MATERIALS

Background of Invention

The present invention relates to post-synthesis processing of diamond and related super-hard materials.

Summary of Invention

In the context of the present invention super-hard materials are defined as those materials having a Vickers hardness of no less than 2000 kg/mm ² . These materials include a range of diamond materials, cubic boron nitride materials (cBN), sapphire, and composites comprising the aforementioned materials. For example, diamond materials include chemical vapour deposited (CVD) single crystal and polycrystalline synthetic diamond materials of a variety of grades, high pressure high temperature (HPHT) synthetic diamond materials of a variety of grades, natural diamond material, and diamond composite materials such as polycrystalline diamond which includes a metal binder phase (PCD) or silicon cemented diamond (ScD) which includes a silicon/silicon carbide binder phase.

In relation to the above, it should be noted that while super-hard materials are exceedingly hard, they are generally very brittle and have low toughness. As such, these materials are notoriously difficult to process into a product after the raw material is synthesized. Any processing method must be sufficiently aggressive to overcome the extreme hardness of the super-hard material while at the same time must not impart a large degree of stress or thermal shock to the material which would cause macroscopic fracturing of the material due to its brittle nature and low toughness. Furthermore, for certain applications it is important that surface and sub-surface damage at a microscopic scale, such as microcracking, is minimized to avoid deterioration of functional properties which may result from such surface and subsurface damage including, for example, optical scattering, increased optical absorption, decreased wear resistance, and increased internal stress resulting in a decrease in coherence time for quantum spin defects near the processed surface.

There is narrow operating window for achieving successful processing of super-hard materials and many available processing methods fall outside this operating window. For example, most processing methods are not sufficiently aggressive to process super-hard materials to any significant extent in reasonable time-frames. Conversely, more aggressive processing techniques tend to impart too much stress and/or thermal shock to the super-hard material thus causing cracking and material damage or failure.

Certain processing methods have operational parameters which can be altered so as to move from a regime in which no significant processing of a super-hard material is achieved into a regime in which processing is achieved but with associated cracking and damage or failure of the super-hard material. In this case, there may or may not be a transitional window of parameter space in which processing can be achieved without cracking and damage or failure of the super-hard material. The ability to operate within a suitable window of parameter space in which processing can be achieved without cracking and damage or failure of the super-hard material will depend on the processing technique, the size of any transitional operating window for such a technique, and the level of operation parameter control which is possible to maintain processing within the window of parameter space in which processing can be achieved without cracking and damage or failure of the super-hard material.

In light of the above, it will be appreciated that post-synthesis processing of super-hard materials is not a simple process and, although a significant body of research has been aimed at addressing this problem, current processing methods are still relatively time consuming and expensive, with processing costs accounting for a significant proportion of the production costs of super-hard material products.

Post synthesis processing may comprise one or more of the following basic processes: surface processing to remove material from the surface of the as-grown super-hard material in order to increase surface flatness, decrease surface roughness, remove surface defects, and/or attain a target thickness for the super-hard material; surface processing to achieve a fine surface finish where minimal material is removed from the super-hard product, i.e. polishing; and cutting of the super-hard material into target shapes and sizes for particular product application.

In principle there are two basic forms of mechanical surface processing: (i) a two-body process in which abrasive particles are embedded/fixed in one body which is moved against a second body to process the second body; and (ii) a three-body process in which one body is moved relative to a second body to be processed and free abrasive particles, constituting a third body, are disposed between the first and second bodies in order to achieving surface processing of the second body.

