 123.61, 123.11, 122.79, 118.51, 117.83, 112.10, 109.56, 109.51, 47.74, 47.70, 39.48, 31.01, 29.69, 28.82, 14.02, 25.51, 24.40, 23.040, 10.90. Elem. anal. calcd for C₃₅H₃₈N₂OS: C, 78.61%; H, 7.16%; N, 5.24%; S, 6.00; found: C, 78.79%; H, 6.98%; N, 5.31%; S, 6.11%.
Synthesis of 5-[6-(4-[bis-(4-methoxy-phenyl)-amino]-phenyl)-9-(2-ethylhexyl)-9H-carbazol-3-yl]-thiophene-2-carbaldehyde (3c). The compound was synthesized by following similar synthetic procedure as 3b. 3,6-Dibromo-9-(2-ethylhexyl)-9H-carbazole (0.497 g, 1.14 mmol) and 1c (0.35 g, 0.81 mmol) were reacted under Suzuki reaction conditions for 8 hours. After which, 5-formyl-thiophene-2-boronic acid (0.35 g, 2.24 mmol) and Pd(PPh₃)₄ (0.469 g, 0.04 mmol) were added and the mixture was stirred at 80 °C for another 12 hours. The purified compound, 3c was isolated as orange oil (0.102 g, 18% yield). MS (EI) m/z: 692.5 (M⁺), 692.31 (calcd). ¹H NMR (300 MHz, CDCl₃, ppm): δ = 9.88 (s, 1H, –CHO), 8.42 (b, 1H, –CAR), 8.29 (b, 1H, –CAR), 7.69–7.79 (m, 3H), 7.54 (d, J = 8.7 Hz, 2H) 7.40–7.46 (m, 3H), 7.05–7.13 (m, 6H), 6.86 (d, J = 8.9 Hz, 4H), 4.19 (d, J = 6.9 Hz, 2H, –CH₂–), 3.82 (s, 6H, –OMe), 2.09 (m, 1H, –CH–), 1.30–1.42 (m, 8H, –CH₂–), 0.90–0.96 (m, 6H, –CH₃). ¹³C NMR (75.4 MHz, CDCl₃, ppm): δ = 182.5, 156.3, 155.7, 147.5, 141.9, 141.1, 141.0, 140.5, 137.8, 133.9, 132.8, 127.6, 126.3, 125.4, 124.4, 124.0, 123.5, 123.0, 122.8, 121.2, 118.4, 118.2, 114.6, 109.6, 109.5, 55.4, 39.4, 30.9, 29.6, 28.7, 24.3, 23.0, 13.9, 10.8. Elem. anal. calcd for C₄₅H₄₄N₂O₃S: C, 78.00%; H, 6.40%; N, 4.04%; S, 4.63; found: C, 78.19%; H, 6.09%; N, 4.12%; S, 4.53%.
Synthesis of 5-[7-(4-[bis-(4-methoxy-phenyl)-amino]-phenyl)-9-(2-ethylhexyl)-9H-carbazol-2-yl]-thiophene-2-carbaldehyde (4). 2,7-Dibromo-9-(2-ethylhexyl)-9H-carbazole (0.5 g, 1.14 mmol) and 1c (0.345 g, 0.80 mmol) were reacted under Suzuki reaction conditions for 8 hours. After which, 5-formyl-thiophene-2-boronic acid (0.35 g, 2.24 mmol) and Pd(PPh₃)₄ (0.07 g, 0.06 mmol) were added and the mixture was stirred at 80 °C for another 12 hours. The crude product was purified using column chromatography and compound 4 was isolated as orange oil (0.114 g, 14% yield). MS (EI) m/z: 692.4 (M⁺), 692.31 (calcd). ¹H NMR (300 MHz, DMSO-d6, ppm): δ = 9.91 (s, 1H, –CHO), 8.09 (d, J = 8.1 Hz, 2H), 7.78 (d, J = 3.9 Hz, 1H), 7.65 (b, 1H), 7.45–7.57 (m, 6H), 7.03–7.14 (m, 6H), 6.88 (d, J = 8.7 Hz, 4H), 4.22 (d, J = 6.9 Hz, –CH₂), 3.82 (s, 6H, –OMe), 2.10–2.14 (t, 1H, –CH), 1.26–1.43 (m, 8H, –CH₂), 0.85–0.98 (m, 6H, –CH₃). ¹³C NMR (75 MHz, DMSO-d6, ppm): δ = 182.6, 167.7, 155.9, 155.9, 148.2, 148.0, 142.5, 141.9, 141.5, 140.8, 139.6, 137.5, 133.7, 130.0, 127.8, 127.7, 126.6, 123.9, 123.7, 120.9, 120.7, 118.6, 117.6, 114.7, 106.7, 55.4, 47.3, 39.4, 30.9, 28.7, 24.4, 23.0, 13.9, 10.9. Elem. anal. calcd for C₄₅H₄₄N₂O₃S: C, 78.00%; H, 6.40%; N, 4.04%; S, 4.63; found: C, 78.13%; H, 6.31%; N, 4.16%; S, 4.71%.

