Efficient dye sensitised photovoltaic devices employing nanocrystalline metal oxide films were first reported in 1991, see B. O'Regan and M. Gratzel, Nature. 1991,353,737.

Such cells can achieve solar to electrical energy conversion efficiencies of up to 10%.

There is currently extensive commercial R&D aiming to develop photovoltaic devices based upon this design. centred around Prof. Gratzel's group in EPFL. Switzerland.

Previous investigations have searched for the identity of the optimum sensitiser dye for this technology.

An important limitation in the design of dye sensitised PV devices currently being commercialised is the requirement for a electrically conducting liquid component (an electrolyte). It has been proposed to replace this liquid with a solid state electrolyte analogue. The important requirements for such solid state analogues are good device energy conversion efficiency, good stability and low sealing requirements. Gratzel and co- workers have reported that an organic material, OMeTAD developed for the Xerographic industry, is one such material. although the efficiencv of the solid state device produced with this material was limited to < 0.8 %, see Back et al., Nature 1998.395.583. Other proposals have considered conducting polymers, but have to date achieved efficiencies even less than the above, see Murakosh. et al., Chem Letts., 1997,471. Polymer gel electrolytes have also been shown to achieve high efficiencies, however such system retain a solvent phase and therefore still require sealing, and cannot therefore be regarde as truly solid state, see Cao et al.. J. Phys. Chem, There has been a report of the use of solid state ionic commercial rubber. but efficiencies achieved were verv low (O 1%), see Nogueira et al.. Abstracts IPS-12.

Various photovoltaic and battery cells are described in US-A-5. 441. 827. US-A- 5.438.556. US-A-4.5'0.086. WO-A-97/08719 and WO-A-93/'0569

Viewed from one aspect the present invention provides a photovoltaic device including an ionic conducting material as a solid state electrode.

Ionic conducting materials have been found to provide good energy conversion efficiencies, good stability and require little or no sealing.

Preferably the ionic conducting material comprises molecular, oligomer or polymeric components.

The use of ionic conducting materials as an electrode in a photovoltaic device is particularly well suited to photovoltaic devices including a nanocrystalline particle film.

The nanocrystalline particle film could have various forms, but it is preferably a nanocrystalline metal oxide film.

Metal Oxide films, such as titanium oxide, are well suited to having electron injection into the titanium oxide conduction band followed by electron conduction through the titanium oxide particles to a electrode.

In the context of a photovoltaic device including a nanocrystalline particle film, the ionic conducting material is preferably intercalated with the nanocrystalline particle film such that it can act as an electrode to carry away a hole following the absorption of a photon and the injection of an electron into the conduction band of the nanocrystalline particle.

Conversion efficiencies of such devices are significantly improved when the nanocrystalline particle film is dye sensitised.

The suitability of the ionic conducting material for use as an electrode is significantly enhanced when the ionic conducting material is doped.

The conductivity of the ionic material may be further enhanced by the addition of electronic hole transporting components, including, for example triaryl amine compounds.

Such hole transporting components may or may not be covalently bound to other components of the ionic conducting material.

The effectiveness of the ionic conducting material as an electrode is further enhanced when it is mixed with redox active components, including salts, such as lithium iodide.

The ionic conducting material can have various forms. A particularly preferred ionic conducting material is polyethyleneoxide and varients thereof.

Surprisingly high efficiency, high stability and low sealing requirements have been found in photovoltaic devices including polyethyleneoxide, iodine and lithium iodide as a mix providing the role of an electrode.

A particularly suitable disposition for the various components of the photovoltaic devices is one in which the ionic conducting material and the nanocrystalline particle film are disposed on one of a conducting glass substrate and a conducting plastic substrate.

Incident photons can pass through the transparent plate where they are absorbed by the photovoltaic device leading to electron transport via the nanocrystalline particle film and hold transport via the ionic conducting material.

The present invention is particularly well suited for use in photovoltaic devices including iodine and that are unsealed. Iodine is normally considered relatively volatile and it is surprising that these devices show good stability obviating the need for sealing.

The ionic conducting polymer and nanocrystalline particle film layer can have various thickness'. Preferably the thickness of this layer is between five and fifteen micrometers and still more preferably the thickness of the layer is substantially ten micrometers.

The nanocrystalline particle film could be formed of various different substances providing that they have the correct physical and electrical properties. A particularly preferred option is the use of titanium oxide nanoparticles. Preferably such nanoparticles are between five and fifty nanometers in diameter, and still more preferably substantially ten manometers in diameter. An alternative option is the use of zinc oxide nanoparticles.

Viewed from another aspect the present invention provides a method of making a photovoltaic device, said method comprising the steps of : forming a porous nanocrystalline particle film; and forming an ionic conducting material intercalated with said nanocrystalline particle film, said ionic conducting polymer film serving as a solid state electrode.

It is surprising that the ionic conducting material with its relatively high molecular mass is able to intercalate effectively with a porous nanocrystalline particle field and so provide good function as an electrode within the photovoltaic device.