The latter three-body approach to surface processing is known as lapping and it is this approach which is conventionally used to remove macroscopic quantities of surface material from super-hard materials. Three-body lapping, as opposed to a two-body surface processing technique, is preferred for removing macroscopic quantities of surface material from super- hard materials as it has been found that lapping is more efficient at removing surface material from a super-hard material without imparting a large degree of stress or thermal shock to the material which would cause macroscopic fracturing of the material due to its brittle nature and low toughness. In contrast, when it is desired to achieve a fine surface finish without removing macroscopic quantities of material then a two-body processing technique may be utilized. As such, conventionally lapping is used to remove material from the surface of an as-grown super-hard material in order to increase surface flatness, decrease surface roughness, remove surface defects, and/or attain a target thickness for the super-hard material. Subsequently, if a fine surface finish is required, the super-hard material is polished and this may be performed using a two-body surface processing technique in which abrasive material is fixed in a polishing wheel such as via resin bonding. Polishing may also be achieved using an iron or steel wheel which is diamond impregnated and this is known as scaife polishing. Although scaife polishing generally utilizes free diamond abrasive particles these are of a small size relative to pores within the iron or steel wheel and are thus embedded/fixed into the wheel thus effecting a two-body processing as opposed to a true three body lapping process.

The most appropriate surface processing technique will depend on the end application, the type of surface finish required for the end application, and commercial considerations including an evaluation of the cost of a particular processing technique versus the commercial value of the product obtain after such processing. Surface processing parameters of interest may include one or more of: roughness; flatness; curvature; surface/sub-surface crystal damage; speed; cost; precision; and repeatability. Prior to discussing embodiments of the invention in more detail, it may be pertinent to clarify the distinction between flatness and roughness, particularly in the context of synthesis and processing of super-hard materials such as synthetic diamond materials. In this regard, a skilled person will understand that flatness and roughness are two different characteristics of a surface and particular applications will be sensitive to either one or both of these characteristics. For example, a smooth curved surface has low roughness but it not fiat as illustrated in Figure 1 whereas a rough non-curved surface may be flat but have a high degree of roughness as illustrated in Figure 2. Roughness is generally the deviation of a surface from a smooth target profile measured on a microscopic scale relative to the scale of the surface area whereas flatness is generally the deviation of a surface from a smooth target profile measured on a macroscopic scale relative to the scale of the surface area. The two parameters are thus distinguished by the method of measuring deviations from a smooth surface profile with roughness being measured by a technique which determines deviations from the smooth surface profile at a microscopic scale and flatness/curvature being measured by a technique which determines deviations from a smooth surface profile at a macroscopic scale. For a wafer of material having two opposing surfaces then surface parallelism may often also be an important additional parameter.

In light of the above, it will be evident that a surface which has low roughness may still deviate significantly from a smooth target profile due to macroscopic deviations from the smooth target profile. For example, Figure 3 shows a schematic illustration of a wafer of super-hard material which has a surface profile which is bowed from a targeted smooth fiat configuration. Furthermore, if the target profile is fiat then a low surface roughness surface can still deviate significantly from a smooth fiat target profile due to non-perpendicular surface processing leading to a sloped or wedge-shaped profile as illustrated in Figure 4. Similar deviations to those illustrated in Figures 3 and 4 for fiat surface can also occur when a curved surface profile is desired.

Such deviations from a target profile may be caused by stress introduced into the super-hard material during synthesis and/or during surface processing which can lead to bowing in a super-hard material Furthermore, macroscopic deviations from a smooth target profile may also result due to non-uniform processing. The aforementioned issues are typically more problematic for super-hard materials when compared to less hard materials for a number of reasons as discussed below.

First, synthesis conditions for super-hard materials are often extreme, e.g. very high pressures and/or temperatures, leading to stress in the synthesized super-hard material which can cause bowing.

Secondly, the extreme hardness of super-hard materials typically requires a large amount of energy to be imparted to process a surface of the material and this generates heat leading to the generation of thermal stress during processing which can again lead to bowing.

Thirdly, due to the extreme hardness of super-hard materials, abrasive particles can be broken down into smaller particles during processing of a super-hard material which can result in differential processing of, for example, central and outer regions of a wafer of super-hard material.

Fourthly, the extreme hardness of super-hard materials typically requires a large force to be applied to the super-hard material during processing and if this force isn't uniformly applied across the surface of the super-hard material during process a sloped or wedge-shaped profile can result.

