Rigid triarylamine donor–π–acceptor porphyrin dyes and their application in dye-sensitized solar cells

Abstract

Three donor–π–acceptor porphyrin dyes bearing a variety of rigid triarylamine donor groups were synthesized for application as photosensitizers in TiO₂ based dye-sensitized solar cells (DSSCs). Compared with the “naked” porphyrin ZnP, i.e. having no triarylamine moiety, broadened and red-shifted spectral features were exhibited by the triaylamine porphyrin sensitizers ISB-ZnP, CZ-ZnP, and IDB-ZnP where ISB = 5-phenyliminostilbene, CZ = 5-phenylcarbazole, and IDB = 5-phenyliminodibenzyl. Percentage power conversion efficiencies (η) and incident photon-to-current conversion efficiencies (%IPCE) in DSSC devices show the trend ISB-ZnP < ZnP < CZ-ZnP < IDB-ZnP. Inferior performance of the ISB-ZnP dye is attributed to its weaker adsorption to the TiO₂ film, which is roughly half that of ZnP. In contrast, introduction of the CZ and IDB electron donors is demonstrated to promote a better performance than the “naked” ZnP porphyrin. The best performance was observed for the IDB-ZnP device reaching a power conversion efficiency of η = 3.62% under AM 1.5 irradiation conditions with a corresponding %IPCE maximizing at 48% for both Soret band (450 nm) and Q band (570 nm) photoexcitation.

---

Introduction

Due to their low cost and high power conversion efficiencies, dye-sensitized solar cells (DSSCs) have attracted significant attention as alternatives to conventional photovoltaic devices based on silicon.^1,2 In the early years of DSSC development research was focused mainly on the development of metal complexes as sensitizers inspired by the N3 and N719 dyes.^3–5 While ruthenium based sensitizers have offered great redox stability in DSSC devices, there has always been environmental concerns with respect to their large scale application.⁶ For this reason organic photosensitizers, inspired by the porphyrin, chlorin and bacteriochlorin pigments found in the light harvesting antenna of photosystem II, have earned much interest.^7–12 Organic photosensitizers have a greater capacity for light absorption than the metal-to-ligand charge-transfer (MLCT) based visible absorption of ruthenium based photosensitizers and in recent years, due to enhanced steric and electronic molecular engineering, have competed successfully with the most efficient metal based DSSC devices.¹³ A wide variety of organic dyes such as porphyrins,^14–18 coumarins,^19–21 perylene,^22,23 phthalocyanines,²⁴ triarylamines,^25,26 and carbazoles^26,27 have been developed as successful alternatives to ruthenium(II) based sensitizers for use in DSSCs, however, the porphyrin chromophore remains the most widely studied system due to the combination of its highly evolved synthetic access and favorable optical, photophysical, electrochemical properties.^28,29 Not only does the porphyrin chromophore exhibit multiple strong absorption bands in the visible region, but its redox and photophysical properties can be finely tuned by functionalization at either of its four meso or eight beta positions.^30–32 The most celebrated application of porphyrins in a DSSC to date is the work of Yella et al. where the ethynyl conjugated triphenylamine porphyrin YD-o-C8 donor–π–acceptor system has achieved an overall power conversion efficiency of η = 12.3% in full sunlight.^13,18 In this particular case the strong tendency of the planar porphyrin macrocycle to π-aggregate at the TiO₂ surface was resolved by introducing bulky alkyl substituents to provide considerable steric hindrance while also maximizing photocurrent generation by preventing charge recombination from the reduced TiO₂ conduction band and the oxidized [Co(bpy)₃]³⁺ mediator.³³ Similarly, Aljarilla,³⁴ Fan,³⁵ and Zegkinoglou³⁶ have synthesized porphyrin dyes bearing the carbazole moiety as a redox active substituent in a donor–π–acceptor structure. Wang has recently introduced the triarylamines IDB (5-phenyl-iminodibenzyl) and ISB (5-phenyl-iminostilbene) as semi-rigid electron donating groups in conjugation with the simple thienyl cyanoacrylic acid acceptor.³⁷ A respectable power conversion efficiency of η = 5.83% was observed for the ISB donor leading to the conclusion that the semi-rigid donor structure contributes to a greater charge injection efficiency and an overall improvement in DSSC performance. In the current study we wanted to test the hypothesis that expansion of the covalent aryl bridging linkage of the carbazole donor would lead to an overall improvement in DSSC performance utilizing a donor–π–acceptor porphyrin design. Three new porphyrin dyes are here presented with the rigid triarylamine electron donors CZ (5-phenyl-carbazole), ISB and IDB introduced at the meso position of porphyrin ring (Fig. 1). The “naked” photosensitizer zinc(II)-15-(4-carboxyphenyl)-10,20-diphenylporphyrin (ZnP) lacking a meso-functionalized triarylamine electron donor was studied as a reference system. Structural, electronic, photophysical and computational properties of these systems were investigated by ¹H-NMR and mass spectroscopies, UV-vis electronic absorption and fluorescence emission spectroscopies, time-dependent density functional theory (TDDFT), cyclic voltammetry, photocurrent–voltage and electrochemical impedance measurements.

Fig. 1 Molecular structures of the ZnP, CZ-ZnP, IDB-ZnP and ISB-ZnP dyes.

Results and discussion

Synthesis

Three rigid triarylamine porphyrin dyes were synthesized to afford the target photosensitizers CZ-ZnP, IDB-ZnP and ISB-ZnP alongside the reference dye ZnP (Scheme 1). The porphyrin precursors dipyrromethane, 5-phenyldipyrromethane and 5-(4-methoxycarbonylphenyl) dipyrromethane were synthesized according to the literature.³⁸ The rigid triarylamine aldehydes 1, 2 and 3 were synthesized by a Buchwald–Hartwig type C–N coupling reaction between p-iodobenzaldehyde and the appropriate diarylamine in 52%, 62% and 67% yield, respectively, with a Pd(OAc)₂/DPEPhos catalyst in THF. This latter method proved more efficient than the previously reported Vilsmeier–Haack strategy.³⁷ The meso-substituted porphyrin intermediates 4, 5 and 6 were subsequently prepared by a standard MacDonald type 2 + 2 condensation reaction. Although the yields of 4, 5 and 6 were low due to scrambling reactions, sufficient quantities of sensitizer were isolated to proceed with the subsequent zinc(II) insertion and ester hydrolysis steps. The ZnP reference dye was synthesized in a similar manner.

Scheme 1 The synthetic routes for CZ-ZnP, IDB-ZnP, ISB-ZnP and ZnP. Reaction conditions: (i) Pd(OAc)₂, Cs₂CO₃, 18-crown-6, DPEPhos, THF(II) (a) TFA, CH₂Cl₂ (b) DDQ, NEt₃ (iii) Zn(OAc)₂·2H₂O, CHCl₃, CH₃OH(IV) (a) aq. KOH/THF/C₂H₅OH, (b) aq. HCl.

