Collector grid, electrode structures and interrconnect structures for photovoltaic arrays and methods of manufacture

This application is a Continuation-in-Part of U.S. patent application Ser. No. 11/824,047 filed Jun. 30, 2007, entitled Collector Grid, Electrode Structures and Interconnect Structures for Photovoltaic Arrays and other Optoelectric Devices, which is a Continuation-in-Part of U.S. application Ser. No. 11/404,168 filed Apr. 13, 2006, entitled Substrate and Collector Grid Structures for Integrated Photovoltaic Arrays and Process of Manufacture of Such Arrays, which is a Continuation-in-Part of U.S. application Ser. No. 10/776,480 filed Feb. 11, 2004, entitled Methods and Structures for the Continuous Production of Metallized or Electrically Treated Articles, now abandoned, which is a Continuation-in-Part of U.S. patent application Ser. No. 10/682,093 filed Oct. 8, 2003 entitled Substrate and Collector Grid Structures for Integrated Series Connected Photovoltaic Arrays and Process of Manufacture of Such Arrays, which is a Continuation-in-Part of U.S. patent application Ser. No. 10/186,546 filed Jul. 1, 2002, entitled Substrate and Collector Grid Structures for Integrated Series Connected Photovoltaic Arrays and Process of Manufacture of Such Arrays, now abandoned, which is a Continuation-in-Part of U.S. patent application Ser. No. 09/528,086, filed Mar. 17, 2000, entitled Substrate and Collector Grid Structures for Integrated Series Connected Photovoltaic Arrays and Process of Manufacture of Such Arrays, and now U.S. Pat. No. 6,414,235, which is a Continuation-in-Part of U.S. patent application Ser. No. 09/281,656, filed Mar. 30, 1999, entitled Substrate and Collector Grid Structures for Electrically Interconnecting Photovoltaic Arrays and Process of Manufacture of Such Arrays, and now U.S. Pat. No. 6,239,352. The entire contents of the above identified applications are incorporated herein by this reference.

This invention teaches novel structure and methods for achieving efficient collection and conveyance of power from photovoltaic power generating devices.

Photovoltaic cells have developed according to two distinct methods. The initial operational cells employed a matrix of single crystal silicon appropriately doped to produce a planar p-n junction. An intrinsic electric field established at the p-n junction produces a voltage by directing solar photon produced holes and free electrons in opposite directions. Despite good conversion efficiencies and long-term reliability, widespread energy collection using single-crystal silicon cells is thwarted by the high cost of single crystal silicon material and interconnection processing.

A second approach to produce photovoltaic cells is by depositing thin photovoltaic semiconductor films on a supporting substrate. Material requirements are minimized and technologies can be proposed for mass production. Thin film photovoltaic cells employing amorphous silicon, cadmium telluride, copper indium gallium diselenide, dye sensitized solar cells (DSSC), printed silicon inks and the like have received increasing attention in recent years.

