[0001] This invention relates to a method to form a dielectric layer on a silicon surface with the purpose of passivating the newly formed Si/dielectric interface to increase the conversion efficiency of a photovoltaic device, for example a solar cell, made from the silicon wafer. It is particularly concerned with coated silicon wafers suitable for use in photovoltaic cells which convert energy from light impinging on the front face of the cell into electrical energy. (The front face of a photovoltaic cell is the major surface facing the light source and the opposite major surface is the back surface.)

[0002] Photovoltaic devices or solar cells are typically configured as a cooperating sandwich of p- and n-type semiconductors, wherein the n-type semiconductor material exhibits an excess of electrons, and the p-type semiconductor material exhibits an excess of holes. Such a structure, when appropriately located electrical contacts are included, forms a working photovoltaic cell. Sunlight incident on photovoltaic cells is absorbed in the semiconductor creating electron/hole pairs. Some solar cells have a p-type base (with a n- type emitter) and are referred to as p-type solar cells. Others have an n-type base (with a p- type emitter) and are referred to as n-type solar cells. By way of a natural internal electric field created by sandwiching p- and n-type semiconductors, electrons created in the p-type material flow to the n-type material where they are collected, resulting in a DC current flow between the opposite sides of the structure when the same is employed within an appropriate, closed electrical circuit.

[0003] Photovoltaic cells are widely used as solar cells for providing electricity from impinging sunlight. Significant cost reduction of silicon solar cells requires a high throughput, low cost, and reliable industrial process on thin silicon wafer substrates. The thickness of the silicon wafer processed in mass production of solar cells has progressively decreased and is now about 180μιη; it is expected to be about 100μιη by 2020. This imposes significant modifications to the architecture of the solar cell because of cell bowing and loss of conversion efficiency. Cell bowing may result from a mismatch of the coefficients of thermal expansion of materials used in the cell. Loss in conversion efficiency may result from the larger number of photo-generated minority carriers reaching the back of thinner cells during their lifetime and being subject to back surface recombination through interfacial defects.

[0004] Present industrial surface conditioning and back surface passivation processes do not meet the requirements for yield and performance on thin substrates. The currently dominating technology of Aluminum Back Surface Field (AI-BSF) cell architecture, has reached its limits, particularly because of excessive cell bowing with wafers below about 200μιη following the high temperature (800 °C+) co-firing step generally used in solar cell production. Alternatives are required, particularly for back surface passivation.

[0005] One alternative solution relies on the use of dielectric layers for the passivation of the back surface, at least one of the layers of the stack being hydrogen rich to be used as an hydrogen source for dangling bonds passivation.

[0006] A paper by M. Tucci et al in Thin Solid Films (2008), 516(20), pages 6939-6942 describes thermal annealing after the sequential deposition by plasma enhanced chemical vapor deposition (PECVD) of a stack of hydrogenated amorphous silicon and hydrogenated amorphous silicon nitride to ensure stable passivation.

[0007] WO-A-2007/055484 and WO-A-2008/07828 disclose an alternative stack made of a silicon oxy-nitride (SiOxNy) passivation layer and a silicon nitride anti-reflective layer deposited on the back of the cell for surface passivation and optical trapping. The

passivation layer is 10-50nm thick while the anti reflective layer is 50-1 OOnm thick.

[0008] WO-A-2006/1 10048 (US-A-2009/056800) discloses the deposition of a thin hydrogenated amorphous silicon or hydrogenated amorphous silicon carbide film, followed by the deposition of a thin hydrogenated silicon nitride film, preferably by PECVD (Plasma Enhanced Chemical vapor deposition) prior to a final anneal at high temperature in forming gas.

[0009] US-A-2010/0323503 describes depositing a thin (0.1 to 10nm) amorphous hydrogenated silicon layer on the surface to be passivated and converting it to Si0 ₂ by rapid thermal processing in an oxygen environment at between 750 °C and 1200°C for 5 seconds to 30 minutes.

