Field

The present invention relates to conductive pastes which are particularly suitable for use in solar cells.

Background

Screen printed conductive (e.g. silver) pastes are routinely used as conductive tracks for solar cells, such as silicon solar cells. The pastes typically comprise conductive (e.g. silver) powder, glass frit, and sometimes one or more additional additives, all dispersed in an organic medium. The glass frit has several roles. During firing, it becomes a molten phase and so acts to bond the conductive track to the semiconductor wafer. However, the glass frit is also important in etching away the anti-reflective or passivation layer (usually silicon nitride) provided on the surface of the semiconductor wafer, to permit direct contact between the conductive track and the semiconductor. The glass frit is typically also important in forming an ohmic contact with the semiconductor emitter.

The quality of the contact between the conductive track and the semiconductor wafer is instrumental in determining the efficiency of the final solar cell. The best glass frits need to be optimised to flow at the correct temperature, and to provide the correct degree of etching of the antireflective layer. If too little etching is provided, then there will be insufficient contact between the semiconductor wafer and the conductive track, resulting in a high contact resistance. Conversely, excessive etching may lead to deposition of large islands of silver in the semiconductor, disrupting its p-n junction and thereby reducing its ability to convert solar energy into electrical energy.

There remains a need for conductive pastes e.g. for solar cells, which offer a good balance of properties. In particular, there remains a need for conductive pastes for solar cells which provide an excellent (lowered) contact resistance without negatively influencing the p-n junction of a solar cell, and which include glass frit which flows at a suitable temperature for firing the conductive paste during manufacture of a solar cell. Much recent attention has focussed on improving the glass frit materials included in conductive pastes for photovoltaic cells, to provide a good balance of properties. As outlined above, prior art conductive paste compositions for use in photovoltaic cell applications typically comprise three main components: (i) a conductive (e.g. silver) powder; (ii) a glass frit; and (iii) an organic carrier medium in which the conductive powder and glass frit are disposed for printing of the conductive paste.

The present applicant has surprisingly found that the glass frit component of such conductive pastes can be replaced with a mixture of several crystalline metal compound (e.g. metal oxide) particles. This is described in the present applicant’s earlier patent applications published as WO2017/081448 and WO2017/081449. These conductive pastes comprise: (i) a conductive (e.g. silver) powder; (ii) a mixture of crystalline metal compounds in powder form (rather than glass frit); and (iii) an organic carrier medium in which the conductive powder and crystalline metal compound powder are disposed.

Summary of Invention

The present invention builds on the applicant’s previous work on developing glass-free conductive compositions for photovoltaic cells and particularly focusses on providing conductive pastes which are: (i) environmentally friendly; (ii) low cost; (iii) highly efficient in photovoltaic cell applications; and (iv) more readily optimized to new photovoltaic cell designs when compared to glass frit compositions, enabling more rapid innovation of next generation photovoltaic cells and cutting development time to market.

(i) Environmental Considerations

A particular advantage of using crystalline particles of different metal compounds in place of glass frit is that it removes the glass forming step from the process of manufacturing a conductive paste. The glass forming step typically has high energy demands, since it requires the glass precursors to be heated to temperatures above the melting point of the crystalline materials used to manufacture the glass. Glasses are typically used in conductive pastes due to their relatively low softening and melting points. Typically, glasses used in conductive pastes flow at temperatures in the range of about 400-700 °C. The present inventors have surprisingly found that despite the considerably higher melting point of at least some of the substantially crystalline metal compounds used in the pastes of the present invention, these mixtures still exhibit similar flow and melt behaviour to glass frits, which enables them to be used with a similar firing profile and manufacturing method as pastes comprising glass frit.

In addition to reduced energy demand, conductive pastes described herein are intended to be environmentally friendly from a chemical perspective. In this regard, many prior art glass frits used in conductive pastes include lead (Pb) to provide a glass with suitable properties (e.g. low melting point) for use in photovoltaic cells. However, lead is highly toxic. As such, lead-free glass compositions have been developed. The present compositions seek to be crystalline and lead free to combine the environmental advantages of low energy demand and low chemical toxicity.

(ii) Cost Considerations

By avoiding glass forming steps in the paste manufacturing process energy costs can be reduced. Fabrication times are also reduced thus reducing operational time costs. In terms of capital investment costs, glass forming equipment is not required thus reducing capital investment requirements for equipment. Furthermore, since bulky glass forming equipment is not needed, less space is required for the manufacturing process, thus reducing manufacturing footprint and reducing capital investment for land and building space.

It should also be noted that the glass frit forming process inevitably introduces various impurities into the composition. It is possible to reduce such contamination by utilizing specialized glass forming equipment which is expensive. Even then, some contamination is inevitable, and this may impact on consistency and end performance. By avoiding the requirement for a glass frit in the conductive paste composition, the number of production steps where contamination can occur is significantly reduced.

(Hi) Performance Considerations

While reducing costs and environmental impact is important, the conductive pastes must, of course, perform at least as well as prior art glass frit-based pastes in terms of producing highly efficient photovoltaic cells. Most preferably the conductive pastes should improve upon the performance characteristics of such prior art glass frit-based conductive pastes. In this regard, it has previously been identified that providing tellurium in the glass frit results in a better functional performance in photovoltaic cell applications. It has been found by the present applicant that providing tellurium in the composition is also advantageous when using crystalline metal compounds in place of the glass frit. However, it is important to note that when the crystalline conductive paste is deposited and fired to form a conductive component, the crystalline components do not react to form a tellurium containing glass of the type used in the prior art pastes. While glass phases may form when the paste is fired, the tellurium alloys with the silver in the paste composition rather than forming a tellurium glass of the type used in the prior art conductive pastes. As such, it should be appreciated that the crystalline pastes as described herein, i.e. pastes based on a mixture of particles of different crystalline metal compounds, do not merely comprise the crystalline compounds in the same ratios used as starting materials for the glass frits in the prior art paste compositions. Furthermore, the mechanisms by which the crystalline pastes react when fired to form a conductive component are different to the mechanisms by which glass frit-based pastes react to form a conductive component. As such, crystalline conductive pastes have different compositional requirements to glass frit conductive pastes for optimized performance, even though they share some common features such as use of tellurium.

