WO2009030281A1

WO2009030281A1 - Method of forming thin film solar cell

Info

Publication number: WO2009030281A1
Application number: PCT/EP2007/059422
Authority: WO
Inventors: Ayodhya Nath Tiwari; Kaia Ernits; Christopher John Hibberd; Marc Roland Kaelin; Maxim Ganchev
Original assignee: ETH Zürich
Priority date: 2007-09-07
Filing date: 2007-09-07
Publication date: 2009-03-12

Abstract

The method for producing a ternary or greater compound thin film comprises the steps: - deposition of a binary or higher group INA chalcogenide thin film on a substrate, - incorporation of group IB atoms into said chalcogenide layer by bringing said chalcogenide layer into contact with a group IB atoms, containing media, and - heat treatment of the resulting mixed layer to convert it into a homogenous phase.

Description

Method of forming thin film solar cell

The present invention refers to the production of a ternary or greater compound thin film with properties suitable for use as a light absorbing layer in a photon to electricity conversion device at least containing elements from group I, III, VI of the periodic table of the elements. More specifically, the present invention refers to the production of chalcopyrite layers for thin film solar cells consisting of a stack of layers supported upon a substrate.

Background of the invention The basic properties, components and structure of thin film solar cells are taught by references [1] and [2]. Many patents and academic articles have been published relating to this category of solar cells and the technology and methods for producing it, testifying to the technological relevance of the device. Of the methods used for producing these solar cells, the most successful to date as judged by the efficiency of energy conversion is that described in [3], which reports an energy conversion efficiency of 19.5% with a Cu(In₅Ga)Se₂ compound semiconductor light absorbing layer. The method used to produce the absorber layer was that of co- evaporation of elemental metals, in a vacuum chamber, onto a substrate held at a high temperature for the duration of the deposition.

In an alternative approach to depositing the same light absorbing film, one or more layers of material are first deposited and then converted into a single layer with the desired morphology, phase and composition or compositional gradient by a conversion process consisting of one or more sequential treatments. This approach is sometimes referred to in the art as the two stage process. The initially deposited layer or layers are referred to as precursor layers and may be layers of elements (both metals and non- metals), binary and higher metallic alloys, binary and higher oxides, binary and higher non-oxide compounds (including chalcogenides and salts) and mixtures of these materials. More precisely, the material may be present in the form of continuous films or as particulates, which may be intercalated with another material in some embodiments of the process. Additionally, the precursor may include material that is not required in the light absorbing layer; this material is preferentially removed during the conversion process. The conversion process may contain a treatment which supplies energy to facilitate phase changes, mixing and recrystallization within the precursor layer or layers. This energy is preferably supplied by heating within a furnace using lamps, resistive elements or inductive elements. The heating is performed within an atmosphere that ensures the desired layer is formed. Such an atmosphere may be at a pressure other than atmospheric pressure and may contain an inert gas, one or more chalcogen hydride gases, one or more chalcogen vapours, a reducing gas and/or an oxidising gas. An example of an industrial process is taught in reference [4]. The conversion process may also contain a treatment within which unwanted materials that are present in or upon the layer are removed by etching. Such an etching stage may utilise either a chemical etchant or an electrochemical process to remove the unwanted material.

In some embodiments of the two stage process, the precursors are deposited using vacuum equipment. For example the methods reported in the articles [5,6,7] use sputtering and thermal evaporation to deposit stacked elemental films, where the resulting layers are annealed, sometimes in the presence of a chalcogen, to form the desired light absorbing layer.

