CN111816346A - Printing pastes for improving the material properties of metal particle layers - Google Patents

Abstract

Embedding pastes for use with semiconductor devices are disclosed. The slurry contains noble metal particles, embedded particles, and an organic vehicle, and can be used to improve the material properties of the metal particle layer. A specific formation has been developed to screen-print and sinter directly onto the layer of dry metal particles to make a sintered multilayer stack. Sintered multilayer stacks are tailored to produce solderable surfaces, high mechanical strength and low contact resistance. In some embodiments, the sintered multilayer stack can be etched through the dielectric layer to improve adhesion to the base layer. Such pastes can be used to increase the efficiency of silicon solar cells, particularly polycrystalline and monocrystalline silicon back surface field (BSF), and passivated emitter and rear contact (PERC) photovoltaic cells. Other applications include integrated circuits, and more generally, electronic devices.

Description

Printing pastes for improving the material properties of metal particle layers

This application is a divisional application of the application dated November 24, 2016, the application number is 201611044245.X, and the invention title is "Printing paste for improving the material properties of the metal particle layer".

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims US Provisional Patent Application 62/259,636 filed on November 24, 2015, US Provisional Patent Application 62/318,566 filed on April 5, 2016, and US Provisional Patent Application 62/ filed on August 5, 2016 371,236 and priority to US Provisional Patent Application 62/423,020, filed November 16, 2016, the entire contents of which are incorporated herein by reference.

Statement of Government Support

This invention has received government support based on Contract No. IIP-1430721 awarded by the NSF. The government may have certain rights in this invention.

technical field

The present invention relates to intercalation pastes comprising precious metal particles, intercalation particles and an organic vehicle.

Intercalation pastes can be used to improve the power conversion efficiency of solar cells. The silver-based embedding paste was printed on the aluminum layer, which had moderate peel strength after firing and was then soldered to the tabbing ribbon. This paste is particularly well suited for silicon-based solar cells, which use aluminum back surface fields (BSFs). Typically, 85-92% of the back surface area of the silicon wafers of commercially produced mono- and poly-crystalline silicon solar cells is covered by a layer of aluminum particles, which forms the back surface field and makes ohmic contact with the silicon . The remaining 5-10% of the back silicon surface is covered by a silver back marker layer, which does not create a field and does not make ohmic contact with the silicon wafer. The rear marking layer is mainly used to solder the marking tape to electrically connect the solar cells.

When the silver layer is in direct contact with the silicon substrate on the backside of the solar cell, instead of the aluminum particle layer contacting the substrate, the absolute benchmark for the conversion efficiency of the solar cell is estimated to decrease by 0.1% to 0.2%. Therefore, it is highly desirable to cover the entire rear of the solar cells with a layer of aluminium particles and still be able to use the marker tape to connect the solar cells together. In the past, researchers have attempted to print silver directly on top of a layer of aluminum particles, but during high temperature firing in air, the aluminum and silver layers interdiffusion and cause the layer surface to become oxidized and lose solderability sex. Some researchers have attempted to alter atmospheric conditions to reduce oxidation; however, front-side silver pastes perform best in an oxidizing atmosphere, such as dry air, and overall solar cell efficiency decreases after processing in an inert atmosphere. Other researchers have tried lowering the peak firing temperature of the wafer to reduce interdiffusion, but the front-side silver paste requires a high peak firing temperature (ie, greater than 650°C) to sinter the silicon oxynitride for ohmic interaction with the silicon base layer. touch. Recently, researchers have used ultrasonic soldering of tin alloys directly on top of aluminum to produce solderable surfaces. This technique has achieved adequate peel strengths (ie, 1-1.5 N/mm), but requires additional equipment and uses large amounts of tin, which adds expense. Additionally, the use of ultrasonic soldering on brittle materials such as aluminum and silicon wafers increases wafer cracking and reduces process throughput.

There is a need to develop printable pastes that can modify the material properties of the underlying metal particle layer during firing. For example, precious metals containing pastes, which can be printed directly on aluminum and fired using standard solar cell processing conditions, can improve solar cell efficiency. These slurries reduce Ag/Al interdiffusion, thereby maintaining solderability to the marking tape. It is required that the paste is screen printable and acts as a drop-in replacement, which does not incur additional significant outlay and can be integrated into existing production lines immediately.

SUMMARY OF THE INVENTION

Fired multilayer stacks are disclosed. In one embodiment of the invention, the stack has a base layer, a layer of metal particles on at least a portion of the surface of the base layer, a layer of modified metal particles on at least a portion of the surface of the base layer, and a modified insert directly on at least a portion of the layer of modified metal particles Floor. The modified intercalator has a solderable surface facing away from the base layer. The layer of modified metal particles includes the same metal particles as the layer of metal particles and at least one material from the modified intercalation. The modified intercalation comprises precious metals and materials selected from the group consisting of: antimony, arsenic, barium, bismuth, boron, cadmium, calcium, cerium, cesium, chromium, cobalt, gallium, germanium, indium, iron, lanthanum, hafnium, lead , lithium, magnesium, manganese, molybdenum, niobium, phosphorus, potassium, rhenium, selenium, silicon, sodium, strontium, sulfur, tellurium, tin, vanadium, zinc, zirconium, combinations thereof, alloys thereof, oxides thereof, synthesis thereof , and other combinations. In one arrangement, the modified intercalation layer comprises a noble metal and a material selected from the group consisting of: bismuth, boron, indium, lead, silicon, tellurium, tin, vanadium, zinc, combinations thereof and alloys thereof, oxides thereof, composites, and other combinations thereof.

In one embodiment of the present invention, the modified intercalation layer has two phases: a precious metal phase and an intercalation phase. Greater than 50% of the solderable surface of the modified intercalation may contain noble metal phases. The layer of modified metal particles may include the metal particles discussed above and at least one material from the intercalation phase. The intercalation phase includes a material selected from the group consisting of: antimony, arsenic, barium, bismuth, boron, cadmium, calcium, cerium, cesium, chromium, cobalt, gallium, germanium, indium, iron, lanthanum, hafnium, lead, lithium, Magnesium, manganese, molybdenum, niobium, phosphorus, potassium, rhenium, selenium, silicon, sodium, strontium, sulfur, tellurium, tin, vanadium, zinc, zirconium, combinations thereof, alloys thereof, oxides thereof, compositions thereof, and its other combinations. The noble metal phase includes at least one material selected from the group consisting of: gold, silver, platinum, palladium, rhodium, and alloys, composites, and other combinations thereof.

In another embodiment of the present invention, the modified intercalation has two sublayers: an intercalation sublayer directly on at least a portion of the modified metal particle layer, and an intercalation sublayer directly on at least a portion of the subintercalation precious metal sublayer. The solderable surface of the modified intercalation contains a noble metal sublayer. The layer of modified metal particles may include the metal particles discussed above and at least one material from the subintercalation. Possible materials for the subintercalation are the same as described above for the intercalation phase. Possible materials for the noble metal sublayer are the same as described above for the noble metal phase.

In another embodiment of the present invention, the sintered multilayer stack has as its modified metal particle layer a modified aluminum particle layer. It has a modified intercalation layer with two sublayers: a bismuth-rich sublayer directly on the modified aluminum particle layer; and a silver-rich sublayer directly on the bismuth rich sublayer. The solderable surface of the improved intercalation includes a silver rich sublayer. The modified aluminum particle layer includes aluminum particles and may further include at least one material selected from the group consisting of aluminum oxide, bismuth, and bismuth oxide.

In one arrangement, at least one dielectric layer is directly on at least a portion of the surface of the base layer. The dielectric layer includes at least one material selected from the group consisting of: silicon, aluminum, germanium, gallium, hafnium, and oxides, nitrides, composites, and combinations thereof. In another arrangement, the aluminum oxide dielectric layer is directly on at least a portion of the surface of the base layer and the silicon nitride dielectric layer is directly on the aluminum oxide dielectric layer.

In one arrangement, a solid (eg, eutectic) compound layer is directly on the surface of the base layer. The solid composite layer includes one or more metals selected from the group consisting of aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, titanium, and one or more materials selected from the group consisting of silicon , oxygen, carbon, germanium, gallium, arsenic, nitrogen, indium and phosphorus.

A portion of the base layer adjacent to the base layer surface may be doped with at least one material selected from the group consisting of: aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, titanium, steel, and combinations thereof.

In one embodiment of the invention, a portion of the sintered multilayer stack has a variable thickness. The sintered multilayer stack can have an average peak-to-valley height greater than 12 μm.

At least 70 wt % (weight percent) of the solderable surface of the modified intercalation layer may include materials selected from the group consisting of: silver, gold, platinum, palladium, rhodium, and alloys, composites, and other combinations thereof.

The base layer may include at least one material selected from the group consisting of: silicon, silicon dioxide, silicon carbide, aluminum oxide, sapphire, germanium, gallium arsenide, gallium nitride, and indium phosphide. Alternatively, the base layer may comprise a material selected from the group consisting of: aluminum, copper, iron, nickel, titanium, steel, zinc, and alloys, composites, and other combinations thereof. The metal particle layer may comprise a material selected from the group consisting of: aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, titanium, steel and alloys, composites, and other combinations thereof. Precious metals may include materials selected from the group consisting of: silver, gold, platinum, palladium, rhodium, and alloys, composites, and other combinations thereof.

The metal particle layer may have a thickness between 0.5 μm and 100 μm and/or a porosity between 1 and 50%. The modified intercalation layer may have a thickness between 0.5 μm and 10 μm. Sintered multilayer stacks can have contact resistances between 0 and 5 mOhm, as determined by power line measurements.

There may also be a marking tape directly on at least a portion of the solderable surface of the modified interlayer. In one arrangement, the peel strength between the marker tape and the modified interlayer is greater than 1 N/mm.

In another embodiment of the invention, a sintered multilayer stack has a base layer, a layer of metal particles on at least a portion of the base layer, a layer of modified metal particles on at least a portion of the base layer, and a layer of modified metal particles directly on at least a portion of the layer of modified metal particles Improved intercalation. The modified intercalation has two sublayers: a subintercalation directly on at least a portion of the modified metal particle layer, and a noble metal sublayer directly on at least a portion of the subintercalation. The layer of modified metal particles includes metal particles and at least one material from the subintercalation. Possible materials for subintercalation have been described above.

In another embodiment of the present invention, a sintered multilayer stack has a silicon base layer, a layer of aluminum particles on at least a portion of the base layer, a layer of modified aluminum particles on at least a portion of the base layer, and a modified intercalation layer directly on the layer of modified aluminum particles Floor. The modified intercalation has two sublayers: a bismuth-rich sublayer directly on the modified aluminum particle layer, and a silver-rich sublayer directly on the bismuth-rich sublayer. The modified aluminum particle layer includes at least one material selected from the group consisting of aluminum, aluminum oxide, bismuth, and bismuth oxide.

In one embodiment of the invention, the solar cell has a silicon based layer, at least one front dielectric layer directly on at least a portion of the front surface of the silicon based layer, a plurality of fine gridlines ( fine grid line), at least one front busbar layer in electrical contact with at least one of the plurality of fine grid lines, a layer of aluminum particles on at least a portion of the back surface of the silicon-based layer, and on the back surface of the silicon-based layer part of the rear tabbing layer. The rear marker layer includes a layer of modified aluminum particles on a portion of the rear surface of the silicon-based layer, and a modified intercalation layer directly on at least a portion of the layer of modified aluminum particles. The modified intercalation layer has a solderable surface facing away from the silicon base layer. The layer of modified aluminum particles includes aluminum particles and at least one material from the modified intercalation. Possible materials for improved intercalation have been described above. The layer of aluminium particles may have a thickness between 1 μm and 50 μm and/or a porosity between 3 and 20%. The rear marking layer may have a thickness between 1 μm and 50 μm. The silicon-based layer can be a single crystal silicon wafer, with a p-type substrate or an n-type substrate. The silicon-based layer can be a polysilicon wafer, with a p-type substrate or an n-type substrate.

In one embodiment of the invention, the modified intercalation includes two phases: a noble metal phase and an intercalation phase. More than 50% of the solderable surface can be made from the noble metal phase. The modified aluminum particle layer includes aluminum particles and at least one material from the intercalation phase. Possible materials for the embedded phase have been described above. Possible materials for the noble metal phase have been described above.

In another embodiment of the invention, the modified intercalation comprises two sublayers: a subintercalation directly on at least a portion of the modified metal particle layer, and a noble metal sublayer directly on at least a portion of the subintercalation. The solderable surface contains a noble metal sublayer. The layer of modified aluminum particles includes aluminum particles and at least one material from the subintercalation. Possible materials for subintercalation have been described above. Possible materials for the noble metal sublayer have been described above.

In another embodiment of the present invention, the modified intercalation layer includes two sublayers: a bismuth-rich sublayer directly on the modified aluminum particle layer, and a silver-rich sublayer directly on the bismuth-rich sublayer. The modified aluminum particle layer further includes at least one material selected from the group consisting of aluminum oxide, bismuth, and bismuth oxide. In one arrangement, the layer of modified aluminum particles further comprises bismuth and/or bismuth oxide, and the weight ratio of bismuth to bismuth plus aluminum (Bi:(Bi+Al)) is at least higher in the layer of modified aluminum particles than in the layer of aluminum particles 20%. The bismuth-rich sublayer may have a thickness of between 0.01 μm and 5 μm or between 0.25 μm and 5 μm.

In one arrangement, the at least one back dielectric layer is directly on at least a portion of the back surface of the silicon-based layer. The back dielectric layer includes one or more of the following: silicon, aluminum, germanium, hafnium, gallium, and oxides, nitrides, composites, and combinations thereof. The back dielectric layer may include silicon nitride. In another arrangement, the aluminum oxide back dielectric layer is directly on at least a portion of the back surface of the silicon base layer and the silicon nitride back dielectric layer is directly on the aluminum oxide back dielectric layer. In one arrangement, the solid aluminum-silicon eutectic layer is directly on the silicon-based layer. In one arrangement, a portion of the silicon base layer adjacent to the back surface of the silicon base layer further includes a back surface field, and the back surface field is doped p-type to between 10 ¹⁷ and 10 ²⁰ atoms per cm ³ .

In one embodiment of the invention, a portion of the rear marking layer has a variable thickness and may have an average peak-to-valley height greater than 12 μm.

There may be a marking tape directly on at least a portion of the solderable surface of the modified intercalator. The solderable surface may be silver rich. The solderable surface may contain at least 75 wt % silver. The tape soldered to the silver rich solderable surface can have a peel strength greater than 1 N/mm.

A portion of the layer of modified aluminum particles may have a variable thickness. A portion of the modified aluminum particle layer may have an average peak-to-valley height greater than 12 μm. The contact resistance between the rear marking layer and the aluminum particle layer can be between 0 and 5 mOhm, as determined by power line measurements.

In another embodiment of the invention, a solar cell has a silicon based layer, at least one front dielectric layer directly on at least a portion of the front surface of the silicon based layer, a plurality of fine grid lines on a portion of the front surface of the silicon based layer , at least one front bus layer in electrical contact with at least one of the plurality of fine grid lines, a layer of aluminum particles on at least a portion of the back surface of the silicon based layer, and a back marker layer on a portion of the back surface of the silicon based layer. The rear marking layer has a solderable surface. The rear marker layer includes a modified aluminum particle layer on at least a portion of the rear surface of the silicon-based layer, a bismuth-rich sublayer directly on at least a portion of the modified aluminum particle layer, and a silver-rich sublayer directly on at least a portion of the bismuth-rich sublayer Floor. The modified aluminum particle layer includes aluminum particles and at least one material selected from the group consisting of aluminum oxide, bismuth, and bismuth oxide.

In another embodiment of the present invention, a solar cell module has a front sheet, a front encapsulant layer on the rear surface of the front sheet, and first silicon solar cells and second silicon solar cells on the front encapsulant layer Silicon solar cells. Each silicon solar cell can be any silicon solar cell described herein. The solar cell module also has a first cell interconnect that includes a first marker strip, a back sheet in electrical contact with both the front bus layer of the first silicon solar cell and the rear marker layer of the second silicon solar cell (rear sheet), rear encapsulant layer on the rear surface of the rear sheet. A first portion of the rear encapsulation layer is in contact with the first silicon solar cell and the second silicon solar cell, and a second portion of the rear encapsulation layer is in contact with the front encapsulation layer.

The first cell interconnect may also include a junction box in contact with the back sheet. The junction box may contain at least one bypass diode. There may also be at least one bus strip connected to the first marker strip.

In one embodiment of the present invention, a paste is disclosed. The slurry contains between 10 wt % and 70 wt % noble metal particles, at least 10 wt % intercalating particles and organic vehicles. The embedded particles include one or more selected from the group consisting of low temperature base metal particles, crystalline metal oxide particles, and glass frit particles. The weight ratio of intercalated particles to noble metal particles may be at least 1:5.

The noble metal particles may include at least one material selected from the group consisting of gold, silver, platinum, palladium, rhodium, and alloys, composites, and other combinations thereof. The noble metal particles may have a D50 between 100 nm and 50 μm and a specific surface area between 0.4 and 7.0 m ² /g. A portion of the noble metal particles may have, for example, spherical, platelet and/or elongated shapes. The noble metal particles can have a unimodal size distribution or a multimodal size distribution. In one embodiment, the noble metal particles are silver and have a D50 between 300 nm and 2.5 μm and a specific surface area between 1.0 and 3.0 m ² /g.

The embedded particles may have a D50 between 100 nm and 50 μm and a specific surface area between 0.1 and 6.0 m ² /g. A portion of the embedded particles may have, for example, spherical, platelet and/or elongated shapes. The embedded particles can have a unimodal size distribution or a multimodal size distribution.