UV-vis electronic absorption and fluorescence emission spectroscopy

UV-vis electronic absorption spectra of the porphyrin dyes recorded in dichloromethane are presented in Fig. 2a. All of the dyes exhibit typical features of a porphyrin chromophore with a strongly absorbing Soret band (S₀ → S₂) band in the 410–430 nm spectral region with associated molar extinction coefficients in the range of 2.34 × 10⁵–5.16 × 10⁵ M⁻¹ cm⁻¹. The less intense Q(0,1) and Q(0,0) bands (vibronic bands within the S₀ → S₁ electronic transition) are observed in the spectral range 530–570 nm with molar extinction coefficients varying from 0.40 × 10⁴–7.68 × 10⁴ M⁻¹ cm⁻¹ (Table 1). As anticipated, due to introduction of the electron rich triarylamine substitutents, noticeable red-shifts of the Soret (7–10 nm) and Q (6–51 nm) bands are observed in each of the CZ-ZnP, IDB-ZnP and ISB-ZnP porphyrins relative to ZnP. The red-shift of the Soret bands is slight most likely due to the orthogonal orientation between the meso-aryl and central porphin ring systems. The same Soret bands do show a significant decrease in their molar extinction coefficients however, relative to the ‘naked’ ZnP system, with a concomitant broadening of their full-width half-maxima (fwhm). This suggests significant vibronic coupling between the triarylamine donors and the porphyrin ring system and is also favorable for enhanced light harvesting over a wider spectral range than ZnP (vide infra).³⁹ Interestingly, ISB-ZnP shows the strongest decrease and broadening of its Soret band for the series (ε = 3.54 × 10⁵ M⁻¹ cm⁻¹; fwhm = 1336 cm⁻¹) likely due to its extended π-conjugation across the back of the ISB donor. Fig. 2b shows the normalized absorption spectra of the porphyrin photosensitizers adsorbed on optically transparent 5 μm thick mesoporous TiO₂ films. The Soret and Q bands on TiO₂ films are comparable to their solution phase spectra in Fig. 1a with a slight broadening possibly due to monolayer aggregation at the TiO₂ interface.⁴⁰ Encouragingly, from a light harvesting perspective, relatively strong Q band absorption is observed for all dyes adsorbed on the mesoporous TiO₂ films.

Fig. 2 Absorption spectra of the porphyrin dyes recorded at 298 K in (a) dichloromethane and (b) adsorbed on optically transparent 5 μm thick TiO₂ films.

Table 1 UV-vis electronic absorption and fluorescence emission data

Dyes	Absorption/nm (ε/× 10^5 M^−1 cm^−1)^a	Emission/nm (Φfl)^b^,^c
a Recorded in CH2Cl2 at room temperature.b Recorded in acetonitrile under 1 atm N2 at room temperature.c Fluorescence data is corrected for excitation intensity (λexc = 555 nm) and detector response.
ZnP	414 (5.16), 542 (0.22)	600, 649 (0.0013)
CZ-ZnP	420 (4.54), 548 (0.19), 586 (0.04)	606, 658 (0.0016)
IDB-ZnP	422 (2.34), 552 (0.19), 593 (0.77)	613, 663 (0.0016)
ISB-ZnP	423 (3.54), 551 (0.26), 592 (0.10)	614, 664 (0.0020)

---

Fluorescence emission spectra were recorded in acetonitrile solvent to best reproduce the solvation conditions in a DSSC device. Typical for a meso-tetraphenyl porphyrin system, two distinct fluorescence bands are observed which arise from the Q(0,0)* and Q(0,1)* vibrationally resolved transitions (Fig. 3). All triarylamine substituted porphyrins show a stronger Q(0,0)* band relative to the ZnP system resulting in a slight increase in their fluorescence quantum yields. This may be as a result of increased molecular rigidity relative to the “naked” ZnP system. The ISB-ZnP system displays the largest quantum yield of the series at Φ_fl = 0.0020 (Table 1). Such low fluorescence quantum yields are typical of the porphyrin macrocycle which has a strong tendency for non-radiative decay due to vibrational freedom of the macrocycle.

Fig. 3 Corrected fluorescence emission spectra of all porphyrins recorded in argon degassed acetonitrile at 298 K. Peak areas are relative to the calculated quantum yields (re. Table 2). A schematic (right) is included to explain the origin of the higher energy Q(0,0)* and lower energy Q(0,1)* vibrationally resolved emission bands within the S₁ → S₀ electronic transition. Q-band (S₀ → S₁) and Soret (aka B) band (S₀ → S₂) transitions are also included for reference.

Computational analysis

To help guide interpretation of electrochemical data (vide infra), in correlation with the UV-vis electronic absorption spectra, computational analysis was conducted on each of the porphyrin photosensitizers. Geometry optimization was first carried out by density functional theory (DFT) using the polarizable continuum model (PCM) with the dielectric constant of dichloromethane⁴¹ where the rB3LYP hybrid exchange–correlation functional^42,43 and 6-31g** basis set^44,45 was used for all atoms. A vibrational frequency analysis coupled with the PCM model was carried out in order to confirm the minimum-energy geometry in solution. Electronic transitions contributions and oscillator strengths were subsequently determined by time-dependent density functional theory (TD-DFT).⁴⁶ Molecular orbital energies with frontier orbital images are presented in Fig. 4 with an exemplar TD-DFT predicted UV-vis spectrum presented in Fig. 5 for the IDB-ZnP dye. A comprehensive summary of all TD-DFT data is provided in the ESI† for each dye. A closer look at the DFT optimized dye geometries is presented in Fig. 6. Interestingly, all triarylamine porphyrins show a slight ruffling of the porphyrin macrocycle relative to ZnP which likely contributes to the slight broadening and red-shift of their Soret bands.⁴⁷ Furthermore, comparison of the triarylamine donor optimized geometries in Fig. 6 illustrates the non-planarity of the rigid IDB and ISB aryl groups relative to CZ. This effect may inhibit π-delocalization onto the IDB and ISB phenyl groups enhancing their electron donating capacity towards the porphyrin ring. This hypothesis is consistent with the trend in oxidation potentials observed experimentally via cyclic voltammetry studies below.

Fig. 4 Molecular orbital energy level diagram for ZnP and the rigid triarylamine porphyrin dyes calculated using DFT (rB3LYP/6-31g**/dichloromethane PCM). HOMO and LUMO levels are highlighted in blue and red, respectively, with theoretical band gaps (eV) included for comparison. Electron occupancy is not displayed, and only select frontier orbital images are included for clarity.

Fig. 5 Overlay of theoretical (line & curve) and experimental spectra for the IDB-ZnP dye calculated/recorded in dichloromethane PCM/solvent, respectively. Electronic transitions and major contributions responsible for the Soret and Q band absorptions are also highlighted.

