Effective co-sensitization using D–π–A dyes with a pyridyl group adsorbing at Brønsted acid sites and Lewis acid sites on a TiO₂ surface for dye-sensitized solar cells

Abstract

Effective and convenient co-sensitization to enhance dye coverage on TiO₂ electrodes for dye-sensitized solar cells have been achieved successfully by employing two kinds of D–π–A dyes with a pyridyl group capable of adsorbing at the Brønsted acid sites and the Lewis acid sites on a TiO₂ surface.

---

TiO₂-based dye-sensitized solar cells (DSSCs) have been widely recognized as one of the most promising photovoltaic cells because of their interesting construction and operational principles, decorative natures, and low cost of production.^1–3 To develop high-performance DSSCs, many kinds of ruthenium (Ru) dyes, porphyrin dyes, phthalocyanine dyes and organic dyes having a carboxyl group as an anchoring group have been developed as dye sensitizers during the last two decades.^1–7 These dyes are usually bound to the surface of TiO₂ through a carboxyl group, which can form a bidentate bridging linkage at Brønsted acid sites (surface-bound hydroxy groups, Ti–OH) of the TiO₂ surface. To further improve the photovoltaic performance of DSSCs, co-sensitization employing two kinds of organic dyes,⁸ phthalocyanine dye and organic dye,⁹ porphyrin dye and organic dye,¹⁰ and Ru-complex dye and organic dye¹¹ has been utilized as an effective approach to achieve high surface coverage of TiO₂ electrode, leading to improvement of light-harvesting efficiency (LHE), suppression of charge recombination between the injected electrons in TiO₂ and I₃⁻ ions in the electrolyte, and prevention of the aggregation between dye molecules. However, the co-sensitization using two kinds of dyes with carboxyl group often causes competitive adsorption between the two kinds of dyes, leading to lowering of LHE due to the decrease in the amount of the dye adsorbed on the TiO₂ electrode, and thus resulting in reduce of power conversion efficiency.

On the other hand, we have designed and developed so far new-type of D–π–A dye sensitizers NI-5, NI-6‡ and YNI-2‡ with pyridyl group as an electron-withdrawing-injecting anchoring group (Scheme 1),¹² which can adsorb onto TiO₂ electrode through the formation of coordinate bonding between the pyridyl group of the dye and the Lewis acid site (exposed Tiⁿ⁺ cations) on the TiO₂ surface. Moreover, we found that the D–π–A dye SAT-1‡ with benzo[4,5]thieno[2,3-c]pyridine as thiophene-fused pyridine ring,¹³ is predominantly adsorbed on the TiO₂ surface through the hydrogen bonding at Brønsted acid sites. For this reason, quite recently, to obtain insight into the effective combination of two different dyes to overcome the competitive adsorption in the co-sensitization method, we have prepared co-sensitized DSSCs employing Ru-complex dye (black dye) with carboxyl group and D–π–A dye sensitizers (NI-5 or YNI-2) with pyridyl group.¹⁴ The conversion efficiency of co-sensitized DSSCs was improved compared with that of DSSC based on only black dye, NI-5 or YNI-2. Thus, the previous study provided a new effective co-sensitization method using both Brønsted acid sites and Lewis acid sites on the TiO₂ surface to improve the efficiency of DSSCs, that is, a successful strategy for developing efficient co-sensitized DSSCs is to separate the adsorption sites on the TiO₂. However, the co-adsorption of black dye and NI-5 or YNI-2 to the TiO₂ electrode was carried out by a stepwise adsorption method, because the NCS ligands of black dye would undergo ligand substitution by a pyridyl group of NI-5 or YNI-2 in a mixed solution. In addition, an acid-base reaction between the dye with carboxyl group and the dye with pyridyl group in the mixed solution may cause a decrease in the amount of the dye adsorbed on the TiO₂ electrode. Thus, the stepwise dye adsorption is required as long as the dye with carboxyl group and the dye with pyridyl group are employed as co-sensitization dyes, so that the current co-sensitization method has still disadvantage in the use as convenient co-adsorption method.