In other two-stage processes, the precursor layers are deposited from solution baths. For example [8] teaches that a layer with composition close to that of CuInSe₂ or one of its alloys with gallium and sulphur may be deposited by electrodeposition and then annealed to form the desired light absorbing layer with reported efficiencies of up to 10.6%. Further the use of electroless deposition is described [9] where an additional vacuum step was required to produce cell efficiencies of up to 12.4%. An alternative embodiment of the two-stage method uses spray techniques to deposit layers from solutions containing all of the required elements onto a heated substrate whereupon the constituents of the solution are pyrolysed to form the desired light absorbing layer whilst the unwanted materials are removed as volatile reaction products. Solar cells based upon CuInSe₂ layers produced by this method were reported to yield efficiencies of up to 5% [10]. In a two stage method [1 1], first layers of copper indium oxide were deposited by spray deposition and then converted into the desired light absorbing layer by annealing with selenium vapour. Another alternative embodiment of the two-stage method uses metal salts as precursor materials [12] where inorganic salts of copper, indium and gallium are mixed in a solution with a binder and then doctor blade coated onto a substrate which was subsequently annealed with selenium vapour. A further group of embodiments of the two stage method are concerned with the deposition of particulate precursors. In [13] a method of depositing mixed metal particles containing one or more metal oxides by different deposition techniques is described. Similarly, [14] teaches a method of depositing metal oxide particles. The presence of oxides in the precursors requires the involvement of aggressive and toxic gasses to obtain the desired light absorbing layer. It is known by those skilled in the art that within chalcopyrite thin films some of the constituent species are mobile and may diffuse over time, particularly at elevated temperatures. In particular, copper is known to be an extremely mobile element within such films and some methods have arisen to take advantage of this fact. For example, in the method reported in the article [15], indium selenide layers are deposited and a copper flux is provided onto the layer held at 200°C and the selenium flux is provided onto the resulting layer held at 600°C. The resulting solar cell is reported to have an efficiency of 13.7%. Reference [16] showed that the resulting films may still contain binary chalcogenide phases, mainly copper selenide. These secondary phases can be removed by chemical etching and solar cells of reported efficiencies of 15-16% were prepared with this approach. In the process described in [1] a group III chalcogenide layer is exposed to a flux of selenium and copper, which diffuses into the layer. After obtaining a slightly copper rich composition, the surface is again exposed to a flux of group IHA vapors and selenium to eliminate copper rich phases and record efficiencies up to 19.5% were obtained. These examples show the potential of CuInSe₂ material formed by the diffusion of copper into group III - selenide layers, however the methods rely on expensive vacuum equipment in common with other technologies.

Finally a few methods were reported where solutions provide ions of a specific type to be incorporated in or exchanged with the near surface part of an immersed layer. [17] uses such a method for exchanging near surface Ga atoms of a chalcopyrite compound with Al atoms. [18] teaches that a layer of p-type Cu(In, Ga)Se₂ may be immersed in a solution containing metal ions of group II with the result that the metal ions are incorporated into the surface region of the film, forming an n-type semiconducting layer that forms a p-n junction with the Cu(In, Ga)Se₂. P-n junctions formed in this manner, also called partial electrolyte treatment, were used in the manufacture of 18% efficient thin film solar cells.

Similarly, in H-VI compounds copper solutions were used for converting surface near regions; [19] discloses a simple way of forming a p-n junction on top of a CdS layer by immersion into a solution containing a cuprous compound. The quality of the resulting Cu_xSy-CdS junction could be significantly improved. [20] reports that a copper layer forms on CdTe films immersed in an aqueous solution of copper ions. Upon annealing, this layer is shown to diffuse into the film, forming a two phase structure containing CdTe and CuTe.

It is clear from this description of the state of the art that methods for the production of thin film solar cells that avoid the use of expensive plant have been developed, however the resulting cells are still significantly inferior to the methods that use a conventional vacuum technology plant. As such, there is a need to develop new methods of production for said solar cells using methods devised to reduce the cost of the completed device. It is also clear that methods exist for exchanging ions between a solution and a thin film for the formation of p-n junctions and light absorbing layer/conductive layer structures. The present method provides a means of fulfilling the aforementioned need using a new and novel method of thin film/solution ion exchange.

Summary of the Invention

One object of the present invention is to provide a low-cost large area scalable method for the production of a thin film suitable for use as a light absorbing layer in a thin film solar cell. Furthermore, an object is to avoid some of the aforementioned problems, namely, the use of complex equipment and the maintenance of high processing temperatures for extended durations, both of which tend to increase the cost of said solar cells. Accordingly, the present invention proposes a process according the wording of claim one.

Accordingly, the present invention proposes a method for producing a ternary or greater thin film, said method comprising the steps:

- deposition of a binary or higher group IHA chalcogenide thin film onto a substrate, incorporation of group IB atoms into said chalcogenide layer by bringing said chalcogenide layer into contact with group IB ions containing medium, and

- homogenising at least partially of the resulting mixed layer by a heat treatment in controlled atmosphere.

