Field of the Invention The invention relates to an electro-conductive paste comprising Ag particles with a multimodal diameter distribution in the preparation of electrodes in solar cells, particularly in the preparation of electrodes in MWT solar cells, particularly in the preparation of the metal wrap through, or plug, electrode in such solar cells. In particular, the invention relates to a solar cell precursor, a process for preparing a solar cell, a solar cell and a module comprising solar cells.

Background of the Invention

Solar cells are devices that convert the energy of light into electricity using the photovoltaic effect. Solar power is an attractive green energy source because it is sustainable and produces only non-polluting by-products. Accordingly, a great deal of research is currently being devoted to developing solar cells with enhanced efficiency while continuously lowering material and manufacturing costs. When light hits a solar cell, a fraction of the incident light is reflected by the surface and the remainder transmitted into the solar cell. The transmitted photons are absorbed by the solar cell, which is usually made of a semiconducting material, such as silicon which is often doped appropriately. The absorbed photon energy excites electrons of the semiconducting material, generating electron-hole pairs. These electron-hole pairs are then separated by p-n junctions and collected by conductive electrodes on the solar cell surfaces. Figure 1 shows a minimal construction for a simple solar cell.

Solar cells are very commonly based on silicon, often in the form of a Si wafer. Here, a p-n junction is commonly prepared either by providing an n-type doped Si substrate and applying a p-type doped layer to one face or by providing a p-type doped Si substrate and applying an n- type doped layer to one face to give in both cases a so called p-n junction. The face with the applied layer of dopant generally acts as the front face of the cell, the opposite side of the Si with the original dopant acting as the back face. Both n-type and p-type solar cells are possible and have been exploited industrially. Cells designed to harness light incident on both faces are also possible, but their use has been less extensively harnessed.

In order to allow incident light on the front face of the solar cell to enter and be absorbed, the front electrode is commonly arranged in two sets of perpendicular lines known as "fingers" and "bus bars" respectively. The fingers form an electrical contact with the front face and bus bars link these fingers to allow charge to be drawn off effectively to the external circuit. It is common for this arrangement of fingers and bus bars to be applied in the form of an electro- conductive paste which is fired to give solid electrode bodies. A back electrode is also often applied in the form of an electro-conductive paste which is then fired to give a solid electrode body.

Another approach to solar cell preparation seeks to increase the proportion of incident light which is absorbed by the front face by means of back contacting of the front electrode. In so called MWT ("Metal Wrap Through") solar cells, electrodes on the front face of the solar cell are contacted with the back face by means of channels joining the front and back face which contain electrode material, often known as a metal wrap through electrode or a plug electrode.

A typical electro-conductive paste contains metallic particles, inorganic reaction system, and an organic vehicle. There is a need in the state of the art for solar cells with improved properties, in particular for MWT solar cells with improved properties. Summary of the Invention

The invention is generally based on the object of overcoming at least one of the problems encountered in the state of the art in relation to solar cells, in particular in relation to metal wrap through solar cells, and in particular in relation to the mechanical and electrical properties of the metal wrap through electrode.

More specifically, the invention is further based on the object of providing a metal wrap through electrode which exhibits high physical adhesion with the Si surface of the channel in an MWT solar cell, preferably whilst simultaneously exhibiting other favourable electrical and physical properties of the solar cell.

A contribution to achieving at least one of the above described objects is made by the subject matter of the category forming claims of the invention. A further contribution is made by the subject matter of the dependent claims of the invention which represent specific embodiments of the invention.

Detailed Description A contribution to achieving at least one of the above described objects is made by a solar cell precursor at least comprising as precursor parts:

0 a wafer with at least one hole with a Si surface;

ϋ) an electro-conductive paste at least comprising as paste constituents: a) metallic particles;

b) an inorganic reaction system;

c) an organic vehicle; and

d) an additive;

comprised by the hole, wherein the metallic particles have a multimodal distribution of particle diameter. Above and in what follows, unless otherwise specified, all diameter distributions and maxima and d ₅ o, d ₁₀ , d ₉₀ values thereof are as described in the test methods.

In one embodiment of the solar cell precursor according to the invention, the metallic particles have a bimodal distribution of particle diameter.

In one embodiment of the solar cell precursor according to the invention, the metallic particles have a diameter distribution with at least two maxima in the range from about 0.1 to about 15 μιη, preferably in the range from about 0.2 to about 12 μηι, more preferably in the range from about 0.3 to about 10 μπι, wherein at least one pair selected from the at least two local maxima are separated by at least about 0.3 μιιι, preferably by at least about 0.5 μπι, more preferably by at least 1 μπι. In some cases separations of maxima in diameter distributions of up to about 20 μπι have been employed. In another embodiment of the solar cell precursor according to the invention, the metallic particles have a multi-modal diameter distribution with at least two maxima, wherein at least one pair selected from the at least two maxima are separated by at least about 1 μιη, preferably at least about 2 μπι, more preferably at least about 3 μιη. In some cases separations of maxima in diameter distributions of up to about 20 μπι have been employed.

In one embodiment of the solar cell precursor according to the invention, the metallic particles have a diameter distribution with a Sarle's bimodality coefficient b greater than about 0.8, more preferably greater than about 0.85, most preferably greater than about 0.9. The value of Sarle's bimodality coefficient cannot be greater than 1.

In one embodiment of the solar cell precursor according to the invention, the metallic particles are Ag particles.

In one embodiment of the solar cell precursor according to the invention, the inorganic reac- tion system is present in the paste in a range from about 0.1 to about 5 wt. %, preferably in a range from about 0.3 to about 3 wt. %, more preferably in a range from about 0.5 to about 2 wt. %.

In one embodiment of the solar cell precursor according to the invention, at least one hole is a channel joining the front face and back face of the wafer.

In one embodiment of the solar cell precursor according to the invention, the Si surface in at least one hole comprises at least one p-type doped section and at least one n-type doped section.

In one embodiment of the solar cell precursor according to the invention, the paste is in direct contact with the Si surface of the hole.

A contribution to achieving at least one of the above described objects is made by a process for preparing a solar cell precursor according to the invention, comprising at least the step of combining at least two different portions of Ag particles with values of d ₅₀ which differ by at least about 1 μπι, preferably at least about 2 μιη, more preferably at least about 3 μπι, in order to achieve the multimodal particle distribution. In some cases portions of Ag particles with values of d ₅₀ which are separated by up to about 20 μιη have employed.

A contribution to achieving at least one of the above described objects is made by a process for the preparation of a solar cell at least comprising the steps:

i) provision of a solar cell precursor according to the invention; ii) firing of the solar cell precursor to obtain a solar cell.

In one embodiment of the process according to the invention, the provision according to step i) at least comprises the steps:

a) provision of a Si wafer with a back doped layer and a front doped layer of opposite doping types;

b) making of at least one hole in the wafer; c) introduction of an electro-conductive paste into at least one hole to give a precursor according to the invention;

A contribution to achieving at least one of the above described objects is made by a solar cell obtainable by the process according to the invention.

In one embodiment of the invention, the solar cell at least comprises as precursor parts:

i) a wafer with at least one hole with a Si surface;

ii) an electrode comprised by the hole,

wherein the electrode has a higher concentration of glass at the surface where the electrode meets the Si surface than in the bulk of the electrode.

A contribution to achieving at least one of the above described objects is made by a solar cell at least comprising as solar cell parts:

i) a wafer with at least one hole with a Si surface;

ii) an electrode comprised by the hole,

wherein the metallic particles present in the electrode have a multi-modal diameter distribution with at least two maxima, wherein at least one pair selected from the at least two maxima are separated by at least about 1 μπι, preferably at least about 2 μπι, more preferably at least about 3 μιη. In some cases separations of maxima in diameter distributions of up to about 20 μπι have employed.

A contribution to achieving at least one of the above described objects is made by a module comprising at least one solar cell according to the invention and at least a further solar cell.

A contribution to achieving at least one of the above described objects is made by a module comprising at least one solar cell according to the invention and at least a further solar cell. Wafer

Preferred wafers according to the invention are regions, among other regions of the solar cell, capable of absorbing light with high efficiency to yield electron-hole pairs and separating holes and electrons across a boundary with high efficiency, preferably across a so called p-n junction boundary. Preferred wafers according to the invention are those comprising a single body made up of a front doped layer and a back doped layer.

It is preferred for that wafer to consist of appropriately doped tetravalent elements, binary compounds, tertiary compounds or alloys. Preferred tetravalent elements in this context are Si, Ge or Sn, preferably Si. Preferred binary compounds are combinations of two or more tetravalent elements, binary compounds of a group III element with a group V element, binary compounds of a group II element with a group VI element or binary compounds of a group IV element with a group VI element. Preferred combinations of tetravalent elements are combina- tions of two or more elements selected from Si, Ge, Sn or C, preferably SiC. The preferred binary compounds of a group III element with a group V element is GaAs. It is most preferred according to the invention for the wafer to be based on Si. Si, as the most preferred material for the wafer, is referred to explicitly throughout the rest of this application. Sections of the following text in which Si is explicitly mentioned also apply for the other wafer compositions described above.

Where the front doped layer and back doped layer of the wafer meet is the p-n junction boundary. In an n-type solar cell, the back doped layer is doped with electron donating n-type dopant and the front doped layer is doped with eleptron accepting or hole donating p-type dopant. In a p-type solar cell, the back doped layer is doped with p-type dopant and the front doped layer is doped with n-type dopant. It is preferred according to the invention to prepare a wafer with a p-n junction boundary by first providing a doped Si substrate and then applying a doped layer of the opposite type to one face of that substrate. Doped Si substrates are well known to the person skilled in the art. The doped Si substrate can be prepared in any way known to the person skilled in the art and which he considers to be suitable in the context of the invention. Preferred sources of Si substrates according to the invention are mono-crystalline Si, multi-crystalline Si, amorphous Si and upgraded metallurgical Si, mono-crystalline Si or multi-crystalline Si being most preferred. Doping to form the doped Si substrate can be carried out simultaneously by adding dopant during the preparation of the ' Si substrate or can be carried out in a subsequent step. Doping subsequent to the preparation of the Si substrate can be carried out for example by gas diffusion epitaxy. Doped Si substrates are also readily commercially available. According to the invention it is one option for the ini- rial doping of the Si substrate to be carried out simultaneously to its formation by adding dopant to the Si mix. According to the invention it is one option for the application of the front doped layer and the highly doped back layer, if present, to be carried out by gas-phase epitaxy. This gas phase epitaxy is preferably carried out at a temperature in a range from about 500 °C to about 900 °C, more preferably in a range from about 600 °C to about 800 °C and most pref- erably in a range from about 650 °C to about 750 °C at a pressure in a range from about 2 kPa to about 100 kPa, preferably in a range from about 10 to about 80 kPa, most preferably in a range from about 30 to about 70 kPa.