Fig. 6 Side-on perspective view of DFT optimized geometries for ZnP and the rigid triphenylamine–porphyrin photosensitizers illustrating the non-planar geometry of both IDB and ISB triarylamine donors.

Electrochemical properties

The electrochemical properties of all porphyrins were investigated by cyclic voltammetry (CV) to assess their potential for application in a TiO₂ based DSSC device. Critical for this application, the excited state oxidation potential of the photosensitizer must be more negative than the TiO₂ semiconductor conduction band (E_CB = −0.50 V vs. NHE) for an exergonic charge-transfer photocurrent generation. Similarly, the excited state reduction potential of the photosensitizer must be more positive than the 2I⁻ → I₂˙⁻ oxidation reaction within the DSSC I₃⁻/I⁻ redox mediator solution to facilitate dye regeneration.^48,49 Unfortunately the 2I⁻ → I₂˙⁻ oxidation reaction can only be estimated as <+0.93 V vs. NHE.^49,50 In fact, the iodide oxidation potential is often misquoted as the triiodide reduction E(I₃⁻ → 3I⁻) = +0.35 V which occurs at the Pt counter electrode and is responsible for the observed V_OC based upon the relationship V_OC = E(I₃⁻ → 3I⁻) − E_Fermi(TiO₂).^48,49 Of course, reversible oxidation and reduction processes are also favoured to impart increase stability and high turnover number of the photosensitizer in a functional device. Cyclic voltammograms of all porphyrins recorded in N,N-dimethylformamide are presented in Fig. S21† and their data summarized in Table 2 in reference to the SCE electrode. All triarylamine porphyrins show a reversible first oxidation potential in contrast to the quasi-reversible nature of the ZnP first oxidation at +0.76 V. Oxidation of both IDB-ZnP and ISB-ZnP occurs at +0.74 V implying that they are slightly more electron rich than CZ-ZnP which displays its first oxidation potential at +0.82 V. A second quasi-reversible oxidation is observed for all triarylamine porphyrins in the range +0.94 to +1.02 V. All porphyrins in the series display three reductions, the first of which occurs in the narrow range of −1.41 V to −1.43 V implying that there is little change in energy of the porphyrin π* based LUMO energy level upon introduction of the various triarylamine electron donating groups. This is consistent with DFT calculations due to the limited π-conjugation across the orthogonal porphyrin and meso-aryl substituents.

Table 2 Electrochemical properties of the porphyrin dyes^a

Dyes	Eox (V vs. SCE)	Ered (V vs. SCE)	Ebg (eV)^b	E0–0 (eV)^c
a Recorded in 0.1 M Bu4NPF6 N,N-dimethylformamide electrolyte.b Electrochemical band-gap (Ebg = Eox − Ered).c Optical S0(v0) → S1(v0) band-gap.
ZnP	+1.15, +0.96, +0.76	−1.43, −1.80, −1.97	2.19	2.11
CZ-ZnP	+1.02, +0.82	−1.41, −1.79, −1.95	2.22	2.08
IDB-ZnP	+0.94, +0.74	−1.43, −1.83, −1.96	2.17	2.06
ISB-ZnP	+0.96, +0.74	−1.42, −1.83, −1.96	2.17	2.06

---

The photosensitizer excited-state oxidation, E(S⁺/S*), and reduction, E(S⁻/S*), potentials can be estimated by consideration of the absorbed photon energy according to eqn (1) and (2).

E(S⁺/S*) = E(S⁺/S) − ΔE_0–0(S/S*) (1)

E(S⁻/S*) = E(S⁻/S) + ΔE_0–0(S/S*) (2)

here, ‘S’ symbolizes the ground state photosensitizer, ‘S⁺’ and ‘S⁻’ the oxidized and reduced forms of the photosensitizer, S* the singlet excited state of the photosensitizer, and ΔE_0–0 the 0–0 energy of the S₀ → S₁ electronic transition. ΔE_0–0 may be estimated by determining the point of intersection between the normalized Q(0,0) absorption and Q(0,0)* emission bands. Alternatively, the electrochemically determined HOMO–LUMO band-gap E_bg may be used although this is often overestimated relative to optical band gap (Table 3). Energy level alignment of all porphyrins is summarized in Fig. 7. Here sufficient driving forces are observed for excited-state charge-transfer oxidation to the TiO₂ conduction band, ΔG = ∣E(S⁺/S*) − E_CB(TiO₂)∣, and the subsequent dye regeneration by the iodide redox mediator.³⁴

Table 3 Photovoltaic performance data of the four dyes in a DSSC device^a

Dyes	Γ (10^−8 mol cm^−2)	VOC (mV)	JSC (mA cm^−2)	FF	η (%)
a Light source: 100 mW cm^−2 AM 1.5G simulated solar light. Working area: 0.159 cm^2. Thickness: 15 mm. Dye bath: 0.3 mM EtOH solution. Electrolyte: 0.06 M LiI, 0.03 M I2, 0.6 M 1,2-dimethyl-3-propylimidazolium iodide (DMPII), 0.5 M 4-tertbutylpyridine (TBP) in acetonitrile solution.
ZnP	6.7	660	5.92	0.770	3.01
CZ-ZnP	5.2	680	7.00	0.729	3.47
IDB-ZnP	5.7	670	6.93	0.780	3.62
ISB-ZnP	3.5	640	5.20	0.767	2.55

---

Fig. 7 Energy level alignment of the porphyrin dyes relative to the TiO₂ conduction band (E_CB) and the iodide oxidation potential. Ground and excited state redox potentials are distinguished in blue and red, respectively. All potentials are referenced vs. NHE (SCE = +0.241 vs. NHE).

Photovoltaic studies

Dye-sensitized solar cells were fabricated according to standard methods using an FTO/TiO₂ sensitized photoanode, FTO/Pt cathode and a I₃⁻/I⁻ acetonitrile redox mediator electrolyte. Photocurrent density–voltage (J–V) curves are presented in Fig. 8. The surface coverage parameters of each dye (Γ, mol cm⁻²) and detailed parameters of photovoltaic performance, i.e. short-circuit current density (J_SC), open-circuit voltage (V_OC), fill factor (FF), and power conversion efficiency (η), are summarized in Table 3. The reference ZnP photosensitizer shows a power conversion efficiency of η = 3.01% while the CZ-ZnP and IDB-ZnP dyes show the best performance with η = 3.47% and 3.62%, respectively. This is primarily due to an increase in photocurrent with CZ-ZnP maximizing at J_SC = 7.00 mA cm⁻². Although IDB-ZnP displays a slightly lower photocurrent (J_SC = 6.93 mA cm⁻²) its greater fill-factor (FF = 0.780) allows it to top the series of dyes studied. Interestingly, the ISB-ZnP photosensitizer displays a poorer performance than even the ZnP reference system with a power conversion efficiency of just 2.55%. The lower V_OC of ISB-ZnP of 640 mV is likely due to a lowering of the TiO₂ Fermi-level consistent with the poorer photocurrent observed (J_SC = 5.92 mA cm⁻²). To gain further insight into the photocurrent generation of each device their incident photon-to-current conversion efficiency (IPCE) spectra were measured (Fig. 9). Consistent with the J–V plots, a lower photocurrent is observed across the entire visible spectrum for ISB-ZnP with a similar trend observed across the series, i.e. ISB-ZnP < ZnP < CZ-ZnP < IDB-ZnP. The improved light harvesting performance of CZ-ZnP and IDB-ZnP is consistent with their greater surface coverage relative to ISB-ZnP. In fact, both CZ-ZnP and IDB-ZnP display 48% IPCE maxima upon irradiation of both their Soret and Q-bands.