[0010] WO-A-2006/097303 and US-A-2009/0301557 describe a method for the production of a photovoltaic device, for instance a solar cell, by depositing a dielectric layer on the rear surface of a semiconductor substrate and depositing a passivation layer comprising hydrogenated SiN on top of the dielectric layer and forming back contacts through the dielectric layer and the passivation layer.

[0011] US5372842 discloses a method to convert a hydrogen silsesquioxane film on a substrate into Si0 ₂ 2 by converting the resin into a pre-ceramic silicon oxide by heating in an inert atmosphere and converting the pre-ceramic silicon oxide into silicon oxide ceramic by heating at 400 °C in an atmosphere consisting of oxygen and oxygen mixed with an inert gas.

[0012] US6191 183 discloses a method to convert a hydrogen silsesquioxane film on conductive substrate by exposing it to high-energy radiation such as an electron beam.

[0013] US5336532 discloses a method to convert a hydrogen silsesquioxane film on a substrate into a silicon dioxide-containing ceramic by heating the pre-ceramic coating to a temperature between 40 ° to 400 °C in the presence of ozone to enhance the rate of ceramification.

[0014] US4753855 and US4756977 disclose methods to convert a hydrogen

silsesquioxane film on a substrate into Si0 ₂ by exposition at temperatures between 200 °C and 1000 °C in an inert atmosphere.

[0015] EP460868 discloses a method to convert a hydrogen silsesquioxane film on a substrate into Si0 ₂ by exposing the coating to an amine to catalyze the conversion of the coating at 20-400 °C for < 2 h to a ceramic coating.

[0016] US5059448 discloses a method to convert a hydrogen silsesquioxane film on a substrate into Si0 ₂ by rapidly heating the film for several seconds in an oxidizing

atmosphere to a temperature ranging from 400-900 °C, using high intensity radiation of various frequencies (UV, visible or IR). A platinum catalyst is present in the resin film.

[0017] A process according to the present invention for the production of a silicon wafer coated with a passivation layer of a silicon oxide comprises the steps of

(i) coating the wafer with a silicon-containing polymer having Si-H groups so that the layer of coating produced contains less than 5% carbon atoms (calculated as the proportion of carbon atoms in the layer of silicon-containing polymer coating to total atoms excluding hydrogen as measured by X-ray photoelectron spectroscopy)

(ii) thermally treating the layer of coating produced at a temperature above 350 °C for 5 to 120 seconds, during which treatment the silicon-containing polymer having Si-H groups is subject to a maximum temperature in the range 700 to 1020 °C, thereby converting the layer of silicon-containing polymer having Si-H groups into a silicon dioxide layer, and

(iii) hydrogenating the silicon dioxide layer thus produced.

[0018] The excellent passivation of the silicon dielectric interface when produced according to the present invention is demonstrated by the low surface recombination velocity (lower than 300cm/s) and the high minority carrier lifetime (longer than 50 \s).

[0019] The invention includes a photovoltaic device comprising a silicon wafer coated with a passivation layer of a hydrogenated silicon oxide by the process of the invention.

[0020] The proportion of carbon atoms to total atoms excluding hydrogen in the coating layer of silicon-containing polymer having Si-H groups is measured by X-ray photoelectron spectroscopy (XPS). X-ray photoelectron spectroscopy (XPS) involves irradiation of a sample with soft X-rays, and the energy analysis of photo-emitted electrons that are generated close to the sample surface. XPS has the ability to detect all elements (with the exception of hydrogen and helium) in a quantitative manner from an analysis depth of 10nm (or less). In addition to elemental information XPS can also be used to probe the chemical state of elements through the concept of binding energy shift. XPS analysis can be performed using an Axis Ultra spectrometer (Kratos Analytical).

