WO2014173811A1

WO2014173811A1 - Charge mediators for sensitised solar cells

Info

Publication number: WO2014173811A1
Application number: PCT/EP2014/057905
Authority: WO
Inventors: Krishnan Ravindranathan THAMPI; Praveen Kumar SUROLIA; Owen Joseph BYRNE
Original assignee: University College Dublin, National University Of Ireland, Dublin
Priority date: 2013-04-23
Filing date: 2014-04-17
Publication date: 2014-10-30
Also published as: GB201307320D0

Abstract

The invention concerns the use of a perovskite oxide as a charge mediator in a sensitised solar cell (SSC), the perovskite oxide having the formula AB(X)_3-δ, wherein A is an alkali metal ion, an alkaline earth metal ion, a transition metal ion, a p-block element cation or a rare earth metal ion; B is an alkaline earth metal ion, a transition metal ion, a p-block element cation or a rare earth metal ion; X is an oxygen anion;and δ is zero to 0.3.The perovskite oxide may be used as the main charge transfer relay mediator in the absence of added iodine and is optionally deposited on an electrode or dispersed in an ionic liquid, ionic solid,eutectic ionic melt or solid state hole conductor.

Description

Charge mediators for sensitised solar cells

This invention relates to charge mediators for sensitised solar cells (SSC) and to SSC containing same. In particular, the invention relates to perovskite oxide-based charge mediators for SSC.

The use of electrolytes containing solvents, iodide/triiodide redox couple and other additives has created difficulties for the wider practical applications of SSC due to solvent volatility, sealing failure, leakage, water and oxygen ingress, difficult manufacturing conditions, and often the need for extra processing steps. Problems also arise with hole conductors, particularly with regard to their stability in air and in humid conditions.

Two prior SSC, containing perovskite-halide type compounds have been reported. A lab synthesized pervoskite type material (CsSnl₃._xF_x) has been used as a charge mediator based on a hole conduction mechanism but requires a critical synthetic procedure and conditions, and had problems with air and humidity sensitivity [Chung, I., Lee, B., He, J., Chang, R. P. & Kanatzidis, M. G. All-solid- state dye-sensitized solar cells with high efficiency, Nature 485, 486-489, doi:10.1038/nature11067 (2012).]. An organometal halide perovskite, methyl ammonium lead iodide chloride (CH₃NH₃Pbl₂CI) has been used as a light sensitizer coupled with 2,27,7'-tetrakis-(N,N-di-p-methoxyphenyl-amine)- 9,9'-spirobifluorene (spiro-MeOTAD) as charge mediator [Lee, M. M., Teuscher, J., Miyasaka, T., Murakami, T. N. & Snaith, H. J. Efficient hybrid solar cells based on meso-superstructured organometal halide perovskites, Science 338, 643-647, doi:10.1126/ science.1228604 (2012)

It is an object of the invention to avoid or minimise the disadvantages of the prior art. The invention is based on the discovery that perovskite oxides may be used in SSC as the main charge transfer relay mediator in the absence of added iodine and optionally deposited on an electrode or dispersed in an ionic liquid, ionic solid, eutectic ionic melt or solid state hole conductor.

According to the invention there is provided the use of a perovskite oxide as a charge mediator in a sensitised solar cell (SSC), the perovskite oxide having the formula ΑΒ(Χ)₃._δ, wherein A is an alkali metal ion, an alkaline earth metal ion, a transition metal ion, a p-block element cation or a rare earth metal ion;

B is an alkaline earth metal ion, a transition metal ion, a p-block element cation or a rare earth metal ion;

X is an oxygen anion; and

δ is zero to 0.3.

As used herein, the term "charge mediator" is intended to mean a medium that transfers or relays charge between two locations, e.g., between two poles of an electrochemical device.

As used herein, the term "sensitizer" is intended to mean a photon harvesting material including, but not limited to, dye molecules, quantum dots and nanostructured semiconductor materials.

As used herein, the term "ionic liquid" is intended to mean a salt in the liquid state.

As used herein, the term "ionic solid" is intended to mean an ionic compound in its solid form.

As used herein, the term "eutectic ionic melt" is intended to mean a mixture of ionic liquids and/or ionic solids, which when mixed together, form a liquid at a melting temperature that is lower than the melting temperature of the individual components of the melt.

Preferably, A is an alkali metal ion selected from lithium, sodium, potassium and cesium; or an alkaline earth metal ion selected from magnesium, calcium, strontium and barium; or a transition metal ion selected from iron and palladium; or a p-block element cation selected from aluminium, silicon, gallium, germanium, tin, lead, bismuth and thallium; or a rare earth metal ion selected from lanthanum, cerium, praseodymium, neodymium, samarium, gadolinium, erbium and ytterbium.

More preferably, A is barium, calcium, strontium, lanthanum, cerium, praseodymium, neodymium or gadolinium, most preferably lanthanum. A may be doped with 1 or 2 elements selected from an alkali metal ion, an alkaline earth metal ion, a transition metal ion, a rare metal ion, and a p-block element cation. Preferred dopants include barium, calcium, strontium, lanthanum, cerium, praseodymium, neodymium and gadolinium. In a preferred embodiment, A is lanthanum doped with strontium.

Preferably, B is an alkaline earth metal ion selected from magnesium, calcium, strontium and barium; or a transition metal ion selected from scandium, yttrium, titanium, zirconium, manganese, iron, cobalt, vanadium, niobium, tantalum, ruthenium, nickel and copper; or a rare earth metal ion selected from lanthanum, cerium, praseodymium, neodymium samarium, gadolinium, erbium and ytterbium; or a p-block element cation selected from aluminium, silicon, gallium, germanium, tin, lead, bismuth and antimony.

