NL2019617B1

NL2019617B1 - Interdigitated back-contacted solar cell with p-type conductivity

Info

Publication number: NL2019617B1
Application number: NL2019617A
Authority: NL
Inventors: Eugène Bende Evert; Wu Yu; Anker John
Original assignee: Tno
Priority date: 2017-09-22
Filing date: 2017-09-22
Publication date: 2019-03-28

Abstract

A back-contacted solar cell based on a silicon substrate of p-type conductivity has a front surface for receiving radiation and a rear surface. The rear surface is provided with a tunnel oxide layer and a doped polysilicon layer of n-type conductivity. The tunnel oxide layer and the patterned doped polysilicon layer of n-type conductivity form a patterned layer stack provided with gaps in the patterned layer stack. An Al-Si alloyed contact is arranged within each of the gaps, in electrical contact with a base layer of the substrate, and one or more Ag contacts are arranged on the patterned doped polysilicon layer and in electrical contact with the patterned doped polysilicon layer.

Description

Interdigitated back-contacted solar cell with p-type conductivity

Field of the invention

The present invention relates to a back-contacted p-type solar cell. In addition, the invention relates to a solar panel or photovoltaic module comprising at least one such solar cell.

Background A conventional p-type IBC (Interdigitated Back-Contacted) solar cell has on its rear side regions of either a p-type base or diffused n-type emitter. It is a challenge to select a dielectric layer that can passivate both regions equally well. Most dielectric layers have a surface charge. Well-known passivating dielectric layer are aluminum oxide (alumina, AI2O3) with a negative surface charge and hydrogenated amorphous silicon nitride (a-SiNx:H) with a positive surface charge.

If AI2O3 is applied over the entire rear surface of a p-type IBC solar cell this has a beneficial effect on the surface recombination at the p-type base region. In that case electrons (minority carriers) are repelled while holes (majority carriers) are attracted, which is denoted as “accumulation”. Since the minority carrier concentration is the limiting factor it determines the surface recombination rate. In this way, the surface recombination rate at the p-type base surface can be reduced, which contributes to a high open-circuit voltage. However, at the n-type region (the emitter) the mechanism works in the opposite direction. Here, holes (minority carriers) are attracted and electrons (majority carriers) are repelled. This increases surface recombination, since also here the minority carrier concentration, which are in this area the holes, sets the rate of recombination.

If the doping concentration of the n-type emitter decreases, which is usually beneficial for the reduction of surface recombination, the effect becomes stronger. The worst case occurs when the product of the concentration of electrons Ne and the capture cross section oe for electrons equals those parameters Nh, Oh of holes (i.e. Ne oe = Nh Oh).

This situation is denoted as “depletion”, since the sum of concentration of electrons and holes is in that case minimal.

Hoex et al., in “Surface passivation of phosphorus-diffused n+-type emitters by plasma-assisted atomic-layer deposited AI2O3”, Phys. Status Solidi RRL 6, No. 1, 4-6 (2012) / DOI 10.1002/pssr.201105445, show a minimum in implied Voc of an nonmetallized solar cell that illustrates this situation (Figure 3, ibid.).

Further reduction of the doping concentration leads to “inversion” in the emitter, which means that the electron concentration becomes lower than the hole concentration. This can be beneficial for surface recombination reduction, but leads to a non-functional solar cell.

If, on the contrary, a dielectric with a positive surface charge, like a-SiNx:H, is deposited on the p/n+ rear surface of the conventional p-type IBC cell, the physics work in the opposite way. The positive surface charge has beneficial effect on the emitter surface recombination and, in general, the surface recombination current jo decreases with decreasing doping surface concentration (which in practice means an increasing emitter sheet resistance).

However, on the lowly doped p-type substrate, typically having a resistivity of 1-3 Ω-cm, the strong positive surface charge leads to inversion. The implies that base surface region has become a surface-charge induced (i.e., non-diffused) emitter which extends towards the base contact. It is well known that this leads to shunts, causing extreme poor cell results. A possible solution to the above problem is to create a dielectric layer with a (near-) zero surface charge. However, it has shown to be quite difficult to process these dielectric layers. Moreover, the quality of the passivation does not depend anymore on the field effect but only on the so-called “chemical passivation”, which is in general much harder to realize. An example of such a layer is intrinsic amorphous silicon of which the chemical surface recombination is very good provided that the wet chemical pretreatment is of a very high standard. However, such a layer is known not to be stable at high temperatures and therefore not suitable in combination with mainstream firing process of screen-printed metal contacts.

It is an object of the present invention to overcome or mitigate one or more of the problems of the prior art.

