WO1987001513A2

WO1987001513A2 - Gallium arsenide solar cell system

Info

Publication number: WO1987001513A2
Application number: PCT/US1986/001845
Authority: WO
Inventors: M. Edmund Ellion; George Wolff
Original assignee: Hughes Aircraft Company
Priority date: 1985-09-09
Filing date: 1986-09-08
Publication date: 1987-03-12
Also published as: EP0236447A1; IL79993A0; WO1987001513A3

Abstract

An N-on-P gallium arsenide solar cell, and solar cell arrays incorporating N-on-P gallium arsenide solar cells. In the N-on-P gallium arsenide solar cell, the n-type gallium arsenide faces the sun. The solar cell arrays utilize geometrically alternating, laterally disposed N-on-P solar cells and P-on-N solar cells. Series electrical connections between adjacent solar cells in the array are accomplished by connecting the lower faces of a first and a second solar cell, the upper faces of the second and a third solar cell, and so forth in an alternating pattern. The connections between the upper faces can be eliminated entirely by placing pairs of the N-on-P and P-on-N solar cells in lateral contact to form a module, extending a connector grid across the upper surface of the module, and using strip connectors to joint the bases of adjacent modules.

Description

GALLIUM ARSENIDE SOLAR CELL SYSTEM

BACKGROUND OF THE INVENTION

This invention relates to solar cells, and, more particularly, to N-on-P gallium arsenide solar cells and solar cell arrays utilizing N-on-P solar cells.

Semiconductor solar cells are utilized - to convert light energy to useable electrical voltages and currents. Briefly, a typical semiconductor solar cell includes an interface between n-type and p-type transparent semiconductor materials. Light shining on the interface creates hole- electron pairs in addition to those otherwise present, and the minority charge carriers migrate across the interface in opposite directions. There is no compensating flow of majority carriers, so that a net flow of electrical charge results. A useful electrical current is then obtained in an external electrical circuit by forming ohmic contacts to the materials on either side of the interface.

Semiconductor solar cells may be produced from a wide variety of semiconductor materials. Silicon solar cells are most widely used, but it has been found that cells fabricated from p-type and n-type gallium arsenide are particularly promising. Such solar cells have higher beginning- of -life efficiency and lower degradation with time and temperature in a space environment, as compared with silicon solar cells. Gallium arsenide solar cells are therefore particularly attractive, and have already found limited use. It is expected that gallium arsenide solar cells will find increased future application, in both space and on earth, particularly if the efficiency of solar cell arrays can be improved and inexpensive fabrication techniques are developed.

A gallium arsenide solar cell is fabricated by depositing the appropriate semiconductor layers onto a substrate, and then adding additional components to complete the cell. More specifically, a conventional

P-on-N gallium arsenide solar cell is fabricated by epitaxially depositing a layer of n-type gallium arsenide onto a single crystal gallium arsenide substrate, and depositing a layer of p-type gallium arsenide over the layer of n-type gallium arsenide. A

P+ layer of gallium aluminum arsenide is deposited over the layer of p-type gallium arsenide to limit surface recombination of charge carriers. A transparent cover of glass is applied over the gallium aluminum arsenide to protect the active semiconductor components from physical contact and radiation damage such as encountered in a space environment. The p-type gallium arsenide faces the sun during operation of the cell, as indicated by the terminology "P-on-N" solar cell. The individual solar cells, typically measuring about 2 centimeters by 4 centimeters in lateral dimensions, are joined together in large arrays to produce useable electrical voltages and currents.

The arrays may have as many as ten thousand individual solar cells. Since the electrical output of each individual P-on-N solar cell is only about 0.9 volts, in an array a number of P-on-N solar cells are connected in a series fashion to provide an electrical voltage which is the sum of the voltages of the indi- vidual series-connected solar cells.