It is known to the person skilled in the art that Si substrates can exhibit a number of shapes, surface textures and sizes. The shape can be one of a number of different shapes including cuboid, disc, wafer and irregular polyhedron amongst others. The preferred shape according to the invention is wafer shaped where that wafer is a cuboid with two dimensions which are similar, preferably equal and a third dimension which is significantly less than the other two dimensions. Significantly less in this context is preferably at least a factor of about 100 smaller.

A variety of surface types are known to the person skilled in the art. According to the invention Si substrates with rough surfaces are preferred. One way to assess the roughness of the substrate is to evaluate the surface roughness parameter for a sub-surface of the substrate which is small in comparison to the total surface area of the substrate, preferably less than about one hundredth of the total surface area, and which is essentially planar. The value of the surface roughness parameter is given by the ratio of the area of the subsurface to the area of a theoretical surface formed by projecting that subsurface onto the flat plane best fitted to the subsurface by mmimising mean square displacement. A higher value of the surface roughness parameter indicates a rougher, more irregular surface and a lower value of the surface rough- ness parameter indicates a smoother, more even surface. According to the invention, the surface roughness of the Si substrate is preferably modified so as to produce an optimum balance between a number of factors including but not limited to light absorption and adhesion of fingers to the surface. The two larger dimensions of the Si substrate can be varied to suit the application required of the resultant solar cell. It is preferred according to the invention for the thickness of the Si wafer to lie below about 0.5 mm more preferably below about 0.3 mm and most preferably below about 0.2 mm. Some wafers have a minimum size of about 0.01 mm or more. It is preferred according to the invention for the front doped layer to be thin in comparison to the back doped layer. It is preferred according to the invention for the front doped layer to have a thickness lying in a range from about 0.1 to about 10 μπι, preferably in a range from about 0.1 to about 5 μπι and most preferably in a range from about 0.1 to about 2 μπι. A highly doped layer can be applied to the back face of the Si substrate between the back doped layer and any further layers. Such a highly doped layer is of the same doping type as the back doped layer and such a layer is commonly denoted with a + (n ⁺ -type layers are applied to n-type back doped layers and p ⁺ -type layers are applied to p-type back doped layers). This highly doped back layer serves to assist metallisation and improve electro-conductive proper- ties at the substrate/electrode interface area. It is preferred according to the invention for the highly doped back layer, if present, to have a thickness in a range from about 1 to about 100 μιη, preferably in a range from about 1 to about 50 μπι and most preferably in a range from about 1 to about 15 μιη. Dopants

Preferred dopants are those which, when added to the Si wafer, form a p-n junction boundary by introducing electrons or holes into the band structure. It is preferred according to the inven- tion that the identity and concentration of these dopants is specifically selected so as to tune the band structure profile of the p-n junction and set the light absorption and conductivity profiles as required. Preferred p-type dopants according to the invention are those which add holes to the Si wafer band structure. They are well known to the person skilled in the art. All dopants known to the person skilled in the art and which he considers to be suitable in the context of the invention can be employed as p-type dopant. Preferred p-type dopants according to the invention are trivalent elements, particularly those of group 13 of the periodic table. Preferred group 13 elements of the periodic table in this context include but are not limited to B, Al, Ga, In, Tl or a combination of at least two thereof, wherein B is particularly preferred. Preferred n-type dopants according to the invention are those which add electrons to the Si wafer band structure. They are well known to the person skilled in the art. All dopants known to the person skilled in the art and which he considers to be suitable in the context of the invention can be employed as n-type dopant. Preferred n-type dopants according to the invention are elements of group 15 of the periodic table. Preferred group 15 elements of the periodic table in this context include N, P, As, Sb, Bi or a combination of at least two thereof, wherein P is particularly preferred.

As described above, the various doping levels of the p-n junction can be varied so as to tune the desired properties of the resulting solar cell.

According to the invention, it is preferred for the back doped layer to be lightly doped, preferably with a dopant concentration in a range from about 1 x 10 ¹³ to about 1 x 10 ¹⁸ cm ^"3 , preferably in a range from about 1 x 10 ¹⁴ to about 1 x 10 ¹⁷ cm ^"3 , most preferably in a range from about 5 x 10 ¹⁵ to about 5 x 10 ¹⁶ cm ^"3 . Some commercial products have a back doped layer with a dopant concentration of about 1 x 10 ¹⁶ . It is preferred according to the invention for the highly doped back layer (if one is present) to

1 7

be highly doped, preferably with a concentration in a range from about 1 x 10 to about 5 x 10 cm , more preferably in a range from about 5 x 10 to about 5 x 10 cm ^" , and most

1 R

preferably in a range from about 1 x 10 to about 1 x 10 cm ^" .

Electro-conductive Paste

Preferred electro-conductive pastes according to the invention are pastes which can be applied to a surface and which, on firing, form solid electrode bodies in electrical contact with that surface. The constituents of the paste and proportions thereof can be selected by the person skilled in the art in order that the paste have the desired properties such as sintering and printa- bility and that the resulting electrode have the desired electrical and physical properties. Metallic particles can be present in the paste, primarily in order that the resulting electrode body be electrically conductive. In order to bring about appropriate sintering through surface layers and into the Si wafer, an inorganic reaction system can be employed. An example composition of an electrically-conductive paste which is preferred in the context of the invention might comprise: i) metallic particles, preferably at least about 50 wt. %, more preferably at least about 70 wt.% and most preferably at least about 80 wt. %; inorganic reaction system, preferably in a range from about 0.1 to about 5 wt. %, preferably in a range from 0.3 to 3 wt. %, more preferably in a range from 0.5 to 2 wt. %.; iii) organic vehicle, preferably in a range from about 5 to about 40 wt. %, more preferably in a range from about 5 to about 30 wt. % and most preferably in a range from about 5 to about 15 wt. %; v) additives, preferably in a range from about 0.1 to about 22 wt. %, more preferably in a range from about 0.1 to about 15 wt. % and most preferably in a range from about 5 to about 10 wt. %, wherein the wt. % are each based on the total weight of the electro-conductive paste and add up to 100 wt. %.

In order to facilitate printability of the electro-conductive paste, it is preferred according to the invention that the viscosity of the electro-conductive paste lies in a range from about 10 to about 50 Pa s, preferably in a range from about 20 to about 40 Pa s.

Metallic particles

Preferred metallic particles in the context of the invention are those which exhibit metallic conductivity or which yield a substance which exhibits metallic conductivity on firing. Metallic particles present in the electro-conductive paste give metallic conductivity to the solid electrode which is formed when the electro-conductive paste is sintered on firing. Metallic particles which favour effective sintering and yield electrodes with high conductivity and low contact resistance are preferred. Metallic particles are well known to the person skilled in the art. All metallic particles known to the person skilled in the art and which he considers suitable in the context of the invention can be employed as the metallic particles in the electro-conductive paste. Preferred metallic particles according to the invention are metals, alloys, mixtures of at least two metals, mixtures of at least two alloys or mixtures of at least one metal with at least one alloy.

Preferred metals which can be employed as metallic particles according to the invention are Ag, Cu, Al, Zn, Pd, Ni or Pb and mixtures of at least two thereof, preferably Ag. Preferred alloys which can be employed as metallic particles according to the invention are alloys containing at least one metal selected from the list of Ag, Cu, Al, Zn, Ni, W, Pb and Pd or mix- hires or two or more of those alloys. In one embodiment according to the invention, the metallic particles comprise a metal or alloy coated with one or more further different metals or alloys, for example copper coated with silver.

In one embodiment according to the invention, the metallic particles are Ag. In another embodiment according to the invention, the metallic particles comprise a mixture of Ag with Al.

As additional constituents of the metallic particles, further to above mentioned constituents, those constituents which contribute to more favourable sintering properties, electrical contact, adhesion and electrical conductivity of the formed electrodes are preferred according to the invention. All additional constituents known to the person skilled in the art and which he considers to be suitable in the context of the invention can be employed in the metallic particles. Those additional substituents which represent complementary dopants for the face to which the electro-conductive paste is applied are preferred according to the invention. When forming an electrode interfacing with an n-type doped Si layer, additives capable of acting as n-type dopants in Si are preferred. Preferred n-type dopants in this context are group 15 elements or compounds which yield such elements on firing. Preferred group 15 elements in this context according to the invention are P and Bi. When forming an electrode interfacing with a p-type doped Si layer, additives capable of acting as p-type dopants in Si are preferred. Preferred p- type dopants are group 13 elements or compounds which yield such elements on firing. Preferred group 13 elements in this context according to the invention are B and Al.

It is well known to the person skilled in the art that metallic particles can exhibit a variety of shapes, surfaces, sizes, surface area to volume ratios, oxygen content and oxide layers. A large number of shapes are known to the person skilled in the art. Some examples are spherical, angular, elongated (rod or needle like) and flat (sheet like). Metallic particles may also be present as a combination of particles of different shapes. Metallic particles with a shape, or combination of shapes, which favours advantageous sintering, electrical contact, adhesion and electri- cal conductivity of the produced electrode are preferred according to the invention. One way to characterise such shapes without considering surface nature is through the parameters length, width and thickness. In the context of the invention the length of a particle is given by the length of the longest spatial displacement vector, both endpoints of which are contained within the particle. The width of a particle is given by the length of the longest spatial displacement vector perpendicular to the length vector defined above both endpoints of which are contained within the particle. The thickness of a particle is given by the length of the longest spatial displacement vector perpendicular to both the length vector and the width vector, both defined above, both endpoints of which are contained within the particle. In one embodiment according to the invention, metallic particles with shapes as uniform as possible are preferred i.e. shapes in which the ratios relating the length, the width and the thickness are as close as possible to 1, preferably all ratios lying in a range from about 0.7 to about 1.5, more preferably in a range from about 0.8 to about 1.3 and most preferably in a range from about 0.9 to about 1.2. Examples of preferred shapes for the metallic particles in this embodiment are therefore spheres and cubes, or combinations thereof, or combinations of one or more thereof with other shapes. In another embodiment according to the invention, metallic particles are preferred which have a shape of low uniformity, preferably with at least one of the ratios relating the dimensions of length, width and thickness being above about 1.5, more preferably above about 3 and most preferably above about 5. Preferred shapes according to this embodiment are flake shaped, rod or needle shaped, or a combination of flake shaped, rod or needle shaped with oth- er shapes.

