GB2405030A

GB2405030A - Bifacial thin film solar cell

Info

Publication number: GB2405030A
Application number: GB0318906A
Authority: GB
Inventors: Ayodhya N Tiwari; Gennadiy Khrypunov; Fiodor Kurdzesau; Derk Leander Batzner; Alessandro Romeo
Original assignee: ETH; Loughborough University
Current assignee: ETH; Loughborough University
Priority date: 2003-08-13
Filing date: 2003-08-13
Publication date: 2005-02-16
Also published as: GB0318906D0

Abstract

A thin film photo-voltaic device formed on a transparent substrate 41 comprises a p-type semiconductor layer 44, and an n-type semiconductor layer 43 forming a p-n junction at the interface with the p-type layer 44. Transparent conducting oxide (TCO) contacts 42 46 are provided on the p-type 44 and n-type 43 layers to form a device which converts radiation 40 47 incident on both front and back surfaces into electricity. The surface 45 of the p-type layer 44 contacting the TCO contact 46 is modified such that a buffer layer between the p-type layer 44 and the TCO contact 46 is not required. The photo-voltaic device can be stacked in tandem formation with other photo-voltaic devices to gather radiation at different wavelengths.

Description

THIN FILM PHOTO-VOLTAIC DEVICE

The present invention relates to a photo-voltaic device for generating electricity when illuminated by a light S source. In particular, but not exclusively, the present invention relates to a solar cell having good stability, which is double sided and transparent to optical radiation at both a front and back surface, and which may be stacked in a tandem arrangement.

As today's supplies of non-renewable energy sources become depleted more efforts are being made to provide renewable energy sources. One area of research is the development of photo-voltaic cells. In such photo voltaic devices light from the sun or other light source which is incident on the photo-voltaic device is converted into electricity. In many of these photo- voltaic devices a pen junction is formed between an n- type and p-type semi-conductor. When a photon of light with an energy greater than the band gap of the semi- conductor passes through the cell it may be absorbed. It is well known that when this absorption occurs an À.r. electron/hole pair is produced. These carriers may À diffuse to the depletion region of the pen junction 25 before they recombine. In this case a quantum of charge 2. is caused to flow. This is the origin of a solar cell's photo current and hence a generated electricity source. À: À..

À *. Many photo-voltaic devices currently use a single-crystal of silicon. However cells produced in this manner are expensive if a large size (that is a cell having a large surface area) is required. A further type of photo voltaic device which has been developed is the thin film cadmium telluride/cadmium sulphide solar cell. These types of device have many advantages over the single- crystal silicon cells. For example since the cell is made from polycrystalline materials on glass or other substrates the overall construction process is far cheaper than construction with bulk silicon. Also the material properties of the semi-conductors mean that thin layers of the semi-conductors can be provided using a variety of techniques such as chemical vapour deposition (CVD), chemical bath deposition and physical vapour deposition (PVD).

In view of the above observations polycrystalline CdTe/CdS thin film solar cells are one of the most important types of photo-voltaic devices which provide a cost-effective and clean generation of solar electricity for terrestrial applications.

In addition recent measurements of the photo-voltaic performance of CdTe solar cells irradiated with high energy protons and electrons have proved their excellent radiation hardness for space applications as well. In . contrast it is well known that the performance of À . conventional silicon or GaAs based space solar cells À severely deteriorate when irradiated with high energy electrons and protons. Development of flexible and lightweight CdTe solar cells together with highly stable performance and low production cost makes them : interesting for low cost and easily deployable solar À .. power generators in space. As is well known the advantage of power generation in space is that there is no atmosphere to reduce the intensity of light from our sun. Hence in space a spectrum is referred to as AMO (Air Mass O). By contrast for most terrestrial applications the accepted standard is AMl.5 conditions.

-

Another problem with known photo-voltaic devices is their stability, that is their efficiency over time reduces quite quickly. This means that photo-voltaic devices must be changed relatively frequently in detectors or power generators, or it would not deliver the rated power.

