Efficient aqueous dye-sensitized solar cell electrolytes based on a TEMPO/TEMPO⁺ redox couple

Abstract

Aqueous electrolyte-based dye-sensitized solar cells (DSSCs) have recently emerged and shown to be a promising eco-friendly photovoltaic device. In the present study, we, for the first time, have developed 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) and TEMPO⁺ tetrafluoroborate salt as a redox couple in an aqueous electrolyte for DSSCs. With the hydrophobic dye LEG4 as a light absorber, we have achieved a power conversion efficiency of 4.3% and a record open circuit voltage of 955 mV in the device. This is attributed to the high formal redox potential of TEMPO/TEMPO⁺ (0.71 V vs. NHE) in water. In addition, despite the wide use of surfactants in previous studies, we have clearly shown that the addition of surfactants to the electrolyte is detrimental to solar cell performance. Therefore, the use of surfactants in aqueous DSSC electrolytes should be avoided or used with caution.

---

Mesoporous dye sensitized solar cells (DSSCs) have attracted considerable attention as a potential alternative to conventional inorganic photovoltaics for more than 20 years.^1,15 Countless research efforts have eventually led to the current 13% record efficiency in DSSCs.² The presence of acetonitrile (ACN) as the electrolyte solvent is, however, problematic for long-term stability and environmental reasons because it has well-known limitations such as flammability, toxicity and volatility. Continuous efforts have been made to explore alternative solvents, including polymer-gel³ and ionic liquid electrolytes,^4,5 but, to date, the most environmentally friendly option, water, has been avoided because of the vague consideration that it is detrimental for solar cells performance and stability.⁶ However, recent studies indicate that water-based DSSCs could actually reach an appreciable efficiency.^7–14 Redox couples previously applied in non-aqueous electrolyte DSSCs have been successfully developed for water-based systems, such as I⁻/I₃⁻,^9–11,14 ferricyanide/ferrocyanide,¹² and Co(bpy)₃^2+/3+.^8,13 Moreover, new redox couples⁷ and dyes^7,11,14 have been specifically designed to achieve functional water-based redox electrolytes for DSSCs. However, despite the achieved progress, all the current aqueous DSSCs still suffer from a low voltage and poor stability compared to its counterpart in non-aqueous electrolytes.

In DSSCs, one of the main energy losses is due to the dye regeneration by the redox species in the electrolyte, which is determined by the HOMO level of the dye molecules and the redox potential of the redox species in the electrolyte. Therefore, it is desirable to further explore redox couples with a more positive redox potential in aqueous electrolytes, in order to reduce this energy loss and eventually increase the device performance. Furthermore, the effects of surfactants, which were frequently used in previous studies,^7,8,10–13 are still not clearly understood and make the system different from non-aqueous systems.

Herein, we, for the first time, report the use of TEMPO (see Fig. 1a) and TEMPO⁺ as a redox couple in aqueous electrolytes for DSSCs. The TEMPO/TEMPO⁺ redox couple is known for its fast kinetics,^16,17 and more importantly it has a high positive redox potential, which is essential for achieving a high open-circuit voltage (V_oc) and efficiency in the aqueous system. In addition, the effects of surfactants on the solar cell performance and its origin were also investigated and are discussed here. These findings will be important for the future development of efficient environmentally friendly aqueous DSSCs.

Fig. 1 (a) The structure of LEG4 and TEMPO/TEMPOBF₄, (b) current–voltage curves of solar cells with different amount of SDS in the electrolytes (0, 2, 6 wt%) under standard AM 1.5G, 100 mW cm⁻² illumination. Inset graphs are the corresponding IPCE results. The detailed device fabrication and characterization setup is given in ESI.† The electrolyte composition was 0.15 M TEMPO, 0.05 M TEMPOBF₄ (synthesized according to literature²¹), 0.1 M LiClO₄, 0.2 M 1-methylbenzimidazole (NMBI), and indicated SDS amount (wt%) in water.

The formal redox potential of TEMPO/TEMPO⁺ in water was determined to be 0.71 V vs. NHE by cyclic voltammetry, which is slightly more negative than the value determined in acetonitrile (ACN) i.e., 0.88 V (ref. 18) vs. NHE (Fig. S1†). The diffusion coefficient was determined to be 4.4 × 10⁻⁶ cm² s⁻¹ (see inset graph in Fig. S1†), which is comparable to that of cobalt complexes in acetonitrile in the reported DSSCs.^2,19 This suggests that the diffusion coefficient is sufficient and that mass transport should not lead to current limitations.