The low temperature base metal particles may include materials selected from the group consisting of bismuth, tin, tellurium, antimony, lead, and alloys, composites, and other combinations thereof. In one embodiment, the low temperature base metal particles comprise bismuth and have a D50 between 1.5 and 4.0 μm and a specific surface area between 1.0 and 2.0 m ² /g.

In one embodiment of the invention, at least some of the low temperature base metal particles have a bismuth core particle surrounded by a single shell comprising a material selected from the group consisting of: silver, nickel, nickel-boron, tin, tellurium , antimony, lead, molybdenum, titanium, and alloys, composites, and other combinations thereof. In another embodiment of the present, at least some of the low temperature base metal particles have a bismuth core particle surrounded by a single shell comprising a material selected from the group consisting of: silicon oxide, magnesium oxide, boron oxide, and any combination thereof.

Crystalline metal oxide particles may include oxygen and metals selected from the group consisting of: bismuth, tin, tellurium, antimony, lead, vanadium, chromium, molybdenum, boron, manganese, cobalt, and alloys, composites, and other combinations thereof .

The glass frit includes a material selected from the group consisting of: antimony, arsenic, barium, bismuth, boron, cadmium, calcium, cerium, cesium, chromium, cobalt, fluorine, gallium, germanium, indium, hafnium, iodine, iron, lanthanum , lead, lithium, magnesium, manganese, molybdenum, niobium, potassium, rhenium, selenium, silicon, sodium, strontium, tellurium, tin, vanadium, zinc, zirconium, their alloys, their oxides, their composites, and other combinations thereof .

The slurry may have a solids loading between 30 wt% and 80 wt%. The embedded particles may make up at least 15 wt% of the slurry. In one arrangement, the slurry includes 45 wt% Ag particles, 30 wt% bismuth particles, and 25 wt% organic vehicle. In another arrangement, the slurry includes 30 wt% Ag particles, 20 wt% bismuth particles and 50 wt% organic vehicle. The slurry may have a viscosity between 10,000 to 200,000 cP at 25°C at a shear rate of 4 seconds (sec) ⁻¹ .

In one embodiment of the present invention, a co-firing method of forming a sintered multilayer stack is described. The method comprises the steps of: a) coating at least a portion of the surface of the base layer with a layer of wet metal particles, b) drying the layer of wet metal particles to form a layer of dry metal particles, and c) directly coating at least a portion of the layer of dry metal particles with a wet plug layers to form the multi-layer stack, d) drying the multi-layer stack, and e) co-firing the multi-layer stack to form the sintered multi-layer stack.

In another embodiment of the present invention, a sequential method of forming a sintered multilayer stack is described. The method comprises the steps of: a) coating at least a portion of the surface of the base layer with a layer of wet metal particles, b) drying the layer of wet metal particles to form a layer of dry metal particles, c) firing the layer of dried metal particles to form a layer of metal particles, d) directly applying a wet intercalation layer to at least a portion of the metal particle layer to form a multi-layer stack, e) drying the multi-layer stack, and f) firing the multi-layer stack to form a sintered multi-layer stack.

In one arrangement, the wet intercalation has between 10 wt % and 70 wt % noble metal particles, at least 10 wt % intercalated particles and an organic vehicle for both the co-fired method and the sequential method. The embedded particles may include one or more selected from the group consisting of low temperature base metal particles, crystalline metal oxide particles, and glass frit. The wet metal particle layer may include metal particles selected from the group consisting of: aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, titanium, steel and alloys, composites, and other combinations thereof.

In one arrangement, for both the combined firing method and the sequential method, there is an additional step before step a). An additional step includes depositing at least one dielectric layer on at least a portion of the surface of the base layer. In this arrangement, step a) comprises directly coating at least a portion of the dielectric layer with a layer of wet metal particles.

For both the combined firing method and the sequential method, each coating step may contain a method selected from the group consisting of: screen printing, gravure printing, spray deposition, slot coating, 3D printing and inkjet printing. In one arrangement, step a) comprises screen printing through a patterned screen to produce a layer of wet metal particles of variable thickness.

For both the combined firing method and the sequential method, steps b) and d) may comprise drying at a temperature below 500°C for between 1 second and 90 minutes, or at a temperature between 150°C and 300°C Between 1 second and 60 minutes. Step e) may comprise rapid heating in air to a temperature greater than 600°C for between 0.5 seconds and 60 minutes, or rapid heating in air to a temperature greater than 700°C for 0.5 to 3 seconds.

In one arrangement, for both the combined firing method and the sequential method, an additional step f) involves soldering a marker tape on a portion of the sintered multilayer stack.

The low temperature base metal particles, crystalline metal oxide particles, glass frit and layers of metal particles are described in detail above.

In another embodiment of the present invention, a method of making a solar cell comprises the steps of: a) providing a silicon wafer, b) coating at least a portion of the backside of the silicon wafer with a layer of wet aluminum particles, c) drying the layer of wet aluminum particles to form aluminum particle layer, d) directly coating at least a portion of the aluminum particle layer with wet intercalation to form a multilayer stack, e) drying the multilayer stack, f) coating the front surface of the silicon wafer with a plurality of fine grid lines and at least a front bus layer, g) drying the plurality of fine grid lines and at least one front bus layer to form a structure, and h) co-firing the structure to form a silicon solar cell.

Wet intercalation has been described above.

In one arrangement, there are additional steps between step a) and step b). Additional steps include depositing at least one dielectric layer on at least a portion of the rear surface of the silicon wafer. In this arrangement, step b) comprises directly coating at least a portion of the dielectric layer with a layer of wet aluminum particles.

Each coating step may comprise a method selected from the group consisting of: screen printing, gravure printing, jet deposition, slot coating, 3D printing, and ink jet printing. In one arrangement, step b) comprises screen printing through a patterned screen to produce a layer of wet aluminium particles of variable thickness.

For both the combined firing method and the sequential method, steps e) and g) may comprise drying at a temperature below 500°C for between 1 second and 90 minutes, or at a temperature between 150°C and 300°C Between 1 second and 60 minutes. Step h) may comprise rapid heating in air to a temperature greater than 600°C for between 0.5 seconds and 60 minutes, or rapid heating in air to a temperature greater than 700°C for 0.5 to 3 seconds.

Low temperature base metals, crystalline metal oxide particles and glass frit have been described in detail above.

Description of drawings

The foregoing and other aspects will be readily appreciated by those skilled in the art when the following description of illustrative embodiments is read in conjunction with the accompanying drawings. The figures are not drawn to scale. The drawings are illustrative only and are not intended to be exhaustive or to limit the invention.

1 is a schematic cross-sectional view of a multilayer stack prior to firing, in accordance with an embodiment of the present invention.

Figure 2 is a schematic cross-sectional view of a sintered multilayer stack in accordance with an embodiment of the present invention.

Figure 3 is a schematic cross-sectional view of a sintered multilayer stack in which the intercalation layers have separated phases.

Figure 4 is a schematic cross-sectional view of a sintered multilayer stack in which the intercalation layer has a phase divided into two sublayers.

5 is a schematic cross-sectional view of a portion of the sintered multilayer stack shown in FIG. 2 in accordance with an embodiment of the present invention.

6 is a scanning electron microscope (SEM) cross-sectional view of a co-fired multilayer stack in accordance with an embodiment of the present invention.

7 is a scanning electron microscope (SEM) cross-sectional view of a co-sintered multilayer stack with a silver-bismuth frit layer.

8 is a scanning electron microscope (SEM) cross-sectional view (in SE2 mode) of an aluminum particle layer on a silicon base layer.

FIG. 9 is a scanning electron microscope (SEM) cross-sectional view (in InLens mode) of the aluminum particle layer on the silicon-based layer shown in FIG. 8 .

10 is a scanning electron microscope (SEM) cross-sectional view (in InLens mode) of a portion of a silicon solar cell comprising a co-sintered multilayer stack.

11 is a scanning electron microscope (SEM) cross-sectional view (in SE2 mode) of the portion of the silicon solar cell shown in FIG. 10 comprising a co-sintered multilayer stack.

Figure 12 shows energy dispersive x-ray (EDX) spectra from a layer of aluminum particles and from a layer of modified aluminum particles in accordance with an embodiment of the present invention.

Figure 13 is an EDX spectrum of the surface of a rear marker layer comprising an aluminum-bismuth intercalation, in accordance with an embodiment of the present invention.

Figure 14 shows the x-ray scattering pattern from a co-sintered multilayer thin film stacked on the back marker layer of a silicon solar cell.

15 is a schematic cross-sectional view of a multilayer thin film stack including a dielectric layer prior to firing, in accordance with an embodiment of the present invention.

16 is a schematic cross-sectional view of a sintered multilayer thin film stack including a dielectric layer in accordance with an embodiment of the present invention.

Figure 17 is a plan view optical micrograph of a co-sintered multilayer thin film stack that has been bent.

Figure 18 is a screen design (not drawn to scale) that may be used during deposition of a wet metal particle layer in accordance with an embodiment of the present invention.

19 is a schematic cross-sectional view of a layer of dry metal particles having variable thicknesses deposited using the wire mesh shown in FIG. 18, in accordance with an embodiment of the present invention.

20 is a schematic cross-sectional view of a layer of modified metal particles having variable thicknesses deposited using the wire mesh shown in FIG. 18 and subsequently co-sintered in accordance with an embodiment of the present invention.

FIG. 21 is a plan view optical micrograph of the co-sintered multilayer stack shown in FIG. 20 .

Figure 22 is a cross-sectional SEM image of a portion of a sintered multilayer stack with variable thickness.

Figure 23 is a cross-sectional SEM image of a portion of a thin film of aluminum particles on a silicon-based layer having a flat thickness.

Figure 24 is a surface topology scan of a sintered multilayer stack with variable thickness.

Figure 25 is a surface topology scan of a layer of sintered aluminum particles.

Figure 26 is a schematic diagram showing the front (or illuminated) side of a silicon solar cell.

FIG. 27 is a schematic diagram showing the rear side of a silicon solar cell.

28 is a schematic cross-sectional view of a solar cell module including a sintered multilayer stack in accordance with an embodiment of the present invention.

29 is a scanning electron microscope (SEM) cross-sectional view of the backside of a solar cell including a sintered multilayer stack and soldered marker tape, in accordance with an embodiment of the present invention.

Figure 30 is a plot of the transmission line measurements of a conventional silver-on-silicon back-mark layer.

Figure 31 is a plot of transmission line measurements of a silver-bismuth intercalation on an aluminum particle layer that can be used as a back marker layer on silicon.

Detailed ways

Preferred embodiments have been shown in the context of sintering an embedded paste on a layer of metal particles. However, those skilled in the art will readily appreciate that the materials and methods disclosed herein have applications in a variety of contexts where good electrical contact with semiconductor or conductor materials is required, particularly good adhesion, high performance and low Cost is important.

All publications referenced herein are hereby incorporated by reference in their entirety for the purpose as if fully set forth herein.

Disclosed herein is the composition and use of an intercalation paste comprising precious metal particles and intercalation particles, which can be printed on a layer of metal particles to alter the properties of the layer of metal particles after they are sintered into a sintered multilayer stack . In one embodiment of the invention, the embedding paste is used to provide a solderable surface on the metal particle layer, which is not solderable by itself. Embedding pastes can also be used to improve adhesion in fired multilayer stacks or to alter the interaction of the metal particle layer with the underlying substrate. Embedding pastes are widely applicable to many applications, including transistors, light emitting diodes, and integrated circuits; however, the examples disclosed below will primarily focus on photovoltaic cells.

Definitions and Methods

6|60探测器来执行。关于操作条件的细节被描述于每个分析。烧结多层堆叠的截面SEM图像通过离子铣削(ion milling)而准备。薄环氧树脂层被涂在烧结多层堆叠的顶部且干燥至少30分钟。该样本随后传送至JEOL IB-03010CP离子铣削机，在5kV和120uA操作8小时，以从样本边缘除去80微米。铣削过的样本在SEM/EDX之前被存储在氮气套箱中。Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDX) (collectively referred to as SEM/EDX) used here used a Zeiss Gemini Ultra-55 Resolving Field Emission Scanning Electron Microscope, equipped with Bruker

6|60 detectors to perform. Details about operating conditions are described for each analysis. Cross-sectional SEM images of the sintered multilayer stack were prepared by ion milling. A thin epoxy layer was applied on top of the sintered multilayer stack and allowed to dry for at least 30 minutes. The sample was then transferred to a JEOL IB-03010CP ion milling machine operating at 5kV and 120uA for 8 hours to remove 80 microns from the edge of the sample. Milled samples were stored in nitrogen jackets prior to SEM/EDX.

The term "drying" describes a heat treatment, at or below 500°C, or below 400°C, or below 300°C, for a period between 1 second and 90 minutes or any included therein scope. The paste is typically applied to the base layer by screen printing or other deposition methods to create a "wet" layer. The wet layer can be dried to reduce or remove volatile organic materials, such as solvents, resulting in a "dry" layer.

The term "firing" describes heating at a temperature above 500°C, above 600°C, or above 700°C for a period between 1 second and 60 minutes or any range subsumed therein. The term "dried layer" describes a dried layer that has been sintered.

The term "multilayer stack" is used herein to describe a base layer having two or more layers of different materials thereon. A "fired multilayer stack" is a multilayer stack whose layers have been dried and sintered. There are multiple ways to fire this multilayer stack. The term "co-firing" is used to describe a single firing process of a multilayer stack. For example, during silicon solar cell fabrication, a layer of aluminum particle paste is first applied to a base layer and dried. Subsequently, a post-marking paste layer is coated on a portion of the dried aluminum particle layer, which is then dried, resulting in a dried aluminum particle layer and a dried post-marking layer. During co-firing, both dry layers are fired simultaneously in one step. The term "sequential firing" is used to describe the process of multiple firing of a multilayer stack. During sequential processing, the metal particle slurry is applied to the base layer, dried and then sintered. The embedding paste is then coated on a portion of the dried and sintered metal particle paste (referred to as the metal particle layer). Subsequently, the entire multilayer stack is dried and sintered a second time. It should be noted that embodiments of the present invention that describe co-sintered multilayer stacks or structures are also applicable to multilayer stacks or structures that have been sequentially sintered.

The term "intercalation" as used herein is used to describe the penetration of porous materials. In the context of the embodiments described herein, the term "intercalation" describes the penetration of material from intercalating particles in an intercalation layer into an adjacent layer of porous dry metal particles during the firing process, It results in a coating (partial or full) of embedded particle material on at least a portion of the metal particles. The term "modified metal particle layer" as used herein is used to describe this layer of sintered metal particles into which the material from the embedded particles has penetrated.

In describing the relationship between adjacent layers, the preposition "on" as used herein means that the layers may or may not be in direct physical contact with each other. For example, a layer over a base layer means that the layer is positioned directly adjacent to the base layer or indirectly over or adjacent to the base layer. A particular layer is indirectly over or adjacent to a base layer means that there may or may not be one or more additional layers between the particular layer and the base layer. In describing the relationship between adjacent layers, this use of the preposition "directly on" means that the layers are in direct physical contact with each other. For example, a layer between a layer above a base layer means that the layer is positioned directly adjacent to the base layer.

When the metal particle layer mainly contains metal A particles, it may be referred to as a "metal A particle layer". For example, when the metal particle layer mainly contains aluminum particles, it may be referred to as an aluminum particle layer. When the improved metal particle layer mainly contains metal A particles, it may be referred to as an "improved metal A particle layer". For example, when the layer of modified metal particles mainly contains aluminum particles, it may be referred to as a layer of modified aluminum particles.

The term "solderable surface" is known in the art. "Solderable surface" means a surface that can be soldered to the ribbon. One of ordinary skill in the art is familiar with changes to solderable surfaces. Examples of materials that produce solderable surfaces include, but are not limited to, tin, cadmium, gold, silver, palladium, rhodium, copper, zinc, lead, nickel, alloys thereof, combinations thereof, compositions thereof, and mixtures thereof. In one embodiment, a surface is solderable when at least 70 wt % of the surface comprises materials such as silver, gold, platinum, palladium, rhodium, and alloys, composites, and other combinations thereof.

The particles described herein can take on a variety of shapes, sizes, specific surface areas, and oxygen content. Particles may be spherical, needle-like, angular, dendritic, fibrous, flake-like, granular, irregular and nodular, as defined by ISO 3252. It should be understood that the term "spherical" as used herein refers to a generally spherical shape, and may include spherical, granular, nodular, and sometimes irregular shapes. The term "flake" means flakes, and sometimes angular, fibrous, and irregular shapes. The term "elongated" refers to needle-like, and sometimes angular, dendritic, fibrous and irregular shapes, as defined in ISO 3252:1999. Particle shape, morphology, size and size distribution generally depend on the synthesis technique. A set of particles can include a combination of particles of different shapes and sizes.

Spherical or elongated particles are typically described by their D50, specific surface area and particle size distribution. The D50 value is defined as a value at which half of the number of particles have diameters below this value and half of the number of particles have diameters above this value. Measuring the particle diameter distribution is typically performed using a laser diffraction particle size analyzer, such as a Horiba LA-950. For example, spherical particles are dispersed in a solvent where they are well separated and the spread of transmitted light is directly related to the size distribution from smallest to largest diameter. A common method to represent laser diffraction results is to report the D50 value based on the volume distribution. The statistical distribution of particle size can also be measured using a laser diffraction particle size analyzer. It is common for noble metal particles to have a unimodal or multimodal particle size distribution. In a unimodal distribution, the particle size is monodisperse and the D50 is in the center of the monomodal distribution. A multimodal particle size distribution has more than one peak (or apex) in the particle size distribution. A multimodal particle size distribution can increase the tap intensity of the powder, which typically results in a higher green film density.

