Patent Application
20060162768
This invention relates to optoelectronic devices, and, more specifically, to low bandgap, monolithic, multi-bandgap solar photovoltaic (SPV) and thermophotovoltaic (TPV) cells for converting solar and/or thermal energy to electricity as well as for related photodetector devices for detecting light signals and light emitting diode (LED) devices for converting electricity to light and/or infrared (IR) radiant energy.
It is well known that the most efficient conversion of radiant energy to electrical energy with the least thermalization loss in semiconductor materials is accomplished by matching the photon energy of the incident radiation to the amount of energy needed to excite electrons in the semiconductor material to transcend the bandgap from the valence band to the conduction band. However, since solar radiation and blackbody radiation usually comprise a wide range of wavelengths, use of only one semiconductor material with one bandgap to absorb such radiant energy and convert it to electrical energy will result in large inefficiencies and energy losses to unwanted heat.
Ideally, there would be a semiconductor material with a bandgap to match the photon energy for every wavelength in the radiation. That kind of device is impractical, if not impossible, but persons skilled in the art are building monolithic stacks of different semiconductor materials into devices commonly called tandem converters and/or monolithic, multi-bandgap or multi-bandgap converters, to get two, three, four, or more bandgaps to match more closely to different wavelengths of radiation and, thereby, achieve more efficient conversion of radiant energy to electrical energy. Essentially, the radiation is directed first into a high bandgap semiconductor material, which absorbs the shorter wavelength, higher energy portions of the incident radiation and which is substantially transparent to longer wavelength, lower energy, portions of the incident radiation. Therefore, the higher energy portions of the radiant energy are converted to electric energy by the larger bandgap semiconductor materials without excessive thermalization and loss of energy in the form of heat, while the longer wavelength, lower energy portions of the radiation are transmitted to one or more subsequent semiconductor materials with smaller bandgaps for further selective absorption and conversion of remaining radiation to electrical energy.
Semiconductor compounds and alloys with bandgaps in the various desired energy ranges are known, but that knowledge alone does not solve the problem of making an efficient and useful energy conversion device. Defects in crystalline semiconductor materials, such as impurities, dislocations, and fractures provide unwanted recombination sites for photogenerated electron-hole pairs, resulting in decreased energy conversion efficiency. Therefore, high-performance, photovoltaic conversion cells comprising semiconductor materials with the desired bandgaps, often require high quality, epitaxially grown crystals with few, if any, defects. Growing the various structural layers of semiconductor materials required for a multi-bandgap, tandem, photovoltaic (PV) conversion device in a monolithic form is the most elegant, and possibly the most cost-effective, approach.
Epitaxial crystal growth of the various compound or alloy semiconductor layers with desired bandgaps is most successful, when all of the materials are lattice-matched (LM), so that semiconductor materials with larger crystal lattice constants are not interfaced with other materials that have smaller lattice constants or vice versa. Lattice-mismatching (LMM) in adjacent crystal materials causes lattice strain, which, when high enough, is usually manifested in dislocations, fractures, wafer bowing, and other problems that degrade or destroy electrical characteristics and capabilities of the device. Unfortunately, the semiconductor materials that have the desired bandgaps for absorption and conversion of radiant energy in some energy or wavelength bands do not always lattice match other semiconductor materials with other desired bandgaps for absorption and conversion of radiant energy in other energy or wavelength bands. Therefore, fabrication of device quality, multi-bandgap, monolithic, converter structures is difficult, if not impossible, for some portions of the radiation frequency or wavelength spectrum.
This problem has been particularly difficult to solve in the infrared (IR) portion of the spectrum, where options for suitable, commercially available substrates on which to grow thin films with the necessary bandgaps for absorption and conversion of the infrared radiation to electrical energy are very limited, and where compatible, i.e., lattice-matched, semiconductor materials with the different bandgaps needed to absorb and convert different portions of the infrared spectrum efficiently are also quite limited.
For example, the group III-V family of semiconductor alloys include some of the best materials for fabricating photovoltaic converters with bandgaps in a range of about 0.35 eV to 1.65 eV to absorb and convert infrared (IR) radiation with wavelengths in a range of about 3.54 μm to 0.75 μm. Group III-V alloys comprise combinations of binary compounds formed from Groups III and V of the Periodic Table. These binary compounds can be alloyed together into various ternary or quaternary compositions to obtain any desired bandgap in the range of 0.35 eV to 1.65 eV. These alloys also have direct bandgaps (i.e., no change in momentum is required for an electron to cross the bandgap between the valance band and the conduction band), which facilitate efficient absorption and conversion of radiant energy to electricity. However, InP, which has a lattice constant of 5.869 Å (sometimes rounded to 5.87 Å) and a bandgap of 1.35 eV, is one of only a few feasible, commercially available substrate materials with a lattice constant even close to those lower bandgap Group III-V alloys i.e., InP-based or related ternary and quaternary compounds. The lowest bandgap Group III-V alloy that can be lattice-matched to the 5.869 Å lattice constant of an InP substrate is Ga0.47In0.53As, which has a bandgap of about 0.74 eV, which leaves a significant range of lower frequency, longer wavelength (>1.67 μm), infrared (IR) radiation that cannot be absorbed and converted to electricity in monolithic converters in which the semiconductor absorption materials are lattice-matched to the substrate.
While the current unavailability of efficient and cost-effective solar photovoltaic (SPV) converters, especially multi-bandgap, monolithic, converter devices, capable of absorbing and converting infrared (IR) radiation in wavelengths greater than 1.67 μm leaves substantial amounts of energy in the solar spectrum to remain unconverted to electricity, in state-of-the-art SPV's, it is an even greater problem for thermophotovoltaic (TPV) devices. Infrared (IR) radiation of wavelengths greater than 1.67 μm comprises a substantial amount of the energy radiated from blackbodies, and thermophotovoltaic (TPV) converters are intended to absorb and convert as much radiant energy from blackbodies to electric power as possible. Therefore, solutions to these problems, especially if such solutions could enable fabrication of monolithic converters with multiple bandgaps in infrared (IR) energy ranges, they would facilitate capture of more electric energy from solar and/or blackbody radiation.
