WO1987001512A1

WO1987001512A1 - High efficiency photovoltaic assembly

Info

Publication number: WO1987001512A1
Application number: PCT/US1986/001841
Authority: WO
Inventors: M. Edmund Ellion
Original assignee: Hughes Aircraft Company
Priority date: 1985-09-09
Filing date: 1986-09-08
Publication date: 1987-03-12
Also published as: IL79995A0; EP0236495A1

Abstract

A high efficiency photovoltaic assembly, wherein a semiconductor solar cell array includes at least two semiconductor solar cells, each cell being characterized by a different incident light wavelength for optimal conversion efficiency of the cell, and means for directing an incident light beam toward the solar cells so that each cell receives the component of the incident light beam including its respective light wavelength for optimal conversion efficiency. The solar cell array may include a light wavelength analyzer such as a prism or diffraction grating for splitting the light beam into component rays of various wavelengths, which rays are then directed toward those solar cells which can optimally convert the respective wavelengths. The solar cell array could alternatively include a structure which directs the entire incident light beam serially through a succession of solar cells, wherein successive pairs of cells are arranged in a non-coplanar, spaced apart fashion.

Description

. . .

HIGH EFFICIENCY PHOTOVOLTAIC ASSEMBLY

BACKGROUND OF THE INVENTION

This invention relates to photovoltaic assemblies, and, more particularly, to solar cell arrays utilizing solar cells optimally sensitive to differing wavelengths of light.

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 con¬ tacts to the materials on either side of the interface. In general terms, a photovoltaic solar cell is fabricated by depositing the appropriate semiconductor layers onto a substrate, and then adding additional components to complete the cell. As an example, 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. The interface between the p-type gallium arsenide and the n-type gallium arsenide forms the basic solar cell active structure. External ohmic electrical contacts to the n-type and p-type layers are applied, and a voltage is measured across the contacts when light is directed against the interface. Optionally, a P+ layer of ...

gallium aluminum arsenide may be deposited over the layer of p-type gallium arsenide to limit recombination of charge carriers. To protect the solar cell from physical contact and radiation damage such as encount- ered in a space environment, it is conventional to apply a transparent cover of glass over the solar cell components.

All known types of solar cells are cha¬ racterized by an electrical current output which is ; dependent upon the wavelength of the light that is incident upon the solar cell, as may be determined in a laboratory experiment wherein the light wavelength is slowly varied and the output current is measured. This effect is thought to arise because of the quantum : nature of the conversion process wherein light photons of a particular wavelength or energy promote electron, transitions in the semiconductor materials used in forming the solar cell. That is, the number of excess charge carriers is dependent upon the wavelength or energy of the light photons, and the current produced by the solar cell in turn depends upon the number of excess charge carriers. As an example, the gallium arsenide solar cell discussed above has a band gap energy of about 1.4 eV. The conversion of light to > electrical current is optimized when the wavelength of the incident light is about 0.6 to about 0.9 micrometers. Light of lesser or greater wavelengths may have a minor effect on the production of electrical current, but for the most part is either reflected or transformed into unuseable heat energy.

Although it is possible to control the wavelength of the incident light in a laboratory environment, in actual commercial practice the wave¬ length of the light is determined by a light source which ordinarily is not controllable. Specifically, solar cell arrays are furnished incident sunlight, which is a white light having a broad range of constituent wavelengths from many parts of the visible and invisible spectrum. Accordingly, most of the light falling upon a solar cell is outside the range of optimum sensitivity and conversion efficiency of the solar cell, so that the conversion efficiencies of most solar cells with incident sunlight are relatively low, on the order of about 5% to about 15%.

There have been attempts to increase the con¬ version efficiency of solar cells by incorporating multiple junctions into the cells. Thus, for example, a gallium aluminum arsenide solar cell could be fabricated epitaxially over a gallium arsenide solar cell. The gallium aluminum arsenide solar cell exhibits a band gap of about 1.8 eV, so that shorter wavelengths of light would be optimally converted by the gallium aluminum arsenide solar cell (as compared with the wavelengths optimally converted by the gallium arsenide solar cell) . With such a multi- junction structure, conversion efficiencies of about 20% may be attained. It may be envisioned that even further solar cells could be epitaxially deposited into a single inte- grated structure, to obtain even higher efficiencies.