More preferably, B is selected from manganese, iron, cobalt, germanium, lanthanum, cerium, praseodymium, neodymium, gadolinium, titanium and zirconium, most preferably cobalt.

B may be doped with 1 or 2 elements selected from an alkaline earth metal ion, a transition metal ion, a rare earth metal ion and a p-block element cation. Preferred dopants for B include manganese, iron, cobalt, germanium, lanthanum, cerium, praseodymium, neodymium, gadolinium, titanium and zirconium.

A and B cannot be the same. However, if both of A and B are doped, they may be doped with the same or different dopants.

A particularly preferred perovskite oxide is La-|._xSr_xCo0₃.₅ (LSCO) wherein x is 0.1 to 0.7 and δ represents nonstoichiometry and typically has a value of zero to 0.3, preferably zero to 0.15.

LSCO has been used for solid oxide fuel cell (SOFC) cathode applications, due to its mixed ionic- electronic conductivities at high temperatures. However, due to its lower material stability, LSM (La₀.8Sro.₂ Mn0₃.₅) is often preferred over LSCO for SOFC applications, despite the latter's lower ionic conductivity (LSM~10^"7-10^"8S.cm ^~1 ; LSCO~0.20-0.25S.cm^~1 at 800-900°C). At temperatures below 100°C, its ionic conductivity is considered very small and not much numerical information is available on this parameter at ambient temperatures owing to the fact that most of its applications are hitherto reported at temperatures above 500°C. At lower temperatures it is considered as a semiconductor with most of its conductivity attributed to the electronic component. However, it should be noted that these perovskite cathode materials show an increasing oxygen anionic (0^2~) conductivity related performance enhancement under electric load due to the fact that the anion vacancies are increasingly generated and transported under current somewhat through an auto- catalytic or activation process facilitating oxygen anion vacancy movements. This is most often noticed in SOFC experiments where the domain SOFC attains its ultimate performance only after induction under the potentiostatic bias or electric load for several hours.

The perovskite oxide used herein is preferably deposited on an electrode or dispersed in an ionic liquid, ionic solid, eutectic ionic melt or solid state hole conductor. Such dispersions form part of the present invention. In such dispersions, the perovskite oxide transfers greater than 50%, preferably greater than 60%, more preferably greater than 70%, of the charge between two locations, e.g. between two poles of an electrochemical device. No iodine is added to the charge mediator. Preferably, substantially no iodine is present in the charge mediator. However, small amounts of iodine could be formed from iodide present in the dispersant. Preferred ionic liquids include 1-methyl-3-propylimidazolium iodide (PMII), 3-hexyl-1- methylimidazolium iodide (HMII), 1-butyl-3-methylimidazolium iodide (BMII), 3-hexyl-1 ,2- dimethylimidazolium iodide (DMHII) and 1-ethyl-3-methylimidazolium tetracyanoborate (EMITCB).

Preferred ionic solids include 1 ,3-dimethylimidazolium iodide (DMII), 1-ethyl-3-methylimidazolium iodide (EMM), 1-allyl-3-methylimidazolium iodide (AMII), 1 ,2-dimethyl-3-propylimidazolium iodide (DMPII) and N-methyl-N-butyl pyrrolidinium iodide (P_{1 4}l).

Preferred eutectic ionic melts include DMII mixed with EMM, and DMII mixed with EMM and AMII. Most preferred are the ionic liquids PMII and HMII or the eutectic ionic melt DMII/EMII/AMII. The perovskite oxide is preferably dispersed in the ionic liquid, ionic solid, eutectic ionic melt or solid state hole conductor at a concentration of from about 1 to 20% (w/w), preferably from about 7 to 10% (w/w).

The charge mediator is preferably used in the SSC at a temperature less than 100°C.

The sensitizer in the SSC may be selected from dye molecules, quantum dots and nanostructured semiconductor materials.

The invention also provides a sensitised solar cell (SSC) comprising a perovskite oxide-based charge mediator as defined above.

The SSC may be a dye sensitized solar cell (DSSC), a photoelectrochemical cell or a photogalvanic cell.

In one embodiment, the sensitizer is a dye molecule and the SSC is a dye sensitized solar cell (DSSC). In a preferred embodiment of the DSSC, the charge mediator comprising the perovskite oxide is dispersed on a dye-adsorbed working electrode which is operatively associated with a counter electrode.

In another preferred embodiment of the DSSC, the charge mediator comprising the perovskite oxide is deposited on a working electrode and/or on a counter electrode.

Examples of depositions include but are not limited to: screen-printing, stencil-printing, inkjet- printing, scrubbing onto the surface, thermal evaporation, sputtering, spraying, dip coating, spin coating, dropping, and chemical bath deposition, from dispersions.