In some embodiments of the invention, the particles may have a flake or elongated shape as defined above. The flakes may have a diameter between 1 μm and 100 μm or between 1 μm and 15 μm and a thickness between 100 nm and 500 nm. The elongated shape may have a diameter between 200 nm and 100 nm and a length greater than 1 μm. In another embodiment of the present invention, the particle shape is not limited; any particle shape can be used as long as its largest diameter is not greater than 50 μm, 5 μm or 1 μm.

The specific surface area of the particles can be measured using the Brunauer-Emmett-Teller (BET) method according to DIN ISO 9277, 2003-05. The specific surface areas of the particles disclosed herein, and particularly the silver and bismuth particles, were determined by the following test method: BET measurements were performed using a TriStar 3000 (from Micromeritics Instruments, Inc.), which operates based on physisorption analytical techniques. Sample preparation includes degassing to remove absorbed molecules. Nitrogen was the analytical gas and helium was used to determine the void volume of the sample tube. Micromeritics provided silica alumina for use as a reference material, along with preparation procedures and test conditions. The measurement begins by adding a known mass of reference material to the sample tube and mounting the sample tube on the BET device manifold. The thermally stable dosing manifold, sample tubes, and dedicated tubes for measuring saturation pressure (P _o ) were evacuated. When a sufficient vacuum is reached, the manifold is filled with helium (a non-absorbing gas) and the sample port is opened to determine the warm free space of the sample at room temperature. The sample tube with reference material was immersed in liquid nitrogen and cooled to around 77K and free space analysis was performed again. The adsorption saturation pressure was measured using a _Po tube, followed by nitrogen dosing into the manifold above atmospheric pressure. Nitrogen pressure and temperature are recorded, and then the sample port is opened, allowing nitrogen to be absorbed on the sample. After some time, the port is closed, allowing the absorption to reach equilibrium. The amount absorbed is the amount of nitrogen removed from the manifold minus any residual nitrogen in the sample tube. Measurement points along the absorption isotherm are used to calculate the specific area in m ² /g of the reference material; this procedure is repeated for any sample of interest, such as the particles described herein.

The particles described herein have significant thermal properties: melting point and/or softening point, both depending on the crystallinity of the material. The melting point of the particles can be determined by performing differential scanning calorimetry using a DSC2500 differential scanning calorimeter manufactured by TA Instruments and using the method described in ASTM E794-06 (2012). Melting points of crystalline materials can also be determined using a heated stage and x-ray diffraction. As the crystalline material is heated above its melting point, the diffraction peaks begin to disappear. The softening point is the temperature at which amorphous or vitreous particles begin to soften. The softening point of the glass particles can be determined using a dilatometer. The softening point can also be obtained by the fiber extension method described in ASTM C338-57.

Materials for the manufacture of sintered multilayer stacks

In one embodiment of the present invention, the base layer, metal particle paste, and embedding paste form a sintered multilayer stack. The base layer can be a solid, planar or rigid material. In one embodiment, the base layer includes at least one material selected from the group consisting of: silicon, silicon dioxide, silicon carbide, aluminum oxide, sapphire, germanium, gallium arsenide, gallium nitride, and indium phosphide. Such base layers are commonly used for the deposition of layers that make up transistors, light emitting diodes, integrated circuits and photovoltaic cells. The base layer may also be conductive and/or flexible. In another embodiment, the base layer includes at least one material selected from the group consisting of aluminum, copper, iron, nickel, titanium, steel, zinc, and alloys, composites, and other combinations thereof.

In one embodiment of the present invention, the metal particle slurry includes metal particles and an organic vehicle. In one arrangement, the metal particle paste also includes an inorganic binder, such as glass frit. In one arrangement, conventional, commercially available metal particle slurries are used. Metal pastes containing aluminum commonly used on silicon solar cells are sold by Ruxing Technology (eg RX8252H1), Monocrystal (eg EFX-39) and GigaSolar Materials (eg M7). The metal particles can include at least one of aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, titanium, or alloys, composites, or other combinations thereof. In various arrangements, the metal particles have a D50 of between 100 nm and 100 μm, between 500 nm and 50 μm, between 500 nm and 200 μm, or any range subsumed therein. The metal particles can have spherical, elongated, or flake-like shapes, and can have unimodal or multimodal size distributions. The glass frit may be included in the metal particle paste in small amounts (ie, less than 5 wt%). In one embodiment, the metal particle paste includes 70 wt% to 80 wt% aluminum particles, less than 2 wt% glass frit, and an organic vehicle.

In one embodiment of the present invention, the intercalation paste includes noble metal particles, intercalation particles, and an organic vehicle. The term "solids loading" may be used in conjunction with a slurry to describe the amount and proportion of precious metal and embedded particulate solids in the slurry. The slurries described herein also include an organic vehicle, although this is not often explicitly stated.

Embedded paste ingredients

In one embodiment of the invention, as described herein, the noble metal particles comprise at least one material selected from the group consisting of gold, silver, platinum, palladium, and rhodium, and alloys, composites, or other combinations thereof. In one embodiment, the noble metal particles comprise between 10 and 70 wt% of the slurry. In various embodiments, the noble metal particles have a D50 of between about 100 nm and 50 μm, between 300 nm and 10 μm, between 300 nm and 5 μm, or any range subsumed therein. In various embodiments, the noble metal particles have a specific surface area ranging from about 0.4 to 7.0 m ² /g or from about 1 to 5 m ² /g or any range subsumed therein. The noble metal may have an oxygen content of up to 2 wt%; the oxygen may be mixed homogeneously throughout the particle, or the oxygen may be found in the oxide shell, which has a thickness of up to 500 nm. The noble metal particles can have spherical, elongated, or flake-like shapes, and have unimodal or multimodal size distributions. Silver particles are commonly used in metallization pastes in the solar industry. In a typical embodiment, at least some of the noble metal particles are silver, have a D50 between 300 nm and 2.5 μm and a specific surface area between 1 and 3 m ² /g.

The term "embedded particle" is used to describe a particle that is deformable when heated and, when positioned adjacent to a porous layer of other metal particles, can at least partially sandwich the layer of porous metal particles and phase from the other metal particles based on the effect of heating separation. In various arrangements, the embedded particles have a D50 of between 50 nm and 50 μm, between 50 nm and 10 μm, between 300 nm and 5 μm, or any range subsumed therein. In one embodiment, the intercalating particles have a D50 of between 300 nm and 3 μm. In various embodiments, the embedded particles have a specific surface area ranging from about 0.1 to 6 m ² /g, about 0.5 to 3 m ² /g, or 0.5 to 4 m ² /g, or any range subsumed therein. According to one embodiment, the embedded particles are flake-shaped and have a specific surface area of about 1.0 to 3.0 m ² /g. The embedded particles can have spherical, elongated, or flake-like shapes, and can have unimodal or multimodal size distributions.

There are three groups of particles here that can be used as intercalating particles: lo temperature base metal particles (LTBM), crystalline metal oxide particles and glass frit particles. In some arrangements, the embedded particles include only low temperature base metal particles, or crystalline metal oxide particles or glass frit. In other arrangements, the embedded particles are a mixture of two or more particles from these groups. It is desirable that the elements embedded in the particles have low solubility and do not alloy with elements in adjacent metal particle layers.

In one embodiment, the embedded particles are low temperature base metal particles. As used herein, the term "low temperature base particle" (LTBM) is used to describe particles including, exclusively or essentially, any base metal or metal alloy having a low temperature melting point, ie, a melting point below 450°C. In some arrangements, the LTBM also contains up to 2 wt% oxygen; the oxygen may be mixed homogeneously throughout the particle, or the oxygen may be found in an oxide shell having a thickness of up to 500 nm and coated or partially coated with the particle. In some arrangements, the melting point of the LTMB is lower, eg, below 350°C or below 300°C. In one embodiment of the invention, the LTBM is made exclusively or substantially from bismuth, tin, tellurium, antimony, lead, or alloys, composites, or other combinations thereof. In one embodiment, the intercalated particles contain only bismuth and have a D50 between 1.5 and 4 μm and a specific surface area between 1 and 2 m ² /g.

In another embodiment, the LTBM intercalated particles are bismuth core particles surrounded by a metal or metal oxide shell. In another embodiment, the LTBM intercalated particles are bismuth core particles surrounded by a single shell composed of silver, nickel, nickel alloys such as nickel boron, tin, tellurium, antimony, lead, molybdenum, titanium, combinations thereof, and/or Other combinations were made. In another embodiment, the LTBM intercalation particles are bismuth core particles surrounded by a single shell, which is silicon oxide, magnesium oxide, boron oxide, or any combination thereof. Any of these shells may have a thickness ranging from 0.5 nm to 1 μm, or 0.5 nm to 200 nm, or any range subsumed therein.

In another embodiment, the intercalated particles are crystalline metal oxide particles. Metal oxides are compounds having at least one oxygen atom (the anion has an oxidation state of -2) and at least one metal atom. Many metal oxides contain multiple metal atoms, which may all be the same or may include multiple metals. A wide range of metal to oxygen atomic ratios are possible, as will be understood by those skilled in the art. When metal oxides form ordered periodic structures, they are crystalline. Such crystalline metal oxides can disperse x-ray radiation in peak patterns of different intensity characteristics of their crystalline structure. In one embodiment, the crystalline metal oxide particles consist solely of or consist essentially of oxides of at least one of the following metals: bismuth, tin, tellurium, antimony, lead, vanadium, chromium, molybdenum, boron, manganese, cobalt, and alloys, composites or other combinations thereof.

For the structures disclosed herein and described in more detail below, as the crystalline metal oxide particles are heated, if at low temperatures below the temperature at which significant interdiffusion occurs between metal particles or within the structure between different compositions, It is useful that they start to melt (ie, reach their melting point ( _TM )). The melting point of the crystalline material in the mixed layer can be determined using thermal order and x-ray diffraction; as the sample is heated above its melting point, the diffraction peak decreases and then disappears. In some typical embodiments, boron (III) oxide (B ₂ O ₃ , _TM = 450° C.), vanadium (V) oxide (V ₂ O ₅ , _TM = 690° C.), tellurium (IV) oxide (TeO ₂ , _TM = 733° C.) and bismuth(III) oxide (Bi ₂ O ₃ , _TM = 817° C.) can deform during the firing process and become embedded in the adjacent layer of porous metal particles, resulting in improved layer of metal particles. In a typical embodiment, the intercalated particles are crystalline bismuth oxide with a D50 between 50 nm and 2 μm and a specific surface area between 1 and 5 m ² /g. In another embodiment, the crystalline metal oxide particles also contain small amounts (ie, less than 10 wt%) of one or more additional elements that can adjust the melting point of the particles. Such additional elements may include, but are not limited to, silicon, germanium, lithium, sodium, potassium, magnesium, calcium, strontium, cesium, barium, zirconium, hafnium, vanadium, niobium, chromium, molybdenum, manganese, iron, cobalt, rhenium, Zinc, cadmium, gallium, indium, carbon, nitrogen, phosphorus, arsenic, antimony, sulfur, selenium, fluorine, chlorine, bromine, iodine, lanthanum and cerium.

In another embodiment, the embedded particles are glass frit. In one embodiment, the glass frit consists only of or consists essentially of oxygen and a combination of at least one of the following elements: silicon, boron, germanium, lithium, sodium, potassium, magnesium, calcium, strontium, cesium, barium, zirconium, Hafnium, vanadium, niobium, chromium, molybdenum, manganese, iron, cobalt, rhenium, zinc, cadmium, gallium, indium, tin, lead, carbon, nitrogen, phosphorus, arsenic, antimony, bismuth, sulfur, selenium, tellurium, fluorine, Chlorine, bromine, iodine, lanthanum, cerium, oxygen, and alloys, complexes, and other combinations thereof. It is useful if the glass frit has a softening point below 900°C or below 800°C to effectively deform during firing. In a typical embodiment, the intercalated particles are bismuth silicate glass frit with a D50 between 50 nm and 2 μm and a specific surface area between 1 and 5 m ² /g.

The term "organic vehicle" describes a mixture or solution of organic chemicals or compounds that assists in dissolving, dispersing and/or suspending solid components in a slurry. For the embedding pastes described herein, many different organic vehicle mixtures can be used. Such organic vehicles may or may not contain thixotropes, stabilizers, emulsifiers, thickeners, plasticizers, surfactants and/or other common additives.

The components of organic vehicles are well known to those skilled in the art. The main components of the organic vehicle include one or more binders and one or more solvents. Binders can be polymeric or monomeric organic components, or "resins," or a mixture of both. Polymeric binders can have various molecular weights and various polydispersity indices. Polymeric adhesives may comprise a combination of two different monomeric units, known as copolymers, wherein the monomeric units may be either alternating or in bulk (block copolymers). Polysaccharides are commonly used polymeric binders and include, but are not limited to, alkyl cellulose and alkyl derivatives such as methyl cellulose, ethyl cellulose, propyl cellulose, butyl cellulose, ethyl hydroxyethyl Cellulose bases, cellulose derivatives and mixtures thereof. Other polymeric binders include, but are not limited to, polyesters, polyethylenes, polypropylenes, polycarbonates, polyurethanes, polyacrylates (including polymethacrylates and polymethylmethacrylates), polyethylenes (including polychlorides) ethylene, polyvinylpyrrolidone, polyvinyl butyral, polyvinyl acetate), polyamides, polyglycols (including polyethylene glycols), phenolic resins, polyterpenes, derivatives thereof, and combinations thereof. The organic vehicle binder may comprise between 1 and 30 wt% binder.

Solvents are organic species that are typically removed from the slurry during processing by thermal means such as evaporation. In general, solvents that can be used in the slurries described herein include, but are not limited to, polar, non-polar, protic, aprotic, aromatic, non-aromatic, chlorinated, and non-chlorinated solvents. Solvents that can be used in the slurries described herein include, but are not limited to, alcohols, diols (including ethylene glycol), polyols (including glycerol), mono- and polyethers, mono- and polyesters, alcohol ethers, Alcohol esters, mono- and disubstituted adipates, mono- and polyacetates, ether acetates, glycol acetates, glycol ethers (including ethylene glycol monobutyl ether, diethylene glycol Alcohol monobutyl ether, triethylene glycol monobutyl ether), ethylene glycol ethyl ether acetate (including ethylene glycol monobutyl ether acetate), linear or branched saturated and unsaturated alkyl chains (including butane, pentane, hexane, octane, and decane), terpenes (including alpha-, beta-, gamma- and 4-terpineol), 2,2,4-trimethyl-1,3-pentane Diol monoisobutyrate (also known as texanol ^™ ), 2-(2-ethoxyethoxy)ethanol (also known as carbitol ^™ ), derivatives, combinations and mixtures thereof.

来改变浆料流变。在另一个示例中，可通过改变粘合剂和触变剂且考虑将在退火期间发生的峰值烧制温度、烧制轮廓(firing profile)和气流而增加或降低有机载体的碳含量。还可包括些微的添加剂。这种添加剂包括但不限于，触变剂和表面活化剂。这种添加剂是本领域熟知的，且可通过常规实验确定这种成分的有用量，以最大化装置效率和可靠性。在一个实施方式中，金属化浆料具有在25℃且在4秒^-1的剪切速度下具有10,000至200,000cP之间的粘度，使用温度受控的Brookfield RVDV-II+Pro粘度计测量。In one arrangement, the organic vehicle includes between 70-100 wt% solvent. The ratio and composition of binder, solvent and any additives can be adjusted to achieve the desired dispersion or suspension of the slurry particles, the desired carbon content and/or the desired rheological properties, as will be understood by those skilled in the art of. For example, by adding a thixotropic agent such as Thixatrol

to change the slurry rheology. In another example, the carbon content of the organic support can be increased or decreased by changing the binder and thixotropic agent and taking into account the peak firing temperature, firing profile and gas flow that will occur during annealing. Minor additives may also be included. Such additives include, but are not limited to, thixotropic agents and surfactants. Such additives are well known in the art, and useful amounts of such ingredients can be determined through routine experimentation to maximize device efficiency and reliability. In one embodiment, the metallization slurry has a viscosity between 10,000 and 200,000 cP at 25°C and a shear rate of 4 sec ^-1 , as measured using a temperature controlled Brookfield RVDV-II+Pro viscometer.

Embedded paste formulation

Exemplary compositional ranges for embedded slurries in accordance with some embodiments of the present invention are shown in Table I. In various embodiments, the intercalation slurry has a solids loading of between 30 wt% and 80 wt%, a noble metal particle composition of between 10 wt% and 70 wt% of the intercalation slurry, at least 10 wt%, 15 wt%, 20 wt% of the intercalation slurry %, 25 wt %, 30 wt % or 40 wt % of intercalated particles, and the weight ratio of intercalated particles to noble metal particles is at least 1:5. In an exemplary embodiment, the noble metal particle content is 50 wt% and the embedded particle composition is at least 10 wt% of the embedded slurry. In various embodiments, the weight ratio of embedded particles to noble metal particles in the embedded slurry is at least 1:5, or 2:5, or 3:5 or 1:1 or 5:2.