Scheme 1 Chemical structures of D–π–A dye sensitizers NI-5, NI-6, YNI-2 and SAT-1.

In this study, to provide an innovative approach for effective and convenient co-sensitization in DSSCs, we developed a new co-sensitization method employing two kinds of D–π–A dyes with pyridyl group capable of adsorbing at the Brønsted acid sites and the Lewis acid sites on TiO₂ surface, that is, one-step co-adsorption of SAT-1 and NI-6 or YNI-2 to the TiO₂ electrode using the mixed solution. We have demonstrated that effective and convenient co-sensitization of SAT-1 and NI-6 or YNI-2 on the TiO₂ electrode without competitive adsorption was achieved successfully due to the site-selective adsorption behaviour of the two kinds of D–π–A dyes with pyridyl group, resulting in enhancement of dye coverage on TiO₂ electrode.

The dyes SAT-1, NI-6 and YNI-2 showed the absorption maximum (λ^abs_max) at around 410 nm (ε = 38 400 M⁻¹ cm⁻¹), 395 nm (ε = 48 100 M⁻¹ cm⁻¹) and 378 nm (ε = 46 100 M⁻¹ cm⁻¹), respectively (Fig. 1a), which is assigned to the intramolecular charge-transfer (ICT) excitation from electron donor moiety (diphenylamino group for NI-6 and SAT-1, and carbazole unit for YNI-2) to electron acceptor moiety (pyridyl group). This result shows that the dyes SAT-1 and YNI-2 have complementary absorption properties, compared with the case of the dyes SAT-1 and YNI-2. The absorption bands of co-adsorbed TiO₂ film with SAT-1 and NI-6 or YNI-2 show are broadened, compared with the absorption bands of SAT-1, NI-6 and YNI-2 in THF (Fig. 1b). It is worth noting that the co-adsorbed TiO₂ film with SAT-1 and NI-6 shows broad absorption band at the onset compared with that of SAT-1 and YNI-2.

Fig. 1 (a) Absorption spectra of SAT-1, NI-6 and YNI-2 in THF and (b) absorption spectra of co-sensitizers (SAT-1 + NI-6 and SAT-1 + YNI-2) adsorbed on TiO₂ film (3 μm).