The SSC of the invention may be an open or closed cell. The SSC of the invention is suitable for use in indoor light harvesting applications and in outdoor sunlight harvesting applications. The invention further provides a perovskite oxide as defined hereinabove deposited on an electrode or dispersed in an ionic liquid, ionic solid, eutectic ionic melt or solid state hole conductor. The perovskite oxide is preferably La-|._xSr_xCo0₃.₅ as defined hereinabove, especially Lao.sSro^CoOs-s_, wherein δ is as defined hereinabove. In the accompanying drawings, Figure 1 illustrates photo-current, photo-voltage (J-V) characteristics of a DSSC prepared with various LSCO based electrolyte combinations simulated under AM1.5 sunlight with irradiance powers of 1000 W/m² (a)-(c); effect of Ar on the cell efficiency and current density of the cell (d); Figure 2 illustrates the proposed O^2" ion conduction (a) and hole transfer mechanism (b): Senaris-Rodriguez, M. A. et al. Magnetic and Transport Properties of the System Lai_-xSr_xCo0₃-₅ (0 < x < 0.50). Journal of Solid State Chemistry 118, 323-336,

doi: 10.1006/jssc.1995.1351 (1995); Figure 3 illustrates stability testing of a batch of open cells containing Lao.sSro^CoOs-s (LSCO)/BMII electrolyte stored in dark conditions in ambient air atmosphere during 650 hours; and Figure 4 is a diagrammatic illustration of solar cells consisting of a working electrode (WE) operatively associated with a counter electrode (CE) and an electrolyte material (E).

The invention is illustrated in the following Examples.

Example 1

Chemicals and Materials

Ti0₂ P25 powder (average size 20 nm), ethyl cellulose (Fluka, #46080 and #46070) and anhydrous terpineol (Sigma-Aldrich, 86480) were used to prepare transparent absorption layer paste as described in Ito, S. et al. Fabrication of screen-printing pastes from Ti0₂ powders for dye-sensitised solar cells. Progress in Photovoltaics: Research and Applications 15, 603-612, doi: 10.1002/pip.768 (2007). Scattering layer paste (particle size 150-250 nm, WER 2-0) was purchased from Dyesol Ltd. The following materials were used as received and without further purification; Lao.sSro^CoOs-s (PRAXAIR Surface Technologies), N-719 dye (Dyesol Ltd), 1-methyl-3-propylimidazolium iodide (PMII, Merck 4.94202.0025), acetonitrile (ACN, Sigma Aldrich 271004-1 L) and Co(N0₃)₂.6H₂0 (BDH Chemicals Ltd.). Preparation of electrodes and DSSC assembly

The working electrodes were prepared using the standard screen-printing method as described inlto, S. ef al. Fabrication of screen-printing pastes from Ti0₂ powders for dye-sensitised solar cells.

Progress in Photovoltaics: Research and Applications 15, 603-612, doi:10.1002/pip.768 (2007). The fluorine doped tin oxide (FTO) conducting glass was cleaned and immersed in 40mM TiCI₄ solution at 70°C for 30min in order to make a non-porous dense blocking Ti0₂ under layer. The blocking layer helps to reduce charge recombination. Layers of Ti0₂ paste were printed on this TiCI₄ treated FTO- glass using a TIFLEX manual screen printer. Overall, 7 layers of transparent absorption layer paste and 2 layers of scattering layer paste were printed, yielding a total thickness of 15 microns. In between two consecutive printing steps, the printed films were kept in an ethanol chamber for 6min each to reduce any surface irregularities, followed by drying at 125°C for 6min. The fully prepared working electrodes were finally sintered with gradual heating in oven at 325°C (5min), 375°C (5min), 450°C (15min) and 500°C (30min). The obtained electrodes were once more treated with TiCI₄ and sintered again at 500°C for 30min. The printed area was of 6mm diameter circles resulting in a total area of 0.28cm² per cell. These Ti0₂ electrodes were immersed into 0.3mM N-719 dye solution in acetonitrile : tert-butylalcohol : THF (volume ratio, 4.5:4.5: 1 ) for 20 hours.

The counter electrodes were prepared by the deposition of a Pt catalyst via a drop of H₂PtCI₆ solution (2mg Pt in 1 ml ethanol) followed by a heat treatment at 400°C for 15min. A drop of electrolytes prepared as described in tables 1-4 below was placed on the dye adsorbed working electrode. Both working and counter electrodes were sandwiched together with clips using 50μιη thick Bynel polymer gasket as a spacer to make an open DSSC. Closed DSSC were also prepared using 50 and δθμιτι thick hot melt Bynel gaskets for comparison purposes (Tables 1 and 2).

The commercially available perovskite material La₀.₈Sr₀.₂CoO₃-5 (LSCO) was dispersed in PMII as 7% (w/w), denoted LSC1 and used as electrolyte without using any other additives, as shown in Table 1. The other combinations of electrolytes tested are described in Tables 1-3. A Newport 91195A-1000 solar simulator with Newport 69920 Arc Lamp Power Supply was used for the electro-optical characterization of the prepared cells. A Newport 81088A Air Mass Filter was placed before the output of the solar simulator to simulate AM 1.5 spectrum with irradiance powers of 1000 W/m². I-V measurements including current/voltage (l-V) curves, open circuit voltage (Voc), short circuit current density (Jsc) and fill factor (FF) were recorded with a GAMRY Instruments Potentiostat.

Table 1. Photovoltaic parameters of the DSSC devices made with the listed electrolytes as below at 1 sun (1000 W.nrf²) incident intensity of AM 1.5 solar light

DSSC Composition of electrolyte Voc (V) Jsc FF η (%) Sample

(mA.cm-²)

LSC1 Open cell, 7% (w/w, LSCO/PMII) 0.775 6.50 0.47 2.38

LSC1 After 5 min warm up 0.747 8.10 0.57 3.43

LSC1 After 10 min warm up 0.743 8.28 0.57 3.50

LSC2 Closed cell, 7% (w/w, 0.680 2.06 0.66 0.92

LSCO/PMII), Vacuum back filling

through hole

LSC3 Closed cell, 7% (w/w, LSCO/PMII), A 0.648 4.33 0.61 1.69 drop of electrolyte was put on the dye

adsorbed Ti0₂ and electrodes were

sealed after that (Bynel 50μπι thick)

