Transparent Conductive Oxide In Silicon Heterojunction Solar Cells

Abstract

Devices and methods for reducing optical losses in transparent conductive oxides (TCOs) used in silicon heterojunction (SHJ) solar cells while enhancing series resistance are disclosed herein. In particular, the methods include reducing the thickness of TCO layers by about 200% to 300% and depositing hydrogenated dielectric layers on top to form double layers of antireflection coating. It has been discovered that the conductivity of a thin TCO layer can be increased through a hydrogen treatment supplied from the capping dielectric during the post deposition annealing. The optimized cells with ITO/SiO.sub.x:H stacks achieved more than 41 mA/cm.sup.2 generation current on 120-micron-thick wafers while having approximately 100 Ohm/square sheet resistance. Further, solar cells and methods may include integration of ITO/SiO.sub.x:H stacks with Cu plating and use ITO/SiN.sub.x/SiO.sub.x triple layer antireflection coatings. The experimental data details the improved optics and resistance in cell stacks with varying materials and thicknesses.

Claims

1. A solar cell, comprising: a silicon base layer; an emitter layer disposed on a first side of the silicon base layer, wherein the emitter layer comprises amorphous silicon; a first antireflective coating layer disposed on the emitter layer, wherein the first antireflective coating layer comprises a transparent conducting oxide; and a second antireflective coating layer disposed on the first antireflective coating layer, wherein the second antireflective coating layer comprises a hydrogenated silicon oxide.

2. The solar cell of claim 1, wherein the first antireflective coating layer is hydrogenated by the second antireflective coating layer after annealing such that conductivity of the first antireflective coating layer is increased.

3. The solar cell of claim 2, wherein conductivity of the first antireflective coating layer is increased by about 20% to about 40%.

4. The solar cell of claim 1, wherein the second antireflective coating layer has an atomic percentage of hydrogen between about 10% and about 40%.

5. The solar cell of claim 4, wherein the atomic percentage of hydrogen is about 25%.

6. The solar cell of claim 1, further comprising: a conducting grid disposed on the second antireflective coating layer.

7. The solar cell of claim 6, wherein the conducting grid comprises at least one of silver, copper, and nickel.

8. The solar cell of claim 1, wherein the silicon base layer includes a crystalline-Si substrate.

9. The solar cell of claim 1, wherein the silicon base layer includes an epitaxially formed crystalline-Si thin film.

10. A method for fabricating a solar cell, comprising: (a) preparing a silicon (Si) base layer; (b) depositing an emitter layer on a first surface of the Si base layer, wherein the emitter layer comprises an amorphous silicon; (c) depositing a first antireflective coating layer on the emitter layer, wherein the first antireflective coating layer comprises a transparent conducting oxide; and (d) depositing a second antireflective coating layer on the first antireflective coating layer, wherein the second antireflective coating layer comprises a hydrogenated silicon oxide.

11. The method of claim 10, further comprising: (e) annealing the solar cell such that the first antireflective coating layer is hydrogenated by the second antireflective coating layer thereby increasing conductivity of the first antireflective coating layer.

12. The method of claim 11, wherein conductivity of the first antireflective coating layer is increased by about 20% to about 40%.

13. The method of claim 10, wherein the second antireflective coating layer has an atomic percentage of hydrogen between about 10% and about 40%.

14. The method of claim 13, wherein the atomic percentage of hydrogen is about 25%.

15. The method of claim 10, further comprising: depositing a conducting grid on the second antireflective coating layer.

16. The method of claim 15, wherein the conducting grid comprises at least one of silver, copper, and nickel.

17. The method of claim 10, wherein the silicon base layer includes a crystalline-silicon substrate.

18. The method of claim 10, wherein the silicon base layer includes an epitaxially formed crystalline-silicon thin film.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0032] FIG. 1 is a representation of one example of a layer stack of a Si heterojunction solar cell, in accordance with the present disclosure.

[0033] FIG. 2 is a graph showing experimental results for sheet resistance as a function of oxygen flow and sputtering pressure across three different cell stack arrangements, in accordance with the present disclosure.

[0034] FIG. 3 is a graph showing experimental results for generation current as a function of oxygen flow and sputtering pressure across three different cell stack arrangements, in accordance with the present disclosure.