A further problem with known photo-voltaic devices is that each device is sensitive to only a portion of the spectrum. This means that energy in the form of electro- magnetic radiation having a wavelength which is not captured and converted in the photo-voltaic device is wasted.

It is an aim of embodiments of the present invention to at least partly mitigate the above-mentioned problems.

According to a first aspect of the present invention there is provided a photo-voltaic device for generating electricity comprising a p-type semiconductor layer; an A. e-type semi-conductor layer forming a pen junction at an À interface with said p-type semi-conductor layer; a À transparent front contact for providing electrical contact to said e-type layer; and a back contact for À À . providing electrical contact to said p-type layer; wherein said back contact comprises a layer of À.

transparent conducting oxide (TCO) on a back surface of À .. said p-type semi-conductor layer.

According to a second aspect of the present invention there is provided a method for manufacturing a photo voltaic device comprising the steps of providing a first layer of transparent conducting oxide (TCO) on a glass or polymer substrate; providing an e-type semi-conductor layer on said first TCO layer; providing a p-type semi- conductor layer on said e-type semi-conductor layer; modifying a back surface of said p-type semi-conductor layer; and providing a second TCO layer on the modified back surface of said p-type semi-conductor.

Embodiments of the present invention provide a photo- voltaic device which has an enhanced stability over known II-VI compound semiconductor based photo-voltaic devices.

This is achieved because a back contact to the device is formed from transparent conducting oxide rather than having a copper based back contact which is prone to poor stability.

Embodiments of the present invention provide a bi-facial photo-voltaic device. This means that optical radiation incident on both the back and front side of the device can be converted into electricity. This can improve the overall efficiency of the device.

*., Embodiments of the present invention provide a photo voltaic device which can be stacked in a tandem formation with other photo-voltaic devices. In this way each device in the tandem stack can be tailored to gather radiation at different portions of the spectrum. In this way overall efficiency can be greatly improved. À: '

À .. Embodiments of the present invention will now be described hereinafter by way of example only with reference to the accompanying drawings in which: Figure l illustrates a conventional photo-voltaic device; - ) Figure 2 illustrates a superstrata configuration of a photovoltaic device; Figure 3 illustrates a substrate configuration of a photovoltaic device; Figure 4 illustrates a bi-facial photo-voltaic device; Figure 5 illustrates a photo-voltaic device; Figure 6 illustrates the current-voltage (I_V) curves of the photovoltaic device when light radiation is incident on the front or back side; and Figure 7 illustrates the improved stability of a photo- voltaic device according to an embodiment of the present invention.

In the drawings like-referenced numerals refer to like parts.

Figure l illustrates a conventional polycrystalline cadmium telluride (CdTe)/cadmium sulphide (CdS) thin film À solar cell lo. Such solar cells provide one of the most important photo-voltaic devices so far used for À À.

terrestrial (or ground based) applications. The CdTe/CdS solar cell is based around the heterojunction formed between e-type CdS and p-type CdTe. As illustrated in . Figure l the solar cell lo includes a glass layer ll which forms a substrate. This may be formed from ordinary window glass due to its excellent transparency, strength and cost effectiveness. The glass substrate acts as protection for the active layers set out below.

The glass substrate also provides mechanical strength and lo may include an anti-reflective coating on its front surface 12. Polymer film could also be used instead of glass as a substrate. Below the back surface of the substrate a layer of transparent conducting oxide (TCO) 13 is formed. This may be a doped tin oxide (SnOx:F) layer or an indium- tin oxide (ITO) layer or a doped zinc oxide (ZnO:Al,ZnO:Ga,ZnO:In,ZnO:Br, ZnO:B) layer or a cadmium stannate (CTO) layer or a zinc stannate (ZTO) layer or any other suitable TOO. The ITO layer 13 acts as a front contact to the photo-voltaic device.