[0021 ] The silicon wafer substrate which is coated is generally crystalline and can be mono-crystalline or multi-crystalline silicon. A mono-crystalline wafer can for example be a float-zone (FZ) silicon wafer, a Czochralski process (CZ) silicon wafer or a quasi-mono type silicon wafer. The silicon wafer can for example be 100 μιη to 400 μιη thick (provided in the specification from the wafer manufacturer). An example of a preferred silicon wafer is an FZ silicon wafer of bulk lifetime greater than 500 ts and resistivity 1 -5 Ω.οιη. The wafer may have both faces, one face or none of its faces chemically polished. The non-polished surfaces are preferably textured.

[0022] We have found that the silicon-containing polymer having Si-H groups is readily converted by thermal treatment into a dielectric silicon dioxide layer suitable for forming a passivated silicon dielectric interface. In one preferred process the silicon-containing polymer having Si-H groups comprises hydrogen silsesquioxane resin.

[0023] The hydrogen silsesquioxane resin is a silicone resin substantially of the empirical formula HSi0 _3/2 . It can be prepared by the hydrolysis of trichlorosilane HSiCI ₃ . The hydrogen silsesquioxane resin generally has a cage-like molecular structure. The hydrogen silsesquioxane resin can be used alone so that the silicon-containing polymer having Si-H groups contains no carbon, or can be mixed with a silicon-containing polymer having hydrocarbon groups bonded to silicon.

[0024] A silicon-containing polymer having hydrocarbon groups bonded to silicon with which the hydrogen silsesquioxane resin is mixed can for example be another

silsesquioxane resin such as methyl silsesquioxane resin of empirical formula CH ₃ Si0 ₃ /2. It may be preferred that the silicon-containing polymer having hydrocarbon groups also has Si- H groups, for example a silsesquioxane resin having Si-H groups and Si-CH ₃ groups or an organopolysiloxane containing RHSi0 _2/2 siloxane units where R represents a hydrocarbyl group, alternatively an alkyl group having 1 to 6 carbon atoms. The silicon-containing polymer having hydrocarbon groups can for example be a poly(methylhydrogensiloxane). The proportion of carbon atoms to total atoms excluding hydrogen as measured by XPS in any mixture of silicon-containing polymers applied in step (i) is less than 5%.

[0025] The silicon-containing polymer having Si-H groups can alternatively be a silsesquioxane resin having Si-H groups and hydrocarbon groups bonded to silicon. , for One example of a resin having Si-H groups and hydrocarbon groups bonded to silicon is a hydrogen methyl silsesquioxane resin comprising HSi0 _{3 2} units and CH ₃ Si0 _{3 2} units. Such a resin can be prepared by hydrolysis of a mixture of HSiCI ₃ and CH ₃ SiCI ₃ . Alternative examples of a resin having Si-H groups and hydrocarbon groups bonded to silicon are resins comprising HSi0 _3/2 units and (CH ₃ )HSi0 ₃₂ units and resins comprising HSi0 ₃₂ units, (CH ₃ )HSi0 _2/2 units and CH ₃ Si0 _{3 2} units. Such resins can be prepared by hydrolysis of a mixture of HSiCI _3, (CH ₃ )HSiCI ₂ and optionally CH ₃ SiCI ₃ . The proportion of carbon atoms to total atoms excluding hydrogen as measured by XPS in any resin or mixture of silicon- containing polymers applied in step (i) is less than 5%.

[0026] In step (i) of the process of the invention the wafer is preferably coated by applying a solution of the silicon-containing polymer having Si-H groups to the wafer. The solvent can then be evaporated from the coating of silicon-containing polymer solution on the wafer. The solution of silicon-containing polymer having Si-H groups can for example be a solution in a volatile siloxane solvent having a degree of polymerisation of less than 10 siloxane units per molecule such as hexamethyldisiloxane, octamethyltrisiloxane and/or

decamethyltetrasiloxane, particularly a solution of hydrogen silsesquioxane resin in such a siloxane solvent. The solution of silicon-containing polymer having Si-H groups can for example have a concentration of 1 to 50% by weight, alternatively 3 to 25%, for example a 25% solution of hydrogen silsesquioxane resin in a blend of hexamethyldisiloxane and octamethyltrisiloxane which may be used without further dilution or alternatively may be diluted with a volatile siloxane solvent such as octamethyltrisiloxane and/or

decamethyltetrasiloxane. The solution of silicon-containing polymer having Si-H groups, for example hydrogen silsesquioxane resin, can alternatively be a solution in an aliphatic ketone such as methyl isobutyl ketone, methyl ethyl ketone or methyl isoamyl ketone. A further alternative is a 14% solution of hydrogen silsesquioxane resin in methyl isobutyl ketone.