Despite significant improvements in individual cell conversion efficiencies for both single crystal and thin film approaches, photovoltaic energy collection has been generally restricted to applications having low power requirements. One factor characteristic of many optoelectric devices and photovoltaic cells in particular is that electrical energy is produced over a relatively expansive surface area. Thus a challenge to implementing bulk power systems is the problem of economically collecting the photogenerated power from an expansive surface. In particular, photovoltaic cells can be described as high current, low voltage devices. Typically individual cell voltage is less than about two volts, and often less than 0.6 volt. The current component is a substantial characteristic of the power generated. Efficient power collection from expansive photovoltaic cell surfaces must minimize resistive losses associated with the high current characteristic. In the specific case of most photovoltaic cells, the upper surface is normally formed by a transparent conductive oxide (TCO). However, these TCO layers are relatively resistive compared to pure metals and have a surface resistivity on the order of 10 to 100 ohms per square. Thus the conductive surface itself is limited in its ability to collect and transport current and efforts must be made to minimize resistive losses in transport of current through the TCO layer. This problem increases in severity as individual cell sizes increase. One solution is to simply reduce individual cell size (and thus accumulated current from an individual cell) to a point where the transparent conductive oxide alone can handle the current. Where larger individual cell sizes are the norm, it is common practice to augment the transparent conductive oxide with a current collector structure comprising a pattern of highly conductive traces extending over substantially the entire surface from which current is to be collected. Often the structure is in the form of a grid or lattice pattern. The current collector structure reduces the distance that current must be transported by the transparent conductive oxide before it reaches a highly conductive conveyance path off the surface. Thus the current collector structure collects current from a surface having relatively low surface conductivity. Many current collector structures or grids are conventionally prepared by first applying metal wires, fused silver filled pastes or silver filled ink traces to the cell surface and then covering the surface with a sealing material in a subsequent operation. These highly conductive traces may lead to a collection buss such as a copper foil strip which also functions as a tab extending to the back electrode of an adjacent cell. The wire approach requires positioning and fixing of multiple fine fragile wires which makes mass production difficult and expensive. Silver pastes are expensive and require high fusion temperatures which not all photovoltaic semiconductors can tolerate. A silver filled ink, as compared to a fuseable paste, is simply dried or cured at mild temperatures which do not adversely affect the cell. However, this ink approach requires the use of relatively expensive inks because of the high loading of finely divided silver particles. In addition, batch printing on the individual cells is laborious and expensive. Finally, the silver filled ink is relatively resistive compared to a fuseable silver paste or metal wire. Typical silver filled inks have intrinsic resistivities in the range 0.00002 to 0.01 ohm-cm.

Thus there remains a need for improved materials and structure for collecting the current from the top light incident surface of photovoltaic cells.

Normally one envisions a photovoltaic power collection device much larger than the size of an individual cell. Therefore, an arrangement must be supplied to collect power from multiple cells. This is normally accomplished by interconnecting multiple cells in series. In this way, voltage is stepped through each cell while current and associated resistive losses are minimized. Such interconnected multi-cell arrangements are commonly referred to as “modules” or “arrays”. However, it is readily recognized that making effective, durable series connections among multiple small cells can be laborious, difficult and expensive. Regarding traditional crystalline silicon cells, the individual cells are normally discrete and comprise rigid wafers approximately 200 micrometers thick and approximately 230 square centimeters in area. A common way to convert multiple such cells into modules is to use a conventional “string and tab” arrangement. In this process multiple discrete cells are arranged in “strings” and the topside current collector electrodes of cells are connected to backside electrodes of adjacent cells using “tabs” or ribbons of conductive material. The cell connections often involve tedious manual operations such as soldering and handling of multiple interconnected cells. Next, unwieldy flexible leads from the terminal cells must be directed and secured in position for outside connections, again a tedious operation. Finally, weight and assembly concerns limit the ultimate size of the module. These limitations impede adoption of the modules for large scale power generation.

In order to approach economical mass production of modules of series connected individual cells, a number of factors must be considered in addition to the type of photovoltaic materials chosen. These include the substrate employed and the process envisioned. Since thin films can be deposited over expansive areas, thin film technologies offer additional opportunities for mass production of interconnected modules compared to inherently small, discrete single crystal silicon cells. Thus a number of U.S. patents have issued proposing designs and processes to achieve series interconnections among thin film photovoltaic cells. Many of these technologies comprise deposition of photovoltaic thin films on glass substrates followed by scribing to form smaller area individual cells. Multiple steps then follow to electrically connect the individual cells in series while maintaining the original common glass substrate. These “common” substrate approaches have come to be known as “monolithic integration”. Examples of these proposed processes are presented in U.S. Pat. Nos. 4,443,651, 4,724,011, and 4,769,086 to Swartz, Turner et al. and Tanner et al. respectively. While expanding the opportunities for mass production of interconnected cell modules compared with inherently discrete approaches for crystal silicon cells, monolithic integration employing common glass substrates must inherently be performed on an individual batch basis. In addition, many monolithic approaches are not compatible with the use of a current collector grid and therefore cell sizes (in the direction of current flow) are constrained. Typically, cell widths for monolithic integration between 0.5 cm. and 1.0 cm. are taught in the art. However, as cell widths decrease, the width of the area between individual cells (interconnect area) should also decrease so that the relative portion of inactive surface of the interconnect area does not become excessive. These small cell widths demand very fine interconnect area widths, which dictate delicate and sensitive techniques to be used to electrically connect the top TCO surface of one cell to the bottom electrode of an adjacent series connected cell. Furthermore, achieving good stable ohmic contact to the TCO cell surface has proven difficult, especially when one employs those sensitive techniques available when using the TCO only as the top collector electrode.