In additional to different compositional requirements, it has also been found that the crystalline metal compound particles should be processed to a smaller particle size than the glass frit particles typically used in prior art compositions. As such, for improved performance in photovoltaic cell applications, crystalline conductive pastes differ from glass frit based pastes in two different ways: (i) the crystalline metal compound particles are processed to a smaller particle size than the glass frit particles typically used in prior art compositions; and (ii) the crystalline conductive pastes have different chemical compositional requirements than glass frit based conductive pastes to achieve high performance in a photovoltaic cell.

It is a particular focus of the present specification to describe a conductive paste for solar cell applications which has a modified crystalline composition for improved solar cell performance when compared to previous crystalline compositions.

(iv) Innovation Considerations The crystalline conductive pastes as described in this specification are more readily optimized to new photovoltaic cell designs when compared to glass frit compositions, enabling more rapid innovation of next generation photovoltaic cells and cutting development time to market for solar cell manufacturers. This is because there is no requirement to develop and manufacture new glass compositions for introduction into conductive paste formulations when developing new pastes optimized for a new cell design. Rather, it is only required to mix different metal compound compositions. As such, the rate of iterating through different compositional variants to optimize the paste for a new cell design is significantly increased. Further still, it is possible to use ratios of metal compounds which may not be considered and/or would not be suitable for fabrication of glass frit. This has the potential to widen the scope of the compositions which can be used in the conductive pastes. As such, it is considered that using a crystalline conductive paste approach should aid solar cell manufacturers in terms of the scope and rate at which next generation cells can be developed.

Conductive Paste Compositions

With all of the above in mind, the present specification describes conductive pastes which have the following characteristics: (i) based on a fine crystalline metal compound powder rather than a relatively coarse glass frit powder; (ii) lead (Pb) free; and (iii) have a crystalline metal compound powder chemical composition which provides improved functionality in photovoltaic cell applications when compared to previous crystalline conductive pastes. In particular, a conductive paste is provided comprising: a solids portion comprising a silver powder and a crystalline metal compound powder; and an organic carrier medium in which the silver powder and crystalline metal compound powder are dispersed, wherein the solids portion has a glass content less than 1 wt%, wherein the crystalline metal compound powder has a lead content less than 0.5 wt% calculated as PbO; wherein the crystalline metal compound powder has a particle size with a Dgo £ 5 pm and a D ₅₀ £ 2 pm, and wherein the crystalline metal compound powder has a composition comprising at least: 40 to 60 wt% Te calculated as TeC>2;

12 to 25 wt% Bi calculated as B12O3; and

10 to 25 wt% Zn calculated as ZnO, expressed as percentage weights (wt%) relative to a total weight of the crystalline metal compound powder in the conductive paste excluding any silver compounds, and/or wherein the crystalline metal compound powder has a composition comprising at least:

20 to 40 mol% Te calculated as TeC>2;

2 to 6 mol% Bi calculated as B1 ₂ O ₃ ; and

14 to 29 mol% Zn calculated as ZnO, expressed as molar percentage (mol%) of the crystalline metal compound powder in the conductive paste excluding any silver compounds.

The conductive paste may comprise a solids portion forming 85 to 95 wt% of the conductive paste with the organic carrier forming 5 to 15 wt% of the conductive paste. Furthermore, the solids portion may include 1 to 5 wt% of crystalline metal compound powder and 95 to 99 wt% of silver powder.

Preferably the Te, Bi, and Zn are provided as TeC>2, B1 ₂ O ₃ , and ZnO in which case the wt% and mol% values correspond to the amount of Te0 ₂ , B1 ₂ O ₃ , and ZnO in the crystalline metal compound powder. However, it should be noted that the Te, Bi, and Zn could be provided as different compounds to Te0 ₂ , B1 ₂ O ₃ , and ZnO respectively. In which case, the wt% and mol% values of those different compounds should be converted to the equivalent amount of Te0 ₂ , B1 ₂ O ₃ , and ZnO according to the amount of metal that the compounds provide. It is conventional to define frit compositions in terms of their metal oxide content even when starting materials may include other types of compound such as carbonates, phosphates, etc. As such, specifying a metal content in terms its equivalent oxide content is a well-known method of defining compositional information for inorganic frits. In certain examples, the crystalline metal compounds each comprise only one type of metal (e.g. a simple metal (M) oxide of general formula M _x O _y ) although it is also envisaged that one or more of the crystalline compounds could comprise more than one type of metal (M1, M2) such as a complex metal oxide of general formula M1xM2yOz. Furthermore, it will be noted that the calculated values for the crystalline metal compound composition are defined excluding any silver compounds. This is because it is possible to provide some of the silver content of the paste as a silver compound rather than as metallic silver. This will effectively dilute and shift the values of the other components of the crystalline metal compound powder while effectively providing a corresponding paste composition. As such, when the crystalline metal compound powder includes one or more silver compounds, these compounds should be excluded from the calculation of wt% or mol% values of the other components. Such silver compounds may decompose on firing to yield silver metal and thus contribute to the metallic silver component of the fired product rather than the inorganic oxide component.

Another way of taking into account the possible presence of silver compounds is to use concentration (weight, atomic, or molar) of silver as a reference and use the ratio of other elements relative to silver content to define the composition of the paste. As an illustrative example, if a paste comprises 90 wt% silver and 2 wt% crystalline metal oxides, and the crystalline metal oxides comprise 50 wt% TeC>2, the weight ratio of TeCVAg = 1 :90. Similarly, the molar/atomic ratio = 1/160 : 90/108 = 0.00625 : 0.83 = 0.625 : 83. The calculations can include all sources of a particular element. For example, metallic Te, TeC>2, tellurium in AgTe2 and so on. That is, the ratio can be expressed as SUM(all sources of a particular element) : SUM(all sources of Ag). This approach enables equivalent paste compositions to be compared and identified and it is intended that this specification covers such equivalent compositions. In this regard, the conductive pastes as described herein include a crystalline metal compound powder having a composition comprising at least the following components expressed as a molar ratio defined by SUM(all sources of a particular element) / SUM(all sources of Ag): 0.0060 to 0.0090 Te/Ag; 0.0012 to 0.0026 Bi/Ag; and 0.0029 to 0.0074 Zn/Ag. It will be appreciated that the metal compound compositions can be defined in terms of this molar ratio and/or as wt% and/or mol% values, and all these values have been included in the present specification for clarity and completeness.

