PHOTOVOLTAIC CELL
This application relates to a regenerative photo- electrochemical cell which is beneficially suitable for converting incidental light into electrical energy. The cell basically consists of a single substrate provided with current collector areas which are electronically insulated in relation to each other. One collector area is in electrical contact with a semi-conductor functional layer facing direct exposure to incidental light whereas the other collector area is electronically insulated from the semi-conductor functional layer by ion-permeable means.
The prior art relating to regenerative photovoltaic cells is crowded and diverse and, among others, possessed of detailed and precise knowledge concerning physical phenomena which can play a role within the framework of light conversion cells. EP-A-0.333.641 describes photo-electrochemical cells having a polycrystalline metal oxide semi-conductor with a monomolecular chromophore layer in its surface region whereby the metal oxide semi-conductor surface has a roughness factor of greater than 20, preferably greater than 150. At least one of the electrodes of the '641 reference shall be transparent to the incidental light. The photo-electrochemical cell in accordance with this reference is of the dual substrate "sandwich-type" configuration having oppositely charged current collectors attached to distinct substrates. The semi-conductor collector is provided with sintered (electrically conductive) nano-cells, such as Ti02. Kavan et al., J. of The Electrochemical Society,
February 1996, pages 394-400, have reported that sintering can
reduce the (BET) surface area of nanocrystalline Ti02 and henceforth its photo-electrochemical conversion functionality. Pechy et al., J. Che . Soc, Chem. Commun., 1995, pages 65-66, describe the preparation of phosphonated polypyridyl ligands which are reported to exhibit excellent charge-transfer sensitizer properties for nanocrystalline Ti02 film application. These characteristics render the phosphonated polypyridyl ligands particularly attractive for application in molecular photovoltaic devices.
Hanprasopwattana et al., Langmuir 1996, 12, 3173- 3179, describe methods of uniformly distributing monolayer to multilayer films of titania on mono-disperse silica spheres. The titania coatings so deposited are amorphous. Heating to temperatures around 500°C is needed to convert the amorphous coating into polycrystalline Ti02. Kay et al.. The Journal of Physical Chemistry 1996, Vol. 98, pages 952-959, report on the mechanism of photosensitization of transparent Ti02 electrodes with chlorophyll derivatives, particularly induced photocurrent phenomena. Gratzel et al.. Current Science, Vol. 66, No. 10, 25 May 1994, describe properties and requirements of efficient dye- sensitized photoelectrochemical cells for direct conversion of visible light to current. The authors conclude that actual cell technology (at the date of the paper) is not sufficient and cannot lead to commercially viable executions. McEvoy et al.. Solar Energy Materials and Solar Cells 32 (1994) 221-227, summarize general principles and the historical development sequence concerning photovoltaic technologies.
While significant efforts have been extended towards the development of photovoltaic cell technology which can be used commercially, these efforts have not yielded
commercially viable executions. Past approaches were almost quantitatively directed to sandwich-type structures which are known to be difficultly manufacturable, in part because of constraints inherent to producing effective conductor areas.
It is therefore a main object of this invention to make available photovoltaic cells which can be manufactured economically and which are capable of efficiently converting incidental light into electrical energy. It is another object of this invention to make available cell structures which do not require light transparent electrode arrangements and which can be manufactured economically and operated efficiently commensurate with continuity requirements. It is still another object of this invention to provide photoelectrochemical cells which do not require light transparent electrode arrangements and which are constructed onto a single substrate.
This invention, in part, relates to regenerative photovoltaic cell comprising current collector areas which are deposited onto a (single) substrate, said areas being physically insulated in relation to each other, one of the areas being in electrical contact with a dye coated semi-conductor functional layer, said semi-conductor layer facing direct exposure to incident light and comprising electronically-interconnected nano-cells, the other current collector area being physically insulated from the nano-cells by ion-permeable means, said collector areas being in electrical contact with an electrolyte medium. The term "direct" light exposure defines an arrangement whereby no photovoltaic (PV) functional element, such as conductive glass and/or a conductive cover, is present in between the semi-conductor, comprising the electrolyte, and the source of (incident) light. It is understood that (non-PV
functional) protective cover, such as glass or plastic or other non-functional means can be used to e.g. protect, and secure the integrity of, the electrolyte upon exposure to atmospheric surroundings .