Table I

Embedding paste formulation, in weight percent (wt%)

Slurry Type precious metal particles embedded particles organic carrier Embedding paste (range I) 10-70 10-50 20-70 Embedded Paste (Scope II) 20-50 10-35 30-60 Embedded paste A 50 12.5 37.5 Embedded paste B 45 30 25 Embedded paste C 45 30 25 Embedded paste D 30 20 50

In one embodiment of the present invention, for solar cell applications, the intercalation paste comprises between 20 and 50 wt % noble metal particles (ie, intercalation paste range II in Table I) and between 10 and 35 wt % intercalating particles, It may include LTBM, crystalline metal oxides, glass frits, or combinations thereof. In one embodiment, the intercalating particles are metallic bismuth particles. Intercalation Paste A (Table I) may contain 50 wt% silver particles, 12.5 wt% bismuth particles, and 37.5 wt% organic vehicle, resulting in a 1:4 (weight) ratio of intercalation particles to precious metal particles. Intercalation Paste C (Table I) may contain 45 wt% silver particles, 30 wt% bismuth particles, and 25 wt% organic vehicle, resulting in a 1:1.5 (weight) ratio of intercalation particles to noble metal particles. When the intercalation paste includes silver and bismuth particles, use the annotation Ag:Bi.

In another embodiment, the embedded particles are glass frit. Intercalation Paste B (Table I) may contain 45 wt% silver particles, 30 wt% bismuth-based glass frit, and 25 wt% organic vehicle, resulting in a 1:1.5 (weight) ratio of intercalation particles to precious metal particles. In another embodiment, the embedded particles are a mixture of LTBM, crystalline metal oxide particles and glass frit. Intercalation Paste D (Table I) may contain 30 wt% silver particles, 15 wt% metallic bismuth particles, 5 wt% high lead content glass frit, and 50 wt% organic vehicle. The formulation of the embedding paste can be adjusted to achieve the desired bulk resistance, contact resistance, layer thickness and/or peel strength for a particular metal layer.

In another embodiment of the present invention, a method of forming an intercalation slurry includes the steps of providing precious metal particles, providing intercalating particles, and mixing the precious metal particles and intercalating particles together in an organic vehicle. In one arrangement, the embedded particles are added to the organic carrier and mixed in a planetary mixer (eg, Thinky AR-100), followed by the noble metal particles (and additional organic carrier, if desired) added and mixed in the planetary mixer mix. The embedded slurry may or may not be subsequently ground, eg, by using a three roll mill (eg, Exakt 50I). In one arrangement, the intercalation paste comprises between 10 and 70 wt % noble metal particles and greater than 10 wt % intercalating particles.

Method of forming a sintered multilayer stack

In one embodiment of the invention, the sintered multilayer stack includes a base layer having at least one metal particle layer and at least one intercalation layer thereon. In one embodiment, the sintered multilayer stack is formed using a combined firing process comprising the steps of: applying a layer of metal particles to the surface of the base layer, drying the layer of metal particles, applying an intercalation layer directly on a portion of the dried layer of metal particles, The intercalation is dried, and the multilayer stack is then co-fired. In another embodiment, a sintered multilayer stack is formed using a sequential firing process comprising the steps of: coating the surface of the base layer with a layer of metal particles, drying the layer of metal particles, firing the layer of metal particles, sintering a portion of the layer of metal particles The intercalation layer is directly coated on top, the intercalation layer is dried and the multilayer stack is then fired. In one embodiment, during firing, a portion of the intercalation layer penetrates into the metal particle layer, thereby converting the metal particle layer to a modified metal particle layer. In some embodiments, each coating step includes a method independently selected from the group consisting of: screen printing, gravure printing, jet deposition, slot coating, 3D printing, and inkjet printing. In one embodiment, the metal particle layer is applied to a portion of the base layer by screen printing a metal particle paste, and the intercalation layer, after being dried, is directly applied to a portion of the metal particle layer by screen printing the embedding paste . In one embodiment, a portion of the base layer surface is covered by at least one dielectric layer, and the metal particle layer is coated on a portion of the dielectric layer.

Dry and sintered multilayer stacks

1 is a schematic cross-sectional view showing a multilayer stack 100 prior to co-sintering, in accordance with an embodiment of the present invention. The dry metal particle layer 120 is directly on a portion of the base layer 110 . Intercalation layer 130, consisting of intercalated particles and noble metal particles, is directly on a portion of dry metal particle layer 120, as described above. In various embodiments of the present invention, the intercalation layer 130 has an average thickness of between 0.25 μm and 50 μm, between 1 μm and 25 μm, between 1 μm and 10 μm, or any range subsumed therein. In one embodiment of the present invention, the intercalation layer 130 includes noble metal particles, intercalating particles, and an optional organic binder (which may remain in the intercalation layer 130 after drying). The noble metal particles and intercalated particles may be homogeneously distributed in the intercalation layer 130 prior to co-firing. In one arrangement, the noble metal particles and embedded particles are not deformed after drying (and prior to firing), maintaining their original size and shape.

In one embodiment of the invention, the dry metal particle layer 120 is porous and includes at least one of aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, titanium, and alloys, composites, or other combinations thereof. In one arrangement, prior to co-firing, the dried metal particle layer 120 includes metal particles, and may or may not include an organic binder, and may or may not include non-metal particles, such as glass frits. Metal particles typically do not deform after drying (and before firing), retaining their original size and shape.

During firing, the intercalated particles from the intercalation layer 130 are embedded in a portion of the dried metal particle layer 120 of the adjacent intercalation layer 130 (shown below in FIG. 1 ). The portion of the dry metal particle layer 120 adjacent to the intercalation layer 130 and into which the embedded particulate material penetrates is referred to as the "modified metal particle layer" for the purposes of this disclosure. After firing, the remainder of the dried metal layer 120, which is not adjacent to the intercalation and has no or only traces of intercalated metallic material penetrated therein, is referred to as the "metal particle layer" for the purposes of this disclosure. Purpose. In one arrangement, the particles in the dried metal particle layer 120 may be sintered or melted during firing such that the metal particle layer has a different morphology and less porosity than the dried metal particle layer 120 . The modification and sintering of the multilayer stack that occurs during firing will be discussed in more detail below.

FIG. 2 is a schematic cross-sectional view illustrating sintered multilayer stack 200 (the structure 100 of FIG. 1 after it has been sintered) in accordance with an embodiment of the present invention. The sintered multilayer stack 200 includes a modified (due to firing) metal particle layer 222 of at least a portion of the adjacent base layer 210 , and a modified (due to firing) intercalation layer 230 of the adjacent modified metal particle layer 222 . During firing, at least a portion of the noble metal particles and the intercalated particles in the intercalation layer (shown as 130 in FIG. 1 prior to firing) form phases that are phase separated from each other. The noble metal particles can sinter or melt, changing morphology and reducing the porosity of the modified intercalation layer 230 . At least a portion of the embedded particles melt and flow or embed the adjacent modified metal particle layer 222 , with at least a portion of the noble metal particles (which may be sintered or melted) moving toward the solderable surface 230S of the modified intercalation layer 230 . The modified metal particle layer 222 includes metal particles into which the particle-embedded material from the intercalation layer (prior to firing, shown as 130 in FIG. 1 ) has penetrated, modifying the dry metal layer (prior to firing, at Material properties of a portion shown as 120) in FIG. 1 to form the modified metal particle layer 222. The material from the embedded particles can loosely connect the metal particles filled in the modified metal particle layer 222, or it can coat the metal particles in the modified metal particle layer 222 that are already in contact with each other.

In some arrangements, there is also a metal particle layer 220 into which little or only trace amounts of embedded particle material have penetrated. In one arrangement, the metal particle layer 220, which is not in direct contact with the modified intercalation layer 230, does not contain an increased concentration of elements from the intercalated particles. In some arrangements, metal particle layer 220 and modified metal particle layer 222 form a mixture having base layer 210 or doped base layer 210 during co-firing (not shown). Although FIG. 2 indicates a sharp boundary between the metal particle layer 220 and the modified metal particle layer 222, it should be understood that the boundary is generally not sharp. In some arrangements, the boundary is determined by modifying the lateral spread of the intercalation 230 material into the metal particle layer 220 during co-firing.

In some embodiments of the present invention, the material in the modified intercalation layer 230 in FIG. 2 is a phase divided into a phase comprising material from intercalated particles and a phase comprising a noble metal. FIG. 3 is a schematic cross-sectional view showing a sintered multilayer stack 390 (equivalent to the structure 200 of FIG. 2 ) with the modified intercalation layer 330 having a separated phase. Sintered multilayer stack 390 (in multilayer stack region 350 only) includes modified (during firing) metal particles in multilayer stack region 350 between a portion of base layer 300 and (during firing) modified intercalation layer 330 Layer 322. The metal particle layer 320 including the metal particles 392 is on the base layer 300 of the adjacent multilayer stack region 350 .

The modified intercalation layer 330 includes two phases: a noble metal phase 335 and an embedded phase 333, and has a solderable surface 335S. The majority (at least greater than 50%) of the solderable surface consists of the noble metal phase 335. In some arrangements, the noble metal phase 335 and the embedded phase 333 are not completely phase separated during firing, so that there is still some embedded phase 333 on the solderable surface 335S. The modified metal particle layer 322 includes metal particles 392 and a portion of the material from the embedded phase 333 . There is an interface 322I between the modified intercalation layer 330 and adjacent metal particles 392 in the modified metal particle layer 322 . Interface 322I may not be smooth and depends on the size and shape of metal particles 392 and firing conditions. In embodiments where the optional glass frit is already included in the dried metal particle layer (120 in FIG. 1) prior to firing, the modified metal particle layer 322 and the metal particle layer 320 may also include a small amount of glass frit (not shown in Figure 1). shown), which constitutes less than 3 wt% of the layer. .

In other embodiments, the phase separation of the materials in the intercalation layer 230 of FIG. 2 is modified to form a layered structure. FIG. 4 is a schematic cross-sectional view showing a sintered multilayer stack 400 (equivalent to structure 200 of FIG. 2 ) including an intercalation layer with two sub-layers. Sintered multilayer stack 400 (in multilayer stack region 450 only) includes modified (during firing) metal in multilayer stack region 450 between a portion of base layer 410 and modified (during firing) intercalation layer 430 Particle layer 422 . The metal particle layer 420 containing the metal particles 402 is on the base layer 410 of the adjacent multi-layer stack region 450 .

The modified intercalation layer 430 includes two sublayers: a subintercalation layer 433 directly on the modified metal particle layer 422 , and a noble metal sublayer 435 directly on the subintercalation layer 433 . The noble metal sublayer 435 has a solderable surface 435S. The modified metal particle layer 422 contains metal particles 402 and some material 403 from the subintercalation layer 433 . There is an interface 422I between the modified intercalation layer 430 (or sub-intercalation layer 433 ) and the topmost metal particles 402 in the modified metal particle layer 422 . In embodiments where the optional glass frit is already included in the dried metal particle layer (120 in FIG. 1 ) prior to firing, the modified metal particle layer 422 and the metal particle layer 420 may also include a small amount of glass frit (not shown in FIG. 1 ). shown), which constitutes less than 3 wt% of the layer.

Cross-sectional SEM images were used to identify the layers and measure the layer thicknesses in the multilayer stack. The average layer thickness of each layer in the multilayer stack is obtained by averaging at least ten thickness measurements, each at least 10 μm separation, across the cross-sectional images. In various embodiments of the present invention, the metal particle layer (eg, 220 in FIG. 2 ) has or is comprised between 0.5 μm and 100 μm, between 1 μm and 50 μm, between 2 μm and 40 μm, between 20 μm and 30 μm any range of average thicknesses. This metal particle layer on the base layer is typically smooth, with a minimum and maximum layer thickness within 20% of the average metal particle layer thickness over a 1x1 mm area. In addition to cross-sectional SEM, layer thickness and variation over the described area can be accurately measured using an Olympus LEXTOLS4000 3D laser measuring microscope and/or a profilometer, eg a Veeco Dektak 150.

In a typical embodiment, the metal particle layer (eg, 220 in FIG. 2 ) is made of sinterable aluminum particles and has an average thickness of 25 μm. The porosity of the metal particle layer can be measured in the range between 0.01 kPa and 2 Mpa using a mercury porosimeter, eg CE Instruments Pascal 140 (low pressure) or Pascal 440 (high pressure). The layer of sintered metal particles may have a porosity of between 1% and 50%, between 2% and 30%, between 3% and 20%, or any range contained therein. Sintered metal particle layers made from aluminum particles and used in solar applications can have a porosity between 10% and 18%.

The thicknesses of the subintercalation and noble metal sublayers, eg schematically shown as 433 and 435 respectively in FIG. 4, were measured in the actual multilayer stack using cross-sectional SEM/EDX. The sublayers are distinguished in the SEM due to contrasting differences between the intercalation and noble metal phases. EDX mapping is used to identify the interface location, shown as 432I in FIG. 4 . In various embodiments, the noble metal sublayer has a thickness of between 0.5 μm and 10 μm, between 0.5 μm and 5 μm, between 1 μm and 4 μm, or any range subsumed therein. In various embodiments, the subintercalation layer has a thickness of between 0.01 μm and 5 μm, between 0.25 μm and 5 μm, between 0.5 μm and 2 μm, or any range subsumed therein.

In one embodiment of the present invention, the modified intercalation comprises two phases: a noble metal phase and an intercalation phase. This structure is shown in detail in FIG. 4 . Typically, the embedded phase is not solderable, so it is useful to ensure solderability if the solderable surface 230S contains a substantial portion of the precious metal phase. In various arrangements, the solderable surface contains greater than 50%, greater than 60%, or greater than 70% noble metal phase. In one arrangement, the solderable surface of the modified intercalator contains a majority of the precious metal(s). The plan view EDX was used to determine the concentration of elements on the surface of the modified intercalation. SEM/EDX was performed using the equipment disclosed above and at an accelerating voltage of 10 kV, with a 7 mm sample working distance and 500X magnification. In various embodiments, at least 70 wt %, at least 80 wt %, at least 90 wt %, at least 95 wt %, or at least 98 wt % of the solderable surface 230S of the modified intercalation layer 230 comprise gold, silver, platinum, palladium, rhodium, and the like One or more of alloys, composites, and other combinations. Firing conditions, intercalation particle and noble metal particle types and sizes all reflect the degree of phase separation in the modified intercalation morphology.

The modified metal particle layer (shown as 222 in Figure 2) contains a higher concentration of embedded particle material than the metal particle layer (shown as 220 in Figure 2). Comparison of EDX spectra obtained from cross-sections of the modified metal particle layer and the metal particle layer in the actual multilayer stack can be used to determine the concentration of material from the modified intercalation that has been embedded in the modified metal particle layer. The SEM/EDX apparatus described above, operating at 20 kV with a working distance of 7 mm, was used to measure metal (eg, bismuth) from embedded particles versus total metal (eg, bismuth plus aluminum) in cross-sectional samples of modified metal particle layers. ) ratio. The weight ratio (embedded metal to total metal ratio) is called the IM:M ratio. Baseline EDX analysis was performed in the region of the metal particle layer, which was at least 500 μm away from the modified metal particle layer to ensure reproducible measurements. A second EDX spectrum was obtained from the modified metal particle layer and the spectra were compared. In the determination of the IM:M ratio, only peaks from metal elements were considered (ie, peaks from carbon, sulfur and oxygen were ignored). When analyzing ratios, precious metals and any metallic elements from the base layer are excluded, preventing unreliable results. In one embodiment, when the dry metal particle layer (shown as 120 in FIG. 1 ) contains aluminum particles and the intercalation layer 130 contains bismuth and silver particles, the metal particle layer (ie, after firing) contains approximately 1 wt % bismuth and greater than 98 wt% aluminum, with a Bi:(Al+Bi)(IM:M) ratio of 1:99. Other modified metal particle layers with an embedded metal composition of less than 0.25 wt% are not considered in the calculation of the IM:M ratio. In various other embodiments, the IM:M ratio is 1:106, 1:1000, ¹ :100, 1:50, 1:25, or 1:10.

It should be noted that the base layers can be somewhat rough, which results in a rough interface with them as well. 5 is a schematic cross-sectional view showing a portion of this base layer 510, modified metal particle layer 522, and modified intercalation layer 530, in accordance with an embodiment of the present invention. There is a non-planar interface 501B between the base layer 510 and the modified metal particle layer 522 . There is a non-planar interface 522B between the modified metal particle layer 520 and the modified intercalation layer 530 . Line 502 indicates the deepest intrusion of sublayer 510 into modified metal particle layer 522 . Line 504 indicates the deepest penetration of modified intercalation layer 530 into modified metal particle layer 522 . The region of modified metal particle layer 522 between lines 502 and 504 may be referred to as sample region 522A. In determining the IM:M ratio in the modified metal particle layer 522, it is useful to limit this analysis to the sample region 522A, thereby avoiding spurious results due to interface roughness.

In exemplary embodiments, the IM:M in the modified metal particle layer is 20%, 50% higher, 100% higher, 200% higher than in the metal particle layer (in a region at least 500 μm away from the modified metal particle layer) , 500% higher, or 1000% higher. In an exemplary embodiment, the intercalation comprising bismuth particles is on the layer of aluminum particles, and the layer of modified metal particles (as analyzed in the sample area, eg shown as 522A in FIG. 5 ) comprises 4 wt% bismuth and 96 wt% aluminum with a Bi:(Al+Bi) (or IM:M) ratio of 1:25. The Bi:(Al+Bi) ratio in the improved metal particle layer was 400% higher than that in the metal particle layer.

When the intercalation layer contains crystalline metal oxides and/or glass frits, which contain more than one metal, the intercalated metal components are quantified by EDX and summed to determine the IM:M ratio. For example, if the frit contains both bismuth and lead, then the ratio is defined as (Bi+Pb):(Bi+Pb+Al).