A variety of surface types are known to the person skilled in the art. Surface types which favour effective sintering and yield advantageous electrical contact and conductivity of produced electrodes are favoured for the surface type of the metallic particles according to the invention.

Another way to characterise the shape and surface of a metallic particle is by its surface area to weight ratio, also known as specific surface area. The specific surface area can be determined using the BET method. The lowest value for the surface area to weight ratio of a particle is embodied by a sphere with a smooth surface. The less uniform and uneven a shape is, the higher its surface area to weight ratio will be. In one embodiment according to the invention, metallic particles with a high specific surface area ratio are preferred, preferably in a range from about 0.1 to about 25 m ² /g, more preferably in a range from about 0.5 to about 20 m ² /g and most preferably in a range from about 1 to about 15 m ² /g. In another embodiment according to the invention, metallic particles with a low specific surface area are preferred, preferably in a range from about 0.01 to about 10 m /g, more preferably in a range from about 0.05 to about 5 m /g and most preferably in a range about 0.10 to about 1 m /g.

It is preferred according to the invention that the diameter distribution of the metallic particles be selected so as to reduce the occurrence of areas of low Ag density in the electrode in the resultant solar cell. The person skilled in the art may select the diameter distribution of the metallic particles to optimise advantageous electrical and physical properties of the resultant solar cell. It is preferred according to the present invention for the metallic particles to exhibit a multimodal diameter distribution. In one embodiment, the metallic particles have a bimodal diameter distribution. In another embodiment, the metallic particles have a diameter distribution with at least two maxima in the range from about 0.1 to about 15 μπι, preferably in a range from about 0.2 to about 12 μπι, more preferably in a range from about 0.3 to about 10 μιη, at least one pair selected from the at least two maxima being separated by at least about 0.3 μηι, preferably by at least about 0.5 μπι, more preferably by at least about 1 μπι. In a further embodiment the difference between at least one pair selected from the at least two maxima are separated by at least about 1 μπι, preferably at least about 2 μπι and more preferably at least about 3 μπι. The diameter distribution may be achieved in any manner which the skilled person considers suitable in the context of the invention. It is preferred for the diameter distribution to be achieved by combining portions of metallic particles with different values of d ₅ o- In one embodiment of the invention, it is preferred for the metallic particles in the solar cell precursor to be prepared by combining at least two different portions of metallic particles with values of d ₅₀ which differ by at least about 1 μιη, preferably at least about 2 μπι, more preferably at least about 3 μπι, in order to achieve the multimodal particle distribution.

The metallic particles may be present with a surface coating. Any such coating known to the person skilled in the art and which he considers to be suitable in the context of the invention can be employed on the metallic particles. Preferred coatings according to the invention are those coatings which promote improved printing, sintering and etching characteristics of the electro-conductive paste. If such a coating is present, it is preferred according to the invention for that coating to correspond to no more than about 10 wt. %, preferably no more than about 8 wt. %, most preferably no more than about 5 wt. %, in each case based on the total weight of the metallic particles.

In one embodiment according to the invention, the metallic particles are present as a proportion of the electro-conductive paste more than about 50 wt. %, preferably more than about 70 wt. %, most preferably more than about 80 wt. %.

Inorganic Reaction System

An inorganic reaction system is present in the electro-conductive paste according to the invention in order to bring about etching and sintering. Effective etching is required to etch any additional layers which may have been applied to the Si wafer and thus lie between the silicon wafer and the applied electro-conductive paste as well as to etch into the Si wafer to an appropriate extent. Appropriate etching of the Si wafer means deep enough to bring about good mechanical but not to bring about good electrical contact between the electrode and the silicon wafer and thus lead to a high contact resistance. Preferred inorganic reaction systems are glass frits.

In one embodiment of the solar cell precursor according to the invention, the inorganic reaction system is glass frit.

Preferred glass frits in the context of the invention are powders of amorphous or partially crystalline solids which exhibit a glass transition. The glass transition temperature T _g is the temperature at which an amorphous substance transforms from a rigid solid to a partially mobile undercooled melt upon heating. Methods for the determination of the glass transition temperature are well known to the person skilled in the art. The etching and sintering brought about by the glass frit occurs above the glass transition temperature of the glass frit and the glass transi- tion temperature must lie below the desired peak firing temperature. Glass frits are well known to the person skilled in the art. All glass frits known to the person skilled in the art and which he considers suitable in the context of the invention can be employed as the glass frit in the electro-conductive paste.

In the context of the invention, the inorganic reaction system present in the electro-conductive paste preferably comprises elements, oxides, compounds which generate oxides on heating, other compounds, or mixtures thereof. Preferred elements in this context are Si, B, Al, Bi, Li, Na, Mg, Pb, Zn, Gd, Ce, Zr, Ti, Mn, Sn, Ru, Co, Fe, Cu, Ba and Cr or mixtures of two or more from this list. Preferred oxides which can be comprised by the invention in the context of the invention are alkali metal oxides, alkali earth metal oxides, rare earth oxides, group V and group VI oxides, other oxides, or combinations thereof. Preferred alkali metal oxides in this context are sodium oxide, lithium oxide, potassium oxide, rubidium oxides, caesium oxides or combinations thereof. Preferred alkali earth metal oxides in this context are beryllium oxide, magnesium oxide, calcium oxide, strontium oxide, barium oxide, or combinations thereof. Preferred group V oxides in this context are phosphorous oxides, such as P ₂ 0 _5j bismuth oxides, such as Bi ₂ 0 ₃ , or combinations thereof _. Preferred group VI oxides in this context are tellurium oxides, such as Te0 ₂ , or Te0 _3j selenium oxides, such as Se0 ₂ , or combinations thereof. Preferred rare earth oxides are cerium oxides, such as Ce0 ₂ and lanthanum oxides, such as La ₂ 0 ₃ . Other preferred oxides in this context are silicon oxides, such as Si0 ₂ , zinc oxides, such as ZnO, aluminium oxides, such as A1 ₂ 0 ₃ , germanium oxides, such as Ge0 ₂ , vanadium oxides, such as V ₂ 0 _5; niobium oxides, such as Nb ₂ 0 ₅ , boron oxide, tungsten oxides, such as W0 ₃ , molybdenum oxides, such as M0O3, and indium oxides, such as ln ₂ 0 ₃ , further oxides of those elements listed above as preferred elements, or combinations thereof. Preferred oxides are also mixed oxides containing at least two of the elements listed as preferred elemental constituents of the frit glass, or mixed oxides which are formed by heating at least one of the above named oxides with at least one of the above named metals. Mixtures of at least two of the above-listed oxides and mixed oxides are also preferred in the context of the invention. As mentioned above, where the inorganic reaction system is a glass frit, it must have a glass transition temperature below the desired firing temperature of the electro-conductive paste. According to the invention, preferred glass frits have a glass transition temperature in a range from about 250 °C to about 800 °C, preferably in a range from about 300 °C to about 750 °C and most preferably in a range from about 350 °C to about 700 °C.

It is well known to the person skilled in the art that inorganic reaction system particles can exhibit a variety of shapes, surface natures, sizes, surface area to volume ratios and coating layers. A large number of shapes of inorganic reaction system particles are known to the per- son skilled in the art. Some examples are spherical, angular, elongated (rod or needle like) and flat (sheet like). Inorganic reaction system particles may also be present as a combination of particles of different shapes. Inorganic reaction system particles with a shape, or combination of shapes, which favours advantageous sintering, adhesion, electrical contact and electrical conductivity of the produced electrode are preferred according to the invention.

Another way to characterise the shape and surface of an inorganic reaction system particle is by its surface area to weight ratio, also known as specific surface area. The specific surface area can be determined using the BET method. The lowest value for the surface area to weight ratio of a particle is embodied by a sphere with a smooth surface. The less uniform and uneven a shape is, the higher its surface area to weight ratio will be. In one embodiment according to the invention, inorganic reaction system particles with a high specific surface area ratio are preferred, preferably in a range from about 15 to about 1 m ² /g, more preferably in a range from about 13 to about 1.5 m /g and most preferably in a range from about 8 to about 2 m /g. In another embodiment according to the invention, inorganic reaction system particles with a low specific surface area are preferred, preferably in a range from about 10 to about 0.01 m /g, more preferably in a range from about 5 to about 0.02 m ² /g and most preferably in a range about 2 to about 0.04 m ² /g.

The average particles diameter d ₅₀ , and the associated parameters d ₁₀ and d9 ₀ are characteristics of particles well known to the person skilled in the art. It is preferred according to the inven- tion that the average particle diameter d ₅₀ of the inorganic reaction system lies in a range from about 0.1 to about 10 μπι, more preferably in a range from about 0.5 to about 7 μιη and most preferably in a range from about 0.8 to about 5 μπι. The determination of the particles diameter d ₅₀ is well known to the person skilled in the art.

In one embodiment of the invention, the inorganic reaction system is present in the paste in a range from about 0.1 to about 15 wt. %.

In some cases it is preferred for the glass frit to be present in low concentrations. In one embodiment of the solar cell precursor according to the invention, the inorganic reaction system is present in the paste in a range from about 0.1 to about 5 wt. %, preferably in a range from 0.3 to 3 wt. %, more preferably in a range from 0.5 to 2 wt. %.

