BACKGROUND
The present invention relates to methods of forming interdigitated back contact (IBC) layers that may be used in photovoltaic (PV) devices. IBC solar cells are the most efficient Si solar cells today. In IBC solar cells, both junctions that collect positive and negative photocarriers are on the back of the cell, while the front has no grid and thus no shading losses. However, the dopant patterning required for these back junctions is challenging. Heavily doped p-type and heavily doped n-type regions should be separated by a highly resistive gap that may be made of intrinsic polysilicon or another dielectric material, otherwise the cell may be shunted. Further, the doped regions should be several hundred microns wide with well-defined edges, in order not to shunt across the undoped gap. Thus, reliable, precise, and inexpensive dopant patterning is needed for IBC solar cells.
Related art dopant patterning techniques include ion implantation through a mask, plasma immersion ion implantation (PIII) through a mask, and printed spin-on doping. However, these techniques can be costly, and may not be clean enough to enable high-efficiency IBC solar cells. Therefore, it would be advantageous to provide a simple, inexpensive, and precise dopant pattering technique that is compatible with solar cell processing.
SUMMARY
Exemplary embodiments of the invention provide methods of forming IBC layers. According to an aspect of the invention, a method includes forming a first layer having alternating regions of n-type amorphous hydrogenated silicon and p-type amorphous hydrogenated silicon on a second layer including intrinsic amorphous hydrogenated silicon. The method also includes annealing the first layer and the second layer such that dopants from the first layer diffuse into the second layer, and the first layer and the second layer crystallize into polysilicon. The annealing may be performed at a temperature greater than 650° C.
Before the forming of the first layer, the second layer may be deposited on a third layer comprising silicon oxide. The first layer may be formed by using a first shadow mask to deposit the regions of the n-type amorphous hydrogenated silicon, and using a second shadow mask to deposit the regions of the p-type amorphous hydrogenated silicon.
Before the annealing, the first layer may include gaps between adjacent regions of the n-type amorphous hydrogenated silicon and the p-type amorphous hydrogenated silicon. After the annealing, the second layer may include alternating regions of n-type polycrystalline silicon, intrinsic polycrystalline silicon, and p-type polycrystalline silicon. The first layer may be formed by plasma-assisted chemical vapor deposition (PECVD).
In an aspect, disclosed is a method for using a deposited mask to form a layer of polycrystalline silicon with alternating doping types in a photovoltaic device having interdigitated back contact layers, the method includes blanketing with i/a-Si:H+SiNx; patterning and etching; removing photoresist; adding p/a-Si:H+SiNx; blanketing with n/a-Si:H+SiNx; precrystallizing; crystallizing a-Si:H into polySi; and diffusing a dopant. In an embodiment, the method also includes blanketing with i/a-Si:H+SiNx+i/aSi:H.
In another aspect, disclosed is a method for using a shadow mask to form a layer of polycrystalline silicon with alternating doping types in a photovoltaic device having interdigitated back contact layers, the method includes depositing i/a-Si:H; patterning a n/a-SiH deposition; patterning a p/a-Si:H deposition; crystallizing; and diffusing a dopant.
Other objects, advantages, and novel features of the present invention will become apparent from the following detailed description of the invention when considered in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIGS. 1A, 1B, 1C, 1D, 1E, 1F, 1G, 1H, 1I and 1J depict a method of using a deposited mask to form a layer of polycrystalline silicon with alternating doping types according to a first embodiment of the invention. FIG. 1A depicts a step of blanketing i/a-Si:H+SiNx. FIG. 1B depicts a step of patterning and etching. FIG. 1C depicts a step of removing photoresist (PR) and adding p/a-Si:H+SiNx. FIG. 1D depicts a step of patterning. FIG. 1E depicts a step of etching and liftoff. FIG. 1F depicts a step of blanketing with n/a-Si:H+SiNx. FIG. 1G depicts a step of patterning. FIG. 1H depicts a step of etching and liftoff. FIG. 1I depicts a step of precrystallizing. FIG. 1J depicts a step of crystallizing a-Si:H into polySi and dopant diffusion.