Fig. 8 Photocurrent–voltage (J–V) plots of the porphyrin DSSC devices recorded under AM 1.5G simulated irradiation.

Fig. 9 Percentage incident photon-to-current conversion efficiencies (%IPCE) recorded under monochromatic irradiation at 10 nm step intervals.

Electrochemical impedance spectroscopy

To gain further insight into the influence of the rigid triarylamine substituents on DSSC photovoltaic performance electrochemical impedance spectroscopy was conducted to study interfacial charge-transfer resistance.⁵¹ EIS analyses was performed in the dark at an applied bias of −0.6 V making the electronic and steric properties of each dye the only variable between each device. Nyquist plots are presented in Fig. 10. An electrochemical impedance Nyquist plot of a DSSC device typically contains three semicircles.^51–53 The high-frequency (leftmost) semicircle, associated with the Pt → I₃⁻ interfacial charge-transfer processes, is usually of weak intensity (low resistance) and is here obscured by the larger mid-frequency semicircle. The latter is attributed to TiO₂ → I₃⁻ interfacial charge recombination, here of prime importance. Finally, the low-frequency (right-most) semicircle, attributed to I₃⁻/I⁻ mass-transfer resistance, is also weak and is again mostly obscured by the intense mid-frequency signal.

Fig. 10 Electrochemical impedance Nyquist plots of porphyrin DSSC devices under a −0.6 V applied bias in dark conditions.

Dominance of the mid-frequency semicircle in the Nyquist plot confirms that TiO₂ → I₃⁻ interfacial charge transfer is rate-determining and significantly influencing the overall device efficiency. Under the experimental conditions employed (−0.6 V bias in the dark), observation of a high resistance to TiO₂ → I₃⁻ interfacial charge-recombination is favourable. This informs on the porosity of the TiO₂/porphyrin monolayer and its capacity for preventing performance deteriorating TiO₂ → I₃⁻ charge-recombination. Consistent with having the lowest V_OC, power conversion efficiency and %IPCE of the series ISB-ZnP displays the least resistance to charge-recombination. Overall the Nyquist plots show that the TiO₂ → I₃⁻ charge-recombination resistance increases in the order ISB-ZnP < ZnP < IDB-ZnP < CZ-ZnP. An identical trend is observed across the series for both J_SC and V_OC which suggests a similar trend in the TiO₂ Fermi level according to the relationship E_F = V_OC − E(I₃⁻/I⁻). An increased J_SC and V_OC is anticipated for CZ-ZnP based upon its greater oxidation potential and hence greater driving force for dye regeneration via iodide oxidation. However, considering these electronic factors, a slightly higher power conversion efficiency is still observed for the IDB-ZnP photosensitizer consistent with its greater surface coverage and larger fill factor relative to CZ-ZnP. Therefore the trend in observed photovoltaic parameters and charge-recombination resistance for the triarylamine porphyrin sensitizers here studied goes beyond their electronic properties and can be attributed to a combination of dye surface coverage and possibly even triarylamine steric effects at the TiO₂–photosensitizer interface.

Conclusions

In summary, three new porphyrin sensitizers containing the 5-phenyl-carbazole (CZ), 5-phenyl-iminodibenzyl (IDB) and 5-phenyl-iminostilbene (ISB) electron donors have been designed and synthesized for application in DSSCs. It was found that the CZ-ZnP and IDB-ZnP dyes exhibit an improved photovoltaic performance compared to the “naked” ZnP dye, which lacks a meso-substituted triarylamine electron donor. In contrast, the ISB-ZnP triarylamine photosensitizer displays a lower efficiency than ZnP. Using a combination of J–V, %IPCE and electrochemical impedance analysis the trend in DSSC device performance was found to correlate extremely well with the J_SC, V_OC and TiO₂ → I₃⁻ charge-recombination properties. As the electronic properties of each dye studied in solution were not deemed significantly different to cause such an effect, the differences observed in device performance are best attributed to surface coverage and steric factors which together control porosity and I₃⁻ mass-transfer in the mesoporous TiO₂/porphyrin film. In conclusion, this study confirms the critical importance of both the electronic and structural characteristics of triarylamine electron donors when designing a DSSC photosensitizer and highlights the IDB triarylamine donor as a promising alternative in the future design of original and versatile organic photosensitizers.

Experimental section

Materials and reagent

All reagents and solvents were obtained from commercial sources and used without further purification, unless otherwise noted. THF was dried prior to use by distillation over sodium using benzophenone as an indicator. CH₂Cl₂ was also freshly distilled prior to use. All chromatographic separations were carried out on silica gel (300–400 mesh or silica H) using CH₂Cl₂:petroleum ether solvent systems.

Analytical measurements

¹H-NMR spectra were recorded in CDCl₃ or DMSO on Bruker AV400 spectrometer. The chemical shifts were reported in parts per million (δ) relative to the appropriate reference signal: residual chloroform (δ_H 7.26 ppm) or DMSO (the quintet centered at 2.50 ppm). High resolution MALDI-TOF mass spectra were recorded on a Bruker Autoflex TOF/TOF III instrument. UV-vis electronic absorption spectra were recorded on a Shimadzu UV-1800 instrument. Although, acetonitrile is more suitable for mimicking DSSC conditions, molar extinction coefficients were determined in CH₂Cl₂ as complete dissolution of solids was not possible in acetonitrile. Emission spectra were recorded on a PTI Quantamaster 40 fluorimeter. Fluorescence quantum yields for all samples (Φ_fl) were calculated by the optically dilute technique in argon degased acetonitrile solutions with ZnTPP used as a reference (Φ_ref = 0.033 in toluene) for all samples according to eqn (3).