Summary of the invention

The object is achieved by a back-contacted solar cell based on a silicon substrate of p-type conductivity having a front surface for receiving radiation and a rear surface; in which the rear surface is provided with a tunnel oxide layer and a doped polysilicon layer of n-type conductivity; the tunnel oxide layer and the doped polysilicon layer of n-type conductivity forming a patterned layer stack, is provided with gaps in the patterned layer stack; an Al-Si contact is arranged within each of the gaps, in electrical contact with a base layer of the substrate, and one or more Ag contacts is arranged on the patterned doped polysilicon layer and in electrical contact with the patterned doped polysilicon layer.

According to the invention, there is provided a p-type IBC solar cell with a patterned stack layer of tunnel oxide and n-type poly silicon emitter region. The tunnel-oxide is typically 0.5-2 nm thick and can be made by wet chemical oxidation (e.g. with a HNO3 solution) or by gas-based oxidation at elevated temperatures.

In such a solar cell, the passivating properties are determined by the silicon oxide surface of the tunnel oxide layer. Also, the passivation of dangling bonds of the silicon substrate surface by hydrogen atoms plays a key role.

In one embodiment where the rear surface of the p-type IBC solar cell has alternating structure of n-type polysilicon emitter regions and base region, a dielectric with a negative surface charge, like AI2O3 (or a stack comprising AI2O3 layer and a-SiNx:H layer), can be deposited. Then the presence of the dielectric layer causes the desired accumulation at the base region surface, whereas the tendency towards depletion is not detrimental since, first of all, the doping concentration in n-type polysilicon is rather high and, secondly, the tendency towards depletion takes place at a distance of typically 10-20 nm and therefore does not affect the surface recombination at the tunnel oxide since the polysilicon layer is typically one order of magnitude thicker. In the case of n-type poly silicon, the important function of the dielectric layer is to provide hydrogen that will migrate during an elevated temperature step (usually taking place during firing of the contacts of the solar cell) towards the tunnel oxide and thereby strongly reduces the surface recombination.

In another embodiment the solar cell’s rear surface has an interdigitating electrode pattern comprising alternating areas of n-type polysilicon and intrinsic polysilicon. This can typically be made by subsequently depositing the tunnel oxide and the intrinsic polysilicon layer. Thereafter a diffusion barrier pattern is applied and a diffusion of phosphorous atoms into the wafer surface takes place, e.g. by POCI3 diffusion or by ion implant. This is then followed by removal of the diffusion barrier. In this embodiment the surface passivation is at the tunnel oxide over the full area. The hydrogenation of the tunnel oxide where the hydrogen atoms will be located at the dangling bonds of the silicon wafer surface is resulting in a very high degree of surface passivation. Since a polysilicon layer, of about 200 nm, is covering the tunnel oxide and the surface charge of the dielectric only affects carrier concentration levels to 10-20 nm at the outer poly silicon surface, the sign of the surface charge is not relevant anymore. The only role of the dielectric is for both the intrinsic and the n-type poly silicon to provide hydrogen. In this embodiment therefore the choice of the dielectric layer type is free. This means the dielectric layer can for example be AI2O3, a-SiNx:H or a stack of these dielectric layers.

The present invention also relates to a solar panel or photovoltaic module comprising one or more of solar cells as described above.

Advantageous embodiments are further defined by the dependent claims.

Brief description of the drawings

Embodiments of the present invention will be described in more detail hereinafter, by way of example only, with reference to the accompanying drawings which are schematic in nature and therefore not necessarily drawn to scale. Furthermore, corresponding reference signs in the drawings relate to corresponding or substantially similar elements.

In the drawings, Figure 1 schematically shows a cross-section of a solar cell in accordance with an embodiment of the present invention;

Figure 2 schematically shows a cross-section of a solar cell according to an present invention;

Figure 3 schematically shows a cross-section of a solar cell according to an embodiment of the present invention;

Figure 4 schematically shows a plane view of a rear surface of a solar cell according to an embodiment of the present invention;

Figure 5 schematically shows a cross-section of the solar cell of Figure 4;

Figure 6 schematically shows a plane view of a rear surface of a solar cell in accordance with an embodiment of the present invention;

Figure 7 schematically shows a plane view of a rear surface of a solar cell according to an embodiment of the present invention, and

Figure 8 shows a cross-section of a solar panel comprising at least one solar cell according to an embodiment of the invention.

Description of embodiments

The solar cell of the present invention is based on a p-type semiconductor substrate with positive and negative polarity contacts arranged on the rear surface of the substrate. The positive contacts directly connect to the p-type base layer of the substrate, while the negative contacts are connected to n-type doped areas on the rear surface. Typically, the positive contacts are embodied as metal-alloy contacts, the negative contacts are embodied as metal or metallic contacts.