In order to investigate effective co-sensitization condition employing two kinds of D–π–A dyes with pyridyl group, the co-adsorbed TiO₂ electrode was prepared under various concentration of dye solution (0.05 mM, 0.1 mM or 1 mM) and by a stepwise adsorption method (firstly immersed into SAT-1 solution and then NI-6 or YNI-2 solution, or firstly immersed into NI-6 or YNI-2 solution and then SAT-1 solution) or a one-step co-adsorption method using the mixed dye (SAT-1 and NI-6 or YNI-2) solution. Consequently, it was found that the best photovoltaic performance for co-sensitized DSSCs was obtained in the co-adsorbed TiO₂ electrode prepared by one-step co-adsorption method using 0.1 mM mixed dye solution, that is, one-step co-adsorption method is the most effective and convenient co-adsorption method for co-sensitized DSSCs. Thus, the co-adsorption of SAT-1 and NI-6 or YNI-2 to the TiO₂ electrode was carried out by one-step co-adsorption method using 0.1 mM mixed dye (SAT-1 and NI-6 or YNI-2) solution (Table 1, see ESI† for details of experimental procedures). It is worth mentioning here that the adsorption amounts of SAT-1 and NI-6 on the co-adsorbed TiO₂ electrode are 6.8 × 10¹⁶ and 7.1 × 10¹⁶ molecules per cm², respectively, which are equivalent to or slightly higher than that (3.8 × 10¹⁶ molecules per cm² for SAT-1 and 6.9 × 10¹⁶ molecules per cm² for NI-6) on the individually-adsorbed TiO₂ electrode. Also, the adsorption amounts of SAT-1 (3.6 × 10¹⁶ molecules per cm²) and YNI-2 (1.0 × 10¹⁷ molecules per cm²) on the co-adsorbed TiO₂ electrode are equivalent of that (1.1 × 10¹⁷ molecules per cm² for YNI-2) on the individually-adsorbed TiO₂ electrode. As reported previously,^12,13 the FTIR spectra of these D–π–A dyes with pyridyl group adsorbed on TiO₂ nanoparticles demonstrated that the dyes NI-6 and YNI-2 are predominantly adsorbed on the TiO₂ surface through coordinate bonding at Lewis acid sites, that is, the band at around 1610 cm⁻¹ can be assigned to pyridyl group coordinated to the Lewis acid sites on the TiO₂ surface (Fig. 2). On the other hand, the dye SAT-1 is predominantly adsorbed on the TiO₂ surface through the hydrogen bonding at Brønsted acid site, that is, the band at 1598 cm⁻¹ can be assigned to the hydrogen-bonded pyridyl group to Brønsted acid sites on the TiO₂ surface. Thus, to elucidate the adsorption states of the two kinds of dyes on co-adsorbed TiO₂ nanoparticles, we measured the FTIR spectra (SAT-1 + NI-6 and SAT-1 and YNI-2 in Fig. 2) of the co-adsorbed TiO₂ nanoparticles with SAT-1 and NI-6 or YNI-2. In the FTIR spectra of the co-adsorbed TiO₂ nanoparticles, the two characteristic bands were clearly observed at around 1598 and 1610 cm⁻¹, which can be assigned to the hydrogen-bonded pyridyl group to Brønsted acid sites and the coordinated pyridyl group to the Lewis acid sites, respectively, on the TiO₂ surface. Thus, the FTIR spectral measurement revealed that the two kinds of D–π–A dyes with pyridyl group on the co-adsorbed TiO₂ nanoparticles are adsorbed on the TiO₂ surface through the hydrogen bonding at Brønsted acid site for SAT-1 and the coordinate bonding at Lewis acid sites for NI-6 and YNI-2. This result clearly indicates that effective and convenient co-sensitization without competitive adsorption was achieved successfully by employing the two kinds of D–π–A dyes with pyridyl group possessing the site-selective adsorption behaviour.

Table 1 Photovoltaic performances of DSSCs based on SAT-1, NI-6 or YNI-2 and co-sensitized DSSCs based on SAT-1 and NI-6 or YNI-2^a

Dye	Jsc	Voc/mV	ff	η (%)	Adsorption amount dye on TiO2 electrode^b (molecules per cm^2)
a Under a simulated solar light (AM 1.5, 100 mW cm^−2).b The 9 μm thick TiO2 electrode was immersed into 0.1 mM dye (SAT-1, NI-6 or YNI-2) solution in THF or 0.1 mM mixed dye (SAT-1 and NI-6 or YNI-2) solution in THF for 15 hours enough to adsorb the dye sensitizers. The amount of adsorbed dye (only SAT-1, NI-6 or YNI-2) on TiO2 nanoparticles was determined form the calibration curve by absorption spectral measurement of the concentration change of the dye solution before and after adsorption. For the adsorption amounts of dye on the co-adsorbed TiO2 electrode, dye solution after the co-adsorption was chromatographed on silica gel plate (dichloromethane–ethyl acetate = 5 : 1 and then dichloromethane as eluent for SAT-1 and NI-6, and dichloromethane–ethyl acetate = 5 : 1 as eluent for SAT-1 and YNI-2) to isolate the dye from the other dye, and then the adsorption amount of each dye was determined form the calibration curve by the absorption spectral measurement of the dye solution.
SAT-1	2.85	481	0.57	0.79	3.8 × 10^16
NI-6	4.76	523	0.59	1.47	6.9 × 10^16
YNI-2	5.64	543	0.66	2.02	1.1 × 10^17
SAT-1 + NI-6	4.52	505	0.61	1.39	6.8 × 10^16 (SAT-1), 7.1 × 10^16 (NI-6)
SAT-1 + YNI-2	5.56	561	0.64	1.99	3.6 × 10^16 (SAT-1), 1.0 × 10^17 (YNI-2)

---

Fig. 2 FTIR spectra of (a) SAT-1, NI-6 and co-sensitizers (SAT-1 + NI-6) and (b) SAT-1, YNI-2 and co-sensitizers (SAT-1 + YNI-2) adsorbed on TiO₂ nanoparticles.