However, the fabrication of such composite semi¬ conductor solar cells is limited by fabrication difficulties such as obtaining epitaxial deposition of different semiconductor materials. Even if complex composite structures are fabricated, the individual junctions are inherently connected in series in such a structure, so that the electrical current output of the structure is limited to the lowest current produced by any of the junctions connected in series. The output current in excess of this minimum value, produced by other junctions in series, is dissipated as heat and cannot be used. There therefore exists a continuing need for an improved photovoltaic assembly wherein higher efficiencies of conversion of the incident light to electrical output are obtained. Such solar cell arrays 5 would desirably produce high electrical currents and would be fabricated and transported readily. It is also desirable that such arrays be capable of incorporation into compact structures which are not readily disabled by laser or other beam damage in a space environment. 10. The present invention fulfills this need, and further provides related advantages.

SUMMARY OF THE INVENTION

The present invention resides in a semi¬ conductor solar cell array which achieves enhanced

15 light output conversion efficiencies and high electrical current outputs, in a structure which may be readily fabricated and used in the operating environ¬ ment. The solar cell array allows the interconnection of the individual solar cells in a series fashion to ^•20 achieve the desired electrical voltages, and the inter¬ connection of the series-connected cells in a parallel fashion for high output currents. The solar cell arrays of the invention may be made in a compact form which is difficult to target and damage by iaser beam

25 attack. These and other advantages may be obtained utilizing single-junction technologies already well established or in advanced development, do not require the development of new ulti- junction technologies and fabrication procedures, and are not limited by the

30 series-connection configuration of multi- junction devices. In accordance with the invention,^' a semi¬ conductor solar cell array comprises at least two semiconductor solar cells, each solar cell having a different incident light wavelength of optimal con¬ s' version of light energy to electricity; and means for directing an incident light beam toward the solar cells, so that each cell receives the component of the incident light beam having its respective light wavelength of optimal conversion. In one approach, the 0 means for directing includes a light wavelength ana¬ lyzer for splitting the incident light beam into component rays of various wavelengths, and means for positioning each of the solar cells to receive the component ray containing its respective wavelengths of 5 optimal conversion. In another embodiment, the means for directing includes means for orienting the solar cells so that the incident light beam falls upon the solar cells serially, and further so that successive pairs of cells are not parallel to each other. Any Q desired number of different types of solar cells may be used together in such arrays, but as a practical matter, about four types of solar cells having different band gaps are usually enough to obtain an acceptably high conversion efficiency. The individual 5 solar cells may be electrically joined in any desired series or parallel arrangement to achieve particular output voltages and electrical currents.

As indicated, one embodiment provides a means for analyzing an incident light beam into component 0 rays of various wavelengths, and at least two semiconductor solar cells, each cell having a different incident light wavelength of optimum conversion of light energy to electricity, with each cell being positioned to receive from the light analyzer the 35 component ray containing its respective wavelength of optimal conversion of light to electricity. The means for analyzing is conveniently a prism or diffraction grating which splits a beam of light into its component wavelengths to form a spectrum, and then the individual solar cells are positioned to receive the portion of the spectrum that is optimally converted by the 5 respective solar cell.