The resulting structure may further be completed into a thin film solar cell by any method known to those skilled in the art, including an optional surface etching step, a partial electrolyte treatment, buffer layer deposition, transparent conducting electrode deposition, anti reflection layer, metal grid structure deposition.

The term substrate may refer to any material known to those skilled in the art as being suitable for the role such as, but not limited to, glass, metal or polyimide and additionally may include one or more thin films, such as e.g., but not limited to, a metallic electrode, a transparent conducting oxide electrode, a highly reflective layer, a diffusion barrier, a layer of dopant material or a layer designed to alter the interfacial properties between two such layers. Additionally, the substrate may provide material that is incorporated into the light absorbing layer, including but not limited to the case of an alkali dopant diffusing from soda lime glass.

For the deposition of the group III chalcogenide compound any known method to produce thin films of chalcogenide compounds may be used including but not restricted to evaporation, spray pyrolysis, chemical vapour deposition, chemical bath deposition, electrodeposition, electroless deposition, sol-gel methods, atomic layer deposition, paste coating and chalcogenization.

In the method of the invention, copper or other group IB metals such as e.g. silver or gold are incorporated into a binary chalcogenide thin film or binary chalcogenide alloy thin film from its surroundings, which may be a liquid such as a solution or a gas. In the case of a solution it may be maintained at a temperature less than or equal to its boiling point at the pressure of the container and can use different type of agitation such as ultrasonic or magnetic stirring. Said pressure may be other than atmospheric pressure. The incorporation of copper ions into the chalcogenide layer may be thought of as an exchange of cations between the chalcogenide layer and the surrounding media followed by diffusion of the ions or as an addition into the amorphous or microcrystalline structure of the chalcogenide layer. The media from which the group IB atoms are incorporated into the chalcogenide layer may contain ingredients other than the source of group IB atoms, including but not limited to reagents to regulate the pH, chelating agents and ions of metals other than copper. Additionally, the concentration of group III metal ions in the media may increase throughout the process as they may leave the chalcogenide thin film to make vacancies by which the copper ions are incorporated into the thin film.

The stoichiometry of the light absorbing layer may be set during the incorporation of copper ions into the film by adjusting the concentration of ions in the media, the temperature at which the process is performed and the duration of the immersion of the chalcogenide layer into the process media.

After the immersion in the group IB ion containing media an optional surface cleaning treatment with an adequate solvent agent can be performed. As an example potassium chloride solution can be used to remove copper or copper salt rests.

The heat treatment in inert or chalcogen vapour or gas containg atmosphere allows the group IB atoms to diffuse into the layer and recrystallizes the material mainly into the chalcopyrite phase. The used heating parameters and atmosphere influences the compositional gradient as known to those skilled in the art.

Typically such treatments lead to a gallium accumulation towards the back contact of the layer structure. For optimal device performance it is advantageous to increase the bandgap in the near surface part of the layer, e.g. by replacing some of the near surface indium atoms with gallium or aluminium or a part of the selenium atoms with sulphur. Therefore an additional immersion step in adequate solutions for ion exchange in the near surface part may be applied.

In the case of non-stoichiometric compositions impurity binary chalcogenide phases may still remain in the layer. Copper rich impurity phases are known to degrade the solar cell parameters. Those skilled in the art know that this phase may effectively be removed via etching, e.g. by using a solution of potassium cyanide, which additionally improves the surface properties of the layer. Alternatively group IB rich phases may be neutralized by addition of group III chalcogenides as for example by chemical bath deposition, spray pyrolysis or evaporation. For the better understanding the present invention shall be described in more details with reference to examples of embodiments according to the present invention and with reference to the attached figures, in which

Figure 1 through Figure 5 show cross sections of a solar cell throughout stages of the method of the invention, Figure 6 shows an x-ray photoelectron spectroscopy (XPS) depth profile of a binary chalcogenide layer after it has undergone the treatment to incorporate copper into the layer,

Figure 7 shows an x-ray photoelectron spectroscopy (XPS) depth profile of a layer prepared in an identical manner after it has undergone heat treatment in the presence of selenium vapour, and

Figure 8 shows the resulting Cu concentration in function of diffusion time measured with energy dispersive X-ray analysis after Ion in-diffusion and subsequent selenization, where the used solution was prepared with ethyleneglycol, wherein CuCl and KCl were dissolved at 17O⁰C, Figure 9 shows x-ray diffraction (XRD) patterns of a layer throughout various stages of the method of the present invention.