Figure 1 schematically shows a cross-section of a solar cell in accordance with an embodiment of the present invention.

According to an embodiment of the invention, a solar cell 1 comprises a silicon substrate 10 of p-type conductivity. For example, the silicon substrate is doped with Boron as dopant species.

The silicon substrate 10 has a front surface F and a rear surface R.

On the rear surface R, a stack of a tunnel oxide layer 12 and a doped poly-silicon layer 14 is arranged. The layer stack 12, 14 is patterned with gaps A to the surface of the silicon substrate. The tunnel oxide layer 12 consists typically of silicon dioxide and typically has a thickness of about 2 nanometer or less.

The patterned doped poly-silicon layer 14 has an n-type conductivity, opposite to the p-type conductivity of the silicon substrate. The thickness of the n-type patterned poly-silicon layer is between about 10 and about 300 nm.

Within the pattern of the patterned n-type doped poly-silicon layer gaps A are present. In the gaps A, a contact 16 of a first conductivity type is positioned that connects to the silicon substrate 10, i.e., the base of the solar cell. In this embodiment, the contact of first conductivity type is a metal-alloy contact, typically an Al-Si contact. Optionally this contact comprises Boron. Such an Al-Si contact consists in general of a layered structure comprising an Al head 160, an Al-Si alloy layer 161 and an Al doped BSF region 162. Optionally, layers 160, 161 and 162 comprise Boron. The BSF region 162 typically is in contact with the silicon matrix of the substrate and envelopes the Al-Si alloy layer 161. The Al head 160 is positioned on top of the Al-Si alloy layer at the rear surface R and extending therefrom. The lateral size of an Al-Si contact at the substrate level is smaller than the lateral size of a gap A to avoid contact with the bordering region of n-type doped polysilicon.

Contacts 18 of the second conductivity type, opposite to the first conductivity type, are arranged on and in electrical contact with the patterned n-type doped polysilicon layer 14. According to the invention, the contacts 18 of the second conductivity type are embodied as metal contacts, in an embodiment Ag contacts or alternatively transition metal (e.g., Ni-Cu plated, potentially by Light Induced Plating) contacts. The lateral size of an Ag contact 18 is smaller than the lateral size of the patch of the patterned n-type doped poly-silicon on which the Ag contact is positioned.

The patterned n-type doped poly-silicon layer 14 and the openings A are covered by a hydrogenated dielectric layer 20. Such hydrogenated dielectric layer 20 improves the passivation of the rear surface, i.e., the interface between the tunnel oxide and the surface of the silicon substrate 10, by providing hydrogen to mask electrically active faults at the rear surface R, (e.g., dangling bonds at the silicon surface). The hydrogenated dielectric layer 20 can comprise a layer of hydrogenated silicon nitride SiNx:H 24, a layer of alumina (AI2O3) 22, or a dielectric layer stack of these layers 22, 24. In such a dielectric layer stack, the alumina layer 22 is positioned in-between the rear surface of the substrate and the SiNx:H layer 24, since the SiNx:H layer may have a positive surface charge.

Both the Al-Si contacts 16 and the Ag contacts 18 extend through the hydrogenated dielectric layer(s) to be in electrical contact with the base of the silicon substrate 10 and the patterned n-type doped poly-silicon layer 14, respectively.

In an embodiment, the hydrogenated dielectric layer is configured to have a negative surface charge. Such negative surface charge influences the distribution of minority and majority charge carriers in the silicon substrate which as explained in the introductory part beneficially provides a reduction of the recombination effect.

In an embodiment, the front surface F which is to receive radiation is covered by an AI2O3 layer 30 , bordering the silicon substrate 10 of p-type conductivity, and a SiNx:H layer 32 on top of the AI2O3 layer.

In another embodiment, the front surface F which is to receive radiation is covered by an intrinsic poly-silicon layer 30A and an anti-reflecting coating layer 32A, for example silicon nitride. Since intrinsic poly-silicon is known to parasitically absorb blue light, a solar cell with such an intrinsic poly-silicon layer 30A could be used as a bottom solar cell in a tandem device where the bottom cell mainly harvests Near Infra-Red light and the parasitic absorption of i-poly is then relatively low.

In case the front surface F is configured with only an anti-reflecting coating, the solar cell 1 will suffer less from parasitic blue light absorption

Figure 2 schematically shows a cross-section of a solar cell 2 according to an embodiment of the present invention.

The solar cell 2 according to this embodiment comprises a silicon substrate 10 of p-type conductivity.

