Glass comprising molybdenum and lead in a solar cell paste

Abstract

In general, the invention relates to electro-conductive pastes comprising a glass which comprises molybdenum and lead as a constituent of a solar cell paste, and the use of such in the preparation of photovoltaic solar cells. More specifically, the invention relates to electroconductive pastes, precursors, processes for preparation of solar cells, solar cells and solar modules. The invention relates to an electro-conductive paste at least comprising as paste constituents: a) metallic particles; b) a glass; c) an organic vehicle; and d) an additive; wherein the glass comprises the following: i) Pb in the range from about 1 to about 94 wt. %; ii) Mo in the range from about 2 to about 30 wt. %; iii) O in the range from about 1 to about 50 wt. %; with the wt. % in each case being based on the total weight of the glass.

Claims

1. An electro-conductive paste comprising as paste constituents: a) metallic particles; b) a glass; c) an organic vehicle; and d) an additive; wherein the glass comprises the following: i) Pb in the range from about 1 to about 94 wt. %; ii) Mo in the range from about 5 to about 10 wt. %; iii) O in the range from about 1 to about 50 wt. %; and iv) Te with the wt. % in each case being based on the total weight of the glass.

2. The paste according to claim 1, wherein the Pb is present in the glass in oxidation state in the range from +2, to +4.

3. The paste according to claim 1, wherein the Mo is present in the glass in oxidation state in the range from +3 to +6.

4. The paste according to claim 1, wherein the metallic particles are Ag particles.

5. The paste according to claim 1, wherein the glass is present in the paste in the range from about 1 wt. % to about 6 wt. %, based on the total weight of the paste.

6. A precursor comprising the following: a. a wafer; b. an electro-conductive paste according to claim 1 superimposed over the wafer.

7. The precursor according to claim 6, wherein the wafer is an Si wafer.

8. The precursor according to claim 6, wherein the wafer comprises at least one n-doped region and at least one p-doped region.

9. The precursor according to claim 8, wherein the paste is superimposed over a p-doped region.

10. The precursor according to claim 8, wherein the paste is superimposed over an n-doped region.

11. The precursor according to claim 8, wherein the paste is superimposed over the front face of the wafer.

12. The precursor according to claim 8, wherein the paste is superimposed over the back face of the wafer.

13. A process for the preparation of a solar cell at least comprising the steps: i) provision of a precursor according to claim 8; ii) firing of the precursor to obtain a solar cell.

14. A solar cell obtainable by the process according to claim 13.

Description

DESCRIPTION OF THE DRAWINGS

(1) 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,

(2) FIG. 1 shows a cross sectional view of the minimum layer configuration for a solar cell,

(3) FIG. 2 shows a cross sectional view a common layer configuration for a solar cell,

(4) FIGS. 3a, 3b and 3c together illustrate the process of firing a front side paste,

(5) FIG. 4 shows the positioning of cuts for the test method below to measure specific contact resistance.

(6) FIG. 1 shows a cross sectional view of a solar cell 100 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 104, a back doped layer 106, a p-n junction boundary 102, a front doped layer 105 and a front electrode 103, wherein the front electrode penetrates into the front doped layer 105 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 106 and the front doped layer 105 together constitute a single doped Si wafer 101. In the case that 100 represents an n-type cell, the back electrode 104 is preferably a silver electrode, the back doped layer 106 is preferably Si lightly doped with P, the front doped layer 105 is preferably Si heavily doped with B and the front electrode 103 is preferably a mixed silver and aluminium electrode. In the case that 100 represents a p-type cell, the back electrode 104 is preferably a mixed silver and aluminium electrode, the back doped layer 106 is preferably Si lightly doped with B, the front doped layer 105 is preferably Si heavily doped with P and the front electrode 103 is preferably a silver and aluminium electrode. The front electrode 103 has been represented in FIG. 1 as consisting of three bodies purely to illustrate schematically the fact that the front electrode 103 does not cover the front face in its entirety. The invention does not limit the front electrode 103 to those consisting of three bodies.