Fifthly, the extreme hardness of super-hard materials typically requires a large force do be applied to the super-hard material during processing and this can lead to development of non- uniformities in the surface of the processing wheel over time, e.g. deviations from a flat processing surface, which can be result in corresponding non-uniformities in the surface of the surface of the super-hard materials being processed.

In addition to the above, even if a smooth, flat surface can be achieved by a particular surface processing method, or a combination of surface processing methods, problems may still exist in terms of crystal defects/damage being imparted into the crystal surface and sub-surface. Figure 5 illustrates a super-hard material having a smooth and flat surface but where micro- cracks have been formed in a surface region due to forces imparted during the processing of the surface to a high degree of smoothness and flatness. This is a particular problem for super-hard materials due to the high hardness and low toughness of such materials. Several different methods are available for measuring surface and sub-surface crystal damage. For example, one technique involves applying a revealing plasma etch to the processed surface which preferentially etches cracked or damaged regions to form etch pits which can then be counted to evaluate the density of defects at the processed surface.

In light of the above, it is evident that while it is desirable for many applications to form low roughness and highly flat or precisely curved surfaces without imparting defects/damage into the crystal structure there are many problems associated with forming such surfaces in super- hard materials and surface processing times are very long.

It is an aim of embodiments of the present invention to provide a more cost effective surface process methodology for such super-hard materials. In particular, it is an aim to reduce the processing time to achieve a target surface flatness and/or roughness for such super-hard materials. Further still, certain embodiments of the present invention can also achieve a target surface roughness without imparting as much surface and sub-surface crystal damage into the super-hard material.

Summary of Invention

Typically, when processing non-super-hard materials, the processing wheel is much more abrasive than the material being processed. In this case, the processing rates are high, removal of surface material is relatively easy, wear of the processing wheel is not a major issue, and large surface areas of material can be readily processed to a high degree of precision.

In contrast, surface processing of super- hard materials, particularly large area components such as full wafers of poly crystalline CVD diamond material, can take a long time due to very slow material removal rates (e.g. less than 1 μητ/ητ) adding significant cost to the total fabrication process. It has been found that the rate of material removal is highly dependent on subtle changes to the surface profiles of the processing surfaces used to process a super-hard material. Typically, a super-hard material will undergo a lapping process to remove material from the surface of the super-hard material to move towards a desired target surface finish. The lapping process will then typically be followed by a polishing process to reduce the surface roughness of the lapped surface to a desired level.

It has been noted that during lapping of a super-hard material the lapping surface undergoes significant wear during processing and tends to move away from a starting profile (which is typically flat but can be a desired convex or concave target shape for certain products). This will result in a lapped surface which deviates from a desired target profile such as a perfectly flat surface finish. For example, if the lapping wheel is worn to have a small convex curvature during operation (by preferential wear at peripheral regions of the lapping wheel which is used for processing a super-hard material), then the lapped surface of the super-hard material will have a small concave curvature which is complimentary to the final profile of the processing surface after the lapping process is completed.