[0035] FIG. 4 is a graph showing experimental results for external quantum efficiencies as a function of wavelength across three different cell stack arrangements, in accordance with the present disclosure.

[0036] FIG. 5 is a graph showing experimental results for external quantum efficiencies as a function of wavelength across two different cell stack arrangements, in accordance with the present disclosure.

DETAILED DESCRIPTION

[0037] The present disclosure provides systems and methods for enhancing the performance of Si heterojunction (SHJ) solar cells. Further, the disclosure provides methods for improving the optical response of SHJ solar cells through the use of hydrogenated Si oxide (SiO.sub.x:H) layers adjacent to transparent conducting oxide (TCO) layers to allow for hydrogen transfer between the antireflective coating layers, thereby increasing the conductivity of the transparent conducting oxide layer.

[0038] As shown in FIG. 1, a SHJ solar cell stack 100 comprises a silicon base layer 102 at the center of the SHJ solar cell stack 100. The silicon base layer 102 includes a first surface 104 and a second surface 106 and may be made of crystalline Si (c-Si) or any other suitable substrate, such as n-type Czochralski (CZ) silicon wafer, for example. Top (or front) layers 108 are disposed on the first surface 104 of the silicon base layer 102. Rear (or bottom) layers 110 are disposed on the second surface 106 of the silicon base layer 102. The top layers 108 absorb incident sunlight shining upon the SHJ solar cell stack 100 and may include antireflective layers to increase any absorption occurring therethrough. The rear layers 110 may include rear mirrors to reflect the sunlight back toward the junction. Adjacent the first surface 104 of the Silicon base layer 102, a first emitter layer 112 may be disposed. A second emitter layer 114 may be disposed on the silicon base layer 102 adjacent the second surface 106. The first and second emitter layers 112, 114 may be made of amorphous Si (a-Si) or any other suitable material for forming the heterojunction of the solar cell. As, depicted in FIG. 1, on top of and below the first and second emitter layers 112, 114, first and second antireflective coating layers 116, 118 may be deposited. The first and second antireflective coating layers 116, 118 may be a transparent conductive oxide (TCO) material, such as indium tin oxide (ITO), GaInO (GIO), GaInSnO (GITO), ZnInO (ZIO), and/or ZnInSnO (ZITO). It is also contemplated that the first and second antireflective coating layers 116, 118 may be different thicknesses and may be optimized to provide different responses to impingent photons. It is also contemplated that the first and second antireflective coating layers 116, 118 may be doped with different materials or may be different materials.

[0039] Third and fourth antireflective coating layers 120, 122, respectively, may be deposited onto outer surfaces of first and second antireflective coating layers 116, 118 on either side of the SHJ solar cell stack 100. The third and fourth antireflective coating layers 120, 122 may comprise a hydrogenated Silicon oxide (SiO.sub.x:H) layer. Alternatively, the third and fourth antireflective coating layers 120, 122 may comprise other suitable hydrogenated dielectric materials. After deposition of the all of the antireflective coating layers, the SHJ solar cell stack 100 may then be annealed. Advantageously, the annealing process may cause hydrogen from the third and fourth antireflective coating layers 120, 122 (i.e., hydrogenated dielectric layers) 120,122 to diffuse into the TCO of the first and second antireflective coating layers 116, 118. This hydrogenation of the TCO layers provided in the systems and methods disclosed herein provides the SHJ solar cell stack 100 with enhanced properties, such as increased conductivity, optical response, and efficiency. Further details of this beneficial discovery are described below and in the experiments section.

[0040] The method may include preparing ITO/SiO.sub.x:H stacks on the front and on the rear sides of a SHJ cell to reduce optical losses without compromising with series resistance or recombination. This structure is advantageous because SiO.sub.x:H on the front may serve as a second antireflection coating to allow thinner TCO absorbing less light. At the same time, conductivity of a thin TCO can be maintained at a sufficient level due to the effect of hydrogen treatment, where hydrogen is supplied from SiO.sub.x:H layer during post-deposition annealing. At the rear side thin-ITO/SiO.sub.x:H/Ag stack may be a superior rear mirror when compared to a conventional thick-TCO/Ag stack with less parasitic absorption in 800-1200 nm range.