(7) FIG. 2 shows a cross sectional view of a common layer configuration for a solar cell 200 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 104, a back passivation layer 208, a highly doped back layer 210, a back doped layer 106, a p-n junction boundary 102, a front doped layer 105, a front passivation layer 207, an anti-reflection layer 209, front electrode fingers 214 and front electrode bus bars 215, wherein the front electrode fingers penetrate through the anti-reflection layer 209 and the front passivation layer 207 and into the front doped layer 105 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. In the case that 200 represents an n-type cell, the back electrode 104 is preferably a silver electrode, the highly doped back layer 210 is preferably Si heavily doped with P, the back doped layer 106 is preferably Si lightly doped with P, the front doped layer 105 is preferably Si heavily doped with B, the anti-reflection layer 209 is preferably a layer of silicon nitride and the front electrode fingers and bus bars 214 and 215 are preferably a mixture of silver and aluminium. In the case that 200 represents a p-type cell, the back electrode 104 is preferably a mixed silver and aluminium electrode, the highly doped back layer 210 is preferably Si heavily doped with B, the back doped layer 106 is preferably Si lightly doped with B, the front doped layer 105 is preferably Si heavily doped with P, the anti-reflection layer 209 is preferably a layer of silicon nitride and the front electrode fingers and bus bars 214 and 215 are preferably silver. FIG. 2 is schematic and the invention does not limit the number of front electrode fingers to three as shown. This cross sectional view is unable to effectively show the multitude of front electrode bus bars 215 arranged in parallel lines perpendicular to the front electrode fingers 214.

(8) FIGS. 3a, 3b and 3c together illustrate the process of firing a front side paste to yield a front side electrode. FIGS. 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 to without more detailed consideration.

(9) FIG. 3a illustrates a wafer before application of front 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 311, a back doped layer 106, a p-n junction boundary 102, a front doped layer 105 and additional layers on the front face 312. The additional layers on the back face 311 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 312 can comprise any of a front passivation layer, an anti-reflection layer or none of the above.

(10) FIG. 3b shows a wafer with electro-conductive paste applied to the front face before firing 300b. In addition to the layers present in 300a described above, an electro-conductive paste 313 is present on the surface of the front face.

(11) FIG. 3c shows a wafer with front electrode applied 300c. In addition to the layers present in 300a described above, a front side electrode 103 is present which penetrates from the surface of the front face through the additional front layers 312 and into the front doped layer 105 and is formed from the electro-conductive paste 313 of FIG. 3b by firing.

(12) In FIGS. 3b and 3c, the applied electro-conductive paste 313 and the front electrodes 103 are shown schematically as being present as three bodies. This is purely a schematic way of representing a non-complete coverage of the front face by the paste/electrodes and the invention does not limit the paste/electrodes to being present as three bodies.

(13) FIG. 4 shows the positioning of cuts 421 relative to finger lines 422 in the wafer 420 for the test method below to measure specific contact resistance.

(14) Test Methods

(15) 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) (temperature of 298.15 K and an absolute pressure of 100 kPa) apply.

(16) Viscosity

(17) 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.sup.−1 within 48 s and 50 equidistant measuring points and further increased to 150 s.sup.−1 within 312 s and 156 equidistant measuring points. After a waiting time of 60 s at a shear rate of 150 s.sup.−1, the shear rate was reduced from 150 s.sup.−1 to 20 s.sup.−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.sup.−1 of the downward shear ramp.

(18) Specific Surface Area

(19) BET measurements to determine the specific surface area of particles are made in accordance 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 Aluminum oxide CRM BAM-PM-102 available from BAM (Bundesanstalt für Materialforschung und-prüfung) 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.sub.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 sample, 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.sub.2 5.0 with a molecular cross-sectional area of 0.162 nm.sup.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.sup.2/g.