If it is desired to then polish the slightly concave lapped surface of the super-hard material to achieve a target surface roughness two options seem feasible: (i) use a flat polishing wheel in which case the edge regions of concave lapped surface will need to be worn away prior to the polishing wheel contacting and polishing a central region; or (ii) use a polishing wheel which has a small convex curvature which complimentary to the concave lapped surface such that the entire lapped surface is contacted and polished from the initiation of the polishing process. Intuitively, one would expect that option (ii) would provide the fastest route to achieving the target surface roughness. Surprisingly, it has been found that neither of the aforementioned options provides the optimal process for achieving the desired level of surface roughness in the shortest time scale. Rather, it has been found that a target surface roughness can be achieved in less than half the time required following option (ii) by utilizing at least two different polishing wheels including one which is slightly less convex and one which is slightly more convex than a convex surface which would provide a perfect match to the slightly concave surface of super-hard material. That is, rather than using a polishing wheel having a convex surface profile which has a perfect complimentary match to the surface of the super-hard material to be polished, it has been found that polishing times can be reduced dramatically by using two different polishing wheels which are slightly more and less convex than would be intuitively selected. For example, it has been found that in one application polishing polycrystalline CVD diamond wafers a polishing time of over 100 hours following option (ii) is reduced to less than 50 hours using two different polishing wheels having slight less and slightly more convex surface profiles. Furthermore, it has been found that the order in which the less and more convex polishing wheels are utilized does not significantly change the processing time. This again seems counter-intuitive. However, it is useful for industrial applications as a set of polishing wheels can be provided to an operator and it does not matter in which order the polishing wheels are utilized to obtain the desired surface finish at reduced processing times. Providing slightly less and more convex polishing wheels ensures that each polishing wheel works a slightly different area of the slightly concave lapped surface of the super-hard material. Furthermore, if the polishing wheels are sufficiently close in shape the polished areas blend into each other to form a uniform finish in a shorter timescale than if the entire lapped surface of the super-hard material is polished at the same time using a single complimentary polishing wheel. In this regard, it is found that the centre and the edge of a lapped surface of a super-hard material are the areas where most flaws are found and therefore these are the areas to focus on polishing to ensure a good homogeneous finish. If a perfectly complimentary polishing wheel is used then this can lead to extended polishing times as to remove, for example, lapping damage at the edge and/or centre of the lapped surface of the super-hard material requires material from the whole lapped surface to be removed in a uniform manner until all the damage in central and edge regions is removed. In contrast, using slightly more and less convex polishing wheels focuses polishing in central and edge regions such that the damage in these regions can be removed and the surface finish blended into a less damaged intermediate region of the lapped surface of the super-hard material. As such, the methodology as described herein has been found to be an optimal way of achieving a low surface roughness, polished super-hard material in significantly shorter timescales. Accordingly, the present invention provides cost efficient processing, faster more predictable throughput, and fewer machines required for a given capacity. Furthermore, by reducing polishing times it can also be possible to produce a polished super-hard material product which has less surface and sub-surface crystal damage caused by the polishing process.

While the invention has been described above in relation to lapping using a three-body process followed by polishing using a two-body process, in principle the basic methodology of using differently shaped processing surfaces to reduce surface processing times is not limit to this specific combination of processing methods. As such, in accordance with the present invention there is provided a method of processing a super-hard material having a Vickers hardness of no less than 2000 kg/mm ² , the method comprising:

(a) a first processing step in which a processing surface is moved over a surface of the super-hard material such that an abrasive removes material from the surface of the super-hard material; (b) a measuring step in which a surface profile of either the processing surface or the surface of the super-hard material is measured after performing the first processing step to yield a measured end surface profile of the first processing step; and

(c) a second processing step for removing material from the surface of the super-hard material, the second processing step comprising the use of at least two processing sub-steps using different processing surfaces for processing the surface of the super-hard material after the first processing step, the second processing step including the use of a first processing surface which has a first surface profile which is measured to have a larger radius of curvature than said measured end surface profile of the first processing step and a second processing surface which is measured to have a second surface profile which has a smaller radius of curvature than said measured end surface profile of the first processing step.

Brief Description of the Drawings

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

Figure 1 shows a smooth curved surface;

Figure 2 shows a flat rough surface;

Figure 3 shows a schematic illustration of a wafer of super-hard material which has bowed from a targeted flat configuration due to stress induced during synthesis and/or surface processing and/or due to non-uniform processing;

Figure 4 shows a schematic illustration of a wafer of super-hard material which has a wedge shaped profile deviating from a targeted flat configuration due to non-uniform processing;

Figure 5 shows a schematic illustration of a wafer of super-hard material which has a smooth flat surface profile which comprises surface damage in the form of microcracks introduced during surface processing; Figure 6 shows a flow diagram illustrating a multi-step surface processing technique for processing a super-hard material which uses a flat polishing surface which is not according to the present invention;

Figure 7 shows a flow diagram illustrating another multi-step surface processing technique for processing a super- hard material which uses a complimentary polishing surface which is not according to the present invention;

Figure 8 shows a flow diagram illustrating a multi-step surface processing technique for processing a super-hard material according to an embodiment of the present invention which uses polishing surfaces which are slightly more and slightly less convex than a complimentary polishing surface; and

Figures 9 and 10 show schematic illustrations of a lapping apparatus for performing an initial surface processing step.