To accomplish the series electrical connection, the upper layer (i.e. p-type gallium arsenide) of a first solar cell is connected to the lower layer (i.e. n-type gallium arsenide) of a laterally adjacent second solar cell, and this connection approach is repeated from the second to the third solar cell, and so forth. This connection approach requires that the laterally adjacent solar cells be spaced a sufficient distance apart, typically two millimeters, that a connector can be inserted between the laterally adjacent cells. λ ^Mz" connector is used for making the connection, with the upper leg of the "z" soldered to the top of the first solar cell and the lower leg soldered to the bottom of the laterally adjacent second solar cell. The active area of the cell that is available to produce electricity is reduced by the area covered or shaded by the connector attachment to the top surface. The necessary lateral spacing of the cells which permits the insertion of the "z" connector also reduces the electrical efficiency of the array. (The term "efficiency" is used here to mean the electrical output of the array per unit area of the array.) The geometrical limitations thereby imposed on the efficiency of the solar cell arrays, due to the spacing needed between adjacent cells to insert the connectors, and the area shaded by the end connections, can significantly reduce the overall efficiency of the array in terms of electrical output per unit area of array.

There therefore exists a continuing need for improving the efficiency of a solar cell array. A more efficient solar cell array would desirably utilize the benefits achievable through the use of known solar cell materials of construction. The solar cell array also should not be less resistant to radiation damage in a solar space environment than existing types of arrays constructed from the same materials of construction.

The present invention fulfills this need, and further provides related advantages. SUMMARY OF THE INVENTION

The present invention resides in an improved, gallium arsenide solar cell, and solar cell arrays using the improved solar cell. The improved solar cell incorporates a modified construction which achieves substantially the same performance as conventional gallium arsenide solar cells, and in addition allows the interconnection of solar cells in an array to be accomplished in a manner which increases the overall output efficiency of the array. Arrays employing the improved solar cell require less complex interconnection hardware and procedures, allow closer packing of the individual solar cells to achieve increased electrical output per unit area of the array, and provide decreased problems resulting from differential thermal expansion of the individual cells in the array. The solar cell arrays of the invention can otherwise be used in a fashion identical with that of existing solar cell arrays, and have the same resistance to radiation damage.

In accordance with the invention, an N-on-P gallium arsenide solar cell comprises a single crystal gallium arsenide substrate, a single crystal layer of p+ gallium aluminum arsenide epitaxially overlying the substrate, a single crystal layer of p-type gallium arsenide epitaxially overlying the layer of gallium aluminum arsenide, and a single crystal layer of n-type gallium arsenide epitaxially overlying the layer of p-type gallium arsenide. In operation, the n-type gallium arsenide layer faces the sun. This N-on-P cell achieves similar electrical output and resistance to radiation damage as conventional P-on-N gallium arse¬ nide solar cells. The N-on-P solar cell is covered with a transparent cover such as glass to protect it and allow handling. The layer of p-type gallium arse¬ nide is typically about 0.5 micrometers thick, and the layer of n-type gallium arsenide about 10 micrometers thick. In accordance with another aspect of the invention, a solar cell array comprises an N-on-P first solar cell, a P-on-N second solar cell laterally adjacent said first solar cell, and an electrical connector extending from the p-type layer of the first solar cell to the n-type layer of the second solar cell. The p-type layer of the first solar cell is directly adjacent the n-type layer of the second solar cell, since the p and n-type layers are inverted as between the first and second solar cells. This inversion of the active layers allows the electrical connector between the adjacent cells to extend laterally between the cells, rather than from the top layer of one cell to the bottom layer of the next cell. Preferably, the connector is a flat metallic strip which extends between the surfaces of the laterally adjacent layers. The solar cells are preferably gallium arsenide solar cells. In another embodiment, and in the alternative, the n-type layer of the first solar cell would be connected to the p-type layer of the second solar cell, rather than the p-type layer of the first solar cell being connected to the n-type layer of the second solar cell.

In yet another aspect of the invention, a solar cell module comprises an N-on-P solar cell laterally joined to a P-on-N solar cell, and a connector grid extending across the top surfaces of both cells. A solar cell array is formed by placing such modules laterally adjacent each other and electrically connecting adjacent bottom surfaces of the p-layer of the N-on-P solar cell to the n-layer of the adjacent P-on-N solar cell of the adjacent module. When the adjacent modules are connected in this fashion, no top connections are required, so that there is no shading of a portion of each solar cell, with consequent reduced efficiency, due to top connections.