Organic Vehicle

Preferred organic vehicles in the context of the invention are solutions, emulsions or dispersions based on one or more solvents, preferably an organic solvent, which ensure that the constituents of the electro-conductive paste are present in a dissolved, emulsified or dispersed form. Preferred organic vehicles are those which provide optimal stability of constituents within the electro-conductive paste and endow the electro-conductive paste with a viscosity allowing effective line printability. Preferred organic vehicles according to the invention comprise as vehicle components:

(i) a binder, preferably in a range from about 1 to about 10 wt. %, more preferably in a range from about 2 to about 8 wt. % and most preferably in a range from about 3 to about 7 wt. %;

(ii) a surfactant, preferably in a range from about 0 to about 10 wt. %, more preferably in a range from about 0 to about 8 wt. % and most preferably in a range from about 0.01 to about 6 wt. %;

(ii) one or more solvents, the proportion of which is determined by the proportions of the other constituents in the organic vehicle; (iv) additives, preferably in range from about 0 to about 15 wt. %, more preferably in a range from about 0 to about 13 wt. % and most preferably in a range from about 5 to about 11 wt. %,

wherein the wt. % are each based on the total weight of the organic vehicle and add up to 100 wt. %. According to the invention preferred organic vehicles are those which allow for the preferred high level of printability of the electro-conductive paste described above to be achieved.

Binder

Preferred binders in the context of the invention are those which contribute to the formation of an electro-conductive paste with favourable stability, printability, viscosity, sintering and etching properties. Binders are well known to the person skilled in the art. All binders which are known to the person skilled in the art and which he considers to be suitable in the context of this invention can be employed as the binder in the organic vehicle. Preferred binders according to the invention (which often fall within the category termed "resins") are polymeric binders, monomelic binders, and binders which are a combination of polymers and monomers. Polymeric binders can also be copolymers wherein at least two different monomelic units are contained in a single molecule. Preferred polymeric binders are those which carry functional groups in the polymer main chain, those which carry functional groups off of the main chain and those which carry functional groups both within the main chain and off of the main chain. Preferred polymers carrying functional groups in the main chain are for example polyesters, substituted polyesters, polycarbonates, substituted polycarbonates, polymers which carry cyclic groups in the main chain, poly-sugars, substituted poly-sugars, polyurethanes, substituted polyurethanes, polyamides, substituted polyamides, phenolic resins, substituted phenolic resins, copolymers of the monomers of one or more of the preceding polymers, optionally with other co-monomers, or a combination of at least two thereof. Preferred polymers which carry cyclic groups in the main chain are for example polyvinylbutylate (PVB) and its derivatives and poly-terpineol and its derivatives or mixtures thereof. Preferred poly-sugars are for example cellulose and alkyl derivatives thereof, preferably methyl cellulose, ethyl cellulose, propyl cellulose, butyl cellulose and their derivatives and mixtures of at least two thereof. Preferred polymers which carry functional groups off of the main polymer chain are those which carry amide groups, those which carry acid and/or ester groups, often called acrylic resins, or polymers which carry a combination of aforementioned functional groups, or a combination there- of. Preferred polymers which carry amide off of the main chain are for example polyvinyl pyr- rolidone (PVP) and its derivatives. Preferred polymers which carry acid and/or ester groups off of the main chain are for example polyacrylic acid and its derivatives, polymethacrylate (PMA) and its derivatives or polymethylmethacrylate (PMMA) and its derivatives, or a mixture thereof. Preferred monomeric binders according to the invention are ethylene glycol based monomers, terpineol resins or rosin derivatives, or a mixture thereof. Preferred monomeric binders based on ethylene glycol are those with ether groups, ester groups or those with an ether group and an ester group, preferred ether groups being methyl, ethyl, propyl, butyl, pen- tyl hexyl and higher alkyl ethers, the preferred ester group being acetate and its alkyl derivatives, preferably ethylene glycol monobutylether monoacetate or a mixture thereof. Alkyl cel- lulose, preferably ethyl cellulose, its derivatives and mixtures thereof with other binders from the preceding lists of binders or otherwise are the most preferred binders in the context of the invention.

Surfactant

Preferred surfactants in the context of the invention are those which contribute to the formation of an electro-conductive paste with favourable stability, printability, viscosity, sintering and etching properties. Surfactants are well known to the person skilled in the art. All surfactants which are known to the person skilled in the art and which he considers to be suitable in the context of this invention can be employed as the surfactant in the organic vehicle. Preferred surfactants in the context of the invention are those based on linear chains, branched chains, aromatic chains, fluorinated chains, siloxane chains, polyether chains and combinations thereof. Preferred surfactants are single chaine'd double chained or poly chained. Preferred surfactants according to the invention have non-ionic, anionic, cationic, or zwitterionic heads. Pre- ferred surfactants are polymeric and monomeric or a mixture thereof. Preferred surfactants according to the invention can have pigment affinic groups, preferably hydroxyfunctional car- boxylic acid esters with pigment affinic groups (e.g., DISPERBYK ^® -108, manufactured by BYK USA, Inc.), acrylate copolymers with pigment affinic groups (e.g., DISPERBY ^® -116, manufactured by BYK USA, Inc.), modified polyethers with pigment affinic groups (e.g., TEGO ^® DISPERS 655, manufactured by Evonik Tego Chemie GmbH), other surfactants with groups of high pigment affinity (e.g., TEGO ^® DISPERS 662 C, manufactured by Evonik Tego Chemie GmbH). Other preferred polymers according to the invention not in the above list are polyethyleneglycol and its derivatives, and alkyl carboxylic acids and their derivatives or salts, or mixtures thereof. The preferred polyethyleneglycol derivative according to the invention is poly(ethyleneglycol)acetic acid. Preferred alkyl carboxylic acids are those with fully saturated and those with singly or poly unsaturated alkyl chains or mixtures thereof. Preferred carboxylic acids with saturated alkyl chains are those with alkyl chains lengths in a range from about 8 to about 20 carbon atoms, preferably C ₉ H ₁₉ COOH (capric acid), CnH2 ₃ COOH (Laurie acid), C ₁₃ H ₂₇ COOH (myristic acid) C ₁₅ H ₃₁ COOH (palmitic acid), C ₁₇ H ₃₅ COOH (stearic acid) or mixtures thereof. Preferred carboxylic acids with unsaturated alkyl chains are C ₁₈ H3 ₄ O ₂ (oleic acid) and C ₁₈ H ₃₂ O ₂ (linoleic acid). The preferred monomelic surfactant according to the invention is benzotriazole and its derivatives.

Solvent

Preferred solvents according to the invention are constituents of the electro-conductive paste which are removed from the paste to a significant extent during firing, preferably those which are present after firing with an absolute weight reduced by at least about 80% compared to before firing, preferably reduced by at least about 95% compared to before firing. Preferred solvents according to the invention are those which allow an electro-conductive paste to be formed which has favourable viscosity, printability, stability and sintering characteristics and which yields electrodes with favourable electrical conductivity and electrical contact to the substrate. Solvents are well known to the person skilled in the art. All solvents which are known to the person skilled in the art and which he considers to be suitable in the context of this invention can be employed as the solvent in the organic vehicle. According to the inven- tion preferred solvents are those which allow the preferred high level of printability of the electro-conductive paste as described above to be achieved. Preferred solvents according to the invention are those which exist as a liquid under standard ambient temperature and pressure (SATP) (298.15 K, 25 °C, 77 °F), 100 kPa (14.504 psi, 0.986 atm), preferably those with a boiling point above about 90 °C and a melting point above about -20 °C. Preferred solvents according to the invention are polar or non-polar, protic or aprotic, aromatic or non-aromatic. Preferred solvents according to the invention are mono-alcohols, di-alcohols, poly-alcohols, mono-esters, di-esters, poly-esters, mono-ethers, di-ethers, poly-ethers, solvents which comprise at least one or more of these categories of functional group, optionally comprising other categories of functional group, preferably cyclic groups, aromatic groups, unsaturated-bonds, alcohol groups with one or more O atoms replaced by heteroatoms, ether groups with one or more O atoms replaced by heteroatoms, esters groups with one or more O atoms replaced by heteroatoms, and mixtures of two or more of the aforementioned solvents. Preferred esters in this context are di-alkyl esters of adipic acid, preferred alkyl constituents being methyl, ethyl, propyl, butyl, pentyl, hexyl and higher alkyl groups or combinations of two different such alkyl groups, preferably dimethyladipate, and mixtures of two or more adipate esters. Preferred ethers in this context are diethers, preferably dialkyl ethers of ethylene glycol, preferred alkyl constituents being methyl, ethyl, propyl, butyl, pentyl, hexyl and higher alkyl groups or combinations of two different such alkyl groups, and mixtures of two diethers. Preferred alcohols in this context are primary, secondary and tertiary alcohols, preferably tertiary alcohols, terpin- eol and its derivatives being preferred, or a mixture of two or more alcohols. Preferred solvents which combine more than one different functional groups are 2,2,4-trimethyl-l,3-pentanediol monoisobutyrate, often called texanol, and its derivatives, 2-(2-ethoxyethoxy)ethanol, often known as carbitol, its alkyl derivatives, preferably methyl, ethyl, propyl, butyl, pentyl, and hexyl carbitol, preferably hexyl carbitol or butyl carbitol, and acetate derivatives thereof, preferably butyl carbitol acetate, or mixtures of at least 2 of the aforementioned. Additives in the organic vehicle

Preferred additives in the organic vehicle are those additives which are distinct from the aforementioned vehicle components and which contribute to favourable properties of the elec- tro-conductive paste, such as advantageous viscosity, sintering, electrical conductivity of the produced electrode and good electrical contact with substrates. All additives known to the person skilled in the art and which he considers to be suitable in the context of the invention can be employed as additive in the organic vehicle. Preferred additives according to the invention are thixotropic agents, viscosity regulators, stabilising agents, inorganic additives, thickeners, emulsifiers, dispersants or pH regulators. Preferred thixotropic agents in this context are car- boxylic acid derivatives, preferably fatty acid derivatives or combinations thereof. Preferred fatty acid derivatives are C ₉ H ₁₉ COOH (capric acid), CnH ₂₃ COOH (Laurie acid), C ₁₃ H ₂₇ COOH (myristic acid) C ₁₅ H ₃₁ COOH (palmitic acid), C ₁₇ H ₃₅ COOH (stearic acid) C ₁₈ H ₃₄ 0 ₂ (oleic acid), C ₁₈ H ₃ 20 ₂ (linoleic acid) or combinations thereof. A preferred combina- tion comprising fatty acids in this context is castor oil.

