Four Junction Inverted Metamorphic Multijunction Solar Cell with Two Metamorphic Layers

Abstract

A multijunction solar cell including an upper first solar subcell having a first band gap; a second solar subcell adjacent to the first solar subcell and having a second band gap smaller than the first band gap; a first graded interlayer adjacent to the second solar subcell; the first graded interlayer having a third band gap greater than the second band gap; and a third solar subcell adjacent to the first graded interlayer, the third subcell having a fourth band gap smaller than the second band gap such that the third subcell is lattice mismatched with respect to the second subcell. A second graded interlayer is provided adjacent to the third solar subcell; the second graded interlayer having a fifth band gap greater than the fourth band gap; and a lower fourth solar subcell is provided adjacent to the second graded interlayer, the lower fourth subcell having a sixth band gap smaller than the fourth band gap such that the fourth subcell is lattice mismatched with respect to the third subcell.

Claims

1.-20. (canceled)

21. A multijunction solar cell comprising: one or more upper solar subcells; a first upper barrier layer below the one or more upper solar subcells, the first upper barrier layer being composed of InGa(Al)P; a first graded interlayer composed of InGaAlAs, the first graded interlayer being below and adjacent the first upper barrier layer and having a band gap that is constant throughout the thickness of the first graded interlayer; a first lower barrier layer below and adjacent the first graded interlayer, the first lower barrier layer being composed of InGaP and having a composition different from that of the first upper barrier layer; at least one middle solar subcell below the first lower barrier layer; a second graded interlayer composed of InGaAlAs and disposed below the at least one middle solar subcell and having a band gap that is constant throughout the thickness of the second graded interlayer; and at least one lower solar subcell below the second graded interlayer.

22. The multijunction solar cell of claim 21 wherein the first upper barrier layer and the first lower barrier layer help prevent threading dislocations from propagating, respectively, into adjacent ones of the solar subcells.

23. The multijunction solar cell of claim 21 wherein the band gap of the first graded interlayer is 1.5 eV.

24. The multijunction solar cell of claim 23 wherein the band gap of the second graded interlayer is 1.1 eV.

25. The multijunction solar cell of claim 24 wherein: the one or more upper solar subcells include a first solar subcell having a first band gap in the range of 1.8 to 2.1 eV, and a second solar subcell below the first solar subcell and having a second band gap in the range of 1.35 to 1.45 eV; the at least one middle solar subcell includes a third solar subcell having a band gap in the range of 0.9 to 1.1 eV; and the at least one lower solar subcell includes a fourth solar subcell having a band gap in the range of 0.6 to 0.8 eV.

26. The multijunction solar cell of claim 21 wherein: the one or more upper solar subcells include a first solar subcell having a first band gap, and a second solar subcell disposed below the first solar subcell and having a second band gap smaller than the first band gap; the first graded interlayer has a third band gap greater than the second band gap; the at least one middle solar subcell includes a third solar subcell having a fourth band gap smaller than the second band gap such that the third subcell is lattice mismatched with respect to the second subcell, and wherein the first graded interlayer provides a gradual transition in lattice constant from the second solar subcell to the third solar subcell; the second graded interlayer has a fifth band gap greater than the fourth band gap; and the at least one lower solar subcell includes a fourth solar subcell having a sixth band gap smaller than the fourth band gap such that the fourth solar subcell is lattice mismatched with respect to the third solar subcell, and wherein the second graded interlayer provides a gradual transition in lattice constant from the third solar subcell to the fourth solar subcell.

27. The multijunction solar cell of claim 21 wherein the first upper barrier layer is n-type, and the first lower barrier layer is n+ type.

28. The multijunction solar cell of claim 21 wherein the first upper barrier layer is composed of InGaP.

29. The multijunction solar cell of claim 21 further including a second upper barrier layer disposed below the at least one middle solar subcell, and disposed above and adjacent to the second graded interlayer.

30. The multijunction solar cell of claim 29 wherein the second upper barrier layer is composed of GaInP.

31. A method of manufacturing a solar cell comprising: providing a first substrate; depositing over the first substrate a first sequence of layers of semiconductor material forming one or more first solar subcells; depositing over the first sequence of layers a first upper barrier layer composed of InGa(Al)P; depositing over the first upper barrier layer a first graded interlayer composed of InGaAlAs, the first graded interlayer being adjacent the first upper barrier layer and having a band gap that is constant throughout the thickness of the first graded interlayer; depositing over the first graded interlayer a first lower barrier layer such that the first lower barrier layer is adjacent the first graded interlayer, the first lower barrier layer being composed of InGaP and having a composition different from that of the first upper barrier layer; depositing over the first lower barrier layer at least one second solar subcell; depositing over the at least one second solar subcell a second graded interlayer composed of InGaAlAs and having a band gap that is constant throughout the thickness of the second graded interlayer; and depositing over the second graded interlayer at least one third solar subcell.

32. The method of claim 31 wherein the first upper barrier layer and the first lower barrier layer help prevent threading dislocations from propagating, respectively, into adjacent ones of the solar subcells.

33. The method of claim 31 wherein depositing a first graded interlayer includes depositing a graded interlayer having a band gap of 1.5 eV.

34. The method of claim 33 wherein depositing a second graded interlayer included depositing a graded interlayer having a band gap of 1.1 eV.