On the rear surface R, a tunnel oxide layer 12 is arranged. On the tunnel oxide layer 12, a patterned n-type doped poly-silicon layer 14 is arranged.

Within the pattern of the patterned n-type doped poly-silicon layer, regions of intrinsic poly-silicon 26 are present that border on the n-type doped poly-silicon 14. In this embodiment, the tunnel oxide layer 12 is also present between the intrinsic poly silicon 26 and the rear surface R of the silicon substrate.

In the intrinsic poly-silicon region 26, a metal-alloy contact 16; 160,161,162 of a first conductivity type, typically an Al-Si contact, is positioned that connects to the silicon substrate 10, i.e., the base of the solar cell. The lateral size of an Al-Si contact is smaller than the lateral size of the intrinsic poly-silicon region 26.

Metal contacts 18 of the second conductivity type such as Ag contacts are arranged on and in electrical contact with the patterned n-type doped poly-silicon layer 14, as described above with reference to figure 1.

In this embodiment of the solar cell 2, the patterned n-type doped poly-silicon layer 14 and the intrinsic poly-silicon regions 26 are covered by a layer of hydrogenated silicon nitride SiNx:H24 which beneficially provides hydrogen for passivation of the rear surface.

Both the Al-Si contacts 16; 160,161,162 and the Ag contacts 18 extend through the hydrogenated dielectric layer(s) to be in electrical contact with the base of the silicon substrate 10 and the patterned n-type doped poly-silicon layer 14, respectively.

Figure 3 schematically shows a cross-section of a solar cell 3 according to an embodiment of the present invention.

The solar cell shown in figure 3 may be regarded as a variant of the solar cell 2 as described above with reference to figure 2.

Within the pattern of the patterned n-type doped poly-silicon layer, regions of intrinsic poly-silicon 26 are present that border on the n-type doped poly-silicon 14. The tunnel oxide layer 12 is also present between the intrinsic polysilicon 26 and the rear surface R of the silicon substrate.

The Ag contact 18 is positioned as shown in figure 2 on the patterned n-type doped poly-silicon 14.

In the region of intrinsic poly-silicon 26, the Al-Si contact 16; 160,161,162 is arranged in electrical contact with the base layer of the silicon substrate 10.

In this embodiment of the solar cell, the intrinsic poly-silicon and the n-type doped poly-silicon are covered with an alumina layer 22 that functions as hydrogenated dielectric layer that provides hydrogen for surface passivation at the tunnel oxide.

The Al-Si contact 16; 160,161,162 is located between flanking bounding elements 28 that consist of a material characterized as an inert material with respect to the alumina dielectric, aluminum and Al-Si alloy. The bounding elements 28 are arranged on the stack of alumina 22 and intrinsic polysilicon 26 which shows to be a stable configuration during the heat treatment to form the Al-Si contacts 16.

In an embodiment, the bounding elements material comprises aluminium-oxide particles. Also, the bounding elements material may comprise a material based on aluminum oxide or aluminium nitride.

The bounding elements are configured to prevent poor Al-BSF formation at the edges of the Al. This reduces contact recombination. This is in particular important if the Al-paste is a firing-through paste with etchant particles, e.g. a glass frit.

Since the material of the bounding elements does not react with the dielectric layer or the silicon substrate at the annealing temperature, the bounding elements act as an inert template during the anneal step. Typically, the anneal step is done between about 660°C and about 800 °C.

In addition, as the aperture (i.e., the width between the bounding elements) determines the surface area where Al reacts with Si, the bounding elements do not require a high aspect ratio and may have a rounded cross-section or sloped side walls like a mound or embankment. The bounding elements can be printed in a similar manner as aluminum based paste, e.g. by screen printing.

Figure 4 schematically shows a plane view of a rear surface of a solar cell in accordance with the embodiment of the solar cell 2 shown in figure 2. On the rear surface R, the Al-Si contacts 16; 160,161,162 and Ag contacts 18 are laid out as interdigitating electrodes. The intrinsic poly-silicon layer 26 and the n-type doped polysilicon layer 14 are laid-out in corresponding interdigitating patterns.

In the intrinsic poly-silicon layer 26 an elongated opening 34 to the silicon substrate has been created by a laser scribe or by printing an etching paste. The Al-Si contact 16; 160,161,162 fdls the elongated opening 34 and is in electrical contact with the underlying silicon substrate as shown in Figure 2.

On the intrinsic poly-silicon layer 26 the Al head portion 160 of the Al-Si contact extends beyond the boundaries of the laser scribed elongated opening 34 in both transverse and longitudinal directions Y; X.