Compounds TC1 to TC5 were synthesized using similar procedures and the synthesis of TC1 was described in detail.

Synthesis of 3-(5-[4-(N,N-dihexylamino)phenyl]thiophene-2-yl)-2-cyano-acrylic acid (TC1). Compound 2a (0.268 g, 0.72 mmol) was dissolved in 8 mL glacial acetic acid. To this, cyanoacetic acid (0.122 g, 1.44 mmol) and ammonium acetate (0.111 g, 1.44 mmol) were added. The mixture was stirred and refluxed for 8 h. After cooling to room temperature, the mixture was poured into cold water (20 mL), filtered and washed with water (2 × 20 mL) followed by methanol (2 × 20 mL). The crude product was purified by column chromatography using DCM and methanol (8 : 2) mixture as eluent to give a red solid (0.264 g, yield 83.4%). MS (MALDI-TOF) m/z: 439.3137 (M⁺), 438.6 (calcd). ¹H NMR (300 MHz, DMSO-d6, ppm): δ = 8.09 (s, 1H, –CHO), 7.60–7.63 (dd, 1H), 7.42–7.50 (2 d, 2H), 7.32–7.35 (dd, 1H), 6.58–6.67 (2 d, 2H), 3.26–3.30 (t, 4H), 1.51 (broad, 4H), 1.28 (s, 12H), 0.85–0.86 (2 s, 6H). ¹³C NMR (75.4 MHz, DMSO-d6, ppm): δ = 164.67, 150.19, 150.23, 148.61, 141.36, 136.97, 133.72, 127.42, 121.54, 120.35, 119.78, 112.37, 111.93, 50.46, 42.90, 31.51, 28.94, 27.16, 26.44, 22.49, 14.23. FT-IR (KBr): 3422 (OH stretch), 2926, 2855 (sp³ CH stretch), 2210 (C≡N stretch), 1608 (C=O stretch). Elem. anal. calcd for C₂₆H₃₄N₂O₂S: C, 71.19%; H, 7.81%; N, 6.39%; S, 7.31; found: C, 71.24%; H, 7.49%; N, 6.19%; S, 7.51%.
Synthesis of 3-(5-[4-(pyrrolidin-1-yl)phenyl]thiophene-2-yl)-2-cyano-acrylic acid (TC2). Compound 2b (0.192 g, 0.75 mmol) and cyanoacetic acid (0.132 g, 1.55 mmol) were reacted according to general Knoevenagel condensation procedure to yield product (TC2), red solid (0.145 g, yield 60%). MS (MALDI-TOF) m/z: 322.7128 (M⁺), 324.4 (calcd). ¹H NMR (300 MHz, DMSO-d6, ppm): δ = 7.94 (s, 1H), 7.55–7.57 (d, J = 3.63 Hz, 1H), 7.51–7.54 (d, J = 8.7 Hz, 2H), 7.33–7.35 (d, J = 3.93 Hz, 1H), 6.58–6.61 (d, J = 8.73 Hz, 2H), 1.97 (broad m, 4H). ¹³C NMR (75.4 MHz, DMSO-d6, ppm): δ = 148.40, 136.65, 134.16, 127.33, 121.56, 112.44, 47.75, 25.42. FTIR (KBr): 3441.26 (OH stretch), 2210.59 (C≡N stretch), 1631.63 (C=O stretch), 1118.67 (C–O stretch). Elem. anal. calcd for C₁₈H₁₆N₂O₂S: C, 66.64%; H, 4.97%; N, 8.64%; S, 9.88; found: C, 66.78%; H, 4.86%; N, 8.48%; S, 9.70%.