The subscript “s” refers to the sample and the subscript “ref” to the reference sample, A is the absorbance at the excitation wavelength, I is the integrated emission area, η_s is the solvent refractive index of the unknown sample and η_ref is the solvent refractive index of the reference sample. Excitation and emission slits were both set at 2 nm and an excitation wavelength of 555 nm was used for all samples. Cyclic voltammetry was conducted on a CH Instruments 620D potentiostat for all complexes. A standard three electrode cell was used with a supporting electrolyte of 0.1 M Bu₄NPF₆ in spectrophotometric grade N,N-dimethylformamide under an atmosphere of argon. The electrode assembly consisted of a glassy carbon disc working electrode (3 mm diameter), a Pt wire counter electrode and a non-aqueous reference electrode to minimize IR drop. The latter consisted of a Ag wire in the same 0.1 M Bu₄NPF₆ N,N-dimethylformamide electrolyte but separated by a porous vycor frit. The ferrocenium/ferrocene redox couple was used as a pseudo reference and was added to each porphyrin electrolyte solution at the end of every experiment for in situ calibration.

Computational analysis

Geometry optimization was first carried out using density functional theory (DFT) with the rB3LYP hybrid exchange–correlation functional^42,43 and 6-31g** basis set^44,45 for all atoms as implemented in the Gaussian ‘09 program⁵⁴ using the polarizable continuum model (PCM) with the dielectric constant of dichloromethane.⁴¹ A vibrational frequency analysis coupled with the PCM model was carried out in order to confirm the minimum-energy geometry in solution. Subsequent time-dependent density functional theory (TD-DFT) calculations were carried out on all compounds.⁴⁶ Finally, the percent contributions of atomic orbitals to molecular orbitals were calculated using the Gausssum software.⁵⁵

Absorption spectra on TiO₂ films

Absorption spectra of the sensitizers deposited on optically transparent TiO₂ films were measured with a Shimadzu UV-1800 spectrometer. The TiO₂ films with a typical thickness of 5 μm were dipped into 0.3 M ethanol solution of dyes for 15 min, then the sensitized films were rinsed with ethanol three times and air-dried before measurement of their absorption spectra.

Measurement of surface coverage (Γ)

TiO₂ films with an area of 0.16 cm² and 16 mm thickness were dipped into 0.3 M ethanol dye solutions for 2 h and then thoroughly washed with ethanol to remove any excess physisorbed dye. The quantity of chemisorbed dye was estimated by desorption in 3 mL of 0.1 M EtONa solution. The absorption spectra of the EtONa solutions were then measured to obtain the surface coverage (Γ) via molar extinction coefficient data according to the standard method.⁵⁶

Fabrication of DSSCs and photovoltaic measurements

A TiO₂ sol–gel was prepared from P25 following a literature procedure.⁵⁷ The FTO/TiO₂ electrodes were prepared by screen printing on a pre-cleaned FTO glass. The film was then annealed to 325 °C for 5 min, 375 °C for 5 min, 450 °C for 15 min, and 500 °C for 15 min. Finally, the TiO₂ film was post-treated with an aqueous TiCl₄ solution (50 mM) at 70 °C for 30 min, rinsed with ethanol, and annealed again at 500 °C for 30 min. After the film was cooled to 100 °C, the TiO₂ electrode was immersed into an ethanol solutions of the dye (0.3 mM) for 2 h. The sensitized electrode was rinsed thoroughly with ethanol and then air-dried. A sandwich cell consisting of the porphyrin sensitized TiO₂ working electrode and a Pt foil counter electrode was fabricated. The electrolyte (0.06 M LiI, 0.03 M I₂, 0.6 M 1,2-dimethyl-3-propylimidazolium iodide, and 0.5 M 4-tert-butylpyridine in acetonitrile) was injected between both working and counter electrodes. Upon fabrication the porphyrin based DSSCs were tested under simulated AM 1.5G irradiation (100 mW cm⁻²) by photocurrent–voltage (J–V) analysis on a Keithley 2400 Source meter (solar AAA simulator, Oriel China, calibrated with a standard crystalline silicon solar). The percentage power conversion efficiency (η) was calculated according to eqn (4):where J_SC is the short-circuit current density (mA cm⁻²), V_OC is the open-circuit voltage (V), FF is the fill-factor and I₀ the light flux (AM 1.5G). Photocurrent action experiments were performed on a commercial apparatus (Q Test Station 2000 IPCE Measurement System, CROWNTECH, USA) and incident photon-to-current conversion efficiencies were calculated according to eqn (5):where I_SC(λ) is the monochromatic short circuit current, I₀(λ) is the monochromatic light flux incident on the 0.16 cm² DSSC area and λ is the monochromatic excitation wavelength. Electrochemical impedance spectra of the DSSCs were recorded on a CH Instruments 760D potentionstat at a forward bias of −0.6 V under dark conditions at room temperature. The spectra were scanned in a frequency range of 0.1–10⁶ Hz and an AC amplitude of 10 mV at room temperature.

Synthesis of the photosensitizers

Synthesis of 1, 2 & 3. The rigid triarylamine aldehydes 1, 2 and 3 were prepared by coupling p-iodobenzaldehyde with the appropriate diarylamine which proved more efficient than the previously reported Vilsmeier–Haack procedure.³⁷ Analytical data was identical to that reported in the literature and is summarize following the general procedure outlined for 1. A solution of carbazole (334 mg, 2 mmol), 4-iodobenzaldehyde (557 mg, 2.4 mmol), palladium(II) acetate (22.4 mg, 0.1 mmol), DPEPhos (bis[(2-diphenylphosphino)phenyl]ether) (80.7 mg, 0.15 mmol), cesium carbonate (1.629 g, 2.5 mmol) and 18-crown-6 (26.4 mg, 0.1 mmol) in THF (50 mL) was refluxed for 5 h under argon atmosphere. The inorganic components were removed by filtration after cooling, the solvent was distilled under reduced pressure, then the crude product was purified by column chromatography to yield a white solid 1 (282 mg, 52%). ¹H NMR (400 MHz, CDCl₃) δ: 10.12 (s, 1H, –CHO), 8.19–8.12 (m, 4H, Ar–H), 7.80 (d, J = 8.2 Hz, 2H, Ar–H), 7.51 (d, J = 8.2 Hz, 2H, Ar–H), 7.44 (t, J = 7.6 Hz, 2H, Ar–H), 7.34 (t, J = 7.4 Hz, 2H, Ar–H).

2 a pale yellow solid (371 mg, 62%). ¹H NMR (400 MHz, CDCl₃) δ 9.74 (s, 1H, –CHO), 7.62 (d, J = 8.8 Hz, 2H, Ar–H), 7.42–7.36 (m, 2H, Ar–H), 7.32 (s, 6H, Ar–H), 6.64 (d, J = 8.8 Hz, 2H, Ar–H), 3.01 (s, 4H, Ar–H).

3 a light red solid (398 mg, 67%). ¹H NMR (400 MHz, CDCl₃) δ: 9.70 (s, 1H, –CHO), 7.61–7.45 (m, 8H, Ar–H), 7.44–7.37 (m, 2H, Ar–H), 6.87 (s, 2H, Ar–H), 6.36 (d, J = 8.8 Hz, 2H, –CH=CH–).