The use of laterally adjacent, geometrically alternating N-on-P and P-on-N solar cells in a solar cell array allows increased ease of construction of the solar cell array, since laterally adjacent n-type and p-type layers can be connected directly together, without the need for top-to-bottom connectors such as

"z" type connectors. Consequently, the solar cells may be more tightly packed in the solar cell array, leading to higher array efficiencies. Modules may be constructed to be joined in an array which has no top connections that shade a portion of the active area of the array. Other features and advantages of the present invention will become apparent from the following more detailed description, taken in conjunction with the accompanying drawings, which illustrate, by way of example, the principles of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGURE 1 is an elevational view of a conventional P-on-N solar cell;

FIGURE 2 is an elevational view of an N-on-P solar cell;

FIGURE 3 is an elevational view of a conventional solar cell array incorporating P-on-N solar cells;

FIGURE 4 is a top plan view of the solar cell array of FIGURE 3;

FIGURE 5 is an elevational view of a solar cell array utilizing alternating P-on-N and N-on-P solar cells; FIGURE 6 is a top plan view of the solar cell array of FIGURE 5;

FIGURE 7 is an elevational view of a solar cell array utilizing geometrically alternating P-on-N and N-on-P solar cells, and eliminating the top connections; and

FIGURE 8 is a top plan view of the solar cell array of FIGURE 7.

DETAILED DES CRIPTION OF THE PREFERRED EMBODIMENT

A conventional P-on-N gallium arsenide solar cell 10, as illustrated in FIGURE 1, is prepared in the following manner. The solar cell 10 is fabricated on a gallium arsenide single crystal substrate 12, which may be prepared by any of several well established techniques. Most commonly, a gallium arsenide single crystal is fabricated by the horizontal Bridgeman technique. Substrates suitable for use in the preparation of solar cells are prepared by slicing wafers about 200 micrometers thick from the solidified single crystal. The surface orientation of the wafers is typically selected to be about (100) (cubic Miller indices notation) . Gross damage induced during the slicing procedure is removed by polishing the wafer on successively finer grits of metallographic polishing paper, finishing with a 4/0 diamond paste. The polished wafer is then etched in a solution of hydrogen peroxide and ammonium hydroxide to remove any residual polishing damage.

A single crystal layer 14 of n-type gallium arsenide is epitaxially deposited overlying the gallium arsenide single crystal substrate 12. The n-type gallium arsenide of the layer 14 preferably has a net charge carrier concentration of about 10¹⁸ per cubic centimeter and a thickness of about 10 micrometers.

A single crystal layer 16 of p-type gallium arsenide is next epitaxially deposited overlying the layer 14 of n-type gallium arsenide. Preferably, the p-type gallium arsenide in the layer 16 has a net charge carrier concentration of about 10¹⁸ per cubic centimeter, with a thickness of about 0.5 micrometers or less. The junction between the layer 14 of n-type gallium arsenide and the layer 16 of p-type gallium arsenide provides a basic solar cell, but the operation of this solar cell is adversely affected by surface charge recombination at the top surface of the layer 16, unless charge recombination is inhibited. Con- ventionally, to inhibit surface charge recombination, a p+ gallium aluminum arsenide single crystal inhibitory layer 18 is epitaxially deposited on top of the layer 16 of p-type gallium arsenide. The p+ gallium aluminum arsenide layer 18 preferably has a net charge carrier concentration of about 2 x 10¹⁸ per cubic centimeter, with a thickness of about 0.1 micrometers. A typical composition of gallium aluminum arsenide having these characteristics is about Ga_{0> 7}Al_Q>3As.