The co-sensitized DSSCs were prepared by using the co-adsorbed TiO₂ electrode with SAT-1 and NI-6 or YNI-2, Pt-coated glass as a counter electrode, and an acetonitrile solution with iodine (0.05 M), lithium iodide (0.1 M), and 1,2-dimethyl-3-propylimidazolium iodide (0.6 M) as electrolyte. The photocurrent–voltage (I–V) characteristics were measured under simulated solar light (AM 1.5, 100 mW cm⁻²). The incident photon-to-current conversion efficiency (IPCE) spectra and the I–V curves are shown in Fig. 3. The photovoltaic performance parameters of DSSCs based on only SAT-1, NI-6 or YNI-2 and co-sensitized DSSCs based on SAT-1 and NI-6 or YNI-2 are collected in Table 1. The IPCE spectrum of co-sensitized DSSC based on SAT-1 and NI-6 is broadened at the onset compared with that of co-sensitized DSSC based on SAT-1 and YNI-2 (Fig. 3a), which are in good agreement with the absorption spectra of co-adsorbed TiO₂ film. The maximum IPCE value of co-sensitized DSSC based on SAT-1 and NI-6 (52% at 424 nm) or SAT-1 and YNI-2 (59% at 425 nm) is lower than that of DSSC based on only NI-6 (64% at 422 nm) or equivalent of that of DSSC based on only YNI-2 (61% at 424 nm), but is higher than that of DSSC based on only SAT-1 (32% at 424 nm). The I–V curves (Fig. 3b) show that the short-circuit photocurrent density (J_sc) and solar energy-to-electricity conversion yield (η) of co-sensitized DSSC based on SAT-1 and NI-6 (4.52 mA cm⁻² and 1.39%) or SAT-1 and YNI-2 (5.56 mA cm⁻² and 1.99%) is slightly lower than that of DSSC based on only NI-6 (4.76 mA cm⁻² and 1.47%) or equivalent of that of DSSC based on only YNI-2 (5.64 mA cm⁻² and 2.02%), but is higher than that of DSSC based on only SAT-1 (2.85 mA cm⁻² and 0.79%). These results indicate that the lowering of photovoltaic performance of co-sensitized DSSC relative to that of DSSC based on only NI-6 or YNI-2 may be attributed to a decrease in electron injection efficiency by electron transfer and/or energy transfer between the two kinds of dyes due to poorly complementary absorption properties. Moreover, it is worth mentioning here that the open-circuit photovoltage (V_oc) value (561 mV) of co-sensitized DSSC based on SAT-1 and YNI-2 is higher than that of SAT-1 and NI-6 (505 mV). Thus, electrochemical impedance spectroscopy (EIS) analysis was performed to study the electron recombination process in co-sensitized DSSCs based on SAT-1 and NI-6 or YNI-2 in the dark under a forward bias of −0.60 V with a frequency range of 10 mHz to 100 kHz. The large semicircle in the Nyquist plot, which corresponds to the mid frequency peaks in the Bode phase plots, represents the charge recombination between the injected electrons in TiO₂ and I₃⁻ ions in the electrolyte, that is, the charge-transfer resistances at the TiO₂/dye/electrolyte interface. The Nyquist plots (Fig. 4a) show that the resistance value for the large semicircle for co-sensitized DSSC based on SAT-1 and YNI-2 (32 Ω) is larger than that of SAT-1 and NI-6 (26 Ω), indicating that the electron recombination resistance of co-sensitized DSSC based on SAT-1 and YNI-2 is higher than that of SAT-1 and NI-6. The electron recombination lifetimes (τ_e) expressing the electron recombination between the injected electrons in TiO₂ and I₃⁻ ions in the electrolyte, extracted from the angular frequency (ω_rec) at the frequency peak in the Bode phase plot (Fig. 4b) using τ_e = 1/ω_rec, are 4 ms for co-sensitized DSSC based on SAT-1 and NI-6 and 10 ms for co-sensitized DSSC based on SAT-1 and YNI-2, respectively, which is consistent with the V_oc values in the co-sensitized DSSCs. Consequently, the co-adsorption of SAT-1 and YNI-2 with two pyridyl groups can efficiently suppress charge recombination between the injected electrons in the CB of TiO₂ and I₃⁻ ions in the electrolyte, arising from the approach of I₃⁻ ions to the TiO₂ surface, which is responsible for the higher V_oc value for co-sensitized DSSC based on SAT-1 and YNI-2.