In various embodiments, the sintered multilayer stack further includes a solid mixed layer formed from the interaction between the dried metal particle layer and the metal particles in the base layer during firing. Solid mixed layers may include, but are not limited to, alloys, eutectics, composites, mixtures, or combinations thereof. In one arrangement, the layer of modified metal particles and the base layer form a solid mixed (multi) region at their interface. The solid mixed (multi) region may contain one or more alloys. The solids mixing (multi) zone can be continuous (one layer) or semi-continuous. Depending on the composition of the base layer and metal particle layer, (poly)alloys or other mixtures formed may include aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, titanium, silicon, oxygen, carbon, germanium, gallium, arsenic, One or more of indium and phosphorous. For example, aluminum and silicon can form a eutectic above 660°C, which upon cooling brings a solid aluminum-silicon (Al-Si) eutectic layer at the silicon interface. In an exemplary embodiment, the solid mixture layer is a solid Al-Si eutectic layer formed on a portion of the silicon base layer. The formation and morphology of solid Al-Si eutectic layers are well known in silicon solar cells. In another embodiment, the base layer is doped with at least one of aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, titanium, and alloys, compositions, and other combinations thereof. In one example, aluminum is a p-type dopant in silicon, and during firing, aluminum in a layer of aluminum particles from an adjacent base layer provides more aluminum dopant to form a height in the silicon base layer p-type doped regions, which are known as back surface fields.

Depending on atmospheric conditions, the embedded particles undergo multiple phase changes as they melt and become embedded in the layer of modified metal particles in the sintered multilayer stack. Depending on the materials in the layer of modified metal particles and the base layer, the embedded particles also form a crystalline mixture as they are embedded in the layer of modified metal particles. This crystalline mixture can improve the cohesion between the metal particles in the metal particle layer, prevent the interdiffusion of specific elements, and/or reduce the electrical contact resistance between the metal layers in the sintered multilayer stack. In one embodiment, the modified intercalation and modified metal particle layers comprise crystals consisting of at least one of bismuth and oxygen, silicon and silver, and alloys, composites, and other combinations thereof.

In one embodiment, the noble metal phase includes at least one material selected from the group consisting of gold, silver, platinum, palladium, rhodium, and alloys thereof, composites thereof, and other combinations thereof. In one arrangement, the noble metal phase essentially comprises one or more of these materials. When one of these materials constitutes the majority of the noble metal phase, the noble metal phase is described as rich in that material. For example, if the noble metal phase, noble metal layer, or noble metal sub-layer contains a substantial portion of silver, it may be referred to as a silver-rich region, silver-rich layer, or silver-rich sub-layer, respectively.

The intercalation phase contains elements from intercalated particles, and may also contain elements from the external environment (eg, oxygen) and, to a lesser extent, elements from adjacent metal particle layers and noble metal particles of a nearby base layer that have been integrated during firing. The broad array of elements that can be in the intercalation phase depends on whether low temperature base metals, crystalline metal oxides and/or glass frits are used as intercalating particles. In one embodiment (when the intercalating particles are low temperature base metals only), the intercalating phase comprises at least one material selected from the group consisting of: bismuth, boron, tin, tellurium, antimony, lead, oxygen, and alloys thereof, Synthetic and other combinations. In another embodiment (when the intercalating particles are only crystalline metal oxides), the intercalating phase comprises at least one material selected from the group consisting of: bismuth oxide, tin, tellurium, antimony, lead, vanadium chromium, molybdenum , boron, manganese, cobalt and their alloys, composites and other combinations. In another embodiment (when the intercalating particles are glass frits only), the intercalating phase comprises oxygen and at least one of the following elements: silicon, boron, germanium, lithium, sodium, potassium, magnesium, calcium, strontium, cesium, barium, zirconium, hafnium, vanadium, niobium, chromium, molybdenum, manganese, iron, cobalt, rhenium, zinc, cadmium, gallium, indium, tin, lead, carbon, nitrogen, phosphorus, arsenic, antimony, bismuth, sulfur, selenium, Tellurium, fluorine, chlorine, bromine, iodine, lanthanum, cerium, and alloys, composites, and other combinations thereof. When one of these materials makes up the bulk of the embedded region, the embedded region is described as rich in that material. For example, if an intercalation region, intercalation, or subintercalation contains a substantial portion of bismuth, it may be referred to as a bismuth-rich region, a bismuth-rich layer, or a bismuth-rich sublayer, respectively.

Examples and applications of sintered multilayer stacks

Most metal particles including aluminum, copper, iron, nickel, molybdenum, tungsten, tantalum, and titanium cannot be soldered using mildly activated (RMA) fluxes and tin-based solders after firing. In solar cells and other devices, however, solder ribbons are highly desirable to make electrical contact with layers of metal particles, such as layers of aluminum particles. As disclosed herein, inventive embedding pastes containing precious metals, such as silver and gold, can be used on a layer of metal particles and sintered in air to produce a highly solderable surface. This is in contrast to other attempts to increase the solderability of the metal particle layer by adding precious metals, which usually interdiffuse with the metal particle layer (eg, aluminum) due to the firing of the precious metal based on the multilayer stack, resulting in the inclusion of very little precious metal Solderable surface for good soldering. For example, firing a commercially available silver post-marking paste containing less than 10 wt% glass frit on a layer of aluminum particles does not result in a solderable surface. These layers experience significant silver-aluminum interdiffusion during the firing step and result in a non-solderable silver-aluminum surface.

As disclosed herein, intercalation can be used to modify the material properties of the metal particle layer to 1) block the diffusion of noble metals and provide a solderable surface, 2) mechanically strengthen the metal particle layer, and 3) aid in etching the metal particles The layers below the layer. In one embodiment of the present invention, an intercalation paste is used to form a multi-layer stack comprising precious metal particles made of silver and intercalated particles made of bismuth metal or bismuth-based glass frits, and an adjacent layer of metal particles comprising aluminum particles . The sintered multilayer stack was formed by: screen printing an aluminum paste (usually used in solar cell applications) on a bare silicon wafer, drying the sample at 250°C for 30 seconds, screen printing embedding on a portion of the dried aluminum particle layer Slurry, drying samples at 250°C for 30 seconds, and co-firing samples, using spike fire profiles with peak temperatures between 700°C and 820°C and ramp up and cooling rates greater than 10°C/sec. All drying and firing steps were performed using a Despatch CDF7210 furnace, which is commonly used in silicon solar manufacturing.

SEM/EDS analysis was used to determine the elemental composition of various regions in the fired multilayer stack of polished cross-sections and to study the intercalation process. SEM/EDX was performed using the previously described apparatus using two different modes of operation. SEM micrographs were taken using a Zeiss Gemini Ultra-55 analytical field emission SEM using two modes called SE2 and Inlens. The SE2 mode was operated at 5-10 kV and a working distance of 5-7 mm, using a SE2 second electron detector and a scan cycle time of 10 seconds. Brightness and contrast were varied between 0 and 50% and between 0 and 60%, respectively, in order to maximize the contrast between the embedded regions and Al particles. The Inlens mode was operated at 1-3 kV and a working distance of 3-7 mm, using an InLens second electron detector and a scan cycle time of 10 seconds. To shoot BSF in Inlens mode, the brightness is set to 0% and the contrast is set to around 40%.

In one embodiment of the present invention, an intercalation paste comprising 10-15 wt% intercalation particles blocks interdiffusion between noble metal (ie, silver) and metal particles (ie, aluminum). Intercalation Paste A (shown in Table I) contained 12.5 wt% bismuth particles and 50 wt% Ag, resulting in a 1 :4 weight ratio of intercalation particles to noble metal particles. Sintered multilayer stacks were prepared as described above. The SEM of the sintered multilayer stack was performed in SE2 mode, using the equipment described above, at an accelerating voltage of 5 kV, a working distance of 7 mm and a magnification of 4000 times.

Figure 6 is a SEM cross-sectional view of a co-sintered multilayer stack. The modified intercalation layer 630 is directly on the modified metal particle layer 622 . The modified intercalation layer 630 includes a bismuth-rich sublayer 632 (intercalation phase), which includes bismuth oxide, and a silver-rich sublayer 634 (noble metal phase). The modified metal particle layer 622 includes aluminum particles 621 and intercalation phase material 623 that have diffused out of the bismuth-rich sublayer 632 . The bismuth-rich sublayer 632 is directly on the aluminum particles 621 , at least near the interface region 631 . The bismuth-rich sublayer 632 appears to prevent interdiffusion of silver from the modified intercalation layer 630 and aluminum from the modified metal particle layer 622 during the co-firing process. FIG. 6 is an example of the layered structure described above in FIG. 4 . The silver rich sublayer 634 provides a highly solderable surface (away from the modified metal particle layer 622). The embedded phase material 623 does not penetrate far into the modified metal particle layer 622 . The modified metal particle layer 622 contains mostly aluminum particles, which are weakly sintered together after co-firing and have poor mechanical strength. There is not enough bismuth available to penetrate deeply into the modified metal particle layer 622, and the bismuth-rich sublayer 632 can apply pressure to the modified metal particle layer 622, which can mechanically weaken the co-sintered multilayer stack. The peel strength of this co-sintered multilayer stack is below 0.4 N/mm (Newtons per mm) with the dominant failure mechanism between Al particles. Existing solar industry standards requiring peel strengths greater than 1 N/mm are considered for commercial viability.

Embedding Paste B (shown in Table I) used glass frit as the embedding particles to achieve a solderable surface. Intercalation Paste B contained 30 wt% bismuth-based glass frit (intercalation) particles and 45 wt% Ag, resulting in a 1:1.5 weight ratio of intercalation particles to noble metal particles. The glass frit mainly contained bismuth and had a glass transition temperature of 387°C and a softening point of 419°C. The SEM of the sintered multilayer stack was performed in SE2 mode, using the equipment described above, at an accelerating voltage of 5 kV, a working distance of 7 mm and a magnification of 4000 times. 7 is a SEM cross-sectional view of this co-sintered multilayer stack in accordance with an embodiment of the present invention. The improved aluminum particle layer 722 is an improved metal particle layer including the aluminum particles 730 . During co-firing, the bismuth-based frit is not completely phase-separated from the Ag particles, resulting in an improved intercalation 750 with two phases: a noble metal phase 721 and a bismuth-based intercalation phase 740, similar to the figures above shown in 3. Surface 750S on modified intercalation layer 750 contains more than 50% noble metal phase 721 . Surface 750S can be soldered using fluxes commonly used in the solar cell industry (eg, Kester 952S, Kester 951 and Alpha NR205). The overall peel strength of the sintered multilayer stack is below 0.5 N/mm, which may be due to the low penetration of the bismuth-based intercalation phase 740 into the improved aluminum particle layer 722 . In general, the morphology of the modified intercalation can be modified by changing the composition and loading of the embedded particles in the intercalation.

Embedding pastes block interdiffusion of elements and strengthen the underlying metal particle layer

The examples above show two paste formulations designed to block interdiffusion between noble technology (ie, silver) and metal particles (ie, aluminum), but their sintered layers lack sufficient mechanical strength when soldered strength. Intercalation Paste C (shown in Table I) contained 30 wt% bismuth particles and 45 wt% silver particles (ie, Ag:Bi intercalation paste), resulting in a 1:1.5 weight ratio of intercalated particles to noble metal particles. The increased intercalation particle content in the slurry results in a higher concentration of intercalation material in the modified metal particle layer and results in a mechanically stronger sintered multilayer stack. Embedding Paste C is used as a drop-in replacement for commercial silver post-marker paste during manufacture of BSF, polycrystalline, p-type solar cells. Embedding paste C may also be referred to as silver-on-aluminum (Ag-on-Al), postmark, floating postmark, or logo embedding paste. A set of characterization tools were used on the obtained sintered multilayer stacks to assess the IM:M (intercalation metal:metal) ratio, noble metal surface coverage, and to determine whether crystals formed in the intercalated regions.

The effect of intercalation on the metal particle layer is best demonstrated by first illustrating the morphology of the sintered aluminum particle layer on the unintercalated silicon-based layer. 8 is a scanning electron microscope (SEM) cross-sectional view in SE2 mode of this sintered aluminum particle layer 822 on a silicon base layer 810, along the region of a silicon solar cell that does not contain intercalation. The layer of sintered aluminum particles 822 is approximately 20 μm thick and contains aluminum particles 821 and a small amount of inorganic binder (ie, glass frit) 840 . InLens mode SEM of the same layer of aluminum particles is shown in FIG. 9 . In the InLens mode, the aluminum particle layer 922, aluminum particles 921, and silicon base layer 910 are clearly visible, in addition to the back surface field region 970 and the solidified aluminum-silicon (Al-Si) eutectic layer 980.

After co-firing, the effect of intercalation on producing a layer of modified metal particles can be understood with reference to FIG. 10 . Figure 10 is an InLens SEM interfacial view for the same silicon solar cell in the image shown in Figure 8, but along the area containing the co-sintered multilayer stack made using intercalation paste C (shown in Table 1). . Co-sintered multilayer stack 1000 includes modified intercalation layer 1030 , modified aluminum particle layer 1022 , cured Al-Si eutectic layer 1080 , aluminum doped back surface field (BSF) region 1070 , and silicon base layer 1010 . In an exemplary embodiment, the BSF in the silicon base layer is doped p-type to between 10 ¹⁷ and 10 ²⁰ per cm ³ .

The SE2 mode SEM of the co-sintered multilayer stack of FIG. 10 is shown in FIG. 11 . While the InLens mode clearly shows the BSF region, the SE2 mode is the preferred mode to reflect the bismuth (intercalation phase) in the modified aluminum particle layer. Co-sintered multilayer stack 1100 includes modified intercalation layer 1130 , modified aluminum particle layer 1122 and silicon based layer 1110 . The silver sublayer 1134 and the bismuth subintercalation 1132 in the modified intercalation layer 1130 can also be seen. The BSF region and the solidified Al-Si eutectic layer cannot be clearly seen in this image. The modified aluminum particle layer 112 contains a substantial amount of bismuth intercalation material 1103 that intercalates around the aluminum particles 1102 during co-firing. In some instances, the contrast between bismuth and silver may not be strong enough to clearly identify the sublayer and the degree of bismuth intercalation in the layer of aluminum particles. Elemental maps in these example cross-sections were obtained using SEM/EDX to fully locate silver and bismuth in the co-fired multilayer stack.

The amount of embedded metal (ie, bismuth) due to intercalation in the modified aluminum particle layer can be determined by comparing EDX spectra, from the modified aluminum particle layer region and the aluminum particle layer region in the same cross-sectional sample. This is most useful if the regions are more than 1 μm apart from each other. The way in which this comparison is made has been described above as the IM:M or Bi:(Bi+Al) ratio. This analysis is useful in determining whether intercalation pastes are used in solar cell fabrication. Metallization layers in solar cells contain a limited subset of metals including, for example, aluminum, silver, bismuth, lead, and zinc. In commercial solar cells, the layer of aluminum particles contains aluminum almost entirely.

In one example, the embedded particles in the embedded paste C contain only bismuth, and the metal particles in the metal particle layer are mostly aluminum. Comparing the ratio Bi:(Bi+Al) of bismuth to bismuth plus aluminum in an aluminum particle layer (ie, which does not interact with the intercalation paste) and the modified aluminum particle layer is useful in determining whether the intercalation paste is incorporated into a solar cell metric. The EDX spectra for these two layers were measured for approximately three minutes, using the equipment described above, at an accelerating voltage of 20 kV and a working distance of 7 mm. The EDX spectrum for sintered aluminum particle layer 822 in FIG. 8 was collected from region 898. The EDX spectrum for the modified aluminum particle layer 1122 in FIG. 11 was collected from region 1199. Elemental quantification was performed on these spectra using Bruker QuantaxEsprit 2.0 software for automatic element identification, background subtraction and peak fitting. The EDX spectrum is shown in Figure 12. The aluminum and bismuth metal peak areas were quantified and the wt% for both layers was calculated from the EDX spectra in Figure 12 and summarized in Table II below. No significant amounts of any other metals were identifiable in the EDX spectrum. The aluminum particle layer EDX spectrum shown in Figure 12A yields a Bi:(Bi+Al) wt% ratio of 1:244, and the modified aluminum particle layer spectrum shown in Figure 12B yields a Bi:(Bi of 1:4 +Al) wt% ratio, as shown in Table II. The Bi:(Bi+Al) wt % ratio in the modified aluminum particle layer 1122 is about 62 times higher than that of the sintered aluminum particle layer 822 without the Ag:Bi intercalation contact. In various embodiments, the Bi:(Bi+Al) ratio in the sintered multilayer stack is at least 20%, or at least 50%, or at least 2x higher, or at least 2x higher in the modified aluminum particle layer than in the sintered aluminum particle layer 5x or at least 10x or at least 50x.

Table II:

Al-Bismuth EDX Quantification and Results Bi:(Bi+Al)wt% Ratio

Al Bi Bi:(Bi+Al) ratio aluminum particle layer 40.290 0.166 1:244 Improved aluminum particle layer 43.641 14.974 1:3.91

Plan view EDX can be used to determine elemental concentrations on the surface of the rear marker layer in silicon solar cells. In plan view, the EDX probe area surface to a depth of about 4 μm or less, making this a useful technique for identifying the degree of interdiffusion in joint stacked multilayer stacks: higher noble metal concentrations mean that there are more Less interdiffusion, and lower noble metal concentration means more interdiffusion here. Figure 13 is a plan view EDX spectrum taken from the surface of a rear marker layer comprising Ag:Bi intercalation in accordance with an embodiment of the present invention. EDX spectra were collected using a SEM operating at an accelerating voltage of 10 kV, a working distance of 7 mm and a magnification of 500X. The main peak between 3.5 and 4 keV and the smaller peak at 0.3 keV are all identified as silver. The remaining small peaks in the spectrum are identified as follows: carbon at 0.3 keV (convoluted with a small silver peak); oxygen at 0.52 keV; aluminum at 1.48 keV; and bismuth at 2.4 keV. Elemental quantification was performed automatically using Bruker Quantax Esprit 2.0 software to subtract background, identify elemental peaks, and then fit the peak intensities of the x-ray energies. Standard weight percents for each element are shown in Table III below. The overall silver coverage on the surface of the rear logo layer was 96.3 weight percent (wt%).