Additives in the electro-conductive paste

Preferred additives in the context of the invention are constituents added to the electro- conductive paste, in addition to the other constituents explicitly mentioned, which contribute to increased performance of the electro-conductive paste, of the electrodes produced thereof or of the resulting solar cell. All additives known to the person skilled in the art and which he considers suitable in the context of the invention can be employed as additive in the electro- conductive paste. In addition to additives present in the vehicle, additives can also be present in the electro-conductive paste. Preferred additives according to the invention are thixotropic agents, viscosity regulators, emulsifiers, stabilising agents or pH regulators, inorganic additives, thickeners and dispersants or a combination of at least two thereof, whereas inorganic additives are most preferred. Preferred inorganic additives in this context according to the invention are Mg, Ni, Te, W, Zn, Mg, Gd, Ce, Zr, Ti, Mn, Sn, Ru, Co, Fe, Cu and Cr or a com- bination of at least two thereof, preferably Zn, Sb, Mn, Ni, W, Te and Ru or a combination of at least two thereof, oxides thereof, compounds which can generate those metal oxides on firing, or a mixture of at least two of the aforementioned metals, a mixture of at least two of the aforementioned oxides, a mixture of at least two of the aforementioned compounds which can generate those metal oxides on firing, or mixtures of two or more of any of the above mentioned.

Solar Cell Precursor

A contribution to achieving at least one of the above described objects is made by a solar cell precursor. MWT type solar cell precursors are preferred in the context of the invention. In such a precursor, a first electro-conductive paste as described above is present in at least one channel joining the front and back of a wafer. A second electro-conductive paste, which may be the same as or different to the first paste, is present on the front of the solar cell and makes contact with the paste in the channel. A third electro-conductive paste which may be the same as or different to the first and second pastes is present on the back face of the solar cell and does not make contact with the first paste.

In one embodiment of the solar cell precursor according to the invention, a further electro- conductive paste is present on the front face of the wafer. In a further embodiment, an additional electro-conductive paste is present on the back side of the wafer.

In this application, the terms channel is often used in place of the more general term hole. All descriptions referring to channels apply equally to holes.

In one embodiment of the solar cell precursor according to the invention, at least one hole is a channel joining the front face and the back face of the wafer.

The making of a hole in the Si wafer may be effected by any means known to the skilled person and which he considers suitable in the context of the invention. Preferred methods according to the invention are cutting, etching, heating, melting, burning, boring, piercing, stamping or drilling, such as mechanical drilling or laser drilling, or a combination of at least two thereof, preferably drilling, more preferably laser drilling.

The hole in the Si wafer can have a number of cross-sectional shapes. Any cross-sectional shape known to the skilled person and which he considers suitable in the context of the invention can be employed. Preferred cross-sectional shapes for the hole according to the invention are polygonal and non-polygonal. Preferred polygonal shapes are triangle, square, oblong or other regular or irregular polygons. Preferred non-polygonal shapes are circular or oval or other non-polygonal shapes. The preferred cross-sectional shape according to the invention is cir- cular.

The effective cross-section diameter of the hole in the Si wafer can be altered by the skilled person so as to maximum the performance of the resultant solar cell. Where the cross-sectional shape is not circular, the effective diameter is given as the diameter of a hypothetical circle with an area equal to the cross-sectional area of the hole. It is preferred in the context of the invention that the cross-sectional area of the hole be in a range from about 10 to about 500 μη , more preferably in a range from about 20 to about 300 μηι, most preferably in a range from about 30 to about 250 μπι. Figure 1 exemplifies a common layer configuration for a solar cell according to the invention (excluding additional layers which server purely for chemical and mechanical protection). Individual layers can be added to or omitted from this common layer configuration or individual layers can indeed perform the function of more than one of the layers shown in figure 1. In one embodiment of the invention, a single layer acts as both anti-reflection layer and passivation layer. Figure 2 exemplifies another common layer configuration, wherein a number of layers shown in figure 1 have been omitted.

Optionally, one or more of the layers present on the front face of the wafer, such as the front doped layer, the front passivation layer, and the anti-reflection layer, may extend further along the surface of the channel than into the bulk of the front face. In this context, additional cover- age of the surface of the channel is taken as meaning surface coverage which extends into the channel by a distance greater by at least 5 % than the median thickness of the layer in the bulk of the front face. Additional coverage of the surface of the channel may correspond to full or partial coverage of the additional surface of the channel. The presence or absence of additional surface coverage by front face layers on the surface of the channel can be achieved by altering the sequence in which:

- the front doped layer is applied to the wafer;

additional layers, such as passivation layer, and anti-reflection layer are applied to the wafer;

channels are made in the wafer;

faces are cut from the wafer.

In one embodiment of the present invention, no front face layers have additional coverage on the surface of the channel.

In one embodiment of the solar cell precursor according to the invention, the paste is in direct contact with the Si surface of the hole. In one embodiment of the solar cell precursor according to the invention, the Si surface in at least one hole comprises at least a p-type doped section and an n-type doped section.

As exemplified in figure 5, a preparation of a wafer in which no front face layers have additional coverage on the surface of the channel can be achieved by the following sequence of steps:

front doped layer applied;

front passivation layer and anti-reflection layer applied;

channel made;

back and side faces cut to remove front doped layer, passivation layer and anti- reflection layer therefrom. In one embodiment of the invention, the front doped layer is present with additional coverage of the surface of the channel. In one aspect of this embodiment, a passivation layer is also present with additional coverage of on the surface of the channel. In a further aspect of this embodiment an anti-reflection layer is present with additional coverage on the surface of the channel. As exemplified in figure 4, a preparation of a wafer in which front doped layer, passivation layer and anti-reflection layer are all present on the entire surface of the channel can be achieved by the following sequence of steps: channel made;

front doped layer applied;

front passivation layer and anti-reflection layer applied;

back and side faces cut to remove front doped layer, passivation layer and anti- reflection layer therefrom.

Process for producing a Solar Cell

A contribution to achieving at one of the aforementioned objects is made by a process for producing a solar cell at least comprising the following as process steps:

i) provision of a solar cell precursor as described above, in particular combining any of the above described embodiments; and

ii) firing of the solar cell precursor to obtain a solar cell.

It is preferred according to the invention for formation of the p-n junction in the wafer to precede the making of the hole. In one embodiment, the provision according to step i) at least comprises the steps:

a) provision of a Si wafer with a back doped layer and a front doped layer of opposite doping types;

b) making of at least one hole in the wafer; c) introduction of an electro-conductive paste into at least one hole to give a precursor according to the invention;

Printing

It is preferred according to the invention that each of the front, back and plug electrodes be applied by applying an electro-conductive paste and then firing that electro-conductive paste to obtain a sintered body. The electro-conductive paste can be applied in any manner known to the person skilled in that art and which he considers suitable in the context of the invention including but not limited to impregnation, dipping, pouring, dripping on, injection, spraying, knife coating, curtain coating, bmshing or printing or a combination of at least two thereof, wherein preferred printing techniques are ink-jet printing, screen printing, tampon printing, offset printing, relief printing or stencil printing or a combination of at least two thereof. It is preferred according to the invention that the electro-conductive paste is applied by printing, preferably by screen printing. It is preferred according to the invention that the screens have mesh opening with a diameter in a range from about 20 to about 100 μπι, more preferably in a range from about 30 to about 80 μπι, and most preferably in a range from about 40 to about 70 μπι. As detailed in the solar cell precursor section, it is preferred for the electro-conductive paste applied to the channel to be as described in this invention. The electro-conductive pastes used to form the front and back electrodes can be the same or different to the paste used in the channel, preferably different, and can be the same as or different to each other.

Firing It is preferred according to the invention for electrodes to be formed by first applying an electro-conductive paste and then firing said electro-conductive paste to yield a solid electrode body. Firing is well known to the person skilled in the art and can be effected in any manner known to him and which he considers suitable in the context of the invention. Where glass frit is used as the inorganic reaction system, it is preferred for firing to be carried out above the glass transition temperature of the glass frit. According to the invention the maximum temperature set for the firing is below about 900 °C, preferably below about 860 °C. Firing temperatures as low as about 820 °C have been employed for obtaining solar cells. It is preferred according to the invention for firing to be car- ried out in a fast firing process with a total firing time in a range from about 30 s to about 3 minutes, more preferably in a range from about 30 s to about 2 minutes and most preferably in a range from about 40 s to about 1 minute. The time above about 600 °C is most preferably in a range from about 3 to about 7 s. Firing of electro-conductive pastes on the front face, back face and in the hole can be carried out simultaneously or sequentially. Simultaneous firing is appropriate if the electro-conductive pastes have similar, preferably identical, optimum firing conditions. Where appropriate, it is preferred according to the invention for firing to be carried out simultaneously. Solar Cell

A contribution to achieving at least one of the above described objects is made by a solar cell obtainable by a process according to the invention. Preferred solar cells according to the invention are those which have a high efficiency in terms of proportion of total energy of incident light converted into electrical energy output and which are light and durable.

Anti-reflection Coating

According to the invention, an anti-reflection coating can be applied as the outer and often as the outermost layer before the electrode on the front face of the solar cell. Preferred anti- reflection coatings according to the invention are those which decrease the proportion of incident light reflected by the front face and increase the proportion of incident light crossing the front face to be absorbed by the wafer. Anti-reflection coatings which give rise to a favourable absorption/reflection ratio, are susceptible to etching by the employed electro-conductive paste but are otherwise resistant to the temperatures required for firing of the electro-conductive paste, and do not contribute to increased recombination of electrons and holes in the vicinity of the electrode interface are favoured. All anti-reflection coatings known to the person skilled in the art and which he considers to be suitable in the context of the invention can be employed. Preferred anti-reflection coatings according to the invention are SiN _x , Si0 ₂ , Α1 ₂ θ3, Ti0 ₂ or mixtures of at least two thereof and/or combinations of at least two layers thereof, wherein SiN _x is particularly preferred, in particular where an Si wafer is employed.