Fig. 3 (a) IPCE spectra and (b) I–V curves of DSSCs based on SAT-1, NI-6 or YNI-2 and co-sensitized DSSCs based on SAT-1 and NI-6 or YNI-2.

Fig. 4 (a) Nyquist plots and (b) Bode phase plots of co-sensitized DSSCs based on SAT-1 and NI-6 or YNI-2.

In conclusion, co-sensitized DSSCs employing two kinds of D–π–A dyes with pyridyl group capable of adsorbing at the Brønsted acid sites and the Lewis acid sites on TiO₂ surface have been developed. We have demonstrated that effective and convenient co-sensitization without competitive adsorption was achieved successfully by one-step co-adsorption using the mixed solution due to the site-selective adsorption behaviour of the two kinds of D–π–A dyes with pyridyl group. Thus, this study provides that the co-sensitization method employing two kinds of D–π–A dyes with pyridyl group possessing bonding ability to both Brønsted acid sites and the Lewis acid sites on TiO₂ surface is one of the most promising strategy to enhance dye coverage on TiO₂ electrode and LHE for DSSCs, leading to improvement in photovoltaic performances of co-sensitized DSSCs.

Notes and references

1. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737.
2. A. Mishra, M. K. R. Fischer and P. Bäuerle, Angew. Chem., Int. Ed., 2009, 48, 2474.
3. (a) Z. Ning and H. Tian, Chem. Commun., 2009, 5483; (b) Z. Ning, Y. Fu and H. Tian, Energy Environ. Sci., 2010, 3, 1170.
4. (a) Y. Ooyama and Y. Harima, Eur. J. Org. Chem., 2009, 18, 2903; (b) Y. Ooyama and Y. Harima, ChemPhysChem, 2012, 13, 4032.
5. A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson, Chem. Rev., 2010, 110, 6595.
6. (a) L.-L. Li and E. W.-G. Diau, Chem. Soc. Rev., 2013, 42, 291; (b) S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. E. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazeeruddin and M. Grätzel, Nat. Chem., 2014, 6, 242; (c) A. Yella, C.-L. Mai, S. M. Zakeeruddin, S.-N. Chang, C.-H. Hsieh, C.-Y. Yeh and M. Grätzel, Angew. Chem., Int. Ed., 2014, 53, 2973.
7. (a) P. Y. Reddy, L. Giribabu, C. Lyness, H. J. Snaith, C. Vijaykumar, M. Chandrasekharam, M. Lakshmikantam, J.-H. Yum, K. Kalyanasundaram, M. Grätzel and M. K. Nazeeruddin, Angew. Chem., Int. Ed., 2007, 46, 373; (b) S. Mori, M. Nagata, Y. Nakahata, K. Yasuta, R. Goto, M. Kimura and M. Taya, J. Am. Chem. Soc., 2010, 132, 4054.
8. (a) D. Kuang, P. Walter, F. Nuësch, S. Kim, J. Ko, P. Comte, S. M. Zakeeruddin, M. K. Nazeeruddin and M. Grätzel, Langmuir, 2007, 23, 10906; (b) J.-H. Yum, S.-R. Jang, P. Walter, T. Geiger, F. Nuësch, S. Kim, J. Ko, M. Grätzel and M. K. Nazeeruddin, Chem. Commun., 2007, 4680; (c) H. Li, Y. Wu, Z. Geng, J. Liu, D. Xu and W. Zhu, J. Mater. Chem. A, 2014, 2, 14649; (d) R. Agosta, R. Grisorio, L. De Marco, G. Romanazzi, G. P. Suranna, G. Gigli and M. Manca, Chem. Commun., 2014, 50, 9451.