Synthesis of 3-(5-[6-(4-[bis(4-methoxy-phenyl)-amino]-phenyl)-9-(2-ethylhexyl)-9H-carbazol-3-yl]-thiophene-2-yl)-2-cyano-acrylic acid (TC3). The compound 3c (0.088 g, 0.13 mmol), and cyanoacetic acid (0.022 g, 0.25 mmol) were under general Knoevenagel condensation procedure gave a maroon solid (0.038 mg, 40% yield). MS (MALDI-TOF) m/z: 759.5 (M⁺), 759.31 (calcd). ¹H NMR (300 MHz, DMSO-d6, ppm): δ = 8.76 (b, 1H, –CAR), 8.60 (b, 1H, –CAR), 8.48 (s, 1H, =CH), 8.05 (b, 1H, –CAR), 7.90–7.94 (dd, 1H), 7.78–7.83 (m, 2H), 7.65–7.72 (m, 4H), 7.10 (d, J = 9 Hz, 4H), 6.94–6.99 (m, 6H), 4.35 (d, Hz = 7.4, 2H, –CH₂), 3.80 (s, 6H, –OMe), 2.05 (m, 1H, –CH), 1.25–1.39 (m, 8H, –CH₂), 0.80–0.93 (m, 6H, –CH₃). ¹³C NMR (75.4 MHz, DMSO-d6, ppm): δ = 164.1, 156.0, 154.5, 147.5, 141.9, 141.8, 140.7, 140.6, 140.4, 134.0, 133.9, 133.2, 133.2, 132.1, 132.0, 127.8, 127.7, 126.7, 125.3, 123.8, 123.3, 123.3, 123.0, 122.9, 120.7, 120.6, 115.3, 55.6, 47.2, 30.5, 28.3, 24.0, 22.8, 14.1, 11.0. FT-IR (KBr): 3442.3 (OH stretch), 3035.8 (=CH stretch), 2927.4 (–CH stretch), 2216.1 (CN stretch), 1685.9 (CO stretch), 1577.7 (C=C alkene stretch), 1506.1 (C=C aromatic stretch) and 1240.0 (–C–O stretch). Elem. anal. calcd for C₄₈H₄₅N₃O₄S: C, 75.86%; H, 5.97%; N, 5.53%; S, 4.22; found: C, 75.68%; H, 6.11%; N, 5.63%; S, 4.37%.
Synthesis of 3-(5-[6-(4-pyrrolidin-1-yl-phenyl)-9-(2-ethylhexyl)-9H-carbazol-3-yl]-thiophene-2-yl)-2-cyano-acrylic acid (TC4). The compound 3b (0.1 g, 0.19 mmol) and cyanoacetic acid (0.033 g, 0.40 mmol) were reacted under general Knoevenagel condensation procedure gave product TC4 as red solid (0.032 g, yield 85%). MS (MALDI-TOF) m/z: 601.2777 (M⁺), 601.8 (calcd). ¹H NMR (300 MHz, DMSO-d6, ppm): δ = 8.52 (s, 1H), 8.74 (s, 1H), 8.00–8.01 (d, 1H), 8.44 (s, 1H), 7.85–7.88 (d, 1H), 7.79–7.80 (d, 1H), 7.72–7.75 (d, 1H), 7.64–7.66 (d, J = 8.37 Hz, 2H), 7.58–7.61 (d, 2H), 6.65–6.68 (d, J = 8.55 Hz), 4.29–4.31 (d, 2H), 3.17 (overlapped), 1.99 (b, 2H), 1.23 (m, 9H), 0.79–0.87 (m, 6H). ¹³C NMR (75.4 MHz, DMSO-d6, ppm): δ = 148.39, 147.06, 141.90, 141.70, 134.00, 127.96, 127.62, 124.62, 123.63, 123.38, 112.47, 110.85, 47.78, 30.53, 29.38, 28.44, 25.35, 24.01, 22.87, 14.20, 11.05. FTIR (KBr): 3443.25 (–OH stretch), 2958.03, 2923.32 and 2852.80 (sp³ –CH stretch), 2214.97 (C≡N stretch), 1684.88 (C=O stretch), 1610.03 (C=C stretch), 1063.72 (C–O stretch). Elem. anal. calcd for C₃₈H₃₉N₃O₂S: C, 75.84%; H, 6.53%; N, 6.98%; S, 5.33; found: C, 75.71%; H, 6.63%; N, 6.73%; S, 5.44%.