The thickness of anti-reflection coatings is suited to the wavelength of the appropriate light. According to the invention it is preferred for anti-reflection coatings to have a thickness in a range from about 20 to about 300 nm, more preferably in a range from about 40 to about 200 nm and most preferably in a range from about 60 to about 90 nm.

Passivation Layers According to the invention, one or more passivation layers can be applied to the front and/or back side as outer or as the outermost layer before the electrode, or before the anti-reflection layer if one is present. Preferred passivation layers are those which reduce the rate of electron/hole recombination in the vicinity of the electrode interface. Any passivation layer which is known to the person skilled in the art and which he considers to be suitable in the context of the invention can be employed. Preferred passivation layers according to the invention are silicon nitride, silicon dioxide and titanium dioxide, silicon nitride being most preferred. According to the invention, it is preferred for the passivation layer to have a thickness in a range from about 0.1 nm to about 2 μιη, more preferably in a range from about 10 nm to about 1 μπι and most preferably in a range from about 30 nm to about 200 nm.

Electrodes

It is preferred for the plug electrode in the solar cell to have an accumulation of glass at the boundary where the electrode meets the Si wafer, preferably in the form of an electrically insu- lating layer of glass. In one embodiment of the solar cell according to the invention, the solar cell at least comprises as solar cell parts:

i) a wafer with at least one hole with a Si surface;

ii) an electrode comprised by a hole,

wherein the concentration of glass in the electrode is higher at the surface at which the electrode contacts the Si surface than in the main body of the electrode.

Additional Protective layers In addition to the layers described above which directly contribute to the principle function of the solar cell, further layers can be added for mechanical and chemical protection. The cell can be encapsulated to provide chemical protection. Encapsulations are well known to the person skilled in the art and any encapsulation can be employed which is known to him and which he considers suitable in the context of the invention. According to the invention, transparent pol- ymers, often referred to as transparent thermoplastic resins, are preferred as the encapsulation material, if such an encapsulation is present. Preferred transparent polymers in this context are for example silicon rubber and polyethylene vinyl acetate (PVA).

A transparent glass sheet can be added to the front of the solar cell to provide mechanical pro- tection to the front face of the cell. Transparent glass sheets are well known to the person skilled in the art and any transparent glass sheet known to him and which he considers to be suitable in the context of the invention can be employed as protection on the front face of the solar cell. A back protecting material can be added to the back face of the solar cell to provide mechanical protection. Back protecting materials are well known to the person skilled in the art and any back protecting material which is known to the person skilled in the art and which he considers to be suitable in the context of the invention can be employed as protection on the back face of the solar cell. Preferred back protecting materials according to the invention are those having good mechanical properties and weather resistance. The preferred back protection ma- terials according to the invention is polyethylene terephthalate with a layer of polyvinyl fluoride. It is preferred according to the invention for the back protecting material to be present underneath the encapsulation layer (in the event that both a back protection layer and encapsulation are present).

A frame material can be added to the outside of the solar cell to give mechanical support. Frame materials are well known to the person skilled in the art and any frame material known to the person skilled in the art and which he considers suitable in the context of the invention can be employed as frame material. The preferred frame material according to the invention is aluminium.

Solar panels

A contribution to achieving at least one of the above mentioned objects is made by a module comprising at least a solar cell obtained as described above, in particular according to at least one of the above described embodiments, and at least one more solar cell. A multiplicity of solar cells according to the invention can be arranged spatially and electrically connected to form a collective arrangement called a module. Preferred modules according to the invention can take a number of forms, preferably a rectangular surface known as a solar panel. A large variety of ways to electrically connect solar cells as well as a large variety of ways to mechanically arrange and fix such cells to form collective arrangements are well known to the person skilled in the art and any such methods known to him and which he considers suitable in the context of the invention can be employed. Preferred methods according to the invention are those which result in a low mass to power output ratio, low volume to power output ration, and high durability. Aluminium is the preferred material for mechanical fixing of solar cells according to the invention. Description of the Drawings

The invention is now explained by means of figures which are intended for illustration only and are not to be considered as limiting the scope of the invention. In brief,

Fig. 1 shows a cross sectional view a common layer configuration for a solar cell,

Fig. 2 shows a cross sectional view of the minimum layer configuration for a solar cell,

Fig. 3a, 3b and 3c together illustrate the process of firing a front side paste.

Figures 4a, 4b, 4c, 4d and 4e together illustrate a process for preparing a wafer with a hole, a front doped layer and optional additional front layers, wherein the front doped layer and the optional additional front layers cover the surface of the hole.

Figures 5a, 5b, 5c, 5d and 5e together illustrate a process for preparing a wafer with a hole, a front doped layer and optional additional front layers, wherein the front doped layer and the optional additional front layers do not cover the surface of the hole.

Figure 6 shows an SEM image of a plug electrode on which circles corresponding to silver content have been drawn according to the algorithm in the test method for particle diameter in electrode. Figure 7 shows a schematic for a cross sectional cut of wafer with six samples for analysis of glass content in the electrode.

Figure 8 shows an exemplary form for a bimodal diameter distribution of silver particles in a plug electrode. Figure 1 shows a cross sectional view of a common layer configuration for a solar cell 100 according to the invention (excluding additional layers which serve purely for chemical and mechanical protection). Starting from the back face and continuing towards the front face the solar cell 200 comprises a back electrode 107, a back passivation layer 112, a highly doped back layer 111, a back doped layer 104, a p-n junction boundary 102, a front doped layer 103, a front passivation layer 110, an anti-reflection layer 109, and front electrode 106, wherein the front electrode penetrates through the anti-reflection layer 109 and the front passivation layer 110 and into the front doped layer 103 far enough to form a good electrical contact with the front doped layer, but not so far as to shunt the p-n junction boundary 102. A hole electrode 105 is present in a hole joining the front and back faces of the solar cell. This electrode is preferably an Ag electrode according to the invention. In the case that 100 represents an n-type cell, the back electrode 107 is preferably a silver electrode, the highly doped back layer 111 is preferably Si heavily doped with P, the back doped layer 104 is preferably Si lightly doped with P, the front doped layer 103 is preferably Si heavily doped with B, the anti-reflection lay- er 109 is preferably a layer of silicon nitride and the front electrode 106 is preferably a mixture of silver and aluminium. In the case that 100 represents a p-type cell, the back electrode 107 is preferably a mixed silver and aluminium electrode, the highly doped back layer 111 is preferably Si heavily doped with B, the back doped layer 104 is preferably Si lightly doped with B, the front doped layer 103 is preferably Si heavily doped with P, the anti-reflection layer 109 is preferably a layer of silicon nitride and the front electrode 106 is preferably silver. Figure 2 is schematic and shows only one hole. The invention does not limit the number of back electrodes 107, front electrodes 106, holes, or hole electrodes 105.

Figure 2 shows a cross sectional view of a solar cell 200 and represents the minimum required layer configuration for a solar cell according to the invention. Starting from the back face and continuing towards the front face the solar cell 100 comprises a back electrode 107, a back doped layer 104, a p-n junction boundary 102, a front doped layer 103 and a front electrode 106, wherein the front electrode penetrates into the front doped layer 103 enough to form a good electrical contact with it, but not so much as to shunt the p-n junction boundary 102. The back doped layer 104 and the front doped layer 103 together constitute a single doped Si wafer 101. A hole electrode 105 is present in a hole joining the front and back faces of the solar cell. This electrode is preferably an Ag electrode according to the invention. In the case that 200 represents an n-type cell, the back electrode 107 is preferably a silver electrode, the back doped layer 104 is preferably Si lightly doped with P, the front doped layer 103 is preferably Si heavily doped with B and the front electrode 106 is preferably a mixed silver and aluminium electrode. In the case that 200 represents a p-type cell, the back electrode 107 is preferably a mixed silver and aluminium electrode, the back doped layer 104 is preferably Si lightly doped with B, the front doped layer 103 is preferably Si heavily doped with P and the front electrode 106 is preferably a silver electrode. This diagram is schematic and the invention does not limit the number of front electrodes 105, back electrodes 107, holes or hole electrodes 105.

Figures 3 a, 3 b and 3 c together illustrate the process of firing a precursor without front layers covering the surface of the hole to obtain a solar cell. Figures 3a, 3b and 3c are schematic and generalised and additional layers further to those constituting the p-n junction are considered simply as optional additional layers without more detailed consideration.

Figure 3a illustrates a wafer before application of front electrode and hole electrode, 300a. Starting from the back face and continuing towards the front face the wafer before application of front electrode 300a optionally comprises additional layers on the back face 316, a back doped layer 104, a p-n junction boundary 102, a front doped layer 103 and additional layers on the front face 314. The additional layers on the back face 316 can comprise any of a back electrode, a back passivation layer, a highly doped back layer or none of the above. The additional layer on the front face 314 can comprise any of a front passivation layer, an anti-reflection layer or none of the above.

Figure 3b shows a wafer with electro-conductive pastes applied to the front face and to a hole before firing 300b. In addition to the layers present in 300a described above, an electro- conductive paste 306 is present on the surface of the front face and an electro-conductive paste 305 is present in the hole. Figure 3 c shows a wafer with front electrode applied 300c. In addition to the layers present in 300a described above, a front side electrode 106 is present which penetrates from the surface of the front face through the additional front layers 314 and into the front doped layer 103 and is formed from the electro-conductive paste 306 of figure 3b by firing. A hole electrode 105, which has been formed from the electro-conductive paste 305 of figure 3b by firing, is present in the hole.

In figures 3 b and 3 c, only one hole, one front electrode and one hole electrode are shown. This diagram is schematic and the invention does not limit the number of holes, front electrodes or hole electrodes.

Figures 4a, 4b, 4c, 4d and 4e together illustrate a process for preparing a wafer with a hole, a front doped layer and optional additional front layers, wherein the front doped layer and the optional additional front layers cover the surface of the hole.

Figure 4a shows a wafer which will later constitute the back doped layer 104.

Figure 4b shows a wafer 104 after a hole 315 has been made. Figure 4c shows a wafer 104 with a hole 315 after what will later constitute the front doped layer 103 has been applied. This layer 103 is present on each side of the wafer and on the surface of the hole.