The conductive paste composition defined herein represents an improvement over those compositions disclosed in the applicant’s earlier patent applications published as WO2017/081448 and WO2017/081449. In this regard, it is noted that these earlier applications contain disclosures of broad ranges for individual components of pastes which may overlap with the ranges given for individual components of pastes according to the present specification. However, there is no disclosure of the specific combination of features of the present pastes or that the area of parameter space defined by the present specification leads to an improvement in performance. One major difference is that the compositions as described herein have significantly more zinc than all the exemplified compositions disclosed in WO2017/081448 and WO2017/081449. That is, the prior art compositions which have been exemplified include a zinc oxide content significantly below 10 wt% whereas the compositions of the present specification have more than 10 wt%, preferably more than 13 wt% zinc oxide, and optionally more than 15 wt%. It has been found that compositions having tellurium, bismuth, and zinc content combined in the presently defined ranges result in better functional performance of a solar cell produced using these compositions.

The conductive paste as described herein is lead-free. By lead-free we mean that the lead content of the crystalline metal compound powder, calculated as PbO, is less than 0.5 wt%, less than 0.1 wt%, less than 0.05 wt%, less than 0.01 wt%, or less than 0.005 wt%.

The conductive paste as described herein is also glass-free. By glass-free we mean that the glass content of the solids portion is less than 1 wt%, less than 0.5 wt%, less than 0.25 wt%, less than 0.1 wt%, less than 0.05 wt%, less than 0.01 wt%, or less than 0.005 wt% with respect to total weight of the solids portion.

The Te (tellurium) content of the crystalline metal compound powder calculated as wt% TeC>2 (excluding silver compounds as previously discussed) is: at least 40, 42, 44, 46, or 48 wt%; no more than 60, 58, 56, 54, 52, or 50 wt%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively, the Te content of the crystalline metal compound powder calculated as mol% TeC>2 (excluding silver compounds) is: at least 20, 22, 24, 26, 28, or 30 mol%; no more than 40, 38, 36, 34, 33, or 32 mol%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively still, the Te content of the crystalline meta compound powder calculated as a molar ratio defined by SUM(all sources of Te) / SUM(all sources of Ag) is: at least 0.0060, 0.0064, 0.0068, 0.0072, or 0.0074; no more than 0.0090, 0.0086, 0.0082, 0.0078, or 0.0076; or within a range defined by any combination of the aforementioned lower and upper limits The Bi (bismuth) content of the crystalline metal compound powder calculated as wt% B12O3 (excluding silver compounds) is: at least 12, 13, 14, or 15 wt%; no more than 25, 23, 21 , 19, or 17 wt%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively, the Bi content of the crystalline metal compound powder calculated as mol% B12O3 (excluding silver compounds) is: at least 2.0, 2.5, 3.0, or 3.4 mol%; no more than 6.0, 5.0, 4.0, or 3.8 mol%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively still, the Bi content of the crystalline meta compound powder calculated as a molar ratio defined by SUM(all sources of Bi) / SUM(all sources of Ag) is: at least 0.0012, 0.0014, 0.0016, or 0.0018; no more than 0.0026, 0.0024, 0.0022, or 0.0020; or within a range defined by any combination of the aforementioned lower and upper limits.

The Zn (zinc) content of the crystalline metal compound powder calculated as wt% ZnO (excluding silver compounds) is: at least 10, 12, 13, 14, 15 or 16 wt%; no more than 25, 23, 21 , 19, or 18 wt%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively, the Zn content of the crystalline metal compound powder calculated as mol% ZnO (excluding silver compounds) is: at least 14, 16, 18, 20 or 21 mol%; no more than 29, 27, 25, 24, or 23 mol%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively still, the Zn content of the crystalline meta compound powder calculated as a molar ratio defined by SUM(all sources of Zn) / SUM(all sources of Ag) is: at least 0.0029, 0.0035, 0.0040, 0.0045, or 0.0050; no more than 0.0074, 0.0070, 0.0065, 0.0060, or 0.0055; or within a range defined by any combination of the aforementioned lower and upper limits.

The crystalline metal compound powder may further comprise an Li (lithium) content, calculated as wt% UO2 (excluding silver compounds) of: at least 5, 6, 7, or 8 wt%; no more than 15, 13, 11, 10, or 9 wt%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively, the crystalline metal compound powder may further comprise an Li content, calculated as mol% UO2 (excluding silver compounds) of: at least 25, 28, 29, or 30 mol%; no more than 45, 40, 35, 34, 33, 32, or 31 mol%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively still, the Li content of the crystalline meta compound powder calculated as a molar ratio defined by SUM(all sources of Li) / SUM(all sources of Ag) is: at least 0.0080, 0.0100, 0.0120, 0.0140, or 0.0150; no more than 0.0240, 0.0220, 0.0200, 0.0180, or 0.0170; or within a range defined by any combination of the aforementioned lower and upper limits.

The Li may be provided as two different compounds, e.g. U2CO3 and U3PO4. It has been found that while other lithium and phosphorus sources may be used, a mixture of U2CO3 and U3PO4 is advantageous from a processing perspective.

The crystalline metal compound powder may further comprise an Mg (magnesium) content, calculated as wt% MgO (excluding silver compounds) of: at least 0.5, 1.0, 1.5, or 2.0 wt%; no more than 4.0, 3.5, 3.0, or 2.8 wt%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively, the crystalline metal compound powder may further comprise an Mg content, calculated as mol% MgO (excluding silver compounds) of: at least 2, 3, 4, or 5 mol%; no more than 10, 9, 8, or 7 mol%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively still, the Mg content of the crystalline meta compound powder calculated as a molar ratio defined by SUM(all sources of Mg) / SUM(all sources of Ag) is: at least 0.0003, 0.0005, 0.0007, 0.009, or 0.0010; no more than 0.0024, 0.0020, 0.0018, 0.0016, or 0.0014; or within a range defined by any combination of the aforementioned lower and upper limits.

