METHOD FOR FORMING DIAMOND PRODUCT

Abstract

A method for forming a diamond product. Diamond material is provided and a damage layer comprising sp.sup.2 bonded carbon is formed in the material. The presence of the damage layer defines a first diamond layer above and in contact with the damage layer and a second diamond layer below and in contact with the damage layer. The damage layer is electrochemically etched to separate it from the first layer, wherein the electrochemical etching is performed in a solution containing ions, the solution having an electrical conductivity of at least 500 μS cm.sup.−1, and wherein the ions are capable of forming radicals during electrolysis. The diamond product is also described.

Claims

1. A method for forming a diamond product comprising: providing a diamond material; forming a damage layer comprising sp.sup.2 bonded carbon, wherein the presence of the damage layer defines a first diamond layer above and in contact with the damage layer and a second diamond layer below and in contact with the damage layer; and electrochemically etching the damage layer to separate therefrom the first layer, wherein the electrochemical etching is performed in a solution containing ions, the solution having an electrical conductivity of at least 500 μS cm.sup.−1, and wherein the ions are capable of forming radicals during electrolysis.

2. The method according claim 1, wherein the ions are selected from any of persulfate ions, sulfate ions, oxalate ions, chloride ions, carbonate ions and metal ions.

3. The method according to claim 1, wherein the etching occurs at an etch rate selected from any of at least 2 mm.sup.2 hr.sup.−1, at least 5 mm.sup.2 hr.sup.−1, at least 10 mm.sup.2 hr.sup.−1, 20 mm2hr−1, 30 mm2hr.sup.−1, and at least 40 mm.sup.2hr.sup.−1.

4. The method according to claim 1, wherein etch rate is measured by aligning the diamond material comprising the damage layer with electrodes in an electrochemical cell to allow imaging of the diamond material and monitoring the material loss of the diamond material over time.

5. The method according to claim 1, wherein the electrochemical etching is performed in an electrochemical cell comprising at least an anode, a cathode and an electrolyte, and the ions are present in the electrolyte.

6. The method according to claim 1, wherein the ions are present during the electrochemical etching at a concentration of between 0.01 M and an upper limit of the solubility of the ions in the electrolyte.

7. The method according to claim 1, wherein the electrochemical etching is performed at a temperature selected from any of at least 50° C., at least 70° C. and at least 90° C.

8. The method according to claim 1, wherein the first and second diamond layers comprise diamond formed by chemical vapour deposition (CVD).

9. The method according to claim 1, wherein the diamond material is single crystal diamond material.

10. The method according to claim 9, wherein the single crystal diamond material has a largest linear dimension selected from any of at least 6 mm, at least 8 mm, at least 10 mm, at least 15 mm and at least 20 mm.

11. The method according to claim 1, wherein the diamond material is polycrystalline diamond material.

12. The method according to claim 1, wherein the diamond material comprises diamond produced by any of a chemical vapour deposition, CVD, process and a high pressure high temperature, HPHT, process.

13. The method according to claim 1, wherein the diamond material is doped.

14. The method according to claim 13, wherein the diamond material is doped with any of nitrogen, boron, phosphorus, and silicon.

15. The method according to claim 1, wherein the damage layer comprises graphitic material.

16. The method according to claim 1, further comprising forming the damage layer by an ion implantation process.

17. The method according to claim 1, further comprising attaching a support structure to a surface of the diamond material prior to the step of electrochemically etching the damage layer, wherein the support structure and the means of attaching the support structure are formed from materials that are substantially inert to a solution in which the electrochemical etching occurs.

18. A diamond product obtained by the method according to claim 1.

19. The diamond product according to claim 18, wherein the diamond product comprises any of single crystal diamond material, polycrystalline diamond material, boron-doped single crystal diamond material and boron-doped polycrystalline diamond material.

20. Use of the diamond product according to claim 18 in any of an optical product, an electronic product and an electrochemical product.

21. A diamond product comprising polycrystalline diamond material, wherein the diamond product has an average thickness of no greater than 30 μm and the polycrystalline diamond material has an average lateral grain size as measured by scanning electron microscope imaging of no more than 5 μm.

22. The diamond product according to claim 21, wherein the average lateral grain size is selected from any of greater than 10 μm, greater than 15 μm, and greater than 20 μm.