In preferred executions of the invention, the current collector areas are constructed/deposited onto a single substrate. The collector areas can be located in adjacent position on the same side of the substrate or said current collector areas can be arranged on both sides of the substrate. Flexible substrates can be used within the preferred executions of this invention. The dye is capable of injecting electrons into the semi-conductor layer upon exposure to light having a wave-length in the range of from 300 nm to about 900 nm, more preferably from about 400 nm to about 700nm. The, preferably flexible, substrate can comprise metals or mineral materials or organic polymers such as polyethylene therephthalate (PET) , polyalkylene such as polyethylene, polypropylene, polybutylene polystyrene and combinations thereof, glass, ceramics, cloth, fibers, binder resins and combinations of such materials. The particulate material in the semi-conductor functional layer can be impregnated with an electrolyte. The substrate generally is represented by an electronically insulating material, or a combination of such insulating materials, and (substrate) is in physical contact with and/or can serve as support for at least one of the electrodes. The other current collector electrode is insulated from the semi-conductor via physical methods such as patterned printing or ion permeable physical separation such as porous materials including a ceramic, a porcelain or a polymer matrix. Another optional design might involve metallization of both sides of a porous insulating membrane, with no short circuit
between opposite sides and the functionally active material being deposited on one side, the counter electrode on the other side, the electrolyte filling the porous membrane. Another method would be suitably coated wire electrodes to either side of the porous membrane, as above. The single-substrate photovoltaic cell execution of this invention is preferably and generally covered by a transparent protective layer, such as plastic, glass, which layer is substantially impermeable to water, organic vapors and oxygen.
A cross-section of the photovoltaic cell having current collector areas on one side of a substrate is represented in Fig. 1.
A cross-section of the photovoltaic cell having having current collector areas on opposite sides of a substrate is represented in Fig. 2.
Figure 3 represents a planar view of an interdigitated execution of a photovoltaic device in accordance with the invention herein consisting of two cells connected in series.
The single-substrate of the inventive cell can be represented by carrier materials which are known to be suitable for the intended carrier functionality. Examples of suitable substrate are represented by the materials recited hereinafter. The substrate generally can be represented by flexible or rigid, mineral or organic, natural or synthetic polymers for example polyethylene therephthalate (PET) , polyalkylene such as polyethylene, polypropylene and polybutylene and mixtures thereof, glass, ceramic, cloth, fibers, binder resins and
combinations of any two or more of the preceding materials. The ultimate selection of the substrate material will depend upon the final execution of the cell. Flexible substrates based on the foregoing organic polymeric material can be desirable. Other suitable flexible substrates are polystyrene, aramid, kevlar and comparable materials. The substrate generally exhibits physical stability under conventional operating conditions. The substrate shall, in the event it constitutes part of the (protective) outer layer of the cell, preferably be substantially impermeable towards water, organic vapors and oxygen to thus preserve substantially unaltered the physical properties of the substrate and of the photo-electrochemical system. The substrate shall be permeable to the electrolyte in the event said substrate is provided, for example, with current collector areas on opposite sides i.e. the substrate is not subject to direct exposure and alteration.
The substrate is provided with current collector areas which areas are electronically insulated in relation to each other. One area is in contact with/covered by a dye coated semi-conductor functional layer comprising electronically- interconnected nano-cells. This functional layer faces direct exposure to incidental light. A conductive layer is deposited onto the substrate underneath the semi-conductor functional layer. The conductive layer can be represented by conductive paints, polymers or (pure) metals having established electrically conductive properties. Examples of suitable metals include titanium, tantalum, molybdenum, stainless steel (302; 304; 316; 321; 310; 17-7PH;), Hastelloy C-276, tungsten, inconel, zirconium and comparable metals or metal alloys or combinations thereof well-known in the relevant domain. The terms: 302; 304; 316; 321; 310; 17-7PH; Hastelloy C-276; and
SUBSTITUTESHEET (fiULE26)
iconel have the meaning given in PERRY' s CHEMICAL ENGINEERINGS' HANDBOOK, 6th Edition, R.H. Perry, D.W. Green, J.O. Maloney, Mc Graw Hill Book Company, 1984.