The layers 14, 16 and 18 are deposited by techniques well known in the art, preferably liquid phase epitaxial growth or vapor phase metal oxide chemical vapor deposition. The liquid phase epitaxial growth of gallium arsenide solar cells is accomplished by dipping substrates of gallium arsenide into a gallium melt saturated with gallium arsenide. This process is performed with the melt at 750 "C in a sealed, nitrogen-purged system. Vapor . phase metal oxide chemical vapor deposition takes place when tri-metal gallium in a gaseous form mixes with arsine gas. The mixture decomposes into gallium arsenide and is deposited on a gallium arsenide substrate at 750 ' C in a low-pressure chamber.

To faciliate electrical connection to the upper and lower surfaces of the solar cell 10, a metallic connector grid 20 is deposited on an upper face 22 of the solar cell 10. The metal is deposited in a vacuum environment using conventional sputtering techniques. The individual lines of the connector grid 20 are spaced about two millimeters apart, so that electron charge carriers may readily diffuse through the semiconductor layers 14, 16 and 18 to be collected by the individual elements of the connector grid 20. If the individual elements are too widely spaced, they cannot readily collect the electrons.

A transparent cover 24 is attached to the upper face 22. The composition and thickness of the transparent cover 24 are selected to optimize the electrical performance of the solar cell 10. Preferably, the transparent cover 24 is a silica glass such as Corning Glass type 7940, having a thickness of about 200 micrometers. The transparent cover 24 performs three important functions. ^" First, the cover 24 allows light to pass through to the layers 14 and 16. Second, the cover 24 supports the remaining elements of the solar cell 10. Third, the cover 24 protects the remaining elements of the solar cell 10 from physical damage and radiation in a space environment.

The transparent cover 24 may be bonded to the layer 18 by any suitable technique, such as by a transparent adhesive or electrostatic bonding. It is preferable that the bonding technique permit retention of the bond at temperatures as high as about 200 * C, to resist delamination in subsequent processing and in use. High temperature adhesives of high molecular weight compounds such as carborane εiloxane polymer have been found to be operable.

In accordance with one aspect of the present invention, and as illustrated in FIGURE 2, an N-o^ή-P gallium arsenide solar cell 26 is prepared ._r by fur¬ nishing a gallium arsenide single crystal substrate 28 substantially identical to the substrate . 12 previously described. A single crystal layer 30 of p+ gallium aluminum arsenide . is epitaxially deposited overlying the substrate 28, to inhibit surface charge recom¬ bination. A single crystal layer 32 of p-type gallium arsenide is then epitaxially deposited overlying the layer 30 of p+ gallium aluminum arsenide. A single crystal layer 34 of n-type gallium arsenide is then epitaxially deposited overlying the layer 32 of p-type gallium arsenide. A connector grid 36 is deposited upon an upper face 38 of the layer 34 of n-type gallium arsenide. Finally, a transparent cover 40 is fastened over the connector grid 36 and the layer 34.

The characteristics of the layers 30, 32 and 34, the methods for depositing the layers 30, 32 and 34, the geometrical arrangement, structure and method of depositing the connector grid 36, and the structure and method of attaching the transparent cover 40 in relation to the N-on-P gallium arsenide solar cell 26 are all substantially identical to the corresponding aspects of the conventional P-on-N gallium arsenide solar cell 10, described in the preceding paragraphs.

Only the order of depositing the active layers 30, 32 and 34 to form the N-on-P solar cell 26 differs from the order of depositing the layers 14, 16 and 18 to form the conventional P-on-N solar cell 10 is different.

It has been found that the N-on-P solar cell 26 exhibits nearly identical electrical performance and resistance to radiation damage as the P-on-N solar cell 10. In fact, the electrical performance of the N-on-P gallium arsenide solar cell degrades slightly less with the passage of time, as compared with a conventional P-on-N gallium arsenide solar cell. The maximum power available from the conventional P-on-N solar cell will degrade approximately 20 percent when exposed to 1 MEV electrons at a fluence of 10"¹- , which is equivalent to almost 5 years in synchronous orbit.

Individual P-on-N solar cells 10 each produce a voltage output of about 0.9 volts, which is too low a voltage for any practical application in a space environment. The P-on-N solar cells 10 are therefore conventionally hooked together in a series fashion to obtain an output voltage equal to the sum of the voltages produced by the individual solar cells 10.