23. A composite body comprising a single crystal diamond plate having an average thickness of no more than 50 μm, and a support structure attached to a surface of the single crystal diamond plate by a glue that is substantially inert to ions capable of forming radicals during electrolysis.

24. The composite body according to claim 23, wherein the glue comprises an epoxy.

25. The composite body according to claim 23, wherein the single crystal diamond plate has an average thickness selected from any of no more than 25 μm, no more than 10 μm, no more than 5 μm and no more than 1 μm.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0062] FIG. 1 is a visual representation of the electrochemical etch process of an HPHT single crystal diamond material from t=0 to t=120 minutes. Each image is separated by approximately 8 minutes of etching in 0.3% v/v (0.05 M) H.sub.2SO.sub.4;

[0063] FIG. 2 is an exemplary etch profile from in-situ transmission microscope imaging in a 4.1 mm CVD sample in 0.25 M solution of K.sub.2SO.sub.4;

[0064] FIG. 3(a) is a schematic of an electrochemical cell used in step (b) showing the arrangement of the diamond material and the electrodes;

[0065] FIG. 3(b) is a photograph of the apparatus used in step (b) showing the microscope camera, flow system and backlight;

[0066] FIG. 3(c) is an image captured from the microscope camera during etching;

[0067] FIG. 4 is a comparison of bulk etch rates of different diamond materials in 0.25M solutions of supporting electrolytes as described in Example 1;

[0068] FIG. 5 is a plot of bulk etch rate versus sulfate concentration for a 3.5 mm CVD single crystal diamond material as described in Example 3;

[0069] FIG. 6 is a graph showing the bulk etch rates for a 4.1 mm CVD single crystal diamond material in solutions comprised of K.sub.2SO.sub.4 and H.sub.2SO.sub.4, with constant sulfate concentrate but decreasing pH;

[0070] FIG. 7 shows microscope images of the etching of boron doped polycrystalline diamond. Each image is separated by 25 minutes of etching from t=0, as described in Example 4;

[0071] FIG. 8 is microscope images of the etching of optical grade polycrystalline diamond. Each image is separated by 25 minutes of etching from t=0, as described in Example 5;

[0072] FIG. 9 shows Electron Paramagnetic Resonance (EPR) data for various salt solutions after electrolysis;

[0073] FIG. 10 is a flow diagram showing exemplary steps for forming a diamond product; and

[0074] FIG. 11 illustrates schematically exemplary steps for forming the diamond product.

DETAILED DESCRIPTION

Materials and solutions

[0075] All solutions were prepared from Milli-Q water (Millipore Corp.) with a resistivity of 18.2 MΩ cm at 25° C.

[0076] The electrolytes used comprised 0.25M salt in DI water. The salts tested were Potassium Nitrate (KNO.sub.3, 99.97%, Sigma Aldrich, UK), Potassium Chloride (KCI, ≥99%, Sigma Aldrich, UK), Potassium Sulphate (K.sub.2SO.sub.4, Pure, Acros Organics, US), Potassium Carbonate (K.sub.2CO.sub.3, ACS reagent, Sigma Aldrich, UK), and Potassium Oxalate (K.sub.2C.sub.2O.sub.4, >98.5%, Sigma Aldrich, UK).

[0077] Sulphuric Acid (H.sub.2SO.sub.4, >96%, Merck, UK) was used at 0.3% v/v (0.05 M) in DI water.

Diamond Sample Preparation

[0078] CVD and HPHT single crystal diamond plates (Element Six Technologies Ltd., Harwell, UK) having a thickness of approximately 500 μm were used as the diamond material for all studies, unless otherwise stated. The front face of each sample was mechanically polished to ˜nm roughness, and the rear face lapped to ˜μm roughness. The polished face was then implanted (Ion Beam Centre, University of Surrey, UK) with 2×10.sup.16 carbon atoms (2 MeV) per square centimetre, to produce an end of range damage layer ˜500 nm below the surface, as calculated by SRIM Simulations.

[0079] Implanted samples were then annealed at 1300° C. for two hours, to convert the damage layer generated by ion implantation to a sp.sup.2 bonded carbon layer.

[0080] The data shown in FIG. 2 was collected using a 3.5×3.5 mm square CVD plate.

[0081] The data shown in FIG. 4 was collected using 3.5 and 4.1 mm square CVD plates, as well as 4.2 mm square HPHT plates.