The dye coated semi-conductor functional layer collects the electrons/electricity generated upon exposure to incident light. The semi-conductor functional layer comprises a structure capable of exhibiting semi-conductor or quasi-semi- conductor properties. The layer can be represented by a dye coated metaloxide semi-conductor. The semi-conductor component which is preferably characterized by a polycrystalline structure can be selected from transition metal oxides or of an element selected from the fourth, fifth or sixth group or sub-group of the periodic system of elements, in particular titanium, zirconium, hafnium, strontium, zinc, indium, tin, antimony, yttrium, lanthanum, vanadium, niobium, tantalum, chromium, molybdenum, tungsten, oxides of zinc, iron, nickel or silver, mixed oxides, and/or oxide mixtures of two or more of these metal oxides. In a preferred execution the semi-conductor layer comprises a polycrystalline Ti02, or niobium dioxide, most preferably the anatase form of Ti02# layer deposited onto a conductive layer. The Ti02 semi-conductor layer can be prepared in accordance with the SOL-GEL process (Stalder and Augustynski, J. Electrochem. Soc. 1979, 126, 2007) as applied in EP-A- 0.333.641, page 2, lines 48-56. The semi-conductor particles herein are nano-crystalline having a particle diameter in the range of from 1 to 200 nm, preferably from 2 to 100 nm, particlurly from 10 to 60nm.
The semi-conductor layer is coated with a chromophore dye. The coating is preferably a monomolecular layer. The chromophore dye releases photoexcited electrons under
the influence of incidental light. The electrons originated from the photoexcited dye are irreversibly injected into the semiconductor functional lattice and are consequently separated spatially and by a potential barrier from the oxidized dye coating. The spectral sensitivity can vary depending, among others, upon the chemical structure of the dye. The selection of a suitable dye can henceforth be optimized to thereby take into considerations relevant light parameters with a view to enhance the conversion of light to electrical energy. Chromophore dyes which can be used in regenerative photovoltaic cells are well known in the relevant domain. Suitable chromophore dyes can, for example, be represented by a transition metal complex of the type ruthenium tris(RuL3), ruthenium bis(RuL2), osmium tris(OsL3), osmium bis(OsL2) or ruthenium cis diaqua bipyridyl complex of the type RuL2 (H20) 2, such as ruthenium cis-diaqua bis (2,2'-bipyridyl-4, ' ) -dicarboxylate . A preferred class of photosensitizer dyes for use herein has the formula (X)nRuLLι where n is 1 or 2, Ru is ruthenium, each X independently is selected from Cl, SCN, H20, Br, I, CN, -NCO and SeCN, and L and Li can be a ligand having a formula appearing in WO 94/04497 page 2, last paragraph to the end of page 3. More preferably X is X' where X' is Cl, CN, -NCO or -SCN. Specifically preferred species include cis (X) 2bis (2, 2' -bipyridyl-4, 4' -dicarboxylate) - ruthenium(II) complexes where X is Cl, Br, I, CN and SCN. The thiocyanato compound is eminently useful. The chromophore dye can be attached to the surface of the semi-conductor by che isorbtion, adsorption or comparable techniques capable of yielding adequate and economically viable conversion of light to electrical energy. It has been established that good results can be obtained in the event the chromophore dye is attached to the metal oxide semi-conductor by means of carboxylic acid ligands. The chromophore dye layer can also be represented by metal or
non-metal complexes of phthalocyanin or porphyrin.
The semi-conductor functional layer is impregnated with an electrolyte solution. The electrolyte participates regeneratively in the charge transport by acting as an electron donator in relation to the chromophore dye to thus compensate for the loss of photo-excited electrons which have irreversibly entered the semi-conductor functional layer e.g. the titaniumdioxide polycrystalline particles deposited onto the conductive layer. The electrolyte solution for use within the photovoltaic cell herein can be represented by lithium iodide, bromide, hydroquinone and other comparable species which are well known in the relevant domain and have found application. The electrolyte is used in conjunction with a suitable solvent such as acetonitrile or similar organic nitriles and polynitriles and in conjunction with iodine. The electrolyte serves for the transport of electrical charges between the counter electrode and the working electrode as for example described by Brian 01Regan et al.. Nature, Vol. 353, pages 737- 740.
Fig.l represents a cross-section of a photovoltaic cell execution of this invention having current collector areas on one side of the substrate. The cell 10 comprises a substrate 1. The substrate can basically be made from all materials having established functionality in the domain of regenerative cells. The substrate is preferably substantially impermeable towards water, organic vapors and oxygen. The substrate can be represented by flexible or rigid executions of mineral or organic polymeric materials such as glass, polyethylene therephthalate (PET) , polyalkylenes particularly polyethylene, polypropylene, polybutylene and
mixtures of such polyalkylenes, cloth, fibers, binder resins and combinations thereof. It is understood the selection of the substrate can vary depending upon the intended application of the claimed arrangement. The guiding criteria shall be that the substrate exhibits desirable physical stability under the selected usage conditions.