DSSCs were fabricated using the hydrophobic dye LEG4 (ref. 20) (Fig. 1a) as a sensitizer and the TEMPO electrolyte with the addition of different concentrations (wt%) of the surfactant sodium dodecyl sulfate (SDS). The typical solar cell current–voltage characteristic curves are shown in Fig. 1b together with the incident photon to current efficiency (IPCE) as an inset. The averaged solar cell parameters are summarized in Table 1. Without the use of any surfactants, solar cells exhibit power conversion efficiencies (PCE) as high as 4.3%, which is among the best values reported for DSSCs with aqueous electrolytes.^7–14 Especially, a high V_oc above 940 mV was achieved for all the investigated solar cells, which is remarkably 100 mV higher than all the previously reported aqueous systems.^7–14 The high V_oc is due to the high positive redox potential of the TEMPO redox couple, compared to 0.12 V for TT⁻/DTT,⁷ 0.42 V for Co(bpy)₃^2+/3+,⁸ 0.54 V for I⁻/I₃⁻ (ref. 22) and 0.47 V for ferricyanide/ferrocyanide²³ in water. The low short-circuit photocurrent density (J_sc) and IPCE (maximum at 40% shown in the inset graph of Fig. 1b), are attributed to a significant recombination between TEMPO⁺ and electrons injected into TiO₂, as shown in previous studies.^17,18 The good fill factor (FF) was related to the high heterogeneous electron transfer constant of TEMPO, which indicates the fast regeneration reaction of the mediator on the counter electrode, as previously shown by Hiroyuki Nishide et al.¹⁶

Table 1 Solar cell performance from current–voltage measurements of DSSCs with the three aqueous TEMPO electrolytes in this study

	Voc, mV	Jsc, mA cm^−2	Fill factor	PCE, %
0 wt% SDS	955 ± 5	5.78 ± 0.35	0.75 ± 0.00	4.14 ± 0.23
2 wt% SDS	948 ± 2	5.04 ± 0.12	0.73 ± 0.01	3.48 ± 0.05
6 wt% SDS	945 ± 5	4.09 ± 0.17	0.72 ± 0.01	2.82 ± 0.09

---

In previous studies, the addition of surfactants was shown to be critical to the performance of aqueous DSSC.^10,11 It led to improvements that were attributed to the increased solubility of a phase separated compound,¹¹ and more importantly to the improved wetting behavior of electrolytes and dye molecules.¹⁰ However, in the present study, despite the use of the hydrophobic dye LEG4, the addition of SDS decreased the V_oc, J_sc and FF. Eventually, it resulted in 16% and 30% efficiency losses with 2 and 6 wt% SDS in the electrolyte, respectively. The slight decrease of V_oc with the addition of SDS was caused by an extra introduction of Na⁺, which is known as a potential-dependent cation.^24–26 Charge extraction measurements under an open circuit situation (shown in Fig. S2(a)†) confirmed that with the addition of SDS in the electrolytes more charges were extracted under the same V_oc. This indicates a lower conduction band edge with more SDS in the electrolytes, with the assumption that the trap distribution in TiO₂ does not change with the addition of Na⁺.²⁴ The almost identical electron lifetime under same charge extraction amount (shown in Fig. S2(b)†) could exclude electron lifetime effects on V_oc.

We further studied the effects of surfactant on J_sc through linear potential sweep voltammetry and impedance measurements on symmetrical cells consisting of two counter electrodes and an electrolyte in between.^27,28 Fig. 2a shows that the increase in SDS concentration results in a decrease in diffusion limiting current (13, 8 and 5 mA cm⁻² for 0, 2 and 6 wt% SDS, respectively) in linear sweep voltammetry. The apparent concentration of limiting redox species (TEMPO⁺ in our case) is related to the limiting current according to equation: j_c = 2nFcD/δ,^27,28 where j_c is the diffusion limiting current, n is the number of exchanged electrons, c is the concentration of limiting species, F is Faraday's constant, D is the diffusion coefficient and δ is the diffusion length. If we consider constant diffusion coefficients with the addition of SDS, then it implies that the effective TEMPO⁺ concentration decreased 2 times for 6 wt% SDS electrolytes compared to 0 wt%. The limitation in the diffusion process with the addition of SDS in electrolytes was further confirmed by impedance measurements.