Detailed Description

Figures 1 to 5 have already been discussed in the background section of this specification and serve to illustrate some different types of surfaces which can be generated during synthesis and processing of a super-hard material. Particular problems in processing super-hard materials have also been discussed in the background section and the desire to develop improved surface processing methodology for super-hard materials in order to achieve low roughness surfaces finishes, preferably also having low surface and sub-surface crystal damage, in a cost effective and reduced time scale.

Figure 6 shows a flow diagram illustrating a multi-step surface processing technique for processing a super-hard material which comprises lapping and polishing techniques which is not according to the present invention. The method comprises:

(a) a first processing step (which in the illustration is a lapping process) in which a first processing surface 60 is moved over a surface of the super-hard material 62 such that a first abrasive 64 removes material from the surface 66 of the super- hard material 62, said first processing surface 60 having a first surface profile which deviates from a smooth flat, convex, or concave shape (in the illustration this is a convex surface forming a concave surface on the super-hard material 62); and

(b) a second processing step (which in the illustration is a polishing process) in which a second processing surface 68 is moved over the surface 66 of the super-hard material 62 such that a second abrasive 69 removes material from the surface 66 of the super-hard material, said second processing surface 68 having a second surface profile which is flat.

In the method illustrated in Figure 6, it should be noted that deviations in the surface flatness of the processing surfaces have been exaggerated for illustrative purposes. The first processing step is a three-body lapping process in which the first abrasive is supported in a fluid introduced between the first processing surface and the surface of the super-hard material while the second processing step is a two-body polishing process in which the second abrasive is embedded in the second processing surface. For example, the second processing step may comprise resin bond polishing in which the second abrasive is bonded to the second processing surface with a resin.

In the illustrated arrangement the first processing step causes differential wear of the first processing surface 60 which develops a convex surface profile such that the super-hard material 62 develops a concave surface profile. If the subsequent polishing step uses a flat polishing wheel then edge regions of the super-hard material are required to be polished back prior to central regions of the super-hard material being polished which can be time consuming.

Figure 7 shows a flow diagram illustrating another multi-step surface processing technique for processing a super-hard material which comprises lapping and polishing techniques which is not according to the present invention. The method comprises:

(a) a first processing step (which in the illustration is a lapping process) in which a first processing surface 70 is moved over a surface of the super-hard material 72 such that a first abrasive 74 removes material from the surface 76 of the super- hard material 72, said first processing surface 70 having a first surface profile which deviates from a smooth flat, convex, or concave shape (in the illustration this is a convex surface forming a concave surface on the super-hard material 72); and (b) a second processing step (which in the illustration is a polishing process) in which a second processing surface 78 is moved over the surface 76 of the super-hard material 72 such that a second abrasive 79 removes material from the surface 76 of the super-hard material, said second processing surface 78 having a second surface profile which is convex and complimentary to the concave surface of the super-hard material after the lapping step.

In the method illustrated in Figure 7, it should be noted that deviations in the surface flatness of the processing surfaces have been exaggerated for illustrative purposes. The first processing step is a three-body lapping process in which the first abrasive is supported in a fluid introduced between the first processing surface and the surface of the super-hard material while the second processing step is a two-body polishing process in which the second abrasive is embedded in the second processing surface. For example, the second processing step may comprise resin bond polishing in which the second abrasive is bonded to the second processing surface with a resin.

In the illustrated arrangement the first processing step causes differential wear of the first processing surface 70 which develops a convex surface profile such that the super-hard material 72 develops a concave surface profile. If the subsequent polishing step uses a complimentary convex polishing wheel then the entire surface of the super-hard material is contacted and polished at the same time. However, since it has been found that lapping tends to cause increased surface damage in central and peripheral regions of the super-hard material then it has been found that polishing the entire lapped surface of the super-hard material at the same time to remove lapping damage from central and peripheral regions is not the most efficient approach and again can be time consuming.