In another embodiment, at least two semi¬ conductor solar cells having different incident light wavelengths of optimal conversion efficiency are supported in a mounting structure in a serially non- α coplanar arrangement, and there is provided means for directing an incident light beam serially from one of the solar cells to the next. Each solar cell then extracts energy from the incident light beam at its optimal conversion efficiency, passing onto the next , solar cell the portion of the incident light beam that is not converted to electricity. The individual solar cells may not be positioned in a layered, parallel arrangement, since internal reflection rapidly increases .the temperature of the solar cells to a point 0 where their conversion efficiency is drastically reduced. Successive pairs of solar cells may therefore not be parallel to each other, although, for example, a first cell and a third cell could be parallel to each other in a manner such that the light beam cannot be 5 reflected directly between the two. Preferably, the incident light beam may be directed serially from one solar cell to the next with mirrors, preferably in the form of silvered back surfaces of the solar cells themselves. A particularly desirable angular 0 orientation between the successive semiconductor solar cells is 45 *, inasmuch as geometrically regular mounting structures can be fabricated in such orientations. One such mounting structure has a hollow triangular elevational crosssection, with at least one 5- solar cell on each side thereof. Another such solar cell array has a hollow parallelogram elevational cross-section. AU types of single-junction semiconductor solar cells may be utilized in conjunction with the present invention. Some such solar cells, such as silicon and gallium arsenide, are already well established and in commercial use. Other types of single junction solar cells are known, but development work remains before they are commercially practical. The present invention allows the use of the solar cells already developed without substantial modification, and without the development of multi-junction structures in fabrication techniques. As other single-junction solar cells become commercially practical, these may be utilized in conjunction with the present invention also. With the arrays of the present invention, substantially improved photovoltaic assembly and solar cell array operating characteristics can be achieved. Moreover, the arrays of the present invention may be made in a compact form, and utilized within surrounding support structure, which improves the survival characteristics of the solar cell arrays in a normal space environment and when subjected to attack in a space environment. 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 DES CRIPTION OF THE DRAWINGS

FIGURE 1 is an elevational view of a con¬ ventional single-junction gallium arsenide solar cell;

FIGURE 2 is an elevational view of a dual- 5 junction solar cell;

FIGURE 3 is an elevational view of a pair of single-junction solar cells arranged one above the other in a coplanar fashion;

FIGURE 4 is an elevational view of a solar I0> cell array wherein the incident light is split into its component rays and the component rays are directed to a number of solar cells;

FIGURE 5 is an end elevational view of a solar cell triangular mounting structure and the solar cells 15 mounted thereupon;

FIGURE 6 is an end elevational view of a solar cell parallelogram mounting structure and the solar cells mounted thereupon; and

FIGURE 7 is a schematic sectional view of a 20 solar cell array of the present invention mounted in conjunction with a parabolic reflector and lens to form a photovoltaic power supply. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

FIGURE 1 illustrates a conventional single- junction solar cell, here depicted as a gallium arse¬ nide solar cell 10 for purposes of illustration. The solar cell 10 comprises a single crystal gallium arsenide substrate 12, upon which the active elements of the solar cell are fabricated. Epitaxially overlying the substrate 12 is a single crystal layer of n-type gallium arsenide 14. Epitaxially overlying the ; single crystal layer 14 of n-type gallium arsenide is a layer of p-type gallium arsenide 16. The layers 14 and 16 together comprise an active gallium arsenide solar cell 18, with the interface between the layers 14 and 16 being the single- junction solar cell 18. A glass 5 window 20 is typically attached over the solar cell 18 to protect it from radiation and to support the solar cell 18. Other semiconductor layers may optionally be added, such as a P+ type gallium aluminum arsenide layer epitaxially overlying the p-type gallium arsenide 0 layer 16, as for the purpose of inhibiting surface charge recombination.

Such a solar cell 18 has a band gap of about 1.4 electron volts, and is most sensitive to incident light radiation of about 0.6 to about 0.9 micrometers. 5 That is, light having such wavelengths is converted to electricity by the single-junction solar cell 10 with optimal efficiency. Light having other wavelengths may be converted to electrical energy, but at significantly reduced efficiencies. Although the gallium arsenide O solar cell 10 has been described in detail for the purpose of illustration, other solar cells based on other technologies such as silicon or cadmium telluride are similarly limited to particular incident light wavelengths of optimal conversion efficiency, although 5 the optimal wavelength ranges for other solar cells will be different from that of a gallium arsenide solar cell because of the differing band gaps.