[0041] The effect of hydrogen doping of ITO from hydrogenated plasma-enhanced vapor deposition (PECVD) films was recently reported by Ritzau et al., where the observed increase of ITO conductivity during a post deposition annealing in a standard SHJ cell was attributed to hydrogen effusion from underlying a-Si films. Hydrogen was also reported to promote crystallinity of certain TCOs, such as indium oxide (IO) and indium cerium oxide (ICO), for example, if supplied to the chamber during their deposition. Thus, it has been discovered that hydrogen treatment of post deposited ITO films not only increases free carrier density, but also may improve mobility.

[0042] Finally, the SHJ solar cell stack 100 of FIG. 1 may include a conductive grid 124 deposited on the top outer surface of third antireflective coating layer 120. This conductive grid 124 may be made of any conductive material such as Ag, Cu, and/or Ni, for example. At the bottom of the SHJ solar cell stack 100, there may be included a rear contact 126. This rear contact 126 may provide a conductive material for transferring electricity from the SHJ solar cell stack 100 after conversion from light. The rear or bottom layers 110 in the SHJ solar cell stack 100 may all act as rear mirrors for the solar cell. Thus, in one aspect, the present disclosure provides a system and method for enhanced operation of a SHJ solar cell that is structured similar to and as described.

[0043] In another aspect, the methods may include reducing the thickness of TCO layers by about 200% to 300% and depositing hydrogenated dielectric layers on top to form double layers of antireflection coating. It has been discovered that the conductivity of a thin TCO layer can be increased through a hydrogen treatment supplied from the capping dielectric during the post deposition annealing. The optimized cells with ITO/SiO.sub.x:H stacks achieved more than 41 mA/cm.sup.2 generation current on 120-micron-thick wafers while having approximately 100 Ohm/square sheet resistance. Further, the solar cells and methods disclosed herein may include integration of ITO/SiO.sub.x:H stacks with Cu plating and use ITO/SiN.sub.x/SiO.sub.x triple layer antireflection coatings. The experimental data described in detail below shows the improved optics and resistance in cell stacks with varying materials and thicknesses.

[0044] The disclosed systems and methods may be easily incorporated into the production of Si heterojunction solar cells, thereby enhancing the operational performance of the solar cell.

EXAMPLES

[0045] The following Examples are provided in order to demonstrate and further illustrate certain embodiments and aspects of the present disclosure and are not to be construed as limiting the scope of the disclosure.

Example 1

[0046] The following section details the results and protocol undertaken to enhance the conductivity and improve the optical response in Si heterojunction solar cells through the use of a second hydrogenated Si oxide antireflective coating layer capping the first transparent conducting oxide antireflective coating layer. This experiment was conducted to evaluate the combined effects of varying the thicknesses of the stack layers while using hydrogenated Si oxide antireflective coating layers to increase the conductivity of the adjacent transparent conducting oxide antireflective coating layer.

[0047] The starting ITO films may have a thickness of about 30-50 nm and a relatively high oxygen content with very low carrier density. Previously, such films with 150-200 nm thickness would be used at the rear side of the cell as IR mirrors. The experimental results confirm that the conductivity of these films is increased after SiO.sub.x:H deposition and annealing. This approach may achieve around 41-41.3 mA/cm.sup.2 generation current (J.sub.gen) on 120 μm thick double side textured wafer or base layer, while keeping the sheet resistance (R□) of the front ITO film equal to 100 Ω/□.

[0048] In this experiment, SHJ cells with metal electrodes deposited on ITO were coated with a PECVD SiO.sub.x:H film. Although not preventing probing the cells, this approach may make soldering Cu ribbons to the busbars challenging. To overcome these potential challenges, an additional patterning process may be used in a solar cell with an ITO/SiO.sub.x:H stack in order to form a front metal grid. In a non-limiting example method for producing one of the disclosed SHJ solar cell arrangements, SiO.sub.x:H layers may be patterned by the lift off of the screen printed resist and Cu may be plated in the openings by light induced plating.

