Patent Application
20110203655
Copper indium gallium selenide (CIGS) based photovoltaic devices can be made from high temperature vacuum processes. Using glass as a substrate can be beneficial as the glass plate provides mechanical support as well as a smooth surface for high quality film growth. However, a heat treating process to strengthen the glass substrate after CIGS deposition can give rise to some other problems, such like CIGS degradation, equipment and process cost increase. Furthermore, exposure of CIGS to ambient results in CIGS surface oxidation, in turn diminishing device performance.
Photovoltaic devices can include multiple layers formed on a substrate (or superstrate). Copper indium gallium selenide (CIGS) based photovoltaic devices can be made from high temperature vacuum processes, such as co-evaporation, reaction of stacked elemental layers, or selenization of metal precursors. For example, a photovoltaic device can include a transparent conductive oxide (TCO) layer, a buffer layer, a semiconductor layer, and a conductive layer formed adjacent to a substrate. The semiconductor layer can include a semiconductor window layer and a semiconductor absorber layer, which can absorb photons. The semiconductor absorber layer can include CIGS. Each layer in a photovoltaic device can be created (e.g., formed or deposited) by any suitable process and can cover all or a portion of the device and/or all or a portion of the layer or substrate underlying the layer. For example, a “layer” can mean any amount of any material that contacts all or a portion of a surface.
Copper indium (di) selenide (CIS) and their alloys with gallium (CIGS) have the requisite optoelectronic properties (high optical absorption coefficient, favorable bandgaps, suitable resistivity etc.) for photovoltaic (PV) applications. Moreover, polycrystalline CIS and CIGS solar cells can outperform their monocrystalline counterparts. A variety of techniques were used in the preparation of these materials. The best cell efficiencies can be obtained on materials prepared with the co-evaporation of elements technique. Co-evaporation of CIGS thin film via two-stage and three-stage processes has been widely used.
From a product perspective, using glass as a substrate in a CIGS photovoltaic device can be beneficial as the glass plate provides mechanical support as well as a smooth surface for high quality film growth. In addition, if a sodium-containing glass is used, it can serve as the source of sodium during film growth. However, glass can fracture when its surfaces or edges are placed into tension. Under these conditions, inherent surface or edge fissures may propagate into visible cracks. The basic principle employed in a heat treating process is to create an initial condition of surface and edge compression. This condition is achieved by first heating the glass, then cooling the surfaces rapidly. This leaves the center glass thickness relatively hot compared to the surfaces. As the center thickness then cools, it forces the surfaces and edges into compression. In some embodiment, glass can only withstand certain mechanical loads and the heat as well as cool-down rates need to be considered as well. The mechanical strength of a glass plate is a strong function of the thermal history of that plate. Glass is typically produced in an annealed condition (with minimal internal stress). Annealed glass tends to be fragile since mechanical strain causes micro-cracks present on the glass surface (called Grifith cracks) to propagate. Grifith crack defects cause the working tensile strength of soda-lime glass plate to typically be around 2000 psi, which is lower than the theoretical strength of bulk glass.
In order to overcome this fragility problem, glass plate is often quench strengthened by heating to a temperature sufficient to soften the glass (e.g., to 650° C., known as softening point), and then cooling rapidly (e.g., to 450° C., known as annealing point). The rapid cooling process can cause the exterior of the glass to harden while the core is still viscous and can relax to accommodate the thermal contraction of the hardened surface. Subsequent hardening of the core (center) of the glass below the annealing temperature causes it to solidify and then shrink, which puts the surfaces of the glass in compression. The resulting built-in compressive surface stress can allow the glass plate to withstand greater bending stresses before the tensile strength limit is reached. This strengthening process can be called “tempering.” The rapid cooling from the softening point to the annealing point can be accomplished in practice by directing high pressure jets of cooling gas onto the glass surface from both sides. In some cases liquid cooling is even used in order to maximize the “temper,” or surface compressive stress. In order to achieve rapid cooling heat transport via conduction is necessary. The latter could be achieved via a purge with helium or nitrogen gas at high flow rates.
Quench strengthening of the glass substrate can cause Se to desorb from the deposited CIGS film surface during the gas-flow cooling process, leaving behind Se vacancies in the lattice. The latter can reduce the voltage and efficiency of the resulting device. Thus, cool down is typically in a slight Se overpressure atmosphere down to 300° C., at which point the desorption loss of Se out of the CIGS surface becomes negligible. This is due to the fact that the partial pressure of Se above the CIGS becomes very small. Due to the heat capacity of the glass long vacuum cool down times are required making in-line CIGS coaters very long, or cool down sequences batch ovens very time consuming. This in turn increases production cost.
Air exposure of the CIGS absorber layer results in oxidation of the surface. Some of the resulting oxides can be removed in a subsequent process step, possibly in combination with the deposition of a buffer layer.
Another aspect of CIGS device processing is related to the fact that the band gap of the chalcopyrite semiconductor can be modified via interchanging elements. The most common two approaches are exchanging a portion of In with Ga or some of the Se with S. In conversion of metal or stacked elemental layers for CIGS formation it is difficult to accomplish the optimal surface band gap gradient via Ga and a combination of Ga and S grading is used. In a multi-stage co-evaporation process band gap grading via Ga alone is possible.
As discussed above, if a Na-containing glass is used as the substrate the Na in the glass can serve as the source of Na during film growth. The addition of Na during CIGS formation has been found to be critical for high-efficient devices. More recently it has been shown that a Na treatment step post CIGS growth can also improve the quality of the CIGS absorber and subsequent PV device.