[0082] The data shown in FIG. 5 was collected using 3.5 mm square CVD plates.

[0083] The data shown in FIG. 6 was collected using 4.1 mm square CVD plates.

[0084] The data shown in FIG. 8 was collected using a 5 mm square BDD CVD plate.

Implant Characterisation

[0085] A Stopping and Range of Ions in Matter (SRIM) simulation was used to calculate the approximate depth and width of the end-of-range damage layer from ion implantation, giving each sample a damage region with a thickness of around 1 μm.

Acid Cleaning

[0086] Acid cleaning of diamond samples was carried out in a mixture of concentrated H.sub.2SO.sub.4 (Merck, >96%) saturated with KNO.sub.3 and heated (200° C.) for 30 minutes. This was followed by another 30 minutes in H.sub.2SO.sub.4 at 200° C. and then washing in ultrapure water (>18 MΩ cm, Millipore corp.).

Electrochemical Cell

[0087] The electrochemical etching step (b) took place in a custom designed 3D printed cell (Lulzbot Taz 6). This cell was designed to allow positive alignment of the diamond material comprising a damage layer and electrodes and to provide a light path for transmission imaging of samples undergoing etching (FIG. 3(a) and (b)).

[0088] A digital USB microscope (Veho Discovery VMS-001, 20-90× Magnification) was used to capture time lapse images of etch progress (FIG. 3(c)). Unless otherwise stated the electrodes used were 2 cm of 0.75 mm diameter platinum wire (for a total area of 0.5 cm.sup.2), with a spacing of 6 mm. A variable DC power supply (Elektro-Automatik GmbH, EA-PS 9750-04) was used to apply a potential between the two electrodes, operating in a potential limiting mode at 30 V, with a current dependent on the solution composition. To maintain constant solution composition and temperature, a flow system was used with a temperature-controlled reservoir of between 5-90° C. Electrolyte was circulated from this reservoir into the cell and returned via an outflow.

Image Analysis

[0089] A MATLAB script was used to analyse all time lapse data. Each frame from the video capture was extracted. The region of interest is defined on the first image and the script then calculates what proportion of that area corresponds to brighter etched regions, which match RGB values in the script. This proportion is then converted to an area, which is then plotted versus the etch time, to create etch profiles. Etch rates are compared in Table 1.

TABLE-US-00001 TABLE 1 Etch Rate Comparison Solution Substrate Etch Rate (mm.sup.2 hr.sup.−1) ½ Saturated Boric Acid 3.5 mm square HPHT 0.8 ± 0.04 (n = 3) FIG. 4  0.25M KNO.sub.3 3.5 mm square CVD 0.36  0.25M KNO.sub.3 4.1 mm square CVD 0.40  0.25M KNO.sub.3 4.2 mm square HPHT 0.34  0.25M K.sub.2CO.sub.3 4.2 mm square HPHT 7.23  0.25M KCl 3.5 mm square CVD 5.22  0.25M KCl 4.1 mm square CVD 4.84  0.25M KCl 4.2 mm square HPHT 7.64  0.25M K.sub.2C.sub.2O.sub.4 4.2 mm square HPHT 9.10  0.25K.sub.2SO.sub.4 3.5 mm square CVD 10.26  0.25K.sub.2SO.sub.4 4.1 mm square CVD 9.84  0.25K.sub.2SO.sub.4 4.2 mm square HPHT 12.62 FIG. 5  0.25M KNO.sub.3 3.5 mm square CVD 0.36  0.03M K.sub.2SO.sub.4 3.5 mm square CVD 4.92  0.06M K.sub.2SO.sub.4 3.5 mm square CVD 10.22  0.125M K.sub.2SO.sub.4 3.5 mm square CVD 10.36  0.25M K.sub.2SO.sub.4 3.5 mm square CVD 10.26 FIG. 6  0.25M K.sub.2SO.sub.4 4.1 mm square CVD 9.84  0.20M K.sub.2SO.sub.4 and 4.1 mm square CVD 17.82  0.05M H.sub.2SO.sub.4  0.25 M K2SO4 and 4.1 mm square CVD 16.04  0.25M H2SO4  0.25M H.sub.2SO.sub.4 4.1 mm square CVD 20.96 Temperature (0.05M H.sub.2SO.sub.4) 10 3.5 mm square HPHT 4.89 ± 0.79 (n = 3) 30 3.5 mm square HPHT 9.23 ± 2.8 (n = 3)  50 3.5 mm square HPHT 14.97 ± 1.2 (n = 3)  70 3.5 mm square HPHT 42.23 ± 6.9 (n = 3) 

