The first dye-sensitized solar cell with p-type LaOCuS nanoparticles as a photocathode

Abstract

Layered LaOCuS oxysulfide is a well-known wide band gap p-type semiconductor that has attracted strong interests for transparent electronics. We report here that nanoparticles of this material can also be used as a substitute for the widely used NiO compound to fabricate photocathodes for p-type dye sensitized solar cells.

---

Transparent conducting oxides (TCOs) have an exceptional propensity to couple two antagonist properties: a high electrical conductivity and a high optical transparency. Such characteristics open up the door to numerous optoelectronic applications such as electrochromic glasses, touchscreen, anti-glare, flat panel displays, anti-static shields, low emissivity glasses, defrosting windows and photovoltaics.¹ Commonly used TCOs are n-type materials belonging to the ZnO–SnO₂–In₂O₃–Ga₂O₃–CdO classes of compounds. They have successfully reached the level of industrial development, as they can be found in many devices in everyday life. Nevertheless, new applications for optoelectronics require p-type TCOs such as CuAlO₂,² CuGaO₂,³ SrCu₂O₂,⁴ which have recently emerged in the literature. These p-type metal oxides are also candidates for a new generation of dye-sensitized solar cells (p-DSSCs), which are much less investigated than conventional Grätzel cells (Fig. 1).⁵ The most used and most investigated p-type semiconductor (p-SC) for this technology is undoubtedly NiO, while recently, doped and undoped CuAlO₂,⁶ CuGaO₂,⁷ CuCrO₂,⁸ and NiCo₂O₄ ⁹ have been proposed to replace it. p-DSSCs are based on a reverse operation principle to the Grätzel cell, in which the photoanode (consisting of an n-type semiconductor such as TiO₂,¹⁰ ZnO¹¹) is replaced by a photocathode made of a dye-sensitized p-type semiconductor (p-SC, e.g. NiO¹²). Upon light excitation, a hole is injected from the dye excited state (S*) into the valence band of the p-SC and the reduced dye (S⁻) donates its electron to a redox shuttle, which diffuses to the Pt counter electrode (Fig. 1). p-DSSCs still have quite low performances compared to Grätzel cells, with a recent record 2.51% photovoltaic conversion (η) currently held by Perera et al.¹³ It is accepted that the photovoltaic performances of p-DSSCs can be greatly enhanced if NiO could be replaced by a new p-SC displaying higher transparency, higher hole mobility and deeper valence band potential.⁵ In this context, we have investigated LaOCuS oxysulfide as a potential material to prepare p-DSSCs. Indeed, LaOCuS exhibits a higher conductivity, a slightly higher charge mobility (∼0.6 cm² V⁻¹ s⁻¹)¹⁴ and a higher transparency in the visible range (60–70%)¹⁵ (see Fig. S1 in the ESI†) than NiO, making it an attractive substitute for NiO for this application. Here, we report the first investigation on this material for application in p-DSSCs.

Fig. 1 Schematic operating principle of a p-type dye sensitized solar cell. S = sensitizer, M = redox mediator.