FIGS. 2A, 2B, 2C, 2D, 2E, 2F, 2G, 2H, 2I and 2J depict a method of using a deposited mask to form a layer of polycrystalline silicon with alternating doping types according to the first embodiment of the invention. FIG. 2A depicts a step of blanketing i/a-Si:H+SiNx+i/aSi:H. FIG. 2B depicts a step of patterning and etching. FIG. 2C depicts a step of removing PR and adding p/a-Si:H+SiNx. FIG. 2D depicts a step of patterning. FIG. 2E depicts a step of etching and liftoff. FIG. 2F depicts a step of blanketing with n/a-Si:H+SiNx. FIG. 2G depicts a step of patterning. FIG. 2H depicts a step of etching and liftoff. FIG. 2I depicts a step of precrystallizing. FIG. 2J depicts a step of crystallizing a-Si:H into polySi and dopant diffusion.
FIGS. 3A, 3B, 3C, 3D, 3E, 3F, 3G, and 3H, depict a method of using yet another method of using a deposited mask to form a layer of polycrystalline silicon with alternating doping types according to the first embodiment of the invention. FIG. 3A depicts a step of blanketing i/a-Si:H+SiNx+i/aSi:H. FIG. 3B depicts a step of patterning and etching. FIG. 3C depicts a step of removing PR and adding p/a-Si:H+SiNx. FIG. 3D depicts a step of patterning. FIG. 3E depicts a step of etching and liftoff. FIG. 3F depicts a step of blanketing with n/a-Si:H+SiNx. FIG. 3G depicts a step of patterning. FIG. 3H depicts a step of etching and liftoff. FIG. 3I depicts a step of precrystallizing. FIG. 3J depicts a step of crystallizing a-Si:H into polySi and dopant diffusion.
FIGS. 4A, 4B, 4C, and 4D depict a method of using a shadow mask to form a layer of polycrystalline silicon with alternating doping types according to a second embodiment of the invention. FIG. 4A depicts a step of i/a-Si:H deposition. FIG. 4B depicts a step of patterned n/a-SiH deposition. FIG. 4C depicts a step of patterned p/a-Si:H deposition. FIG. 4D depicts a step of crystallization and dopant diffusion.
FIGS. 5A and 5B depict the passivation quality of intrinsic polycrystalline Si/SiO2 structures formed on n-Cz wafers according to exemplary embodiments of the invention. FIGS. 5A and 5B depict passivation quality of intrinsic poly-Si/SiO2 structures on KOH-planarized n-Cz wafers. FIG. 5A depicts intrinsic a-Si:H deposition temperature series. FIG. 5B depicts a-Si:H thickness series. The passivation is characterized by the iVoc after an 850° C. crystallization anneal and H-passivation by Al2O3.
FIGS. 6A and 6B depict characteristics of high performance p-type and n-type passivated contacts that are produced according to exemplary embodiments of the invention. FIG. 6A depicts symmetric sandwich structures as precursors to passivated contacts. FIG. 6B depicts iVoc of these structures after crystallization/drive-in anneal at various temperatures, followed by passivation with Al2O3 as H-source layer.
FIGS. 7A and 7B depict characteristics of an amorphous hydrogenated silicon film that was produced according to the second embodiment of the invention. FIG. 7A depicts a-Si:H film deposited on quartz through the BSF finger mask of the IBC cell. FIG. 7B depicts spatial edge fidelity of the BSF finger resolved by optical microreflectance scan.
FIG. 8 depicts an image of an IBC cell that was produced according to the first embodiment of the invention. FIG. 8 depicts an image of premetallized, lithography-defined IBC cell with BSF (blue region labelled as A), emitter (yellow region labelled as B), and gap (pink region labelled as C) clearly defined.
DETAILED DESCRIPTION
Exemplary embodiments of the invention provide methods of forming layers having alternating regions of n-type polycrystalline silicon and p-type polycrystalline silicon. A first layer having alternating regions of n-type amorphous hydrogenated silicon and p-type amorphous hydrogenated silicon is formed on a second layer of intrinsic amorphous hydrogenated silicon. The first layer may be formed by various methods, as discussed below with regard to the first and second embodiments. The first layer and the second layer are then annealed, such that dopants from the first layer diffuse into the second layer, and the first layer and the second layer crystallize into polysilicon.
