Lindqvist polyoxometalates as electrolytes in p-type dye sensitized solar cells

Abstract

The development of new redox couples provides a clear strategy to improve power conversion efficiency (PCE) in p-type dye-sensitized solar cells (p-DSSCs) through enabling improvements in open-circuit voltage (V_OC). In this work we report the use of molybdenum and tungsten containing Lindqvist polyoxometalates (POMs), [TBA]₂Mo₆O₁₉ and [TBA]₂W₆O₁₉, as an alternative to the traditionally used I⁻/I₃⁻ redox mediator in p-DSSCs. POM electrolytes are cheap and easy to synthesize, air stable and transparent, making them suitable for tandem solar cell applications. Electrolytes were evaluated using a simple testing device to benchmark the respective devices. Up to a 5-fold increase in V_OC was observed for p-DSSCs employing electrolytes with POMs upon comparison with traditionally used I₃⁻/I⁻, while short-circuit photocurrents with the same order of magnitude were observed under identical dilute conditions.

---

Introduction

With an ever increasing energy demand and the prospect of climate change as a consequence of greenhouse gas emission, alternatives to fossil fuels are needed. Solar energy has the most potential for providing this energy in a sustainable fashion.¹ Dye-sensitized solar cells (DSSCs) are considered as a promising low-cost alternative to traditional crystalline silicon solar cells.² Furthermore, they can be made from sustainable materials and can absorb diffuse light, making them suitable for small scale appliances.^3,4

n-Type DSSCs based on mesoporous TiO₂ have achieved an efficiency of 11.9% with traditional iodide electrolytes, with greater power conversion efficiencies (PCEs) made possible through the use of alternative redox couples, enabling PCE improvements through increases in open-circuit voltage (V_OC) without sacrificing short-circuit photocurrent (J_SC).^5–8 Complementary improvements in p-type DSSCs (p-DSSC), based on mesoporous NiO semiconductors, aim to enable the fabrication of highly efficient tandem DSSC devices. However, p-DSSCs exhibit much lower PCEs than n-type DSSCs.^9–12

Besides engineering the mesoporous semiconductor and dye of the p-DSSC, the development of the redox mediator within the electrolyte of the p-DSSC provides a clear pathway to device optimization, as evidenced by several groups using cobalt pyridyl complexes,^13,28 disulfides/thiolates,¹⁴ cobalt diaminoethane complexes¹⁵ and iron acetylacetonate complexes.¹⁶ Traditionally used I₃⁻/I⁻ couples have been scrutinized due to their ambiguous mechanism as an electrolyte and parasitic absorption of visible light. The absorption of visible light by the I₃⁻/I⁻ couple removes photons utilizable by the dye, reducing J_SC, while the radical I₂˙⁻ formed upon visible light excitation, recombines with a photogenerated hole at the semiconductor–electrolyte interface,¹⁷ compromising the V_OC of the device. Importantly, devices based on I₃⁻/I⁻ electrolytes suffer from low V_OC, as their redox potential is not energetically matched in an optimal fashion with the optoelectronic properties of the dye. The redox potential of the I₃⁻/I⁻ electrolyte also limits its solar fuel applications, to drive for example CO₂-reduction catalysts.

Bach et al. demonstrated an elegant solution to improving V_OC in p-DSSCs by employing [Co(en)₃]^3+/2+ and [Fe(acac)₃]^0/1− redox mediators that exhibit higher redox potentials than I₃⁻/I⁻ couples, enabled by appropriate matching of dye electronics, leading to higher PCEs.^13–16 It is important to note that the increase in redox potential is solely beneficial for p-type DSSCs and not for tandem cells. The cobalt and iron based electrolytes are however unstable in air and therefore have to be prepared under inert conditions, complicating device fabrication. Also, the cobalt and iron based electrolytes compete for visible light absorption with the dyes, making them less suitable for tandem solar cell applications.

Polyoxometalates (POMs) are well-known anionic metal oxide clusters that can host several heteroatoms.¹⁸ POMs are widely used because of their electrochemical properties. Recent applications included water oxidation catalysts, as components in redox flow batteries, lithium ion batteries and super capacitors.^19–22 POMs have also been used as co-adsorbents on NiO surfaces in p-type DSSCs, which showed a decreased recombination rate and increased V_OC.²³ To our knowledge, POMs have not been employed as redox mediators in p-DSSCs.

