The disclosed system and method relate to a semiconductor device and, more particularly, to a semiconductor device including a tunnel junction that has a n-doped tunnel layer and a p-doped tunnel layer, where the p-doped tunnel layer is constructed of aluminum gallium arsenide antimonide (AlGaAsSb).
Wafer bonding technology may be used to monolithically join two materials with different lattice structures together. Wafer bonding technology has great potential. For example, joining gallium arsenide (GaAs) or indium phosphide (InP) based materials to other semiconductor materials may result in the integration of optical, photovoltaic, and electronic devices and enhance the performance of computers, solar cells, light emitting diodes and other electronic devices. In one specific example, a five junction (5J) cell, which is created by bonding a three junction (3J) GaAs-based cell with a two junction (2J) InP-based cell, results in a terrestrial solar cell having an efficiency of about 39% and a space solar cell having an efficiency of about 36%.
One requirement for an InP-based multi junction solar cell is a high transparency (which is also referred to as bandgap) tunnel junction. The tunnel junctions currently available that are employed in InP-based multi junction solar cells may sometimes absorb high amounts of light, or have very low peak tunneling currents. For example, one type of tunnel junction that is currently available includes a n-doped InP layer and a p-doped InAlGaAs layer. However, this tunnel junction may not always be easy to grow. This is because compounds containing a large amount of indium, such as InAlGaAs, are typically challenging to dope p-type. Furthermore, this type of tunnel junction may have a limited peak tunnel current as well. In another approach, a tunnel junction having a p-doped gallium arsenide antimonide (GaAsSb) layer and a n-doped indium gallium arsenide (InGaAs) layer may be used. This tunnel junction has a relatively high peak tunnel current, but both layers of this tunnel junction may also absorb light that is intended for active junctions of the solar cell. Thus, there exists a need in the art for a semiconductor device having a tunnel junction that is relatively easy to dope, has a relatively high transparency, and a relatively high peak tunnel current.
In one embodiment, a tunnel junction for a semiconductor device is disclosed. The tunnel junction includes a n-doped tunnel layer and a p-doped tunnel layer. The p-doped tunnel layer is constructed of aluminum gallium arsenide antimonide (AlGaAsSb).
In another embodiment, a semiconductor device is disclosed. The semiconductor device includes a first subcell, a second subcell, and a tunnel junction for electrically connecting the first subcell and the second subcell together in electrical series. The tunnel junction includes a n− doped tunnel layer and a p-doped tunnel layer. The p-doped tunnel layer is constructed of aluminum gallium arsenide antimonide (AlGaAsSb) and is doped with carbon.
In yet another embodiment, a method of constructing a photovoltaic device is disclosed. The method includes growth of a n-doped tunnel layer and a p-doped tunnel layer. The p-doped tunnel layer is constructed of aluminum gallium arsenide antimonide (AlGaAsSb).
Other objects and advantages of the disclosed method and system will be apparent from the following description, the accompanying drawings and the appended claims.
The first photovoltaic cell 22 may include a first emitter and base 20. In one exemplary embodiment, the first emitter and base 20 is an indium gallium arsenide phosphide (GaInPAs) emitter. In another embodiment, the first emitter and base 220 may be III-V material such as, but not limited to, aluminum arsenide antimonide (AlAsSb), AlGaAsSb, aluminum indium arsenide (AlInAs), indium phosphide (InP), aluminum gallium indium arsenide (AlGaInAs), gallium indium arsenide (GaInAs), or gallium arsenide antimonide (GaAsSb). In one embodiment, the first emitter and base 20 includes a separate emitter layer and base layer (not shown), where the emitter layer is nearest to incident light.
In the non-limiting embodiment as shown, the first photovoltaic cell 22 includes a bandgap of 1.1 eV. In another embodiment, the first photovoltaic cell 22 may include a bandgap of from about 0.73 to 2.45 eV. In yet another embodiment, the first photovoltaic cell 22 may include a bandgap of from about 1.0 to 1.1 eV and may be included in a three or more junction solar cell. The first photovoltaic cell 22 may be sensitive to a first-photoactive-subcell-layer wavelength. As used herein, the term wavelength may mean a single discrete wavelength, or, wavelength may include a range of wavelengths at which the layer material achieves a good light-to-electricity conversion efficiency.