35. The method of claim 34 wherein: depositing the one or more first solar subcells includes: depositing a first solar subcell having a first band gap in the range of 1.8 to 2.1 eV; and depositing over the first solar subcell another solar subcell having a second band gap in the range of 1.35 to 1.45 eV; depositing the at least one second solar subcell includes depositing a solar subcell having a band gap in the range of 0.9 to 1.1 eV; and depositing the at least one third solar subcell includes depositing a solar subcell having a band gap in the range of 0.6 to 0.8 eV.

36. The method of claim 31 wherein: depositing the one or more first solar subcells includes: depositing a first solar subcell having a first band gap; and depositing over the first solar subcell another solar subcell having a second band gap smaller than the first band gap; depositing the first graded interlayer includes depositing a graded interlayer having a third band gap greater than the second band gap; depositing the at least one second solar subcell includes depositing a solar subcell having a fourth band gap smaller than the second band gap, wherein the solar subcell having the fourth band gap is lattice mismatched with respect to the solar subcell having the second band gap, and wherein the first graded interlayer provides a gradual transition in lattice constant from the solar subcell having the second band gap to the solar subcell having the fourth band gap; depositing a second graded interlayer includes depositing a graded interlayer having a fifth band gap greater than the fourth band gap; and depositing the at least one third solar subcell includes depositing a solar subcell having a sixth band gap smaller than the fourth band gap such that the solar subcell having the sixth band gap is lattice mismatched with respect to the solar subcell having the fourth band gap, and wherein the second graded interlayer provides a gradual transition in lattice constant from the solar subcell having the fourth band gap to the solar subcell having the sixth band gap.

37. The method of claim 31 wherein depositing a first upper barrier layer includes depositing a barrier layer composed of an n-type layer, and wherein depositing a first lower barrier layer includes depositing a barrier layer composed of an n+ type layer.

38. The method of claim 31 wherein depositing a first upper barrier layer includes depositing a barrier layer composed of InGaP.

39. The method of claim 31 further including depositing a second upper barrier layer over the at least one second solar subcell prior to depositing the second graded interlayer.

40. The method of claim 39 wherein depositing a second upper barrier layer includes depositing a barrier layer composed of GaInP.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The invention will be better and more fully appreciated by reference to the following detailed description when considered in conjunction with the accompanying drawings, wherein:

[0030] FIG. 1 is a graph representing the bandgap of certain binary materials and their lattice constants;

[0031] FIG. 2 is a cross-sectional view of the solar cell of the present invention after an initial stage of fabrication including the deposition of certain semiconductor layers on the growth substrate;

[0032] FIG. 3 is a cross-sectional view of the solar cell of FIG. 2 after the next sequence of process steps;

[0033] FIG. 4 is a cross-sectional view of the solar cell of FIG. 3 after the next sequence of process steps;

[0034] FIG. 5 is a cross-sectional view of the solar cell of FIG. 4 after the next sequence of process steps;

[0035] FIG. 6 is a cross-sectional view of the solar cell of FIG. 5 after the next process step;

[0036] FIG. 7 is a cross-sectional view of the solar cell of FIG. 6 after the next process step in which a surrogate substrate is attached;

[0037] FIG. 8A is a cross-sectional view of the solar cell of FIG. 7 after the next process step in which the original substrate is removed;

[0038] FIG. 8B is another cross-sectional view of the solar cell of FIG. 8A with the surrogate substrate on the bottom of the Figure;

[0039] FIG. 9 is a simplified cross-sectional view of the solar cell of FIG. 8B after the next process step;

[0040] FIG. 10 is a cross-sectional view of the solar cell of FIG. 9 after the next process step;

[0041] FIG. 11 is a cross-sectional view of the solar cell of FIG. 10 after the next process step;

[0042] FIG. 12 is a cross-sectional view of the solar cell of FIG. 11 after the next process step;

[0043] FIG. 13A is a top plan view of a wafer in which the solar cells are fabricated;

[0044] FIG. 13B is a bottom plan view of a wafer in which the solar cells are fabricated;

[0045] FIG. 14 is a cross-sectional view of the solar cell of FIG. 12 after the next process step;

[0046] FIG. 15 is a cross-sectional view of the solar cell of FIG. 14 after the next process step;

[0047] FIG. 16 is a cross-sectional view of the solar cell of FIG. 15 after the next process step

[0048] FIG. 17 is a top plan view of the wafer of FIG. 16 depicting the surface view of the trench etched around the cell;

[0049] FIG. 18A is a cross-sectional view of the solar cell of FIG. 16 after the next process step in a first embodiment of the present invention;

[0050] FIG. 18B is a cross-sectional view of the solar cell of FIG. 16 after the next process step in a second embodiment of the present invention;

[0051] FIG. 19 is a cross-sectional view of the solar cell of FIG. 18 after the next process step;

[0052] FIG. 20 is a graph of the doping profile in a base and emitter layers of a subcell in the metamorphic solar cell according to the present invention;

[0053] FIG. 21 is a graph that depicts the current and voltage characteristics of an inverted metamorphic multijunction solar cell according to the present invention;

[0054] FIG. 22 is a diagram representing the range of band gaps of various GaInAlAs materials as a function of the relative concentration of Al, In, and Ga;

[0055] FIG. 23 is a graph representing the Ga mole fraction versus the Al to In mole fraction in GaInAlAs materials that is necessary to achieve a constant 1.5 eV band gap;

[0056] FIG. 24 is a graph representing the mole fraction versus lattice contstant in GaInAlAs materials that is necessary to achieve a constant 1.5 eV band gap band gap;

DESCRIPTION OF THE PREFERRED EMBODIMENT

[0057] Details of the present invention will now be described including exemplary aspects and embodiments thereof. Referring to the drawings and the following description, like reference numbers are used to identify like or functionally similar elements, and are intended to illustrate major features of exemplary embodiments in a highly simplified diagrammatic manner. Moreover, the drawings are not intended to depict every feature of the actual embodiment nor the relative dimensions of the depicted elements, and are not drawn to scale.