U.S. Pat. No. 5,479,032 issued to S. Forrest et al., teaches that one or more ternary InxGa1-xAs alloys with x>0.53, i.3., with band-gaps less than 0.75 eV, can be grown epitaxially on an InP substrate by using intervening, graded layers of InAsyP1-y between the InP substrate and the InxGa1-xP (x>0.53) layers. However, those Forrest et al., patent teachings, which were directed to pixel detection of near infrared radiation incident on a focal plane for telecommunications applications, are not useful in SPV and TPV applications.
Accordingly, a general object of the present invention is to provide a monolithic, multi-bandgap, photovoltaic converter for absorbing and converting infrared (IR) radiation of multiple wavelengths to electricity.
A more specific object of this invention is to provide a photovoltaic converter with at least one bandgap less than 0.74 eV to absorb infrared radiation in wavelengths longer than 1.67 μm and convert it to electricity.
An even more specific object of this invention is to provide a electric device quality, multi-bandgap, monolithic, photovoltaic converter that has at least one lattice-matched (LM), double-heterostructure (DH) with a bandgap less than 0.74 eV to absorb infrared (IR) energy in wavelengths longer than 1.67 μm and convert it to electricity.
Another specific object of the invention is to provide a device quality, multi-bandgap, monolithic, photovoltaic device with at least one lattice-matched (LM), double-heterostructure (DH) with a bandgap less than 0.74 eV, which is not lattice-matched to an InP substrate, but including a lattice constant transition layer or layers, which is transparent to infrared radiation wavelengths longer than about 1.67 μm, positioned somewhere between such lattice-matched (LM), double-heterostructure (DH) and the InP substrate.
Still another of this invention object is to provide a lattice constant transition layer or layers, which is transparent to infrared (IR) radiation wavelengths longer than about 1.67 μm, positioned between two subcells in a multi-bandgap, monolithic device, where the two subcells are not lattice-matched to each other and at least one of the subcells has a bandgap, which is less than the bandgap of the other subcell and is less than 0.74 eV.
Another object of the present invention is to provide one or more subcells with bandgaps less than 0.74 eV on an InP substrate.
Another object of the present invention is to provide a bifacial, monolithic, integrated, module QI) comprising multiple subcells, at least one subcell of which absorbs and converts radiation wavelengths less than 0.92 μm to electricity.
Another object of the present invention is to provide a bifacial, monolithic, integrated, module (MIM) comprising multiple subcells, at least one subcell of which absorbs and converts radiation wavelengths less than 1.67 μm to electricity.
Another specific object of this invention is to provide a method of voltage-matching a plurality of subcell circuits that have subcells with different bandgaps less than or equal to 1.35 eV.
Additional objects, advantages, and novel features of the invention are set forth in part in the description that follows and will become apparent to those skilled in the art upon examination of the following description and figures or may be learned by practicing the invention. Further, the objects and the advantages of the invention may be realized and attained by means of the instrumentalities and in combinations particularly pointed out in the appended claims.
To achieve the foregoing and other objects and in accordance with the purposes of the a present invention, as embodied and broadly described herein, a method of this invention may comprise growing one or more subcell(s) that has a lattice constant greater than 5.869 Å, either alone or in combination with other subcells, on an InP substrate by using a lattice constant transition material between the InP substrate and the subcell(s) that have the lattice constants greater than 6.869 Å. The lattice constant transition material can be InAsyP1-y, where y is graded either continuously or in discrete stepped increments from one (1) to a value at which the InAsyP1-y has a lattice constant that matches the lattice constant of at least one of the subcells with a lattice constant greater than 5.869 Å. The subcell bandgap is lower than the bandgap of the InP substrate and lower than the bandgap of the InAsyP1-y, lattice constant transition material. Additional subcells with even lower bandgaps can also be added, and, if any of such additional subcells has an even greater lattice constant that cannot be matched to the first subcell, then one or more additional lattice constant transition layers can also be added. All of the subcells can be grown on only one side of the substrate (monofacial) or one or more subcells can be grown on the front side of the substrate while one or more other subcells can be grown on the back side (bifacial), using whatever lattice constant transition layers are necessary to accommodate the subcell(s) on each side of the substrate.
Isolation layers can be used between subcells for independent electrical connection of the subcells, although, in bifacial embodiments, the substrate can be insulating or semi-insulating to serve as an isolation layer. Alternately, tunnel junctions can be used for intra-cell current flow between subcells. Either the monofacial or bifacial subcell structures can be made in monolithic, integrated, modules (MIMs), which are particularly useful for voltage-matching a plurality of such subcells, although the bifacial embodiments are particularly suitable for such MIM structures and voltage matching. On the other hand, the monofacial embodiments are particularly useful in ultra-thin devices in which the substrate is removed.
To achieve the foregoing and other objects and in accordance with the purposes of the a present invention, as embodied and broadly described herein, of this invention may also comprise a monolithic, multi-bandgap, photovoltaic converter that has a first subcell comprising GaInAs(P) with a first bandgap and a first lattice constant, a second subcell comprising GaInAs(P) with a second bandgap and a second lattice constant, wherein the second bandgap is less than the first bandgap and the second lattice constant is greater than the first lattice constant, and further, wherein the second lattice constant is equal to a lattice constant of a InAsyP1-y alloy with a bandgap greater than the first bandgap, and a lattice constant transition material positioned between the first subcell and the second subcell, said lattice constant transition material comprising InAsyP1-y alloy with a lattice constant that changes gradually from the first lattice constant to the second lattice constant.