The crystalline metal compound powder may further comprises a P (phosphorus) content, calculated as wt% P2O5 (excluding silver compounds) of: at least 2.0, 3.0, or 4.0 wt%; no more than 7.0, 6.0, or 5.0 wt%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively, the crystalline metal compound powder may further comprise a P content, calculated as mol% P2O5 (excluding silver compounds) of: at least 1, 2, or 3 mol%; no more than 6, 5, or 4 mol%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively still, the P content of the crystalline meta compound powder calculated as a molar ratio defined by SUM(all sources of P) / SUM(all sources of Ag) is: at least 0.0007, 0.0009, 0.0011, or 0.0013; no more than 0.0024, 0.0022, 0.0020, or 0.0018; or within a range defined by any combination of the aforementioned lower and upper limits. As previously discussed, the phosphorus is advantageously provided in the form of U3PO4.

The crystalline metal compound powder may further comprise an Na (sodium) content, calculated as wt% Na ₂ 0 (excluding silver compounds) of: at least 0.1, 0.3, 0.5, or 0.8 wt%; no more than 1.5, 1.0, or 0.9 wt%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively, the crystalline metal compound powder may further comprise an Na content, calculated as mol% Na ₂ 0 (excluding silver compounds) of: at least 0.2, 0.5, 0.8, or 1.0 mol%; no more than 2.0, 1.7, 1.5 or 1.3 mol%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively still, the Na content of the crystalline meta compound powder calculated as a molar ratio defined by SUM(all sources of Na) / SUM(all sources of Ag) is: at least 0.0001, 0.0003, 0.0005, or 0.0006; no more than 0.0012, 0.0010, 0.0008, or 0.0007; or within a range defined by any combination of the aforementioned lower and upper limits.

The crystalline metal compound powder may further comprise a W (tungsten) content, calculated as wt% WO ₃ (excluding silver compounds) of: at least 1.0, 1.3, or 1.6 wt%; no more than 5.0, 3.5, or 2.0 wt%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively, the crystalline metal compound powder may further comprise a W content, calculated as mol% WO ₃ (excluding silver compounds) of: at least 0.2, 0.4, or 0.6 mol%; no more than 2.0, 1.5, or 1.0 mol%; or within a range defined by any combination of the aforementioned lower and upper limits. Alternatively still, the W content of the crystalline meta compound powder calculated as a molar ratio defined by SUM(all sources of W) / SUM(all sources of Ag) is: at least 0.0001, 0.0002, or 0.0003; no more than 0.0005 or 0.0004; or within a range defined by any combination of the aforementioned lower and upper limits.

The crystalline metal compound powder may be substantially free of boron. For example, the crystalline metal compound powder may have a boron content, calculated as B2O3, of less than 0.1 wt% boron, less than 0.05 wt%, less than 0.01 wt%, or less than 0.005 wt% _.

As previously indicated, the conductive paste may comprise a solids portion forming 85 to 95 wt% of the conductive paste with the organic carrier forming 5 to 15 wt% of the conductive paste. Furthermore, the solids portion may include 1 to 5 wt% of crystalline metal compound powder and 95 to 99 wt% of silver powder. These conductive paste compositions balance printing properties for the paste and electrical performance properties of the fired paste in a solar cell. Again, the wt% of silver powder and crystalline metal compound powder can be shifted by providing at least a portion of the silver in compound form rather than metallic elemental silver. As such, the solids portion may alternatively be defined as comprising 1 to 5 wt% crystalline metal compound powder (excluding silver compounds) and 95 to 99 wt% silver including metallic elemental silver and silver compound powder. Alternatively still, the silver and crystalline metal compound powder content of the solids portion can be expressed as a ratio as previously described.

According to another aspect of the present invention there is provided a method for the manufacture of a surface electrode of a solar cell, the method comprising applying a conductive paste as defined above to a semiconductor substrate and firing the applied conductive paste. The method can use a firing profile in which the temperature of the surface of the applied conductive paste exceeds 500°C for a period of two minutes or less, one minute or less, 30 seconds or less, 20 seconds or less, or 10 seconds or less. Optionally, the firing profile exceeds 400°C for a period of two minutes or less, one minute or less, 30 seconds or less, or 20 seconds or less. Optionally, the firing profile exceeds 300°C for a period of two minutes or less, one minute or less, 30 seconds or less, or 20 seconds or less. Optionally, the firing profile exceeds 200°C for a period of two minutes or less, one minute or less, 30 seconds or less, or 25 seconds or less. Optionally, the firing profile exceeds 100°C for a period of two minutes or less, one minute or less, or 30 seconds or less.

Brief Description of the Drawings

For a better understanding of the present invention and to show how the same may be carried into effect, certain embodiments of the present invention will now be described by way of example only with reference to the accompanying drawings, in which:

Figure 1 shows an example of a firing curve for paste compositions as described herein.

Detailed Description

Conductive pastes of the present invention include an organic medium and a solids portion. The solids portion includes electrically conductive material and an inorganic particle mixture. Each of these will be discussed, as will various methods of using them to make a conductive paste. Inorganic Particle Mixture - Composition

The solids portion of the conductive pastes described herein contains a blend of substantially crystalline inorganic materials in particulate form. This inorganic blend is sometimes referred to as an oxide particle mixture herein. The oxides, carbonates, and other materials as described below can be mixed (for example, by co-milling) and then incorporated into a conductive paste.

Generally, in some aspects of the invention the inorganic particle mixture is made up of two or more different particulate inorganic materials such as metal compounds, e.g. metal oxides, metal carbonates and the like. The particles are substantially crystalline. The mixture may contain non-oxide materials and may be formed from materials which are not oxides.