[0058] The basic concept of fabricating an inverted metamorphic multijunction (IMM) solar cell is to grow the subcells of the solar cell on a substrate in a reverse sequence. That is, the high band gap subcells (i.e. subcells with band gaps in the range of 1.8 to 2.1 eV), which would normally be the top subcells facing the solar radiation, are grown epitaxially on a semiconductor growth substrate, such as for example GaAs or Ge, and such subcells are therefore lattice-matched to such substrate. One or more lower band gap middle subcells (i.e. with band gaps in the range of 1.2 to 1.8 eV) can then be grown on the high band gap subcells.

[0059] At least one lower subcell is formed over the middle subcell such that the at least one lower subcell is substantially lattice-mismatched with respect to the growth substrate and such that the at least one lower subcell has a third lower band gap (i.e. a band gap in the range of 0.7 to 1.2 eV). A surrogate substrate or support structure is then attached or provided over the bottom or substantially lattice-mismatched lower subcell, and the growth semiconductor substrate is subsequently removed. (The growth substrate may then subsequently be re-used for the growth of a second and subsequent solar cells).

[0060] A variety of different features of inverted metamorphic multijunction solar cells are disclosed in the related applications noted above. Some or all of such features may be included in the structures and processes associated with the solar cells of the present invention. However, more particularly, the present invention is directed to the fabrication of a four junction inverted metamorphic solar cell using two different metamorphic layers, all grown on a single growth substrate. In the present invention, the resulting construction includes four subcells, with band gaps in the range of 1.8 to 2.1 eV, 1.3 to 1.5 eV, 0.9 to 1.1 eV, and 0.6 to 0.8 eV respectively.

[0061] FIG. 1 is a graph representing the band gap of certain binary materials and their lattice constants. The band gap and lattice constants of ternary materials are located on the lines drawn between typical associated binary materials (such as the ternary material GaAlAs being located between the GaAs and AlAs points on the graph, with the band gap of the ternary material lying between 1.42 eV for GaAs and 2.16 eV for AlAs depending upon the relative amount of the individual constituents). Thus, depending upon the desired band gap, the material constituents of ternary materials can be appropriately selected for growth.

[0062] The lattice constants and electrical properties of the layers in the semiconductor structure are preferably controlled by specification of appropriate reactor growth temperatures and times, and by use of appropriate chemical composition and dopants. The use of a vapor deposition method, such as Organo Metallic Vapor Phase Epitaxy (OMVPE), Metal Organic Chemical Vapor Deposition (MOCVD), Molecular Beam Epitaxy (MBE), or other vapor deposition methods for the reverse growth may enable the layers in the monolithic semiconductor structure forming the cell to be grown with the required thickness, elemental composition, dopant concentration and grading and conductivity type.

[0063] FIG. 2 depicts the multijunction solar cell according to the present invention after the sequential formation of the three subcells A, B and C on a GaAs growth substrate. More particularly, there is shown a substrate 101, which is preferably gallium arsenide (GaAs), but may also be germanium (Ge) or other suitable material. For GaAs, the substrate is preferably a 15 off-cut substrate, that is to say, its surface is orientated 15 off the (100) plane towards the (111)A plane, as more fully described in U.S. patent application Ser. No. 12/047,944, filed Mar. 13, 2008.

[0064] In the case of a Ge substrate, a nucleation layer (not shown) is deposited directly on the substrate 101. On the substrate, or over the nucleation layer (in the case of a Ge substrate), a buffer layer 102 and an etch stop layer 103 are further deposited. In the case of GaAs substrate, the buffer layer 102 is preferably GaAs. In the case of Ge substrate, the buffer layer 102 is preferably InGaAs. A contact layer 104 of GaAs is then deposited on layer 103, and a window layer 105 of AlInP is deposited on the contact layer. The subcell A, consisting of an n+ emitter layer 106 and a p-type base layer 107, is then epitaxially deposited on the window layer 105. The subcell A is generally latticed matched to the growth substrate 101.

[0065] It should be noted that the multijunction solar cell structure could be formed by any suitable combination of group III to V elements listed in the periodic table subject to lattice constant and bandgap requirements, wherein the group III includes boron (B), aluminum (Al), gallium (Ga), indium (In), and thallium (T). The group IV includes carbon (C), silicon (Si), germanium (Ge), and tin (Sn). The group V includes nitrogen (N), phosphorous (P), arsenic (As), antimony (Sb), and bismuth (Bi).