Synthesis of 4, 5 & 6. The mixture of methyl-4-formylbenzoate (164 g, 1 mmol), 1 (271 mg, 1 mmol), and 5-phenyldipyrromethane (444 mg, 2 mmol) was condensed in dichloromethane (200 mL) with TFA (0.055 mL, 0.74 mmol) at room temperature for 4 h in the dark. DDQ (0.681 g, 3 mmol) was added, after which the mixture was stirred for 6 h at room temperature. TEA (3 mL) was added to quench the reaction. The reaction mixture was stirred for another 1 h at room temperature. The solvent was removed under reduced pressure. The residue was purified by silica gel column chromatography and recrystallization from CH₂Cl₂/n-hexane gave 4 (34 mg, 4%). ¹H NMR (400 MHz, CDCl₃) δ: 9.01 (d, J = 4.8 Hz, 2H, pyrr–H), 8.94 (d, J = 4.8 Hz, 2H, pyrr–H), 8.89 (d, J = 4.7 Hz, 2H, pyrr–H), 8.81 (d, J = 4.7 Hz, 2H, pyrr–H), 8.52–8.43 (m, 4H, Ar–H), 8.37–8.20 (m, 8H, Ar–H), 8.00 (d, J = 8.1 Hz, 2H, Ar–H), 7.88–7.78 (m, 8H, Ar–H), 7.59 (t, J = 7.6 Hz, 2H), 7.41 (t, J = 7.5 Hz, 2H, Ar–H), 4.12 (s, 2H, –COOMe), −2.73 (s, 2H, NH). MALDI-Tof: m/z calcd for C₅₈H₃₉N₅O₂, 837.31; found, 837.61[M]⁺.

5 (52 mg, 6%). ¹H NMR (400 MHz, CDCl₃) δ 8.99 (t, J = 4.6 Hz, 2H, pyrr–H), 8.84 (d, J = 6.5 Hz, 4H, pyr–H), 8.76 (d, J = 4.8 Hz, 2H, pyrr–H), 8.44 (dd, J = 8.2, 3.1 Hz, 2H, Ar–H), 8.29 (dd, J = 8.5 Hz, 2H, Ar–H), 8.26–8.16 (m, 4H, Ar–H), 7.94 (dd, J = 8.6, 1.3 Hz, 2H), 7.82–7.68 (m, 8H, Ar–H), 7.39 (t, J = 7.6 Hz, 4H, Ar–H), 7.34–7.29 (m, 2H, Ar–H), 6.97 (d, J = 8.7 Hz, 2H), 4.11 (d, J = 3.9 Hz, 3H, –COOMe), 3.24 (s, 4H, –CH₂CH₂–), −2.75 (s, 2H, NH). MALDI-Tof: m/z calcd for C₆₀H₄₃N₅O₂, 865.34; found, 865.63[M]⁺.

6 (61 mg, 7%). ¹H NMR (400 MHz, CDCl₃) δ 8.94 (d, 2H, pyrr–H), 8.85–8.78 (m, 4H, pyrr–H), 8.78–8.70 (m, 2H, pyrr–H), 8.43 (dd, J = 8.2, 2.2 Hz, 2H, Ar–H), 8.29 (d, J = 8.2 Hz, 2H, Ar–H), 8.20 (d, J = 6.3 Hz, 2H, Ar–H), 7.87–7.82 (m, 2H, Ar–H), 7.80–7.71 (m, 8H, Ar–H), 7.65–7.55 (m, 4H, Ar–H), 7.48–7.48 (m, 2H, Ar–H), 7.05 (s, 2H, Ar–H), 6.68 (d, J = 8.7 Hz, 2H, –CH=CH–), 4.11 (d, J = 4.4 Hz, 3H, –COOMe), −2.78 (s, 2H, NH). MALDI-Tof: m/z calcd for C₆₀H₄₁N₅O₂, 863.32; found, 863.62[M]⁺.

Synthesis of 7, 8 & 9. To a solution of porphyrin 4, 5 or 6 (0.025 mmol) in CHCl₃ (20 mL) was added zinc(II) acetate dihydrate (22 mg, 0.10 mmol) in CH₃OH (5 mL). The mixture was refluxed for 1 h. The solvent was removed under reduced pressure. The residue was purified by silica gel column chromatography and recrystallization from CH₂Cl₂/n-hexane to give analytically pure product:

7 (21 mg, 93%). ¹H NMR (400 MHz, CDCl₃) δ: 9.12 (d, J = 4.7 Hz, 2H, pyrr–H), 9.05 (d, J = 4.7 Hz, 2H, pyrr–H), 9.00 (d, J = 4.7 Hz, 2H, pyrr–H), 8.91 (d, J = 4.7 Hz, 2H, pyrr–H), 8.45 (dd, J = 11.6, 8.2 Hz, 4H, Ar–H), 8.33 (d, J = 7.3 Hz, 2H, Ar–H), 8.30–8.23 (m, 6H, Ar–H), 7.99 (d, J = 8.2 Hz, 2H, Ar–H), 7.84–7.71 (m, 8H, Ar–H), 7.58 (t, J = 7.6 Hz, 2H, Ar–H), 7.40 (t, J = 7.4 Hz, 2H, Ar–H), 4.11 (s, 3H, –COOMe). MALDI-Tof: m/z calcd for C₅₈H₃₇N₅O₂Zn, 899.22; found, 899.14[M]⁺.

8 (21 mg, 92%) ¹H NMR (400 MHz, CDCl₃) δ 9.12–9.07 (m, 2H, pyrr–H), 8.94 (m, 4H, pyrr–H), 8.86 (d, J = 4.6 Hz, 2H, pyrr–H), 8.41 (dd, J = 8.2, 3.0 Hz, 2H, Ar–H), 8.30 (dd, J = 8.2, 1.6 Hz, 2H, Ar–H), 8.25–8.18 (m, 4H, Ar–H), 7.93 (d, J = 7.7 Hz, 2H, Ar–H), 7.74 (m, 8H, Ar–H), 7.38 (t, J = 7.9 Hz, 4H, Ar–H), 7.31 (dd, J = 11.1, 3.7 Hz, 2H, Ar–H), 6.96 (d, J = 8.6 Hz, 2H, Ar–H), 4.09 (d, J = 3.7 Hz, 3H, –COOMe).3.24 (s, 4H, –CH₂CH₂–). MALDI-Tof: m/z calcd for C₆₀H₄₁N₅O₂Zn, 927.26; found, 927.16[M]⁺.