An approach to achieving a solar cell having sensitivity to a broader range of incident wavelengths is illustrated as a dual-junction semiconductor solar cell 22, in FIGURE 2. A single crystal gallium arsenide substrate 24 is first prepared. A single crystal layer 26 of n-type gallium arsenide epitaxially overlies the substrate, and a single crystal layer 28 of p.-type gallium arsenide epitaxially overlies the layer^* 26 of n-type gallium arsenide. The layers 26 and 28" together form a gallium arsenide single- junction solar, cell. 30. Next, a single crystal layer 32 of n-type- gallium aluminum arsenide epitaxially overlies the layer 2S of p-type gallium arsenide, and a single crystal^', layer 34 of p-type gallium aluminum arsenide epitaxially overlies the layer 32 of n-type gallium aluminum arsenide. The layers 32 and 34 together comprise a gallium aluminum arsenide single-junction solar cell 36. A glass window 38 is then attached over the dual- junction solar cell 22.

The dual-junction solar cell 22 thus comprises two solar cells 30 and 36 arrayed one on top of the other in a series arrangement. The interface between the layers 26 and 28 creates a first voltage, and the interface between the layers 32 and 34 creates a second voltage, with the total voltage produced by the dual- junction solar cell 22 being the sum of the voltages of the individual solar cells 30 and 36. The current which flows through the dual- junction solar cell 22 is limited to the maximum current produced by either the gallium arsenide solar cell 30, or the gallium aluminum arsenide solar cell 36. The band gap of the gallium arsenide solar cell 30 is about 1.4 eV, so that, as previously indicated, this solar cell junction is optimally sensitive to light having wavelengths of about 0.6 to about 0.9 micrometers. The gallium aluminum arsenide solar cell 36 has a band gap of about 1.8 eV, and is therefore optimally sensitive to light having wavelengths of from about 0.3 to about 0.6 micrometers. The two solar cells 30 and 36 together are sensitive to light having wavelengths of from about 0.3 to about 0.9 micrometers. However, as noted, the maximum current produced by the dual-junction solar cell 22 is limited to the lesser of the current flows produced by the gallium arsenide solar cell 30 and the gallium aluminum arsenide solar cell 36. Performance limitations with the multi- junction solar cell 22 arise because the individual single junction solar cells 30 and 36 which comprise the multi- junction solar cell 22 are necessarily linked in a series fashion, and cannot be individually connected by external connections. The problem of the multi- junction solar cell

22, arising because the individual component solar cells are necessarily linked in series, can be overcome * by physically separating t e component solar cells. FIGURE 3 illustrates a pair of singlejuncti n solar cells 40 and 42, each of the same general type as illustrated in FIGURE 1 but constructed of different materials of different optimal conversion efficiencies, stacked one above the other to form a stacked solar cell 44 having a gap 50 therebetween. The individual single- junction solar cells 40 and 42 may be furnished with external connections independent of each other, thereby avoiding the internal series linking of solar cells found in the dual- junction solar cell 22. A beam of light incident upon the stacked solar cell 44 is indicated by the numeral 46. As the beam of light 46 passes through the first solar cell 40, the first solar cell responds most efficiently to a particular wavelength of light. The remainder of the beam of light 46 passes from the first solar cell 40 and into the second solar cell 42. However, a portion of a transmitted beam 48 is reflected from the top surface of the second solar cell 42. A portion of the reflected beam travels back into the first solar cell 40, and a portion is reflected from the bottom surface of the first solar cell 40. Multiple reflections in the gap 50 can then occur, resulting in heating of the solar cells 40 and 42. Under particularly demanding operating conditions, the heat build-up in the solar cells 40 and 42 due to the beams reflected in the gap 50 may heat the solar cells 40 and 42 above their optimum operating temperatures, resulting in a loss of efficiency of the conversion of light to electrical current. Thus, the simple stacked solar cell 44 does not present a viable solution to the problem of increasing the efficiency of solar cell arrays through increased sensitivity to a range of wavelengths. In accordance with one preferred embodiment of the present invention, FIGURE 4 illustrates a semiconductor solar cell array 52 having an analyzer, here illustrated as a prism 54, for splitting an incident beam of light 56 into a spectrum of rays 58 of varying wavelengths. Three solar cells 60, 62 and 64 are placed into the path of the rays 58 so as to intercept and receive light waves of different wavelengths.