Synthesis of 3-(5-[7-(4-[bis(4-methoxy-phenyl)-amino]-phenyl)-9-(2-ethylhexyl)-9H-carbazol-2-yl]-thiophene-2-yl)-2-cyano-acrylic acid (TC5). The compound 4 (0.1 g, 0.14 mmol) and cyanoacetic acid (0.024 g, 0.29 mmol) were reacted under general Knoevenagel condensation procedure gave an orange solid (0.062 mg, 57% yield). MS (MALDI-TOF) m/z: 759.5 (M⁺), 759.31 (calcd). ¹H NMR (300 MHz, DMSO-d6, ppm): δ = 8.16–8.21 (m, 2H), 8.03 (s, 1H), 7.86 (s, 1H), 7.73 (d, 3H), 7.65 (d, J = 8.7 Hz, 2H), 7.55 (d, J = 8.2 Hz, 1H), 7.46 (d, J = 8.2 Hz, 1H), 7.08 (d, J = 8.9 Hz, 4H), 6.88–6.96 (m, 6H), 4.22 (d, J = 7.1 Hz, –CH₂), 3.76 (s, 6H, –OMe), 2.04 (t, 1H, –CH), 1.18–1.33 (m, 8H), 0.76–0.92 (m, 6H, –CH₃). ¹³C NMR (75.4 MHz, DMSO-d6, ppm): δ = 156.2, 149.4, 148.2, 142.4, 141.6, 141.6, 140.5, 139.8, 138.6, 136.7, 135.9, 133.0, 130.7, 128.1, 127.1, 127.0, 124.6, 122.7, 121.3, 121.2, 120.8, 120.1, 120.0, 118.2, 117.6, 115.3, 107.1, 106.7, 55.6, 30.5, 28.3, 24.1, 22.9, 14.1, 11.2. FT-IR (KBr): 3437.0 (OH stretch), 3034.1 (=CH stretch), 2927.9 (–CH stretch), 2211.4 (CN stretch), 1602.9 (CO stretch), 1506.4 (C=C aromatic stretch) and 1240.9 (–C–O stretch). Elem. anal. calcd for C₄₈H₄₅N₃O₄S: C, 75.86%; H, 5.97%; N, 5.53%; S, 4.22; found: C, 75.61%; H, 5.91%; N, 5.75%; S, 4.33%.

Conclusions

A series of organic dyes with multi-donor–acceptor (push–push–pull) configuration and different architecture were synthesized, characterized and primarily tested for improving the photovoltaic efficiency. The results showed that the push–push–pull system exhibited interesting performance due to the enhanced electron donor ability of the N,N-di(methoxyphenyl)phenylamine group at different positions (2,6- vs. 3,7-) of the carbazole. TC3 showed high values for molar absorptivity (ε = 33 718 M⁻¹ cm⁻¹) and open circuit voltage (V_OC = 748 mV) with an overall power conversion efficiency of 4.08%. The results reported here indicate that further improvements of the organic sensitizers could be achieved through systematic changes in molecular design via judicious choice of the number of donor groups, their chemical nature, and fine-tuning the overall geometry of the molecular dyes to prevent aggregation and reduce the dark current.

Acknowledgements

We greatly acknowledge the support of department of chemistry, National University of Singapore for providing funding (R-143-000-570-112) and technical support and NUS computer centre for high performance computation. Also, the authors thank Ms Sriramulu Deepa for helping to revise the paper.

Notes and references

1. C.-H. Chen, Y.-C. Hsu, H.-H. Chou, K. R. J. Thomas, J. T. Lin and C.-P. Hsu, Chem.–Eur. J., 2010, 16, 3184–3193.
2. B. O'Regan and M. Gratzel, Nature, 1991, 353, 737–740.
3. M. Gratzel, Nature, 2001, 414, 338–344.
4. T. W. Hamann, R. A. Jensen, A. B. F. Martinson, H. Van Ryswyk and J. T. Hupp, Energy Environ. Sci., 2008, 1, 66–78.
5. K.-L. Wu, C.-H. Li, Y. Chi, J. N. Clifford, L. Cabau, E. Palomares, Y.-M. Cheng, H.-A. Pan and P.-T. Chou, J. Am. Chem. Soc., 2012, 134, 7488–7496.
6. M. K. Nazeeruddin, S. M. Zakeeruddin, R. Humphry-Baker, M. Jirousek, P. Liska, N. Vlachopoulos, V. Shklover, C.-H. Fischer and M. Grätzel, Inorg. Chem., 1999, 38, 6298–6305.