The first subcell is preferably a lattice-matched, double-heterostructure, comprising homojunction layers of GaInAs(P) clad by InAsyP1-y cladding layers wherein the InAsyP1-y cladding has a value for y in a range of o≦y<1, such the InAsyP1-y cladding layers of the first subcell have a lattice constant equal to the first lattice constant. The second subcell is is also preferably a lattice-matched, double-heterostructure comprising homojunction layers of GaInAs(P) clad by InAsyP1-y cladding layers, wherein the InAsyP1-y cladding has a value for y in a range of o≦y<1, such that the InAsyP1-y cladding layer of the second subcell have a lattice constant equal to the second lattice constant. Either a tunnel junction or an isolation layer is also positioned between subcells. The InP substrate can be doped with deep acceptor atoms to make the substrate more electrically insulating, and, in bifacial structures, this feature allows the substrate to serve as an electrical isolation between subcells positioned on opposite sides of the substrate.
A plurality of the monolithic, multi-bandgap, photovoltaic converters can also be grown on a common substrate in a monolithic, integrated, module (MIM), comprising the plurality of monolithic, multi-bandgap, photovoltaic converters, each of which comprises: (i) a first subcell with a first bandgap and a first lattice constant; (ii) a second subcell with a second bandgap and a second lattice constant, wherein the second bandgap is less than the first bandgap and the second lattice constant is greater than the first lattice constant; and (iii) a lattice constant transition material positioned between the first subcell and the second subcell, said lattice constant transition material having a bandgap at least as large as the first bandgap and a lattice constant that changes from the first lattice constant to the second lattice constant. Either monofacial structures or bifacial structures can be grown in MIM configurations, but the bifacial structure is particularly suited to MIM applications. The subcells in MIM structures can be isolated for independent electrical connection, or tunnel juctions can be provided. Isolated, independently connected, subcells are particularly adaped for voltage-matching in MIM structures. There can be more subcell stacks on one side of the substrate than the other to facilitate such voltage-matching, where the subcells on one side of the substrate are lower bandgap than subcell on the other side of the substrate.
The substrates can also be removed to provide ultra-thin photovoltaic devices.
The accompanying drawings, which are incorporated in and form a part of the specification, illustrate the preferred embodiments of the present invention, and together with the descriptions serve to explain the principles of the invention.
In the drawings:
A schematic diagram of principle components of a monofacial embodiment of a low-bandgap, monolithic, multi-bandgap (tandem) photovoltaic (PV) converter 10 according to this invention is shown in
In the monofacial embodiment or approach of this invention illustrated in
The monofacial PV converter 10 illustrated in
Referring again to the exemplary monofacial, monolithic, multi-bandgap, photovoltaic (PV) converter 10 illustrated in
A preferred lattice constant transition layer 20 according to this invention is a ternary InAsyP1-y material in which the proportion of As is gradually increased, either continuously or in discrete increments as will be discussed in more detail below. A significant feature of this invention is that the InAsyP1-y lattice constant transition layer 20 is transparent to infrared (IR) radiation wavelengths longer than those absorbed by the ternary GaxIn1-xAs or optional quaternary GaxIn1-xAsyP1-y in the first subcell 22, so there is virtually no loss of energy, or production of heat, in the lattice constant transition layer 20.
The monofacial, monolithic, multi-bandgap, photovoltaic (PV) converter 10 illustrated in
The back-surface reflector (BSR) or other spectral control element 28, which can also function as an electrode contact or lateral current flow element, is deposited on the second subcell 24, as will be described in more detail below. A spectral control layer 30 would usually be deposited on the front side 27 of the substrate 26 either to minimize reflection of incident radiation R, e.g., an anti-reflective coating (ARC), as is well-known to persons skilled in the art, especially for SPV converter applications, or to reflect all incident radiation R with wavelengths lower than those absorbable by the lowest bandgap subcell 24, especially for the TPV converter applications used for generating electricity and not heat. These structures and functions will be discussed in more detail below. The terms front and back, as used in this description, relate to the direction in which incident radiation propagates into and through a device or layers in a device. Therefore, radiation is incident first on the front of a device or layer and propagates toward the back of the device or layer.
In converter 10, the preferred substrate 26 comprises InP, because, as explained above: (i) InP has a lattice constant (5.87 Å), which is one of a few commercially available bulk, single crystal materials that are close to the lattice constants of Group III-V alloys that have bandgaps less than 0.74 eV (for absorbing infrared radiation wavelengths longer than about 1.67 μm); (ii) InP has a bandgap of about 1.35 eV (see
Subject to accommodations for a contact, buffer, cladding, optical control elements, and/or other auxiliary layers (not shown in
A particularly preferred subcell 22 lattice-matched to the InP substrate 26 comprises InP/Ga0.47In0.53As with a bandgap of about 0.74 eV, because, as shown by the lines 12, 14 in
An example second subcell 24 for use in conjunction with a first subcell 22 described above, therefore, can be a quaternary GaxIn1-xAsyP1-y or a ternary GaxIn1-xAs with a bandgap as low as 0.52 eV. A particular preferred example second cell 24 comprises a lattice-matched, double-heterostructure InAsyP1-y/GaxIn1-xAs with a bandgap 19 of 0.55 eV and a lattice constant 21 of about 5.952 Å.
