Patent Application
20130048064
High-efficiency multijunction solar cells are fabricated from materials with different band gaps. In a typical multijunction solar cell, individual single-junction cells with different energy band gaps (Eg) are stacked on top of each other. Sunlight falls first on the material having the largest band gap, and the highest energy photons are absorbed. Photons not absorbed in the first or top cell are transmitted to the second cell, which absorbs the higher energy portion of the remaining solar radiation, while remaining transparent to the lower energy photons. In theory, any number of cells can be used in multijunction devices. There is a desire to make multijunctions solar cells with four or more cells. However, to date, only two or three cells have been functionally designed.
Multijunction solar cells may be made in one of two ways, monolithically or mechanically stacked. Monolithic multijunction solar cells are typically made by sequentially growing all the necessary layers of materials for two or more cells and the necessary interconnection between the cells. Ideally these materials can be grown epitaxially, but for some material combinations, this is impossible or undesirable. Growing four solar cell junctions on the same substrate requires lattice-mismatched epitaxy, and the associated dislocations can degrade the performance of the fourth solar cell, such that the resulting device performs more poorly than existing three junction devices.
Another approach is to spectrally split the light and send the spectrally split light to different junctions grown on different substrates. This approach is inherently complex, and optical losses may reduce the device efficiency to below the level of existing three junction solar cell devices.
A third option is direct semiconductor bonding used to bond together solar cells that have been grown on different substrates. To date, bonds with adequate electrical conductivity and mechanical integrity for concentrated photovoltaics (CPV) applications do not exist.
Yet another solution is to mechanically stack sub-cells in such a manner that the entire stack of sub-cells converts incident light into electricity. Many different combinations of solar cells have been created using mechanical stacks. However, most mechanically stacked multijunction solar cells have poor thermal conductivity and optical coupling between the upper and lower subcells. In principle, this approach enables the use of a wide range of materials and therefore, very high conversion efficiencies. In practice, it is important to minimize the electrical resistivity and optical reflectivity losses at each bonded interface in the mechanical stack. For most applications, it is also important that heat from the upper solar cells can easily pass through the bonded interface and lower solar cells to reach a heat sink beneath the lower cells.
The foregoing examples of the related art and limitations related therewith are intended to be illustrative and not exclusive. Other limitations of the related art will become apparent to those of skill in the art upon a reading of the specification and a study of the drawings.
Exemplary embodiments presented in this disclosure can be more easily understood and further advantages and uses thereof more readily apparent, when considered in view of the description of the following figures in which:
a-j illustrate a fabrication sequence for fabricating a solar cell with an interfacial metallization pattern of spaced-apart pillars sandwiched between two solar cells.
In accordance with common practice, the various described features are not drawn to scale but are drawn to emphasize features relevant to the presented embodiments. Reference characters denote like elements throughout figures and text.
In the array-of-metal-pillars arrangement 230, each pillar may carry current (shown as arrows 240) collected from a small portion of the total area. As the spacing between pillars 230 is decreased, the total amount of current collected by each pillar decreases. Because of current-crowding, perimeter length of pillars affects R series. Therefore, the optimal shape may be a rectangular cross section, as shown. However, the pillars 230 may be any shape, such as circular, oval, triangular, discontinuous line segments, etc.
An interfacial grid line array (such as shown at 100) may appear to be optimal, because it maximizes the amount of metal at the interface with no apparent shadow loss, assuming a perfect geometry with no alignment or lithography related losses and substantially perfect normal-incident light. However, inclusion of shadow losses, and therefore, loss of light and subsequent current to bottom cell(s), due to lithography and alignment errors may favor an interfacial pillar geometry (such as shown at 200).