Figure 4d shows the wafer 104 with a hole 315 and front doped layer 103 after additional front layers 314 have been applied. This layer 314 is present on each side of the wafer and on the surface of the hole.

Figure 4e shows the wafer 104 with a hole 315, front doped layer 103 and additional front layers 314 after cutting back and sides so as to leave the front doped layer 103 and additional front layers 314 present on the front face and surface of hole only. Figures 5a, 5b, 5c, 5d and 5e together illustrate a process for preparing a wafer with a hole, a front doped layer and optional additional front layers, wherein the front doped layer and the optional additional front layers do not cover the surface of the hole.

Figure 5a shows a wafer which will later constitute the back doped layer 104.

Figure 5b shows a wafer 104 after what will later constitute the front doped layer 103 has been applied. This layer 103 is present on each side of the wafer.

Figure 5c shows the wafer 104 with front doped layer 103 after additional front layers 314 have been applied. This layer 314 is present on each side of the wafer and on the surface of the hole. Figure 5d shows the wafer 104 with front doped layer 103 and additional front layers 314 after a hole 315 has been made. The front doped layer 103 and additional front layers 314 are present on all sides of the wafer but not on the surface of the hole.

Figure 5e shows the wafer 104 with front doped layer 103, additional front layers 314 and hole 315 after cutting back and sides so as to leave the front doped layer 103 and additional front layers 314 present on the front face only.

Figure 6 displays part of an exemplary electron micrograph cross-sectional cut of a processed wafer exhibiting silver particles. The area corresponding to silver content 601, in contrast to area corresponding to non-silver content 602, was identified and filled with circles of decrementing diameter according to the algorithm given in the test method for determining particle diameter distribution in the electrode. For purposes of clarity, Figure 6 shows the image at the point where the fitting algorithm has been partially completed, for diameters decremented from 50 μπι down to 0.5 μπι. Figure 6 shows an exemplary portion of the area to be analysed ac- cording to the test method (1 mm ² ). The diameters of the marked circles of that portion are as follows:

15=0.82μηι, 16=7.15μηι, 17=0.8μπι, 18=2.38μηι ₅ 19=1.37μηι, 20=0.99μπι, 21=1.02μπι,

29=0.92μπι, 30=0.92μηι, 31=0.86μπι, 32=1.1 Ιμηι, 33=0.87μηι, 34=1.47μπι, 35=0.66μτη,

36=1.21μπι, 37=0.65μηι, 38=4.3μιη, 39=2.41μιη, 40=0.83μπι, 41=1.54μπι, 42=1.06μπι,

57=1.38μηι, 58=3.02μηι ₅ 59=2.3 Ιμπι, 60=0.95μπι ₅ 61=0.74μηι, 62=1.66μπι, 63=0.86μπι,

Figure 7 shows schematically a cross sectional cut 700 of a cell 100 with a plug electrode 105. The cut is made so at to contain the axis of the plug 704. Three 1 mm samples 701 are taken from this cut along the axis of the plug 704 at ¼, ½, and ¾ of the distance from the front face to the back face. Three 1 mm ² samples 702 are taken along the boundary where the plug electrode 105 meets the wafer 101 at ¼, ½, and ¾ of the distance from the front face to the back face. The mid point 703 between the front face and the back face of the wafer is also shown in figure 7.

Figure 8 shows an exemplary bi-modal diameter distribution of silver particles in a plug electrode as determined by the test method. Local maxima 801 are present at about 1 μπι and at about 5.5 μπι giving a corresponding separation Δ of about 4.5 μηι. Measurements were taken at 0.1 μπι intervals in a range from 0 to 50 μπι (for clarity, only the portion of the graph up to 18 μπι is shown). The graph has been displayed in a normalised form in which the frequencies sum to 1.

Test Methods The following test methods are used in the invention. In absence of a test method, the ISO test method for the feature to be measured being closest to the earliest filing date of the present application applies. In absence of distinct measuring conditions, standard ambient temperature and pressure (SATP) as a temperature of 298.15 K (25 °C, 77 °F) and an absolute pressure of 100 kPa (14.504 psi, 0.986 atm) apply.

Specific Surface Area

BET measurements to determine the specific surface area of silver particles are made in ac- cordance with DIN ISO 9277:1995. A Gemini 2360 (from Micromeritics) which works according to the SMART method (Sorption Method with Adaptive dosing Rate), is used for the measurement. As reference material Alpha aluminium oxide CRM BAM-PM-102 available from BAM (Bundesanstalt fur Materialforschung und -priifung) is used. Filler rods are added to the reference and sample cuvettes in order to reduce the dead volume. The cuvettes are mounted on the BET apparatus. The saturation vapour pressure of nitrogen gas (N ₂ 5.0) is determined. A sample is weighed into a glass cuvette in such an amount that the cuvette with the filler rods is completely filled and a minimum of dead volume is created. The sample is kept at 80°C for 2 hours in order to dry it. After cooling the weight of the sample is recorded. The glass cuvette containing the sample is mounted on the measuring apparatus. To degas the sam- pie, it is evacuated at a pumping speed selected so that no material is sucked into the pump. The mass of the sample after degassing is used for the calculation. The dead volume is determined using Helium gas (He 4.6). The glass cuvettes are cooled to 77 K using a liquid nitrogen bath. For the adsorptive, N ₂ 5.0 with a molecular cross-sectional area of 0.162 nm ² at 77 K is used for the calculation. A multi-point analysis with 5 measuring points is performed and the resulting specific surface area given in m /g.

Glass Transition Temperature

The glass transition temperature T _g for glasses is determined using a DSC apparatus Netzsch STA 449 F3 Jupiter (Netzsch) equipped with a sample holder HTP 40000A69.010, thermo- couple Type S and a platinum oven Pt S TC:S (all from Netzsch). For the measurements and data evaluation the measurement software Netzsch Messung V5.2.1 and Proteus Thermal Analysis V5.2.1 are applied. As pan for reference and sample, aluminium oxide pan GB 399972 and cap GB 399973 (both from Netzsch) with a diameter of 6.8 mm and a volume of about 85 μΐ are used. An amount of about 20 to about 30 mg of the sample is weighted into the sample pan with an accuracy of 0.01 mg. The empty reference pan and the sample pan are placed in the apparatus, the oven is closed and the measurement started. A heating rate of 10 K/min is employed from a starting temperature of 25°C to an end temperature of 1000°C. The balance in the instrument is always purged with nitrogen (N2 5.0) and the oven is purged with synthetic air (80% N2 and 20% 02 from Linde) with a flow rate of 50 ml/min. The first step in the DSC signal is evaluated as glass transition using the software described above and the determined onset value is taken as the temperature for T _g .

Viscosity

Viscosity measurements were performed using the Thermo Fischer Scientific Corp. "Haake Rheostress 600" equipped with a ground plate MPC60 Ti and a cone plate C 20/0,5° Ti and software "Haake RheoWin Job Manager 4.30.0". After setting the distance zero point, a paste sample sufficient for the measurement was placed on the ground plate. The cone was moved into the measurement positions with a gap distance of 0.026 mm and excess material was removed using a spatula. The sample was equilibrated to 25 °C for three minutes and the rotational measurement started. The shear rate was increased from 0 to 20 s ^"1 within 48 s and 50 equidistant measuring points and further increased to 150 s ^"1 within 312 s and 156 equidistant measuring points. After a waiting time of 60 s at a shear rate of 150 s ^"1 , the shear rate was re- duced from 150 s ^"1 to 20 s ^"1 within 312 s and 156 equidistant measuring points and further reduced to 0 within 48 s and 50 equidistant measuring points. The micro torque correction, micro stress control and mass inertia correction were activated. The viscosity is given as the measured value at a shear rate of 100 s ^"1 of the downward shear ramp. Sheet Resistance For measuring the sheet resistance of a doped silicon wafer surface, the device "GP4-Test Pro" equipped with software package "GP-4 Test 1.6.6 Pro" from the company GP solar GmbH is used. For the measurement, the 4 point measuring principle is applied. The two outer probes apply a constant current and two inner probes measure the voltage. The sheet resistance is deduced using the Ohmic law in Ω/square. To determine the average sheet resistance, the measurement is performed on 25 equally distributed spots of the wafer. In an air conditioned room with a temperature of 22 ± 1 °C, all equipment and materials are equilibrated before the measurement. To perform the measurement, the "GP-Test.Pro" is equipped with a 4-point measur- ing head (part no. 04.01.0018) with sharp tips in order to penetrate the anti-reflection and/or passivation layers. A current of 10 mA is applied. The measuring head is brought into contact with the non metalized wafer material and the measurement is started. After measuring 25 equally distributed spots on the wafer, the average sheet resistance is calculated in Ω/square. Particle Size Determination (d ₁₀ , d ₅₀ , d ₉ o and particle distribution of powder)

Particle size determination for particles is performed in accordance with ISO 13317-3:2001. A Sedigraph 5100 with software Win 5100 V2.03.01 (from Micromeritics) which works according to X-ray gravitational technique is used for the measurement. A sample of about 400 to 600 mg is weighed into a 50 ml glass beaker and 40 ml of Sedisperse PI 1 (from Micromeritics, with a density of about 0.74 to 0.76 g/cm ³ and a viscosity of about 1.25 to 1.9 mPa s) are added as suspending liquid. A magnetic stirring bar is added to the suspension. The sample is dispersed using an ultrasonic probe Sonifer 250 (from Branson) operated at power level 2 for 8 minutes while the suspension is stirred with the stirring bar at the same time. This pre-treated sample is placed in the instrument and the measurement started. The temperature of the suspension is recorded (typical range 24°C to 45°C) and for calculation data of measured viscosity for the dispersing solution at this temperature are used. Using density and weight of the sample (10.5 g/cm ³ for silver) the particle size distribution function is determined. d ₅₀ , d ₁₀ and d ₉₀ can be read directly from the particle distribution function. For the evaluation of multi- modal size, distribution plots of mass frequency against diameter are generated and the peak maxima are determined therefrom.