[0027] The solution of silicon-containing polymer having Si-H groups can for example be applied to the silicon wafer by spin coating, slot die coating, spray coating such as ultrasonic spray coating, dip coating, angle-dependent dip coating, flow coating, capillary coating, roll coating or tampon printing. Some of these methods are appropriate for the coating of a single side of the silicon wafer substrate at a time; others can be used for the simultaneous coating of both sides of the substrate. The most appropriate method may be selected depending upon the type of solar cell architecture required and the need for single-side or double-side coating.

[0028] The wafer is preferably coated in step (i) with an amount of silicon-containing polymer having Si-H groups such that the silicon dioxide layer produced in step (ii) has a thickness of 10nm to 500nm. This generally requires a dry film thickness of silicon-containing polymer in the range 20nm to 750nm. Unless otherwise indicated all film thicknesses are measured using a spectroscopic ellipsometer (e.g. as sold under the tradename UVsel from Jobin-Yvon)

[0029] The spin coating process consists in dispensing a defined volume of solution on a substrate that is, or will be, submitted to spinning. The silicon wafer substrate is placed on a chuck, made of aluminum or Teflon, in a spin coater such as that sold by Chemat

Technology as model KW-4A and held in place by vacuum suction. The solution of silicon- containing polymer having Si-H groups can be dispensed in static mode (the substrate is not spinning during the dispensing stage) or in dynamic mode (the substrate is subject to low speed spinning while dispensing the solution). The spinning process consists of first spinning the substrate at low speed (200-600rpm) for a short time (2-10s) and then spinning the substrate at high rate (1000-1 OOOOrpm) for a longer time (10s-60s) to spread the solution evenly over the wafer substrate. The thickness of the resulting coating will depend upon the solid content of the resin solution and the spinning rate during the second spinning step. Coatings of dry film thickness in the range 40 to 500nm are generally produced from hydrogen silsesquioxane resin solutions of concentration in the range 10 to 25% by weight. The spin coating process has the advantage of providing a very homogeneous coating in terms of thickness, with thickness variation typically in the <±1 % to ±6%, although it has the disadvantages of long duration time and low product usage.

[0030] Spray coating is a suitable process for coating the silicon wafer substrate. An example of a suitable spray nozzle is a Burgener ARI MIST HP Serial 14.547 nebulizer. The solution of silicon-containing polymer having Si-H groups can be deposited by scanning the wafer at with the nebulized spray. Spraying has the advantage of relatively high coating rate while producing a homogeneous coating (±8%). A relatively dilute solution is preferred for spraying, for example a 4-10% by weight solution of the hydrogen silsesquioxane resin.

[0031] Slot die coating is another suitable process for coating the silicon wafer substrate. In the slot die process, the coating is squeezed out by gravity or under pressure through a slot and onto the substrate. The slot-die coater is a pre-metered coating method in which a precision pump delivers the coating solution to the slot die so that all of the coating solution metered to the die is applied to the web. Slot die coating also has the advantages of extremely high thickness homogeneity (<±1 %) and a relatively high coating rate. A relatively dilute solution is preferred for slot die coating, for example a 1 -15% by weight solution of the hydrogen silsesquioxane resin.