Table III

Standard weight percent of elements on the surface of the rear marking layer

Intercalations comprising silver and bismuth can form multiple single crystal phases when sintered on the dried aluminum-based metal particle layer. XRD can be used to differentiate between sintered multilayer stacks using bismuth particles in the intercalation and sintered multilayer stacks using conventional silver-based marking tapes with less than 10 wt% glass frit as inorganic binder. XRD was performed using a Bruker ZXS D8 Discover GADDS x-ray diffractometer equipped with a VANTEC-500 area detector and a cobalt x-ray source operating at 35kV and 40mA. A sintered multilayer stack XRD pattern on the rear marker tape of a silicon solar cell is shown in FIG. 14 . Diffraction patterns were measured in two 25° frames combined for a total window of 25-80° in 2Θ using the cobalt Kα wavelength. Each frame was measured under x-ray exposure for 30 minutes. No background subtraction was performed on the two diffraction patterns of Figure 14. Patterns were normalized to fit the maximum peak and 0.01 background was added to the data to plot as log(intensity).

The XRD diffraction pattern shows that the sintered metal stack formed with Ag:Bi intercalation, or the back marker layer in a solar cell, has a different pattern than the one formed without bismuth. XRD pattern A is from a co-sintered multilayer stack on the back marker layer of a silicon solar cell. The co-sintered multilayer stack includes a modified intercalation layer, formed using an intercalation paste containing approximately 45 wt% silver, 30 wt% Bi, and 25 wt% organic vehicle (as used above for paste C in Table I). Peak 1410 is identified as silver and peak 1420 is bismuth _oxide ( _Bi2O3 ) crystals. XRD pattern B is from a co-sintered multi-layer stack on the back marker layer of a silicon solar cell, formed using a commercially available back marker paste containing less than 10 wt% glass frit, as intercalated on the aluminum particle layer. The co-sintered multilayer stack is dark, indicating significant silver-aluminum interdiffusion. Peak 1450 is identified as the silicon-aluminum eutectic phase. Peak 1460 is identified as the silver-aluminum alloy phase (ie, _Ag2Al ). The silver peak 1410 is observed in pattern A, accompanied by the bismuth oxide mixture, and absent in pattern B, where only silver is observed as part of the silver-aluminum alloy. This further demonstrates that bismuth prevents mutual diffusion in the sintered multilayer stack. In one embodiment, the back marker layer in a silicon solar cell comprises crystals of bismuth and at least one other element, such as silicon, silver, oxides thereof, alloys thereof, composites thereof, or other combinations thereof. In another embodiment, the back marker layer comprises bismuth oxide crystals. In another embodiment, the embedded region undergoes multiple phase transitions during firing.

Intercalation can be etched through the dielectric layer during firing

In some device applications, a dielectric layer is deposited on the base layer surface prior to metal layer deposition, thereby passivating the base layer surface and improving electronic properties. The dielectric layer also prevents interdiffusion of species between the base layer and the adjacent metal particle(s) layers. In some cases, it may be highly desirable to etch through the dielectric layer to form a mixture between the base layer and the metal particle layer to improve electrical conduction between the base layer and the metal particle layer. Glass frits containing bismuth and lead are known to etch through various dielectric layers (eg, silicon nitride) during co-firing of silicon solar cells. In an exemplary embodiment, Intercalation Paste D (from Table I above) comprises approximately 30 wt% silver, 20 wt% intercalation particles (15 wt% metallic bismuth particles, 5 wt% high lead glass frit) and 50 wt% organic vehicle . This embedding paste is particularly useful if etching through the dielectric layer is desired.

15 shows a schematic cross-sectional view of a multilayer stack 1500 including a base layer 1510 coated with at least one dielectric layer 1513 prior to firing, in accordance with an embodiment of the present invention. The dry metal particle layer 1520 is on a portion of the dielectric layer 1513 . Intercalation layer 1530, consisting of intercalated particles and noble metal particles, is directly on a portion of dry metal particle layer 1520, as described above. The noble metal particles and intercalated particles may be homogeneously distributed in the intercalation layer 1530 prior to firing. The dielectric layer includes at least one of silicon, aluminum, germanium, gallium, hafnium, and oxides thereof, nitrides thereof, composites thereof, and combinations thereof. In one arrangement, the dielectric layer 1513 is a 75 nm thick layer of silicon nitride. In another embodiment, there is a second dielectric layer (not shown) between the dielectric layer 1513 and the base layer 1510 . In one arrangement, the second dielectric layer is a 10 nm thick aluminum oxide layer directly on the base layer 1510, and the dielectric layer 1513 is a 75 nm thick silicon nitride layer directly on the aluminum oxide layer. The dry metal particle layer 1520 is formed by depositing a metal particle slurry on the dielectric layer 1513 and then drying. In one arrangement, dry metal particle layer 1520 is 20 μm thick and contains aluminum particles. Intercalation layer 1530 includes intercalated particles, such as glass frit, including lead or bismuth, deposited on dry metal particle layer 1520, covering at least a portion of dry metal particle layer 1520, and then dried.

FIG. 16 is a schematic cross-sectional view showing sintered multilayer stack 1600 (the structure 1500 of FIG. 15 after it has been sintered) in accordance with an embodiment of the present invention. A portion of the base layer 1610 is coated with at least one dielectric layer 1613 . During the co-firing, at least some of the embedded particles in the modified intercalation layer 1630 , which include the glass frit described with reference to FIG. 15 , melt and begin to flow, embedded in the modified metal particle layer 1622 . In one arrangement, material from the glass frit in the modified intercalation layer 1630 penetrates into and passes through the metal particles in the modified metal particle layer 1622 and is etched into the dielectric layer 1613 (1513 before firing), allowing from the modified metal particles Some of the metals of layer 1622 chemically and electrically interact with base layer 1610 to form one or more new mixtures 1614. Other embedded particles (eg, bismuth particles) from the modified intercalation layer 1630 may also be embedded in the modified metal particle layer 1622 and may provide structural support. In one arrangement, as described in more detail above with reference to FIG. 2, at least a portion of the noble metal particles and intercalated particles in the modified intercalation layer 1630 form phases that are phase separated from each other. In some arrangements, there is also a metal particle region 1620 (on dielectric layer 1613) into which little or only trace amounts of embedded particle material penetrate. In an exemplary embodiment, the intercalating particles are bismuth particles and glass frit, and the metal particles are aluminum.

Introduce thickness variation of metal particle layer to reduce bending

Intercalation can cause stress in the underlying modified metal particle layer during firing, which can lead to bending or wrinkling and thus poor layer strength and electrical communication between layers. For example, the intercalation layer may have a different coefficient of thermal expansion than an adjacent layer of modified metal particles, resulting in different expansion or contraction of each layer during firing. Another source of stress in adjacent layers of modified metal particles may be the embedding of molten embedded particle material between the metal particles. These pressures can cause the modified metal particle layer and/or modified intercalation layer to bend or wrinkle. Bending or wrinkling can be described as large, periodic or aperiodic deviations in layer thickness. Typically, this results in delamination between layers. For example, the initial thickness of the stack comprising the intercalation layer and the dry metal particle layer is approximately the same everywhere before the intercalation layer on the dry metal particle layer is sintered. After co-firing, the thickness of the sintered multilayer stack containing the modified intercalation and modified metal particle layers can be up to three times the original thickness in some regions.

Figure 17 is a plan view optical micrograph of a co-sintered multilayer stack in which bending has occurred. Modified intercalation 1730 is visible. The modified intercalation 1730 has been bent; some peak regions 1712 are indicated in FIG. 17 . The adjacent metal particle layer 1720 is not curved and remains smooth or approximately flat. Even if the modified intercalation layer 1730 has been deformed, the mechanical integrity of the co-sintered multilayer stack remains strong with a peel strength greater than 1 N/mm. However, bending can make it challenging to make a good, strong contact between the modified interposer 1730 and the marking tape (not shown) when they are soldered together. The curved surface of the modified interposer 1730 can result in incomplete solder wetting in the area of the modified interposer 1730, which can reduce peel strength and solder bond reliability. It is useful to reduce or eliminate bowing in the co-fired multilayer stack to ensure successful soldering to the marker tape.

Variable thicknesses can be incorporated into the sintered multilayer stack to significantly reduce bending and/or wrinkling of the individual layers. When one or more layers have variable thicknesses, uneven interfaces can be introduced between the layers. One indication of variable thickness is the non-planar interface between the sintered multilayer thin film stacks. A variable thickness is created by patterning a portion of the first layer and then printing the second layer directly on the patterned portion of the first layer, thereby creating a non-planar interface between the two layers. In one arrangement, layers of variable thickness are the result of having been printed using a patterned screen. After firing, the thickness of each layer can be reduced, but firing does not result in a layer of variable thickness becoming a layer of uniform thickness. The variable thickness in a layer can be measured and quantified using cross-sectional SEM and surface topology techniques before and after firing. In various embodiments, a layer may be described as having a variable thickness when it has a thickness variation of at least 20% greater than or at least 20% less than the average thickness of the layer as measured in a 1x1 mm area.

Figure 18 is a variable thickness screen that may be used during deposition of a metal particle slurry to achieve a dry metal particle layer in accordance with an embodiment of the present invention. Screen 1800 has open mesh regions 1810, and patterned regions 1820. Patterned area 1820 includes closed area 1821 and open area 1822 . When the screen is used during printing of the wet metal particle layer, the paste flows through the openings 1822 and the open mesh area 1810 and is blocked by the closed area 1821, which results in the deposited wet metal particle layer having a variable thickness. In one embodiment, the wet metal particle layer is then dried to form a variable thickness dry metal particle layer, and the embedding slurry is deposited directly on the variable thickness dry metal particle layer.

There are several factors that affect the variable thickness in the dry metal particle layer, such as the number of meshes, wire diameter and shape, wire angle relative to the frame, emulsion thickness, and wire mesh design. The mesh size and wire diameter determine the smallest pattern shape and opening that can be printed. The thickness variation in the dry metal particle layer is also affected by the flow of the metal particle slurry, which affects layer sliding. The slurries can be designed with high viscosity and thixotropy to precisely control where they are deposited on the substrate. It is also possible to vary the magnitude of the thickness variation in the metal particle layer by adjusting the emulsion thickness of the wire mesh. The wire mesh can be designed to ensure a continuous layer of dry metal particles on the surface of the base layer, with variable layer thicknesses either overall or only in specific areas. In an exemplary embodiment, the metal particle paste is screen printed using a 230 mesh with an emulsion thickness of 5 μm. In one arrangement, patterned area 1820 has a 100 μm series of 3 mm open area 1822 adjacent to 100 μm closed area 1821 by 3 mm. There are no restrictions on pattern type, period (or lack thereof) or size. Many patterns result in variable thickness, and patterns can be adjusted for a variety of printing conditions and paste formulations.

19 is a schematic cross-sectional view of a layer of dry metal particles of variable thickness deposited on a base layer 1910 using the screen 1800 shown in FIG. 18 in accordance with an embodiment of the present invention. The outer region 1925 of the dried metal particle layer 1920 is formed by depositing a metal particle slurry through the open mesh area 1810 of the wire mesh 1800, and then drying the metal particle slurry. A variable thickness dry metal particle layer 1922 in region 1925 is deposited through patterned region 1820 of screen 1800 and has a variable thickness. The embedding paste is then printed directly on the variable thickness dried metal particle layer 1922 in region 1925 and dried to form intercalation layer 1930.

FIG. 20 is the structure of FIG. 19 deposited on the base layer 2010 using the screen 1800 shown in FIG. 18 (the structure of FIG. 19 ) having variable thickness after it has been co-sintered in accordance with an embodiment of the present invention Schematic cross-sectional view of the dried metal particle layer. Outside region 2025 there is a metal particle layer 2020 (formed from the dried metal particle layer 1920 of Figure 19). As described above, co-firing results in material from the intercalation layer 1930 (FIG. 19) being embedded into the underlying variable thickness dry metal particle layer 1922 (FIG. 19), converting the variable thickness metal particle layer 1922 to a variable thickness The modified metal particle layer 2022 and the conversion intercalation layer 1930 are modified intercalation layers 2030 . In one arrangement, the modified metal particle layer 2022 has patterned thickness variations including, but not limited to, periodic ridges, ridges, edges, and other characteristic shapes. It should be noted that the thickness of the modified intercalation layer 2030 is generally uniform, and that the non-planar interface between the modified intercalation layer and the modified metal particle layer (due to its variable thickness) can be measured by measuring changes in the total layer thickness of the multilayer stack and infer.

FIG. 21 is a plan view optical micrograph of a co-sintered multilayer stack in which the metal particle paste is printed with variable thickness (in some areas) using a screen such as that shown in FIG. 18 . The intercalation layer is printed directly on the variable thickness regions of the metal particle layer, and the multilayer stack is co-sintered to form a modified intercalation layer 2121 on the top surface, with edges on each side of the approximately flat metal particle layer 2120. The metal particle layer 2120 has a flat top surface. The surface of the modified intercalation layer 2121 is uneven, with a pattern that reflects thickness variations in the underlying modified metal particle layer. The surface of the modified intercalation layer 2121 does not show curved or wrinkled symbols, as is clearly visible in the modified intercalation layer 1730 in FIG. 17 . In one embodiment of the invention, a portion of the co-sintered multilayer stack has a variable thickness.

Useful unit of measure to describe variable thickness in order to compare peak and valley thickness to average layer thickness. In any layer, there may be some unintentional thickness variations, but these variations are typically less than 20% of the average layer thickness. A layer can be considered flat (having a uniform thickness) if its thickness varies by less than 20% of the average layer thickness. By careful design of the screen used to print the metal particle paste, it is possible to produce a layer of variable thickness with a thickness variation measured in a 1x1 mm area that is at least 20% greater or at least 20% less than the average thickness of the layer .

The variable thickness in the sintered multilayer stack can be measured from SEM images of polished cross-sectional samples. 22 is a cross-sectional SEM image of a portion of a sintered multilayer stack 2210 having variable thicknesses, in accordance with an embodiment of the present invention. Section samples were prepared and drawn using the methods described above. The sintered multilayer stack 2210 includes a modified intercalation layer 2211 , a modified aluminum particle layer 2212 and a silicon base layer 2213 . Two interfaces on each side of the modified aluminum particle layer 2212 are identified in the image: the interface 2218 between the silicon base layer 2213 and the modified aluminum particle layer 2212, and the interface 2217 between the modified aluminum particle layer 2212 and the modified intercalation layer 2211. Solderable surface 2216 is also shown. For comparison, Figure 23 shows a silicon based layer 2322 with a flat thin film 2321 of aluminum particles that does not have a variable thickness.

The average thickness of the modified aluminum particle layer 2212 in Figure 22 was calculated from the average thickness measurements. The thickness between the two interfaces 2217 and 2218 in Figure 22 is measured at regular intervals (eg, 10 microns) across the sample. Thickness is also measured at local maxima and local minima. Software, such as ImageJ 1.50a, can be used to obtain mean thicknesses as well as minimum and maximum thicknesses. The peaks and valleys seen in a single cross-sectional sample may not be representative of the entire sintered multilayer stack. Therefore, it is useful to perform this measurement on multiple cross-sectional samples, thereby ensuring that the measurement is combined with many peaks and valleys. These methods are known to those skilled in the art.

For the sample shown in Figure 22, the modified aluminum particle layer 2212 had an average thickness of 11.3 μm, a peak thickness of 18.4 μm, and a valley thickness of 5.2 μm. The peak thickness is 64% greater than the average thickness and the valleys are 54% smaller than the average thickness. In various embodiments, the layer of variable thickness has a peak thickness that is at least 20%, at least 30%, at least 40%, or at least 50% greater than the average layer thickness. In various embodiments, the layer of variable thickness has a valley thickness that is at least 20%, at least 30%, at least 40%, or at least 50% less than the average layer thickness.

Solderable surface 2216 of modified interlayer 2211 is approximately parallel to interface 2217 when modified interlayer 2211 is continuous and approximately uniform in thickness. In one embodiment of the invention, all of the measurements described for modifying the layer of aluminum particles 2212 may be performed for the layer of modified aluminum particles 2212 and the combined thickness of the modified interlayer 2211 between the solderable surface 2216 and the interface 2217. The comparison of the thickness measurements for the two combined layers is a good approximation to compare the thickness measurements for the modified aluminum particle layer 2212 only. For the combined layer in Figure 22, the peak thickness is 44% greater than the average overall thickness of 13.2 [mu]m, and the valleys are 43% smaller than the average overall thickness. This alternative method systematically measures the thickness variation in multiple layers of the sintered stack underneath.

For some applications, only a portion of the sintered multilayer stack needs to have a variable thickness. For example, the layer of aluminum particles on the backside of a silicon solar cell is typically flat. It is useful to introduce a variable thickness in the rear logo layer portion (which includes the modified intercalation layer) on the backside of this cell. Comparing the thickness variation in a portion of the back marker layer with the thickness variation in a portion of the surrounding aluminum particle layer can be used to determine whether a layer of variable thickness is used on the backside of a solar cell.

Another useful unit of measure for determining variable thickness in a sintered multilayer thin film stack is the average valley-to-peak height, which is the difference between the average of local maxima and the average of local minima. In cross-sectional SEM images, local maxima and local minima are not guaranteed to be in the image, so surface topology metrology methods such as profilometers, coherent scanning interferometers, and zoom microscopy are more useful. An example of a profilometer is a Bruker or Veeco Dektak 150 or equivalent. Coherent scanning interferometers can be performed using an Olympus LEXT OLS4000 3D measurement microscope. The software that accompanies these methods automatically calculates the average peak-to-valley difference.