The first photovoltaic cell 22 may also include a window layer 28. The window layer 28 may be disposed on a first side 30 of the first emitter and base 20, which would be positioned nearest to incident light L. As used herein, the relative terms top and bottom are used to indicate the surface nearest to and farthest from the incident light L, respectively. Also, when used to compare two layers, upper or above or overlying may refer to a layer closer to the sun, and lower or below or underlying may refer to a layer further from the sun or other source of illumination. The window layer 28 may be an InP, AlGaInAs, AlInAs, AlAsSb, AlGaAsSb, or a GaInPAs composition that provides bandgap energy greater than about 1.1 eV. The window layer 28 has two functions. The first function of the window layer 28 is to reduce minority-carrier recombination (i.e., to passivate) on a front surface 32 of the first photovoltaic cell 22. Additionally, the optical properties of the window material must be such that as much light as possible is transmitted to the first photovoltaic cell 22, and any additional photoactive subcell layers that may be disposed underneath thereof (not shown), where the photogenerated charge carriers may be collected more efficiently. If there is substantial light absorption in the window layer 28, carriers generated in the window layer are less likely to be subsequently collected and hence light absorption in the window degrades overall conversion efficiency.
The semiconductor device 10 may optionally include an antireflection (AR) layer or coating (not shown) disposed on the front surface 32 of the semiconductor device 10 nearest the incident light L, which is shown impinging from the direction indicated by the arrows. In one embodiment, the AR coating may be disposed atop the window layer 28. The AR coating may reduce surface reflections between the optically transparent media above the semiconductor device 10 (such as air, glass, or polymer) and various semiconductor layers of the semiconductor device 10, thereby enabling more photons to enter the semiconductor device 10. The AR coating may be constructed of materials such as, for example, titanium dioxide (TiO2), tantalum pentoxide (Ta2O5), silicon dioxide (SiO2), and magnesium fluoride (MgF2). The thickness of the AR coating may vary, but may range between about 0.04 and 0.35 microns. While an AR coating can be applied to the semiconductor device 10, in other embodiments another subcell may be stacked or applied above the semiconductor device 10.
The first photovoltaic cell 22 may further include a p-doped back surface field (BSF) layer 34 disposed on a bottom surface 36 of the first emitter and base 20. In the exemplary embodiment as illustrated, the p-doped BSF layer 34 is a p-doped InP BSF layer. In another embodiment, the p-doped BSF layer 34 may be an AlGaInAs, GaAsSb, AlAsSb, AlGaAsSb, AlInAs, GaInPAs and their alloys layer. In one embodiment, the BSF layer 34 is lattice-matched to InP. In another embodiment, the BSF layer 34 may be a coherently strained layer with a thickness below a Matthews-Blakeslee critical thickness.
The second photovoltaic cell 24 includes a second emitter and base 40. In the exemplary embodiment as shown, the second emitter and base 40 is a GaInPAs layer having an InP lattice constant. In another embodiment, the second emitter and base 40 may be GaInAs, GaAsSb, AlGaInAs, AlGaAsSb, GaInPAs and their alloys having an InP lattice constant. The second emitter and base 40 may have a bandgap lower than the bandgap of the first emitter and base 20. In the exemplary embodiment as shown in
The second photovoltaic cell may 24 further include a n-doped window layer 42 disposed on a top surface 44 of the second emitter and base 40. In general, the characteristics of the n-doped window layer 42 are similar to the window characteristics of the window layer 28. The n-doped window 42 may include a n-doping concentration of between about 2×1018/cm3 and 2×1019/cm3. In another embodiment, the n-doped window 42 has a n-doping concentration of about 1×1019/cm3 to create a relatively large electric field and to passivate the p-n junction.