The conductive layer 2 is deposited on the substrate and represents a current collector area (-) . The conductive material can be represented by conductive materials including metals such as titanium, tantalum, molybdenum, metal-alloys and combinations thereof all well-known in the relevant domain. Non- metallic conducting materials such as carbon and conductive polymeric materials can also be used beneficially.
A semi-conductor functional layer 3 comprising nanoparticles, interconnected by semi-conductor functional links such as Ti02( is deposited onto the conductive layer. While the semi-conductor component can be selected from a large variety of e.g. metal oxides, a polycrystalline Ti02 is a preferred execution. In another preferred aspect herein, a supplementary layer substantially constituted of semi-conductor oxides or combinations thereof including, for example, Ti02, Sn02, oxides of indium, antimony or combinations thereof, is deposited onto the conductive layer before the depositing of the functional layer 3. This supplementary layer can facilitate and enhance the electrical contact between the conductive layer and the semiconductor layer The supplementary layer containing semiconductor oxides is, within the structure of the Fig. 1 cell, preferably non-permeable to the electrolyte. In still another preferred execution, a minor proportion of semi-conductor metal oxide particles is embedded into the conductive layer.
Preferably from 1% to 10% by weight, expressed versus the conductive material (100%) of said metal oxide particles are embedded in the conductive material, particularly the same metal as in the metal oxide. The semi-conductor functional layer is coated with a chromophore dye, preferably in a monomolecular layer.
The regenerative electrolyte 4 which is within the interstices/pores of the semi-conductor functional layer is preferably represented by lithium iodide although other electrolyte species which have found application in comparable cell technology can also be used beneficially. The electrolyte solution has preferably a concentration in the range of from 1 molar to 10"4 molar, preferably from about 10"α molar to 10"3 molar.
A transparent protective layer 5 is deposited onto the semi-conductor functional/electrolyte impregnated layer. The transparent material can be represented by glass and a variety of polymeric materials which are substantially impermeable to water, organic vapors and oxygen.
The counter electrode (+) 7 is electrically isolated from the semi-conductor functional layer by means of ion-permeable insulating material such as porous-, ceramic-, porcelain-, and/or polymeric coating material 6 or a space filled with the electrolyte. Interdigitated areas could be arranged where 3,4 are only present on top of 2, and the electrolyte on top of both 7 and 3,4, with just a spatial separation between 7 and 2. The ion-permeable insulating material is deposited onto the conductive material layer 7. The conductive material used in 7 can be substantially similar to
the conductive material used in layer 2 or, at least, selected from the conductive materials recited in relation to layer 2. It may be useful to incorporate into conductive layer 7, more preferably to deposit onto its (7) surface, low, catalytic, levels of a metal oxido-reduction catalyst of the platinum family.
The counter electrode 7 or 6 and 7 are insulated from the oppositely charged counter electrode, particularly from the conductive layer by means of suitable insulating materials 8 in a known manner. The insulating material can be represented by known materials such as natural or synthetic polymers, resins, rubber, paint, coating, fibrous material, porcelain and so on.
Fig. 2 represents a cross-section of another preferred cell in accordance with the invention herein. The cell 20 comprises a substrate 16 deposited onto which are current collector areas 12 and 17. The conductive material employed in areas 12 and 17 can correspond, structurally, physically and chemically to the conductive materials 2 and 7 as described in relation to cell 10. The substrate can be represented by ion permeable material such as porous ceramic, porcelain, polymer or fibrous material. In one specific execution, the substrate can be represented by a porous insulating membrane both sides of which have been metallized with no short circuit between opposite sides. The conductive material can, more generally, be represented by metals such titanium and other metals and metal alloys as described above. Non-metallic conducting materials such as carbon and conductive polymeric materials can also be used beneficially.