In this work we report the use of Lindqvist polyoxometalates (POMs) as electrolytes for p-type DSSCs. Lindqvist POMs have a molecular formula M₆O₁₉²⁻ and spherical shape. As we are particularly interested in the development of p-type DSSCs with higher V_OC for the realization of solar fuel devices, a simple testing device was constructed to enable rapid optimization (Fig. 6).

Results and discussion

The Lindqvist POMs used in this work are [TBA]₂Mo₆O₁₉ and [TBA]₂W₆O₁₉. The structure of a typical Lindqvist POM is depicted in Fig. 1.

Fig. 1 Typical Lindqvist M₆O₁₉²⁻ polyoxometalate with the metal atoms depicted in orange and the oxygen atoms depicted in red.

The POMs were synthesized via two simple procedures as previously described by Fournier and Wang et al., with the characterization of the POM structure achieved by single crystal X-ray diffraction.²⁴ The POMs have been made from cheap starting materials, employing abundant elements coupled with air stability, making them amenable to scale-up. The [TBA]₂Mo₆O₁₉ and [TBA]₂W₆O₁₉ POMs demonstrated respective solubilities of 21 g L⁻¹ and 67 g L⁻¹ in acetonitrile, limiting their application in DSSCs. The acetonitrile solution of [TBA]₂Mo₆O₁₉ afforded a faintly yellow solution and [TBA]₂W₆O₁₉ yielded a clear solution. As shown in Fig. 2, the molar absorption coefficients of both POMs are over 100 times lower than that of the traditional I₃⁻/I⁻ electrolyte.

Fig. 2 Extinction coefficient spectra of [TBA]₂W₆O₁₉, [TBA]₂W₆O₁₉ and I₃⁻/I⁻ in acetonitrile, with the inset showing the difference between of [TBA]₂W₆O₁₉ and [TBA]₂W₆O₁₉. Concentration of [TBA]₂W₆O₁₉, [TBA]₂W₆O₁₉ and I₃⁻/I⁻ was 1.7 × 10⁻⁵ M, 3.5 × 10⁻⁵ M and 1.7 × 10⁻⁷ M for [TBA]₂W₆O₁₉, [TBA]₂W₆O₁₉ and I₃⁻/I⁻ respectively.

A combination of cyclic voltammetry and differential pulse voltammetry was employed to determine the redox potential, reversibility and diffusion coefficient of both POMs. Fig. 3 shows the differential pulse voltammogram of both POMs, which exhibit respective redox potentials of −0.40 V and −0.90 V (vs. NHE) for [TBA]₂Mo₆O₁₉ and [TBA]₂W₆O₁₉. The experiments were performed under aerobic conditions, showing the air stability of the POMs. The broad peak separation in the cyclic voltammetry experiments and the bell shaped curve in the differential pulse voltammogram experiments can be explained by the fact that no supporting electrolyte was used and the experiments were done in acetonitrile (see the ESI†).

Fig. 3 Differential pulse voltammogram of [TBA]₂Mo₆O₁₉ and [TBA]₂Mo₆O₁₉ saturated solution in acetonitrile using a glassy carbon working electrode, a platinum counter electrode and a Ag/AgCl reference electrode.

Given that the redox potential of I₃⁻/I⁻, [Co(en)₃]^3+/2+ and [Fe(Acac)₃]^0/1− is reported to be +0.32 V, −0.03 V and −0.20 V vs. NHE respectively,¹⁶ the [TBA]₂W₆O₁₉ POM demonstrates the lowest redox potential to our knowledge. The low redox potential of [TBA]₂W₆O₁₉ should translate into an elevated V_OC in the p-DSSC device, making the POMs an excellent candidate as a redox mediator. Fig. 4 shows an energy level diagram of the components in the p-type DSSCs used in this work. The LUMO of the P1 dye has been reported to be around −0.93 V vs. NHE.^11,25 This means that there is 0.03 V available for electron transfer from the dye to the redox mediator in the case of [TBA]₂W₆O₁₉. Although this should be enough energy to regenerate the [TBA]₂W₆O₁₉ POM, a low J_SC can be expected as the driving force is very small, reducing the rate of this electron transfer process. In the case of [TBA]₂Mo₆O₁₉ there is 0.53 V of driving force available, which should afford lower V_OC values than those of the [TBA]₂W₆O₁₉ POM, but higher J_SC values.