[0049] To optimize ITO/SiO.sub.x:H stacks, J.sub.gen and R□ were measured on complete SHJ solar cells. SHJ cells with the junction on the front side were used. Therefore sheet conductivity between metal fingers on the front was only due to the ITO film. SHJ cells were made on n-type CZ wafers with 3-4 Ω-cm resistivity. The thickness of the wafers after texturing was approximately 120 μm. Amorphous Si films were deposited on both sides of the wafers followed by ITO sputtering and sputtering of the rear Ag contact. A front metal grid was formed either by screen printing of Ag paste to make 10×10 cm.sup.2 cells or by photolithographic patterning of the sputtered Ag film to make 1×4 cm2 cells. Screen printed cells had three 1 mm wide busbars and approximately 100-μm-wide fingers. Photolithographically patterned cells had 20 μm wide fingers and featured a busbar around the perimeter of the cell defining its area. Finally, the cells were capped with dielectric layers and annealed in a muffle furnace at 200° C. for 30 minutes.

[0050] The ITO films were reactively sputtered at room temperature using a MRC 944 tool with a DC power supply. An ITO target with 90/10 In.sub.2O.sub.3/SnO.sub.2 ratio was used. The oxygen flow and deposition pressure were varied during sputtering. The a-Si and SiO.sub.x:H layers were deposited using a Applied Materials P-5000 PECVD tool. The SiO.sub.x:H layers were previously measured by RBS to have approximately 25% hydrogen by the number of atoms. The thicknesses of SiO.sub.x:H and ITO layers were measured using a Woollam M-2000 ellipsometer. The optical modelling was done using Opal 2 and Ray Tracer software developed by PVLightHouse. In simulations, the optical constants of the hydrogenated ITO and SiO.sub.x:H layers measured by ellipsometry were used. The results of the modelling suggested that the optimum combination of ITO and SiO.sub.x:H was about 40-50 nm and 90-110 nm, respectively. These optimum thicknesses were based on a minimum conductivity of the ITO layer as required by the particular grid design. To measure the solar cell performance IV curves (through scanning an applied voltage across the cells and obtaining the current response) of the cells were determined on a flash IV tester from Sinton Instruments using an NREL certified cell as a reference.

Results and Discussion

[0051] 3.1. EQE and R□ of SHJ Cells with ITO/SiO.sub.x:H Stacks

[0052] FIGS. 2-3 show the experimental results of R□ and J.sub.gen as a function of oxygen flow and sputtering pressure varied during ITO sputtering for the font emitter SHJ cells having an 80 nm ITO layer and 50 nm ITO/100 nm SiO.sub.x:H stacks.

[0053] FIG. 2 compares the R□ measured on the p-side-up SHJ solar cells having a 80 nm ITO layer with the R□ measured on the cells having a 50 nm ITO/100 nm SiO.sub.x:H stack on the front side. On the rear side, the cells had a thick-ITO/Ag stack optimized to work as an IR mirror. The results show that capping an ITO layer with a layer of SiO.sub.x:H can increase the conductivity of the TCO layer 2-5 times, depending on the initial oxygen content in the ITO layer and the sputtering pressure at deposition. The same results were observed when the ITO layer was capped by SiN.sub.x:H and a-Si:H layers. These results further advance and are in agreement with the data obtained by Ritzau et al., for the films doped from underlying a-Si:H films. An increase in ITO conductivity of about 20-40% was observed after annealing when the ITO layer is sputtered on a-Si:H films versus being sputtered on glass or crystalline silicon.

[0054] FIG. 3 shows the resulting J.sub.gen as measured on the p-side-up SHJ cells prepared for this experiment. The 50 nm ITO layers that were capped with SiO.sub.x:H demonstrated a superior optical performance when sputtered with a high oxygen dilution. The cells with the optimum ITO layer, such as those sputtered at 5 mTorr pressure and 2 sccm O.sub.2 flow, for example, achieved a 40.5 mA/cm.sup.2 J.sub.gen and about 90-100 Ohm/□ R□. This result is approximately 1 mA/cm.sup.2 higher than the J.sub.gen of the cells with a single optimized 80 nm ITO layer, with almost no loss in sheet resistance. Such R□ is sufficient for use with most front grid designs with an acceptable series resistance loss.