In contrast to the above, Figure 8 shows a flow diagram illustrating another multi-step surface processing technique for processing a super-hard material which comprises lapping and polishing techniques according to the present invention. The method comprises:

(a) a first processing step (which in the illustration is a lapping process) in which a first processing surface 80 is moved over a surface of the super-hard material 82 such that a first abrasive 84 removes material from the surface 86 of the super- hard material 82, said first processing surface 80 having a first surface profile which deviates from a smooth flat, convex, or concave shape (in the illustration this is a convex surface forming a concave surface on the super-hard material 82); and

(b & c) a second processing step, which in the illustration is a polishing process, including two polishing sub-steps including (b) a first polishing step comprising a first polishing surface 88 which is moved over the surface 86 of the super- hard material 82 such that abrasive 89 removes material from the surface 86 of the super-hard material, said first polishing surface 88 having a surface profile which is convex, but less so than would be the case for a complimentary processing surface, and (c) a second polishing step comprising a second polishing surface 98 which is moved over the surface 86 of the super-hard material 82 such that abrasive 99 removes material from the surface 86 of the super-hard material, said second polishing surface 98 having a surface profile which is more convex than would be the case for a complimentary processing surface.

In the method illustrated in Figure 8, it should be noted that deviations in the surface flatness of the processing surfaces have been exaggerated for illustrative purposes. The first processing step (a) is a three-body lapping process in which the abrasive is supported in a fluid introduced between the processing surface and the surface of the super-hard material while the second processing steps (b & c) are two-body polishing processes in which the abrasive is embedded in the processing surfaces. For example, the second processing steps may comprise resin bond polishing in which the abrasive is bonded to the processing surface with a resin, i.e. the polishing wheel of the or each polishing stage in the second processing step comprises abrasive bonded in a resin.

As described in the summary of invention section, embodiments of the present invention use at least two different polishing wheels including one which is slightly less convex and one which is slightly more convex than a convex surface which would provide a perfect match to the slightly concave surface of super-hard material. That is, rather than using a polishing wheel having a convex surface profile which has a perfect complimentary match to the surface of the super-hard material to be polished, it has been found that polishing times can be reduced dramatically by using two different polishing wheels which are slightly more and less convex than would be intuitively selected. Furthermore, it has been found that the order in which the less and more convex polishing wheels are utilized does not significantly change the processing time. This again seems counter-intuitive. However, it is useful for industrial applications as a set of polishing wheels can be provided to an operator and it does not matter in which order the polishing wheels are utilized to obtain the desired surface finish at reduced processing times.

Providing slightly less and more convex polishing wheels ensures that each polishing wheel works a slightly different area of the slightly concave lapped surface of the super-hard material. Furthermore, if the polishing wheels are sufficiently close in shape the polished areas blend into each other to form a uniform finish in a shorter timescale than if the entire lapped surface of the super-hard material is polished at the same time using a single complimentary polishing wheel. In this regard, it is found that the centre and the edge of a lapped surface of a super-hard material are the areas where most flaws are found and therefore these are the areas to focus on polishing to ensure a good homogeneous finish. If a perfectly complimentary polishing wheel is used then this can lead to extended polishing times as to remove, for example, lapping damage at the edge and/or centre of the lapped surface of the super-hard material requires material from the whole lapped surface to be removed in a uniform manner until all the damage in central and edge regions is removed. In contrast, using slightly more and less convex polishing wheels focusses polishing in central and edge regions such that the damage in these regions can be removed and the surface finish blended into a less damaged intermediate region of the lapped surface of the super-hard material. As such, the methodology as described herein has been found to be an optimal way of achieving a low surface roughness, polished super-hard material in significantly shorter timescales. Accordingly, the present invention provides cost efficient processing, faster more predictable throughput, and fewer machines required for a given capacity. Furthermore, by reducing polishing times it can also be possible to produce a polished super-hard material product which has less surface and sub-surface crystal damage caused by the polishing process.