Throughout this description a front and back contact will be referred to as will a front and back surface. It will be understood that the terms are used only to differentiate portions of the devices. A front surface or front contact will refer generally to a surface or contact on the substrate side of the device towards a primary source of illumination 14. The transparent conducting oxide is used to help reduce the series resistance of the device between contacts 15 and 16 which would otherwise arise due to the thinness of the active layers referred to hereafter. À. e. * e À

One of the active layers of the photovoltaic device is a polycrystalline CdS layer 17 which is e-type doped and thus provides one half of a pen junction which is formed where a back surface of the CdS layer 17 contacts a front surface of a p-type semi-conductor layer 18. The e-type Àe - : semi-conductor layer 17 provides a wide band gap material À .. having a band gap of around 2.4 eV at 300k. As such it is transparent to optical radiation down to wavelengths of around 514 nanometers. The CdS layer is often referred to as a window layer for this reason. At wavelengths below this some optical radiation will pass through to the p-type layer due to the thinness of the e-type layer. lo

It will be understood that many different types of e-type doped semiconductor layer may be used to replace the CdS layer referred to above. For example many compounds of a group IIb element (for example Zn) and a group VIb element (for example Se or Te or O) may be used.

The cadmium telluride layer 18 is also polycrystalline but is p-type doped so as to provide a p-type semi- conductor layer. This layer has a smaller band gap which is well suited to optical radiation in the solar spectrum. The CdTe layer is less highly doped than the CdS layer 17 and because of this the depletion region will mostly be within the CdTe layer as is known in the art. The active region of the photo-voltaic device will therefore be within the CdTe layer where most of the carrier generation and collection occurs. A back contact l9 is provided to give a low resistance electrical connection to the CdTe. This layer is usually formed from gold, Mo, Ni, or copper but other metal types may be used.

A It is well known that p-type CdTe is a very difficult À material on which to produce an ohmic contact and as such Alga devices will inevitably result in Schottky diode . 25 (rectifying) characteristics. This is a problem which is well understood by those skilled in the art. Àe

: The active layers of the device lo are formed on top of À .. the glass or polymer substrate ll and therefore this construction is generally referred to as a superstrata configuration.

This superstrata configuration is the most commonly used configuration for high efficiency solar cells. Figure 2 also illustrates the superstrata configuration whilst Figure 3 illustrates another configuration known as a substrate configuration. As shown in Figure 2 a buffer layer may be provided between the back surface of the CdTe layer and between a front surface of the metal layer 19. The buffer layer provides a p-type conducting region of high carrier concentration which facilitates an electrical modification of the p-type absorber layer favourable for tunnelling transport when a metal is applied to form ohmic contact.

As shown in Figure 3 in a less efficient "substrate" type solar cell 30 the CdTe solar cell consists of at least four layers and a similar number of relevant interfaces formed from metal foils or metal coated glass or polymers. In the substrate configuration incident light 14 falls on the front surface of a TOO layer 13. This overlies a front surface of a CdS layer which forms a pen junction at the interface with a CdTe layer. A back contact buffer layer (Te, Sb, Cu. Ni, Sb2Te3, etc.) layer 31 is connected to the back surface of the CdTe and to a . front surface of a substrate 32 formed from metal foil or *a coated glass or polyimide to form ohmic contact with Be. CdTe.

. The TCO layer in direct contact with CdS is called the "front contact", while the metal or metal chalcogenide Be e : layers in direct contact with the CdTe are called the À2'.. "back contact". A variety of TCOs, such as doped tin oxide (eg SnOx:F), indium-tin oxide (ITO), doped zinc oxide (eg ZnO:Al), cadmium stannate (CTO), zinc stannate (ZTO), eta have been used as a front contact in CdTe solar cells. I:

The favourable thermo-physical properties and chemical robustness of CdTe/CdS facilitate simple and low cost manufacturing of solar cells with a variety of deposition methods, such as close space sublimation, vapour transport deposition, high vacuum evaporation, sputtering, electrodeposition, spray, etc. The solar cell efficiency depends on the growth process (temperature) and substrates. High temperature (600 C) grown CdTe layers on expensive "alkali-free" glass yield l0 cells of 15 to 16.5% efficiency. While lower efficiencies of 10 to 14% are obtained from CdTe layers grown on soda lime glass with low temperature (<450 C) processes such as high vacuum evaporation (HVE) and sputtering.