7. A. Hagfeldt and M. Grätzel, Acc. Chem. Res., 2000, 33, 269–277.
8. M. K. Nazeeruddin, F. De Angelis, S. Fantacci, A. Selloni, G. Viscardi, P. Liska, S. Ito, B. Takeru and M. Grätzel, J. Am. Chem. Soc., 2005, 127, 16835–16847.
9. M. K. Nazeeruddin, P. Pechy and M. Gratzel, Chem. Commun., 1997, 1705–1706.
10. S. P. Singh, K. S. V. Gupta, M. Chandrasekharam, A. Islam, L. Han, S. Yoshikawa, M.-a. Haga, M. S. Roy and G. D. Sharma, ACS Appl. Mater. Interfaces, 2013, 5, 11623–11630.
11. A. Mishra, M. K. R. Fischer and P. Bäuerle, Angew. Chem., Int. Ed., 2009, 48, 2474–2499.
12. T. Saga, NPG Asia Mater., 2010, 2, 96–102.
13. H.-H. Chou, Y.-C. Chen, H.-J. Huang, T.-H. Lee, J. T. Lin, C. Tsai and K. Chen, J. Mater. Chem., 2012, 22, 10929–10938.
14. M.-W. Lee, J.-Y. Kim, H. J. Son, J. Y. Kim, B. Kim, H. Kim, D.-K. Lee, K. Kim, D.-H. Lee and M. J. Ko, Sci. Rep., 2015, 5, 7711.
15. M. Grätzel, Inorg. Chem., 2005, 44, 6841–6851.
16. M. Yanagida, T. Yamaguchi, M. Kurashige, K. Hara, R. Katoh, H. Sugihara and H. Arakawa, Inorg. Chem., 2003, 42, 7921–7931.
17. A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson, Chem. Rev., 2010, 110, 6595–6663.
18. N. Koumura, Z.-S. Wang, S. Mori, M. Miyashita, E. Suzuki and K. Hara, J. Am. Chem. Soc., 2006, 128, 14256–14257.
19. H. Wang, X. Zhang, F. Gong, G. Zhou and Z.-S. Wang, Adv. Mater., 2012, 24, 121–124.
20. M. Kimura, H. Nomoto, N. Masaki and S. Mori, Angew. Chem., 2012, 124, 4447–4450.
21. K. R. J. Thomas, N. Kapoor, C.-P. Lee and K.-C. Ho, Chem.–Asian J., 2012, 7, 738–750.
22. K. Hara, K. Sayama, Y. Ohga, A. Shinpo, S. Suga and H. Arakawa, Chem. Commun., 2001, 569–570.
23. K. Hara, T. Sato, R. Katoh, A. Furube, Y. Ohga, A. Shinpo, S. Suga, K. Sayama, H. Sugihara and H. Arakawa, J. Phys. Chem. B, 2002, 107, 597–606.
24. K. Hara, M. Kurashige, Y. Dan-oh, C. Kasada, A. Shinpo, S. Suga, K. Sayama and H. Arakawa, New J. Chem., 2003, 27, 783–785.
25. K. Hara, Y. Tachibana, Y. Ohga, A. Shinpo, S. Suga, K. Sayama, H. Sugihara and H. Arakawa, Sol. Energy Mater. Sol. Cells, 2003, 77, 89–103.
26. K. Hara, K. Miyamoto, Y. Abe and M. Yanagida, J. Phys. Chem. B, 2005, 109, 23776–23778.
27. K. Hara, Z.-S. Wang, T. Sato, A. Furube, R. Katoh, H. Sugihara, Y. Dan-oh, C. Kasada, A. Shinpo and S. Suga, J. Phys. Chem. B, 2005, 109, 15476–15482.
28. K. Hara, M. Kurashige, S. Ito, A. Shinpo, S. Suga, K. Sayama and H. Arakawa, Chem. Commun., 2003, 252–253,  10.1039/b210384b.
29. K. Hara, T. Sato, R. Katoh, A. Furube, T. Yoshihara, M. Murai, M. Kurashige, S. Ito, A. Shinpo, S. Suga and H. Arakawa, Adv. Funct. Mater., 2005, 15, 246–252.
30. T. Kitamura, M. Ikeda, K. Shigaki, T. Inoue, N. A. Anderson, X. Ai, T. Lian and S. Yanagida, Chem. Mater., 2004, 16, 1806–1812.