[0055] Similar performance may be achieved by using a low oxygen ITO film with about 50 Ω/□ R□ at 80 nm thick, capped with a low index dielectric not containing hydrogen, such as thermally evaporated MgO or RF sputtered SiO.sub.x, for example. In this arrangement, the ITO layer is not excessively doped by hydrogen and can provide the necessary transparency. The effect of the hydrogen treatment on the material properties of the ITO layer may guide the design of ITO/dielectric stacks.

Example 2

[0056] Further improvement of the optical response of SHJ cells is possible through reducing parasitic IR absorption at the rear side of the cell. In this experiment, a conventional 200 nm ITO/Ag stack was replaced with a 20 nm ITO/200 nm SiO.sub.x:H/Ag stack. For making a good contact with Ag, SiO.sub.x:H was immersed in HF solution prior to Ag sputtering. Without being bound by theory, it is theorized that this process created local openings in the PECVD SiO.sub.x:H due to its porous structure, enabling the SiO.sub.x:H to make good contact with the sputtered Ag for EQE measurement. Thus, the EQE data represents mostly the ITO/SiO.sub.x:H/Ag cell stack structure with some fraction of ITO/Ag. Note that such structures were prepared only for optical measurements and no solar cells were fabricated using this process flow. Alternatively, a better patterning method could produce functional solar cells with ITO/SiO.sub.x:H/Ag stacks at the rear side.

3.2. Simulation of SHJ Cells with ITO/SiN.sub.x/SiO.sub.x Stacks.

[0057] FIG. 4 shows the results of the EQE with respect to wavelength of p-side-up SHJ cells across three different optical structures.

[0058] FIG. 4 compares the measured EQE of a conventional SHJ cell with the EQE of the cells having ITO/SiO.sub.x:H stacks on the front as well as cells having ITO/SiO.sub.x:H stacks on both the front and rear sides. The results show that a thinner ITO layer together with a lower reflectance double layer of antireflection coating allows for a better response in the 300-500 nm and 700-1050 nm wavelength ranges. Moreover, improving the rear IR mirror allowed for a better IR response in the 1000-1200 nm range, which added another 0.7 mA/cm.sup.2 to the overall J.sub.gen. Interestingly, improving a rear mirror in this experiment yielded equivalent results to using a thicker wafer. For example, SHJ cells with a 20 nm ITO/SiO.sub.x rear mirror made on a 120 μm wafer had a similar IR performance (˜60-70% EQE at 1100 nm) as compared to a 190 μm cell having a conventional 200 nm ITO/Ag stack.

[0059] For some cell designs, optical losses may be further reduced by using even thinner ITO layers. For example, in the cells with 10 μm fingers, the spacing between the fingers may be less than 1 mm, which allows for the use of TCOs with about 200-300 Ω/□ R□. Such fingers may be produced, for example, by laser patterning and Cu plating. As another example of further loss reduction, in cells with a rear emitter, lateral conductivity between the fingers is increased by the conductivity of the bulk of the wafer. In SHJ cells at a maximum power point operation, the bulk can have about 1×10.sup.15 cm.sup.−3 to about 3×10.sup.15 cm.sup.−3 excess carrier density, which provides approximately 100-300 Ω/□ lateral conductivity in 120 μm wafers.

[0060] Thus, the thickness of the ITO layers can be reduced even more, down to about 10-20 nm. The limit for the thickness reduction may depend on the interaction between the metal grid and the doped a-Si layer, as well as other considerations, such as metal contact adhesion and reliability, for example. In an alternative extreme design, the TCO layer can be eliminated altogether and completely excluded from optical losses.

[0061] However, to achieve the desired antireflective properties, in addition to SiO.sub.x, cells with very thin (<50 nm) ITO layers may use dielectrics with higher index, such as a transparent low temperature SiN.sub.x, for example. FIG. 5 is a graph showing the simulated EQE results with respect to wavelength for p-side-up SHJ cells and compares the optical response in 50 nm ITO/SiO.sub.x and 20 nm ITO/SiN.sub.x/SiO.sub.x cell stack structures. FIG. 5 shows the simulated EQE curves for the cell structures with 10 nm ITO/SiN.sub.x/SiO.sub.x stacks. The simulation suggests that the J.sub.gen of such structures may reach 42 mA/cm.sup.2.