The "back contact" on CdTe is an important topic of scientific and industrial interest as it can influence the efficiency and long term stability of solar cells.

The development of efficient and long term stable electrical contact on ptype CdTe is difficult because of both the high electron affinity and energy bandgap of CdTe. Since metals of high work function (5.7) are Arare, alternative methods are applied to develop quasi À ohmic or low resistance contact on p-type CdTe. Most À..

commonly used back contact materials are either metals À 25 (Cu/Au, Ni-P, Cu-graphite paste, Sb, etc) or metal chalcogenides (CuxTe, ZnTe, Sb2Te3, HgTe, eta) that are applied after chemically etching the CdTe surface with Àe.

: either a bromine-methanol or a nitric-phosphoric acid À ace. etch, or by etching with ion beam. On the other hand, when the CdTe layer in CdTe/CdS superstrata solar cells gets in contact with highly conducting TCO, because of pin-holes or incomplete coverage by CdS, shunts are caused which reduce the Voc and FF, thereby reducing the cell efficiency. Therefore, in high efficiency \: superstrata solar cells either a stack of high resistive TCO/low resistive TCO has been used as a front contact or a thick CdS ensuring complete conformal coverage of the conducting TCO is applied for maintaining high Voc and FF. Up until now it has been generally accepted that TCOs form a low resistance ohmic contact with n-CdS and n-CdTe, while they form a rectifying barrier with p-CdTe.

By way of further explanation theory says that for ohmic contact on ptype semi-conductors a metal work function should be higher than the sum of the bandgap and electron affinity of the semi-conductor otherwise a Schottky contact is formed. Therefore for p-type CdTe layers a metal with a work function higher than 5.7 eV is needed since CdTe has a bandgap of 1.45 eV and an electron affinity of 4.3 eV. Most metals therefore form a Schottky contact on p-type CdTe as metals with a work function more than 5 eV are rare. In addition such metals often diffuse from the back contact to the front contact leading to lower efficiency as time increases and therefore to unstable solar cells. In the past an : approach for overcoming this problem is to provide À . surface modification of a back surface of the CdTe layer À.

by etching (for example with bromine-methanol or other À 25 methods) together with the application of a buffer layer.

A. These p-type buffer layers are formed from semi-conductor material (ZnTe:Cu,Sb2Te3 eta) or copper-graphite, copper À.

À. gold. In such a case the copper makes CuxTe by reacting À:. with the Te of the CdTe layer.

However, there are disadvantages in including this buffer layer since most of the metals diffuse into the absorber, preferably via the grain boundaries, thus causing shunts and degrading the cell efficiency with time. Some back contact materials like gold or Sb2Te3 are expensive and therefore less cost effective in production. Moreover the control of the buffer layer thickness is critical in most cases add stringent requirements to processing.

Figure 4 illustrates a photo-voltaic device in accordance with embodiments of the present invention in which the back contact of the device is formed from TCO. As illustrated in Figure 4 primary optical radiation 40 falls on a front surface of a substrate 41 which may be formed from glass or some other polyimide. A TCO layer 42 provides a front contact to the device. The transparent conducting oxide may for example be tin oxide or indium-tin oxide or any other type of transparent conducting oxide as noted hereinabove. An e-type doped semi-conductor layer such as a polycrystalline CdS layer 43 forms a pen junction with a p-type semi-conductor layer such as a polycrystalline CdTe layer 44. The back surface of the p-type semi- conductor layer is modified at the back surface 45 and a further layer of TCO 46 is formed as a back contact to the photo-voltaic device.