More recently, developers have explored depositing wide area films using continuous roll-to-roll processing. This technology generally involves depositing thin films of photovoltaic material onto a continuously moving web. However, a challenge still remains regarding subdividing the expansive films into individual cells followed by interconnecting into a series connected array. For example, U.S. Pat. No. 4,965,655 to Grimmer et. al. and U.S. Pat. No. 4,697,041 to Okamiwa teach processes requiring expensive laser scribing and interconnections achieved with laser heat staking. In addition, these two references teach a substrate of thin vacuum deposited metal on films of relatively expensive polymers. The electrical resistance of thin vacuum metallized layers may significantly limit the active area of the individual interconnected cells.

It has become well known in the art that the efficiencies of certain promising thin film photovoltaic junctions can be substantially increased by high temperature treatments. These treatments involve temperatures at which even the most heat resistant plastics suffer rapid deterioration. Use of a metal foil as a substrate allows high temperature treatments and continuous roll-to-roll processing. However, the subsequent conversion to an interconnected module of multiple cells has proven difficult, in part because the metal foil substrate is electrically conducting.

U.S. Pat. No. 4,746,618 to Nath et al. teaches a design and process to achieve interconnected arrays using roll-to-roll processing of a metal web substrate such as stainless steel. The process includes multiple operations of cutting, selective deposition, and riveting. These operations add considerably to the final interconnected array cost.

U.S. Pat. No. 5,385,848 to Grimmer teaches roll-to-roll methods to achieve integrated series connections of adjacent thin film photovoltaic cells supported on an electrically conductive metal substrate. The process includes mechanical or chemical etch removal of a portion of the photovoltaic semiconductor and transparent top electrode to expose a portion of the electrically conductive metal substrate. The exposed metal serves as a contact area for interconnecting adjacent cells. These material removal techniques are troublesome for a number of reasons. First, many of the chemical elements involved in the best photovoltaic semiconductors are expensive and environmentally unfriendly. This removal subsequent to controlled deposition involves containment, dust and dirt collection and disposal, and possible cell contamination. This is not only wasteful but considerably adds to expense. Secondly, the removal processes are difficult to control dimensionally. Thus a significant amount of the valuable photovoltaic semiconductor is lost to the removal process. Ultimate module efficiencies are further compromised in that the spacing between adjacent cells grows, thereby reducing the effective active collector area for a given module area.

Thus there remains a need for acceptable mass manufacturing processes and articles to achieve effective integrated interconnections among photovoltaic cells.

A further issue that has impeded adoption of photovoltaic technology, especially for bulk power collection in the form of solar farms, involves installation of multiple modules over expansive regions of surface. Traditionally, modules have been mounted individually on supporting mounts, normally at an incline to horizontal appropriate to the latitude of the site. Conducting leads from each module are then physically coupled with leads from an adjacent module in order to interconnect multiple modules. This arrangement results in a string of modules each of which is coupled to an adjacent module. At one end of the string, the power is transferred from the end module to be conveyed to a separate site for further treatment such as voltage adjustment. This arrangement avoids having to run conductive cabling from each individual module to the separate treatment site.

The traditional solar farm installation described in the above paragraph has some drawbacks. First, traditional modules are limited in size due to weight and manufacturing constraints. This fact increases the number of individual modules required to cover a desired surface area. Next, the module itself comprises a string of individual cells. In the conventional module lead conductors in the form of flexible wires or ribbons are attached to an electrode on the two cells positioned at each end of the string in order to convey the power from the module. After mounting the individual modules on their support at the installation site, the respective leads from adjacent modules must be connected in order to couple adjacent modules, and the connection must be protected to avoid environmental deterioration or separation. These are intrinsically tedious manual operations. Finally, since the module leads and cell interconnections are not of high current carrying capacity, the adjacent cells are normally connected in series arrangement. Thus voltage builds up to high levels even at relatively short strings of modules. While not an overriding problem security and insulation must be appropriate to eliminate a shock hazard.