Identical groups of the solar cells joined in series are then hooked together in a parallel fashion to achieve increased electrical currents as required.

FIGURES 3 and 4 illustrate the manner of interconnecting conventional P-on-N solar cells to form a conventional solar cell array 42. To achieve a series connection, the top or p-type layer 16 of one solar cell 10 must be electrically connected to the bottom or n-type layer 14 of the adjacent cell. The type of connectors utilized are termed "z- connectors"

44, because of their shape when viewed in an ele¬ vational view. The z-connectors 44 serve as bus bars to interconnect each of the individual elements of the connector grid 20 together, and also to connect these elements to the adjacent solar cell 10. The convent¬ ional z-connector 44 includes a horizontal portion at each end and an inclined portion extending from the top to bottom of the solar cells. The horizontal portions are soldered or otherwise connected to the surfaces of the solar cell 10 to form the bus bars 46. The area of the attachment under the horizontal portion of the z- σonnector 44, the bus bar 46, is inactive and cannot produce an electrical current, since the junction between the layers 14 and 16 is shaded from the rays of the sun by the bus bar 46. The electrical output of a conventional solar cell- array 42 is therefore reduced below its potential current output due to the spacing between adjacent solar cells, which must be maintained to_., accommodate the z-connector 44, and the inactive area shaded by the bus bar 46 extending along one side of each solar cell. In accordance with another aspect of the invention, P-on-N solar cells and N-on-P solar cells may be geometrically alternated to form solar cell arrays whose electrical current output efficiency is increased because the individual solar cells may be placed more closely together, and because in some configurations the bus bars joined to the upper surfaces of the solar cells may be eliminated.

FIGURES 5 and 6 illustrate a first configuration 48 wherein the z-type connectors are replaced by flat strip connectors 50 disposed between adjacent N-on-P solar cells 26 and P-on-N solar cells 10. Using alternating N-on-P and P-on-N solar cells, series connections may be produced by simply connecting the bottom of the first P-on-N solar cell 10 to the bottom of the first N-on-P solar cell 26, the top of the first N-on-P solar cell 26 to the top of the second P-on-N solar cell 10', the bottom of the second P-on-N solar cell 10' to the bottom of the second N-on-P solar cell 26' and so forth. The connector 50 need not extend from the top of one solar cell to the bottom of the adjacent solar cell, so that the z-type connector 44 can be replaced by the flat strip connector 50. The flat strip connector 50 is more easily installed in the solar cell array, thus saving on installation time and expense. Additionally, the adjacent solar cells 10 and

26 may be positioned more closely together, reducing the linear cell spacing 52 as compared with that of the conventional array 42, with the result that an equivalent number of solar cells may be packed into a tighter area of the solar cell array of the first

^" configuration 48. The electrical current per unit area of the solar cell array is thereby increased by modifying the geometrical packing of the individual solar cells.

FIGURE 6 is a top plan view of the first configuration 48, illustrating the alternating nature of the arrangement of the strip connectors 50, which act as bus bars connecting the connector grids 20 and 36 of the adjacent solar cells. With this first configuration 48, there continues to exist an inactive area under each of the strip connectors 50 connecting the tops of the individual solar cells.

FIGURES 7 and 8 illustrate a second configuration 54 wherein a P-on-N solar cell 10 and an N-on-P solar cell 26 are grouped together as a module 56. In such a module 56, a connector grid 58 extends continuously across the upper surface of the module 56 between the solar cells 10 and 26, electrically connecting the p layer of the P-on-N solar cell 10 to the n layer of the N-on-P solar cell 26. Adjacent modules 56 are connected together only at their lower surfaces, by a strip connector bar 60. The strip connector bar 60 provides an electrical connection from the N-on-P solar cell 26 of a first active pair 56 to the P-on-N solar cell 10 of a second active pair 56, and so on. The connector bars 60 may also partially or fully support all of the weight of the second configuration 54.