I. . Secondary illumination 47 which may be incident at a back . surface of the photo-voltaic device may thus also be À.* converted into electricity. À Àe

I... 25 À The solar cell presented in Figure 4 can be manufactured with already known deposition methods and recipes. CdTe À. . À. surface modification - with chemical and plasma - and I:: subsequent TCO back contact deposition can be selected for specific optimisations. Solar cells have been developed on commercially available 5 x 5 cm2 soda- lime glass substrates coated with an SnOx:F (FTO) layer of about 15 Ohm/sq sheet resistance and about 80% average transmission. A vacuum evaporation method was used to /: grow 500 nm thick CdS and 3000 nm thick CdTe layers at 150 C and 300 C, respectively. For the "CdCl2 treatment", about 200 nm thick layer of CdCl2 was evaporated on the surface of the CdTe and the samples were subsequently annealed in air at 400 C for 30 min. After that the samples were washed in deionised water and the surface of the CdTe layer was etched in a bromine methanol solution.

Samples after etching were immediately transferred into the sputtering system.

An ITO back contact layer is deposited with a rf- magnetron sputtering system (MRC 6031) using a single ITO target. The base pressure in the vacuum system before deposition was 10-6 Torr. The ITO deposition was performed at the Ar/O2 (3 vol % 02) pressure of 6*10-3 Torr and with rf power density of 1500 mW/cm2. The sputtering time was 15 minutes. Under these conditions ITO films with sheet resistance around 12Q/square and transmission around 90% are obtained on glass substrates.

For optimum properties of ITO samples were heated to 250 C prior to the deposition of the ITO layers. After À .. . the deposition small area (3 mm x 4 mm) solar cells were . mechanically scribed. À e.e À e.

The current-voltage (I-V) characteristic of the solar À. cells have been measured at room temperature under simulated AM1.5 illumination and it has been observed À.

that most of the solar cells are in the efficiency range of 7-8%. The photo-voltaic parameters of a typical solar cell, as shown in figure 6 are: VOc=702mV,Jsc=l8.2mA/cm2,FF=o.62=7.9%. When illuminated only from the ITO back contact side the typical cell efficiency is about 1 because of significantly low collection of the charge carriers that are photo generated far away from the CdTe/CdS junction, however this efficiency can be increased by reducing the CdTe thickness and enhancing the diffusion length of minority charge carriers.

The average efficiency of small area solar cells, developed with a similar low temperature process is 10% with Cu/Au back contact on CdTe. Slightly lower efficiency of cells with ITO back contact is because of lower VOC and FF values, which are due to high series resistance, especially due to the resistance of the TCO layers. It should be noted that further optimization of the TCO properties and thickness of etched surface layer are expected to improve the cell efficiency.

As will be appreciated the solar cell as shown in Figure 4 utilises a TCO film as a back contact on the CdTe layer. The cell may be illuminated from either side or from both sides. If illuminated only from the substrate side the cell operates in a manner similar to a conventional superstrata cell. However if illuminated :.., from both sides the photo-voltaic device operates like a À . bi-facial cell. Allowing simultaneous illumination from Àe the primary and secondary side increases efficiency. It À 25 will be appreciated that the cell may be illuminated from À only a back surface. À - .

À. A key difference between this solar cell and prior art À:. solar cells is that there is no requirement for a buffer layer between a modified surface of the p-type CdTe layer and any metal back contact. This is achieved because the step of modifying the back surface of the p-type semi conductor layer provides a tellurium or tellurium-rich CdTe surface with properties sufficient to provide l.

tunnelling. In this way because the e-type TCO is like a semi metal because of its high carrier concentration and electron affinity value, the e-type TCO is sufficient to replace the metal.