The particulate nature means that discrete, separate or individual particles of each inorganic component are present. These are different from the fused, amorphous structures of the glass frits previously used in conductive pastes for photovoltaic cell applications. Since the inorganic particles are substantially crystalline, they do not exhibit a glass transition.

In the solids portion, electrically conductive material and an inorganic particle mixture are present. It may be that these are the only components of the solids portion. The solids portion may therefore consist of only an electrically conductive material and an inorganic particle mixture.

Therefore, in the solids portion the content of amorphous oxide material, or glass, is very low. The glass content of the solids portion may be less than 1 wt%. For example, the glass content of the solids portion may be less than 0.5 wt%, less than 0.25 wt%, less than 0.1 wt%, less than 0.05 wt%, less than 0.01 wt%, or less than 0.005 wt% with respect to total weight of the solids portion. It may be that the solids portion is substantially glass-free. In some embodiments, the solids portion does not include any intentionally added glass and/or any intentionally formed glass phase. It will be understood by the skilled reader that a glass material is not synonymous with an amorphous material, or even an amorphous region within a crystalline material. A glass material exhibits a glass transition. While glasses may include some crystalline domains (they may not be entirely amorphous) these are different from the discrete crystalline particles described herein.

Of course, it will be recognised by the skilled person that some amorphous or glassy phase may be formed even when substantially crystalline raw materials are used due to the nature of the processing conditions used. In aspects of the present invention this is minimised. For example, there may be some surface reaction of the oxide particles induced by milling, or deposition of carbon from organic salts (e.g. formates and/or acetates), metal organics, and/or the organic solvent. However, the lack of glass transition (that is, a non-exhibition of glass transition) may characterise the difference from glass materials.

The inorganic particle mixture includes substantially crystalline particles of different metal compounds. Each metal compound may, for example, be selected from a metal oxide, a metal carbonate, ora metal phosphate. As the skilled person will understand, the metal compounds may include incidental impurities. Such incidental impurities will be present in a metal compound at a very low level (e.g. <1 mol %, or <0.5 mol% with respect to the metal compound in question). Furthermore, processing of the metal compounds (e.g. co-milling) may induce some surface modification or reaction of the compounds. However, in this case the bulk of each particle remains a crystalline metal compound and can still be identified as such by known techniques. The crystalline particles thus differ from glass particles which are formed of a glass material which exhibits a glass transition temperature and/or which is substantially amorphous. Glasses exhibit broad and ill-defined features in X-Ray Diffraction (XRD) analysis compared to crystalline materials which exhibit intense, narrow peaks associated with crystal planes. That said, it should be noted that certain glass materials contain a proportion or component of material which is in a crystalline phase within a bulk amorphous phase. As such, in XRD a glass exhibits broad and ill-defined features but may also exhibit small, narrow peaks corresponding to small amounts of crystalline material within the glass. Glasses can also be characterized by a complex, irregular network comprising more than one different type of metal, although a distinction is drawn between such irregular glass networks and complex crystalline metal oxides which comprise a regular lattice comprising oxygen and more than one metal. Again, the distinction between these materials is clearly evident in XRD analysis. Further still, other techniques can be used to confirm that a material is a glass including: extended X-ray absorption fine structure (EXAFS); X-ray pair distribution function (X-ray PDF); Neutron pair distribution function (Neutron PDF); solid- state nuclear magnetic resonance (NMR); thermal gravimetric analysis (TGA); and differential thermal analysis (DTA). These techniques are known in the art for analysing glass and crystalline materials and distinguishing between the two.

The detailed composition of the crystalline metal compound powder according to the present specification has already been set out in the summary section.

Inorganic particle mixture - Particle Size

Inorganic particle mixtures with certain particle size distributions are surprisingly useful. Control of the particle size distribution is therefore important in the present paste compositions.

The inorganic particle mixture may have a particle size distribution in which either

(a) Dio £ 0.41 pm;

(b) D ₅ o £ 1.6 pm;

(c) Dgo £ 4.1 pm;

(d) (Dso - Dio) £ 1.15 pm;

(e) (Dgo - Dso) £ 2.5 pm;

(f) (Dgo - Dio) £ 3.7 pm; and/or

(g) (D50/D10) £ 3.85.

One or more, two or more, three or more, four or more, five or more or six or more of these requirements may be met in examples of pastes according to the present specification.

In some examples, requirement (a) is met. In some examples requirement (b) is met. In some examples, requirement (c) is met. In some examples, requirement (d) is met. In some examples, requirement (e) is met. In some examples, requirement (f) is met. In some examples, requirement (g) is met. Any combination of these requirements may be met in examples of the invention. Regarding requirement (a), D10 is 0.41 pm or lower, for example 0.4 pm or lower, 0.39 pm or lower, 0.35 pm or lower, 0.32 pm or lower, 0.3 pm or lower, 0.28 pm or lower, 0.25 pm or lower or 0.24 pm or lower.

The value of Dm is preferably 0.4 pm or lower.

Typically, the Dm particle size may be at least 0.1 pm, at least 0.12 pm, at least 0.14 pm, at least 0.17 pm or at least 0.2 pm.

Accordingly, in some embodiments Dm is within the range 0.2 pm £ Dm £ 0.4 pm.

Regarding requirement (b), the D50 of the inorganic particle mixture is preferably less than or equal to 1.6 pm. The D50 may be 1.55 pm or lower, 1.5 pm or lower, 1.45 pm or lower, 1.4 pm or lower, 1.35 pm or lower, 1.3 pm or lower, 1.25 pm or lower, 1.2 pm or lower, 1.15 pm or lower, 1.1 pm or lower, 1.05 pm or lower, 1 pm or lower or 0.95 pm or lower.

The value of D50 is preferably 1.05 pm or lower.

Typically, the D50 particle size may be at least 0.1 pm, at least 0.3 pm, at least 0.5 pm, or at least 0.8 pm.

Accordingly, in some examples D50 is within the range 0.3 pm £ D50 £ 1.05 pm.