[0066] In the preferred embodiment, the emitter layer 106 is composed of InGa(Al)P and the base layer 107 is composed of InGa(Al)P. The aluminum or Al term in parenthesis in the preceding formula means that Al is an optional constituent, and in this instance may be used in an amount ranging from 0% to 30%. The doping profile of the emitter and base layers 106 and 107 according to the present invention will be discussed in conjunction with FIG. 20.

[0067] Subcell A will ultimately become the top subcell of the inverted metamorphic structure after completion of the process steps according to the present invention to be described hereinafter.

[0068] On top of the base layer 107 a back surface field (BSF) layer 108 preferably p+ AlGaInP is deposited and used to reduce recombination loss.

[0069] The BSF layer 108 drives minority carriers from the region near the base/BSF interface surface to minimize the effect of recombination loss. In other words, a BSF layer 18 reduces recombination loss at the backside of the solar subcell A and thereby reduces the recombination in the base.

[0070] On top of the BSF layer 108 is deposited a sequence of heavily doped p-type and n-type layers 109a and 109b that forms a tunnel diode, i.e. an ohmic circuit element that connects subcell A to subcell B. Layer 109a is preferably composed of p++ AlGaAs, and layer 109b is preferably composed of n++ InGaP.

[0071] On top of the tunnel diode layers 109 a window layer 110 is deposited, preferably n+ InGaP. The advantage of utilizing InGaP as the material constituent of the window layer 110 is that it has an index of refraction that closely matches the adjacent emitter layer 111, as more fully described in U.S. patent application Ser. No. 12/258,190, filed Oct. 24, 2008. The window layer 110 used in the subcell B also operates to reduce the interface recombination loss. It should be apparent to one skilled in the art, that additional layer(s) may be added or deleted in the cell structure without departing from the scope of the present invention.

[0072] On top of the window layer 110 the layers of subcell B are deposited: the n-type emitter layer 111 and the p-type base layer 112. These layers are preferably composed of InGaP and In.sub.0.015GaAs respectively (for a Ge substrate or growth template), or InGaP and GaAs respectively (for a GaAs substrate), although any other suitable materials consistent with lattice constant and bandgap requirements may be used as well. Thus, subcell B may be composed of a GaAs, GaInP, GaInAs, GaAsSb, or GaInAsN emitter region and a GaAs, GaInAs, GaAsSb, or GaInAsN base region. The doping profile of layers 111 and 112 according to the present invention will be discussed in conjunction with FIG. 20.

[0073] In previously disclosed implementations of an inverted metamorphic solar cell, the middle cell was a homostructure. In the present invention, similarly to the structure disclosed in U.S. patent application Ser. No. 12/023,772, the middle subcell becomes a heterostructure with an InGaP emitter and its window is converted from InAlP to InGaP. This modification eliminated the refractive index discontinuity at the window/emitter interface of the middle subcell, as more fully described in U.S. patent application Ser. No. 12/258,190, filed Oct. 24, 2008. Moreover, the window layer 110 is preferably is doped three times that of the emitter 111 to move the Fermi level up closer to the conduction band and therefore create band bending at the window/emitter interface which results in constraining the minority carriers to the emitter layer.

[0074] In the preferred embodiment of the present invention, the middle subcell emitter has a band gap equal to the top subcell emitter, and the third subcell emitter has a band gap greater than the band gap of the base of the middle subcell. Therefore, after fabrication of the solar cell, and implementation and operation, neither the emitters of middle subcell B nor the third subcell C will be exposed to absorbable radiation. Substantially all of the photons representing absorbable radiation will be absorbed in the bases of cells B and C, which have narrower band gaps then the emitters. Therefore, the advantages of using heterojunction subcells are: (i) the short wavelength response for both subcells will improve, and (ii) the bulk of the radiation is more effectively absorbed and collected in the narrower band gap base. The effect will be to increase J.sub.sc.

[0075] On top of the cell B is deposited a BSF layer 113 which performs the same function as the BSF layer 109. The p++/n++ tunnel diode layers 114a and 114b respectively are deposited over the BSF layer 113, similar to the layers 109a and 109b, forming an ohmic circuit element to connect subcell B to subcell C. The layer 114a is preferably composed of p++ AlGaAs, and layer 114b is preferably composed of n++ InGaP.

[0076] A barrier layer 115, preferably composed of n-type InGa(Al)P, is deposited over the tunnel diode 114a/114b, to a thickness of about 1.0 micron. Such barrier layer is intended to prevent threading dislocations from propagating, either opposite to the direction of growth into the middle and top subcells B and C, or in the direction of growth into the bottom subcell A, and is more particularly described in copending U.S. patent application Ser. No. 11/860,183, filed Sep. 24, 2007.

[0077] A metamorphic layer (or graded interlayer) 116 is deposited over the barrier layer 115 using a surfactant. Layer 116 is preferably a compositionally step-graded series of InGaAlAs layers, preferably with monotonically changing lattice constant, so as to achieve a gradual transition in lattice constant in the semiconductor structure from subcell B to subcell C while minimizing threading dislocations from occurring. The band gap of layer 116 is constant throughout its thickness, preferably approximately equal to 1.5 eV, or otherwise consistent with a value slightly greater than the bandgap of the middle subcell B. The preferred embodiment of the graded interlayer may also be expressed as being composed of (In.sub.xGa.sub.1-x).sub.y Al.sub.1-yAs, with x and y selected such that the band gap of the interlayer remains constant at approximately 1.50 eV or other appropriate band gap.