In one example embodiment, a profilometer is used to determine the average peak-to-valley height in the same sample, both for sintered multilayer stacks of variable thickness and for aluminum particle layers of uniform thickness. The Veeco Dektak 150 was used to measure surfaces in a 1x1mm area using a 12.5mm radius probe to produce a 3D topological surface map. Figure 24 is a 3D surface topology map of a sintered multilayer stack with variable thickness, and Figure 25 is a 3D surface topology map of (adjacent) layers of aluminum particles with uniform thickness. The brightest regions in the figures indicate local maxima and the darkest regions indicate local minima. Figure 24 shows the thickness variation (from -20.2 μm to 15.9 μm) that would be expected for sintered multilayer stacks comprising variable thickness modified metal particle layers. Figure 25 shows the thickness variation (from -4.9 μm to 5.5 μm) that would be expected for a layer of aluminum particles of uniform thickness. The average peak-to-valley height was calculated using the program Veeco Vision v4.20, which automatically identified and averaged local maxima and minima, and then subtracted the difference. The average peak-to-valley height was 35.54 μm for the sintered multilayer stack of FIG. 24 and 9.51 μm for the aluminum layer of FIG. 25 . In various embodiments, the layers have variable thicknesses when the average peak-to-valley height is greater than 10 μm, greater than 12 μm, or greater than 15 μm, and the layers have a uniform thickness when the average peak-to-valley height is less than 10 μm, less than 12 μm, or less than 15 μm thickness.

In one embodiment of the present invention, when a co-sintered variable thickness multilayer stack modified intercalator, such as the one shown in Figure 20, is soldered to the marking tape, its peel strength is not variable thickness Double the peel strength of sintered multilayer stacks. In one arrangement, the modified intercalation layers on the surface of this variable thickness sintered multilayer stack are soldered to tin based marking tapes and they have greater than 1.5N/mm, or greater than 2N/mm, or greater than 3N /mm peel strength. Thickness variation can be optimized to provide a continuous layer of metal particles and a back surface field on the base layer for silicon solar cells. The thickness variation can be optimized so that the contact resistance of this co-sintered variable thickness multilayer stack is equal to or lower than that of the approximately flat co-sintered multilayer stack. In an exemplary embodiment, the thickness variation in the modified and modified metal particle layer includes regions below 20 μm, 10 μm, 5 μm or 2 μm thick when using the embedding paste to etch through the dielectric layer.

The variable thickness (multi-) layers described above, can be used as the (multi-) composition in any of the sintered multi-layer stacks described herein. Variable thickness (multi) layers, such as variable thickness dry and modified metal particle layers, can be used on any silicon solar cell to reduce bowing of the back marker layer.

Embedding paste as a drop-in replacement in silicon solar cells

系统校准。In one embodiment, an embedded slurry comprising 45 wt % noble metal particles, 30 wt % embedded particles and 25 wt % organic vehicle (Slurry C in Table I above) can be used as a drop in replacement ) to form the back marker layer in silicon solar cells. The fabrication of pn-bonded silicon solar cells is well known in the art. Goodrich et al. provide a complete process flow to fabricate back surface field silicon solar cells, which are referred to as "standard c-Si solar cells". See Goodrich et al., "Wafer-based monocrystalline silicon photovoltaics roadmap: opportunities for further manufacturing cost reduction using known technological improvements", Solar Materials and Solar Cells (2013), pp. 110-135, hereby Incorporated by reference. In one embodiment, a method for making solar cell electrodes includes the steps of: providing a silicon wafer having a portion of the front surface covered in at least one dielectric layer, coating the backside of the silicon wafer with a layer of aluminum particles, drying the layer of aluminum particles, Apply a layer of embedding paste (post-marker) on a part of the layer of aluminum particles, dry the layer of embedding paste, apply a plurality of fine grid lines and at least one front bus layer on the dielectric layer on the front surface of the silicon wafer, and dry And joint firing of silicon wafers. Methods such as screen printing, gravure printing, jet deposition, slot coating, 3D printing and/or inkjet printing can be used to apply the multiple layers. As an example, an Ekra or Baccini screen printer can be used to deposit the aluminum particle layer, the embedding paste layer and the front side gridlines and bussing layers. In another embodiment, the solar cell has at least one dielectric layer covering at least a portion of the rear surface of the silicon wafer. For the PERC (passivated emitter rear cell) architecture, two dielectric layers (ie, aluminum oxide and silicon nitride) are applied to the rear side of the silicon solar cell prior to the application of the aluminum particle layer. Drying the multilayer can be done in a belt furnace at a temperature between 150°C and 300°C for 30 seconds to 15 minutes. In one arrangement, a Despatch CDF 7210 belt furnace was used to dry and cofire silicon solar cells comprising sintered multilayer stacks as described herein. In one arrangement, co-firing is accomplished using a rapid heating technique and heating to a temperature greater than 760°C in air for between 0.5 and 3 seconds, which is a common temperature profile for aluminum back surface field silicon solar cells ( temperature profile). The temperature profile of the wafer is typically using Data with thermocouples attached to the bare wafer

System calibration.

FIG. 26 is a schematic diagram showing the front (or illuminated) side of a silicon solar cell 2600 . Silicon solar cell 26--has a silicon wafer 2610 with at least one dielectric layer (not shown) with fine grid lines 2620 and front bus lines 2630 on top of it. In one embodiment, the dielectric layer on the front side of the silicon wafer includes at least one material selected from the group consisting of silicon, nitrogen, oxygen, aluminum, gallium, germanium, hafnium, composites, and combinations thereof. In another embodiment, the dielectric layer on the front side of the silicon wafer is silicon nitride and is less than 200 nm thick. Commercially available front-side silver metallization pastes known in the art may be used to form fine grid lines 2620 and front bus lines 2630 . It should be noted that the front-side silver layer (ie, fine gridlines 2620 and front bus lines 2630 made from silver metallization paste) may be etched through the dielectric layer and in direct contact with silicon wafer 2610 during co-firing. In one embodiment, the silicon wafer 2610 is single crystalline and doped n-type or p-type. In another embodiment, the silicon wafer 2610 is polycrystalline and doped n-type or p-type. In one exemplary embodiment, the base layer is a polycrystalline p-type silicon wafer with an n-type emitter.

FIG. 27 is a schematic diagram showing the back side of a silicon solar cell 2700 . The back side is coated with a layer of aluminum particles 2730 and has a back marker layer 2740 distributed over the silicon wafer 2710. In one embodiment, the dielectric layer on the backside includes at least one material selected from the group consisting of: silicon, nitrogen, aluminum, oxygen, germanium, gallium, hafnium, composites, and the like on the front surface of the silicon wafer combination. In another exemplary embodiment, the dielectric layer on the front surface of the silicon wafer is silicon nitride and is less than 200 nm thick. In one embodiment, there is no dielectric layer on the backside of the silicon wafer. Commercially available aluminum pastes known in the art can be printed on at least 85%, or at least 90%, or at least 95%, or at least 97% of the total surface area of the silicon wafer backside prior to firing, which can be Described as whole Al coverage. The aluminum particle layer (after co-firing) 2730 has an average thickness between 20 and 30 μm. In various embodiments, the layer of aluminum particles 2730 has a porosity of between 3 and 20%, between 10 and 18%, or any range contained therein. For conventional BSF (back surface field) solar cell architectures, the back marking layer is applied directly to the silicon wafer. However, in order to improve the power conversion efficiency of the solar cell, it is useful to print the rear marking layer on the layer of aluminum particles. In one embodiment, the intercalation layer is directly coated on a portion of the dried aluminum particle layer to form the back marker layer 2740. FIG. 27 shows one possible pattern for the rear logo layer 2740. The intercalation layer and the underlying layer of aluminum particles are finally co-sintered to form a sintered multilayer stack as described herein. In various embodiments, the back marker layer (or modified intercalation layer) 2740 has a thickness of between 1 μm and 20 μm, or between 2 μm and 10 μm, or between 2.5 μm and 8 μm.

The variable thickness metal (aluminum) particle layer previously described above, can be used on the backside of a silicon solar cell to reduce bending of the rear marker layer and improve adhesion and electrical contact. In one embodiment of the invention, a portion of the rear marking layer has a variable thickness. In another embodiment of the present invention, a portion of the layer of modified aluminum particles has a variable thickness. In one arrangement, the back marking layer on the surface of this variable thickness modified aluminum particle layer is soldered to a tin-based marking tape, resulting in greater than 0.7 N/mm, greater than 1.5 N/mm, greater than 2 N/mm mm, or peel strength greater than 3N/mm. Thickness variation can be optimized to provide a continuous layer of metal particles and a back surface field on the base layer for silicon solar cells. In another embodiment, in the area of the rear marker layer, a portion of the combined layers (modified aluminum particle layer and rear marker layer) in that area has a thickness that is at least 20%, 30% greater than the average combined layer thickness measured over a 1x1 mm area. % or 40% of the thickness. In another embodiment, in the area of the rear marker layer, a portion of the combined layers (modified aluminum particle layer and rear marker layer) in that area has a thickness that is at least 20%, 30% smaller than the average combined layer thickness measured over a 1x1 mm area. % or 40% of the thickness.

In one embodiment of the present invention, solar cells comprising any of the sintered multilayer stacks discussed herein may be incorporated into solar modules. There are many possible solar module designs in which such solar cells are used, as will be known to those skilled in the art. The number of solar cells in a module is not intended to be limiting. Typically, 60 or 72 solar cells are combined into a commercially available module, but it is possible to combine more or less, depending on the application (ie, consumer electronics, residential, commercial, utility, etc. ). A module typically contains a bypass diode (not shown), a junction box (not shown), and a support frame (not shown) that does not directly contact the solar cells. Bypass diodes and junction boxes can also be considered components for battery interconnection.

28 is a schematic cross-sectional view showing a portion of a solar cell module in accordance with an embodiment of the present invention. The solar cell module includes at least one silicon solar cell 2840. The front side 2840F of the silicon solar cell 2840 is connected to the first marker tape 2832 (which enters and leaves the page) with the front encapsulation layer 2820 and the front sheet 2810 thereon. The back side 2840B of the silicon solar cell 2840 is connected to the second marker tape 2834, which has the back encapsulation layer 2850 and the back sheet 2860 thereon. Flag strips 2832, 2834 adjacent solar cells are electrically contacted by soldering to the front side of one cell (ie, the front busbar on the front side) and the backside of adjacent solar cells (ie, on the back side) the rear logo tape). A large number of solar cells in a solar module can be electrically connected together using marker tapes as cell interconnects.

Typical cell interconnections include metal flag ribbons soldered to the solar cells and metal bus ribbons connecting the flag ribbons. In one embodiment of the invention, the marking tape is a metal tape with a solder coating. This solder-coated marker tape may have a thickness in the range of 20 to 1000 μm, 100 to 500 μm, 50 to 300 μm, or any range subsumed therein. The width of the solder coated marker tape may be between 0.1 and 10 mm, between 0.2 and 1.5 mm, or any range contained therein. The length of the tape is determined by the application, design and substrate dimensions. The solder coating may have a thickness between 0.5 and 100 μm, between 10 and 50 μm, or any range subsumed therein. The solder coating may contain tin, lead, silver, bismuth, copper, zinc, antimony, manganese, indium, or alloys, composites, or other combinations thereof. The metal marking tape may have a thickness of between 1 μm and 1000 μm, between 50 and 500 μm, between 75 and 200 μm, or any range subsumed therein. Metal marking tapes may comprise copper, aluminum, silver, gold, carbon, tungsten, zinc, iron, tin, or alloys, composites, or other combinations thereof. The width of the metal marker tape may be between 0.1 and 10 mm, between 0.2 and 1.5 mm, or any range contained therein. In one embodiment, the marker tape is a copper tape that is 200 μm thick and 1 mm wide and coated with a 20 μm thick tin:lead (60:40 wt%) solder coating on each side.

The front sheet 2810 in Figure 28 provides some mechanical support for the module and has good optical transmission properties over the portion of the solar spectrum that the silicon solar cells 2840 are designed to absorb. The solar module is positioned such that the front sheet 2810 faces an illumination source, such as sunlight 2890. Front sheet 2810 is typically made from low iron content soda-lime glass. Front encapsulation layer 2820 and rear encapsulation layer 2850 protect silicon solar cell 2840 from electrical, chemical and physical stimuli during operation. The encapsulation is typically in the form of a polymeric sheet. Examples of materials that can be used as encapsulation include, but are not limited to, ethylene vinyl acetate (EVA), poly-ethylene-co-methacrylic acid (ionomer), Polyvinyl butyral (PVB), thermoplastic polyurethane (thermoplastic urethane) (TPU), poly-alpha-olefin (poly-alpha-olefin), poly-dimethylsiloxan (PDMS) , other polysiloxanes (ie silicon) and combinations thereof.

聚氟乙烯(polyvinyl fluoride)(PVF)薄膜典型地用于后片。含氟聚合物(fluoropolymer)和聚对苯二甲酸乙二醇酯(fluoropolymers and polyethylene terephthalate)(PET)也可用于后片中。玻璃片也可被用作后片，其可辅助提供对太阳能模块的结构支撑。支撑框架(未示出)还可被用于改进结构支撑；支撑框架典型地由铝制得。The back sheet 2860 provides protection for the silicon solar cells 2840 from the back side and may or may not be optically transparent. The solar modules are positioned such that the back sheet 2860 faces away from the source of illumination, such as sunlight 2890 . The backsheet 2860 may be a multi-layer structure made from three layers of polymeric film. DuPont ^™

Polyvinyl fluoride (PVF) films are typically used for the backsheet. Fluoropolymers and polyethylene terephthalate (PET) can also be used in the backsheet. Glass sheets can also be used as back sheets, which can assist in providing structural support to the solar module. A support frame (not shown) can also be used to improve structural support; the support frame is typically made of aluminum.

In one embodiment of the present invention, a method for forming a solar cell module is provided. Solder tabs are applied to individual solar cells (including any of the sintered multilayer stacks described herein), either manually or by using an automated tabbing or stringing machine. Subsequently, the individual cells are electrically connected in series by soldering them directly to the marker tape. The resulting structure is called a "cell string". Typically, a plurality of cell strings are arranged on a front encapsulation layer that has been applied to the front sheet. These multiple battery strings are connected to each other using bus ribbons to create an electrical circuit. The bus band is wider than the flag band used in the battery string. When the circuit between all strings is complete, the back encapsulant is applied to the backside of the connected strings and the back sheet is placed on the back encapsulant. The assembly is then sealed using a vacuum lamination process and heated (typically below 200°C) to polymerize the encapsulant. A frame is typically joined around the front panel to provide structural support. Finally, the junction box is connected to the battery interconnect and to the solar module. The bypass diode can be inside the junction box or can be connected inside the module during the battery interconnection process.

In one embodiment of the invention there is provided a method of forming a solar module comprising: a) providing at least one solar cell having a front surface and a back surface; wherein the back surface comprises a sintered multilayer stack, b) marking the back layer and part of the front bus layer to solder a portion of the flag tape to create the cell string, c) optionally solder flag tape to the bus tape to complete the circuit, d) on the front encapsulation layer that has been applied to the front sheet Arranging the battery strings, e) applying the back encapsulation layer to the battery strings and connecting the back sheet to the back encapsulation layer to form the module assembly, f) laminating the module assembly; g) electrically connecting and physically bonding the junction box.

It is possible to disassemble the solar module using the following steps to determine whether the multi-layer stack as described above has been combined. The back sheet and back encapsulation are removed to expose the back surface of the solar cell logo. A fast curing epoxy is applied to the solar cell's logo tape and surrounding back surface. The cells were removed from the module after the epoxy had cured and a diamond saw was used to cut out segments of the marking tape/solar cells. The previously described ion milling machine was used to mill the sections, and SEM/EDX was performed to determine whether the structure was as described in the embodiments of the present invention. Figure 29 is a polished cross-sectional SEM image of the back (unilluminated) side of a solar cell. The samples were from solar cells (which included novel sintered multilayer stacks) that had been incorporated into solar modules and subsequently dismantled as described above. The image shows a metal flag strip 2932 and its solder coating 2931 soldered to the sintered multilayer stack 2902. The structural layers of the sintered multilayer stack 2902 are clearly visible. Just below the solder coating 2931 is a modified intercalation layer 2945, a modified metal particle layer 2944, and a silicon base layer 2941. The layers identified in the drawings can be more easily identified using EDX.

Other PV Cell Architectures

The embedding paste can be used to create a variety of sintered multilayer stacks, which can be used as metallization layers on the front and back sides of many different solar cell architectures. As disclosed herein, embedding pastes and sintering multilayer stacks can be used in solar cell architectures including, but not limited to, BSF silicon solar cells, passivated emitter and rear contact (PERC) solar cells cells, and bifacial and interdigitated back contact solar cells.

The PERC solar cell architecture is an improvement on the BSF solar architecture to reduce back contact surface recombination by using a dielectric barrier between the silicon base layer and the back contact. In a PERC cell, a portion of the backside (ie, unilluminated) of the silicon wafer is passivated to at least one dielectric layer to reduce current carrier recombination. The novel sintered multilayer stacks disclosed herein can be used in PERC solar cells. In one embodiment, the dielectric layer on the backside of the silicon wafer includes at least one of silicon, nitrogen, aluminum, oxygen, germanium, hafnium, gallium, composites, and combinations thereof. In another embodiment, the dielectric layer on the backside of the silicon wafer includes a 10 nm thick aluminum oxide layer on the silicon surface and a 75 nm thick silicon nitride layer on the aluminum oxide layer. Commonly used aluminum pastes designed for PERC cells (eg, single crystal EFX-39, EFX-85) cannot penetrate through the dielectric layer. In order to chemically react the layer of aluminum particles and make ohmic contact with the silicon, a small area of the dielectric layer is removed by laser ablation prior to deposition of the layer of aluminum particles.