Particle Size Determination (d ₁₀ , d ₅ o, d ₉₀ and particle distribution in paste) For the determination of the size distribution of the metal particles in the paste, the following procedure was followed. The organic part is removed by solvent extraction using solvents such as methanol, ethanol, isopropanol, dichloromethane, chloroform, hexane. This can be performed using a Soxhlet apparatus or a combination of dissolution, sedimentation and filtration techniques known to the person skilled in the art. The inorganic part except the metal particles is removed by treatment with aqueous non-oxidizing acids such as hydrochloric acid etc., followed by treatment with bases such as aqueous sodium hydroxide, potassium hydroxide etc. followed by treatment with aqueous hydrofluoric acid. This can be performed using a Soxhlet apparatus or a combination of dissolution, sedimentation and filtration techniques. In a final step, the remaining metal particles are washed with deionized water and dried. Particle size of the resulting powder is measured as described for powders above.

Dopant Level

Dopant levels are measured using secondary ion mass spectroscopy.

Adhesion

The solar cell sample to be tested is secured in a commercially available soldering table M300- 0000-0901 from Somont GmbH, Germany. A solder ribbon from Bruker Spalek (ECu + 62Sn- 36Pb-2Ag) is coated with flux Kester 952S (from Kester) and adhered to the finger line or bus bar to be tested by applying the force of 12 heated pins which press the solder ribbon onto the finger line or bus bar. The heated pins have a set temperature of 280 °C and the soldering preheat plate on which the sample is placed is set to a temperature of 175 °C. After cooling to room temperature, the samples are mounted on a GP Stable-Test Pro tester (GP Solar GmbH, Germany). The ribbon is fixed at the testing head and pulled with a speed of 100 mm/s and in a way that the ribbon part fixed to the cell surface and the ribbon part which is pulled enclose an angle of 45 °. The force required to remove the bus bar/finger is measured in Newton. This process is repeated for contact at 10 equally spaced points along the finger/bus bar, including one measurement at each end. The mean is taken of the 10 results.

Scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDX)

The solar cell is cut in a way that the area of interest is laid open. The cut sample is placed in a container filled with embedding material and oriented such that the area of interest is on top. As embedding material, EpoFix (Struers) is used, mixed according to the instructions. After 8 hours curing at room temperature the sample can be processed further. In a first step the sample is ground with a Labopol-25 (Struers) using silicon carbide paper 180-800 (Struers) at 250 rpm. In further steps the sample is polished using a Rotopol-2 equipped with a Retroforce-4, MD Piano 220 and MD allegro cloth and DP-Spray P 3um diamond spray (all from Struers). Coating with a carbon layer is performed with a Med 010 (Balzers) at a pressure of 2 mbar using a carbon thread 0.27 g/m E419ECO (from Piano GmbH). The examination was performed with a Zeiss Ultra 55 (Zeiss), equipped with a field emission electrode, an accelerating voltage of 20 kV and at a pressure of about 3*10 ^"6 mbar. Images of relevant areas were taken and analysed using image analysis software ImageJ Version 1.46r (Image Processing and anal- ysis in Java, by Wayne Rasband, http://rsb.info.nih.gov/ij). In order to identify Ag particles, the intense Ag _L signal at about 3.4 keV in EDX was used to identify 10 silver particles and the average SEM greyscale intensity for those 10 particles was used to identify further Ag particles from the SEM picture. Glass particles were similarly identified using an analogous SEM/EDX method.

Particle Diameter of Ag in electrode

A cross sectional cut was made through the electrode and processed as described in the SEM test section above to give three square cross sectional samples from within the electrode with an area of 1 mm ² . The samples was taken from a plane containing the plug axis at ¼, ½, and ¾ of the distance between the front face and the back face at the axis of the plug electrode, as described for the first three samples in figure 7. For each sample, the areas corresponding to Ag were identified as described in the SEM section. Circles of decrementing diameter were drawn onto the image according to the following algorithm:

1. Superimpose a square grid with a 0.01 μπι separation onto the image.

2. Each point either corresponds to an area of silver, or not so. Each point which corresponds to silver is initially available to be allocated to circles. Points which do not correspond to silver are not available to be allocated to circles.

3. Start with a diameter of 50 μπι

4. Starting from the top left point in the grid, proceed through the points of the top row from left to right, carrying out step 4a. for each point. Repeat for subsequent rows from top to bottom, arriving finally at the bottom right, all points having been processed.

4a. For each point, if all points within a distance equal to half of the current diameter (initially 50 μιη) of the point are available to be allocated to circles, then:

i.. draw a circle with a diameter equal to the current diameter centred on the current grid point

ii. mark all points within a distance equal to half of the current diameter from the point as unavailable to be allocated to circles iii. Increase by one the cumulative frequency counter for the current value of the diameter (initially set to 0)

5. Upon having proceeded through all of the grid points for a certain value of particle diameter, record the cumulative frequency counter for that value of diameter, decrement the current diameter by 0.1 μπι and carry out step 4. using that value for diameter. Once step 4. has been completed for all values of diameter from 50 μπι down to 0.1 μηι, the algorithm is complete.

Once the circle drawing algorithm had been carried out, the cumulative frequency counter values were multiplied by the square of the corresponding diameter to better correspond to a mass frequency distribution, a best fit curve was fitted to the data using numerical least square regression and the positions of maxima calculated. The result was given as a mean for the three samples. If the standard deviation for the results of the three samples was more than 15 % of the mean value, further samples were taken at the mid point between each adjacent pair of existing samples and mid way between the terminal samples and the adjacent face of the wafer and the mean of all samples given. This process was repeated until the standard deviation was less than 15 % of the mean value.

Glass Content in Electrode

Six square sample surfaces with an area of 1 (mm) ² were cut from the plug electrode as illus- trated in figure 7. All samples were cut from a plane containing the axis of the plug electrode, three along the plug axis, ¼, ½, and ¾ of the distance between the front face and the back face, three at the boundary with the wall of the channel, ¼, ½, and ¾ of the distance between the front face and the back face. All six samples were prepared and analysed for glass content as described above in the SEM test method. For each sample, a square grid was superimposed on the image with a point separation of 0.01 μπι. For each sample, the ratio of the number of points which correspond with glass, as determined by the SEM method, to the total number of points in the sample was returned as the proportion of the sample by weight corresponding to glass content. A mean value was taken for the samples from the plug axis and a mean value was taken from the samples from the electrode boundary. If the standard deviation for the re- suits of the either of the sets of three samples was more than 15 % of the respective mean value, further samples were taken for both sets at the mid point between each adjacent pair of existing samples and mid way between the terminal samples and the adjacent face of the wafer and the mean of all samples in the set given. This process was repeated until the standard deviation was less than 15 % of the mean value for both sets. The glass content at the border of the electrode and the wafer was compared to the glass content at the axis of the plug. Examples

The invention is now explained by means of examples which are intended for illustration only and are not to be considered as limiting the scope of the invention.

Example 1— paste preparation

A paste was made by mixing the following: Organic vehicle (composition given in Table 1); Ag powders from Sigma- Aldrich, product number: 327093, with particles diameter as given in Table 2 in proportions as given in Table 3; glass frit ground to d ₅₀ of 2 μιη, and ZnO (Sigma- Aldrich, article number: 255750) as additive. The paste was passed through a 3-roll mill at progressively increasing pressures from 0 to 8 bar. The viscosity was measured as mentioned above and appropriate amounts of organic vehicle with the composition given in Table 1 were added to adjust the paste viscosity toward a target in a range from about 20 to about 40 Pas. The wt. % of the constituents of the paste are given in Table 3.

Table 1

Table 2

Example 2 - Solar Cell Preparation and Measurement of Adhesion

Pastes were applied to full-square mono-crystalline p-type wafers without an n-type emitter layer and without any passivation or anti-reflective layer. The wafer dimensions were 156 mm x 156 mm, the front side had a textured surface applied by an alkaline etching process. The example paste was screen-printed onto the front side of the wafer as a set of parallel finger lines. Printing was carried out using a ASYS Automatisierungssysteme GmbH Ekra E2 screen printer and a standard H-pattern screen from Koenen GmbH. The screen had 75 finger lines with 110 μηι openings and two 1.5 mm wide bus bars. The Emulsion over mesh was in a range from about 16 to about 20 μπι, the screen had 325 mesh and 20 μπι stainless steel wire. The printing parameters were 1.2 bar squeegee pressure, forward squeegee speed 150 mm/s and flooding speed 200 mm/s and a snap off of 1.5 mm. The device with the printed pattern was then dried in an oven for 10 minutes at 150 °C. The substrates were then fired sun-side up with a Centrotherm Cell & Module GmbH "c-fire" fast firing furnace. The furnace consists of 6 zones. Zone 1 was set to 350 °C, zone 2 to 475 °C, zone 3 to 470 °C, zone 4 to 540 °C, zone 5 to 840 °C and zone 6 to 880 °C. The belt speed was set to 5100 mm/min. The fully processed samples were then tested for adhesion of the back finger lines formed from the example paste, a high value being considered as favourable. I _rev was evaluated at -12 V, a low value being considered as favourable. For each paste, the average result for the adhesion for 6 samples is shown in Table 4. Adhesion was determined as described above.

Table 4

** Results expressed on a scale: ~ very low, - low, + high, ++very high Analysis of content of plug electrode

The maxima of the diameter distribution of Ag in the plug electrode were determined according to the test method. Maxima at about 5.5 μη and about 1 μιη, corresponding to a separation of about 4.5 μπι were observed.

The glass content in the plug electrode was measured at the axis of the plug electrode and at the boundary with the silicon wafer as described in the test method. It was found that the glass content was higher at the boundary where the plug electrode meets the silicon wafer than at the axis of the plug electrode.

Reference list

100 Solar cell

101 Doped Si wafer

102 p-n junction boundary

103 Front doped layer

104 Back doped layer

105 Electrode in channel

106 Front electrode

107 Back electrode

109 Anti-reflection layer

110 Front passivation layer

111 Highly doped back layer

112 Back passivation layer

113 Surface of hole

200 Solar cell

300a Solar cell precursor

300b Solar cell precursor

300c Solar cell precursor

305 Plug paste

306 Front paste

314 Additional layers on front face

315 Hole

316 Additional back layers

401 Wafer

601 Silver particles

602 Non-silver material

700 Cross sectional slice of wafer with plug electrode

701 1 mm ² samples taken along plug axis

702 1 mm ² samples taken along channel wall Mid point between front face and back face of wafer plug axis