The lattice constant transition layer 20, as mentioned above, gradually makes a transition from the lattice constant of the first subcell 22 to the lattice constant of the second subcell 24, preferably while remaining substantially transparent to all infrared radiation wavelengths that are not absorbed by the first subcell 22, as illustrated by the example PV converter 10 of
As also mentioned above, such graded transition of the InAsyP1-y lattice constant transition layer 20 from the lattice constant of the first subcell 22 (e.g., 5.87 Å) to the lattice constant of the second subcell 24 (e.g., 5.952 Å) can be done by increasing the proportion of As on a gradual continuous basis or, preferably, in incremental discrete steps as illustrated by line 23 in the bandgap chart in
A more specific example of the monofacial PV converter 10 of
The substrate 26 is preferably InP doped with a deep acceptor element, such as Fe, (sometimes denoted as InP:Fe or as (Fe) InP) to trap electrons and thereby suppress or prevent free carrier absorption. The substrate 26 can be semi-insulating for isolation or p-type for conducting, as desired for a particular application, and other layers and components are designated as either n-type or p-type, accordingly to provide the n/p junctions 34, 48 needed to convert the incident radiation R to electricity in the subcells 22, 24, respectively. However, p/n junctions would also work, as is understood by persons skilled in the art, so these n-type and p-type designations could be reversed by substituting donor dopants, for acceptor dopants and vice versa, which would be considered equivalent for purposes of the invention.
While the subcells 22, 24 can be simple shallow homojunctions, this invention is particularly conducive to the more efficient, lattice-matched, double-heterostructure subcells 22, 24 illustrated in
The second subcell 24 is also preferably a lattice-matched, double-heterostructure comprising a homojunction 48 formed by n-type and p-type layers 50, 52 of either ternary GaxIn1-xAs or quaternary GaxIn1-xAsyP1-y, but its lattice constant is larger than the lattice constant of the first subcell 22 and of the InP substrate 26, as explained above. Consequently, the second subcell 24 is lattice-mismatched (LMM) in relation to the InP substrate 26 and first cell 22, and it cannot be clad with InP. However, as explained above in relation to the lattice constant transition layer 20, InAsyP1-y can be formulated to have the same lattice constant as the GaxIn1-xAs or GaxIn1-xAsyP1-y homojunction layers 50, 52. Therefore, the passivation and confinement cladding layers 54, 56 of the second subcell 24 comprise InAsyP1-y that is lattice-matched to the GaxIn1-xAs or GaxIn1-xAsyP1-y homojunction layers 50, 52 to form the lattice-matched, double-heterostructure of that subcell 24.
Prior to growing the first subcell 22, a buffer layer 32 of n-InP about 300 Å thick is deposited first on a surface 25 of the InP substrate 26 to begin an epitaxial InP growth layer, if needed. If the InP substrate 26 is doped with a deep acceptor to be electrically insulating or semi-insulating as explained above, then provisions have to be made for a front electrical contact 29 and a conductive layer 33 for accommodating lateral flow of current produced by the subcells 22, 24 to or from the contact 29. Such a conductive layer 33 could be, for example, heavily n-doped InP or any other heavily doped material that is lattice-matched to the InP substrate 26 as well as transparent to all radiation wavelengths that are transmitted by the InP substrate 26. Then, the first subcell 22 comprising the lattice-matched, double-heterostructure of n-Ga0.47In0.53As/p-Ga0.47In0.53As homojunction layers 36, 38 between the two cladding layers 40, 42 of n-InP and p-InP, respectively. As is well-known in the art, semiconductor materials are usually doped with small amounts of elements from an adjacent group of the Periodic Table of the Elements to provide the majority carriers. Therefore, an appropriate donor dopant for the Group III-V semiconductor alloy used in this invention can be, for example, sulphur (S) from Group VI, and appropriate acceptor dopant can be, for example, zinc (Zn) from Group II. The InP buffer layer 32 grown epitaxially on the InP substrate 26 in this example is preferably heavily (10−18-10−20 cm−3) n-type doped with sulfur (S). Then, the InP front cladding layer 40 is grown epitaxially on the buffer layer 32 to a thickness of about 0.01-0.1 μm, but it is more lightly doped n-type with, for example, S to a dopant level of about 1016-1020 cm−3. The Ga0.47In0.53As homojunction layers 36, 38, which lattice-match the InP substrate 26, buffer layer 32, and cladding layer 40, are grown epitaxially. Therefore, the bandgap of the first subcell 22 is about 0.74 eV, which absorbs portions of the incident radiation R with wavelengths of about 1.67 μm and less, as explained above, although other values of x and other formulations would also work in this invention. Lattice-matching quaternary GaxIn1-xAsyP1-y is also possible. The emitter layer 36 of subcell 22 is grown epitaxially to a thickness, preferably in a range of about 0.1-10 μm, and is doped n-type with, for example, S to a dopant level in a range of about 1016-1020 cm−3. The base layer 38 is then grown epitaxially to a thickness of about 0.01-10 μm, and doped p-type to create the n/p junction 34. The p-type dopant, such as Zn in this example, is at a dopant level of about 1016-1020 cm−3. To complete the lattice-matched, double-heterostructure, first subcell 22, the back cladding layer 42 is grown epitaxially on the base layer 38 to a thickness, preferably of about 0.01-0.1 μm, and is p-type doped, for example, with Zn, to a dopant level of about 1016-1020 cm−3 Each of the buffer layer 32, conductive layer 33, and/or cladding layer 40 can all serve any one or more of these functions, individually or together. Therefore, instead of the three distinct layers 32,33,40 shown in
The subcells 22, 24 can be electrically connected together in series, or they can be electrically isolated from each other, as will be described in more detail below. For a monolithic, multi-bandgap, PV device 10 in which the subcells 22, 24 are series connected, a tunnel junction comprising a layer 44 of heavily p-doped Ga0.47In0.53As or GaxIn1-xAsyP1-y followed by a heavily n-doped Ga0.47In0.53As or GaxIn1-xAsyP1-y layer 46 is deposited and grown epitaxially on the back cladding layer 42 of the first subcell 22 to facilitate low-resistive current flow in an ohmic manner between the first subcell 22 and the second subcell 24. Again, if the preferred homojunction layers 36, 38 of subcell 22 comprise Ga0.47In0.53As, as discussed above, then it is also preferable that x=0.47 in the GaxIn1-xAs of the tunnel junction layers 44, 46 in order to lattice-match them with the underlaying InP and Ga0.47In0.53As layers described above, although other values of x and other formulations would also work in this invention. Tunnel junctions are well-known in the art, but, for purposes of this invention, each tunnel junction layer 44, 46 can be about 0.01-0.1 μm thick and doped to a level of about 10−18-10−20 cm−3. Alternative monolithic, multi-bandgap, PV converters with the subcells 22, 24 isolated electrically from each other will be described below.