9. (a) J.-J. Cid, J.-H. Yum, S.-R. Jang, M. K. Nazeeruddin, E. Martínez-Ferrero, E. Palomares, J. Ko, M. Grätzel and T. Torres, Angew. Chem., Int. Ed., 2007, 46, 8358; (b) M. Kimura, H. Nomoto, N. Masaki and S. Mori, Angew. Chem., Int. Ed., 2012, 51, 4371; (c) L. Yu, K. Fan, T. Duan, X. Chen, R. Li and T. Peng, ACS Sustainable Chem. Eng., 2014, 2, 718.
10. (a) T. Bessho, S. M. Zakeeruddin, C.-Y. Yeh, E. W.-G. Diau and M. Grätzel, Angew. Chem., Int. Ed., 2010, 49, 6646; (b) 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; (c) C.-M. Lan, H.-P. Wu, T.-Y. Pan, C.-W. Chang, W.-S. Chao, C.-T. Chen, C.-L. Wang, C.-Y. Lin and E. W.-G. Diau, Energy Environ. Sci., 2012, 5, 6460.
11. (a) R. Ogura, S. Nakane, M. Morooka, M. Orihashi, Y. Suzuki and K. Noda, Appl. Phys. Lett., 2009, 94, 073308; (b) S.-Q. Fan, C. Kim, B. Fang, K.-X. Liao, G.-J. Yang, C.-J. Li, J.-J. Kim and J. Ko, J. Phys. Chem. C, 2011, 115, 7747; (c) H. Ozawa, R. Shimizu and H. Arakawa, RSC Adv., 2012, 2, 3198; (d) L. Han, A. Islam, H. Chen, C. Malapaka, B. Chiranjeevi, S. Zhang, X. Yang and M. Yanagida, Energy Environ. Sci., 2012, 5, 6057; (e) H. Ozawa, Y. Okuyama and H. Arakawa, RSC Adv., 2013, 3, 9175; (f) S. Zhang, A. Islam, X. Yang, C. Qin, K. Zhang, Y. Numata, H. Chen and L. Han, J. Mater. Chem. A, 2013, 1, 4812; (g) C. Qin, Y. Numata, S. Zhang, A. Islam, X. Yang, K. Sodeyama, Y. Tateyama and L. Han, Adv. Funct. Mater., 2013, 23, 3782.
12. (a) Y. Ooyama, S. Inoue, T. Nagano, K. Kushimoto, J. Ohshita, I. Imae, K. Komaguchi and Y. Harima, Angew. Chem., Int. Ed., 2011, 50, 7429; (b) Y. Ooyama, T. Nagano, S. Inoue, I. Imae, K. Komaguchi, J. Ohshita and Y. Harima, Chem.–Eur. J., 2011, 17, 14837; (c) Y. Ooyama, N. Yamaguchi, I. Imae, K. Komaguchi, J. Ohshita and Y. Harima, Chem. Commun., 2013, 49, 2548; (d) Y. Harima, T. Fujita, Y. Kano, I. Imae, K. Komaguchi, Y. Ooyama and J. Ohshita, J. Phys. Chem. C, 2013, 117, 16364.
13. Y. Ooyama, T. Sato, Y. Harima and J. Ohshita, J. Mater. Chem. A, 2014, 2, 3293.
14. N. Shibayama, H. Ozawa, M. Abe, Y. Ooyama and H. Arakawa, Chem. Commun., 2014, 50, 6398.

---

Footnotes

† Electronic supplementary information (ESI) available: Details of experimental procedures. See DOI: 10.1039/c4ra14190c

‡ SAT-1: 4-(5-(benzo[4,5]thieno[2,3-c]pyridin-7-yl)thiophen-2-yl)-N,N-bis(4-ethoxyphenyl)aniline, NI-6: [9-butyl-7-(5-pyridin-4-yl-thiophen-2-yl)-9H-carbazol-2-yl]-diphenyl-amine, and YNI-2: 9-butyl-3,6-bis(5-(pyridin-4-yl)thiophen-2-yl)-9H-carbazole.

---