In a first embodiment, a deposited mask is used to form a layer of heavily doped amorphous silicon with alternating doping types. As shown in FIG. 1A, the method begins with a structure that includes a crystalline silicon (cSi) substrate on which a silicon oxide (SiOx) layer that serves as a tunneling oxide layer is formed. The SiOx layer may have a thickness of approximately 1.5 nm, and may be grown thermally at a temperature of 700° C. Plasma-assisted chemical vapor deposition (PECVD) may be used to perform the various deposition steps discussed below.
In a first step shown in FIG. 1A, an intrinsic amorphous hydrogenated silicon (i/a-Si:H) layer is deposited on the SiOx layer. The i/a-Si:H layer may have a thickness of at least 20 nm. This layer is then covered with a deposited film such as silicon nitride (SiNx), Al2O3, SiOx, or combinations thereof. The deposited film is shown in FIG. 1A as a first SiNx layer. In a second step shown in FIG. 1B, a first photoresist (PR) is patterned via any suitable lithographic technique on a portion of the first SiNx layer. The exposed portion of the first SiNx layer is etched via wet chemistry such as dilute hydrofluoric acid (HF), and the PR is subsequently removed by a solvent such as acetone.
In a third step shown in FIG. 1C, the first PR is completely removed, and a p-type amorphous hydrogenated silicon (p/a-Si:H) layer is deposited on the structure, where portions of the p/a-Si:H layer are in direct contact with the i/a-Si:H layer. The p/a-Si:H layer may be doped with any suitable p-type dopant, such as B. The p/a-Si:H layer may have a thickness of at least 5 nm, and may have a dopant concentration of approximately 1021 cm−3. A second SiNx layer is then deposited on the p/a-Si:H layer.
In a fourth step shown in FIG. 1D, a second PR is patterned on portions of the second SiNx layer to protect the part of the underlying p/a-Si:H layer that is in direct contact with the i/a-Si:H layer. In a fifth step shown in FIG. 1E, portions of the second SiNx layer, the p/a-Si:H layer, and the first SiNx layer that are not covered by the second PR are etched by any suitable method, such as exposing the structure to reactive ion etching (RIE) or wet chemistry (HF). The second PR is then removed.
In a sixth step shown in FIG. 1F, an n-type amorphous hydrogenated silicon (n/a-Si:H) layer is deposited on the structure, and a third SiNx layer is formed on the n/a-Si:H layer. The n/a-Si:H layer may be doped with any suitable n-type dopant, such as P. The n/a-Si:H layer may have a thickness of at least 5 nm, and may have a dopant concentration of approximately 1021 cm−3. In a seventh step shown in FIG. 1G, a third PR is patterned on portions of the third SiNx layer to protect the part of the underlying n/a-Si:H layer that is in direct contact with the i/a-Si:H layer. In an eighth step shown in FIG. 1H, the portions of the third SiNx layer and the n/a-Si:H layer that are not covered by the third PR are etched by any suitable method, such as exposing the structure to reactive ion etching (RIE) or wet chemistry (HF). The third PR is then removed.
In a ninth step shown in FIG. 1I, the remaining sections of the second and third SiNx layers are removed by any suitable method, such as etching with HF. In a tenth step shown in FIG. 1J, the structure is annealed at high temperatures, such as greater than 650° C., which effuses hydrogen and causes dopants from the n/a-Si:H and p/a-Si:H regions to diffuse into the i/a-Si:H layer, while all of the remaining layers in the structure (except for the SiOx and cSi layers) crystallize into polysilicon. This forms alternating regions of n-type polycrystalline silicon (n/polySi) and p-type polycrystalline silicon (p/polySi). The n-type and p-type regions may have the same or different widths. The SiOx layer allows for the high temperatures that are used during annealing to drive the dopants into the i/a-Si:H layer.
The deposited mask approach of the first embodiment may be modified in various ways. For example, as shown in FIGS. 2A through 2J, additional masking layers may be formed on the SiNx layers. The selection of the material for the masking layers is subject to the desired chemistries to be used to etch, since etching selectivity is material dependent. For example, photoresist does not etch with RIE, while a-Si:H does etch with RIE. Similarly, a-Si:H does not etch with HF, while SiNx, Al2O3, and SiOx do etch with HF. The additional masking layers shown in FIG. 2 are intrinsic amorphous silicon (i/a-Si:H) layers. The steps performed in FIGS. 2A-2J may be the same as the steps described above with regard to FIGS. 1A-1J, respectively.