Fig. 2 Electrochemical tests on symmetrical cells. The cells were constructed with two platinized electrodes with 50 μm spacer and the electrolytes shown in Fig. 1. (a) Linear potential sweep voltammogram of the abovementioned cells. Scan range: −0.7 V to 0.7 V; scan rate: 10 mV s⁻¹. (b) Impedance measurements were conducted at an open circuit voltage from 65 kHz to 0.05 Hz at 0 V. The symbols are from experimental data while the solid lines are from the fitting results according to equivalent circuit model shown in the graph. The upper and below graphs show Nyquist and Bode plots, respectively.

Fig. 2b shows the Nyquist and Bode plots of the investigated symmetrical cells, in which the second semicircle was assigned to a Warburg diffusion process. The fitted Warburg resistance using the equivalent circuit model shown in Fig. 2b was determined to be 11.5, 20.4 and 33 Ω for solar cells with 0, 2 and 6 wt% SDS (within 5% fitting error). The three times higher Warburg diffusion resistance for 6 wt% SDS compared to 0 wt% confirmed a stronger diffusion limitation with increased surfactant amount, and it is in good consistence with the results of the linear sweep voltammetry. The explanation for the increased diffusion resistance could be the formation of micelles (with critical micelle concentration (CMC) of SDS reported between 0.06 wt% to 0.23 wt% (ref. 29)) in the electrolyte. The micelles would encapsulate a part of the redox species, especially those with limited solubility in water, and therefore reduce the effective redox species concentration. As a result, the limiting current density for 6 wt% SDS was determined to be only 5 mA cm⁻². Therefore, one could conclude that an even lower current would be the case in real solar cell setups because the presence of the mesoporous TiO₂ will further reduce the limiting current.¹⁰ Another origin could be the increased electrolyte viscosity caused by the addition of SDS in the electrolyte, which was attributed to a transformation in the shape of SDS micelles.³⁰ The abovementioned results indicate that, despite the suggested importance of surfactants in aqueous DSSCs,^10,11 the formation of micelles in water electrolytes can have negative effects on the diffusion process of redox species. This will be more crucial for redox species with low solubility, which is unfortunately often the case for water.^7,8,11,13

The electrochemical properties of TEMPO/TEMPO⁺ in water were studied by differential pulse voltammetry, and they appeared to be quite stable (Fig. S3†). However, the TEMPO solar cells investigated here have poor stability, similar to previously studied aqueous DSSCs.^9–11 The unstable nature of aqueous DSSCs require further investigation in order to develop competitive aqueous DSSCs.

In conclusion, we have developed a new TEMPO/TEMPO⁺ redox couple based aqueous electrolyte for DSSCs. The high positive redox potential of TEMPO/TEMPO⁺ (0.71 V vs. NHE) results in a remarkable high photovoltage (>950 mV) and efficiency (4.3%) in aqueous DSSCs. This confirms the benefits of further exploring redox couples with a high positive redox potential in order to reduce the energy loss during the dye regeneration process. The versatile structure of TEMPO-based derivatives has opened up the possibility for the further development of stable and high performing aqueous redox species. The previously used surfactants, which have been used extensively, were found to impose a diffusion limitation of the redox couple, which led to a significant decrease in the DSSC performance. Therefore, caution should be taken when using surfactants in aqueous electrolytes.

Acknowledgements

The authors acknowledge the Swedish National Research Council, the Swedish Energy Agency as well as the China Scholarship Council (CSC) for their financial support.