Light grey LaOCuS nanoparticles were synthesized according to a slightly modified method of that by Doussier et al.¹⁶ where the dehydration conditions of the LaCl₃·7H₂O precursor were optimized (namely, 75 h at 100 °C in an oven). Scanning electron micrographs on as-obtained samples evidenced plate-shaped crystals with a typical size of 40 nm wide and 5–10 nm thick (see Fig. S2 in the ESI†). This is inherited from the layered structure of the material, commonly described as built upon the regular stacking of [Cu₂S₂] and [La₂O₂] infinite layers. The specific surface area, determined by the Brunauer–Emmett–Teller (BET) method, was calculated at 44(1) m² g⁻¹, i.e. three times lower than that of Inframat© black NiO (158(1) m² g⁻¹)^7a which is the reference material in the domain of p-type DSSCs. Electrochemical impedance spectroscopy (EIS) measurements were carried out on pressed (100 bars) and sintered (3 hours at 450 °C under a nitrogen atmosphere) pellets to determine the flat band potential of LaOCuS in aqueous solution with LiClO₄ (1 M) as supporting electrolyte at a pH of 6.3. Impedance spectra were obtained under an AC voltage (5 mV in amplitude and a frequency range of 1 Hz to 100 kHz) in a potential range from −0.2 to +0.5 V vs. SCE, and were analysed in order to determine the flat band potential (E_fb) using the Mott–Schottky method at relatively high frequencies (1–10 kHz). The faradic surface phenomena were neglected because they are much slower than the charge–discharge ones (capacitive phenomena) at the semiconductor/electrolyte interface. The whole interface semiconductor/electrolyte capacitance (C) was determined according to the imposed potential (E) by modelling the whole electrochemical cell with a simplified Randles-equivalent circuit (Fig. 2a). This is composed of a series resistance R_S (mainly the resistance of the electrolyte and a contact resistance at the interfaces between the copper foil and the pellet, and in between the pellets) and a constant phase element CPE accounting for the non-ideality of the capacitance defined as Z_CPE = 1/(Q(jω)^α). Q, α, and ω stand for the pseudo-capacitance (perfect capacitor, F s^−α units), the deviation towards an ideal capacitance (α difference of 1) and the angular frequency. In the following, C values vs. E were calculated using the C = (R_s^1−αQ)^1/α formula.¹⁷ Assuming that C⁻² ≈ C_SC⁻² with C_SC as the space charge capacity within the space charge layer of the semiconductor, the Mott–Schottky graph was plotted (Fig. 2b) according to the eponymous relationship for p-SC,

where A is the interfacial surface area between the semiconductor electrode and the electrolyte, N_A is the number of acceptors, k is the Boltzmann constant, T is the temperature, e is the electron charge, ε₀ is the vacuum permittivity and ε is the relative permittivity of the semiconductor.

Fig. 2 Randles equivalent circuit of the SC-electrolyte interface (a) and Mott–Schottky plots for a LaOCuS pellet (electrolyte: LiClO₄ in water; pH = 6.3) (b).

The 1/C² vs. E curve is a perfect straight line with a negative slope in agreement with the p-type nature of the LaOCuS powdered sample. The flat band potential (E_fb) of LaOCuS, was extrapolated from the linear region, and is estimated at 0.26 V/SCE, i.e. 0.07 V lower than that of NiO measured under identical conditions (see Fig. S3 in the ESI†).^7a At high voltages (i.e. E higher than 0.05 V/SCE), the 1/C² vs. E curve deviates from the linear behaviour. This is attributed to the surface state contribution in the Helmholtz layer. Indeed, the inverse of the Helmholtz capacity is not negligible anymore and the Mott–Schottky relationship cannot be directly used (C⁻² ≠ C_SC⁻²). It leads to the definition of a “pseudo” flat band potential E′_fb = 0.4 V/SCE (i.e. 0.07 V higher than NiO) directly correlated to the real flat band potential: E_fb = E′_fb + ΔE. This deviation certainly originates from an ohmic drop (ΔE = −0.13 V) through the Helmholtz layer. Indeed, it was noticed that the series resistance changes with the applied potential in the 0.1–0.4 volt domain. This is most probably related to the adsorption of chemical species on the LaOCuS surface or to the progressive dissolution of the inorganic material in contact with water. This would explain the evolution of the C/V curve in the aforementioned voltage domain under cycling and this hypothesis is consistent with the recovery of the initial behavior once the surface was refreshed by polishing.

These intrinsic aforementioned properties of LaOCuS and its flat band potential prompted us to prepare solar cells with LaOCuS particles to assess the possibility to use oxysulfide materials as a substitute for NiO in p-DSSCs. LaOCuS nanoparticles were first dispersed in an organic paste as already described^7b and spread by the doctor blade method onto FTO substrates. The resulting films were sintered at 400 °C for 30 min under an argon atmosphere in order to remove organic compounds and to make contacts between the particles in order to create a hole percolation pathway. Scanning electron micrographs of a LaOCuS film are displayed in Fig. 3. The plan-view reveals a large roughness of the surface of the mesoporous films while examination of the cross-section shows an average thickness of about 3 μm. Dye sensitized photocathodes were then prepared by coating the above films with the PMI-NDI dyad (a push–pull sensitizer, Fig. 3c).¹⁸

Fig. 3 Plan-view (a) and cross-sectional (b) SEM micrographs of LaOCuS on FTO, structure of the PMI-NDI sensitizer (c), cobalt complex redox mediator (d) and a picture of the p-DSSC (e).