9 (21 mg, 92%) ¹H NMR (400 MHz, CDCl₃) δ 9.10–9.05 (m, 2H, pyrr–H), 8.99–8.91 (m, 4H, pyrrole–H), 8.91–8.83 (m, 2H, pyrr–H), 8.44 (dd, J = 8.1, 2.1 Hz, 2H, Ar–H), 8.32 (d, J = 8.1 Hz, 2H, Ar–H), 8.22 (d, J = 6.6 Hz, 4H, Ar–H), 7.86 (d, J = 8.3 Hz, 2H, Ar–H), 7.78 (dt, J = 13.6, 6.9 Hz, 8H, Ar–H), 7.70–7.56 (m, 4H, Ar–H), 7.47 (t, J = 7.5 Hz, 2H, Ar–H), 7.08 (s, 2H, Ar–H), 6.70 (d, J = 8.5 Hz, 2H, –CH=CH–), 4.12 (d, J = 4.2 Hz, 3H, –COOCH₃). MALDI-Tof: m/z calcd for C₆₀H₃₉N₅O₂Zn, 925.23; found, 925.13[M]⁺.

Synthesis of CZ-ZnP, IDB-ZnP & ISB-ZnP. 5,15-Bis(2,4,6-trimethylphenyl)-15-triphenylamino(4-methoxycarbonylphenyl)porphyrin(Zn) (16 mg, 0.018 mmol) and KOH (41 mg, 0.72 mmol) in H₂O (2 mL) were added into THF–EtOH(1 : 1, 20 mL), and the solution was refluxed for 6 h. The mixture was then cooled to room temperature, acidified with concentrated HCl, and extracted with CHCl₃. The crude product left behind was then dried completely and purified with a column to afford the compound 13 (14 mg, 85%). ¹H NMR (400 MHz, DMSO) δ 9.05 (t, J = 4.2 Hz, 2H, pyrr–H), 8.87 (dd, J = 4.5, 2.3 Hz, 2H, pyrr–H), 8.85–8.77 (m, 4H, pyrr–H), 8.48 (d, J = 8.2 Hz, 2H, Ar–H), 8.44–8.28 (m, 6H, Ar–H), 8.21 (t, J = 7.4 Hz, 4H, Ar–H), 8.07 (d, J = 8.2 Hz, 2H, Ar–H), 7.96–7.76 (m, 8H, Ar–H), 7.62 (t, J = 7.5 Hz, 2H, Ar–H), 7.42 (t, J = 7.4 Hz, 2H, Ar–H). MALDI-Tof: m/z calcd for C₅₇H₃₅N₅O₂Zn, 885.20; found, 885.17[M]⁺.

14 (14 mg, 87%). ¹H NMR (400 MHz, DMSO) δ 8.93–8.85 (m, 2H), 8.76 (s, 6H), 8.36 (d, J = 7.8 Hz, 2H), 8.28 (d, J = 7.9 Hz, 2H), 8.18 (d, J = 4.6 Hz, 4H), 7.94 (d, J = 8.4 Hz, 2H), 7.80 (s, 6H), 7.70 (d, J = 7.5 Hz, 2H), 7.50–7.33 (m, 6H), 6.87 (d, J = 8.4 Hz, 2H). MALDI-Tof: m/z calcd for C₅₉H₃₉N₅O₂Zn, 913.23; found, 913.19[M]⁺.

15 (14 mg, 84%). ¹H NMR (400 MHz, DMSO) δ 8.85–8.80 (m, 2H), 8.74 (m, 6H), 8.35 (d, J = 8.2 Hz, 2H), 8.28 (d, J = 8.1 Hz, 2H), 8.16 (dd, J = 1.98, 7.07 Hz, 4H), 7.88–7.73 (m, 10H), 7.72–7.66 (m, 4H), 7.52 (t, J = 7.1 Hz, 1H), 7.15 (s, 2H), 6.57 (d, J = 8.6 Hz, 2H). MALDI-Tof: m/z calcd for C₅₉H₃₇N₅O₂Zn, 911.22; found, 911.16[M]⁺.

Synthesis of ZnP. To a solution of ZnP–CO₂Me⁴⁷ (X mmol) in CHCl₃ (10 mL) was added excess of Zn(OAc)₂·2H₂O (32.9 mg, 150 mmol) in MeOH (3 mL). The solution was refluxed for 1 h. The solvent was removed in vacuo. This crude residue product was added to the solution of KOH (0.08 g, 0.14 mmol) dissolved in THF–EtOH–H₂O (1 : 1 : 0.1, 50 mL). The solution was refluxed for 6 h. The mixture was then cooled to room temperature, acidified with concentrated HCl, and extracted with CH₂Cl₂. The crude product left was then dried completely and purified with a column to afford the compound ZnP (27.4 mg, 85%). ¹H NMR (400 MHz, DMSO) δ: 13.24 (s, 1H, –COOH), 10.35 (s, 1H, meso-H), 9.49 (d, J = 4.4 Hz, 2H, pyrr–H), 8.90 (d, J = 4.4 Hz, 2H, pyrr–H), 8.81 (dd, J = 17.3, 4.6 Hz, 4H, pyrr–H), 8.33 (dd, J = 24.8, 8.1 Hz, 4H, Ar–H), 8.26–8.16 (m, 4H, Ar–H), 7.84 (m, 6H, Ar–H). MALDI-TOF: m/z calcd for C₃₉H₂₄N₄O₂Zn, 644.11; found, 644.01[M]⁺.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (no. 21076147 & no. 21476162), the Natural Science Foundation of Tianjin (no. 10JCZDJC23700), the National International S&T Cooperation Foundation of China (no. 2012DFG41980).