[0032] The solvent can be evaporated from the coating of silicon-containing polymer having Si-H groups at ambient temperature or at elevated temperature. Solvent evaporation can be carried out directly while the coating is applied by using a heated chuck. In one preferred process according to the invention the wafer coated with silicon-containing polymer having Si-H groups is heated at a temperature of 50 °C to 350 °C to partially crosslink the silicon-containing polymer before the thermal treatment step (ii). For hydrogen

silsesquioxane resin the crosslinking, sometimes called 'curing' consists in the siloxane bond rearrangement from a cage-like structure of HSi0 _3/2 to a network structure of HSi0 _3/2 with no or little loss of the SiH groups. Heating can be carried out with a gradual or stepwise increase in temperature, for example 2 minutes each at 150°C, 200 °C and then 350 ^< €, or can be at a single temperature, for example 6 minutes at 150°C. The purpose of the partial crosslinking step is to build a three dimensional network with mechanical integrity and stability for subsequent processing. It is however not a pre-requisite for further processing.

[0033] In step (ii) the coating of silicon-containing polymer having Si-H groups is thermally treated to convert the layer of silicon-containing polymer into a silicon dioxide layer. In this thermal treatment, the silicon-containing polymer having Si-H groups coating is subjected to a temperature above 350 °C for 5 to 120 seconds. During this treatment the silicon- containing polymer having Si-H groups is subject to a maximum temperature in the range 700 °C to 1020°C. Preferably the hydrogen silsesquioxane resin is subject to a maximum temperature in the range 750 to 980 °C. This short time high temperature treatment can for example be achieved using an in-line furnace of the type used by the photovoltaic industry for the thermal contact annealing step of solar cell fabrication, for example in a SOLARIS 150 Rapid Thermal Processing system. In the SSI Rapid Thermal Process (RTP) , the wafer is laid on a carrier made of quartz; a thermocouple is put in direct contact with the wafer and allows high accuracy temperature measurement; wafer is heat-up by two set of IR lamps facing the top and the bottom of the wafer. The IR lamps are computer controlled and the temperature as a function of time is recorded in-line using the signal of the thermocouple.

[0034] The temperature of treatment can be measured by measuring the temperature of the silicon wafer (substrate) with a thermocouple in contact with the surface of the wafer remote from the heat source. Surprisingly we have found that a long thermal treatment is detrimental to the passivation ability of the silicon dioxide layer. The surface recombination velocity increases with longer exposures to high temperature. The coated wafer is preferably raised very rapidly to its maximum temperature, and also cooled rapidly. For example, if the maximum temperature to which the coated wafer is subjected is 850 °C, the coated wafer is preferably above 800 °C for only a very short time, for example 1 to 5 seconds.

[0035] The thermal treatment step (ii) is preferably carried out in an oxidative atmosphere, most preferably an oxygen-containing atmosphere. The atmosphere used for the thermal treatment of step (ii) can for example contain 10 to 100%, alternatively 10 to 50% oxygen. The oxygen is preferably mixed with an inert gas such as nitrogen. Conveniently the oxygen- containing atmosphere can be air.

[0036] The dielectric silicon dioxide layer thus produced is hydrogenated to achieve passivation of the silicon dielectric interface. Hydrogenation preferably takes place at a temperature of 350 °C to 1000°C.

[0037] In one preferred hydrogenation process, the silicon dioxide layer is heated in an atmosphere comprising hydrogen. The atmosphere preferably contains 2 to 20% by volume hydrogen in an inert gas such as nitrogen. This type of hydrogenation process is preferably carried out at a temperature in the range 350 °C to 500 °C, for example at about 400 °C. The time for which hydrogenation is carried out can for example be in the range 10 to 60 minutes or more.

[0038] In an alternative preferred process the silicon dioxide layer is coated with a layer of hydrogenated silicon nitride before heating at a temperature of 400 °C to l OOO i. This type of hydrogenation process is preferably carried out at a temperature in the range 700 °C to 1000°C, for example at 800 °C to 900 °C. Heating at this temperature results in the release of hydrogen from the nitride layer.

[0039] Hydrogenation can alternatively be carried out by exposure of the coated wafer to a hydrogen plasma. In this case hydrogenation needs not be carried out at high temperature. The plasma can be low temperature plasma at below 200 ^e C, and alternatively below 100 ^e C. A preferred type of plasma is a non- local thermal equilibrium atmospheric pressure plasma discharge, including dielectric barrier discharge and particularly a diffuse dielectric barrier discharge such as a glow discharge plasma. A suitable apparatus for generating a non- local thermal equilibrium atmospheric pressure plasma discharge is described in WO- A-2012/010299. This apparatus can be operated with hydrogen as process gas to produce a hydrogen plasma.