Electron Paramagnetic Resonance (EPR)

[0090] Solutions for EPR comprised 0.25 M salt in DI water, including potassium nitrate, potassium chloride, and potassium sulphate (K.sub.2SO.sub.4, Analysis Grade, Sigma Aldrich, UK). Sulphuric acid (H.sub.2SO.sub.4, 96% Ultrapur, Merck, UK) was used at 0.3% v/v.

[0091] Electrolysis of a solution containing 0.05 M H.sub.2SO.sub.4, 0.25 M K.sub.2SO.sub.4, 0.25 M KCl or 0.25 M KNO.sub.3 was performed in a one-compartment cell. 30 V was applied between two platinum electrodes (electrode area, 0.5 cm.sup.2, with the same geometry as those used in the etch cell) placed in solution using a power supply (Elektro-Automatik GmbH, EA-PS 9750-04). After a period of 15 minutes, approximately 20 mg of 5,5-dimethyl-1-pyrroline-N-oxide (DMPO; Enzo life sciences, USA) was added to the mixture and mixed thoroughly. From the resultant solution, an aliquot of the electrolysis mixture was sampled and the EPR spectrum recorded.

[0092] The EPR signals for the DMPO spin adducts were recorded on a X-band spectrometer (EMX; Bruker, Germany) fitted with a HS cylindrical resonator (4119HS/0207; Bruker, Germany) at 298 K. Measurements were performed in a quartz EPR tube with a 1 mm inner diameter (Wilmad® quartz (CFQ) EPR tubes; Sigma-Aldrich, UK). For all measurements, the following spectrometer parameters were used: a non-saturating microwave power of 20 mW; central magnetic field, 3520 G; scan width, 100 G and a modulation amplitude, 0.5 G. All spectra reported are an average of 9 scans. Fitting and simulation of the EPR spectra was performed using the MATLAB package EasySpin (Version 5.2.25, Stoll, S.; Schweiger, A. EasySpin, a Comprehensive Software Package for Spectral Simulation and Analysis in EPR. J. Magn. Reson. 2006, 178 (1), 42-55. https://doi.org/10.1016/j.jmr.2005.08.013.).

EXAMPLE 1

[0093] The effect of different anions on the bulk etch rate of diamond material containing a damage layer of non-diamond carbon

[0094] The electrolytes described above, each having a fixed concentration of 0.25 M of the respective anion, were used as the electrolyte in an electrochemical etching process performed on three different diamond materials, specifically a 3.5 mm square CVD diamond material, a 4.1 mm CVD diamond material and a 3.5 mm HPHT diamond material, all of which had been processed as described above to include a damage layer of sp.sup.2 bonded carbon. The bulk etch rates observed are shown in FIG. 4.

[0095] With the exception of KNO.sub.3 all the electrolytes offered improved etch rates over published processes. The fastest etching electrolyte was K.sub.2SO.sub.4, which had a bulk etch rate over an order of magnitude faster than published processes. This significant increase in bulk etch rate reduces processing times, offering significant savings in process time and cost.

EXAMPLE 2

[0096] The effect of sulfate ion concentration on bulk etch rates of diamond material containing a damage layer of sp.sup.2 bonded carbon.

[0097] Having established that advantageous bulk etch rates were observed when using a electrolyte which comprises sulfate ions, Example 2 was carried out to investigate the effect of sulfate ion concentration on bulk etch rate.

[0098] The bulk etch rate in solutions of K.sub.2SO.sub.4 wherein the concentration of sulfate (SO.sub.4.sup.2−) varied in the range from 0.03 to 0.25 M was measured. The data for 0.25 M KNO.sub.3 was included for comparison (0 M sulfate ions) as this had the slowest bulk etch rate of those measured in Example 1. The conductivity of the 0.25 M KNO.sub.3 solution was 25 mS cm.sup.−1, with the conductivity of the sulfate solutions carrying from 4 to 40 mS cm.sup.−1 across the range. The results are presented in FIG. 5.