It will be appreciated that the photo-voltaic device as illustrated in Figure 4 may be used in a stacked or tandem arrangement.

The tandem arrangement overcomes the limitation for single-junction solar cells which use only a part of the spectrum by stacking two or more photovoltaic junctions with different spectral absorption properties on each other. This is done in a way that the top cell has the highest band gap which allows least absorption of the spectrum. The following cells have more narrow bandgaps utilising the remaining transmitted spectrum. The tandem arrangement can be achieved by mechanical stacking of the single solar cells or by a monolithic growth of heterojunctions.

: In the above described arrangement the CdTe (or CdZnTe) a *., cell will serve as the wide gap top cell and the À.e transparent back contact allows for a high transmission of the non absorbed part of the spectrum which can be À. converted by a second narrow band gap solar cell.

À. Figure 5 illustrates a photo-voltaic device according to À. a further embodiment of the present invention. In this primary illumination 40 is incident upon a glass or polyimide substrate 41 at a front of the photo-voltalc device. A TCO layer 42 is formed at a back surface of the substrate and a pen junction formed at the interface of further layer 43 and 44 formed from CdS and CdTe respectively. The back surface of the CdTe layer is modified in a manner noted above and then a layer 46 of TCO is formed. A reflective or partially reflective layer 50 is formed at the back of the device 51. This reflects light transmitted through the device 51 back into the device so that electricity may be generated from this reflected light. In this way light falling directly into the device or indirectly via reflection from reflecting or partially reflecting surface 50 may be used to generate electricity which improves the overall efficiency of the device and can help in reducing the thickness of the CdTe absorber layer.

According to embodiments of the present invention a configuration for CdTe solar cells is provided where a TCO layer is used as a back contact on CdTe. This solar cell opens a variety of possibilities for applications as it can be illuminated both from the front and backside simultaneously, thus operating like a "bi-facial" cell.

Additionally, these solar cells may optionally be used in tandem solar cells since both the contacts are :.^, transparent and conducting. To improve efficiency a .a.. tandem solar cell may be provided with a thinner and Àe higher band gap II-VI absorber layer for use in top cell.

À.-... 25 À Single junction CdTe/CdS solar cells are interesting for both terrestrial and space applications. Such solar cells À. with TCO contacts are more stable than conventional cells :-'.. with metallic back contacts, because of their chemical robustness and chemical compatibility. Using a low temperature process, which is suitable for the development of flexible cells on polyimide, 7.9% efficient cells on glass substrates can be developed using FTO as a front contact for CdS and ITO as a back contact on CdTe.

The structure provides a superstrate-like structure in which the conventional back contact has been replaced by a TCO layer and the surface of the CdTe layer has been modified by a chemical etching and exposure to a plasma containing Ar and O ions. Some of the most important advantages of this device are: 1. Unique application of TCO both as a front contact for CdS and back contact for CdTe by applying a chemical (etching) and/or a plasma modification of CdTe for engineering the interface properties.

2. Transparent or semitransparent device, depending on the spectral range and CdTe thickness, that can be used for tandem solar cells, especially for being monolithically connected.

3. Superior encapsulation and more stable long term performance because some of the TCOs are chemically more inert and stable (less diffusive) compared to a' most of the metals that have been used in "'. conventional superstrata cells up to now. r ate

4. Highly suitable for industrial manufacturing as TCO Àe thickness control is not stringent, while a strict thickness control of a few nano meter thin buffer (Cu. HgTe, Sb, Sb2Te3, eta) is very important in conventional superstrata cells.