Both the Ag contact interdigitating electrode 18 and the Al-Si contact interdigitating electrode 16 are each connected to respective busbar 118; 116 of n-type and p-type polarity, that run in the transverse direction Y perpendicular with the longitudinal direction of the interdigitating electrodes 18; 16.

According to the invention, both busbars 116, 118 consist of Ag (or Ag alloy). However, the busbars connecting the Al-Si interdigitating electrode can also be made in a similar way as the Al fingers. During manufacturing, the Ag busbars are formed simultaneous with the Ag contacts i.e., the Ag interdigitating electrodes. The Ag busbar for connecting the Al-Si contact (the Al-Si interdigitating electrode) overlaps 216 with a proximal end of the Al-Si interdigitating electrode 16 to obtain an electrical contact. Usually, busbars 116, 118 are provided with Ag interconnection pads.

It is noted that the layout of the interdigitating electrodes 16, 18 and busbars 116, 118 are similar for the solar cells described with reference to figure 1 and figure 3, respectively.

Figure 5 schematically shows a cross-section of a solar cell according to an embodiment of the invention. In this embodiment, the Al-Si contact 16, in particular the Al head 160, is flanked by Ag sub-contacts 181, 182 that creates an electrical contact between the Al head 160 and the top of the intrinsic poly-silicon layer 26, which function as interconnecting pad for solder and electrically conductive adhesive (ECA). As shown in figure 4, Ag busbars are created on both n-type doped poly-silicon 14 and intrinsic poly-silicon 26. This layout allows to use a two-step print of Ag in one step and Al in the other step to create interdigitating electrodes, interconnecting pads and busbars. In the one-step Ag paste print case the Ag paste is firing-through paste that penetrates into the polysilicon. Since the intrinsic poly is not conductive, this will not lead to shunts.

Figure 6 schematically shows a plane view of a rear surface of a solar cell in accordance with an embodiment of the present invention. The layout of the interdigitating electrodes on the rear surface is similar as described above with reference to figure 4. In the embodiment of the solar cell shown in figure 6, the width W of the Al head 160 of the Al-Si contact 16 is larger than the width of the elongated intrinsic poly-silicon 26 surrounding the elongated opening 34 to the silicon substrate. The Al head 160 is configured to overlap a portion of the neighboring n-type doped poly-silicon layer 14.

As it will be appreciated that Al is the limiting factor for the conduction of the metallization (in comparison with Ag). By using a wider Al-Si contact a lower resistance is obtained. Also, by overlapping the n-type doped poly-silicon (the other polarity) the lateral conductance can be improved. Moreover, by varying the Al paste width the alloying process can be tuned. In this way the Al layer size, the alloy layer size and the BSF thickness (i.e., the formation of the layered stack of Al head 160, Al-Si layer 161 and Al-based BSF region 162 and their respective thickness within the Al-Si contact 16) can be tuned for optimal performance.

Figure 7 schematically shows a plane view of a rear surface of a solar cell according to an embodiment of the present invention. In this embodiment, the Al-Si interdigitating electrode 16 is provided with a series of intermittent elongated openings 36 in the intrinsic poly-silicon 26 within the length of the electrode. In comparison to the single elongated opening 34 as shown in figure 4 and figure 6 the series of intermittent openings has a shorter contact length with the underlying silicon substrate 10. As a result, recombination in the contact can beneficially be reduced.

As shown here, the layout of intermittent openings 36 can be combined with an Al head layer 160 partially overlapping the neighboring n-type doped poly-silicon 14 to tune performance.

The invention also relates to a solar panel 800 or photovoltaic module that comprises one or more back-contacted solar cells 1 ;2;3 as described above with reference to the figures 1-7.

The solar panel 800 further comprises a transparent top-plate 802, a top encapsulant 804, a bottom encapsulant 806 and a back sheet 808.

The one or more solar cells 1 ;2;3 are arranged on the backsheet 808, that is provided with a conductor pattern (not shown) to which the busbars 116, 118 that connect to the Al-Si interdigitating electrodes 16 and the Ag interdigitating electrodes, respectively are electrically contacted by solder or conductive adhesive 810. The bottom encapsulant layer 806 is arranged in-between the rear surface R of the one or more solar cells and the back-sheet with contact openings at the locations where the busbars 116; 118 connect to the back-sheet 808.

On top of the solar cells, above the front surfaces F thereof, the top encapsulant 804 is arranged, in-between the solar cells and the top transparent plate 802.

It is noted that the front surface and/or the rear surface R may have a texture of pyramidal shapes.