[0078] In the surfactant assisted growth of the metamorphic layer 116, a suitable chemical element is introduced into the reactor during the growth of layer 116 to improve the surface characteristics of the layer. In the preferred embodiment, such element may be a dopant or donor atom such as selenium (Se) or tellurium (Te). Small amounts of Se or Te are therefore incorporated in the metamorphic layer 116, and remain in the finished solar cell. Although Se or Te are the preferred n-type dopant atoms, other non-isoelectronic surfactants may be used as well.

[0079] Surfactant assisted growth results in a much smoother or planarized surface. Since the surface topography affects the bulk properties of the semiconductor material as it grows and the layer becomes thicker, the use of the surfactants minimizes threading dislocations in the active regions, and therefore improves overall solar cell efficiency.

[0080] As an alternative to the use of non-isoelectronic one may use an isoelectronic surfactant. The term isoelectronic refers to surfactants such as antimony (Sb) or bismuth (Bi), since such elements have the same number of valence electrons as the P atom of InGaP, or the As atom in InGaAIAs, in the metamorphic buffer layer. Such Sb or Bi surfactants will not typically be incorporated into the metamorphic layer 116.

[0081] In an alternative embodiment where the solar cell has only two subcells, and the middle cell B is the uppermost or top subcell in the final solar cell, wherein the top subcell B would typically have a bandgap of 1.8 to 1.9 eV, then the band gap of the interlayer would remain constant at 1.9 eV.

[0082] In the inverted metamorphic structure described in the Wanlass et al. paper cited above, the metamorphic layer consists of nine compositionally graded InGaP steps, with each step layer having a thickness of 0.25 micron. As a result, each layer of Wanlass et al. has a different bandgap. In the preferred embodiment of the present invention, the layer 116 is composed of a plurality of layers of InGaAlAs, with monotonically changing lattice constant, each layer having the same band gap, approximately 1.5 eV.

[0083] The advantage of utilizing a constant bandgap material such as InGaAlAs is that arsenide-based semiconductor material is much easier to process in standard commercial MOCVD reactors, while the small amount of aluminum assures radiation transparency of the metamorphic layers.

[0084] Although the preferred embodiment of the present invention utilizes a plurality of layers of InGaAlAs for the metamorphic layer 116 for reasons of manufacturability and radiation transparency, other embodiments of the present invention may utilize different material systems to achieve a change in lattice constant from subcell B to subcell C. Thus, the system of Wanlass using compositionally graded InGaP is a second embodiment of the present invention. Other embodiments of the present invention may utilize continuously graded, as opposed to step graded, materials. More generally, the graded interlayer may be composed of any of the As, P, N, Sb based III-V compound semiconductors subject to the constraints of having the in-plane lattice parameter greater or equal to that of the second solar cell and less than or equal to that of the third solar cell, and having a bandgap energy greater than that of the second solar cell.

[0085] In another embodiment of the present invention, an optional second barrier layer 117 may be deposited over the InGaAlAs metamorphic layer 116. The second barrier layer 117 will typically have a different composition than that of barrier layer 115, and performs essentially the same function of preventing threading dislocations from propagating. In the preferred embodiment, barrier layer 117 is n+ type GaInP.

[0086] A window layer 118 preferably composed of n+ type GaInP is then deposited over the barrier layer 117 (or directly over layer 116, in the absence of a second barrier layer). This window layer operates to reduce the recombination loss in subcell C. It should be apparent to one skilled in the art that additional layers may be added or deleted in the cell structure without departing from the scope of the present invention.

[0087] On top of the window layer 118, the layers of cell C are deposited: the n+ emitter layer 119, and the p-type base layer 120. These layers are preferably composed of n+ type InGaAs and n+ type InGaAs respectively, or n+ type InGaP and p type InGaAs for a heterojunction subcell, although another suitable materials consistent with lattice constant and bandgap requirements may be used as well. The doping profile of layers 119 and 120 will be discussed in connection with FIG. 20.

[0088] A BSF layer 121, preferably composed of InGaAlAs, is then deposited on top of the cell C, the BSF layer performing the same function as the BSF layers 108 and 113.

[0089] . The p++/n++ tunnel diode layers 122a and 122b respectively are deposited over the BSF layer 121, similar to the layers 114a and 114b, forming an ohmic circuit element to connect subcell C to subcell D. The layer 122a is preferably composed of p++ InGaAlAs, and layer 122b is preferably composed of n++ InGaAlAs.

[0090] FIG. 3 depicts a cross-sectional view of the solar cell of FIG. 2 after the next sequence of process steps. A barrier layer 123, preferably composed of n-type GaInP, is deposited over the tunnel diode 122a/122b, to a thickness of about 1.0 micron. Such barrier layer is intended to prevent threading dislocations from propagating, either opposite to the direction of growth into the top and middle subcells A, B and C, or in the direction of growth into the subcell D, and is more particularly described in copending U.S. patent application Ser. No. 11/860,183, filed Sep. 24, 2007.