The second photovoltaic cell 24 may further include a second BSF layer 48 below the second emitter and base 40, which is similar to the BSF layer 34. The tunnel junction 26 may electrically connect the first photovoltaic cell 22 and the second photovoltaic cell 24 together with one another in electrical series. It should also be appreciated that the tunnel junction 26 is a type-II tunnel junction, which reduces the tunneling energy barrier within the tunnel junction 26. This in turn increases tunneling probability as well as the peak tunneling current of the tunnel junction 26. For example, as seen in
In the embodiment as shown in
It should also be appreciated that the p-doped tunnel layer 60 may be lattice-matched with the p-doped InP BSF layer 34. The inclusion of antimonide in the p-doped tunnel layer 60 allows for lattice-matching with the p-doped InP BSF layer 34. Furthermore, the inclusion of aluminium within the p-doped tunnel layer 60 results in a relatively high bandgap (i.e., transparency) and a low level of light absorption. A relatively high bandgap may be any value greater than about 0.73 eV. In one embodiment, the p-doped tunnel layer 60 may include bandgap ranging from about 0.7 to about 1.4 eV.
The n-doped tunnel layer 62 may also be lattice-matched to InP. In one embodiment, the n-doped tunnel layer 62 is high bandgap III-V semiconductor having an InP lattice constant and that may form type II band alignment with the p-doped tunnel layer 60. In another embodiment, the n-doped tunnel layer 62 may be a highly n-doped InP, aluminium indium phosphide arsenic (AlInPAs), AlAsSb, or AlGaAsSb tunnel layer having a bandgap greater than or equal to 1.35 eV and an InP lattice constant. In one embodiment, the n-doped tunnel layer 62 is an InP tunnel layer having a bandgap of 1.35 eV. The n-doped tunnel layer 62 may be doped with relatively high levels of silicon or tellurium (i.e., Si or Te-doping). That is, the n-doped tunnel layer 62 may include an Si or Te-doping concentration of at least about 1019/cm3.
In one embodiment, the p-doped tunnel layer 60 and the n-doped tunnel layer 62 may be grown sequentially in a metalorganic vapor phase epitaxy (MOVPE) reactor. Furthermore, the semiconductor device 10 as well as various device components (e.g., the window, BSF) are grown in a MOVPE reactor. In another embodiment, the tunnel junction 26 may be grown in a chemical beam epitaxy (CBE), hydride vapor phase epitaxy (HVPE) or atomic layer deposition (ALD) reactor. In the embodiment as shown in
It is to be appreciated that the n-doped tunnel layer 162 may include a lower bandgap and a higher light absorbance than the n-doped tunnel layer 62 (
Referring generally to the figures, the disclosed tunnel junction includes the p-doped tunnel layer constructed of AlGaAsSb, which demonstrates improved performance characteristics when compared to some other tunnel junctions currently available. Specifically, the disclosed p-doped layer may be easier to grow, since AlGaAsSb may be doped more heavily with carbon. In contrast, compounds containing a high amount of indium are typically challenging to dope. In fact, a p-doped InAlGaAs layer may only be capable of being doped to the level of about 1018/cm3, and even this level of doping may be challenging. Furthermore, the disclosed tunnel junction also exhibits relatively high peak tunneling currents. Finally, it should also be appreciated that the disclosed p-doped tunnel layer constructed of AlGaAsSb also exhibits a higher bandgap (i.e., transparency) than some other types of tunnel junctions currently available. Finally, it is to be appreciated that high transparency is especially important in applications where the tunnel junction is placed in the upper portion of a solar cell that is located closer to incident light (i.e., the sun).
While the forms of apparatus and methods herein described constitute preferred embodiments of this invention, it is to be understood that the invention is not limited to these precise forms of apparatus and methods, and that changes may be made therein without departing from the scope of the invention.
This disclosure was made with U.S. Government support under Contract No. ZFM-2-22051-01 awarded by the Department of Energy. The U.S. Government has certain rights in this disclosure.