As may be seen in FIGURES 7 and 8, the second configuration 54 optimizes the output efficiency of an array by reducing both geometrical components of electrical output loss found in conventionail solar cell arrays 42. The loss due to the spacing between adjacent cells is drastically reduced, inasmuch as there is no spacing between the two solar cells comprising each module 56, and a spacing 62 between adjacent modules 56 and 56' may be reduced through use of the flat strip connector 60, in the manner previously discussed in relation to the first configuration 48. Additionally, the need for bus bars at the upper surfaces of the solar cells is eliminated through the use of the continuous connector grid 58. That is, there is no area of the upper surface of any of the solar cells making up the second configuration

54 that is inactive as a result of being shaded from the sun by an upper bus bar connecting adjacent solar cells. It is estimated that the reduction in the cell spacing 52 and spacing 62 in the two configurations can result in an increased efficiency of about 4 percent.

It is further estimated the elimination of the top bus bars in the second configuration illustrated in FIGURES 7 and 8 can result in an increased efficiency of about 6 percent. Thus, the the overall improvement in efficiency, in watts per unit area of array, of the second configuration 54 illustrated in FIGURES 7 and 8 can be as much as 10 percent.

As will now be appreciated, the development of the complementary N-on-P solar cell having substantially the same electrical characteristics and resistance to degradation by radiation allows the construction of large solar cell arrays having increased electrical output per unit area of array, by eliminating the need for top -to-bottom electrical connections between adjacent solar cells of the array.

Assembly of the solar cell array is made easier through the use of modules, each incorporating two solar cells. The modules may be assembled separately, and then joined together using flat connector bars into the solar cell array. Although a particular embodiment of the invention has been described in detail for purposes of illustration, various modifications may be made without departing from the spirit and scope of the invention. Accordingly, the invention is not to be limited except as by the appended claims.

Claims

CLAIMSWhat is claimed is:

1. An N-on-P gallium arsenide solar cell, wherein the n-type gallium arsenide faces the sun, comprising: a single crystal gallium arsenide substrate; a single crystal layer of p+ gallium aluminum arsenide epitaxially overlying said substrate; a single crystal layer of p-type gallium arsenide epitaxially overlying said layer of p+ gallium aluminum arsenide; and a single crystal layer of n-type gallium arsenide epitaxially overlying said layer of p-type gallium arsenide.

2. The solar cell of claim 1, further including a transparent cover over said solar cell.

3. The solar cell of claim 1, wherein said layer of p-type gallium arsenide is about 0.5 micrometers thick and said layer of n-type gallium arsenide is about 10 micrometers thick.

4. A solar cell array, comprising: an N-on-P first solar cell; a P-on-N second solar cell laterally adjacent said first solar cell; and an electrical connector extending from a p-type layer of said first solar cell to an n-type layer of said second solar cell.

5. The solar cell array of claim 4, wherein said connector is a flat metallic strip.

6. The solar cell array of claim 4, wherein said solar cells are gallium arsenide solar cells.

7. The solar cell array of claim 4, further including a second electrical connector joined to an n-type layer of said first solar cell and a third electrical connector joined to a p-type layer of said second solar cell.

8. The solar cell array of claim 4, wherein said second connector and said third connector are flat metallic strips.

9. The solar cell array of claim 7, further including a P-on-N third solar cell having a p-type layer, joined to said n-type layer of said first solar cell by said second connector, and an N-on-P fourth solar cell having an n-type layer joined to said p-type layer of said second solar cell by said third connector.

10. The solar cell array of claim 4, wherein said connector is a grid extending between the layers of said array facing the sun.

11. The solar cell array of claim 4, wherein said first and second solar cells are laterally joined together.

12. A solar cell array, comprising: an N-on-P first solar cell; a P-on-N second solar cell; and an electrical connector extending from an n-type layer of said first solar cell to a p-type layer of said second solar cell.

13. The solar cell array of claim 12, wherein said connector is a flat metallic strip.

14. The solar cell array of claim 12, wherein said solar cells are gallium arsenide solar cells.

15. The solar cell array of claim 12, wherein said connector is a grid extending between the layers of said array facing the sun.

16. The solar cell array of claim 13, wherein said first and said second solar cells are laterally joined together.