5. It can be used as a "bi-facial" solar cell where the device can be illuminated from either or both sides.

According to embodiments of the present invention polycrystalline thin film CdTe/CdS solar cells have been developed in a configuration where a transparent conducting layer of ITO layer has been used for the first time as a back contact on CdTe. Solar cells of 7.9% efficiency may be manufactured developed on SnOx:F coated glass substrates with a low temperature (<450 C) high vacuum evaporation method. After the CdCl2 annealing treatment of CdTe/CdS, a bromine methanol solution was used for etching the CdTe surface prior to the ITO deposition. The unique features of this solar cell with both front and back contacts being transparent and conducting are that the cell can be illuminated from either or both sides simultaneously like a "bi-facialn cell, and it can be used in tandem solar cells. Solar cells with this new process are more stable than the cells with Cu-based back contacts.

Figure 7 illustrates the stability of photo-voltaic devices in accordance with embodiments of the present invention over a period of time under accelerated test condition (cell in open circuit condition is continuously illuminated with AMl.5 radiation and at 80 C). The lower curve 60 illustrates how the efficiency of À. À. À .

À À conventional solar cells deteriorate from a unitary ratio À0..

. of current efficiency divided by starting efficiency to around 0.7 after a time corresponding to half a year or so. By contrast solar cells manufactured in accordance À.

À with the present invention show an increase in efficiency À rather than a slight drop in efficiency. Also efficiency À.

over a long period is maintained and does not go lower À than the initial value. Therefore, the cells with new process are considered stable.

Embodiments of the present invention have been described in detail hereinabove. However it would be understood that the present invention is not limited to the specifics of the examples described but rather modifications may be made without departing from the scope of the present invention. À. À. À À I. À À.. À .e À À Àe À:. À. O Àe a. e À

Claims

CLAIMS: 1. A photo-voltaic device for generating electricity comprising: a

p-type semi-conductor layer; an e-type semi-conductor layer forming a pen junction at an interface with said p-type semi-conductor layer; a transparent front contact for providing electrical contact to said e-type layer; and a back contact for providing electrical contact to said p-type layer; wherein said back contact comprises a layer of transparent conducting oxide (TCO) on a back surface of said p-type semi-conductor layer.
2. The photo-voltaic device as claimed in claim l wherein said back contact is transparent to optical radiation.
3. The device as claimed in claim l or 2 further comprising:

-

À. a layer region of modified p-type semi-conductor I ', material at a back surface of said p-type layer. À À. À:.
4. The device as claimed in any one of claims l to 3 further comprising: a front layer of transparent conducting oxide (TCO) at a front surface of said e-type layer.
5. The device as claimed in claim 4 further comprising: a transparent substrate at a front surface of said front layer of TCO.

-
6. The device as claimed in claim 5 wherein said substrate comprises a glass or polymer material.
7. The device as claimed in any one of claims l to 6 wherein said p-type semi-conductor layer comprises a CdTe thin film layer.
8. The device as claimed in any one of claims l to 7 wherein said e-type semi-conductor layer comprises a CdS thin film layer.
9. The device as claimed in any one of claims l to 6 wherein said p-type semi-conductor layer comprises a ternary or quaternary II-VI compound with a band gap energy in the range 0.9 to 1.8 eV.
lO. The device as claimed in any one of claims l to 6 wherein said e-type semi-conductor layer comprises a ternary or quaternary compound with a band gap energy in the range of 2 to 3.4 eV. À
ll. The device as claimed in any one of claims l to lo wherein said photovoltaic device comprises a thin . film solar cell and a light source for said optical radiation comprises a sun. À:
12. The device as claimed in any one of claims l to lO . : wherein said optical radiation comprises direct or indirect radiation.
13. The device as claimed in any one of claims l to lo wherein said photovoltaic device comprises a bi facial cell for generating electricity when illuminated from both the front and back side simultaneously.
14. The device as claimed in claim 1 further comprising: a metal layer on a back surface of said layer of TCO.
15. The device as claimed in claim 1 further comprising: a conductive paste on a back surface of said layer of TCO.
16. A device as claimed in claim 1 further comprising: a metal grid on said TCO layer.
17. A method for manufacturing a photo-voltaic device comprising the steps of: providing a first layer of transparent conducting oxide (TCO) on a glass or polymer substrate; providing an e-type semi-conductor layer on said first TCO layer; providing a p-type semi-conductor layer on said n type semi-conductor layer; modifying a back surface of said p-type semi À . . conductor layer; and À À'*. providing a second TCO layer on the modified back surface of said p- type semi-conductor. À Àe Àe
18. The method as claimed in claim 17 further comprising : ' the steps of: À a.