LSC4 Open cell, 7% (w/w, LSCO/PMII), Blank 0.542 0.371 0.48 0.10 cell (without dye uptake)

LSC5 Closed cell, only PMII 0.612 0.63 0.14 0.06 Table 2. Photovoltaic parameters of the closed DSSC devices fabricated in different way at 1 sun (1000 W.m^"2) incident intensity of AM1.5 solar light

DSSC Composition of electrolyte Voc (V) Jsc FF η (%) Sample

(mA.cm-²)

LSC6 Closed cell, 7% (w/w, LSCO/PMII), A drop 0.710 8.06 0.50 2.83 of electrolyte was put on the dye

adsorbed Ti0₂ and electrodes were

sealed after that (Bynel 80μπι thick)

LSC7 Closed cell with back hole, 7% (w/w, 0.744 3.12 0.73 1.69

LSCO/PMII), A drop of electrolyte was put

on the dye adsorbed Ti0₂ and electrodes

were sealed after that (Bynel 80μπι thick)

LSC8 The PMII was again filled in LSC7 after 0.751 4.67 0.66 2.30 sealing to make up the damage/loss of

PMII due to temperature during sealing

Table 3. Photovoltaic parameters of the DSSC devices made with the electrolytes prepared with other formulations tried at 1 sun (1000 W.m^"2) incident intensity of AM 1.5 solar light

DSSC Composition of electrolyte Voc (V) Jsc FF η (%) Sample

(mA.cm-²)

LSC9 Closed cell, l₂:PMII (1 :5 molar ratio) 0.539 4.14 0.59 1.30

LSC10 Open cell, l₂:PMII (1 :5 molar ratio), 7% 0.696 2.76 0.66 1.28

(w/w, LSCO/PMII)

LSC1 1 Open cell, 7% (w/w, LSCO/BMII) 0.792 2.79 0.47 1.04

LSC12 Closed cell, Co(N0₃)₂.6H₂0 in 0.771 4.29 0.47 1.55

Acetonitrile (0.1 M)

LSC13 Open cell, 7% (w/w, LSCO/PMII), 0.641 2.60 0.66 1.09

Co(N0₃)₂.6H₂0 (0.1 M)

Table 4. Photovoltaic parameters of the DSSC devices under air, Argon (Ar) and recovery under air at 1 sun (1000 W.m^"2) incident intensity of AM1.5 solar light

The crystal structure of the perovskite La-|._xSr_xCo0₃ can be either "rhombohedral" with a defined space group ^^c for 0.0≤ x≤ 0.5 or "cubic" with a space group Pm3m for 0.55≤ x < 0.7. The material used in the present Examples is rhombohedral. Doped LaCo0₃, and in particular Sr doped La-|._xSr_xCo0₃, has high electronic and ionic conductivities and used as cathode in solid oxide fuel cells. The ionic conductivity is due to the transport of O^2" ions present in the crystal structure. The excess charge created by Sr-doping in lanthanum cobaltite can be compensated either by the creation of holes, i.e. oxidation of Co³⁺ to Co⁴⁺ or by the creation of oxygen vacancies. The oxidation of Co³⁺ is preferable for x<0.5 while creation of oxygen vacancy is preferred for x>0.5. However, none of these possibilities can be ruled out completely at any composition and samples should be described as La-|._xSr_xCo0₃.₅ (See Mineshige, A. ef al. Crystal Structure and Metal-Insulator Transition of La1 -xSrxCo03. Journal of Solid State Chemistry 121 , 423-429,

doi:10.1006/jssc.1996.0058 (1996)). Table 5. Photovoltaic parameters of the DSSC devices made with the electrolytes prepared with the following compositions tested at 1 sun (1000 W.m-2) incident intensity of AM1.5 solar light.

DSSC consists of N719 dye coated Ti0₂ on FTO glass working electrode and platinum coated FTO glass counter electrode with the following electrolyte compositions: LSC01 (X3), LSC02(X3); Lao.7Sro.3Co Oxide (PRAXAIR Surface Technologies, part # PS-PCoLa31 1-2DTT) dispersed in PMII. LSM01, LSM02: Lao.₈Sro.₂Mn Oxide (lab prepared) dispersed in PMII. LSF01, LSF02: (Lao.₈Sro.₂)o.99Fe Oxide (PRAXAIR Surface Technologies, part # PS-PFeLa313-3D50) dispersed in PMII. PMII (1-methyl-3-propylimidazolium iodide, Merck 4.94202.0025).