The solar cells 60, 62 and 64 are chosen and placed so as to be optimally sensitive to the component rays intercepted. The following table illustrates, by way of example, a number of candidate solar cells, their band gaps, the corresponding light wavelengths, and the expected approximate range of optimal sensi- tivities and efficiency in converting light energy to electrical energy: TABLE I

Solar Cell Band Corresponding Expected Incident Material Gap Wavelength Light Range of Op- (eV) (micrometers) timal Sensitivity . micrometers.

Ge 0.75 1.653 1.3-1.6

Si 1.10 1.27 0.8-1.1

InP 1.23 1.008 0.7-1.0

GaAs 1.32 0.939 0.6-0.9

CdTe 1.42 0.873 0.5-0.8

AlSb 1.57 0.789 0.4-0.7

CdS 1.71 0.725 0.4-0.7

Other suitable semiconductor solar cells may also be used in conjunction with the invention, including those now known and those which may be later developed. Since the prism 54 produces rays 58 which . are bent least _. for the longer wavelengths, the semiconductor solar cell 60 might be a long wavelength cell such as those fabricated from doped germanium or silicon. The solar cells 62 might be of intermediate wavelength, such as a solar cell based upon doped indium phosphide or gallium arsenide. The solar cell 64 might be selected from those solar cells having greater sensitivity to short wavelengths, such as aluminum antimonide or cadmium sulfide.

The solar cells 60, 62, and 64 produce electrical outputs independently of any of the other cells in the array 52. Each of the cells may be joined in series with other cells of the same or different types to produce greater voltages, or in parallel with solar cells of the same or different types to produce greater currents. It is not necessary that the solar cells 60, 62 or 64 be compatible in the sense that materials of one be capable of single crystal epitaxial deposition on the materials of the other. The cells 60, 62 and 64 are fabricated separately, using the appropriate procedures.

Two other preferred embodiments of the present invention are illustrated in FIGURES 5 and 6. In these " embodiments of the invention, the incident light is not analyzed into component rays of different wavelengths, but instead is serially directed through a sequence of solar cells having optimal sensisvity and efficiency at various incident wavelengths, so that electrical energy may be converted from the full spectrum of wavelengths making up the incident light beam. The mounting structure supports the solar cells in a non-coplanar arrangement, and means is provided for directing an incident light beam from one solar cell to the next. More specifically, FIGURE 5 illustrates a triangular mounting structure 66 having a hollow tri¬ angular elevational cross-section. A first solar cell 68 is mounted on one side of the triangular support structure 66, so as to be intercepted by an incident Q light beam 70. A second solar cell 72 is mounted on an adjacent side of the triangular support structure 68, so that the portion of the incident light beam 70 which passes through the first solar cell 68 is directed against the second solar cell 72. The second solar 5 cell 72 is serially non-coplanar with the first solar cell 68. As used herein, the term "serially non- coplanar" is used to refer to two solar cells inter¬ cepted by a light beam without passing through any intermediate solar cell, the two solar cells having Q. their active interfaces lying at an angle to each other, and not parallel or in the same plane. In the embodiment illustrated in FIGURE 5, the second solar cell 72, which directly receives the transmitted portion of the incident beam 70 after it passes through 5 the first solar cell 68, lies at an angle of about

45 * with respect to the first solar cell 68. By constructing the array so that successive pairs of solar cells are non-coplanar, i.e., serially non- coplanar, internal reflection, such as described above in relation to the gap 50 and the stacked array 44, is avoided. Means is provided for directing the light beam serially from one of the solar cells to the next solar cell. The transmitted portion of the incident light beam 70 passes directly through the first solar cell 68 to the second solar cell 72. The second solar cell 72 preferably includes a silvered back surface 74, which acts as a mirror to reflect the light beam at an angle equal to its incident angle, thereby forming a first reflected light beam 76.