[0091] A metamorphic layer (or graded interlayer) 124 is deposited over the barrier layer 123 using a surfactant. Layer 124 is preferably a compositionally step-graded series of InGaAlAs layers, preferably with monotonically changing lattice constant, so as to achieve a gradual transition in lattice constant in the semiconductor structure from subcell C to subcell D while minimizing threading dislocations from occurring. The band gap of layer 124 is constant throughout its thickness, preferably approximately equal to 1.1 eV, or otherwise consistent with a value slightly greater than the band gap of the middle subcell C. The preferred embodiment of the graded interlayer may also be expressed as being composed of (In.sub.xGa.sub.1-x).sub.y Al.sub.1-yAs, with x and y selected such that the band gap of the interlayer remains constant at approximately 1.1 eV or other appropriate band gap.

[0092] In the surfactant assisted growth of the metamorphic layer 124, a suitable chemical element is introduced into the reactor during the growth of layer 124 to improve the surface characteristics of the layer. In the preferred embodiment, such element may be a dopant or donor atom such as selenium (Se) or tellurium (Te). Small amounts of Se or Te are therefore incorporated in the metamorphic layer 124, and remain in the finished solar cell. Although Se or Te are the preferred n-type dopant atoms, other non-isoelectronic surfactants may be used as

[0093] A window layer 125 preferably composed of n+ type InGaAlAs is then deposited over layer 124 (or over a second barrier layer, if there is one, disposed over layer 124,). This window layer operates to reduce the recombination loss in the fourth subcell D. It should be apparent to one skilled in the art that additional layers may be added or deleted in the cell structure without departing from the scope of the present invention.

[0094] FIG. 4 depicts a cross-sectional view of the solar cell of FIG. 3 after the next sequence of process steps. On top of the window layer 125, the layers of cell D are deposited: the n+ emitter layer 126, and the p-type base layer 127. These layers are preferably composed of n+ type InGaAs and p type InGaAs respectively, or n+ type InGaP and p type InGaAs for a heterojunction subcell, although another suitable materials consistent with lattice constant and bandgap requirements may be used as well. The doping profile of layers 126 and 127 will be discussed in connection with FIG. 20.

[0095] Turning next to FIG. 5, A BSF layer 128, preferably composed of p+ type InGaAlAs, is then deposited on top of the cell D, the BSF layer performing the same function as the BSF layers 108, 113 and 121.

[0096] Finally a high band gap contact layer 129, preferably composed of p++ type InGaAlAs, is deposited on the BSF layer 128.

[0097] The composition of this contact layer 129 located at the bottom (non-illuminated) side of the lowest band gap photovoltaic cell (i.e., subcell D in the depicted embodiment) in a multijunction photovoltaic cell, can be formulated to reduce absorption of the light that passes through the cell, so that (i) the backside ohmic metal contact layer below it (on the non-illuminated side) will also act as a mirror layer, and (ii) the contact layer doesn't have to be selectively etched off, to prevent absorption.

[0098] It should be apparent to one skilled in the art, that additional layer(s) may be added or deleted in the cell structure without departing from the scope of the present invention.

[0099] FIG. 6 is a cross-sectional view of the solar cell of FIG. 5 after the next process step in which a metal contact layer 123 is deposited over the p+ semiconductor contact layer 122. The metal is preferably the sequence of metal layers Ti/Au/Ag/Au.

[0100] Also, the metal contact scheme chosen is one that has a planar interface with the semiconductor, after heat treatment to activate the ohmic contact. This is done so that (1) a dielectric layer separating the metal from the semiconductor doesn't have to be deposited and selectively etched in the metal contact areas; and (2) the contact layer is specularly reflective over the wavelength range of interest.

[0101] FIG. 7 is a cross-sectional view of the solar cell of FIG. 3 after the next process step in which an adhesive layer 131 is deposited over the metal layer 130. The adhesive is preferably Wafer Bond (manufactured by Brewer Science, Inc. of Rolla, Mo.).

[0102] In the next process step, a surrogate substrate 132, preferably sapphire, is attached. Alternative, the surrogate substrate may be GaAs, Ge or Si, or other suitable material. The surrogate substrate is about 40 mils in thickness, and is perforated with holes about 1 mm in diameter, spaced 4 mm apart, to aid in subsequent removal of the adhesive and the substrate. As an alternative to using an adhesive layer 131, a suitable substrate (e.g., GaAs) may be eutectically or permanently bonded to the metal layer 130.

[0103] FIG. 8A is a cross-sectional view of the solar cell of FIG. 7 after the next process step in which the original substrate is removed by a sequence of lapping and/or etching steps in which the substrate 101, and the buffer layer 103 are removed. The choice of a particular etchant is growth substrate dependent.

[0104] FIG. 8B is a cross-sectional view of the solar cell of FIG. 8A with the orientation with the surrogate substrate 132 being at the bottom of the Figure. Subsequent Figures in this application will assume such orientation.

[0105] FIG. 9 is a simplified cross-sectional view of the solar cell of FIG. 8B depicting just a few of the top layers and lower layers over the surrogate substrate 132.

[0106] FIG. 10 is a cross-sectional view of the solar cell of FIG. 9 after the next process step in which the etch stop layer 103 is removed by a HCl/H.sub.2O solution.

[0107] FIG. 11 is a cross-sectional view of the solar cell of FIG. 10 after the next sequence of process steps in which a photoresist mask (not shown) is placed over the contact layer 104 to form the grid lines 501. As will be described in greater detail below, the grid lines 501 are deposited via evaporation and lithographically patterned and deposited over the contact layer 104. The mask is subsequently lifted off to form the finished metal grid lines 501 as depicted in the Figures.