31. A. Keerthi, D. Sriramulu, Y. Liu, C. T. Yuan Timothy, Q. Wang and S. Valiyaveettil, Dyes Pigm., 2013, 99, 787–797.
32. K. S. V. Gupta, T. Suresh, S. P. Singh, A. Islam, L. Han and M. Chandrasekharam, Org. Electron., 2014, 15, 266–275.
33. R. Y.-Y. Lin, C.-P. Lee, Y.-C. Chen, J.-D. Peng, T.-C. Chu, H.-H. Chou, H.-M. Yang, J. T. Lin and K.-C. Ho, Chem. Commun., 2012, 48, 12071–12073.
34. A. Venkateswararao, K. R. J. Thomas, C.-P. Lee and K.-C. Ho, Tetrahedron Lett., 2013, 54, 3985–3989.
35. W. Xu, B. Peng, J. Chen, M. Liang and F. Cai, J. Phys. Chem. C, 2008, 112, 874–880.
36. D. P. Hagberg, J.-H. Yum, H. Lee, F. De Angelis, T. Marinado, K. M. Karlsson, R. Humphry-Baker, L. Sun, A. Hagfeldt, M. Grätzel and M. K. Nazeeruddin, J. Am. Chem. Soc., 2008, 130, 6259–6266.
37. K. R. Justin Thomas, A. Venkateswararao, C.-P. Lee and K.-C. Ho, Dyes Pigm., 2015, 123, 154–165.
38. G. Sathiyan, E. K. T. Sivakumar, R. Ganesamoorthy, R. Thangamuthu and P. Sakthivel, Tetrahedron Lett., 2016, 57, 243–252.
39. Z. Ning, Q. Zhang, W. Wu, H. Pei, B. Liu and H. Tian, J. Org. Chem., 2008, 73, 3791–3797.
40. A. Keerthi, Y. Liu, Q. Wang and S. Valiyaveettil, Chem.–Eur. J., 2012, 18, 11669–11676.
41. B.-G. Kim, K. Chung and J. Kim, Chem.–Eur. J., 2013, 19, 5220–5230.
42. Y. Wu, X. Zhang, W. Li, Z.-S. Wang, H. Tian and W. Zhu, Adv. Energy Mater., 2012, 2, 149–156.
43. A. C. Khazraji, S. Hotchandani, S. Das and P. V. Kamat, J. Phys. Chem. B, 1999, 103, 4693–4700.
44. R. Yeh-Yung Lin, F.-L. Wu, C.-H. Chang, H.-H. Chou, T.-M. Chuang, T.-C. Chu, C.-Y. Hsu, P.-W. Chen, K.-C. Ho, Y.-H. Lo and J. T. Lin, J. Mater. Chem. A, 2014, 2, 3092–3101.
45. Q. Feng, W. Zhang, G. Zhou and Z.-S. Wang, Chem.–Asian J., 2013, 8, 168–177.
46. G. Boschloo and A. Hagfeldt, Acc. Chem. Res., 2009, 42, 1819–1826.
47. S. M. Feldt, E. A. Gibson, E. Gabrielsson, L. Sun, G. Boschloo and A. Hagfeldt, J. Am. Chem. Soc., 2010, 132, 16714–16724.
48. Y. Liu, J. R. Jennings, Y. Huang, Q. Wang, S. M. Zakeeruddin and M. Grätzel, J. Phys. Chem. C, 2011, 115, 18847–18855.
49. T. Daeneke, A. J. Mozer, Y. Uemura, S. Makuta, M. Fekete, Y. Tachibana, N. Koumura, U. Bach and L. Spiccia, J. Am. Chem. Soc., 2012, 134, 16925–16928.
50. S. M. Feldt, P. W. Lohse, F. Kessler, M. K. Nazeeruddin, M. Gratzel, G. Boschloo and A. Hagfeldt, Phys. Chem. Chem. Phys., 2013, 15, 7087–7097.
51. H. Wang, J.-T. Ryu and Y. Kwon, J. Appl. Polym. Sci., 2011, 119, 377–386.
52. A. Tomkeviciene, J. V. Grazulevicius and V. Jankauskas, Chem. Lett., 2008, 37, 344–345.
53. L. Yu, J. Xi, H. T. Chan, T. Su, L. J. Antrobus, B. Tong, Y. Dong, W. K. Chan and D. L. Phillips, J. Phys. Chem. C, 2013, 117, 2041–2052.
54. A. Keerthi and S. Valiyaveettil, J. Phys. Chem. B, 2012, 116, 4603–4614.
55. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford CT, 2009.
56. R. Dennington, T. Keith and J. Millam, GaussView, Version 5, Semichem Inc., Shawnee Mission, KS, 2009.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra08530j

---