To address the above aspects, a fast quench and oxidation protective cap layer on the CIGS absorber layer can be used. The concept is based on the use of a compatible material, e.g. Na2Se, deposited onto the CIGS absorber layer prior to quench. When a thin, ideally pinhole free, layer of cap layer is deposited, it can prevent subsequent Se loss from the electrically active junction area of the device during quench cool down by providing a Se-rich passivation layer. Use of a Na-containing compound can also serve as a post CIGS source for Na. Furthermore, this cap layer can allow for more rapid quench conditions and protect the CIGS surface from damage and oxidation. Less cool down times required in making in-line CIGS coaters can result in production cost reduction.
In one aspect, a method of manufacturing a structure includes forming a conductive layer adjacent to a substrate, forming a semiconductor absorber layer adjacent to the conductive layer, wherein the semiconductor absorber layer includes copper indium gallium selenide, and forming a protective cap layer adjacent to the absorber layer. The method can include removing the protective cap layer, forming a semiconductor buffer layer adjacent to the semiconductor absorber layer, and forming a transparent conductive oxide layer adjacent to the buffer layer. The method can include heating the substrate to soften the glass substrate after forming the protective cap and cooling the substrate to solidify and strengthen the glass substrate.
Cooling the substrate can include directing a fluid stream of coolant toward a surface of the glass substrate. The fluid stream can include helium. The fluid stream can include nitrogen. The fluid stream can include selenium. The fluid stream can include a gas. The fluid stream can include a liquid. Heating the substrate can include heating the substrate to a softening point temperature. Softening point temperature can be in the range of 500 degree C. to 800 degree C. Cooling the substrate stack can include cooling the substrate to an annealing point temperature. The annealing point temperature can be less than 500 degree C.
Forming the semiconductor absorber layer can include co-evaporating of a plurality of elements. Forming the semiconductor absorber layer can include forming and reacting a plurality of stacked elemental layers. Forming the semiconductor absorber layer can include forming and selenizing a metal precursor layer. Forming the semiconductor absorber layer can include modifying the photovoltaic device band gap by interchanging a plurality of elements. Forming the semiconductor absorber layer can include modifying the photovoltaic device band gap by exchanging a portion of indium with a portion of gallium. Forming the semiconductor absorber layer can include modifying the photovoltaic device band gap by exchanging a portion of selenium with a portion of sulfur.
The protective cap layer can include sodium. The protective cap layer can include a selenide. The protective cap layer can include sodium selenide. The protective cap layer can include a sulfide. The protective cap layer can have a thickness less than 1000 angstrom. The protective cap layer can have a thickness in the range of 200 angstrom to 600 angstrom. The protective cap layer can include a water soluble material. The method can include removing all or a portion of the protective cap layer after the step of cooling the substrate. The step of removing a portion of the protective cap layer can include contacting an aqueous medium to the protective cap layer.
In one aspect, a photovoltaic structure can include a substrate, a conductive layer adjacent to the substrate, a semiconductor absorber layer adjacent to the conductive layer, and a protective cap layer adjacent to the absorber layer. The absorber layer can include copper indium gallium selenide. The photovoltaic device can include a protective cap layer adjacent to the absorber layer.
The protective cap layer can include sodium. The protective cap layer can include a selenide. The protective cap layer can include sodium selenide. The protective cap layer can include a sulfide. The protective cap layer can have a thickness less than 1000 angstrom. The protective cap layer can have a thickness in the range of 200 angstrom to 600 angstrom. The protective cap layer can include a water soluble material. The structure can include a semiconductor buffer layer adjacent to the semiconductor absorber layer. The absorber layer can include copper indium gallium selenide. The structure can include a transparent conductive oxide layer adjacent to the buffer layer.
Referring to
Referring to
The layers included in
Referring to
Protective cap layer 260 can be formed by any suitable method. For example, protective cap layer 260 can be created by any suitable deposition means. After protective cap layer 260 is formed adjacent to semiconductor absorber layer 230, structure 201 can be heated to soften substrate 210 and subsequently cooled (e.g., in a quench process) to solidify and/or strengthen substrate 210, for example in cases where substrate 210 includes glass. Heating structure 201 can include heating substrate 210 to a softening point temperature of substrate 210. The softening point temperature can be any suitable temperature to soften the substrate. For example, the softening point temperature can be in the range of 500 degrees C. to 800 degrees C.
Structure 201 and/or substrate 210 can be subsequently cooled by any suitable method. For example, substrate 210 can be cooled by directing a fluid stream of coolant toward a surface of the substrate. The fluid stream can include a gas and/or a liquid. The fluid stream can include any suitable material. For example, the fluid stream can include helium, nitrogen, or selenium. Substrate 210 can be cooled to a temperature at or below its annealing point. The annealing point can be, for example, 500 degrees C.
In some embodiments, use of a sodium-containing protective cap layer 260 deposited at high temperatures could provide sodium-dosing in case of sodium-free substrates as an alternative of pre-CIGS growth deposition of a sodium-containing precursor onto the back contact coated substrate. Sodium sulfide is also readily soluble in water. Moreover, using sodium sulfide can allow for optimized surface band gap grading with or without the combination of gallium.
Referring to
A number of embodiments of the invention have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the invention. It should also be understood that the appended drawings are not necessarily to scale, presenting a somewhat simplified representation of various preferred features illustrative of the basic principles of the invention.
This application claims priority to U.S. Provisional Patent Application No. 61/306,831, filed on Feb. 22, 2010, which is incorporated by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61306831 | Feb 2010 | US |