In a first embodiment of the method of the present invention according to Figure 1 a clean soda lime glass plate (layer 1) is utilised as substrate and coated with a thin layer of molybdenum (layer 2), preferably 0.5-1 μm in thickness. This layer is most often deposited via DC magnetron sputtering. Onto this substrate, a thin film of a binary chalcogen compound (Figure 2, layer 3) such as In₂Se₃ or an alloy of two or more such compounds, such as (In₅Ga)₂Se₃, is deposited by physical vapour deposition. This layer is preferably between 0.7 μm and 2 μm in thickness. This layer and the substrate upon which it was deposited are then immersed in an aqueous solution containing a copper salt such as Cu(NO₃)₂ and a reagent to reduce the pH below 7, such as acetic acid. The solution is heated to 95°C for the desired duration, preferably between 20 and 60 minutes. After this time the layers and substrate are removed and rinsed in clean water. At this stage the chalcogenide layer (Figure 3, layer 4) contains sufficient copper atoms to produce the desired chalcogenide layer, however the copper atoms are distributed unevenly throughout the depth of the layer, with a peak in concentration towards the surface as shown by x-ray photoelectron spectroscopy (XPS) measurements (Figure 6). The chalcogenide layer containing the copper atoms is then transferred to a furnace where it is heated in the presence of a source of selenium atoms, from a heated crucible of elemental selenium or with a stream of hydrogen-selenide/sulfide. During the desired pattern of heating and cooling, the chalcogenide layer is converted to a chalcopyrite layer (Figure 4, layer 5),such as CuInSe₂,or Cu(In,Ga)Se2 with homogeneous copper concentration throughout its depth as confirmed by x-ray photoelectron spectroscopy (XPS) measurements (Figure 7).

Binary copper selenide impurity phases may still be present in the heat treated layer, especially for the case of copper concentrations exceeding stochiometry. These copper selenides are removed by immersing the layer structure in a solution of 10% potassium cyanide at room temperature for a few seconds. After removal from the cyanide solution the layer structure is washed in clean water.

Finally the structure is fabricated into a completed solar cell, as shown in Figure 5, by the application of a thin (20-200nm) cadmium sulphide film (layer 6) upon the chalcopyrite light absorbing layer, a layer of intrinsic zinc oxide and a layer of aluminium doped zinc oxide (100-lOOOnm combined thickness, layer 7) upon the cadmium sulphide layer and the application of a bilayer grid of nickel and aluminium over a small percentage of the surface (structured layer 8). Possible techniques for the deposition of this layers are chemical bath deposition for the cadmium sulphide, radio- frequency magnetron sputtering for the zinc oxide layers and electron beam evaporation for the metal grid bilayer.

For a better understanding of the conversion reaction according to the present invention X-ray diffractograms were recorded at different stages of the processing. Curve 1 of Figure 8 shows the diffraction patterns of an evaporated indium-selenide layer corresponding to layer 3 of Figure 2. The pattern indicates a hexagonal γ-In2Se3 phase. After immersion of the sample in a copper containing solution according to the present invention, curve 2 of Figure 8 was recorded. Comparison with curve 1 shows extra peaks at around 26.7° and 44.5° (2 theta). These peaks are identified as characteristic of copper rich copper selenide phases, such as Cu₂._xSe.

The X-ray diffractogram of a sample after the heat treatment step is shown in curve 3 of Figure 8. All of the peaks in the diffractogram may be indexed to chalcopyrite CuInSe₂. In another embodiment, the aqueous solution of the first embodiment is replaced with a high molecular weight alcoholic solvent allowing a higher process temperature to be employed at a given pressure. The media is a solution containing ethyleneglycol and Cu⁺ or Cu⁺⁺ salts such as CuCl, CuCl₂, Cu(NOa)₂ or Cu(SO₄)₂.The temperature can be raised to 170°C and higher during immersion of the sample with negligible evaporation of the solvent. The increased temperature leads to much faster in-diffusion of the copper ions.