[0108] As more fully described in U.S. patent application Ser. No. 12/218,582 filed Jul. 18, 2008, hereby incorporated by reference, the grid lines 501 are preferably composed of Pd/Ge/Ti/Pd/Au, although other suitable materials may be used as well.

[0109] FIG. 12 is a cross-sectional view of the solar cell of FIG. 11 after the next process step in which the grid lines are used as a mask to etch down the surface to the window layer 105 using a citric acid/peroxide etching mixture.

[0110] FIG. 13A is a top plan view of a wafer in which four solar cells are implemented. The depiction of four cells is for illustration for purposes only, and the present invention is not limited to any specific number of cells per wafer.

[0111] In each cell there are grid lines 501 (more particularly shown in cross-section in FIG. 9), an interconnecting bus line 502, and a contact pad 503. The geometry and number of grid and bus lines and the contact pad are illustrative and the present invention is not limited to the illustrated embodiment.

[0112] FIG. 13B is a bottom plan view of the wafer with four solar cells shown in FIG. 13A.

[0113] FIG. 14 is a cross-sectional view of the solar cell of FIG. 12 after the next process step in which an antireflective (ARC) dielectric coating layer 130 is applied over the entire surface of the bottom side of the wafer with the grid lines 501.

[0114] FIG. 15 is a cross-sectional view of the solar cell of FIG. 14 after the next process step according to the present invention in which first and second annular channels 510 and 511, or portion of the semiconductor structure are etched down to the metal layer 130 using phosphide and arsenide etchants. These channels define a peripheral boundary between the cell and the rest of the wafer, and leaves a mesa structure which constitutes the solar cell. The cross-section depicted in FIG. 15 is that as seen from the A-A plane shown in FIG. 17. In a preferred embodiment, channel 510 is substantially wider than that of channel 511.

[0115] FIG. 16 is a cross-sectional view of the solar cell of FIG. 15 after the next process step in which channel 511 is exposed to a metal etchant, and that portion of the metal layer 130 located at the bottom of the channel 511 is removed. The depth of the channel 511 is thereby extended to approximately to the top surface of the adhesive layer 131.

[0116] FIG. 17 is a top plan view of the wafer of FIG. 16 depicting the channels 510 and 511 etched around the periphery of each cell.

[0117] FIG. 18A is a cross-sectional view of the solar cell of FIG. 16 after the next process step in a first embodiment of the present invention in which the surrogate substrate 132 is appropriately thinned to a relatively thin layer 132a, by grinding, lapping, or etching. In this embodiment, the thin layer 132a forms the support for the solar cell in applications where a cover glass, such as provided in the second embodiment to be described below, is not required. In such an embodiment, electrical contact to the metal contact layer 130 may be made through the channel 510 or by other via structrures

[0118] FIG. 18B is a cross-sectional view of the solar cell of FIG. 16 after the next process step in a second embodiment of the present invention in which a cover glass 514 is secured to the top of the cell by an adhesive 513. The cover glass 514 preferably covers the entire channel 510, but does not extend to the periphery of the cell near the channel 511. Although the use of a cover glass is the preferred embodiment, it is not necessary for all implementations, and additional layers or structures may also be utilized for providing additional support or environmental protection to the solar cell.

[0119] FIG. 19 is a cross-sectional view of the solar cell of FIG. 18B after the next process step of the present invention in which the adhesive layer 131, the surrogate substrate 132 and the peripheral portion 512 of the wafer is entirely removed, breaking off in the region of the channel 510, leaving only the solar cell with the cover glass 514 (or other layers or structures) on the top, and the metal contact layer 130 on the bottom, which forms the backside contact of the solar cell. The surrogate substrate is preferably removed by the use of the etchant EKC 922. As noted above, the surrogate substrate includes perforations over its surface that allow the flow of etchant through the surrogate substrate 132 to permit its lift off. The surrogate substrate may be reused in subsequent wafer processing operations.

[0120] FIG. 20 is a graph of a doping profile in the emitter and base layers in one or more subcells of the inverted metamorphic multijunction solar cell of the present invention. The various doping profiles within the scope of the present invention, and the advantages of such doping profiles are more particularly described in copending U.S. patent application Ser. No. 11/956,069 filed Dec. 13, 2007, herein incorporated by reference. The doping profiles depicted herein are merely illustrative, and other more complex profiles may be utilized as would be apparent to those skilled in the art without departing from the scope of the present invention.

[0121] FIG. 21 is a graph that depicts the current and voltage characteristics of one of the test solar cells fabricated according to the present invention. In this test cell, the lower fourth subcell had a band gap in the range of approximately 0.6 to 0.8 eV, the third subcell had a band gap in the range of approximately 0.9 to 1.1 eV, the second subcell had a band gap in the range of approximately 1.35 to 1.45 eV and the upper subcell had a band gap in the range of 1.8 to 2.1 eV. The solar cell was measured to have an open circuit voltage (V.sub.oc) of approximately 3.265 volts, a short circuit current of approximately 16.26 mA/cm.sup.2, a fill factor of approximately 82%, and an efficiency of 32.2%.