(20) Specific Contact Resistance

(21) In an air conditioned room with a temperature of 22±1° C., all equipment and materials are equilibrated before the measurement. For measuring the specific contact resistance of fired silver electrodes on the front doped layer of a silicon solar cell a “GP4-Test Pro” equipped with the “GP-4 Test 1.6.6 Pro” software package from the company GP solar GmbH is used. This device applies the 4 point measuring principle and estimates the specific contact resistance by the transfer length method (TLM). To measure the specific contact resistance, two 1 cm wide stripes of the wafer are cut perpendicular to the printed finger lines of the wafer as shown in FIG. 4. The exact width of each stripe is measured by a micrometer with a precision of 0.05 mm. The width of the fired silver fingers is measured on 3 different spots on the stripe with a digital microscope “VHX-600D” equipped with a wide-range zoom lens VH-Z100R from the company Keyence Corp. On each spot, the width is determined ten times by a 2-point measurement. The finger width value is the average of all 30 measurements. The finger width, the stripe width and the distance of the printed fingers to each other is used by the software package to calculate the specific contact resistance. The measuring current is set to 14 mA. A multi contact measuring head (part no. 04.01.0016) suitable to contact 6 neighboring finger lines is installed and brought into contact with 6 neighboring fingers. The measurement is performed on 5 spots equally distributed on each stripe. After starting the measurement, the software determines the value of the specific contact resistance (mOhm*cm.sup.2) for each spot on the stripes. The average of all ten spots is taken as the value for specific contact resistance.

(22) Ag Particles Size Determination (d.sub.10, d.sub.50, d.sub.90)

(23) Particle size determination for Ag 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 P11 (from Micromeritics, with a density of about 0.74 to 0.76 g/cm.sup.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 (density 10.5 g/cm.sup.3 for silver) the particle size distribution is determined and given as d.sub.50, d.sub.10 and d.sub.90.

(24) Dopant Level

(25) Dopant levels are measured using secondary ion mass spectroscopy.

(26) Efficiency, Fill Factor, Open Circuit Voltage, Contact Resistance and Series Resistance

(27) The sample solar cell is characterized using a commercial IV-tester “cetisPV-CTL1” from Halm Elektronik GmbH. All parts of the measurement equipment as well as the solar cell to be tested were maintained at 25° C. during electrical measurement. This temperature is always measured simultaneously on the cell surface during the actual measurement by a temperature probe. The Xe Arc lamp simulates the sunlight with a known AM1.5 intensity of 1000 W/m.sup.2 on the cell surface. To bring the simulator to this intensity, the lamp is flashed several times within a short period of time until it reaches a stable level monitored by the “PVCTControl 4.313.0” software of the IV-tester. The Halm IV tester uses a multi-point contact method to measure current (I) and voltage (V) to determine the cell's IV-curve. To do so, the solar cell is placed between the multi-point contact probes in such a way that the probe fingers are in contact with the bus bars of the cell. The numbers of contact probe lines are adjusted to the number of bus bars on the cell surface. All electrical values were determined directly from this curve automatically by the implemented software package. As a reference standard a calibrated solar cell from ISE Freiburg consisting of the same area dimensions, same wafer material and processed using the same front side layout is tested and the data compared to the certificated values. At least 5 wafers processed in the very same way are measured and the data interpreted by calculating the average of each value. The software PVCTControl 4.313.0 provides values to for efficiency, fill factor, short circuit current, series resistance and open circuit voltage.

(28) Temperature Profile in the Firing Furnace

(29) The temperature profile for the firing process was measured with a Datapaq DQ 1860 A datalogger from Datapaq Ltd., Cambridge, UK connected to a Wafer Test Assembly 1-T/C 156 mm SQ from Despatch (part no. DES-300038). The data logger is protected by a shielding box TB7250 from Datapaq Ltd., Cambridge, UK and connected to the thermocouple wires of the Wafer Test Assembly. The solar cell simulator was placed onto the belt of the firing furnace directly behind the last wafer so that the measured temperature profile of the firing process was measured accurately. The shielded data logger followed the Wafer Test assembly at a distance of about 50 cm to not affect the temperature profile stability. The data was recorded by data logger and subsequently analysed using a computer with Datapaq Insight Reflow Tracker V7.05 software from Datapaq Ltd., Cambridge, UK.

(30) Determination of Elemental Composition

(31) Digestion Procedure:

(32) 1. Acid Digestion Procedure US EPA 3050B:

(33) Mix the sample thoroughly to achieve homogeneity and sieve, if appropriate and necessary, using a USS #10 sieve. All equipment used for homogenization should be cleaned according to the guidance in Sec. 6.0 to minimize the potential of cross-contamination. For each digestion procedure, weigh to the nearest 0.01 g and transfer a 1-2 g sample (wet weight) or 1 g sample (dry weight) to a digestion vessel. For samples with high liquid content, a larger sample size may be used as long as digestion is completed.