[0040] Hydrogenation by coating with a layer of hydrogenated silicon nitride before heating at a temperature of 400 °C to 1000°C has the advantage that the firing step required to form back contacts in a photovoltaic device such as a solar cell can also act as the heating step to effect hydrogenation. The formation of back contacts through silicon nitride and silicon dioxide layers is a known process described for example in US-A-2009/0301557. Contacts are formed by forming holes in the dielectric silicon dioxide layer and silicon nitride layer and depositing a layer of contacting material, thereby filling the holes. The holes may be formed by laser ablation, by applying an etching paste, or by mechanical scribing. The layer of contacting material, for example a metal such as aluminium, can be deposited by evaporation, sputtering, screen printing, inkjet printing, or stencil printing. It can be deposited locally essentially in the holes or as a continuous or discontinuous layer. After the contacting material has been applied, the photovoltaic cell is subjected to a further firing step, for example in the range 600 to 1000°C for 5 to 60 seconds. Even if hydrogenation has been carried out by heating in an atmosphere comprising hydrogen or by exposure of the coated wafer to a hydrogen plasma, the step of forming back contacts will generally be required when forming a solar cell.

[0041] We have found that photovoltaic cells, particularly solar cells, comprising the passivation layer of silicon oxide dielectric material formed by the process of the invention show improved passivation. Passivation can for example be measured by calculating the minority carrier lifetime using a μ-PCD (microwave detected photoconductive decay) device. The minority carrier lifetime is measured after hydrogenation without formation of back contacts. Increased minority carrier lifetime shows improved passivation. A suitable μ-PCD device is for example supplied by SemiLab under the trade mark WT-2000. In the μ-PCD technique, a microwave antenna is placed near the surface of the wafer to direct microwaves at its surface. Some of the microwave signal will enter the semiconductor and another portion will be reflected, depending upon the conductivity of the sample. The wafer conductivity will be modified by means of a laser light pulse that will affect the concentration of the minority carrier and will modify the portion of the reflected microwaves. The time evolution of this portion will be followed as the excess minority carrier decreases and the wafer conductivity comes back to its equilibrium value. The exponential signal decays as a function of the lifetime of the minority carriers and allows determination of the surface recombination rate. A minority carrier lifetime longer then δθμβ indicates effective passivation.

[0042] Another measure of surface passivation is the effective surface recombination velocity S _eff .

[0043] In a doped silicon wafer, used to make a solar cell, some carriers are in excess compared to others. For example in a p-type Si wafer the dopant (typically boron) adds "holes" to the electrical charges resulting in electrons being the minority carriers. However, in the case of a n-type wafer, the dopant e.g. phosphorus adds electrons to the electrical charges and hence the "holes" are the minority carriers.

[0044] Upon light absorption, electrons and holes are produced and due to the fact that either the electrons (p-type silicon) or the holes (n-type silicon) are the minority carriers, the rate of their recombination dictates the likelihood that they will be collected at an electrode, i.e. generate a current in an external circuit. In a "pure" wafer, these minority carriers recombine mostly at the surface of the wafer with majority carriers, due to the "catalytic" presence of defects at these surfaces. Hence, the rate of disappearance is directly linked to the surface recomb

where

D _n = charge carrier diffusion coefficient

T _ef f = effective minority carrier lifetime

T _bu ik = bulk minority carrier lifetime

W = wafer thickness

However, in the case of effective passivation, the surface recombination velocity is low and the equation can be simplified to

The determination of the S _eff requires the knowledge of the T _bU | _k . In experiments described in the present application, high quality silicon float-zone (FZ) wafers have been used with known bulk lifetime in excess of >500μ8 and a T _bu ik value of 2000μ8 (p-type FZ, double side mirror polished, resistivity 1 -5Ω.οιη). The effective surface recombination velocity, S _eff , should be as low as possible. It is considered that achieving a surface recombination velocity lower or equal to 300cm/s leads to very good passivation of the Si/dielectric interface.