Regarding requirement (c), the D90 of the inorganic particle mixture is preferably less than or equal to 4.1 pm. The D90 may be 4 pm or lower, 3.8 pm or lower, 3.6 pm or lower, 3.4 pm or lower, 3.2 pm or lower, 3 pm or lower, 2.8 pm or lower, 2.6 pm or lower, 2.4 pm or lower, 2.2 pm or lower, 2.1 pm or lower, 2 pm or lower or 1.9 pm or lower.

The value of D90 is preferably 2.2 pm or lower.

Typically, the D90 particle size may be at least 1 pm, at least 1.2 pm, at least 1.4 pm, or at least 1.5 pm. Accordingly, in some examples Dgo is within the range 1.4 pm £ Dgo £ 2.2 pm.

Regarding requirement (d), (Dso - Dm) is 1.15 pm or lower, for example 1.1 pm or lower, 1 pm or lower, 0.8 pm or lower, 0.6 pm or lower, 0.59 pm or lower, 0.58 pm or lower, 0.57 pm or lower, 0.56 pm or lower, 0.55 pm or lower, 0.54 pm or lower or 0.53 pm or lower.

The value of (D5 ₀ - Dio) is preferably 0.6 pm or lower.

Typically, the difference between Dso and Dio may be at least 0.1 pm, at least 0.2 pm, at least 0.3 pm, or at least 0.35 pm.

Accordingly, in some examples (Dso - Dio) is within the range 0.3 pm £ (Dso - Dio) £ 0.6 pm.

Regarding requirement (e), (Dgo - Dso) is 2.5 pm or lower, for example 2 pm or lower, 1.75 pm or lower, 1.5 pm or lower, 1.25 pm or lower, 1.15 pm or lower, 1.1 pm or lower, 1.05 pm or lower, 1 pm or lower or 0.95 pm or lower.

The value of (Dgo - Dso) is preferably 1.15 pm or lower.

Typically, the difference between Dgo and Dso may be at least 0.5 pm, at least 0.6 pm, at least 0.7 pm, or at least 0.75 pm.

Accordingly, in some examples (Dgo - Dso) is within the range 0.6 pm £ (Dgo - Dso) £ 1.15 pm.

Regarding requirement (f), (Dgo - Dio), that is, the difference between Dgo and Dio, is preferably less than or equal to 3.7 pm. The value of (Dgo - Dio) may be 3.5 pm or lower, 3 pm or lower, 2.5 pm or lower, 2 pm or lower, 1.8 pm or lower, 1.6 pm or lower, 1.5 pm or lower, 1.45 pm or lower, 1.4 pm or lower, or 1.35 pm or lower.

The value of (Dgo - Dio) is preferably 1.8 pm or lower. Typically, the difference between Dgo and D1 ₀ may be at least 1 p , at least 1.1 pm, at least 1.2 pm, or at least 1.3 pm.

Accordingly, in some examples (Dgo - Dio) is within the range 1.1 pm £ (Dgo - Dio) £ 1.8 pm.

Regarding requirement (g), (D50/D10), that is, the value obtaining by dividing D50 by D10, is less than or equal to 3.85. The value of (D50/D10) may be 3.8 or lower, 3.7 or lower, 3.6 or lower, 3.5 or lower, 3.4 or lower, 3.3 or lower, 3.2 or lower, 3.1 or lower, 3 or lower, 2.8 or lower, or 2.6 or lower.

The value of (D ₅₀ /D ₁₀ ) is preferably 3.6 or lower.

Typically, the ratio between D50 and D10 may be at least 1 , at least 1.5, at least 2, or at least 2.3 pm.

Accordingly, in some examples (D50/D10) is within the range 2.2 £ (D50/D10) £ 3.6.

The particle sizes and distributions described herein may be determined using a laser diffraction method (e.g. using a Malvern Mastersizer 2000).

Inorganic particle mixture - Preparation

The inorganic particle mixture may be prepared by mixing raw materials for the desired metal compounds. Those raw materials may be the oxides, carbonates and so on discussed above. The mixing can be performed in a known manner. Typically, no melting, quenching or other glass production technique is carried out on the inorganic particle mixture.

Mixing or blending the above described materials can lead to an inorganic particle mixture suitable for use in the present invention. Those raw materials may be used in substantially crystalline form.

Mixing or blending techniques are well known in this technical field. The present inventors have found that the co-milling technique is particularly effective in preparing a suitable inorganic (e.g. oxide) particle mixture. Without wishing to be bound by theory, this is believed to be due to its effect on reducing particle size and/or providing a narrow particle size distribution. Alternatively, each component of the inorganic particle mixture may be milled separately (or otherwise processed to provide the desired particle size and/or particle size distribution, if necessary) before being combined to provide the inorganic particle mixture.

Mixing (e.g. co-milling) the raw materials for the inorganic particle mixture may be followed by mixing the resultant blend with an organic medium and an electrically conductive material, for example, in any order. Co-milling may be the only processing carried out on the raw materials for the inorganic particle mixture. For example, no method for glass production may be carried out. It will be understood that alternatively, each component of the inorganic particle mixture could be added separately to the electrically conductive material and the organic medium in order to obtain the conductive paste of the present invention.

For example, the above discussed oxides, carbonates and so on may be blended. Then resultant mixture may then be milled or not. When milled, the process may be carried out, for example, in a planetary mill to provide the desired particle size distribution as discussed above. Wet milling can be carried out in an organic solvent, such as butyldiglycol. A resultant blended powder may then be dried. Sieving may be carried out, to further adjust the particle size distribution.

Conductive Paste

The conductive paste is suitable for forming a conductive track or coating on a substrate. It is particularly suitable for forming a surface electrode on a semiconductor substrate, e.g. in a solar cell. The conductive paste is also suitable for forming an electrode on a thin film solar cell. The conductive paste may be a front side conductive paste.

The conductive paste may comprise a solids portion forming 85 to 95 wt% of the conductive paste with the organic carrier forming 5 to 15 wt% of the conductive paste. Furthermore, the solids portion may include 1 to 5 wt% of crystalline metal compound powder and 95 to 99 wt% of silver powder. These conductive paste compositions balance printing properties for the paste and electrical performance properties of the fired paste in a solar cell. As described in the summary section, a proportion of the silver content may be provided in compound form rather than elemental metallic silver powder.