Fig. 4 Energy-level diagram of the semiconductor, P1-dye and redox couples used in this work.

Cyclic voltammetry presented in Fig. 5 demonstrates that both POMs are electrochemically reversible. The scan-rate dependent measurements provide information on the diffusion constants of the different POMs.¹⁶ Using the Randles–Sevcik equation, the electrochemical data show that [TBA]₂Mo₆O₁₉ and [TBA]₂W₆O₁₉ have a diffusion coefficient of 2.02 (±0.35) × 10⁻⁷ cm² s⁻¹ and 1.08 (±0.09) × 10⁻⁷ cm² s⁻¹ respectively. Earlier studies with the [Fe(acac)₃]^0/1− electrolyte revealed that this species demonstrates a diffusion coefficient of 7.1 (±0.78) × 10⁻⁶ cm² s⁻¹.¹⁶ It was also reported that the J_SC was limited by diffusion, a well-known phenomenon observed when employing redox mediators with greater size.¹⁶ Since the POMs in this work feature lower diffusion coefficients, it is likely that the J_SC of the corresponding p-DSSCs will experience diffusion limitations.

Fig. 5 Scan rate dependent cyclic voltammetry with a glassy carbon working electrode, a platinum counter electrode and a Ag/AgCl reference electrode of (a) [TBA]₂Mo₆O₁₉ and (b) [TBA]₂Mo₆O₁₉. Conversion of potential vs. Ag/AgCl to the NHE is given in the ESI.†

We designed and fabricated a simple device to specifically measure the V_OC in a simple and expedient manner. The device is made from a screen-printed NiO photocathode (1.57 cm²) on FTO glass, subsequently sensitized with the frequently used P1 dye.^25–27 As depicted in Fig. 6, the counter electrode was made from platinum coated FTO. In the device the two electrodes are separated by a 1 cm distance by a Teflon container, featuring a large opening at the top to facilitate the introduction and removal of electrolyte solutions, and washing of the system. The Teflon container has a volume of 3 mL of electrolyte solution. It is expected that the increased distance between the electrodes will amplify diffusion limitations in the system, leading to lower J_SC, but it allows the benchmarking of the V_OC of various devices in an expeditious manner.²⁹

Fig. 6 Simple testing device used in this work to benchmark the open current voltage. Design schematics of the device are shown in the ESI.†

Table 1 summarizes the photovoltaic parameters obtained from the measurement of the J–V properties of the DSSCs. The POM electrolytes are compared with the I₃⁻/I⁻ electrolyte at a reduced concentration (0.01 M as opposed to 1.0 M) due to the limited solubility of the POM materials in MeCN. Interestingly, devices based on the POM electrolytes (entries 1 & 2) show a dramatic increase in V_OC compared the I₃⁻/I⁻ couple (entry 3), corresponding clearly to the differences in redox potentials measured from electrochemical characterization. Also, the elevated V_OC of [TBA]₂W₆O₁₉ compared to [TBA]₂Mo₆O₁₉ is consistent with the slightly higher oxidation potential of [TBA]₂W₆O₁₉. Comparing the two POM-electrolyte devices, the lower J_SC of the [TBA]₂W₆O₁₉ device (entry 1) can be explained by either the smaller driving force for the reduction of [TBA]₂W₆O₁₉ by the photoreduced dye compared to [TBA]₂Mo₆O₁₉ or the greater diffusion coefficient. Given that the J_SC values of all devices have the same order of magnitude when low concentration redox mediators are utilized, the POM materials are promising candidates as electrolytes for p-type DSSCs. Wu et al. showed the impact of the electrolyte recombination resistance on the fill factor.³⁰ It is likely that this is also the cause of the low fill factor in this system. The results in Table 1 also show the effect of increasing the concentration of the electrolyte (entries 3 and 4), when using the I₃⁻/I⁻ couple. Both the V_OC and the short-circuit current increase when the concentration is higher. Since the solubility of the POMs in acetonitrile is low, similar experiments with these electrolytes are not possible. Clearly, more soluble analogues are required for practical applications.