Conductive materials broadly can be applied to
the substrate by conventional techniques well-known in the relevant domain. Examples of suitable metal deposition technologies include vacuum condensation, sputtering, electro- deposition, lamination, painting and combinations of such technologies. Non-metallic conductive materials can be applied, depending upon the physical form of the conductive material and its coating properties. Metal and carbon paints can be applied by means of conventional techniques subject to optimized conditions. Increased temperatures e.g. up to 150 °C or even more can provide benefits. The non-metallic conductive material can broadly be represented by conductive coatings having preferably a conductivity of less than about 10 ohm/square. Such conductive materials can be represented by a polymeric, intrinsically non- conductive matrix, containing a minimal quantity, generally from about 1% to 15% by volume expressed versus the matrix material (100%) of a high aspect ratio fibrous material, for example at least 10, generally more than 40, preferably more than 100, capable of conferring conductivity properties as may be required for application in the cell of this invention. The matrix material can be represented by polymer compounds including styrenic polymers and copolymers, aramid, kevlar, and more generally any polymeric mass adapted and compatible to minor levels of materials capable of conferring conductivity properties. Examples of the like materials are fibrous carbon, metals, metal wires, metal meshes and screens, metal chips and metal powders and similar structures of insulating material coated by a conductive metal layer
Protective layers 11, 15 of cell 20 can structurally, physically and chemically correspond to protective layer 5 of cell 10. Similarly, semi-conductor functional layer 13 and electrolyte 14 of cell 20 correspond to layer 3 and
electrolyte 4 of cell 10. Electrolyte 14 must extend through current collector 12 and substrate 16 to contact current collector/ electrode 17.
The photovoltaic cell herein can be useful for converting incidental light into electrical energy. Generally visible light which can be converted has a wavelength in the range of from 300 nm to around 900 nm. Daylight having a wavelength in the range of from 400 nm to 700 nm is for obvious reasons a prime source of light. It is understood, however, that any particular cell in accordance with this invention can be routinely adapted and optimized to any predominant source of visible light in the range of from 300 nm to 900 nm or more generally to any light source the radiation energy of which can be converted beneficially into electrical energy.
The benefits of the claimed subject matter are illustrated with the aid of the showings below.
Examples .
A slurry of titaniumdioxide nanoparticles was prepared as follows. 40 gs of Ti02 having an average particle diameter of 21 nm and a surface area in the range of from 35-65 m2 (Anatase; DEGUSSA P-25 ™) was mixed with: 60 gs deionized water; 10 drops of a 10% by wt. aqueos nitric acid solution; 10 drops of non- ionic fluoroaliphatic polymeric ester suplied by 3M (Fluorad FC-430™); and 10 drops acetylacetone.
Example 1:
An interdigitated titanium electrode was prepared as follows. A
5cm x 5cm piece of glass was sputter coated, on one side, with titanium up to a layer thickness of 5000 angstroms. Prior to effecting the sputtering, a tape having a width of 1 mm was placed on the glass to thus delineate two sets of interdigitated electrodes. The tape was removed after sputtering and two electrically isolated electrodes were formed on the glass. Using a paintbrush, the "teeth" of one electrode were coated with the anatase slurry described above. The "teeth" of the other electrode were coated with a platinizing solution containing:
0.486 g K2PtCl6; 0.05 g hydroxyethyl cellulose; and 1 drop of a
25% by wt. KOH solution in 10 ml. deionized water. The electrode assembly was then heated to 450°C for 30 minutes under nitrogen.
After cooling to 80°C, the assembly was dipped into a ruthenium dye solution (21.15 mg RuL2(SCN)2 - L = 4, 4'-dicarboxy-2,2'- bipyridyl -, CH3CN (50 mL) and t-BuOH (50 mL) . After rinsing
(CH3CN/t-BuOH) and drying at ambient temperature in a nitrogen stream, the assembly was placed into an "envelope" of transparent heat-sealed polyester with the top of each electrode protruding.
Electrolyte solution (0.3 M Lil; 0.03 M I2 in 3- methoxypropionitrile) was added to cover the electrodes. The cell was then irradiated with a 500 watt photoflood bulb at a distance sufficient to produce a response equivalent to "one sun" on a calibrated silicon solar cell (Edmund Scientific) .
An ammeter connected accross the two electrodes showed a current of 0.5 mA and a voltmeter showed a voltage of 0.38 volt. Correcting for the area of the electrode where the Ti02 coating
had fallen off, current density on the electrode was : 0.42 mA/cm2.
Example 2 :
A slurry of titanium dioxide particles was prepared as follows: 40 gs of Ti02 powder (DEGUSSA PM-25 ™) , as described in Example 1) was finely ground in a mortar with 20 gs of an aqueous solution containing 10 drops, per 100 mis, of 10% by weight nitric acid solution and 3% by weight of polyethylene glycol (average molecular weight 1500; supplied by Janssen Chimica) . More of the aqueous solution was gradually added to the paste and incorporated by further grinding until a total of 100 gs of paste was obtained. The paste was then vigorously ground for a further 10 minutes. Alternatively, the paste was ground further in a planetary ball mill (PULVERISETTE 7 ™ from FRITSCH) .