The porous nanocrystalline particle film could be formed in a number of different ways. Preferably, the nanocrystalline particle film is formed by one of : use of an aqueous sol/gel; thermolysis in a high boiling point solvent; chemical vapour deposition; spray pyrolysis; and spreading and spin coating.

An embodiment of the invention will now be described, by way of example only, with reference to the accompanying drawings in which:

Figure 1 schematically illustrates a photovoltaic device according to one example embodiment of the invention.

Figure 1 shows a photovoltaic device 2. The photovoltaic device 2 includes a porous film of titanium oxide nanocrystalline particles 4. These titanium oxide nanocrystalline particles 4 (or zinc oxide nanoparticles) may be formed by one or more of the use of an aqueous sol/gel, thermolysis in a high boiling point solvent, chemical vapour deposition, spray prolysis and spreading and spin coating. The titanium oxide nanocrystalline particles preferably have a diameter between five and fifty manometers and still more preferably of substantially ten manometers. The titanium oxide nanocrystalline particles 4 are dye sensitised with a dye layer 6. This dye layer is preferably formed of one or more of types of sensitiser dye, such as bipyridyl, porphyrin and phthalocyanine sensitiser dyes.

An ionic conducting polymer (or material in molecular or oligomer form) electrode 8 is intercalated with the porous film of titanium oxide nanocrystalline particles. The ionic conducting polymer 8 is preferably a mixture of polyethyleneoxide, lithium iodide and iodine and varients thereof. The ionic conducting polymer 8 may be further adapted to its role as an electrode for conducting holes by doping with tyiaryl amine components. The porous nanocrystalline titanium oxide particle film 4 and the intercalated ionic conducting polymer 8 preferably form a layer of between five and fifteen micrometers in thickness and still more preferably of substantially ten microrreters in thickness. One face of this layer is bounded by a glass plate 10. A platinised or tin oxide film 12 on the glass oxide plate 10 serves as an electrode. A tin oxide film 14 at the other face of the layer serves as the other electrode.

In operation absorption of a photon by the sensitiser dye results in electron injection into the titanium oxide conduction band. Electron conduction into the conducting glass contact 14 proceeds through the titanium oxide nanocrystalline particles 4 whilst holes are

conducted by the ionic conducting polymer electrode 8 through the pores in the titanium oxide film to the platinum or conducting glass counter electrode 12.

As mentioned previously iodine is mixed in with the ionic conducting polymer electrode. Iodine is a volatile liquid. It would have been expected that the iodine would relatively rapidly evaporate from an unsealed cell rendering it inoperative and an unsealed cell approach unworkable. Surprisingly, it has been found that evaporation from the unsealed cell is slow and that the cell has good stability.

The polyethyleneoxide used as the ionic conducting polymer may have an RMM within the range 10,000 to 50,000 and preferably approximately 20,000. It would have been expected that molecules of such a large size would not have allowed intercalation into the pores between the nanocyrstalline titanium oxide particles and therefore only allowed a poor cell efficiency. Surprisingly, it has been found that even such large molecules yield good efficiency which either indicates that intercalation is occurring or that intercalation is unimportant for the efficiency. The polyethyleneoxide has been found to yield a surprisingly high conductivity when formed at a temperature within the range 50°C to 100°C and preferably approximately 70°C and at a pressure within the range of 0.25kg/cm2 to lkg/cm2 and preferably approximately 0.5kg/cm2.

Examples of the invention have achieved efficiencies approximately twice as efficient as the best solid state dye sensitised device reported to date, with excellent device stability. Further refinement is likely to result in further significant improvements in performance. A summary of results to date is attached.

The invention is based upon the use of an ionic conducting polymer: polyethyleneoxide/LiVI2. The system is solvent free, and stable cell outputs have been observed for in excess of one month without any cell scaling. The efficiencies of the device is further improved by the addition of triaryl amine components as hole conductors.

Solid State Electrolvtes, Progress 31/3/99 HTM Solvent Efficiency Efficiency 0.1 stability 1 sun sun PEG/LiI/I2 Ethanol 0.2% 1% 20% loss over one month PEG/LiI/I2/TPD Acetone 0.4% 1.3% <10% over 1 month PEG/LiI/I2/PC/EC None 0.5% 1.4% Not Done -Efficiencies measure 2 days after fabrication and after evapouration of solvents.

All cells unsealed.

-Cell Type 2 (+TPD) 2.5% efficient at 0.1 sun, 4 hours after fabrication, but reduced to table values over 2 days, efficiency at 0.1 sun then stabilised for one month (1 sun efficiencies dropped to 0.2%) -All cells fabricated with transparent Ti02, no Ti02 backing layer, nonreflective counter electrode (equivalent liquid electrolyte cell efficiency 5-6%).

-Best previous published solid state cell data 0.78% at 0.1 sun (Back et al., Nature 1998) As an alternative to polyethyleneoxide it would be also possible to use molecular hole transporting materials such as TPD and derivatives thereof It will be appreciated that the devices described above may also have bioanalytical applications. Alternatives to metal oxide nanoparticles include semiconductor sensitiser nanoparticles. It is possible that low temperature fabrication techniques would make such material suitable for deposition upon plastics.