Table 1 Photovoltaic parameters of the p-type DSSCs prepared in this work. Data were obtained from I–V curve measurements performed under AM1.5 solar irradiation (100 mW cm⁻²)

Entry	Electrolyte	V OC (mV)	J SC (mA cm^−2)	FF
a 0.5 M of LiTFSI was present as the supporting electrolyte. The data shown are the average and the standard deviation of at least 5 different experiments.
1	0.01 M [TBA]2Mo6O19^a	423 ± 28	0.021 ± 0.002	0.13 ± 0.02
2	0.01 M [TBA]2W6O19^a	541 ± 34	0.010 ± 0.001	0.11 ± 0.02
3	0.01 M I3^−/I^−^a	100 ± 5	0.032 ± 0.003	0.23 ± 0.01
4	1.0 M I3^−/I^−	144 ± 14	0.107 ± 0.013	0.31 ± 0.02
5	1.0 M I3^−/I^− (literature)^26	106	3.010	0.37
6	Co(tbpy)3 (literature)^27	80	0.26	0.26

---

The effectiveness of the simple test cell is clear when comparing our results with those reported in the literature. The V_OC obtained for the test cell are similar to those reported in the literature (entries 4 and 5), and only the J_SC is lower as a result of the large distance between the electrodes. Given the fact that it is really easy to change the electrolyte solution, the test cell is very suitable for benchmarking V_OC.

Conclusions

We have shown that polyoxometalates can be used as electrolytes in p-type dye sensitized solar cells. The POMs presented in this work are easy to prepare based on cheap starting materials. A unique feature of these POMs as electrolytes is that they are transparent and can therefore be suitable for tandem solar cell applications. An increased V_OC is not directly beneficial to tandem solar cell applications, but may be useful for pure p-type solar cells. The POMs are also air stable, allowing for easy preparation of the electrolyte and the DSSC. We also report a simple testing setup that allows us to benchmark the open current potential for different devices thereby facilitating the optimization of electrolyte solutions.

In line with what we expected on the basis of the redox potentials of the POM materials, measured by electrochemistry, we found a 4 to 5-fold increase in V_OC when applying these POM electrolytes in comparison to the devices with the traditional I₃⁻/I⁻ couple. We plan to exploit these voltages in proton and CO₂ reduction for solar fuel applications. The high V_OC is sufficient to drive proton reduction catalysis and CO₂ reduction catalysis. The J_SC have the same order of magnitude under the same conditions. Currently, the low solubility of the POMs is the main hampering factor for increasing the short-circuit current of these cells. Increasing the solubility of these POM materials should therefore be the main focus point when improving the efficiency of these cells. After solubility issues are fixed, traditional DSSCs can be made in order to achieve record efficiencies.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors would like to thank Merck GmbH and the Dutch National Science Foundation (NWO) for funding. The authors would like to give special thanks to Dr J. I. Van der Vlugt for performing the SXRD measurements, Dr R. Detz for fruitful discussions and L. van den Burg for the synthesis of the P1 dye.