Notes and references

1. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740.
2. 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, 1–6.
3. P. Wang, S. M. Zakeeruddin, J. E. Moser, M. K. Nazeeruddin, T. Sekiguchi and M. Grätzel, Nat. Mater., 2003, 2, 402–407.
4. P. Wang, S. M. Zakeeruddin, J.-E. Moser, R. Humphry-Baker and M. Grätzel, J. Am. Chem. Soc., 2004, 126, 7164–7165.
5. Y. Bai, Y. Cao, J. Zhang, M. Wang, R. Li, P. Wang, S. M. Zakeeruddin and M. Grätzel, Nat. Mater., 2008, 7, 626–630.
6. Y. Liu, A. Hagfeldt, X.-R. Xiao and S.-E. Lindquist, Sol. Energy Mater. Sol. Cells, 1998, 55, 267–281.
7. H. Tian, E. Gabrielsson, P. W. Lohse, N. Vlachopoulos, L. Kloo, A. Hagfeldt and L. Sun, Energy Environ. Sci., 2012, 5, 9752.
8. W. Xiang, F. Huang, Y.-B. Cheng, U. Bach and L. Spiccia, Energy Environ. Sci., 2013, 6, 121.
9. C. Law, S. C. Pathirana, X. Li, A. Y. Anderson, P. R. F. Barnes, A. Listorti, T. H. Ghaddar and B. C. O'Regan, Adv. Mater., 2010, 22, 4505–4509.
10. C. Law, O. Moudam, S. Villarroya-Lidon and B. O'Regan, J. Mater. Chem., 2012, 22, 23387.
11. H. Choi, B.-S. Jeong, K. Do, M. J. Ju, K. Song and J. Ko, New J. Chem., 2013, 37, 329.
12. T. Daeneke, Y. Uemura, N. W. Duffy, A. J. Mozer, N. Koumura, U. Bach and L. Spiccia, Adv. Mater., 2012, 24, 1222–1225.
13. C. Dong, W. Xiang, F. Huang, D. Fu, W. Huang, U. Bach, Y.-B. Cheng, X. Li and L. Spiccia, Angew. Chem., Int. Ed., 2014, 53, 6933–6937.
14. V. Leandri, H. Ellis, E. Gabrielsson, L. Sun, G. Boschloo and A. Hagfeldt, Phys. Chem. Chem. Phys., 2014, 16, 19964–19971.
15. A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson, Chem. Rev., 2010, 110, 6595–6663.
16. F. Kato, A. Kikuchi, T. Okuyama, K. Oyaizu and H. Nishide, Angew. Chem., Int. Ed., 2012, 51, 10177–10180.
17. Z. Zhang, P. Chen, T. N. Murakami, S. M. Zakeeruddin and M. Grätzel, Adv. Funct. Mater., 2008, 18, 341–346.
18. W. Yang, N. Vlachopoulos, A. Hagfeldt and G. Boschloo, unpublished work.
19. S. M. Feldt, E. a. Gibson, E. Gabrielsson, L. Sun, G. Boschloo and A. Hagfeldt, J. Am. Chem. Soc., 2010, 132, 16714–16724.
20. E. Gabrielsson, H. Ellis, S. Feldt, H. Tian, G. Boschloo, A. Hagfeldt and L. Sun, Adv. Energy Mater., 2013, 3, 1647–1656.
21. M. a Mercadante, C. B. Kelly, J. M. Bobbitt, L. J. Tilley and N. E. Leadbeater, Nat. Protoc., 2013, 8, 666–676.
22. G. Boschloo and A. Hagfeldt, Acc. Chem. Res., 2009, 42, 1819–1826.
23. D. C. Thornton, K. T. Corby, V. a. Spendel, J. Jordan, A. Robbat, D. J. Rutstrom, M. Gross and G. Ritzler, Anal. Chem., 1985, 57, 150–155.
24. H. Wang and L. M. Peter, J. Phys. Chem. C, 2012, 116, 10468–10475.
25. K. Fredin, J. Nissfolk, G. Boschloo and A. Hagfeldt, J. Electroanal. Chem., 2007, 609, 55–60.
26. D. F. Watson and G. J. Meyer, Coord. Chem. Rev., 2004, 248, 1391–1406.
27. H. Ellis, N. Vlachopoulos, L. Häggman, C. Perruchot, M. Jouini, G. Boschloo and A. Hagfeldt, Electrochim. Acta, 2013, 107, 45–51.
28. L. Kavan, J.-H. Yum and M. Grätzel, Nano Lett., 2011, 11, 5501–5506.
29. E. Fuguet, C. Ràfols, M. Rosés and E. Bosch, Anal. Chim. Acta, 2005, 548, 95–100.
30. K. Chari, B. Antalek, M. Y. Lin and S. K. Sinha, J. Chem. Phys., 1994, 100, 5294–5300.

---

Footnote

† Electronic supplementary information (ESI) available: Details of solar cell fabrication and characterization; cyclic voltammogram of TEMPO at different scan rates; charge extraction and electron lifetime measurements with 0, 2 and 6 wt% SDS in electrolytes; differential pulsed voltammetry (DPV) of the tested TEMPOBF₄ stored in the dark. See DOI: 10.1039/c5ra03248b

---