This was realized by immersing the LaOCuS electrodes into a solution of PMI-NDI (0.3 mM in acetone) for 48 hours. The light red colour of the film attests to the adhesion of the dye on the surface of LaOCuS; the surface concentration is nevertheless low. The components were then assembled as described elsewhere^7b to form the solar cell shown in Fig. 3e. The photocathode/polymer spacer (25 μm, Surlyn)/platinum-coated FTO glass were stacked together and sealed by heating under pressure. The interspaced layer was filled via a drilled hole (by vacuum backfilling technique)¹⁹ with a cobalt-based redox mediator (Fig. 3e), namely a mixture of 0.1 M tris(4,4′-bis-tert-butyl-2,2′-bipyridine)cobalt(II/III) (redox potential around 0.11 V/SCE) and 0.1 M LiClO₄ in propylene carbonate and the hole was isolated with a glass disk by sealing with a hot melt polymer gasket (60 μm, Surlyn). The use of the I⁻/I₃⁻ redox mediator was not considered here due to its reactivity with LaOCuS.

Current–voltage characteristics recorded in the dark and under AM1.5 illumination (1000 W m⁻²) for two p-DSSCs prepared from different LaOCuS batches are depicted in Fig. 4. The examination of the J/V curves clearly evidences the photovoltaic effect of the LaOCuS photocathodes. No photocurrent was observed in the absence of light, while a J_SC of about 0.039 mA cm⁻² measured under illumination supports the effective hole photoinjection in the LaOCuS valence band from the dye as occurs in NiO based p-DSSCs. However, at this stage the J_sc value remains much lower than that measured with NiO-based DSSCs under the same conditions. This is certainly related to the weaker specific surface area of the LaOCuS film and consequently to the lower quantity of dye chemisorbed on its surface. In addition, it is conceivable that the carboxylic binding group on the dye has a lower affinity for LaOCuS than NiO inducing a lower dye coverage of the film.

Fig. 4 Current/voltage characteristics of two solar cells recorded in the dark (dashed line) and under AM1.5 illumination (straight line) with a LaOCuS photocathode, a PMI-NDI dyad as sensitizer, and tris(4,4′-bis-tert-butyl-2,2′-bipyridine)cobalt(II/III)/LiClO₄ in propylene carbonate as a redox mediator.

The open circuit voltage (V_OC ≈ 150 mV) is also lower than the best values obtained with NiO-based devices (285–375 mV) prepared with identical components.^7a,20 Since the flat band potential of NiO and LaOCuS are close, this points to high losses by electron–hole recombinations in the present system decreasing the Fermi level of the SC under illumination.

To sum up, we have demonstrated the successful implementation of an oxysulfide material: LaOCuS as a photocathode in p-DSSCs. The position of its valence band potential is very similar to that of NiO allowing effective hole injection from the photo-excited PMI-NDI sensitizer. However, the attained photovoltaic characteristics of LaOCuS based solar cells (V_OC ≈ 150 mV, J_sc ≈ 0.039 mA cm⁻², ff ≈ 26%, η ≈ 0.002%) (see Table S1 in the ESI† for comparison with a NiO-based DSSC fabricated with the same dye and electrolyte) could be greatly improved with films of higher surface areas and with dyes having higher binding affinities for this semiconductor. In that respect, LaOCuS based p-type DSSCs might become a valuable alternative to NiO-based devices.

Acknowledgements

The authors are indebted to ANR POSITIF for financial support (contract: ANR 2012-PRGE-0016-01). Dr A. Renaud thanks the Région Pays de la Loire for her PhD grant (program Perle2).