Specifically, a pillar arrangement has a similar or lower shadow loss than a grid line arrangement. For example, a 20×20 μm pillar is significantly less sensitive to alignment and fabrication errors than a 5 μm wide grid line. In particular, the sum of the errors may raise the effective shadow loss of each grid line significantly (from 5 μm to 8-11 μm in the above example). For a concentrator grid with a shadow loss of 4% in the top cell(s), the shadow loss of the bottom cell(s) may be in the order of 6 to 8.8%, for normal incidence light. For non-normal light (as from a lens), the shadow loss for the bottom cell may be much higher. Also, a 1 μm mis-alignment of grid lines reduces bonding area by 1 μm from 5 μm to 4 μm, which may result in a 20% reduction. However, for a 20×20 μm pillar, a 1 μm mis-alignment may have less shadow losses and maintain a good bonding area. Accordingly, the pillar arrangement will have a greater metal-to-metal overlap contact area for bonding. The shadow loss for non-normal light should be less for pillars than for grid lines under non-normal light conditions, such as from a lens. Furthermore, the 5 μm wide grid lines may be unrealistic. If 10 μm grid lines are required, then pillars will have a significantly smaller shadow loss.
Although most of the above summary concerns light in a normal-incidence geometry, it may be noted that non-normal light, as from a lens, will likely favor a pillar arrangement. Specifically, given substantially equal shadow loss for normal incidence, pillars should have lower shadow loss for off-normal incidence. At high concentrations, the range of angles can be large, up to approximately 42° for glancing incidence light. This embodiment may minimize electrical and optical losses for a configuration in which metal interconnects are used to carry electrical current from an upper cell(s) across a bonded interface to a lower cell(s).
a-j illustrate a fabrication sequence for fabricating a mechanically stacked, multijunction solar cell 600 with an interfacial metallization pattern of spaced-apart, metal-to-metal pillars 630 and 631, sandwiched between an upper solar cell 650 and a lower solar cell 660, including an optical coupling material 680 that may include a small air gap 670. During fabrication, a layer of photoresist 690 may be added to an optical coupling layer 680 and a top solar cell 650, as shown in
The geometry and dimensions are such that metal-to-metal bonds are made between the upper and lower contacts, and a filler-to-semiconductor or filler-to-filler bond is made over the rest of the interface. Because the metal-to-metal bonds carry electrical current between the upper and lower solar cells, the filler material does not need to perform this function. The filler material, and bonds to it, must, however, be optically transparent to light used by the lower solar cell(s) and have excellent thermal conductivity. In order to accomplish excellent thermal conductivity, the filler material must be in physical contact to the material above and below it. However, physical contact is sufficient, and a strong bond is not necessary. Also, to assist with fabrication limitations, a small air gap is tolerable optically. However, for good thermal conductivity between the solar cells, physical contact between the optical transparent materials and the upper and lower solar cells may be an improvement over an air gap.
While a number of exemplary aspects and embodiments have been discussed above, those of skill in the art will recognize certain modifications, permutations, additions and sub combinations thereof. For example, the metal pillars do not necessarily have to be metal. They can be of any material which can be bonded together with excellent electrical conductivity. It is therefore intended that the following appended claims and claims hereafter introduced are interpreted to include all such modifications, permutations, additions and sub-combinations as are within their true spirit and scope.
Although specific embodiments have been illustrated and described herein, it will be appreciated by those of ordinary skill in the art that any arrangement, which is calculated to achieve the same purpose, may be substituted for the specific embodiment shown. This application is intended to cover any adaptations or variations of the embodiments described herein. Therefore, it is manifestly intended that this invention be limited only by the claims and the equivalents thereof.
This application claims the benefit of and priority to U.S. Provisional Application No. 61/528,668, filed on Aug. 29, 2012 and entitled “MECHANICALLY STACKED MULTIJUNCTION SOLAR CELLS”, which is incorporated herein by reference in its entirety.
The United States Government has rights in this invention under Contract No. DE-AC36-08GO28308 between the United States Department of Energy and the Alliance for Sustainable Energy, LLC, the Manager and Operator of the National Renewable Energy Laboratory.
| Number | Date | Country | |
|---|---|---|---|
| 61528668 | Aug 2011 | US |