[0122] FIG. 22 is a diagram representing the range of band gaps of various GaInAlAs materials as a function of the relative concentration of Al, In, and Ga. This diagram illustrates how the selection of a constant band gap sequence of layers of GaInAlAs used in the metamorphic layer may be designed through the appropriate selection of the relative concentration of Al, In, and Ga to meet the different lattice constant requirements for each successive layer. Thus, whether 1.5 eV or 1.1 eV or other band gap value is the desired constant band gap, the diagram illustrates a continuous curve for each band gap, representing the incremental changes in constituent proportions as the lattice constant changes, in order for the layer to have the required band gap and lattice constant.

[0123] FIG. 23 is a graph that further illustrates the selection of a constant band gap sequence of layers of GaInAlAs used in the metamorphic layer by representing the Ga mole fraction versus the Al to In mole fraction in GaInAlAs materials that is necessary to achieve a constant 1.5 eV band gap.

[0124] FIG. 24 is a graph that further illustrates the selection of a constant band gap sequence of layers of GaInAlAs used in the metamorphic layer by representing the mole fraction versus lattice constant in GaInAlAs materials that is necessary to achieve a constant 1.5 eV band gap band gap.

[0125] It will be understood that each of the elements described above, or two or more together, also may find a useful application in other types of constructions differing from the types of constructions described above.

[0126] Although the preferred embodiment of the present invention utilizes a vertical stack of four subcells, the present invention can apply to stacks with fewer or greater number of subcells, i.e. two junction cells, three junction cells, five junction cells, etc. In the case of four or more junction cells, the use of more than one metamorphic grading interlayer may also be utilized.

[0127] In addition, although the present embodiment is configured with top and bottom electrical contacts, the subcells may alternatively be contacted by means of metal contacts to laterally conductive semiconductor layers between the subcells. Such arrangements may be used to form 3-terminal, 4-terminal, and in general, n-terminal devices. The subcells can be interconnected in circuits using these additional terminals such that most of the available photogenerated current density in each subcell can be used effectively, leading to high efficiency for the multijunction cell, notwithstanding that the photogenerated current densities are typically different in the various subcells.

[0128] As noted above, the present invention may utilize an arrangement of one or more, or all, homojunction cells or subcells, i.e., a cell or subcell in which the p-n junction is formed between a p-type semiconductor and an n-type semiconductor both of which have the same chemical composition and the same band gap, differing only in the dopant species and types, and one or more heterojunction cells or subcells. Subcell A, with p-type and n-type InGaP is one example of a homojunction subcell. Alternatively, as more particularly described in U.S. patent application Ser. No. 12/023,772 filed Jan. 31, 2008, the present invention may utilize one or more, or all, heterojunction cells or subcells, i.e., a cell or subcell in which the p-n junction is formed between a p-type semiconductor and an n-type semiconductor having different chemical compositions of the semiconductor material in the n-type regions, and/or different band gap energies in the p-type regions, in addition to utilizing different dopant species and type in the p-type and n-type regions that form the p-n junction.

[0129] In some cells, a thin so-called intrinsic layer may be placed between the emitter layer and base layer, with the same or different composition from either the emitter or the base layer. The intrinsic layer may function to suppress minority-carrier recombination in the space-charge region. Similarly, either the base layer or the emitter layer may also be intrinsic or not-intentionally-doped (NID) over part or all of its thickness.

[0130] The composition of the window or BSF layers may utilize other semiconductor compounds, subject to lattice constant and band gap requirements, and may include AlInP, AlAs, AlP, AlGaInP, AlGaAsP, AlGaInAs, AlGaInPAs, GaInP, GaInAs, GaInPAs, AlGaAs, AlInAs, AlInPAs, GaAsSb, AlAsSb, GaAlAsSb, AlInSb, GaInSb, AlGaInSb, AIN, GaN, InN, GaInN, AlGaInN, GaInNAs, AlGaInNAs, ZnSSe, CdSSe, and similar materials, and still fall within the spirit of the present invention.

[0131] While the invention has been illustrated and described as embodied in a inverted metamorphic multijunction solar cell, it is not intended to be limited to the details shown, since various modifications and structural changes may be made without departing in any way from the spirit of the present invention.

[0132] Thus, while the description of this invention has focused primarily on solar cells or photovoltaic devices, persons skilled in the art know that other optoelectronic devices, such as, thermophotovoltaic (TPV) cells, photodetectors and light-emitting diodes (LEDS) are very similar in structure, physics, and materials to photovoltaic devices with some minor variations in doping and the minority carrier lifetime. For example, photodetectors can be the same materials and structures as the photovoltaic devices described above, but perhaps more lightly-doped for sensitivity rather than power production. On the other hand LEDs can also be made with similar structures and materials, but perhaps more heavily-doped to shorten recombination time, thus radiative lifetime to produce light instead of power. Therefore, this invention also applies to photodetectors and LEDs with structures, compositions of matter, articles of manufacture, and improvements as described above for photovoltaic cells.

[0133] Without further analysis, the foregoing will so fully reveal the gist of the present invention that others can, by applying current knowledge, readily adapt it for various applications without omitting features that, from the standpoint of prior art, fairly constitute essential characteristics of the generic or specific aspects of this invention and, therefore, such adaptations should and are intended to be comprehended within the meaning and range of equivalence of the following claims.