In yet another embodiment of the invention the media is a solution containing ethyleneglycol, 0.6M copperchloride (CuCl), and 2M potassium chloride (KCl). The Cu+ ions are stabilised by complexation with KCl which further improves the solubility of the CuCl. Figure 9 shows the copper concentration (measured by EDX) before and after the heat treatment which evenly distributes the copper throughout the layer depth. Mainly the whole range of useful copper concentrations can be achieved with the disclosed process. A final desired Cu concentration in the range of 20-25% can be achieved with immersion times of a few minutes.

Thickness measurements before and after immersion treatments showed a slight increase in the layer thickness.

Those versed in the art will recognise that alternative materials may be employed for the layers required for the completion of the solar cell, namely, the glass substrate, the molybdenum rear electrode, the cadmium sulphide buffer layer, the zinc oxide front electrode layers and the metal bilayer collection grid may all be replaced by alternative suitable materials. In a similar vein, the techniques used to deposit these layers will depend on the materials employed and the deposition of the materials specified in the first embodiment is not restricted to the described techniques. Moreover, additional layers may be incorporated into the structure, including but not limited to: one or more barrier layers deposed between the initial substrate layer and the electrode layer with the function of electrically isolating the electrode layer from the substrate and/or with the function of limiting diffusion between these two layers; a layer of material intended to diffuse into the chalcogenide/chalcopyrite layer either during its deposition or during the heat treatment.

Moreover, the replacement of some of the materials specified in the possible embodiments may be performed in order to achieve a conversion of the solar cell geometry from that of a substrate solar cell to that of either a superstrate solar cell or to that of a semi-transparent solar cell wherein light that is not absorbed is free to pass through the cell.

References

[I] Mickelsen, R.,A. and Chen, W.,S., U.S. Patent No. 4335266, Jun. 15, 1982.

[2] Choudary, U.,V., Shing, Y.,H., Potter, R.,R., Ermer, J.,H. and Kapur, K., U.S. Patent No. 461 1091, Sep. 9, 1986.

[3] Contreras, M., A., Ramanathan, K., AbuShama, J., Hasoon, F., Young, D., L., Egaas, B. and Noufi, R. Progress in Photo voltaics: Research and Applications 13 3 (2005) 209-216.

[4] Karg, F. and Probst, V., U.S. Patent No. 5578503, Nov. 26, 1996.

[5] Palm, J., Visbeck, S., Stetter, W., Niesen, T., Fuerfanger, M., Vogt, H., Calwer, H., Baumbach, J., Probst, V. and Karg, F. 21^st European Photovoltaic Solar Energy Conference (2006) 1796- 1800.

[6] Alberts, V. Semiconductor Science and Technology 19 1 (2004) 65-69.

[7] Dejene, F., B. and Alberts, V. Journal of Physics D: Applied Physics 38 (2005) 22- 25.

[8] Taunier, S., Sicx-Kurdi, J., Grand, P. P., Chomont, A., Ramdani, O., Parissi, L., Panheleux, P., Naghavi, N., Hubert, C, Ben-Farah, M., Fauvarque, J. P., Connolly, J., Roussel, O., Mogensen, P., Mahe, E., Guillemoles, J. F., Lincot, D. and Kerrec, O. Thin Solid Films 480-481 (2005b) 526-531

[9] Bhattacharya, R., N., Hiltner, J., F., Batchelor, W., Contreras, M., A., Noufi, R., N. and Sites, J., R. Thin Solid Films 361 - 362 (2000) 396. [10] Duchemin, S., Bougnot, J., El Ghzizal, A. and Belghit, K. Proceedings of the 9th EPVSEC (1989) 476-479.

[I I] Beck, M. E. and Cocivera, M. Thin Solid Films 272 (1996) 71.

[12] Kaelin, M., Rudmann, D., Kurdesau, F. V., Zogg, H., Meyer, T. and Tiwari, A. N. Thin Solid Films 480 - 481 (2005) 486 [13] Eberspacher, C. and Pauls, K., U.S. Patent No. 6268014, JuI. 31, 2001.