[0099] It is clear from FIG. 5 that at sulfate ion concentrations up to 0.06 M, there is a linear increase in the bulk etch rate with increasing sulfate concentration. The bulk etch rate then plateaus above sulfate ion concentrations of greater than 0.06 M but at a bulk etch rate which is significantly faster than has previously been achieved. The initial linear relationship is likely as a result of an increased sulfate ion concentration at the electrode surface, which ultimately results in a higher number of active etchant species being produced, thus supporting the hypothesis in Example 1 that sulfate ions are involved directly in the generation of the active etchant species.

[0100] The plateau in the bulk etch rate can be explained by one or a combination of two factors. First, the concentration of active etchant species at the damage layer interface increases such that the active sites within the damage layer become the rate-limiting step for etching. Second, at higher active etchant species concentrations, the chances of these species combining increases such that the number of active etchant species at the damage layer interface does not actually increase above 0.06 M.

[0101] As no additional electrolyte was added as the concentration of the K.sub.2SO.sub.4 solution was increased, the conductivity of the electrolyte also increases. However, in the region between 60 to 250 mM sulfate ions, there is no noticeable change in rate despite the conductivity increasing from 4 to 42 mS cm.sup.−1. This suggests that conductivity has little effect on bulk etch rate in this regime.

EXAMPLE 3

[0102] The effect of pH on bulk etch rates of diamond material containing a damage layer comprising sp.sup.2 bonded carbon

[0103] To further increase bulk etch rates of diamond material, the effect of pH on sulfate ion-containing electrolyte solutions was investigated. A mixture of K.sub.2SO.sub.4 and H.sub.2SO.sub.4 was used to vary the pH while maintaining a total sulfate ion concentration of 0.25 M.

[0104] The results are presented in FIG. 6.

[0105] It was found that a decrease in pH, and its associated increase in conductivity (42 to 89 mS cm.sup.−1) has the effect of increasing the bulk etch rate. However, as shown in Example 2, changes in conductivity have little effect on the bulk etch rate and so the increase in bulk etch rate can be attributed to the decrease in pH.

[0106] The bulk etch rate observed showed a general upward trend with increasing proton (H.sup.+) concentration suggesting that protons are also involved in the rate limiting step. In particular, the factor 2 increase in the bulk etch rate upon decreasing the pH from 7 to 0.8 demonstrates clearly that the use of more acidic solutions as electrolytes offers further advantageous improvements in the bulk etch rate.

EXAMPLE 4

[0107] The use of single crystal boron doped diamond (BDD) instead of optical grade single crystal diamond

[0108] To investigate whether the approach extends to doped diamond material we investigated the electrochemical etching of 5 mm.sup.2 CVD single crystal diamond material doped with ˜10.sup.19 boron atoms cm.sup.−3. This material was implanted an annealed in the same fashion as all the other material discussed.

[0109] Although it is difficult to process the optical etch data in the normal fashion, due to the less significant colour change, it is obvious by eye that the sp.sup.2 bonded carbon layer is being removed, as seen in FIG. 7. Full separation of the 5 mm.sup.2 plate was achieved in 60 minutes in a 0.25 M H.sub.2SO.sub.4 solution, giving a comparable separation time to non-doped material.

EXAMPLE 5

[0110] The use of polycrystalline optical grade material instead of single crystal optical grade material

[0111] To investigate whether the approach extends to polycrystalline diamond material we investigated the electrochemical etching of 4 mm square CVD optical grade polycrystalline material. This material was implanted an annealed in the same fashion as all the other material discussed.

[0112] It is obvious by eye that the sp.sup.2 bonded carbon layer is being removed, as seen in FIG. 9. Full separation of the 4 mm square plate was achieved in 60 minutes in a 0.25 M H.sub.2SO.sub.4 solution, giving a comparable separation time to non-doped material.

[0113] EPR measurements were obtained from different salt solutions after electrolysis as described above in order to determine the presence of the hydroxyl radical. It is thought that the presence of the hydroxyl radical has an effect on the etching rate of the damage layer. FIG. 9 shows the experimental and fitted data for potassium chloride and potassium sulphate solutions. The peaks shown in the three traces of FIG. 9 are indicative of the hydroxyl radical, and the areas under the peaks are indicative of the quantity of hydroxyl radicals present.