The transparent, lattice constant transition layer 20 of this invention comprising gradually increasing lattice constants is deposited and grown epitaxially on the GaxIn1-xAs or GaxIn1-xAsyP1-y tunnel junction layer 46 in order to make the transition from the lattice constant of the InP substrate 26 and intervening layers described above to a lattice constant that matches the GaxIn1-xAs or GaxIn1-xAsyP1-y of the second subcell 24, which is formulated to provide a desired bandgap Eg2, as described above. According to this invention, the bandgap Eg2 is less than the bandgap Eg1 of the first subcell 24 in the monofacial, inverted PV converter embodiment 10 of
To form the lattice constant transition layer 20, (As) is added to a growing layer of InP in increasing proportions so that the proportion of arsenic (As) increases in the resulting InAsyP1-y material, which increases the lattice constant of the InAsyP1-y. As mentioned above, this change can be accomplished continuously, but it is preferred that the changes in proportions be made in incremental steps. In the InAsyP1-y of the lattice constant transition layer 20 of this example PV converter 10, y varies from zero (where it lattice-matches the Ga0.47In0.53As of the first subcell 22) to about 0.44, where it lattice-matches to the GaxIn1-xAs of the second subcell 24, in which x≈0.26 and the consequent bandgap Eg2 is about 0.55 eV. That example bandgap Eg2=0.55 eV enables the second subcell 24 to absorb infrared radiation R with wavelengths up to about 2.25 μm. In general, the lattice-matching condition of GaxIn1-xAs to InAsyP1-y occurs when the crystal lattices of the epi-layers are fully relaxed, which is where y≈2.143x.
Of course, as mentioned above, the GaxIn1-xAs of the second subcell 24 can have x equal to some other value for a different desired bandgap Eg2, and the y in the InAsyP1-y of the lattice constant transition layer 20 can be varied or customized accordingly to make the necessary corresponding lattice constant transition. Also, as mentioned above, either or both of the subcell materials and/or the lattice constant transition materials could be quaternary GaxIn1-xAsyP1-y with the x and y values customized to desired bandgaps and lattice constants within the physical constraints illustrated by the bandgap vs. lattice parameter chart of
As explained above and shown in
A back surface spectral control element 28, which can also be used as a back electrical contact layer, can be deposited onto the back cladding layer 56 or onto an additional contacting layer (not shown) disposed atop the back cladding layer 56. The nature of the back surface spectral element 28 may depend on the application of the device 10. For example, if the device 10 is a SPV or TVP, the sole purpose of which is to convert radiation to electricity, then the back surface spectral element may comprise a reflector to reflect any remaining, unabsorbed radiation from the second subcell 24 back through the subcells 24, 22. Some of such reflected radiation could be absorbed in this second pass through the subcells, but most of it will continue propagating all the way back through the substrate 26 toward whatever radiator source (not shown) produces the incident radiation R in the first place. Adding such unabsorbed, back-reflected, radiation energy back into the radiator source may enable the radiator source to use such back-reflected energy in the production of new incident radiation R for conversion to electricity in the converter 10. This feature is particularly appropriate for TPV configurations of converter 10 that are applied to convert infrared radiation R produced by a blackbody infrared radiation source (not shown) to electricity. Any radiation reflected back into the blackbody radiator adds energy to the blackbody radiator and thereby tends to raise the temperature of the blackbody radiator, which causes the blackbody radiator to produce more blackbody infrared radiation for the converter 10. Therefore, such back-reflected radiation can help the blackbody radiator to produce more incident radiation R for the device 10 without having to use so much fuel.
On the other hand, some devices 10 are used both for producing electricity and gathering heat for an environment. In those applications, the back surface spectral control element 28 may be a material that is transparent to remaining infrared radiation that is not absorbed by the second subcell 24 so that such remaining infrared can be used as heat someplace behind the device 10.
If the layer 28 is a back surface reflector (BSR), there can be several advantages to designing the last (second) subcell 24 with only one-half of its normal thickness, i.e., one-half the thickness that would be required for full absorption of radiation in the wavelengths that correspond to the bandgap, because any unabsorbed radiation will be reflected by the BSR 28 back into the last subcell 24. The advantages of this kind of design include an enhanced photocurrent, higher operating voltage, and thinner structure that requires less growth time and provides easier device processing. Regardless of its optical characteristics, as described above, the layer 28 can also be a back surface electrical contact. Therefore, it is preferably electrically conductive. An optional, additional metallic contact 45 can also be used on the conductive layer 28 for making an electrical connection, if desired.
The design of the front surface spectral control element 30 on the front surface 27 of the substrate 26 may also depend on usage of the device 10. For example, if the device 10 is to be used only for producing electricity from blackbody radiation, the front surface spectral control element 30 may be a coating layer that transmits only shorter wavelength incident radiation R that can be absorbed and converted to electricity by the subcells 22, 24 and that reflects all longer wavelength incident radiation R back into the blackbody radiator (not shown) for recovery and re-use. On the other hand, if the device 10 is to be used both for producing electricity and heat for an environment, then the front surface spectral control element 30 may be an antireflective coating to enhance transmission of all the incident radiation R into the device 10.