Example 2

Photovoltaic performance of the DSSC

The results obtained with Lao.₈Sro.₂Co0₃-₅ (LSCO) in PMII (7% w/w), denoted as LSC1 electrolyte without any other additives, are shown in Table 1. The other combinations of electrolyte tested are presented in Tables 1-3. The corresponding J-V curves for all these combinations are given in Figure 1 (a-c). The open cell prepared with the electrolyte LSC1 shows an efficiency of 2.38% with a current density of 6.50 mA.cm^"2 and a Voc of 775 mV, upon illumination. The performance improves under continuous illumination of simulated AM1.5G solar light (1 Sun) for the first 10 min. This improvement over time is a characteristic behaviour of these types of perovskites and is akin to auto-generation of increased oxygen anion generation and mobility under increasing electric current (electron supply). Tests were performed that clearly show the increasing current is not due to the heating of the DSSC. This was confirmed by an experimental set up where a water jacket barrier was placed in between the solar beam and the cell to be measured, to ensure the removal of IR radiation responsible for temperature rise in the cell. The cell temperature was maintained steady throughout the experiment. It is to be noted that the change in ionic conductivity of the materials at ambient temperatures is expected to be very low for these materials. The same trend of improvement in cell efficiencies was observed with time, which shows that long time cell operation increases the charge transfer capacity of the LSCO: i.e. its ionic conductivity enhances under electric load. After 10 min, the performance of the cell became steady and did not improve any further, showing that a steady state is reached between electron transfer from the electrodes and the ion charge carrier generation in the electrolyte matrix. Only the peak performances of each cell after 10 min continuous illumination are mentioned in the paper for rest of the cells. The LSC1 shows a typical energy conversion efficiency of 3.5% with the current density of 8.28 mA.cm^"2 and Voc of 743 mV after 10 min. The same electrolyte was then tried for the preparation of closed cells as described LSC2, LSC3 and LSC6-8 in Tables 1 and 2. In the LSC2, the electrolyte was filled by a vacuum back filling method and it showed only 0.92% efficiency. This is due to the large particle size of LSCO, which are in the range of 0.5-1.2μιη and too large for back filling, causing blockage in the back hole. To overcome this problem, LSC3 was prepared where a drop of the electrolyte was placed on the dye adsorbed Ti0₂ and then both the electrodes were sealed using a 50μιη size Bynel gasket. The performance of the cell was better with typical cell efficiency of 1 .69%, but still far behind the LSC1 . The same cell with the technique of LSC3 was prepared with a different gasket thickness, where δθμιτι Bynel was used and denoted as LSC6. LSC6 showed a performance of 2.83%, which is better than any other closed cell but still not up to the mark of open cell LSC1 , which showed 3.5% efficiency. It seems that a minimum of 50μιη spacing between both the electrodes is crucial for a reasonable performance. The melting of δθμιτι Bynel during the sealing process might have reduced the thickness around 50μιη, same as open cell which might have improved the performance of closed cell in this case. However, during the sealing process a slight loss in PMII amount was observed. LSC7 was prepared by using δθμιτι Bynel and a counter electrode with back holes, which showed an efficiency of 1.69%. The PMII was filled in this cell by vacuum filling process to make up the loss of PMII during the sealing process and described as LSC8. The cell immediately improved and showed 2.30% efficiency. This efficiency is still below that of the open cell, which suggested that air or oxygen might be assisting the performance of cell, which was expected from the nature of the electrolyte employed.

For comparison and as a blank, a cell LSC5 was fabricated with just PMII as electrolyte and tested. The cell showed only 0.06% conversion efficiency and confirms that the presence of LSCO is absolutely essential for good performance. LSCO is a black material and to check its role in the photon absorption, a cell LSC4 was prepared without dye on Ti0₂ electrode. Even though LSC4 shows only 0.10% efficiency with 0.371 mA.cm^"2Jsc, it indeed suggested some light absorption; but this could also be due to Ti0₂'s light absorption contribution. This result shows that LSCO does not act as a major photon harvester and it functions mainly as an electrolyte component only. LSCO contains cobalt, and it should be noted that dissolved cobalt complexes have been proven to be useful redox relays (homogeneous systems) in liquid DSSC electrolyte media. The LSCO used herein has Co(lll) and Co(IV) ions in its framework. To further investigate the role of Co ions in the redox mechanism of DSSC, LSC12 was prepared with Co(N0₃)2.6H₂0 in acetonitrile and it showed 1.55% efficiency. Considering this, and to compare this with a heterogeneous electrolyte medium as we have with LSCO in PMII, an electrolyte was prepared with LSCO and Co(N0₃)2.6H₂0 in PMII to create Co(ll), Co(lll) and Co(IV) ions in the system and a cell LSC13 was fabricated. However the combination of all these components shows a relatively poor efficiency of 1.09% as compared to their individual performances.

LSCO with the combination of PMII and iodine (LSC10) was also tried as iodine in PMII (LSC9) itself shows 1.30% efficiency. The addition of iodine (LSC10) does not improve performance further, which suggests the absence of I7I₃ ^" redox couple involvement in the mechanism. A different ionic liquid namely, BMII (1-butyl-3-methylimidazolium iodide) was also used as a replacement of PMII (LSC1 1 ). As shown in table 3, this combination does not show any improvement as compared to LSC1 , which confirms the requirement of a suitable conductive medium for good cell performance. LSCO remains at the surface while PMII can infiltrate through the Ti0₂ pores and provides a contacting medium between LSCO and dye Ti0₂. LSCO without PMII does not show any output. PMII is less viscous and has a higher ionic conductivity than BMII, which could be the reason for its better performance when coupled with LSCO. Also, it points to the fact that improved cell performance can be expected with appropriately tailor-sized nanoparticles of LSCO.