(34) For the digestion of samples for analysis by ICP-AES, add 10 mL of 1:1 HNO.sub.3:HCl, mix the slurry, and cover with a watch glass or vapor recovery device. Heat the sample to 95° C.±5° C. and reflux for 10 to 15 minutes without boiling. Allow the sample to cool, add 5 mL of concentrated HNO.sub.3, replace the cover, and reflux for 30 minutes. If brown fumes are generated, indicating oxidation of the sample by HNO.sub.3, repeat this step (addition of 5 mL of conc. HNO.sub.3) over and over until no brown fumes are given off by the sample indicating the complete reaction with HNO.sub.3. Using a ribbed watch glass or vapor recovery system, either allows the solum tion to evaporate to approximately 5 mL without boiling or heat at 95° C.±5° C. without boiling for two hours. Maintain a covering of solution over the bottom of the vessel at all times.

(35) After the sample has cooled, add 2 mL of water and 3 mL of 30% H.sub.2O.sub.2. Cover the vessel with a watch glass or vapor recovery device and return the covered vessel to the heat source for warming and to start the peroxide reaction. Care must be taken to ensure that losses do not occur due to excessively vigorous effervescence. Heat until effervescence subsides and cool the vessel.

(36) Continue to add 30% H.sub.2O.sub.2 in 1-mL aliquots with warming until the effervescence is minimal or until the general sample appearance is unchanged.

(37) Cover the sample with a ribbed watch glass or vapor recovery device and continue heating the acid-peroxide digestate until the volume has been reduced to approximately 5 mL or heat at 95° C.±5° C. without boiling for two hours. Maintain a covering of solution over the bottom of the vessel at all times.

(38) For the analysis of samples for ICP-AES, add 10 mL conc. HCl to the sample digest and cover with a watch glass or vapor recovery device. Place the sample on/in the heating source and reflux at 95° C.±5° C. for 15 minutes.

(39) Filter the digestate through Whatman No. 41 filter paper (or equivalent) and collect filtrate in a 100-mL volumetric flask. Make to volume and analyze by ICP-AES.

(40) 2. Microwave digestion

(41) The digestion was performed on a commercial Microwave Digestion System using a highpressure quartz vessel (XQ80). The HF-free digestion made the quartz vessel applicable. In each digestion, an accurately weighed amount of 0.05 g coal sample was mixed with 3 ml of acid in a cleaned vessel. After the cap assembles were sealed, the four liners were stuck inside the vessel jackets and then mounted on the rotating tray in a symmetrical pattern to ensure a uniform irradiation. The heating profile was controlled by the predetermined power program. Typically, the temperature rose from room temperature to 200° C. in the first 30 min, and then to about 250° C. in the later 30 min with the autogeneous pressure of about 7.5 MPa.

(42) After the vessel was cooled down and depressurized, the solution was carefully transferred to a to Teflon beaker. Unless stated otherwise, the solution was evaporated in an evacuated bench to a small drop of thick liquid (˜0.1 g) at a temperature below 60° C. by heating with two infrared radiation lamps. This procedure typically took 5 h. The residue was dissolved with 1 M HNO.sub.3, and then passed through a PTFE filter with a pore size of 0.45 μm (Sterile Millex-HV Millipore), and finally diluted with 1 M HNO.sub.3 for the instrumental analysis. Since the decomposition degree of HNO.sub.3 upon the digestion was unknown, the evaporation procedure allowed the digestion solution to be easily adjusted on the background HNO.sub.3 concentration and consequently, the solution could be analyzed by an external calibration method. In all experiments, the polypropylene bottles used for storing the solutions were immersed in 1 M HNO.sub.3 overnight, and rinsed with ultrapure water, and then dried in a Class 100 bench.