As mentioned above, the monofacial PV converter 10 described above and illustrated in
There are a number of reasons that such electrical isolation of the subcells 22, 24 may be desirable in some applications. For example, as mentioned above, current flow through series connected subcells 22, 24 is limited by the lowest photocurrent producing subcell. Therefore, for series connected subcells, a number of subcell design factors, such as bandgaps, thicknesses, doping concentrations, and the like are used to optimize the operating characteristics of the series connected subcells 22, 24, so that electric power production from the tandem combination is maximized. In some designs and applications, however, more efficient conversion of radiant energy to electricity can be accomplished by extracting electric power from the individual subcells 22, 24 separately or independently, or, in some applications, to design the subcells 22, 24 for voltage matching. Such voltage matching techniques with subcells in other devices will be discussed in more detail below in relation to monolithic, integrated module (MIM) devices.
To isolate the subcells 22, 24 electrically from each other, there has to be some material between them that inhibits electric current flow between the subcells 22, 24. However, such electrical isolation material cannot interfere with radiation transmission from one subcell 22 to the other subcell 24. In the PV converter 110 of
An isolation material for isolation layer 39 can be fabricated in a number of ways. One such approach is to fabricate the isolation layer 39 with a high-resistivity semiconductor material that has a high enough bandgap to be transparent to the longer wavelength radiation that is not absorbed in the first subcell 22 and is being transmitted to the second subcell 24. Another such approach is to form the isolation layer 39 as an isolation diode, which, of course, must also be transparent to the radiation being transmitted from the first subcell 22 to the second subcell 24. Also, such high-resistivity material or isolation diode material has to be lattice-matched to the materials in front and in back of it, which, in the position of isolation layer 39 shown in
As mentioned above, InP doped with a deep acceptor element, such as Fe or Cr, is a high-resistivity material and has a bandgap (1.35 eV) that makes it transparent to all radiation that is not absorbed by the first subcell 22. It is also lattice-matched to the InP substrate 26 and to the ternary Ga0.47In0.53As or quaternary GaxIn1-xAsyP1-y of the first subcell 22. Therefore, deep acceptor-doped InP can be used as the high-resistivity, isolation layer 39. Such deep acceptor-doping of other lattice-matched semiconductor materials, such as ternary GaxIn1-xAs, quaternary GaxIn1-xAsyP1-y, or even AlGaInAs in some circumstances, with high enough bandgaps to be transparent to the radiation being transmitted, could also be used to provide suitable high-resistivity materials for the isolation layer 39.
An isolation diode for isolation layer 39 can be provided by one or more doped junctions, such as an n-p junction or n-p-n junctions with high enough reverse-bias breakdown characteristics to prevent current flow between the subcells 22, 24. Again, lattice-matched semiconductor materials, such as InP, GaxIn1-xAs or GaxIn1-xAsyP1-y, or even AlGaInAs, can be doped to provide an isolation diode structure for isolation layer 39.
While a discrete isolation layer 39 is shown in the PV converter 110 of
Of course, with each subcell 22, 24 isolated electrically from each other, some additional provisions for electrical contacts are necessary to extract electric power independently from each subcell 22, 24. Persons skilled in the art will be able to design myriad structures for such contacts, once they understand the principles of this invention. The example additional contacts 27, 42 for this purpose are shown fabricated on lateral current flow layers 39, 41 respectively. Such lateral current flow layers 39, 41 are lattice-matched to their respective subcells 22, 24 and must be transparent to radiation being transmitted from the first subcell 22 to the second subcell 24. Heavily doped GaIn1-xAsyP1-y with 0≦x≦1 and 0≦y≦1 as necessary for lattice matching and transparency can be used for these lateral current flow layers 39, 41.
While the series connected PV converter 10 and isolated or independently connected PV converter 110 described above are illustrated with only two subcells 22, 24, and only one lattice constant transition layer 20 between them, any number of subcells with any number of lattice constant transition layers can be included in a monolithic, multi-bandgap, optoelectronic device according to this invention. To illustrate this principle, a more complex monolithic, multi-bandgap, PV converter 112 is illustrated in
In the PV converter 112, an arbitrary number (five) subcells 114, 116, 118, 120, 122 are illustrated with arbitrary bandgaps Eg1>Eg2>Eg3>Eg4>Eg5. The substrate 124 is InP, and the first and second subcells 114, 116 both have bandgaps Eg1, Eg2 that can be ternary GaInAs or quaternary GaInAsP and are lattice-matched to the InP substrate 124 (see, e.g., lines 130, 12, 14 in
As mentioned above, the numbers and combinations of subcells and lattice constant transition layers as well as the specific example bandgap values shown in the PV converter 112 of
Now, as illustrated in another alternative inverted, monofacial, multi-bandgap, PV converter 140 in
For example, but not for limitation, if it is desired to have the first subcell 22 in the PV converter 140 of
The first subcell 22 with the example desired 0.70 eV bandgap can then be a lattice-matched (LM), double-heterostructure (DH) of, for example, InAsyP1-y/GaxIn1-xAsyP1-y with the same lattice constant, 5.94 Å, as the terminal InAsyP1-y of the lattice constant transition layer 20 (see broken line 60 in