A third solar cell 78 is mounted on the third ' side of the triangular support structure 66, and positioned to intercept the first reflected light beam 76. The third solar cell 78 is also preferably provided with a silvered back surface 80, which reflects the portion of the first reflected light beam 76 not converted to electricity in the third solar cell

78, to form a second reflected light beam 82. Optionally, a fourth solar cell 84 may be mounted on the same side of the triangular support structure 66 as the first solar cell 68, and positioned so that the second reflected light beam 82 passes through the fourth solar cell 84 as it leaves the triangular support structure 66. In this way, a fourth wavelength range may be converted to electrical energy.

The triangular support structure 66 thereby provides a support for four solar cells 68, 72, 78 and

84. These four solar cells are selected as being optimally efficient and responsive to different wavelengths of light, so that each of the four solar cells converts a portion of the total wavelength of the incident light beam 70 to electrical energy. The solar cell array illustrated in FIGURE 5 achieves an efficiency of up to about 50% in converting the light energy of the incident beam 70 to electrical energy, through the use of the four solar cells. The four solar cells are provided with external connections allowing them to be connected with each other or with 5 cells in other arrays, in any selected series or parallel fashion to obtain a desired electrical voltage and current. The means for directing the incident light beam illustrated in FIGURE 5 is the silvered back surfaces 74 and 80. Mirrors, lenses, light pipes, or 0 ; other means for directing the light beams may also be provided.

Another embodiment of the present invention is illustrated in FIGURE 6, wherein a parallelogram-shaped support structure 86 has an elevational cross- section

15 : i . the shape of a hollow parallelogram, with at least one solar cell on each side of the parallelogram. An incident light beam 88 passes through a first solar cell 90 supported by one side of the parallelogram- shaped support structure 86. The first solar cell 90-

20 converts electrical energy from a portion of the wavelengths contained within the incident light beam 88, and transmits the remaining wavelengths out of the first solar cell 90 in a transmitted beam 92. The transmitted beam 92 falls upon a second solar cell 94

25. supported by a second side of the parallelogram-shaped support structure 86. A second portion of the wavelengths contained in the incident light beam 88 is converted to electrical energy by the second solar cell 94.

30 The second solar cell 94 includes a silvered back surface 96, which reflects the unconverted portion of the transmitted beam 92 out of the second solar cell 94, thereby forming a first reflected beam 98. The first reflected beam 98 impinges upon a third solar

35 cell 100, which is supported on a third side of the parallelogram-shaped support structure 86. The energy from a third range of wavelengths of the incident light beam 88 is converted to electrical energy by the third solar cell 100. The third solar cell 100 is provided with a silvered-back surface 102, and the first reflected beam 98 is reflected from the silvered-back surface 102 to form a second reflected beam 104.

The second reflected beam 104 impinges upon a fourth solar cell 106, which is supported by a fourth side of the parallelogram-shaped support structure 86. The fourth solar cell 106 converts the energy in an fourth range of wavelengths of the incident beam 88 to electrical energy.

In the embodiment illustrated in FIGURE 6, the fourth solar cell 106 is provided with a silvered-back surface 108, which reflects the second reflected beam 104 normal to the surface of the fourth solar cell 106, so that the beam traverses back along the path of the second reflected beam 104, the first reflected beam 98, the transmitted beam 92, and the incident beam 88. Thus, the light beam impinges twice upon the solar cells 100, 94 and 90 as it passes through the solar cell array.

It is emphasized that the first solar cell 90 and the second solar cell 92 are serially non-coplanar, the second solar cell 94 and the third solar cell 100 are serially non-coplanar, and the third solar cell 100 and the fourth solar cell 106 are serially non-coplanar.