Further, as shown in FIGS. 3A-3H, the patterning of the doped layers may be modified to provide an undoped gap between adjacent doped layers. For example, the second PR that is deposited in the fourth step shown in FIG. 3D may be widened, such that portions of the first SiNx layer and the additional masking layer remain. The third PR is used as a mask in the seventh step shown in FIG. 3G, such that after the portions are etched away during the eighth step shown in FIG. 3H, a space is formed between the n/a-Si:H and p/a-Si:H regions. In the tenth step shown in FIG. 3J, the structure is annealed at high temperatures, such as greater than 650° C., which effuses hydrogen and causes dopants from the n/a-Si:H and p/a-Si:H regions to diffuse into the i/a-Si:H layer, while the n/a-Si:H, p/a-Si:H, and i/a-Si:H layers crystallize into polysilicon. This forms alternating regions of n-type polycrystalline silicon (n/polySi), intrinsic polycrystalline silicon (i/polySi), and p-type polycrystalline silicon (p/polySi). The steps performed in FIGS. 3A-3J may be the same as the steps described above with regard to FIGS. 1A-1J, respectively.
In a second embodiment, a shadow mask is used to form a layer of heavily doped amorphous silicon with alternating doping types. For example, the shadow mask may be made of a perforated metal sheet or a Si wafer that is patterned by laser-initiated etching. As shown in FIG. 4A, the method begins with a structure that includes a crystalline silicon (cSi) substrate on which a silicon oxide (SiOx) layer that serves as a tunneling oxide layer is formed. The SiOx layer may have a thickness of approximately 1.5 nm, and may be grown thermally at a temperature of 700° C. PECVD may be used to perform the various deposition steps discussed below.
In a first step shown in FIG. 4A, an intrinsic amorphous hydrogenated silicon (i/a-Si:H) layer is deposited on the SiOx layer. The i/a-Si:H layer may have a thickness of at least 20 nm. In a second step shown in FIG. 4B, a first shadow mask (n/shadowmask) is used to deposit an n-type amorphous hydrogenated silicon (n/a-Si:H) layer on regions of the i/a-Si:H layer. The n/a-Si:H layer may be doped with any suitable n-type dopant, such as P. The n/a-Si:H layer may have a thickness of at least 5 nm, and may have a dopant concentration of approximately 1021 cm−3. In a third step shown in FIG. 4C, a second shadow mask (p/shadowmask) is used to deposit a p-type amorphous hydrogenated silicon (p/a-Si:H) layer on other regions of the i/a-Si:H layer. The p/a-Si:H layer may be doped with any suitable p-type dopant, such as B. The p/a-Si:H layer may have a thickness of at least 5 nm, and may have a dopant concentration of approximately 1021 cm−3. The p/a-Si:H regions alternate with the n/a-Si:H regions, and spaces having a width between approximately 50 μm and approximately 100 μm may be formed between adjacent p/a-Si:H and n/a-Si:H regions. Although the shadow masks are shown as being spaced apart from the i/a-Si:H layer in FIGS. 4B and 4C, the shadow masks may be in direct contact with the i/a-Si:H layer during deposition. The shadow masks may provide an edge roughness and pattern fidelity of approximately 10 μm.
In a fourth step shown in FIG. 4D, the structure is annealed at high temperatures, such as greater than 650° C., which effuses hydrogen and causes dopants from the n/a-Si:H and p/a-Si:H regions to diffuse into the i/a-Si:H layer, while the n/a-Si:H, p/a-Si:H, and i/a-Si:H layers crystallize into polysilicon. This forms alternating regions of n-type polycrystalline silicon (n/polySi), intrinsic polycrystalline silicon (i/polySi), and p-type polycrystalline silicon (p/polySi). The n-type and p-type regions may have the same or different widths. The SiOx layer allows for the high temperatures that are used during annealing to drive the dopants into the i/a-Si:H layer.
In both embodiments, the resulting structure may then receive an atomic layer deposited (ALD) Al2O3 layer with an additional SiNx layer deposited by PECVD on the textured front surface. A subsequent forming gas anneal (FGA) at approximately 400° C. may further passivate the structure by hydrogenation and field effects. After removal of the sacrificial Al2O3 layer from the back, the IBC structure may be metallized by thermal evaporated Al through a Si wafer shadow mask or a lithographically defined patterned photoresist. The emitter and back surface field (BSF) contacts, as well as the intrinsic poly-Si gap layers, may be optimized separately using symmetric test structures. These may be deposited on planarized and tunnel-oxidized n-Cz wafers.