(43) Equipment and the Operating Conditions for ICP OES:

(44) The ICP-OES analysis was performed on Optima 7300 DV ICP-OES. The operating conditions were as follows:

(45) TABLE-US-00001 Inductively Coupled Plasma with Optical Emission Spectrometer Model Optima 7300 DV ICP-OES Brand Perkin Elmer Conditions Sets RF Power 1.300 kW Plasma Flow 15.0 L/min Auxiliary Flow 0.2 L/min Nebulizer Flow 0.8 L/min Sample Introduction Sample Uptake 15 s Sample Flow 1.5 mL/min Pump Rate 15 rpm Rinse Time 15 s General Settings Replicates 3
ICP-AES Analysis Procedure:
Test Portion

(46) The test portion may be directly obtained from the test sample or may be diluted from the test sample to accommodate the measurement range or to dilute the matrix. The acidity of the test portion must match the acidity of calibration solutions. Ensure that all elements are present in a non-volatile form. Volatile species must be converted to non-volatile ones e.g. sulphide oxidation by hydrogen peroxide.

(47) Set Up of the Procedure

(48) Adjust the instrumental parameters of the ICP-AES system in accordance with the manufacturer's manual. Use the recommended optimization solution to optimize or check the sensitivity and the stability of the system. Check the wavelength calibration as often as required by the manufacturer. Select wavelengths for measurement and for background subtraction.

(49) Apply the calibration method as per standard procedure.

(50) Sample Measurement

(51) Run one or more calibration solutions and calibration blanks and determine the calibration function. Run the interference check solution(s) to establish interference correction or to check presence of interference. Run all samples including spiked samples if necessary for standard addition calibration Run a calibration blank and a calibration check solution, every 25 samples or less and at the beginning and end of the sample run. Run at least one spiked sample (original sample/digest or aqueous sample) to check recovery.

(52) Run at least one pre digestion blank. Run at least one pre digestion sample in duplicate to check inhomogeneity. Whenever a new of usual sample matrix is encountered check: matrix effects by running the spike sample or matrix effects by running a fivefold diluted sample and inter-element interference analysing at a different wavelength.

(53) Consider a change in calibration strategy.

(54) Calculation

(55) Calculate the element concentration in the aqueous sample:
ρ=(ρ.sub.1−ρ.sub.0)f.sub.df.sub.a

(56) Calculate the element concentration in the digested solid sample:
w=(ρ.sub.1−ρ.sub.0)f.sub.aV/m
where: ρ concentration of the element in the aqueous sample in mg/l; ρ.sub.1 concentration of the element in the test sample in mg/l;) ρ.sub.0 concentration of the element in the blank in mg/l; f.sub.d dilution factor due to digestion of an aqueous sample; in all other cases f.sub.d=1; f.sub.a dilution factor of the test portion; w mass fraction of the element in the solid sample in mg/kg; V volume of the test sample (digest) in 1; m mass of the digested sample in kg.

EXAMPLES

(57) 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.

(58) Paste Preparation

(59) A paste was made by mixing, by means of a Kenwood Major Titanium mixer, the appropriate amounts of organic vehicle (Table 1), Ag powder (PV 4 from Ames Inc. with a d.sub.50 of 2 μm), glass according to the specific invention (Table 2) ground to d.sub.50 of 2 μm, zinc oxide (Sigma Aldrich GmbH, article number 204951). The paste was passed through a 3-roll mill Exact 80 E with stainless steel rolls with a first gap of 120 μm and a second gap of 60 μm with progressively decreasing gaps to 20 μm for the first gap and 10 μm for the second gap several times until homogeneity. 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 16 to about 20 Pas. The wt. % s of the constituents of the paste are given in Table 3.

(60) Glass Synthesis:

(61) The raw materials for respective oxides for the synthesis of glasses were obtained from Sigma Aldrich. These include Lead Oxide, Bismuth (III) Oxide, Tungsten (VI) Oxide, Tellurium (IV) Dioxide, Zinc Oxide, Silver Nitrate, Molybdenum (VI) Oxide, Silicon Dioxide, Titanium (IV) Oxide, Boric Acid, Lithium Carbonate etc. and others as required for the glass composition. The desired proportions of the raw materials totalling 100 gm were thoroughly mixed in an automated pestle and mortar, Model RM200 by Retsch, Germany for 5 to 10 minutes using sintered aluminium oxide mortar and pestle. This raw materials mixture was then transferred to 250 ml high purity alumina crucible. The crucible containing the raw materials mixture was heated in an electric furnace HTK 70/16 from Thermconcept, Germany with MoSi.sub.2 heating elements at temperature ranging between 800 and 1200° C. After melting, the glass melt was soaked for 10 to 120 min Finally the glass melt was poured in a stainless steel bucket with 1 to 5 litre water. The excess water was decanted and the glass was dried at 80-90° C. to get coarse glass frits with size ranging from 0.1 mm to 10 mm.