Notes and references

1. Handbook of Transparent Conductors, ed. D. S. Ginley, co-ed. H. Hisono and D. C. Paine, Springer, 2011, ISBN 978-1-4419-1637-2; J. F. Wager, D. A. Keszler and R. E. Presley, Transparent Electronics, Springer, 2008, ISBN 978-0-387-72341-9, and references therein.
2. H. Kawazoe, M. Yasukawa, H. Kyodo, M. Kurita and H. Yanagi, Nature, 1997, 70, 2701.
3. K. Ueda, T. Hase, H. Yanagi, H. Kawazoe, H. Hosono, H. Ohta, M. Orita and M. Hirano, J. Appl. Phys., 2001, 89, 1790.
4. A. Kudo, H. Yanagi, H. Hosono and H. Kawazae, Appl. Phys. Lett., 1998, 73, 220.
5. (a) F. Odobel, L. Le Pleux, Y. Pellegrin and E. Blart, Acc. Chem. Res., 2010, 43, 1063; (b) F. Odobel and Y. Pellegrin, J. Phys. Chem. Lett., 2013, 4, 2551.
6. (a) A. Nattestad, X. Zhang, U. Bach and Y.-B. Cheng, J. Photonics Energy, 2011, 1, 011103; (b) J. Ahmed, C. K. Blakely, J. Prakash, S. R. Bruno, M. Yu, Y. Wu and V. V. Poltavets, J. Alloy. Comp., 2014, 591, 275.
7. (a) A. Renaud, B. Chavillon, L. Le Pleux, Y. Pellegrin, E. Blart, M. Boujtita, T. Pauporté, L. Cario, S. Jobic and F. Odobel, J. Mater. Chem., 2012, 22, 14353; (b) A. Renaud, L. Cario, P. Deniard, E. Gautron, X. Rocquefelte, Y. Pellegrin, E. Blart, F. Odobel and S. Jobic, J. Phys. Chem. C, 2014, 118, 54; (c) Z. Xu, D. Xiong, H. Wang, W. Zhang, X. Zeng, L. Ming, W. Chen, X. Xu, J. Cui, M. Wang, S. Powar, U. Bach and Y.-B. Cheng, J. Mater. Chem. A, 2014, 2, 2968.
8. D. Xiong, Z. Xu, W. Zhang, W. Chen, X. Xu, M. Wang and Y.-B. Cheng, J. Mater. Chem., 2012, 22, 24760; S. Powar, D. Xiong, T. Daeneke, M. T. Ma, A. Gupta, G. Lee, S. Makuta, Y. Tachibana, W. Chen, L. Spiccia, Y.-B. Cheng, G. Götz, P. Bäuerle and U. Bach, J. Phys. Chem. C, 2014, 118, 16375.
9. Z. Shi, H. Lu, Q. Liu, K. Deng, L. Xu, R. Zou, J. Hu, Y. Bando, D. Golberg and L. Li, Energy Technol., 2014, 2, 517.
10. B. ÓRegan and M. Grätzel, Nature, 1991, 353, 737.
11. K. Keis, E. Magnusson, H. Lindstrom, S.-E. Lindquist and A. Hagfeldt, Sol. Energy Mater. Sol. Cells, 2002, 73, 51.
12. J. He, H. Lindström, A. Hagfeldt and S.-E. Lindquist, J. Phys. Chem. B, 1999, 103, 8940; L. le Pleux, B. Chavillon, Y. Pellegrin, E. Blart, L. Cario, S. Jobic and F. Odobel, Inorg. Chem., 2009, 48, 8245 and references therein.
13. 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.
14. H. Hiramatsu, K. Ueda, H. Ohta, M. Hirano, T. Kamiya and H. Hosono, Appl. Phys. Lett., 2003, 82, 1048.
15. H. Hiramatsu, K. Ueda, H. Ohta, M. Orita, M. Hirano and H. Hosono, Thin Solid Films, 2002, 411, 125.
16. C. Doussier, B. Chavillon, L. Cario and S. Jobic, Inorg. Chem., 2010, 49, 3074.
17. G. J. Brug, A. L. G. Van den Eeden, M. Sluyters-Rehabach and J. H. Sluyters, J. Electroanal. Chem., 1984, 176, 275.
18. A. Morandeira, J. Fortage, T. Edvinsson, L. Le Pleux, E. Blart, G. Boschloo, A. Hagfeldt, L. Hammarström and F. Odobel, J. Phys. Chem. C, 2008, 112, 1721.
19. A. Nattestad, M. Ferguson, R. Kerr, Y.-B. Cheng and U. Bach, Nanotechnology, 2008, 19, 295304/1.
20. 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., Int. Ed., 2009, 48, 4402.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra07859h

---