To test the methods discussed above, symmetric lifetime structures were grown and measured. Before depositing doped source layers for n- and p-type passivated contacts, intrinsic poly-Si/SiO2 structures with high surface passivation were developed. These underwent the same thermal treatments and passivation by ALD Al2O3 as the sandwich-doped layers. The best degree of passivation of i/poly-Si/SiO2 was obtained using a high hydrogen dilution H2/SiH4 ratio during PECVD deposition of i/a-Si:H at a low rate less than 0.5 Å/s. Other factors are the deposition temperature and the i/a-Si:H film thickness, as summarized in FIG. 5. Thicker a-Si:H layers produce better surface passivation but develop blisters during 850° C. annealing. These do not affect the implied open-circuit voltage (iVoc) initially but cause performance loss upon further cell processing. In contrast, depositing a-Si:H at a temperature greater than 400° C. already produces films with blisters (see inset of FIG. 5A), resulting in a low iVoc.
After selecting the i/poly-Si process with a 20 nm thick film deposited at 300° C. under optimal PECVD conditions, heavily doped n-type or p-type a-Si:H was deposited. The dopants were then driven into the rest of the poly-Si by thermal diffusion, with a goal of producing the average dopant concentrations within poly-Si of approximately 2×1020 cm−3. As seen from FIG. 6, high performance p-type and n-type passivated contacts can be produced by this method. Notably, the typically more problematic p-type passivated contacts exhibit an iVoc over 700 mV after this processing. This might be associated with less accumulation of B into the tunneling oxide after diffusion through the poly-Si layer. The n-type contact has a remarkably high iVoc after annealing at 650° C. but deteriorates with increasing the annealing temperature. This is somewhat unexpected since uniformly doped n-type contacts after annealing at 850° C. and passivation typically show an iVoc greater than 730 mV peaking near the 850° C. anneal temperature. The above effect might be due to an overall amount of P that is too low, leading to a reduced field effect passivation.
In-plane 2D spatial patterning of B and P dopants was also investigated. For this purpose, dopant-source layers shown in FIG. 6A were deposited by spatially-masked PECVD using (i) a mechanical mask as described above with regard to the second embodiment and (ii) photolithographically patterned, sacrificial dielectric layers on top of the blanket intrinsic a-Si:H as described above with regard to the first embodiment.
To ensure the absence of edge effects during PECVD, patterned Si wafers were used as shadow masks for the dopant source overlayers in an example based on the second embodiment. FIG. 7A shows the resulting pattern of an approximately 10 nm thick a-Si:H layer deposited on quartz through the n-type finger (BSF) mask of the 2×2 cm2 IBC cell. In this example, the a-Si:H finger edge fidelity was approximately 17 measured as the exponential decay length of the optical microreflectance scan (FIG. 7B). Taking into account that the poly-Si resistivity increases highly nonlinearly with doping below 1019 cm−3, this suggests that the 100 μm wide intrinsic poly-Si gap will remain insulating between the n-type and p-type fingers. This is important because the weak shuntage criterion for an IBC cell (Rshunt/Rchar>100) holds only if the sheet resistivity of the i/poly-Si gap is on the order of MSΩ/□, as the total gap length even in a 2×2 cm2 cell is about 80 cm. Note that unlike the diffused IBC cells, the poly-Si contact based IBC cell is easily shunted due to the ohmic behavior of junctions between heavily doped and defective i-type, n-type, and p-type poly-Si layers.
Patterning dopant source layers was also done using dielectric layers to protect intrinsic a-Si:H during doped a-Si:H PECVD, with lithography steps in between n-type and p-type depositions in an example based on the first embodiment. As shown in FIG. 8, this technique provides excellent edge fidelity. However, this process is more complex than the high-throughput mechanical masking of the second embodiment.
The foregoing disclosure has been set forth merely to illustrate the invention and is not intended to be limiting. Since modifications of the disclosed embodiments incorporating the spirit and substance of the invention may occur to persons skilled in the art, the invention should be construed to include everything within the scope of the appended claims and equivalents thereof.
Patent Application
20180374984