Silver is used as the electrically conductive material in the paste. This is particularly preferable in solar cell applications, e.g. where the paste is intended for contact with an n-type emitter of a solar cell. In some examples, particularly where the paste is intended for contact with a p-type emitter of a solar cell, the conductive material may comprise aluminium, e.g. it may be a blend of silver and aluminium.

The electrically conductive material may be provided in the form of particles, e.g. metal particles. The form of the particles is not particularly limited, but may be in the form of flakes, spherical particles, granules, crystals, powder or other irregular particles, or mixtures thereof.

Typically, the D50 particle size of the electrically conductive material is at least 0.1 pm, at least 0.5 pm, or at least 1 pm. The D50 particle size may be 15 pm or less, 10 pm or less, 5 pm or less, 4 pm or less, 3 pm or less or 2 pm or less. The particle size may be determined using a laser diffraction method (e.g. using a Malvern Mastersizer 2000).

The surface area of the electrically conductive material may be at least 0.1 m ² /g, at least 0.2 m ² /g, at least 0.3 m ² /g, at least 0.4 m ² /g or at least 0.5 m ² /g. For example, it may be 5 m ² /g or less, 3 m ² /g or less, 2 m ² /g or less, 1 m ² /g or less, 0.8 m ² /g or less or 0.7 m ² /g or less.

Organic Medium

The solids portion of the conductive paste is dispersed in organic medium. The solids portion may form 85 to 95 wt% of the conductive paste and the organic carrier may form 5 to 15 wt% of the conductive paste. The organic medium typically comprises an organic solvent with one or more additives dissolved or dispersed therein. The components of the organic medium are typically chosen to provide suitable consistency and rheology properties to permit the conductive paste to be printed onto a semiconductor substrate, and to render the paste stable during transport and storage. Examples of suitable solvents for the organic medium include one or more solvents selected from the group consisting of butyl diglycol, butyldiglycol acetate, terpineol, diakylene glycol alkyl ethers (such as diethylene glycol dibutyl ether and tripropyleneglycol monomethylether), ester alcohol (such as Texanol ®), 2-(2- methoxypropoxy)-1 -propanol and mixtures thereof.

Examples of suitable additives include those dispersants to assist dispersion of the solids portion in the paste, viscosity/rheology modifiers, thixotropy modifiers, wetting agents, thickeners, stabilisers and surfactants.

For example, the organic medium may comprise one or more components selected from the group consisting of rosin (kollophonium resin), acrylic resin (e.g. Neocryl ®), alkylammonium salt of a polycarboxylic acid polymer (e.g. Dysperbik ® 110 or 111), polyamide wax (such as Thixatrol Plus ® or Thixatrol Max ®), nitrocellulose, ethylcellulose, hydroxypropyl cellulose and lecithin.

Typically, the conductive paste is prepared by mixing together electrically conductive material, the components of the inorganic particle mixture and the components of the organic medium, in any order.

Manufacture of a Surface Electrode and Solar Cell

The skilled person is familiar with suitable methods for the manufacture of a surface electrode of a solar cell. Similarly, the skilled person is familiar with suitable methods for the manufacture of a solar cell.

The method for the manufacture of a surface electrode of a solar cell typically comprises applying a conductive paste onto the surface of a semiconductor substrate and firing the applied conductive paste. The conductive paste may be applied by any suitable method. For example, the conductive paste may be applied by printing, such as by screen printing or inkjet printing. The conductive paste may be applied on a semiconductor substrate to form a light receiving surface electrode of a solar cell. Alternatively, the conductive paste may be applied on a semiconductor substrate to form a back-side surface electrode of a solar cell. The solar cell may be an n-type or a p-type solar cell. The paste may be applied onto an n-type emitter (in a p-type solar cell), or onto a p-type emitter (in an n- type solar cell). Some solar cells are known as back junction cells. In this case, it may be preferred that the conductive paste of the present invention is applied to the back side surface of the semiconductor substrate of the solar cell. Such a back-side surface is typically covered with an insulating passivation layer (e.g. SiN layer), similar to the anti-reflective coating applied to the light receiving surface of a solar cell. Alternatively, the conductive paste may be applied to a thin film solar cell or the conductive paste may be applied to a substrate for an electronic device other than a solar cell.

The skilled person is aware of suitable techniques for firing the applied conductive paste. An example firing curve is shown in Figure 1. A typical firing process lasts approximately 30 seconds, with the surface of electrode reaching a peak temperature of about 800°C. Typically, the furnace temperature will be higher to achieve this surface temperature. Firing the applied conductive paste may use a firing profile in which the temperature of the surface of the applied conductive paste exceeds 500°C for a period of two minutes or less. Preferred features of the firing profile for the pastes described herein are shown in Figure 1 and described in the summary section.

The semiconductor substrate of the electrode may be a silicon substrate. For example, it may be a single crystal semiconductor substrate, or a multi crystal semiconductor substrate. Alternative substrates include CdTe. The semiconductor may, for example, be a p-type semiconductor or an n-type semiconductor.

The semiconductor substrate may comprise an insulating layer on a surface thereof. Typically, the conductive paste of the present specification is applied on top of the insulating layer to form the electrode. Typically, the insulating layer will be non-reflective. A suitable insulating layer is SiNx (e.g. SiN). Other suitable insulating layers include Si ₃ N ₄ , Si0 ₂ , Al ₂ 0 ₃ and Ti0 ₂ .

Methods for the manufacture of a p-type solar cell may comprise applying a back-side conductive paste (e.g. comprising aluminium) to a surface of the semiconductor substrate and firing the back-side conductive paste to form a back-side electrode. The back-side conductive paste is typically applied to the opposite face of the semiconductor substrate from the light receiving surface electrode. In the manufacture of p-type solar cells, typically, the back-side conductive paste is applied to the back-side (non-light receiving side) of the semiconductor substrate and dried on the substrate, after which the front-side conductive paste is applied to the front side (light-receiving side) of the semiconductor substrate and dried on the substrate. Alternatively, the front-side paste may be applied first, followed by application of the back-side paste. The conductive pastes are typically co-fired (i.e. the substrate having both front- and back-side pastes applied thereto is fired), to form a solar cell comprising front- and back-side conductive tracks.