3.3. Solar Cell Performance

[0062] Table I below summarizes the performance of the best SHJ solar cells with ITO/SiO.sub.x:H stacks on the front side. The only observed effects on cell performance when capping the ITO layer with the SiO.sub.x:H layer were the sheet resistance reduction and the improved optical response discussed above. The full potential of the stack was realized in 1×4 cm.sup.2 cells having 20 μm fingers patterned photolithographically. The front grid shading in these cells comprised approximately 2%, allowing a J.sub.SC greater than 40 mA/cm.sup.2. Note also that these cells had optimized a-Si layers on the front side allowing for a better response in the visible range. The rear side of the cells had a conventional 200 nm ITO/Ag stack.

TABLE-US-00001 TABLE I The parameters for the record SHJ cells, which were using ITO/SiO.sub.x stacks. V.sub.oc J.sub.gen J.sub.sc pFF FF η Cell type (mV) (mA/cm.sup.2) (mA/cm.sup.2) (%) (%) (%) 10 × 10 cm.sup.2, 727 40.6 37.9 81.6 78.2 21.5.sup.1 screen printing 1 × 4 cm.sup.2 739 41.3 40.6 82.0 78.0 23.4.sup.2 photolith., floating busbar .sup.1confirmed at NREL, .sup.2in-house measurement
3.4. Integration with Cu Plating

[0063] Solar cells with a SiO.sub.x:H layer deposited after the formation of the front metal grid could make soldering the cells problematic. The process flow described below resolves this issue by using SiO.sub.x as a mask for Cu plating. To pattern SiO.sub.x, a resist grid was screen printed on the ITO layer and followed by the deposition of SiO.sub.x on top of the resist. Next, the resist was lifted using an alkaline solution to make openings in the dielectric. Thus, a dielectric mask for Cu plating was formed. With this method, for example, Ni/Cu/Sn could be plated in the openings using light- or field-induced plating. In this experiment, an alternative seeding method was used, which allowed direct Cu/Sn plating on the ITO layer with superior adhesion. The cells processed this way, achieved 20% efficiency on 153 cm.sup.2 area with 729 mV V.sub.OC, 36.1 mA/cm.sup.2 J.sub.SC and 75.9% FF. Thus, presently disclosed are new methods to improve optical response of SHJ cells and to integrate these stacks with Cu plating.

[0064] Thus, the present disclosure provides systems and methods for improving the optical response of SHJ cells using ITO/SiO.sub.x:H stacks and integrating these stacks with Cu plating. Further, conductivity of the sputtered ITO films may be increased by treating the transparent conducting oxide films with hydrogen supplied from PECVD dielectrics. Additionally, SHJ cells with optimized ITO/SiO.sub.x stacks on both sides of the base layer in the solar cell produced about 41 mA/cm.sup.2 J.sub.gen with 100 Ω/□ sheet resistance and some structures achieved 41.3 mA/cm.sup.2. Using transparent conducting oxide antireflective coating layers capped with hydrogenated dielectric layers in ITO/SiN.sub.x/SiO.sub.x stacks with advanced cell designs allowing 300 Ω/□ sheet resistances can further increase J.sub.gen to around 42 mA/cm.sup.2. While the mechanism for increasing conductivity in the antireflective coating layers of transparent conducting oxide capped with hydrogenated dielectrics may not be fully understood, the data suggests that further reducing the thickness of the transparent conducting oxide antireflective coating layers may aid in fully realizing the potential of ITO/SiO.sub.x:H stacks.

INDUSTRIAL APPLICABILITY

[0065] Devices and methods for reducing optical losses in transparent conductive oxides (TCOs) used in silicon heterojunction (SHJ) solar cells has been provided. Thus, improved optics and resistance in cell stacks result in improved generation current and more economically useful silicon heterojunction (SHJ) solar cells.

[0066] Numerous modifications to the present invention will be apparent to those skilled in the art in view of the foregoing description. Accordingly, this description is to be construed as illustrative only and is presented for the purpose of enabling those skilled in the art to make and use the invention and to teach the best mode of carrying out same. The exclusive rights to all modifications which come within the scope of the appended claims are reserved.