[0114] It can be seen from FIG. 9 that the sulphate solution formed far more hydroxyl radicals than the chloride solution. This corresponds with the etch rates shown in FIG. 4, in which diamond in the sulphate solution has a faster etch rate than diamond in the chloride solution.

[0115] Exemplary steps of the method are show in FIG. 10. The following numbering corresponds to that of FIG. 9:

[0116] S1. Diamond material is provided. This may be single crystal or polycrystalline diamond material. If it is single crystal diamond material, it may be natural or HPHT diamond material. Furthermore, the diamond material may be doped, for example with any of nitrogen, boron, phosphorus, and silicon.

[0117] S2. The diamond material is ion implanted so as to form a damage layer comprising sp.sup.2 bonded carbon, wherein the presence of the damage layer defines a first diamond layer above and in contact with the damage layer and a second diamond layer below and in contact with the damage layer.

[0118] S3. As an optional step, the diamond material may be annealed in order to increase the amount of sp.sup.2 carbon in the damage layer.

[0119] S4. The damage layer is electrochemically etched in the presence of ions capable of forming radicals during the etching process. The etching of the damage layer allows the first and second layers to separate. Suitable ions include, but are not limited to, persulfate ions, sulfate ions, oxalate ions, chloride ions, carbonate ions and metal ions.

[0120] Turning now to FIG. 11, some of the steps of FIG. 10 are illustrated schematically (and not to scale).

[0121] FIG. 11a shows a provided diamond material 1, which is subjected to ion implantation.

[0122] FIG. 11b shows a first diamond layer 2 and a second diamond layer 3, each separated by a damage layer 4 comprising sp.sup.2 bonded carbon, formed during the ion implantation step.

[0123] FIG. 11c shows electrochemical etching in the presence of ions capable of forming radicals during electrolysis. The electrochemical etching preferentially attacks the damage layer 4.

[0124] FIG. 11d shows the damage layer 4 having been etched away, allowing the separation of the first layer 2 from the second layer 3.

[0125] When removing a separated layer from the damage layer, the separated layer can be very thin, and difficult to handle without introducing damage to the separated layer. One way to address this issue is to provide a support structure, such as one described in WO 2016/058037, in which a support structure is attached to the diamond material by gluing, brazing, overgrowing polycrystalline diamond, or any suitable means, provided that means is capable of surviving the etching process. For example, a mask made from a material such as diamond may be glued to the diamond material, which allows the separated layer, once removed, to be easily handled.

[0126] When preparing thin plates of polycrystalline CVD diamond, the present invention allows such plates to have a larger average lateral grain size and lower sp.sup.2 carbon quantity than was previously possible. A plate of polycrystalline CVD diamond material usually comprises a nucleation face (the face at which CVD growth started) which comprises small grains of diamond material and significant quantities of non-diamond sp2 carbon (detectable via Raman spectroscopy); and a growth face, which comprises larger grains of diamond material and, if growth conditions are controlled correctly, a lower quantity of non-diamond sp.sup.2 carbon. Diamond grain size increases on moving through such a wafer of polycrystalline CVD diamond material from the nucleation face to the growth face.

[0127] Using prior art techniques to obtain thin plates of polycrystalline CVD diamond, thin plates are grown directly on a substrate and therefore formed close to the nucleation face. Such plates have a small average lateral grain size and relatively high sp.sup.2 carbon content.

[0128] A relatively thick plate can be grown with the growth face having a large average lateral grain size and a low sp2 carbon content. Using the techniques described above, a layer can be removed from the growth surface, leaving a thin layer with a relatively high (greater than 5 μm) average lateral grain size and a thickness of no more than 30 μm. The average lateral grain size may be greater than 10 μm, greater than 15 μm, greater than 20 μm.

[0129] The average lateral grain size of a polycrystalline CVD diamond surface was measured using scanning electron microscopy (SEM). SEM images of a surface of a polycrystalline CVD diamond material indicate boundaries between grains such that individual grains can be identified and counted. Accordingly, an area of the polycrystalline CVD diamond surface can be imaged using SEM, the total number of diamond grains along a line across the image can then be counted, and then the line length can be divided by the number of grains along the line to obtain and an average lateral grain size. A number of lines across the SEM image can be analysed in this manner, optionally in perpendicular directions, and an average value calculated for the lateral grain size across the imaged area.

[0130] While this invention has been particularly shown and described with reference to embodiments, it will be understood to those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as defined by the appendant claims.