References

1. S. Ardo and G. J. Meyer, Chem. Soc. Rev., 2009, 38, 115–164.
2. A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson, Chem. Rev., 2010, 110, 6595–6663.
3. B. O'Regan and M. Gratzel, Nature, 1991, 353, 737–740.
4. M. Gratzel, Inorg. Chem., 2005, 44, 6841–6851.
5. P. G. Bomben, K. C. D. Robson, B. D. Koivisto and C. P. Berlinguette, Coord. Chem. Rev., 2012, 256, 1438–1450.
6. K. Hara, Z. S. Wang, Y. Cui, A. Furube and N. Koumura, Energy Environ. Sci., 2009, 2, 1109–1114.
7. A. Kay and M. Gratzel, J. Phys. Chem., 1993, 97, 6272–6277.
8. W. M. Campbell, A. K. Burrell, D. L. Officer and K. W. Jolley, Coord. Chem. Rev., 2004, 248, 1363–1379.
9. M. K. Panda, K. Ladomenou and A. G. Coutsolelos, Coord. Chem. Rev., 2012, 256, 2601–2627.
10. L.-L. Li and E. W.-G. Diau, Chem. Soc. Rev., 2013, 42, 291–304.
11. M. Urbani, M. Graetzel, M. K. Nazeeruddin and T. Torres, Chem. Rev., 2014, 114, 12330–12396.
12. V. K. Singh, R. K. Kanaparthi and L. Giribabu, RSC Adv., 2014, 4, 6970–6984.
13. 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, 629–634.
14. W. M. Campbell, K. W. Jolley, P. Wagner, K. Wagner, P. J. Walsh, K. C. Gordon, L. Schmidt-Mende, M. K. Nazeeruddin, Q. Wang, M. Gratzel and D. L. Officer, J. Phys. Chem. C, 2007, 111, 11760–11762.
15. T. Bessho, S. M. Zakeeruddin, C.-Y. Yeh, E. W.-G. Diau and M. Graetzel, Angew. Chem., Int. Ed., 2010, 49, 6646–6649.
16. H. Imahori, Y. Matsubara, H. Iijima, T. Umeyama, Y. Matano, S. Ito, M. Niemi, N. V. Tkachenko and H. Lemmetyinen, J. Phys. Chem. C, 2010, 114, 10656–10665.
17. N. Xiang, X. Huang, X. Feng, Y. Liu, B. Zhao, L. Deng, P. Shen, J. Fei and S. Tan, Dyes Pigm., 2011, 88, 75–83.
18. 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. Graetzel, Nat. Chem., 2014, 6, 242–247.
19. Z.-S. Wang, Y. Cui, K. Hara, Y. Dan-oh, C. Kasada and A. Shinpo, Adv. Mater., 2007, 19, 1138–1141.
20. 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.
21. J. Wang, M. Li, D. Qi, W. Shen, R. He and S. H. Lin, RSC Adv., 2014, 4, 53927–53938.
22. Z. Yao, M. Zhang, H. Wu, L. Yang, R. Li and P. Wang, J. Am. Chem. Soc., 2015, 137, 3799–3802.
23. C. Li and H. Wonneberger, Adv. Mater., 2012, 24, 613–636.
24. J. J. He, G. Benko, F. Korodi, T. Polivka, R. Lomoth, B. Akermark, L. C. Sun, A. Hagfeldt and V. Sundstrom, J. Am. Chem. Soc., 2002, 124, 4922–4932.
25. M. Liang and J. Chen, Chem. Soc. Rev., 2013, 42, 3453–3488.
26. R. Z. Li, M. Zhang, C. C. Yan, Z. Y. Yao, J. Zhang and P. Wang, ChemSusChem, 2015, 8, 97–104.
27. N. Koumura and K. Hara, Heterocycles, 2013, 87, 275–301.
28. D. M. Guldi, Chem. Soc. Rev., 2002, 31, 22–36.
29. H. Kurreck and M. Huber, Angew. Chem., Int. Ed. Engl., 1995, 34, 849–866.
30. M. Victoria Martinez-Diaz, G. de la Torrea and T. Torres, Chem. Commun., 2010, 46, 7090–7108.
31. H. Imahori, S. Hayashi, H. Hayashi, A. Oguro, S. Eu, T. Umeyama and Y. Matano, J. Phys. Chem. C, 2009, 113, 18406–18413.
32. J. M. Ball, N. K. S. Davis, J. D. Wilkinson, J. Kirkpatrick, J. Teuscher, R. Gunning, H. L. Anderson and H. J. Snaith, RSC Adv., 2012, 2, 6846–6853.
33. S. M. Feldt, E. A. Gibson, E. Gabrielsson, L. Sun, G. Boschloo and A. Hagfeldt, J. Am. Chem. Soc., 2010, 132, 16714–16724.
34. A. Aljarilla, J. N. Clifford, L. Pelleja, A. Moncho, S. Arrechea, P. de la Cruz, F. Langa and E. Palomares, J. Mater. Chem. A, 2013, 1, 13640–13647.
35. C. Fan, X. Wang, X. Wang and J. Luo, Opt. Mater., 2012, 34, 609–615.
36. I. Zegkinoglou, M.-E. Ragoussi, C. D. Pemmaraju, P. S. Johnson, D. F. Pickup, J. Enrique Ortega, D. Prendergast, G. de la Torre and F. J. Himpsel, J. Phys. Chem. C, 2013, 117, 13357–13364.
37. C. Wang, J. Li, S. Cai, Z. Ning, D. Zhao, Q. Zhang and J.-H. Su, Dyes Pigm., 2012, 94, 40–48.
38. B. J. Littler, Y. Ciringh and J. S. Lindsey, J. Org. Chem., 1999, 64, 2864–2872.
39. H. N. Fonda, J. V. Gilbert, R. A. Cormier, J. R. Sprague, K. Kamioka and J. S. Connolly, J. Phys. Chem., 1993, 97, 7024–7033.
40. J. Rochford, D. Chu, A. Hagfeldt and E. Galoppini, J. Am. Chem. Soc., 2007, 129, 4655–4665.
41. J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005, 105, 2999–3093.
42. A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
43. C. T. Lee, W. T. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789.
44. P. c. Harihara and J. A. Pople, Theor. Chim. Acta, 1973, 28, 213–222.
45. M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon, D. J. Defrees and J. A. Pople, J. Chem. Phys., 1982, 77, 3654–3665.
46. G. Scalmani, M. J. Frisch, B. Mennucci, J. Tomasi, R. Cammi and V. Barone, J. Chem. Phys., 2006, 124, 94107.
47. Y. Zhou, K. T. Ngo, B. Zhang, Y. Feng and J. Rochford, Organometallics, 2014, 33, 7078–7090.
48. G. Boschloo and A. Hagfeldt, Acc. Chem. Res., 2009, 42, 1819–1826.
49. J. G. Rowley, B. H. Farnum, S. Ardo and G. J. Meyer, J. Phys. Chem. Lett., 2010, 1, 3132–3140.
50. J. H. Baxendale, P. L. T. Bevan and D. A. Stott, Trans. Faraday Soc., 1968, 64, 2389–2397.
51. F. Fabregat-Santiago, G. Garcia-Belmonte, I. Mora-Sero and J. Bisquert, Phys. Chem. Chem. Phys., 2011, 13, 9083–9118.
52. Q. Wang, J. E. Moser and M. Gratzel, J. Phys. Chem. B, 2005, 109, 14945–14953.
53. N. Koide, A. Islam, Y. Chiba and L. Han, J. Photochem. Photobiol., A, 2006, 182, 296–305.
54. M. J. Frisch, et al., Gaussian 09, Gaussian Inc., Wallingford, CT, 2009.
55. N. M. O'Boyle, A. L. Tenderholt and K. M. Langner, J. Comput. Chem., 2008, 29, 839–845.
56. M. Yanagida, T. Yamaguchi, M. Kurashige, G. Fujihashi, K. Hara, R. Katoh, H. Sugihara and H. Arakawa, Inorg. Chim. Acta, 2003, 351, 283–290.
57. S. Ito, M. K. Nazeeruddin, P. Liska, P. Comte, R. Charvet, P. Pechy, M. Jirousek, A. Kay, S. M. Zakeeruddin and M. Gratzel, Prog. Photovoltaics, 2006, 14, 589–601.

---

Footnote

† Electronic supplementary information (ESI) available: NMR and mass spectra, TDDFT data, DFT coordinates. See DOI: 10.1039/c5ra04862a

---