References

1. M. Grätzel, Nature, 2001, 414, 338–344.
2. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740.
3. A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Petterson, Chem. Rev., 2014, 110, 4474–4490.
4. F. Bella, S. Galliano, M. Falco, G. Viscardi, C. Barolo, M. Grätzel and C. Gerbaldi, Green Chem., 2017, 19, 1043–1051.
5. R. Komiya in Technical Digest, 21st International Photovoltaic Science and Engineering Conference, 2011, 2 C-5O-08.
6. K. Kakiage, Y. Aoyama, T. Yano, K. Oya, J. Fujisawa and M. Hanaya, Chem. Commun., 2015, 51, 15894–15897.
7. Y. Cao, Y. Liu, S. Zakeeruddin, A. Hagfeldt and M. Grätzel, Joule, 2018, 6, 1108–1117.
8. Q. Yu, Y. Wang, Z. Yi, N. Zu, J. Zhang, M. Zhang and P. Wang, ACS Nano, 2010, 4, 6032–6038.
9. Y. Pellegrin, L. Le Pleux, E. Blart, A. Renaud, B. Chavillon, N. Szuwarski, M. Boujita, L. Cariob, S. Jobic, D. Jacquemina and F. Odobel, J. Photochem. Photobiol., A, 2011, 219, 235–242.
10. M. Yu, G. Natu, Z. Ji and Y. Wu, J. Phys. Chem. Lett., 2012, 3, 1074–1078.
11. P. Qin, J. Wiberg, E. A. Gibson, M. Linder, L. Li, T. Brinck, A. Hagfeldt, B. Albinsson and L. Sun, J. Phys. Chem. C, 2010, 114, 4738–4747.
12. A. Nattestad, A. J. Mozer, M. K. R. Fischer, Y. Cheng, A. Mishra, P. Bäuerle and U. Bach, Nat. Mater., 2010, 9, 31–35.
13. E. A. Gibson, A. L. Smeigh, L. Le Pleux, L. Hammarström, F. Odobel, G. Boschloo and A. Hagfeldt, J. Phys. Chem. C, 2011, 155, 9772–9779.
14. X. Xu, B. Zhang, J. Cui, D. Xiong, Y. Shen, W. Chen, L. Sun, Y. Chengad and M. Wang, Nanoscale, 2013, 5, 7963–7969.
15. S. Powar, T. Daeneke, M. T. Ma, D. Fu, N. W. Duffy, G. Götz, M. Weidelener, A. Mishra, P. Bäuerle, L. Spiccia and U. Bach, Angew. Chem., 2013, 125, 630–633.
16. I. R. Perera, T. Daeneke, S. Makuta, Z. Yu, Y. Tachibana, A. Mishra, P. Bäuerle, C. A. Ohlin, U. Bach and L. Spiccia, Angew. Chem., Int. Ed., 2015, 54, 3758–3762.
17. F. Odobel, Y. Pellegrin, E. A. Gibson, A. Hagfeldt, A. L. Smeigh and L. Hammarström, Coord. Chem. Rev., 2012, 256, 2414–2423.
18. T. Ueda, ChemElectroChem, 2018, 5, 823–838.
19. Y. V. Geletii, B. Botar, P. Kçgerler, D. A. Hillesheim, D. G. Musaev and C. L. Hill, Angew. Chem., Int. Ed., 2008, 47, 3896–3899.
20. H. D. Pratt III and T. M. Anderson, Dalton Trans., 2013, 42, 15650–15655.
21. J. J. Chen, M. D. Symes, S. C. Fan, M. S. Zheng, H. N. Miras, Q. F. Dong and L. Cronin, Adv. Mater., 2015, 27, 4649–4654.
22. P. Gómez-Romero, M. Chojak, K. Cuentas-Gallegos, J. A. Asensio, P. J. Kulesza, N. Casañ-Pastor and M. Lira-Cantffl, Electrochem. Commun., 2003, 5, 149–153.
23. H. El Moll, F. A. Black, C. J. Wood, A. Al-Yasari, M. Reddy Marri, I. V. Sazanovich, E. A. Gibson and J. Fielden, Phys. Chem. Chem. Phys., 2017, 19, 18831–18835.
24. R.-C. Wang and M. Fournier, Inorg. Synth., 1990, 27, 85–96.
25. P. Qin, H. Zhu, T. Edvinsson, G. Boschloo, A. Hagfeldt and L. Sun, J. Am. Chem. Soc., 2008, 130, 8570–8571.
26. 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.
27. N. Li, E. A. Gibson, P. Qin, G. Boschloo, M. Gorlov, A. Hagfeldt and L. Sun, Adv. Mater., 2010, 22, 1759–1762.
28. E. A. Gibson, A. L. Smeigh, L. Le Pleux, J. Fortage, G. Boschloo, E. Blart, Y. Pellegrin, F. Odobel, A. Hagfeldt and L. Hammarström, Angew. Chem., 2009, 48, 4402–4405.
29. A. Yella, S. Mathew, S. Aghazada, P. Comte, M. Grätzel and M. Nazeeruddin, J. Mater. Chem. C, 2017, 5, 2833.
30. Z. Huang, G. Natu, Z. Ji, M. He, M. Yu and Y. Wu, J. Phys. Chem. C, 2012, 116, 26239–26246.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis details, materials and methods. See DOI: 10.1039/c8se00495a

---