Thus there remains a need for improved module form factors and complimentary installation structure to reduce the cost and complexity of achieving large area “utility” scale photovoltaic installations.

In a somewhat removed segment of technology, a number of electrically conductive fillers have been used to produce electrically conductive polymeric materials. This technology generally involves mixing of a conductive filler such as silver particles with the polymer resin prior to fabrication of the material into its final shape. Conductive fillers may have high aspect ratio structure such as metal fibers, metal flakes or powder, or highly structured carbon blacks, with the choice based on a number of cost/performance considerations. More recently, fine particles of intrinsically conductive polymers have been employed as conductive fillers within a resin binder. Electrically conductive polymers have been used as bulk thermoplastic compositions, or formulated into paints and inks. Their development has been spurred in large part by electromagnetic radiation shielding and static discharge requirements for plastic components used in the electronics industry. Other known applications include resistive heating fibers and battery components and production of conductive patterns and traces. The characterization “electrically conductive polymer” covers a very wide range of intrinsic resistivities depending on the filler, the filler loading and the methods of manufacture of the filler/polymer blend. Resistivities for filled electrically conductive polymers may be as low as 0.00001 ohm-cm. for very heavily filled silver inks, yet may be as high as 10,000 ohm-cm or even more for lightly filled carbon black materials or other “anti-static” materials. “Electrically conductive polymer” has become a broad industry term to characterize all such materials. In addition, it has been reported that recently developed intrinsically conducting polymers (absent conductive filler) may exhibit resistivities comparable to pure metals.

In yet another separate technological segment, coating plastic substrates with metal electrodeposits has been employed to achieve decorative effects on items such as knobs, cosmetic closures, faucets, and automotive trim. The normal conventional process actually combines two primary deposition technologies. The first is to deposit an adherent metal coating using chemical (electroless) deposition to first coat the nonconductive plastic and thereby render its surface highly conductive. This electroless step is then followed by conventional electroplating. ABS (acrylonitrile-butadiene-styrene) plastic dominates as the substrate of choice for most applications because of a blend of mechanical and process properties and ability to be uniformly etched. The overall plating process comprises many steps. First, the plastic substrate is chemically etched to microscopically roughen the surface. This is followed by depositing an initial metal layer by chemical reduction (typically referred to as “electroless plating”). This initial metal layer is normally copper or nickel of thickness typically one-half micrometer. The object is then electroplated with metals such as bright nickel and chromium to achieve the desired thickness and decorative effects. The process is very sensitive to processing variables used to fabricate the plastic substrate, limiting applications to carefully prepared parts and designs. In addition, the many steps employing harsh chemicals make the process intrinsically costly and environmentally difficult. Finally, the sensitivity of ABS plastic to liquid hydrocarbons has prevented certain applications. ABS and other such polymers have been referred to as “electroplateable” polymers or resins. This is a misnomer in the strict sense, since ABS (and other nonconductive polymers) are incapable of accepting an electrodeposit directly and must be first metallized by other means before being finally coated with an electrodeposit. The conventional technology for electroplating on plastic (etching, chemical reduction, electroplating) has been extensively documented and discussed in the public and commercial literature. See, for example, Saubestre, Transactions of the Institute of Metal Finishing, 1969, Vol. 47, or Arcilesi et al., Products Finishing, March 1984.

Many attempts have been made to simplify the process of electroplating on plastic substrates. Some involve special chemical techniques to produce an electrically conductive film on the surface. Typical examples of this approach are taught by U.S. Pat. No. 3,523,875 to Minklei, U.S. Pat. No. 3,682,786 to Brown et. al., and U.S. Pat. No. 3,619,382 to Lupinski. The electrically conductive film produced was then electroplated. None of these attempts at simplification have achieved any recognizable commercial application.