[14] Kapur, V.,J., Basol, B.,M., Leidholm, C.,R. and Roe, R.,A., U.S. Patent No. 6127202, Oct. 3, 2000.

[15] Kessler, J., Schmid, D., Zweigart, S., Dittrich, H. and Schock, H. W. 12th European Photovoltaic Solar Energy Conference (1994) 648-652. [16] Walter, T., Braunger, D., Dittrich, H., Koble, Ch, Herberholz, R. and Schock, H. W. Solar Energy Materials and Solar Cells 41-42 (1996) 355-372. [17] Karg, F., U.S. Patent No. 5137835, Aug. 1 1, 1992

[18] Hashimoto, Y., Negami, T., Hayashi, S. and Wada, T., U.S. Patent No. 6107562, Aug. 22, 2000.

[19] Keramidas, B.G., Schefer, J.C. US Patent No 3'416'956, May 16, 1966

[20] Ali,A., Shah,N.,A., Aqili,A.,K.,S. and Maqsood,A. Semiconductor Science and Technology 21 (2006) 1296-1301

Claims

1. Method for producing a ternary or greater compound thin film, said method comprising the steps: - deposition of a binary or higher group HIA chalcogenide thin film on a substrate,

- incorporation of group IB atoms into said chalcogenide layer by bringing said chalcogenide layer into contact with a group IB atoms containing media, and - heat treatment of the resulting mixed layer to convert it into a homogenous phase.

2. Method according to claim 1, characterized in that the deposition is carried out on a substrate such as glass, a polymer foil or a metal foil coated with an insulating layer.

3. Method according to one of the claims 1 or 2, characterized in that prior to the deposition, the substrate is coated with an electrically conductive layer.

4. Method according to one of the claims 1 or 2, characterized in that prior to the deposition the substrate is coated with one or more transparent conductive semiconducting layers.

5. Method according to one of the claims 1 to 4, characterized in that the group III chalcogenide thin film contains one or both of the elements indium and/or gallium.

6. Method according to one of the claims 1 to 5, characterized in that incorporation of copper into said chalcogenide film occurs by exchange of group III ions in said film by copper atoms in a liquid and diffusion of said copper atoms throughout said chalcogenide film.

7. Method according to one of the claims 1 to 6, characterized in that the method comprises the steps:

- deposition of an indium-gallium-selenide compound layer or alloy onto the substrate, - placing the chalcogenide layer into a media containing copper ions for their incorporation into the layer, and

- heat treatment of the resulting mixed layer to convert it into a Cu(In₅Ga)Se₂ phase.

8. Method according to one of the claims 1 to 7, characterized in that the incorporation of copper into the chalcogenide film occurs by exchange of group

III ions in said film by copper ions in an aqueous solution containing a copper salt.

9. Method according to one of the claims 6 to 8, characterized in that an additional source of metal ions of group IB other than copper is added to the solution.

10. Method according to one of the claims 6 to 9, characterized in that the solution is a non-aqueous solution, such as e.g. an organic solution, containing a copper salt.

1 1. Method according to one of the claims 6 to 10, characterized in that said solution is heated to a temperature equal to or below its boiling point at ambient pressure during the immersion of the group III chalcopyrite layer.

12. Method according to one of the claims 6 to 1 1, characterized in that the solution is kept at an ambient pressure greater than atmospheric pressure in order to raise its boiling point.

13. Method according to one of the claims 1 to 12, characterized in that the resulting mixed layer is converted into a light absorbing layer by annealing in a furnace at a temperature less than the melting point or decomposition temperature of the substrate.

14. Method according to one of the claims 1 to 13, which further comprises a step to remove impurity phases formed during the heat treatment by chemical etching.

15. Method according to one of claims 1 to 14, which further comprises a step to exchange In atoms in the layer with Ga or Al atoms.

16. Photovoltaic device comprising at least one layer achievable by a method according to one of the claims 1 to 15.

17. Device according to claim 16, comprising at least a substrate (1) being coated with a film of molybdenum (2), a chalcopyrite layer (5), which is coated by a cadmium sulphide film (6), further, comprising a layer of intrinsic zinc oxide and a layer of aluminium doped zinc oxide (7).