Effect of Oxygen

Open cells performed better as compared to all combinations of closed cells. This could be due to the availability of oxygen. The DSSC system here relies on O^2" ion transport as suggested in Figure 2(a) and the presence of oxygen might help in the charge transfer between Pt counter electrode and the dye. The Pt-counter electrode brings into contact the three phases, i.e., the gaseous oxygen, the solid electrolyte and the electrode itself. The electrons from the counter electrode are taken up by the dissolved atmospheric oxygen as a result of the oxygen anionic transport possible in the LSCO (though very small at room temperature to begin with), without itself being consumed or corroded. Adsorbed oxygen onto LSCO can work similar to the mechanism operated in SOFC, though significantly at smaller levels owing to the temperature difference between SOFC and DSSC. At the surface of the counter electrode-LSCO- 0_2(S), atmospheric oxygen takes up electrons and is reduced to O^2" ion. This O^2" ion is transported through the available oxygen vacancies in LSCO where it is oxidized back to oxygen gas by giving up electrons to dye for regeneration, Figure 2(a). Note that the current densities we are dealing with DSSC (few mA.cm^"2) and SOFC (close to 0.1-1 A. cm^"2) are very different, and consequently in DSSC the demand for charge carrier concentration is much smaller than in SOFC. The residual ionic conductivity of LSCO at room temperature seems to be sufficient to meet the DSSC's demand at 3-4% cell efficiencies. Obviously, it opens up the next course of research, where it should be possible to find out better co-containing perovskite ionic conductors at ambient temperatures and raise the cell efficiencies to higher levels. The flow of charged O^2" ions and release of 0₂ back into the electrolyte by electron donation to dye maintains the overall charge balance. The electronic conductivity seems to be not affecting the cell much as evidenced by a relatively superior Voc of 775 mV, which is better than most commonly reported DSSCs. This again suggests that new perovskites with higher ionic conductivities and very low electronic conduction offer a significant potential for creating an all solid-state DSSC with much higher energy conversion efficiencies.

The role of oxygen in the conduction mechanism was checked by separate experiments. An open cell (LSC14) prepared with the same configuration as of LSC1 , was placed into a sealed glass chamber with a facility for argon gas, "Ar", purging. The performance of this cell was checked in normal atmosphere first and allowed it reach the maximum performance. Ar was then purged through the chamber slowly and the performance of the cell was checked at different time intervals. As the 0₂ present in the chamber and hence in the cell was slowly replaced by Ar, a continuous drop in the current density was observed as shown in Figure 1 (d). After 35 min flow of Ar, the chamber was opened to regain the normal ambient air atmosphere slowly and the performance of the cell was again monitored continuously. As the chamber started to fill with the air, the cell reached back again to its best performance within just 10 min. This experiment confirms the role of oxygen and supports the mechanism shown in Figure 2(a). The oxygen vacancy in the electrolyte may be in equilibrium with the atmospheric oxygen during the charge transfer in the cell. As in fuel cells, in DSSC atmospheric oxygen in contact with LSCO at Pt cathode can convert into O^2" by taking electrons from Pt. This excess charge in term of O^2" can migrate to light sensitizer, where it can release excess negative charge to neutralize the dye and convert back into oxygen molecule. The presence of Ar decreases slowly the oxygen concentration in the cell, prohibiting the charge transfer ionic mechanism; which in turn reduces the cell performance. In the same manner, the closed cell has very limited oxygen available as it is trapped and the cell is likely to be operated under an oxygen diffusion limited regime. This limits the performance of the closed cell. Thus the presence of oxygen in DSSC is taken as an advantage to invent a new type of DSSC for the first time. Such a solid state cell, with tailored newer and DSSC specific perovskite compounds has the potential to make a paradigm shift in DSSC technology. It has the potential to revive the commercial interest in DSSC applications for exteriors and roof tops, since lately DSSC companies are turning their attention to indoor and safe applications due to their inability to withstand harsh weather conditions arising out of the sealing failures and water and oxygen ingress.

Role of Co(lll) and Co(IV) ions

Along with the oxygen anion conduction mechanism, the cobalt present in LSCO can also play a role in the electron transfer. The charge variation due to the replacement of La³⁺ with Sr²⁺ is balanced by Co(IV) generation from Co(lll). The Co ions are arranged in a dynamic short range ordering of high- spin and low-spin states and the hole released as Co(IV) moves in the matrix and electron shuttles from one Co to the next. According to mechanism shown in Figure 2(b), the Co(lll) ion transfers two t₂ electrons equally to six neighbouring Co³⁺ ions and a Co³⁺ ion transfers back two "e" electron equally to each of six neighbouring Co(lll) ions, resulting in no net change in the charge, but interchange of the spin states. However, Co(IV) on the low spin array can only transfer a single t₂ electron to the neighbours and the missing t₂ electron condense as a t₂ hole on one of the neighbouring cobalt centres. This neighbouring cobalt centre interchanges from high-spin to a low- spin state. At the counter electrode, Co(IV) can convert to Co(lll). This then migrates to the dye- electrolyte interlayer through the above process and transfer the electron to the dye cation in order to achieve charge neutrality and closing the electric circuit, Figure 2(b). Applicant has reported for the first time a perovskite oxide-based charge mediator for SSC, operating on an oxygen anion (0^2~) conduction mechanism. Applicant believes, this report is the first of its kind where e.g. a lanthanum strontium cobaltite is used in SSC as a charge mediator material invoking an oxygen anion conduction mechanism. This is also the first report where LSCO assumes the role of a charge mediator rather than an electrode. Further, this is the first time that dissolved oxygen is beneficially used for SSC's operating mechanism. LSCO and other similar perovskites are commercially available, inexpensive and are mass produced through a variety of techniques. The charge mediator used in the invention in a heterogeneous dispersed phase works in open conditions and eliminates the requirement for perfect sealing. The oxygen present in air is constructive for SSC's operation and a new mechanism has been proposed based on the experimental results. A convincing 3.5% efficiency was observed with just LSCO and PMII without any further additives or extra treatment of electrodes or cells. Disclosed herein is a simple, non-polluting, stable and commercially available inorganic mixed conducting material of the perovskite family, for example La₀.₈Sr₀.2CoO3-5 (LSCO), optionally dispersed in an ionic liquid, ionic solid, eutectic ionic melt or solid state hole conductor. This new class of charge mediator is free from evaporation and problems associated with high vapour pressures, and does not require any further organic or inorganic additives. It is expected to be also compatible with glass-frit as well as polymer sealing, two methods highly desired by industrial SSC manufacturers. The extra processes regularly used in SSC fabrication such as use of Ti0₂ paste prepared with pure anatase phase, fluorine plasma etching of Ti0₂ electrode, use of anti-reflecting films and other optimization processes could be used to improve the performance further. Nano-sized tailored LSCO particles will also improve the cell performance due to its better percolation into the Ti0₂ pores and may remove the need for a contacting medium such as PMII. This work opens up an area of an entirely new type of charge mediators for SSC, based on a completely different mechanism and best fit to be manufactured and used in normal environmental conditions. Further, the absence of added iodine makes the charge mediators less corrosive. The design and synthesis of new perovskite compounds with higher ionic conductivities and very low electronic conduction will offer a significant potential for creating an all solid-state SSC with much higher energy conversion efficiencies. Example 3