: annealing said device prior to said modifying step. À -
19. The method as claimed in claim 18 further comprising the steps of: annealing said device via heat treatment optionally (i) in vacuum, (ii) in air, (iii) in a gas, (iv) with CdCl2 vapour, (v) with Cl vapour or (vi) with HCl vapour.
20. The method as claimed in any one of claims 17 to 19 further comprising the steps of: subsequent to said step of providing a second TCO layer, providing a metal layer on a back surface of said TCO layer.
21. The method as claimed in any one of claims 17 to 19 further comprising the steps of: subsequent to said step of providing a second TCO layer, providing optionally (i) a metal grid and/or (ii) a conductive paste on a back surface of said TCO layer.
22. The method as claimed in claim 17 wherein said modifying step comprises the step of: etching the back surface of said p-type semi conductor layer.
23. The method as claimed in claim 22 wherein said etching step comprises washing the back surface of said p-type semi-conductor layer with a bromine methanol *. solution. À A.-

À.
24. The method as claimed in claim 22 wherein said . etching step comprises: washing the back surface of said p-type semi conductor layer with nitric-phosphoric acid. À: À..

:
25. The method as claimed in claim 17 wherein said À - modifying step comprises the step of: cleaning the back surface of said p- type semi conductor layer.
26. The method as claimed in claim 25 wherein said cleaning step comprises heating said device in a vacuum at high temperature or by ion etching.
27. The method as claimed in claim 22 or claim 23 further comprising the steps of: prior to said etching step, washing the back surface of said ptype semi-conductor layer with de-ionised water.
28. The method as claimed in any one of claims 17 to 27 wherein each of the steps of providing an e-type semi-conductor layer and a p-type semiconductor layer comprise the step of growing a respective layer of semi conductor material via vacuum evaporation.
29. The method as claimed in any one of claims 17 to 27 wherein each of the steps of providing an e-type semi-conductor layer and a p-type semiconductor layer comprise the step of growing a respective layer of semi conductor material with a thin or thick film deposition À À À or a growth method. À \
30. The method as claimed in claim 28 À À.

wherein said step of providing an e-type semi conductor layer comprises growing a 500 nm thick layer of CdS via vacuum evaporation and said step of providing a p-type semi-conductor layer comprises growing a 3000 nm thick layer of CdTe via vacuum evaporation.
31. The method as claimed in any one of claims 17 to 30 further comprising the steps of: 23: evaporating a 200 nm thick layer of CdCl2 on a back surface of said p-type semi-conductor layer prior to an t etching step of said modification step.
32. The method as claimed in claim 31 further comprising the steps of: subsequent to said step of evaporation, performing an annealing step at 400 C in air for thirty minutes.
33. The method as claimed in any one of claims 17 to 32 wherein said step of providing said second TCO layer comprises: depositing TCO via an rfmagnetron sputtering system.
34. The method as claimed in any one of claims ll to l9 wherein said first and second TCO layers optionally comprise: (i) a doped tin oxide (SnOx:F) layer; (ii) an indium-tin oxide (ITO) layer; (iii) a doped zinc oxide (ZnO:Al,ZnO:Ga,ZnO:In,ZnO:Br,ZnO:B) layer; (iv) a cadmium : stannate (CTO) layer; or (v) a zinc stannate (ZTO) layer. À..-
35. A method substantially as hereinbefore described with reference to the accompanying drawings. A. 25
36. Apparatus constructed and arranged substantially as hereinbefore described with reference to the accompanying :. .: drawings.