(62) Glass Sizing

(63) The coarse glass frits were first pre-ground in an automated pestle and mortar, Model RM200 by Retsch, Germany for 5 to 10 minutes using sintered aluminium oxide mortar and pestle. This step yielded particles ranging from 100 micron to 2 mm. The pre-ground glass was then transferred in a 250 ml sintered aluminium oxide grinding jar. To this 8 mm size YTZ grinding media in a typical ratio 1:4 to 1:12 (Actual 1:6) (glass: grinding media) was added. Mixture was milled either dry or wet (in water or organic solvent such as alcohol). This jar was placed in planetary ball mill Model PM400 by Retsch, Germany and milled at speeds ranging 30-400 min.sup.−1 (Actual 100 min.sup.−1). The total milling time ranged from 1 hour to 6 hours. In case of wet milling, the slurry was transferred in a glass tray and the dried by solvent evaporation at 80 −90° C. The average particle size obtained by this method ranges from 0.3 micron to 10 micron.

(64) TABLE-US-00002 TABLE 1 Constituents of organic vehicle Organic Vehicle Component Proportion of component 2-(2-butoxyethoxy)ethanol) [solvent] 84 ethyl cellulose (DOW Ethocel 4) [binder] 6 Thixcin ® E [thixotropic agent] 10

(65) TABLE-US-00003 TABLE 2 Glass composition - elemental composition given in wt. % based on the total weight of the glass Example Mo O Te Pb Si P B 1 7 37 0 0 45 4 7 2 7 24 0 40 18 4 7 3 7 22 4 40 16 4 7 Comparative 0 39 0 0 50 4 7

(66) TABLE-US-00004 TABLE 3 Paste composition Paste Ag Organic constituent particles vehicle Glass ZnO Wt. % in paste 85 9 5 1
Solar Cell Preparation and Measurements

(67) Pastes were applied to mono-crystalline Cz-n-type Silicon wafers with a boron doped front face and phosphorous doped back face. The wafers had dimensions of 156×156 mm.sup.2 and a pseudo-square shape. The wafers had an anti-reflect/passivation layer of SiN.sub.x with a thickness of about 75 nm on both faces. The solar cells used were textured by alkaline etching. The example paste was screen-printed onto the p-doped face of the wafer using a semi-automatic screen printer X1 SL from Asys Group, EKRA Automatisierungssysteme set with the following screen parameters: 290 mesh, 20 μm wire thickness, 18 μm emulsion over mesh, 72 fingers, 60 μm finger opening, 3 bus bars, 1.5 mm bus bar width. A commercially available Ag paste, SOL9600A, available from Heraeus Precious Metals GmbH & Co. KG, was printed on the back n-doped face of the device using the same printer and the following screen parameters: 325 mesh, 30 μm wire thickness, 18 μm emulsion over mesh, 156 fingers, 80 μm finger opening, 3 bus bars, 1.5 mm bus bar width. The device with the printed patterns was dried for 10 minutes at 150° C. in an oven after printing each side. The substrates were then fired with the p-doped side up in a Centrotherm DO-FF 8600-300 oven for 1.5 min. For each example, firing was carried out with maximum firing temperature of 800° C. Solar cell efficiency was measured and is given in table 4.

(68) TABLE-US-00005 TABLE 4 electrical properties of solar cells. Example Cell efficiency 1 ∘ 2 + 3 ++ Comparative −− Results displayed as −−: very unfavourable, −: unfavourable, ∘: moderate, +: favourable, ++: very favourable

REFERENCE LIST

(69) 100 Solar cell 101 Doped Si wafer 102 p-n junction boundary 103 Front electrode 104 Back electrode 105 Front doped layer 106 Back doped layer 200 Solar cell 207 Front passivation layer 208 Back passivation layer 209 Anti-reflection layer 210 Highly doped back layer 300 Wafer 311 Additional layers on back face 312 Additional layers on front face 313 Electro-conductive paste 214 Front electrode fingers 215 Front electrode bus bars 420 Wafer 421 Cuts 422 Finger lines