Solar cell performance using perovskite charge mediators measured under 200 lux illumination intensity (suitable as indoor lighting conditions)

The performance of various DSSCs was measured under 200 lux illumination of fluorescent light, i.e., indoor lighting conditions, and the details and results are shown in Table 6 below.

Table 6. DSSC performance measured under 200 lux illumination of fluorescent light i.e. indoor lighting conditions. Performance ranges from 2.8 to 5.1 μνν/cm². Where open circuit voltage (Voc), short circuit current density (Jsc), fill factor (FF), power maximum (Pmax), voltage at maximum power (Vmpp) and current density at maximum power (Jmpp).

LSCO used above is La₀.₈Sro.₂Co0₃.₅.

Example 4

Stability tests

Open Cells stored in air (tested at 1 sun)

A batch of open cells (i.e. not sealed) were fabricated and stored in air atmosphere in the dark. Their performance at 1 sun was measured periodically. Excellent stability is observed over the first -600 hours as shown in Figure 3; an average of 90 % of initial performance is maintained after 624 hours of aging. Figure 3 illustrates stability testing of a batch of open cells containing La₀.₈Sro.₂Co0₃.₅ (LSCO)/BMII electrolyte stored in dark conditions in ambient air atmosphere during 650 hours. Their performance is normalised with respect to the efficiency on Day 1 , measured immediately after fabrication.

Example 5

Preparation and testing of solar cells using perovskite charge mediators deposited at the electrodes

Experimental details:

These solar cells consist of a working electrode (WE) operatively associated with a counter electrode (CE) and an electrolyte material (E), as illustrated in Figure 4. Three types of WE were employed in this study. WE A consists of a layer of DyeSol DSL 18NR-AO paste screen printed with a 43 T mesh, yielding a Ti0₂ thickness of 6-7 μητι after sintering at 500°C. WE B consists of an additional layer of Lao.8Sr₀.2Co03-5 (LSCO) screen printed on top of the Ti0₂ surface followed by additional sintering at 500°C. These electrodes were then immersed in 0.3mM N-719 dye solution in acetonitrile : tert- butylalcohol : THF (volume ratio, 4.5:4.5: 1 ) for 20 hours. WE C consists of a layer of LSCO screen printed directly onto the FTO surface followed by sintering at 500°C.

Counter electrode A (CE A) consisted of platinum coated FTO glass (deposition of a Pt catalyst via a drop of H₂PtCI₆ solution (2mg Pt in 1 ml ethanol) followed by a heat treatment at 400°C for 15min). CE B consisted of an additional layer of LSCO deposited by multiple methods (scrubbing, screen printing). This was followed by an additional heat treatment at 400°C-500°C. CE C consisted of a layer of LSCO deposited directly on the FTO surface, followed by heat treatment at 400°C-500°C.

A drop of the listed liquid fluid component (E) as described in Table 7 below was placed on the WE. Both working and counter electrodes were sandwiched together with clips using 50μιη thick Bynel polymer gasket as a spacer to make an open DSSC.

Figure 4 (on left side) shows a generic overview of the DSSC device structure employed in this study. Various combinations of working electrodes (WE) and counter electrodes (CE), some incorporating LSCO within the electrode structure, were tested. The listed liquid fluid component (E) may also be an ionic liquid such as PMII or BMII. The combinations tested are listed in Table 7 below. Table 7. DSSC performance measured at 1 sun. Various combinations of DSSC device structure incor orating LSCO at both the working and/or counter electrode.

Example 6

Quantum dot sensitized Solar Cell (QDSSC)

SSC was prepared with Ti0₂ (7 absorption + 2 scattering layers) sensitized with CdS QD as working electrode; Pt Counter electrode and Lao.sSro^CoOs-s in PMII as electrolyte material. Open cells prepared as in Example 4 above were tested at 1 sun and the results obtained are given in Table 8 below:

Table 8.

Example 7

Solar Cell prepared with CdS semiconductor photoelectrode

The compatibility of an electrolyte composed of the perovskite material Lao.sSro^CoOs-s (LSCO) dispersed in PMI I at 7% (w/w) was tested with CdS semiconductor photoelectrode instead of regular Ti0₂ photoelectrode. A thin layer of CdS semiconductor was coated over FTO glass by chemical bath deposition method. These working CdS semiconductor photoelectrode without any extra sensitizers were coupled with Pt counter electrode and Lao.sSro^CoOs-s in PMII as electrolyte material to make open solar cell device. The results obtained are given in Table 9 below:

Table 9.

It will be appreciated that the invention described herein may be modified or varied without departing from the scope of the appended claims.

Claims

Use of a perovskite oxide as a charge mediator in a sensitised solar cell (SSC), the perovskite oxide having the formula ΑΒ(Χ)₃._δ, wherein

A is an alkali metal ion, an alkaline earth metal ion, a transition metal ion, a p-block element cation or a rare earth metal ion;

X is an oxygen anion; and

δ is zero to 0.3.