The efficiency of the solar cell may be improved by providing a passivation layer on the back-side of the substrate. Suitable materials include SiNx (e.g. SiN), S1 ₃ N ₄ , S1O ₂ , AI ₂ O ₃ and T1O ₂ . Typically, regions of the passivation layer are locally removed (e.g. by laser ablation) to permit contact between the semiconductor substrate and the back-side conductive track. Alternatively, where pastes of the present specification are applied to the back-side, the paste may act to etch the passivation layer to enable electrical contact to form between the semiconductor substrate and the conductive track.

Where a conductive track is formed on a substrate other than a semiconductor substrate for a solar cell, the way in which the conductive paste is applied to the substrate is not particularly limited. For example, the conductive paste may be printed onto the substrate (e.g. inkjet printed or screen printed) or it may be coated onto the substrate (e.g. dip coated). The firing conditions are also not particularly limited but may be similar to those described above with reference to forming a surface electrode for a solar cell.

Where ranges are specified herein it is intended that each endpoint of the range is independent. Accordingly, it is expressly contemplated that each recited upper endpoint of a range is independently combinable with each recited lower endpoint, and vice versa.

Examples

Inorganic Blend

Crystalline metal compound powders were prepared with a range of compositions according to the present specification. Compositional information is provided for each of the examples in the tables below. Tables 1 and 2 give the composition of each crystalline metal compound component calculated according to its oxide equivalent by wt% (Table 1) or mol% (Table 2). In the examples, Te, Bi, Zn and Ce components were provided as their crystalline oxides, Li was provided as a mixture of carbonate and phosphate, which also provided the phosphorus component, Mg, K, and Na were provided as carbonates, and the W component was provided as H ₂ WO ₄ .

Table 1: Crystalline metal compound powder compositions calculated as wt% of corresponding oxides

Table 2: Crystalline metal compound powder compositions calculated as mol% of corresponding oxides

Inorganic blends were prepared by mixing the crystalline metal compounds using a laboratory mixer to produce a mixed material, followed by wet milling of the mixed material in dipropylene glycol mono methyl ether to produce a co-milled material. The resultant dispersions were then dried in a tray drier and sieved to produce the blended powders.

Paste Preparation

Conductive silver pastes comprising substantially crystalline inorganic particle mixtures were prepared using a commercial silver powder, the inorganic blend as described herein, with the balance being standard organic medium. The pastes were prepared by pre-mixing all the components and passing several times in a triple roll mill, producing a homogeneous paste.

Solar Cell Formation

Monocrystalline PERC wafers with sheet resistance of 130 Ohm/sq, 6 inches size, were screen printed on their back side with commercially available aluminum paste, dried in an IR Mass belt dryer and randomized in groups. Each of these groups was screen printed with a front side silver paste which was one of the conductive pastes described herein and set out in more detail above.

The screens used for the front side pastes had finger opening 50 pm. After printing the front-side the cells were dried in the IR Mass belt dryer and fired in a Despatch belt furnace. The Despatch furnace had six firing zones with upper and lower heaters. The first three zones are programmed around 500°C for burning of the binder from the paste, the fourth and fifth zone are at a higher temperature, with a maximum temperature of 840°C in the final zone (furnace temperature). The furnace belt speed for this experiment was 490 cm/min. The recorded temperature was determined by measuring the temperature at the surface of the solar cell during the firing process using a thermocouple. The temperature at the surface of the solar cell did not exceed 800 °C. This is typical of the firing temperature employed for pastes comprising a glass which typically has a softening point of about 600°C. It is surprising that such good flow behaviour and contact formation are observed for the crystalline inorganic particle mixture of the present specification.

After cooling the fired solar cells were tested in an l-V curve tracer from Halm, model cetisPV-CTL1. The results are provided by the l-V curve tracer, either by direct measurement or calculation using its internal software.

To minimize the influence of contact area, the cells prepared in each set of experiments (experiments A, B, C and D in Table 3 below) were prepared using the same screen for printing, and the same paste rheology. This ensures that the line widths of the compared pastes were substantially identical and had no influence on the measurements. Solar Cell Performance

Fill factor (FF) indicates the performance of the solar cell relative to a theoretical ideal (0 resistance) system. The fill factor correlates with the contact resistance - the lower the contact resistance the higher the fill factor will be. But if the inorganic additive of the conductive paste is too aggressive it could damage the pn junction of the semiconductor. In this case the contact resistance would be low but due to the damage of the pn junction (recombination effects and lower shunt resistance) a lower fill factor would occur. A high fill factor therefore indicates that there is a low contact resistance between silicon wafer and the conductive track, and that firing of the paste on the semiconductor has not negatively affected the pn junction of the semiconductor (i.e. the shunt resistance is high).

The quality of the pn junction can be determined by measuring the pseudo fill factor (SunsVocFF). This is the fill factor independent of losses due to resistance in the cell. Accordingly, the lower the contact resistance and the higher the SunsVoc FF, the higher the resulting fill factor will be. The skilled person is familiar with methods for determining SunsVoc FF, for example as described in Reference 1. SunsVoc FF is measured under open circuit conditions and is independent of series resistance effects.

The efficiency of the solar cell is calculated based on a comparison of solar energy in to electrical energy out. It should be noted that even very small changes in efficiency can be very valuable in commercial solar cells.

Further tests were performed for measuring open circuit voltage (Uoc) and series resistance.

Solar cell test results for several examples are summarized in Table 3:

Table 3: Solar cell test results

It may be noted that while these compositions have not yet been fully optimized for the current cell type, they still show functional performance characteristics equivalent to current glass frit based conductive pastes.

While this invention has been particularly shown and described with reference to certain embodiments, it will be understood to those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as defined by the appended claims.

References

1. A. McEvoy, T. Markvart, L. Castaner. Solar cells: Materials, Manufacture and Operation. Academic Press, second edition, 2013.

21