[0045] The invention is illustrated by the following Examples, in which parts and percentages are by weight.

Example 1

[0046] A 25% solution of hydrogen silsesquioxane resin in a blend of hexamethyldisiloxane and octamethyltrisiloxane was diluted with octamethyltrisiloxane to a resin concentration of 17%. 4" round FZ wafers were coated with the diluted resin solution in a Chemat Technology 'KW-4A' spin coater. 0.5ml of the resin solution was statically dispensed on the wafer substrate, and the substrate was spun for 6s at 300rpm and then for 20s at 2000rpm.

[0047] Solvent was eliminated from the hydrogen silsesquioxane resin solution, and the hydrogen silsesquioxane resin was cured, by heating at 150°C for 120s, then 200 °C for 120s, then 350 °C for 120s.

[0048] The resulting coated wafer was heated in air in a SOLARIS 150 Rapid Thermal Processing system to a target maximum temperature of 800 °C. The measured maximum temperature was 805 °C and the time at 800 °C or above was 2 seconds. The total time at above 350 °C was 20 seconds. The hydrogen silsesquioxane resin coating was converted to a dielectric silicon dioxide layer 250nm thick.

[0049] The dielectric silicon dioxide layer was hydrogenated by exposing the coated wafers to 5% H ₂ in N ₂ gas for 30 minutes at 400 °C.

[0050] The minority carrier lifetime measured using a SemiLab WT-2000 μ-PCD device was 1 1 Ομβ. The effective surface recombination velocity S _eff was calculated as described above and was found to be 135cm/s.

Examples 2 to 5 and Comparative Examples C1 to C4

[0051 ] Example 1 was repeated with the maximum temperature in the SOLARIS 150 Rapid Thermal Processing system being varied as shown in Table 1 . The minority carrier lifetime measured and the effective surface recombination velocity is shown in Table 1 .

Table 1

Example 6

[0052] FZ wafers were coated with hydrogen silsesquioxane resin solution and cured as described in Example 1 and were heated in air in a SOLARIS 150 Rapid Thermal

Processing system to a maximum temperature of 840 °C.

[0053] The dielectric layer of silicon dioxide formed was coated with a layer of

hydrogenated silicon nitride (SiNx:H), known for use as an antireflective coating in solar cells. Hydrogenation of the Si/Si0 ₂ interface was achieved by "firing" of the layered product at 850 °C. The minority carrier lifetime was measured as Ι δθμβ. The effective surface recombination velocity S _eff was calculated as 100cm/s. Examples 7 to 10

[0054] Ά 25% solution of hydrogen silsesquioxane resin in a blend of

hexamethyldisiloxane and octamethyltrisiloxane was diluted with octamethyltrisiloxane to a resin concentration of 2%. 4" round FZ wafers were coated with the diluted resin solution and the hydrogen silsesquioxane resin was cured as described in Example 1 .

[0055] The resulting coated wafers were heated in air in a SOLARIS 150 Rapid Thermal Processing system at various maximum temperatures as shown in Table 2. The total time at above 350 °C in each Example was about 20 seconds. The hydrogen silsesquioxane resin coating was converted to a dielectric silicon dioxide layer 25nm thick.

[0056] The dielectric silicon dioxide layer was hydrogenated as described in Example 1 . For each Example the minority carrier lifetime was measured and the effective surface recombination velocity S _eff calculated. The results are shown in Table 2.

Table 2

[0057] Examples 7 to 10 show that excellent passivation can be achieved using the process of the invention even at very low dielectric silicon dioxide layer thicknesses.