An interdigitated doped tin oxide-coated glass surface was prepared as follows. A 5 cm x 5 cm piece of glass coated with fluorine doped tin oxide (FTO surface conductivity 8 ohms/ square; supplied by Libby-Owens-Ford) was etched with a pointed diamond hand tool in order to form a set of two interdigitated comb-shaped electrodes electrically insulated from each other. A sufficient pressure was applied to the tool to ensure that the conductive FTO layer was fully etched through. Using a paintbrush, the teeth of one electrode were coated with a layer of the Ti02 slurry. The teeth of the other electrode were coated with a platinizing solution as described in Example 1. The coated electrode assembly was allowed to dry slowly in air/under ambient conditions followed by gradually heating the assembly to
450 °C in a flow of hot air. The assembly was kept at that temperature (450 °C) for a period of 20 minutes and then allowed to cool down to 80 °C in a desiccator. The electrode assembly was then immersed overnight in a ruthenium dye solution as described in Example 1, quickly dried in a flow of dry nitrogen and immersed in a glass cell, made from two 7 cm x 7 cm glass sheets sealed on three sides with commercial room temperature vilcanisable (RTV) silicone rubber, containing enough of the electrolyte solution, in accordance with Example 1, to cover all but the top of the electrodes. The protruding top of each electrode was electrically connected to an ammeter. The cell was then irradiated, perpendicularly to the surface of the electrodes with a 500 watt flood lamp. The lamp was positioned in such a way as to produce a response equivalent to one-tenth of a sun on a calibrated silicon solar cell (Edmund Scientific) at the same distance.
The ammeter showed a voltage of 0.41 volt. The current density on the Ti02 electrode was 0.12 mA/cm .22
Example 3:
Using the etching technique described in Example 2 above, a set of four electrodes was scribed onto a 5 cm x 5 cm FTO coated glass sheet. The elctrodes were arranged as shown in Figure 3, the dotted areas being coated with Ti02 paste, hence forming two distinct photoelectrodes, and the striped areas being coated with platinizing solution thus forming two distinct counter- electrodes. The first photoelectrode (21) is electrically insulated from from the first counterelectrode (22) . The first counterelecktrode is electrically connected to the second photoelectrode (23) . The second counterelectrode (24) is
electrically insulated from the second photoelectrode. This arrangement therefore realizes two sets of photoelectrochemical cells, analogous to those described in Example 2, electrically connected in series. This assembly was fired, impregnated with ruthenium dye and mounted in a cell filled with electrolyte solution as described in Example 2.
An ammeter was connected accross (25) to the first photoelectrode (21) and the (26) to the second counterelectrode (24) and the assembly was irradiated to one-tenth sun. The ammeter showed a voltage of 0.74 volt and a current density of 0.09 mA/cm2 of photoelectrode.
Example 4:
A double-sided titanium-glass photovoltaic cell was prepared as follows. A piece of glass, 5 cm x 5 cm and 3 mm thick, was sputter-coated with titanium on both sides to a thickness of about 2000 angstroms and a sheet resistance of 1 ohm/square. A strip of 1.5 cm wide and 5 cm long was cut from the so-prepared metallized glass and used as the electrode pair. Within an area of 1.5 cm x 1.5 cm at the end of the strip, 22 holes (diameter: 0.075 mm) were drilled in evenly spaced rows using a diamond-tipped drill bit. One side of the drilled area was then platinised by spraying with a fine mist of hexachloroplatinic acid in isopropanol at a concentration of 5 itiM followed by heating to 390 °C for 30 minutes. The reverse side of the drilled area was then coated with the anatase slurry as used in Example 1. The so-prepared device was heated to 450 °C for 30 minutes and subsequently soaked overnight in a ruthenium dye solution as described in Example 1 except at a dye concentration of 9 x 10"5 in ethanol.
The dried device was then placed in a heat-sealed polyester bag with the non-coated end protruding. The bag was filled with an electrolyte solution - 0.5 M Lil + 0.05 M iodine in acetonitrile.
An ammeter was connected across the protruding titanium coatings on opposite sides of the glass and when irradiated, as described in Example 1, to produce a response equivalent to one sun, an open circuit voltage of 0.498 V and a short circuit current of 0.184 mA/cm2 were measured.