The triangular support structure 66 and the parralelogram-shaped support structure 86 have the important advantage that light beams cannot be trapped by multiple reflections between adjacent solar cells, to produce high heat loadings on the cells. Addit¬ ionally, the hollow core structure of the support structures 66 and 86 allows heat to be radiated away from the solar cells, or in an atmospheric environment, a coolant to be passed down the center of the support structure. It is particularly desirable that the solar cell arrays of the present invention be resistant to, and protect able from, damage caused by a laser or other energy beam directed against the solar cell array. FIGURE 7 illustrates a preferred photovoltaic assembly that exhibits such a damage resistance. A photovoltaic energy source 110 includes a solar cell array 112 in accordance with the present invention, and means for focusing a beam of incident light 114 upon the solar ; cells of the array 112. A parabolic reflector 116 is aimed at the incident light beam 114, so that components of the beam are reflected toward the focus of the parabola. A lens 118 at the focus directs the focused light beams into a single parallel beam 120, ^" which is then directed into the solar cell array 112, an can be converted to electrical energy in the manner previously described. The solar cell array 112 can be made compact due to its high conversion efficiency. The array 112 is placed in a protective covering 122 which protects the solar cell array 112 from the general radiation environment and also from damage induced by high energy beams directed against the solar- cell array 112. The compact solar cell array 112 is therefore more difficult to target and also more _. defensible than prior solar cell arrays having much greater size.

The solar cell arrays of the present invention are therefore more efficient than conventional solar cell arrays, and can be made more compact. The arrays can be constructed from known technologies, and do not require development of new multi- junction semiconductor solar cells. 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. A semiconductor solar cell array, comprising: at least two semiconductor solar cells, each solar cell having a different incident light wavelength of optimal conversion of light energy to electricity; and means for directing an incident light beam toward said solar cells, so that each cell receives the component of the incident light beam having its respective light wavelength of optimal conversion.

2. The array of claim 1, wherein said means for directing includes a light wavelength analyzer for splitting the incident light beam into component rays of various ! wavelengths; and means for positioning each of said solar cells to receive the component ray containing its respective wavelength of optimal conversion.

3. The array of claim 1, wherein said means for directing includes means for orienting said solar cells so that the incident light beam falls upon said solar cells serially, and further so that successive pairs of

5 said solar cells are not parallel to each other.

4. A semiconductor solar cell array, comprising: means for analyzing an incident light beam into component rays of various wavelengths; at least two semiconductor solar cells, each

5 solar cell having a different incident light wavelength of optimal conversion of light energy to electricity. __2Q_

and each cell being positioned to receive, from said means for analyzing, the component ray containing its respective wavelength of optimal conversion of light to electricity.

5. The array of claim 4, wherein said means for analyzing is a prism.

6. The array of claim 4, wherein said _. means for analyzing is a diffraction grating.

7. The array of claim 4, wherein one of said solar cells is a silicon solar cell.

8. The array of claim 4, wherein one of said solar cells is a gallium arsenide solar cell.

9. The array of cla_^im 4, further including: means for focusing a beam of incident light upon said solar cells.

10. A semiconductor solar cell array, comprising: at least two semiconductor solar cells, each cell having a different incident light wavelength of optimal conversion of light energy to electricity; - a mounting structure for mounting said solar cells in a serially non-coplanar arrangement; and means for directing an incident light beam serially from one of said solar cells to the next.

11. The array of claim 10, wherein said means for directing includes a mirror.

12. The array of claim 11, wherein said mirror is provided as a mirrored back surface on one of said solar cells.

13. The array of claim 10, wherein said serially non-coplanar solar cells have an angular orientation to each other of about 45'.

14. The array of claim 10, wherein one of said solar cells is a silicon solar cell.

15. The array of claim 10, wherein one of said solar cells is a gallium arsenide solar cell.

16. The array of claim 10, wherein said mounting structure has a hollow triangular elevational cross section, with at least one solar cell on each side thereof.

17. The array of claim 10, wherein said mounting structure has a hollow parallelogram elevational cross section, with at least one solar cell on each side thereof.

18. The array of claim 10, further including; means for focusing a beam of incident light upon said solar cells.