Also it is noted that the silicon nitride layer 24 at the rear surface R has a thickness that may be tuned for reflection of radiation that has reached the rear surface R. In particular the tuning of the thickness can be done with respect to a minimal wavelength of the radiation. In an embodiment, the minimal wavelength may be about 900 nm. The thickness may be at least about 80 nm, or about 150 nm or preferably about 200 nm.

The solar cell according to the invention can be created by various methods of manufacturing. Below these methods are described.

Nomenclature FT=firing-through NFT = non firing-through LCO=Local Contact Opening PGR= POC13 Glass Removal SSE=Single Side Etch

The p-type interdigitate back contact (IBC) solar cell is manufactured on boron doped p- type silicon wafer. It has the emitter at rear surface R that does not face to the sunlight and may have or may not have a Front Surface Field (FSF) on the front surface of the silicon substrate. Preferably, the FSF is absent since this saves costly processing steps (e.g. BBr3 diffusion) and moreover, in general, an FSF-free top surface has a lower surface recombination current density. In addition, the high-temperature Boron diffusion step can have detrimental effect on the quality of the base material. Since p-type wafers from an ingot with a resistivity range of 1-3 Ω-cm are standard, an FSF is not really needed to have sufficient conduction in the base. The front Si surface is passivated by a dielectric layer or a stack layer. At the rear surface R, alternating emitter/base-regions are formed by different patterning techniques. A p-type IBC cell with three different types of rear architectures will be described below. These are: I. A rear surface where diffused emitter regions alternate with base regions, where the entire surface is covered with a dielectric layer or a stack of layers with a negative surface charge. II. A rear surface where n-type polysilicon emitter regions alternate with base regions, where the entire surface is covered with a, hydrogen-providing, dielectric layer or stack of layers with a negative surface charge. III. A rear surface where n-type polysilicon emitter regions alternate with intrinsic polysilicon regions, where the entire surface is covered with a, hydrogen-providing, dielectric layer or stack of layers with an arbitrary surface charge.

The solar cell manufacturing process consists of the following steps: 1. Preparing the p-type wafer; 2. Creating a patterned emitter; 3. Applying the front-surface passivation; 4. Applying the rear-surface passivation; 5. Opening of the dielectric; 6. Applying metallization.

These process steps are elaborated below. 1. Preparing the p-type wafer (table 1)

Table 1 Process steps for preparing monocrystalline and multi-crystalline wafers

2. Creating a patterned emitter (table 2).

Table 2 Rear-surface area type-I (alternating bare base region, with classically diffused emitter region)

For rear surface type II and III, a polysilicon layer of typically 10-300 nm is applied. Prior to the polysilicon deposition a very thin (e.g. 0.5-2nm) silicon dioxide layer is created on the silicon surface. This can be realized by in-situ deposition in a Low Pressure Chemical Vapor Deposition (LPCVD) system or in a Plasma Enhanced Vapor Deposition (PECVD) system, or by ex-situ chemical oxidation (such as NAOS, RCA, H2O2, etc ).

On top of the oxide layer an intrinsic poly silicon layer or an n-type, in situ doped, polysilicon with a thickness of typically 10 - 300nm is deposited by LPCVD or by PECVD. In fact, the PECVD deposition mostly leads to deposition of amorphous silicon (a-Si), which can be converted into polysilicon by an elevated temperature step (anneal).

Table 3 Rear-side type II ( alternating bare base region, with n-type polysilicon emitter)

Table 4 Rear-side type III (alternating intrinsic-polysilicon with n-type polysilicon regions)

For a special embodiment of type II and III solar cells, the intrinsic poly silicon layer on the front side can deliberately be kept. An intrinsic poly silicon layer on the front side can be obtained if the ion implant takes place on the rear surface R or if a temporary diffusion barrier (e.g. SiO2 or SiNx) is applied on the front surface F and after POC13 processing is removed. This embodiment can have a very low front-surface recombination due to the good passivating properties of the oxide-intrinsic poly stack. However the intrinsic polysilicon layer will parasitically absorb blue light. Because of this, this solar cell type is targeted as bottom solar cell in a tandem device. 3. Applying the front-surface passivation

For all types I, II, III a (wet) chemical clean step can be applied prior to front coating deposition, which could be RCA-1, RCA-2, HNO3, (buffered) HF, HC1 and different combinations and different sequences.

Front coatings can be AI2O3, or a stack of AhOs-SiNx or SiNx. The preferred coating is the stack of AhO>-SiNx since this is proven as passivating layer on the rear surface of PERC solar cells and the layers of the stack can be optimized for antireflection purposes as well. Typically a stack of 6nm AI2O3 and 80 nm SiNx could work for this purpose. 4. Applying the rear surface passivation

Table 5.