The second subcell 24 in this example can be formulated with a lattice-matched (LM), double-heterostructure (DH) with the same lattice constant of 5.94 Å as the first subcell 22 and still have a bandgap as low as 0.58 eV (see broken lines 60, 66 in
While two lattice-mismatched (LM) subcells 22, 24 and one lattice constant transition layer 20 in any of a variety of ternary and/or quaternary formulations comprising Ga, In, As, and/or P provide wide flexibility in low bandgap designs for efficient absorption and conversion of desired infrared radiation wavelength bands to electricity, this invention also extends to three, four, five, or more subcells and bandgaps with one or more lattice constant transition layers, as needed. For example, there is no theoretical limit to the number of quaternary GaxIn1-xAsyP1-y formulations for different bandgaps between lines 13 and 14 (0.74 eV to 1.35 eV) in
In other words, every ternary GaxIn1-xAs or quaternary GaxIn1-xAsyP1-y with a bandgap in the range between 0.74 eV and 0.355 eV (lines 14 and 31 in
This invention, as mentioned above, also extends to low bandgap, monolithic, multi-bandgap PV converters with more than one lattice constant transition layer. For example, referring again to
As was explained above in relation to the inverted tandem (two-subcell) PV converter 140 in
The energy in the infrared radiation R wavelengths longer than 2.14 μm, which are not absorbed in the second subcell 24 would be wasted in the PV converter 140 embodiment of
Again, as explained above for the first lattice constant transition layer 20, the gradual change of lattice constant in the second lattice constant transition layer 70 can be graded gradually or in discrete stepped increments, the latter of which is preferred. Also, while not shown in detail in
All of the PV converter embodiments 10, 110, 140, 150 described above have been monofacial, i.e., grown on only one face of the substrate. A very significant feature of this invention is that it can also be implemented in bifacial or buried substrate structures, as illustrated diagrammatically by the example low bandgap, monolithic, multi-bandgap, PV converter 80 in
In the example bifacial PV converter 80 in
In this example, any incident radiation R of wavelengths shorter than 1.67 μm will be absorbed by the 0.74 eV bandgap Ga0.47In0.53As in the first subcell 82, and longer wavelength infrared radiation R will pass through the first subcell 82. The InP substrate 84, which has a much higher bandgap of 1.35 eV (line 13 in
As shown by lines 12, 93 in
If the substrate 84 is doped with a deep acceptor element, such as Fe or Cr, to be an insulator or semi-insulator, as explained above, then the first subcell 82 and the second subcell 86 are electrically isolated from each other. Therefore, electricity has to be extracted independently from each subcell 82, 86, as described above for the electrically isolated subcells 22, 24 of the PV converter device 110 in
As explained above, any of a wide range of ternary or quaternary GaInAs(P) alloys with any combinations of bandgaps and lattice constants can be used in subcells of tandem (more than one subcell) stacks of low-bandgap, monolithic, multi-bandgap, optoelectronic devices according to this invention. Another example of such combinations is illustrated in the alternate example bifacial PV converter device 160 in
The third subcell 168 in the PV converter 160 in
Again, if the InP substrate 166 is deep acceptor doped to be an insulator or semi-insulator, the third subcell 168 will be electrically isolated from the first and second subcells 162, 164 and can be connected independently to other PV converters or subcells, such as in a MIM structure (described below). Otherwise, a separate isolation layer (not shown in
Another interesting variation of the bifacial embodiment PV converter 80 in
An illustration of this principle in a slightly more complex bifacial, monolithic, multi-bandgap, solar photovoltaic (SPV) converter device 190, multiples of which can also be incorporated into a MIM structure, is shown in
In the example SPV device 190, the first subcell 192 is shown as a LM, DH, InP subcell with a bandgap of 1.35 eV, while the bandgaps of the second and third subcells 194, 196 have lower bandgaps, e.g., 1.0 eV, LM, DH, quaternary GaxIn1-xAsyP1-y for the second subcell 194 and 0.74 eV, LM, DH, ternary Ga0.47In0.53As for the third subcell 196, which is the lowest bandgap GaInAs that can be lattice-matched to the InP substrate 198 (see lines 13, 14, 210 in
Again, the variations of conductive or highly-resistive substrate 198, isolation and/or tunnel junction layers, electrical contacts, buffer layers, optical control layers, more or fewer subcells, different bandgaps, and the like, as described above for other embodiments, are also applicable to the SPV device 190 described above.
Also, AlInAs in slightly higher bandgaps than the 1.35 eV bandgap of the InP can also be lattice-matched to InP, so a first subcell of such AlInAs lattice-matched to the InP substrate could also be used as part of a bifacial, monolithic, multi-bandgap, PV converter of this invention. Of course, Ga could be added to produce AlGaInAs, if a slightly lower bandgap than AlInAs is desired for either the first subcell or a subsequent subcell of a bifacial PV converter of this invention.
The PV converters described above can be used alone or in combinations with myriad other devices. For example, any of the PV converters, especially the SPV device 190, but also, PV converters 10, 110, 112, 140, 150, 80, 160, can be used for the bottom cell device in a mechanical stack of higher bandgap (higher than the 1.35 eV bandgap of InP) PV converters, such as GaAs based PV converters, in solar cell and other applications. Such other, higher bandgap, PV converters (not shown) can selectively absorb and convert shorter wavelength solar energy to electricity, while the lower bandgap PV converters, e.g., PV converters 10, 110, 112, 150, 80, 160, 190, absorb and convert longer wavelength solar radiation to electricity.