Use according to claim 1 , wherein A is an alkali metal ion selected from lithium, sodium, potassium and cesium; or an alkaline earth metal ion selected from magnesium, calcium, strontium and barium; or a transition metal ion selected from iron and palladium; or a p-block element cation selected from aluminium, silicon, gallium, germanium, tin, lead, bismuth and thallium; or a rare earth metal ion selected from lanthanum, cerium, praseodymium, neodymium, samarium, gadolinium, erbium and ytterbium.

Use according to claim 1 or 2, wherein A is barium, calcium, strontium, lanthanum, cerium, praseodymium, neodymium or gadolinium.

Use according to claim 3, wherein A is lanthanum.

Use according to any preceding claim wherein A is doped with 1 or 2 elements selected from an alkali metal ion, an alkaline earth metal ion, a transition metal ion, a rare metal ion, and a p-block element cation, preferably barium, calcium, strontium, lanthanum, cerium, praseodymium, neodymium or gadolinium.

Use according to claim 5, wherein A is lanthanum doped with strontium.

7. Use according to any preceding claim, wherein B is an alkaline earth metal ion selected from magnesium, calcium, strontium and barium; or a transition metal ion selected from scandium, yttrium, titanium, zirconium, manganese, iron, cobalt, vanadium, niobium, tantalum, ruthenium, nickel and copper; or a rare earth metal ion selected from lanthanum, cerium, praseodymium, neodymium samarium, gadolinium, erbium and ytterbium; or a p- block metal element cation selected from aluminium, silicon, gallium, germanium, tin, lead , bismuth and antimony.

8. Use according to claim 7, wherein B is selected from manganese, iron, cobalt, germanium, lanthanum, cerium, praseodymium, neodymium, gadolinium, titanium and zirconium.

9. Use according to claim 8, wherein B is cobalt.

10. Use according to any preceding claim, wherein B is doped with 1 or 2 elements selected from an alkaline earth metal ion, a transition metal ion, a rare earth metal ion and a p-block element cation, preferably manganese, iron, cobalt, germanium, lanthanum, cerium, praseodymium, neodymium, gadolinium, titanium or zirconium.

1 1. Use according to any preceding claim, wherein the perovskite oxide is La-|._xSr_xCo03.₅, wherein x is 0.1 to 0.7 and δ is zero to 0.3, wherein, optionally, the perovskite oxide is

12. Use according to any preceding claim, wherein the perovskite oxide is deposited on an electrode or dispersed in an ionic liquid, ionic solid, eutectic ionic melt or solid state hole conductor.

13. Use according to claim 12, wherein the ionic liquid is selected from 1-methyl-3- propylimidazolium iodide (PMII), 3-hexyl-1-methylimidazolium iodide (HMII), 1-butyl-3- methylimidazolium iodide (BMII), 3-hexyl-1 ,2-dimethylimidazolium iodide (DMHII) and 1- ethyl-3-methylimidazolium tetracyanoborate (EMITCB); the ionic solid is selected from 1 ,3- dimethylimidazolium iodide (DMII), 1-ethyl-3-methylimidazolium iodide (EMM), 1-allyl-3- methylimidazolium iodide (AMII), 1 ,2-dimethyl-3-propylimidazolium iodide (DMPII) and N- methyl-N-butyl pyrrolidinium iodide (P ₄l); and the eutectic ionic melt is selected from DMII mixed with EMM, and DMII mixed with EMM and AMII. 14. Use according to claim 13, wherein the ionic liquid is PMII or HMII, and the eutectic ionic melt is DMII/EMII/AMII.

15. Use according to any preceding claim, wherein the charge mediator contains substantially no added iodine.

16. Use according to any preceding claim, wherein the charge mediator is used in the SSC at a temperature less than 100°C.

17. Use according to any preceding claim, wherein the sensitizer in the SSC is selected from dye molecules, quantum dots and nanostructured semiconductor materials.

18. Use according to any preceding claim wherein the SSC is selected from a dye sensitized solar cell (DSSC), a photoelectrochemical cell and a photogalvanic cell. 19. A sensitised solar cell (SSC) comprising a perovskite oxide-based charge mediator as defined in any one of claims 1 - 15.

20. A SSC according to claim 19, wherein the sensitizer is a dye molecule and the SSC is a dye sensitized solar cell (DSSC).

21. A SSC according to claim 19 or 20, wherein the charge mediator is dispersed on a dye- adsorbed working electrode which is operatively associated with a counter electrode.

22. A SSC according to claim 19 or 20, wherein the charge mediator is deposited on a working electrode and/or on a counter electrode.

23. A SSC according to any one of claims 19 to 22, which operates on an oxygen anion conduction mechanism.

24. A SSC according to claim 23, which is an open cell.

25. A SSC according to claim 23, which is a closed cell.

26. A SSC according to any one of claims 19 to 25, which is a photoelectrochemical or photogalvanic cell.

27. A SSC according to any one of claims 19 to 26, which is suitable for use in indoor light harvesting applications and/or in outdoor sunlight harvesting applications.

28. A perovskite oxide as defined in any one of claims 1 to 1 1 deposited on an electrode or dispersed in an ionic liquid, ionic solid, eutectic ionic melt or solid state hole conductor.

29. A perovskite oxide according to claim 28, which is La-|._xSr_xCo0₃.₅ as defined in claim 1 1.

30. A perovskite oxide according to claim 28 or 29, which is Lao_.sSro^CoOs-s_.