[00588] The invention may be any one of the following numbered aspects:

1. A process for the production of a silicon wafer coated with a passivation layer of a silicon oxide, comprising the steps of coating the wafer with a silicon-containing polymer having Si-H groups so that the layer of coating produced contains less than 5% carbon atoms (calculated as the proportion of carbon atoms in the layer of silicon-containing polymer coating to total atoms excluding hydrogen as measured by X-ray photoelectron spectroscopy) thermally treating the layer of coating at a temperature above 350 ¾ for 5 to 120 seconds, during which treatment the silicon- containing polymer having Si-H groups is subject to a maximum temperature in the range 700 to 1020°C, thereby converting the layer of silicon-containing polymer having Si-H groups into a silicon dioxide layer, and hydrogenating the silicon dioxide layer thus produced. 2. A process according to aspect 1 , wherein the silicon-containing polymer having Si-H groups comprises hydrogen silsesquioxane resin.

3. A process according to aspect 2 wherein the silicon-containing polymer having Si-H groups contains no carbon.

4. A process according to aspect 2 wherein the hydrogen silsesquioxane resin is mixed with a silicon-containing polymer having hydrocarbon groups bonded to silicon.

5. A process according to aspect 4 wherein the silicon-containing polymer having Si-H groups and hydrocarbon groups bonded to silicon is a poly(methylhydrogensiloxane).

6. A process according to aspect 1 , wherein the silicon-containing polymer having Si-H groups is a silsesquioxane resin having Si-H groups and hydrocarbon groups bonded to silicon.

7. A process according to any of aspects 1 to 6, wherein in step (i) the wafer is coated by applying a solution of the silicon-containing polymer having Si-H groups to the wafer and the solvent is evaporated from the coating of silicon-containing polymer. 8. A process according to any of aspects 1 to 7, wherein the wafer coated with silicon- containing polymer having Si-H groups is heated at a temperature of 50 °C to 350 ¾ to partially crosslink the silicon-containing polymer before the thermal treatment step (ii).

9. A process according to aspect 7 or aspect 8 wherein the solution is a solution of a hydrogen silsesquioxane resin in a volatile siloxane solvent having a degree of polymerisation of less than 10 siloxane units per molecule.

10. A process according to aspect 7 or aspect 8 wherein the solution is a solution of a hydrogen silsesquioxane resin in an aliphatic ketone.

1 1 . A process according to any of aspects 7 to 10 wherein the solution of the silicon- containing polymer having Si-H groups is applied to the wafer by spin coating or slot die coating.

12. A process according to any of aspects 1 to 1 1 wherein the thermal treatment step (ii) is carried out in an oxidative atmosphere.

13. A process according to any of aspects 1 to 12 wherein in step (ii) the hydrogen silsesquioxane resin is subject to a maximum temperature in the range 750 to 980 ^< €.

14. A process according to any of aspects 1 to 13 wherein the wafer is coated in step (i) with an amount of silicon-containing polymer having Si-H groups such that the silicon dioxide layer produced in step (ii) has a thickness of 10nm to 500nm.

15. A process according to any of aspects 1 to 14 wherein in step (iii) the silicon dioxide layer is hydrogenated at a temperature of 350 °C to 1000°C. A process according to aspect 15 wherein the silicon dioxide layer is heated in an atmosphere comprising hydrogen.

A process according to aspect 16 wherein the silicon dioxide layer is heated in a mixture of hydrogen and an inert gas, said mixture comprising 2 to 20% by volume hydrogen.

A process according to aspect 15 wherein the silicon dioxide layer is coated with a layer of hydrogenated silicon nitride before heating at a temperature of 400 °C to 1000 °C.

A process according to any of aspects 1 to 14 wherein in step (iii) the silicon dioxide layer is exposed to a non-local thermal equilibrium atmospheric pressure plasma formed from a gas comprising hydrogen.

A photovoltaic device comprising a silicon wafer coated with a passivation layer of a hydrogenated silicon oxide by the process of any of aspects 1 to 19.

A photovoltaic device according to aspect 20 wherein both faces of the silicon wafer are coated with the passivation layer of a hydrogenated silicon oxide by the process of any of aspects 1 to 19.