It should be noted that the front dielectric layers could be co-deposited with the rear dielectric layers. For instance, for all types (I, II and III) first double-sided AI2O3 followed by double-sided SiNx could be applied. Deposition systems in which halffabricates are levitated and float through the deposition chamber might be a useful for this purpose. It should be noted that, unlike single-side deposition where wrap-around of the deposition can be undesired, the double-sided deposition has more forgiving process conditions.

The rear surface will be opened at the base regions by laser ablation or by printing a chemical etch paste. This is followed by a (wet) chemical clean to remove laser damage and dielectric coating flakes.

Optionally the dielectric is opened in a similar way at the emitter regions, which can later on be used for the contacting of NFT Ag-paste. 5. Opening of the dielectric

Openings in the dielectric layer at the base regions are realized. Per targeted aluminium print area this can be a line shape or a series of smaller lines or dots. Opening can be realized by laser ablation or by selective chemical etching, e.g. by printing an etch paste. These steps can be followed with a chemical clean step. Optionally, local contact openings (LCOs) on the emitter regions can be realized in the same process step. The LCOs can then be used to make a metallic contact to the emitter by (screen)printing of NFT Ag-paste or by (Light Induced) plating. It should be noted that for these types of metallization the metal does not penetrate into the n-type polysilicon emitter or into the classically diffused emitter.

6. Applying metallization

The process for the metallization is listed in Table 6. Three metal types are envisaged for (screen) printing: Al-paste, FT Ag-paste and NFT Ag-paste. The latter will not fire through the dielectric layer. For the NFT pastes, two categories can be identified: The first is co-fired with the Al-paste and the FT Ag-paste and the second is separately printed and fired, mostly at lower firing temperatures. The second paste type is often referred to as “floating busbar” paste.

The NFT paste is used to make contact to the Al-paste. Both the NFT Ag paste and the FT paste can be electrically connected to the metal leads that series connect solar cells in the photovoltaic module by solder or by electrically conductive adhesive.

In all cases the Al paste is printed on the rear dielectric layer such that the LCO in the dielectric layer of the base region is fully covered. The Al paste area might also partially overlap with the emitter region.

The BSF is formed by an Aluminum-Silicon alloy process in which the Aluminum of the screen-printed Al-paste dissolves the silicon of the base wafer material and, optionally, the intrinsic polysilicon during heating up above 660°C (i.e. the melting point of aluminum) and where during the cooling down phase Al-doped silicon is rejected from the melt and recrystallizes, also known as epitaxial growth.

Table 6 shows metallization options for the types I, II and III.

Table 6 Metallization options

After the (screen) print steps, the solar cell 1, 2, 3; I, II, III can undergo a thermal step where the FT Ag-paste penetrates the dielectric layer and contacts the classically diffused emitter, the n-type poly silicon emitter or the intrinsic poly-silicon.

The FT Ag-paste can be selected amongst the pastes used for PERC front-side emitter contacting. Examples of the NFT Ag-paste that are used for PERC solar cells rear surfaces are SOL326, SOL 326S by Heraeus, PV56x by Dupont, BS828B by ENC and GB21 by Gonda.

The BSF is formed by an Aluminum-Silicon alloy process in which aluminum of the screen-printed Al-paste dissolves silicon of the base wafer material and, in case of surface-area type III, the intrinsic polysilicon, during heating up above 660°C and

where during the cooling down phase Al-doped silicon is rejected from the melt and recrystallizes, which also denoted by epitaxial growth.

All metal conductive pastes can be sintered by co-firing at a temperature of (typically 700-840°C). Due to application of three different pastes, three print steps are required to complete the metallization of this type of p-IBC solar cell. For type III the option exists to apply two print step, namely FT Ag-paste and Al-paste. In that case the FT Ag-paste also penetrates the intrinsic polysilicon but since this layer does not conduct it will not lead to shunt.

In the foregoing description, the invention has been described with reference to specific embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the scope of the invention as summarized in the attached claims.

In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiments disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.

Claims

A solar cell provided with rear contacts on the basis of a silicon substrate with p-type conductivity and with a front surface for receiving radiation and a rear surface; wherein the rear surface is provided with a tunnel oxide layer and a doped polysilicon layer with n-type conductivity; wherein the tunnel oxide layer and the doped polysilicon layer with n-type conductivity form a patterned layer stack that is provided with gaps in the patterned layer stack; wherein an Al-Si alloy-based contact is provided within each of the gaps, in electrical contact with a base layer of the substrate, and one or more Ag contacts or transition metal contacts are provided on the patterned doped polysilicon layer and in electrical contact with the patterned doped polysilicon layer.