As mentioned above, the bifacial, monolithic, multi-bandgap, optoelectronic devices of this invention, for example, the bifacial PV converters 80, 160, 190 shown in
In the bifacial MIM PV converter device 230 illustrated in
One advantage of being able to have different numbers of subcell stacks 190′, 190″ on front and back of the substrate 198 is more flexibility to design voltage-matched subcell circuits. A schematic diagram of an equivalent electrical circuit corresponding to the example voltage-matched subcell circuits 262, 264, 266, 270, 272 of the MIM 230 of
However, the front-side subcells 192, 194, 196 of the front-side stacks 190′ can be connected in myriad combinations of series and/or parallel electrical connections, as illustrated in
To illustrate several series and parallel connection options, the bifacial MIM PV converter device 230 in
The back-side subcells 200, 202 are even lower voltage than the front-side subcells 196, but there are more of them than the front-side subcells 192, 194, 196, so the back-side voltage can be matched to the front-side voltage. In the example of
Finally, the front-side subcell circuits 262, 264, 266 are connected in parallel to the back-side subcell circuits 270, 272 at terminal contacts 258, 258′ to add their respective current outputs. Therefore, the bifacial MIM PV converter 230 can be connected electrically to other devices or loads via the two terminal contacts 256, 258, which is a very desirable feature of this invention. Other MIM structures, circuit connections, and advantages can be made according to these principles within this scope of this invention. For example, but not for limitation, the monofacial, monolithic, LM, DH, multi-bandgap, PV converters described above can also be incorporated into MIM structures (not shown), although the bifacial embodiments described above have the advantage of using the substrate 198 as a built-in isolation structure between subcells on the front side and subcells on the back side, as explained above.
Any of the PV converter embodiments 10, 110, 150, described above and shown in
The purpose of the stop-etch layer 98 is to enable the InP substrate 26 and buffer layer 32 to be removed by etching or other selective chemical removal to create an ultra-thin, monolithic, multi-bandgap (tandem) PV converter 100 without etching or damaging any of the first subcell 22. After the several layers of the structure in
Mounting the PV converter 100 on the object 102 can be accomplished with a suitable adhesive or by any other suitable mounting mechanism. An anti-reflective coating 97 can be added to reduce reflection of incident radiation, or layer 97 can be any other optical control material for purposes described above for the PV converter 10.
This ultra-thin, monolithic, multi-bandgap, PV converter 100 enables this device to be used as a solar cell, because elimination of the InP substrate 26 allows all of the incident solar radiation SR to reach the subcells 22, 24, which can convert it to electricity. Otherwise, the InP substrate 26, which has a bandgap of 1.35 eV, would absorb large amounts of solar radiation SR in wavelengths shorter than 0.93 μm, before such solar radiation SR could reach the first subcell 22. There is no n/p junction in the substrate 22, and it cannot convert radiant energy to electricity, so any solar energy absorbed by the substrate 26 would be thermalized and wasted as heat.
Even without the InP substrate, however, there could be significant production of heat in the PV converter 100, when it is used as a solar cell, because there is a substantial amount of energy in higher frequencies (shorter wavelengths) of the solar spectrum, where wavelengths are substantially shorter than the longest wavelength that can be absorbed by the first subcell 22. Therefore, there is significant thermalization of excess energy that is not needed for carriers to transcend the bandgap Eg1 of the first subcell 22, thus a significant production of heat that must be dissipated from the PV converter 100. However, the PV converter 100 is ultra-thin and has no thick substrate, so heat can flow through the PV converter 100 is substantially one-dimensional, and it can flow quickly and easily to the back surface 104. If the object 102 on which the PV converter 100 is mounted is a good heat sink, i.e., good thermal conductivity and sufficient mass and/or surface area to conduct heat away from the PV converter 100, the combination provides very good thermal management and minimizes heat build-up in the PV converter 100.
The ultra-thin, monolithic, multi-bandgap, PV converter 100 can also be grown in a polycrystalline form on less expensive substrates, such as graphite, which is amorphous and does not impose a lattice constant on the first subcell 22, or in single-crystal form on compliant substrate or bonded substrate systems, which provide a lattice constant match to accommodate epitaxial growth. A typical compliant substrate may be made, for example, with an inexpensive substrate material, such as silicon, and with an amorphous oxide of the substrate material followed by a layer of perovskite oxide. Therefore, a first subcell 22 of InP, GaInAs, or GaInAsP will grow with its natural lattice constant. Such first subcell 22 can then be followed by a InAsyP1-y lattice constant transition layer 20 and another, lower bandgap, second subcell 24, as described above. Then, the resulting ultra-thin PV converter 100 is mounted on another object 102 and the compliant substrate is removed.
Compliant substrates can also be used on any of the monofacial PV converter embodiments 10, 110, 112, 140, 150 described above, and they can possibly be used for the bifacial embodiments 80, 160, 190, 230. Possible uses of compliant substrates in the embodiments of this invention depend on the transparency and other properties of the compliant substrate materials and systems being considered and/or applied.
While the description of this invention has focused primarily on photovoltaic converters, persons skilled in the art know that other optoelectronic devices, such as photodetectors and light-emitting diodes (LEDs) are very similar in structure, physics, and materials to PV converters with some minor variations in doping and the like. For example, photodetectors can be the same materials and structures as the PV converters described above, but perhaps more lightly-doped for sensitivity rather than power production. On the other hand LED's can also be much the same 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, apparatus, compositions of matter, articles of manufacture, and improvements as described above for PV converters.
Since these and numerous other modifications and combinations of the above-described method and embodiments will readily occur to those skilled in the art, it is not desired to limit the invention to the exact construction and process shown and described above. For example, accordingly, resort may be made to all suitable modifications and equivalents that fall within the scope of the invention as defined by the claims which follow. The words “comprise,” “comprises,” “comprising,” “include,” “including,” and “includes” when used in this specification and in the following claims are intended to specify the presence of stated features or steps, but they do not preclude the presence or addition of one or more other features, steps, or groups thereof. Also, GaInAs(P) is used as a shorthand, generic term that includes any ternary GaxIn1-xAs and/or quaternary GaxIn1-xAsyP1-y, and similar notation conventions apply to AlGaInAs(P).
The United States Government has rights in this invention under Contract No. DE-AC36-99GO10337 between the United States D Renewable Energy Laboratory, a Division of the Midwest Research Institute.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/US02/16101 | 5/21/2002 | WO | 11/19/2004 |