PERL (passivated emitter with rear locally diffused) and PERT (passivated emitter, rear totally diffused) are two PERC cell architectures that further improve device performance. Both types rely on doping the back of the silicon base layer to further inhibit the recombination of the back contact, which serves a role similar to the back surface field in BSF cells. In a PERL cell, the backside of the silicon base layer is doped around openings in the dielectric that make contact with the back aluminum layer. Doping is typically accomplished by propagating the dopant through a dielectric opening using a boron mixture or aluminum from a layer of aluminum particles in post-composition contacts, similar to the BSF fabrication process. A PERT cell is similar to PERL, but all of the silicon in contact with the back dielectric layer is doped except for the silicon adjacent to the dielectric openings that contact the back contact.

In one embodiment, an intercalation paste, which contains intercalating particles that are not etched through the dielectric layer, is used as a post-marking layer on a PERC, PERL, or PERT cell. A "non-etching" embedding paste is used to provide a solderable silver surface and mechanically strengthen the underlying (modified) aluminum particle layer. The resulting sintered multilayer stack contains a silicon wafer covered with at least one dielectric layer, a layer of modified aluminum particles, and a modified intercalation layer; for PERL or PERT, respectively, silicon is doped only at the dielectric openings or also through the dielectric interface. The use of non-etching embedding pastes can further reduce dielectric layer etching and reduce surface recombination. For example, the busbar paste traditionally used for the back marker layer in PERC cells is printed directly on the dielectric layer and partially etched through the dielectric layer, which increases surface recombination during co-firing.

For cells using post-dielectric layers (ie, PERC, PERL, PERT), in accordance with one embodiment of the present invention, the embedding paste may be modified to etch through the dielectric layer and to assist propagating doping of the silicon regions at the dielectric openings. An "etching" embedding paste (eg, embedding paste D in Table I) is used to provide a solderable silver surface, mechanically strengthen the underlying (modified) aluminum particle layer, and etch through the dielectric layer, exposing the silicon surface to aluminum particles, which can result in aluminum doping to the exposed silicon. The resulting sintered multilayer stack contains a silicon wafer, a layer of modified aluminum particles, and a modified intercalation layer. The sintered multilayer stack may further include Al-doped regions near the silicon surface (similar to the back surface field in BSF cells), and a solid silicon-aluminum eutectic layer at the interface between the silicon wafer and the layer of modified aluminum particles. The use of embedding pastes to etch through (multi) dielectric layers has several advantages. First, it is an inexpensive alternative to the laser ablation procedure that has proven expensive and unreliable in the past. Second, when the wafers are co-sintered, laser ablation often removes tens to hundreds of microns of the silicon-based material and can lead to the formation of large voids between the silicon-based and aluminum particle layers. Sintering the embedding paste does not cause changes to the wafer surface prior to co-sintering, which results in better bond formation, reduced void formation, and better reproducibility than when laser ablation is used.

In accordance with one embodiment of the present invention, an embedding paste can be used to provide a solderable surface for cell construction, depending on the layer of aluminum particles to make ohmic contact with p-type silicon. Examples of these configurations include interdigitated back-contact solar cells, n-type BSF cell architectures, and bifacial solar cells. In one embodiment, intercalation paste C (from Table I) is applied to the Al layer of an interdigitated back-contact solar architecture, such as a Zebra cell. For n-type BSF architectures, which have achieved 20% power conversion efficiency for Al fully covered cells, the embedded paste can replace the traditional back-mark Ag paste that directly contacts the silicon, thus reducing the _Voc of the solar cell. In various n-type wafer based solar cell architectures, embedding pastes can be used on the front side (ie, the illumination side). Embedding pastes can also be used in combination with Al pastes to reduce the cost of bifacial solar cells. Existing bifacial solar cell architectures use Ag pastes that contain small amounts of aluminum (eg, less than 5 wt% Al) to make ohmic contact with the p-type silicon layer. Existing bifacial architectures use almost twice the amount of silver as BSF architectures, which are prohibitively expensive. It is useful to use pure aluminum paste in double-sided architecture, but Al is not solderable. Silver-containing embedding pastes (eg, paste C in Table I) can be printed on Al pastes in a double-sided design and provide mechanical stability and a sinterable surface while reducing Ag usage.

Material properties of sintered multilayer stacks and effects on silicon solar cells.

Material properties of interest in sintered multilayer stacks for solar cells and other electronic devices include solderability, peel strength, and contact resistance.

186)或R(

952)，沉积在标志带上且在70℃干燥。这些熔剂在蚀刻很多当在空气中烧结时形成在铝粒子上的金属氧化物、例如氧化铝(Al₂O₃)时不是有效的。Solderability is the ability to form a strong physical bond between two metal layers by the flow of molten metal solder between the two metal surfaces at temperatures below 400°C. Soldering on the modified intercalation layer of the sintered multilayer stack can be performed after heating to above 650°C in air. Soldering involves the use of fluxes, which are chemical agents that clean or etch one or both surfaces before reflowing the molten solder. Solder fluxes typically used in solar cells, designated RMA (eg,

186) or R(

952), deposited on marker tape and dried at 70°C. These fluxes are not effective in etching many metal oxides, such as alumina (Al ₂ O ₃ ), that form on aluminum particles when sintered in air.

Peel strength is a measure of solder bond strength and an indication of reliability for integrated circuits, light emitting diodes and solar applications. Solder coated with metal strips 0.8 to 20mm wide and 100-300um thick can be dipped into the flux and dried. It was placed on the modified intercalation layer and soldered using a solder iron at a temperature between 200°C and 400°C. The peel strength is the force required to peel off the solder tape at a given peel speed at an angle of 180° to the soldering direction, separated by the width of the solder tape. Solder joints formed during the soldering process have an average peel strength greater than 1 N/mm at 1 mm/sec (eg, a 2 mm marker tape requires greater than 2 N peel force to remove the solder tape). The solar cells are electrically connected via flags, which are soldered to the front bus bar of one cell and the rear flag layer of an adjacent cell. Typically, peel strengths are between 1.5 and 4 N/mm for contacts of marker tapes in commercially available solar cells. When using the sintered multilayer stack as the back marker layer, the dominant failure mode will be near the Al-Si interface, which can be determined using plan view SEM/EDX. In an exemplary embodiment, the peel strength is greater than 1 N/mm when the layer of the (modified intercalated) silver rich sublayer is soldered with a tin based marker tape.

Meier et al. describe how to use four-point probe electrical measurements to determine the resistance of each metallization layer on a complete solar cell. See Meier et al., "Determination of Series Resistance Composition from Measurements on Completed Cells," IEEE (2006), p. 2615, incorporated herein by reference. The bulk resistance of a metallization layer is directly related to the bulk resistance of the material from which it is made. In one embodiment of the invention, the bulk resistance of pure Ag is 1.5x10 ^-8 Ω-m; the pure Ag metallization layer used on industrial solar cells has a bulk resistance 1.5 to 5 times higher than that of pure Ag . Bulk resistance is important for fine grid lines, which must carry current over relatively long (ie, greater than 1 cm) lengths. The resistances of the front bus bars and rear marking layers are less important when the cells are marked in the module.

In most integrated circuit, LED and solar cell architectures, the current from the metal particle layer flows through the modified metal particle layer and into the modified intercalation layer. For sintered multilayer stacks, the contact resistance between these three layers plays an important role in device performance. The contact resistance between these layers in sintered multilayer stacks can be measured using transmission line measurements (TLM) (ref: Meier et al., "Copper backside busbars: Elimination in crystalline silicon solar cells and modules" Ag and full aluminum coverage", IEEE PVSC (2015), pp. 1-6). The TLM is plotted as resistance versus distance between electrodes. TLM is particularly useful for measuring contact resistance, 1) between a layer of metal particles and a layer of modified metal particles, and 2) between a layer of modified metal particles and a modified intercalation layer. The contact resistance of the sintered multilayer stack is the sum of the above-mentioned contact resistances 1) and 2). The contact resistance of the sintered multilayer stack is half the y-intercept value of a linear fit of the resistance versus distance measurements. Resistance between bus bars was measured using a Keithley 2410 Sourcemeter set up with a four point probe, sourcing current between -0.5A to +0.5A and measuring voltage. In various embodiments, the contact resistance of the sintered multilayer stack is between 0 to 5 mOhm, 0.25 to 3 mOhm, 0.3 to 1 mOhm, or any range contained therein. The sheet resistance of the metal particle layer is determined by multiplying the line slope by the electrode length. Contact resistance and sheet resistance are used to numerically determine the transmission length and consequent contact resistivity. The change in series resistance is determined by dividing the contact resistivity by the partial area coverage of the improved intercalation. In various embodiments, the change in series resistance is less than 0.200 Ω-cm ² , less than 0.100 Ω-cm ² , less than 0.050 Ω-cm ² , less than 0.010 Ω-cm ² , or less than 0.001 Ω-cm ² .

The contact resistance between the rear marker layer and the aluminum particle layer affects the series resistance and the power conversion efficiency of the solar cell. This contact resistance can be measured by transmission line measurements. A power line drawing of a conventional silver back marker layer on silicon with a 300 μm overlapping layer of aluminum particles is shown in FIG. 30 . A drawing of the modified intercalation on the aluminum particle layer, the transmission line used as the rear marker layer, is shown in FIG. 31 . The y-intercept value in Figure 31 is 1.11 mOhm, compared to the y-intercept value in Figure 30 of 0.88. The contact resistance between the rear marker (intercalation) layer and the aluminum particle layer was 0.56 mOhm. The contact resistance for the traditional post-mark architecture is 0.44mOhm. In various embodiments, the contact resistance between the rear marker (intercalation) layer and the aluminum particle layer is between 0 and 5 mOhm, between 0.25 and 3 mOhm, or between 0.3 and 1 mOhm or in any range subsumed therein . The sheet resistance of the aluminum layer is determined by multiplying the line slope by the electrode length and is approximately 9 mOhm/square in Figures 30 and 31 .

Although TLM is the preferred method to accurately extract the contact resistance of the sintered multilayer stack (ie, the back marker layer and the aluminum particle layer), it is possible to use the four-point probe method to determine the contact resistance on a complete solar cell. This method is used by first measuring the resistance between the two rear marker layers ( _RAg-to-Ag ) and then moving the probe over the Al particle layer (within 1 mm of the rear marker layer) to obtain _{RAl- to-Al} . Contact resistance is obtained by subtracting R _Ag-to-Ag from R _Al-to-Al and dividing by 2. This is not as precise as a TLM measurement, but when averaging measurements from multiple solar cells, it can be approximated to within 0.50mOhm.

Resistance and sheet resistance are used to numerically determine the transmission length and consequent contact resistivity. In Figure 31, the transmission length of the co-sintered multilayer stack is 5 mm and the contact resistance is 2.2 mΩ. The change in series resistance was estimated by dividing this number by the partial area coverage of the modified intercalation. In FIG. 31 , the estimated change in series resistance is 0.023 Ω-cm ² , which is equal to the change in series resistance of 0.020 Ω-cm ² measured in FIG. 30 calculated for the conventional rear marker layer. Changes in series resistance can be directly measured by fabricating controlled BSF (back surface field) silicon solar cells with full Al coverage and no rear marker layer and fabricating BSF silicon solar cells with full Al coverage and Ag:Bi intercalation. The series resistance of the cell can be obtained from the current-voltage curves at various light intensities, and the difference in series resistance can be attributed to the increased contact resistance between the rear marking layer and the layer of sintered aluminum particles. In various embodiments, the change in series resistance in the solar cell is less than 0.200 Ω-cm ² , less than 0.100 Ω-cm ² , less than 0.050 Ω-cm ² , less than 0.010 Ω-cm ² , or less than 0.001 Ω-cm ² .

One benefit of using intercalation on silicon solar cells is the improvement in open-circuit voltage (V _oc ) brought about by the continuous back surface field formed on the silicon wafer. The _Voc gain can be directly measured by comparing conventional BSF solar cells to BSF solar cells containing Ag:Bi intercalation pastes, as described herein, when both devices have the same back sink surface area. Traditional BSF silicon solar cells are fabricated using a silver-based post-marking paste printed directly on a silicon wafer and surrounded by a layer of aluminum particles. Intercalation layers (eg, fabricated using intercalation paste C) can be used on silicon solar cells with full Al surface coverage. The V _oc of the two solar cells was measured by a current-voltage test at one sunlight intensity. For solar cells with a rear marker surface area greater than 5 ^cm2 , the _Voc can be increased by at least 0.5 mV, at least 1 mV, at least 2 mV, or at least 4 mV compared to a rear marker layer on a conventional silicon architecture when intercalation is used. Finally, the short-circuit current density (J _sc ) and fill factor are also improved when the intercalation architecture is used instead of the traditional post-marker design. Silver does not make ohmic contact with p-type silicon. The silicon marker layer directly on p-type silicon reduces current harvesting, which can be estimated by performing electroluminescence or photoluminescence measurements on complete or incomplete solar cells. The increase in _Jsc can also be measured by testing cells with an embedded configuration versus cells with a back marker layer directly on silicon. Another benefit is an increase in fill factor, which can change positively depending on an increase in _Voc , a decrease in contact resistance, and/or a change in recombination dynamics on the backside of the solar cell.

It is to be understood that the invention described herein may be carried out by various apparatus, materials and arrangements and that various modifications of both apparatus and procedures may be effected without departing from the scope of the invention itself.

Claims (23)

1. A slurry, comprising:

between 10 wt% and 70 wt% noble metal particles;

at least 10 wt% of embedded particles; and

an organic vehicle;

wherein the embedded particles include one or more selected from the group consisting of low temperature base metal particles, crystalline metal oxide particles, and glass frit particles.

2. The slurry of claim 1 wherein the weight ratio of embedded particles to noble metal particles is at least 1: 5.

3. the slurry of claim 1, wherein the noble metal particles comprise at least one material selected from the group consisting of: gold, silver, platinum, palladium, rhodium, and alloys, composites, and other combinations thereof.

4. The slurry of claim 1 wherein the noble metal particles have a D50 of between 100nm and 50 μ ι η.

5. The slurry of claim 1, wherein the noble metal particles have a size of 0.4 to 7.0m²Specific surface area between/g.

6. The slurry of claim 1, wherein a portion of the noble metal particles have a shape selected from the group consisting of spherical, platelet, and elongated.

7. The slurry of claim 1, wherein at least one of the noble metal particles and the embedded particles has a monomodal size distribution.

8. The slurry of claim 1, wherein at least one of the noble metal particles and the insert particles has a multimodal size distribution.

9. The slurry of claim 1 wherein at least some of the noble metal particles are silver and have a D50 between 300nm and 2.5 μ ι η and a D of 1.0 to 3.0m²Specific surface area between/g.

10. The slurry of claim 1, wherein the embedded particles have a D50 between 100nm and 50 μ ι η.

11. The slurry of claim 1, wherein the embedded particles may have a size of 0.1 to 6.0m²Specific surface area between/g.

12. The slurry of claim 1, wherein a portion of the embedded particles have a shape selected from the group consisting of spherical, platelet, and or elongated.

13. The slurry of claim 1, wherein the low temperature base metal particles comprise a material selected from the group consisting of: bismuth, tin, tellurium, antimony, lead, and alloys, composites, and other combinations thereof.

14. The slurry of claim 13, wherein the low temperature base metal particles comprise bismuth and have a D50 between 1.5 and 4.0 μ ι η and 1.0 to 2.0m²Specific surface area between/g.

15. The slurry of claim 1, wherein at least some of the low temperature base metal particles comprise bismuth core particles surrounded by a single shell comprising a material selected from the group consisting of: silver, nickel boron, tin, tellurium, antimony, lead, molybdenum, titanium, and alloys, composites, and other combinations thereof.

16. The slurry of claim 1, wherein at least some of the low temperature base metal particles comprise bismuth core particles surrounded by a single shell comprising a material selected from the group consisting of: silicon oxide, magnesium oxide, boron oxide, and any combination thereof.

17. The slurry of claim 1, wherein the crystalline metal oxide particles comprise oxygen and a metal selected from the group consisting of: bismuth, tin, tellurium, antimony, lead, vanadium, chromium, molybdenum, boron, manganese, cobalt, and alloys, composites, and other combinations thereof.

18. The paste of claim 1, wherein the glass frit comprises a material selected from the group consisting of: antimony, arsenic, barium, bismuth, boron, cadmium, calcium, cerium, cesium, chromium, cobalt, fluorine, gallium, germanium, indium, hafnium, iodine, iron, lanthanum, lead, lithium, magnesium, manganese, molybdenum, niobium, potassium, rhenium, selenium, silicon, sodium, strontium, tellurium, tin, vanadium, zinc, zirconium, alloys thereof, oxides thereof, composites thereof, and other combinations thereof.

19. The slurry of claim 1, wherein the slurry has a solids loading of between 30 wt% and 80 wt%.

20. The slurry of claim 1 wherein the embedded particles comprise at least 15 wt% of the slurry.

21. The paste of claim 1, wherein the paste comprises 45 wt% Ag particles, 30 wt% bismuth particles, and 25 wt% organic vehicle.

22. The paste of claim 1, wherein the paste comprises 30 wt% Ag particles, 20 wt% bismuth particles, and 50 wt% organic vehicle.

23. The slurry of claim 1, wherein the slurry is at 25 ℃ for 4sec^-1At a shear rate ofA viscosity between 10,000 and 200,000 cP.

Images