The solar cell according to claim 1, which comprises an intrinsic polysilicon layer adjacent to the patterned doped polysilicon layer with n-type conductivity and covers the back surface of the silicon substrate in the gap, the tunnel oxide additionally arranged between the intrinsic The polysilicon layer and the back surface of the silicon substrate in the interstices and the Al-Si contacts extend through the intrinsic polysilicon layer and the tunnel oxide layer.

The solar cell of claim 1, wherein a hydrogenated dielectric layer is provided to cover the interstices and the patterned doped polysilicon layer, the Al-Si contacts and the Ag or transition metal contacts each extending through the hydrogenated dielectric layer.

The solar cell of claim 3, wherein the hydrogen-providing dielectric layer has a negative surface charge.

The solar cell according to claim 3 or 4, wherein the hydrogen-providing dielectric layer comprises an Al 2 O 3 layer or a stack of an a-SiN x: H layer and an Al 2 O 3 layer, wherein the Al 2 O 3 layer extends between the a-SiNx: H layer and the interstices and the patterned doped polysilicon layer.

The solar cell of claim 2, wherein a hydrogen-providing dielectric layer is provided to cover the intrinsic polysilicon layer and the patterned doped polysilicon layer, the Al-Si contacts and the Ag or transition metal contacts each extending through the hydrogen-providing dielectric low.

The solar cell of claim 6, wherein the hydrogen-providing dielectric layer comprises an α-SiNx: H layer, an Al 2 O 3 layer or a stack of an α-SiN x: H layer and an Al 2 O 3 layer.

The solar cell according to any of the preceding claims 2-7, wherein an Ag contact body is adjacent to the Al-Si contacts on the intrinsic polysilicon layer.

The solar cell according to any of the preceding claims, wherein the Al-Si contacts and the Ag or transition metal contacts are arranged as interdigitating electrodes between a first bus bar connecting the Ag or transition metal contacts and a second bus bar connecting the Al-Si contacts connects; wherein the bus bars extend along the rear surface in a direction perpendicular to a length of the interdigitating electrodes.

The solar cell according to claim 9, wherein the bus bars each consist of Ag.

The solar cell according to any of claims 1,9, 10, as far as dependent on claim 3, wherein the Ag is based on firing-through Ag paste material.

The solar cell of any one of claims 9-11, wherein the Al-Si interdigitating electrode comprises a plurality of individual Al-Si contacts arranged in the length of the interdigitating electrode, the plurality of Al-Si contacts being interconnected by an elongated body of Al or Al-Si alloy which extends above the plurality of Al-Si contacts.

The solar cell according to any of the preceding claims 9-12, wherein a portion of the Al-Si interdigitating electrode extends laterally over a portion of the adjacent patterned doped polysilicon layer with n-type conductivity.

The solar cell according to claim 2, wherein openings made by means of laser scribe are placed in the dielectric layer above locations of the gaps.

The solar cell according to any of claims 3-8, wherein openings made by means of laser scribe are placed in the hydrogenated dielectric layer above gaps in which the Al-Si contacts are arranged, and openings made by means of laser scribe are placed in the hydrogenated dielectric layer in which the Ag or transition metal contacts are positioned above the patterned doped polysilicon layer with n-type conductivity.

The solar cell according to claim 15, wherein the Ag is based on "ηοη-firing-through" paste material based on Ag.

The solar cell according to any of the preceding claims 1-16, further comprising a dielectric layer on the front surface, wherein the dielectric layer is configured with a negative surface charge.

The solar cell according to any of the preceding claims 1 to 16, further comprising on the front surface an intrinsic polysilicon layer that is covered by a hydrogen-providing dielectric layer.

The solar cell according to claim 6 or 7, wherein the Al-Si contact is bounded at its edges by boundary elements applied to the hydrogen-providing dielectric layer, the boundary elements consisting of a material characterized as an inert material with respect to the material of the hydrogen-providing dielectric layer and with respect to aluminum.

A photovoltaic module comprising at least one solar cell according to any of the preceding claims, a transparent top plate and a backsheet, the at least one solar cell being arranged between the top plate and the backsheet.

A method for manufacturing a backward-facing solar cell based on a silicon substrate with p-type conductivity and having a front surface for receiving radiation and a rear surface; wherein the method comprises: providing on the back surface a layer stack of a tunnel oxide layer and a doped polysilicon layer with n-type conductivity; patterning the layer stack with gaps in the layer stack; applying an Al-Si alloy-based contact within each of the interstices, in electrical contact with a base layer of the substrate, and applying one or more Ag contacts or transition metal contacts to the patterned doped polysilicon layer and into electrical contact with the patterned doped polysilicon layer.