While the invention has been described above in relation to lapping using a three-body process followed by polishing using a two-body process, in principle the basic methodology of using differently shaped processing surfaces to reduce surface processing times is not limit to this specific combination of processing methods. For example, it is also envisaged that the first processing step could comprise a two-body process rather than a three-body lapping process. Furthermore, the second processing step could comprise a chemical-mechanical polishing process (CMP) rather than resin bond polishing. Further still, while the first processing step may consist of a single processing step, in an alternative the first processing step may comprise a plurality of processing steps using sequentially smaller abrasive particles to reduce the surface roughness achieved by the first processing step prior to final polishing in a second processing step.

As such, in accordance with the present invention there is provided a method of processing a super-hard material having a Vickers hardness of no less than 2000 kg/mm ² , the method comprising:

(a) a first processing step in which a processing surface is moved over a surface of the super-hard material such that a first abrasive removes material from the surface of the super-hard material;

(b) a measuring step in which a surface profile of either the processing surface or the surface of the super-hard material is measured after performing the first processing step to yield a measured end surface profile of the first processing step; and

(c) a second processing step in which a second abrasive removes material from the surface of the super-hard material, wherein said second processing step comprises the use of at least two processing sub- steps using different processing surfaces for processing the surface of the super-hard material after the first processing step, the second processing step including the use of a first processing surface which has a first surface profile which has a larger radius of curvature than said measured end surface profile of the first processing step and a second processing surface which has a second surface profile which has a smaller radius of curvature than said measured end surface profile of the first processing step.

The measuring step comprises measuring the surface profile of either the processing surface or the surface of the super-hard material after the first processing step. This could in principle include measuring both the surface profile of the processing surface and the surface of the super-hard material after the first processing step although in practice usually only one or other of these surfaces is required to be measured. The second processing step may utilize the first processing surface followed by the second processing surface or alternatively may utilize the second processing surface followed by the first processing surface. As previously stated, it has been found that the order in which these processing surfaces are used is not critical.

During at least a final stage of the first processing step the surface profile of the processing surface may be controlled to have a deviation from a smooth target profile of no more than 20 μιη or no more than 10 μιη. In the second processing step the first surface profile may be formed to deviate from the measured end surface profile of the first processing step by an amount in a range 2 μιη to 20 μιη or more preferably in a range 5 μιη to 10 μιη. Similarly, in the second processing step the second surface profile may be formed to deviate from the measured end surface profile of the first processing step by an amount in a range 2 μιη to 20 μιη or more preferably in a range 5 μιη to 10 μιη, but in an opposite direction relative to the first surface profile. For example, if the measured end surface profile of the lapping wheel at the end of the first processing step is slightly convex and deviates from a perfectly flat surface profile by 10 μιη then during the second processing step two polishing wheels may be used, one of which deviates from a perfectly flat surface by 0 to 5 μιη, and one of which deviates from a perfectly flat surface by 15 to 20 μιη. This ensures that each polishing wheel works a slightly different area of the super-hard material but also ensures that they are sufficiently close in shape that the polished areas blend into each other to form a uniform finish.

The second processing step may further comprise the use of a third processing surface having a third surface profile which is different to that of the first and second processing surfaces. For example, the third surface profile may have a radius of curvature which lies between that of the first and second surface profiles. For example, in the aforementioned arrangement a set of three polishing wheels may be used, one which is slightly more convex than a complimentary shape, one of which is slightly less convex than a complimentary shape, and a third polishing wheel which is complimentary in shape. The third processing surface may then be applied to the surface of the super-hard material to process the surface of the super- hard material after application of the first and second processing surfaces.

According to certain embodiments, the first processing step is a three-body lapping process in which the abrasive is supported in a fluid introduced between the processing surface and the surface of the super-hard material. Furthermore, according to certain embodiments the second processing step is a two-body polishing process including at least two polishing stages, each polishing stage using a polishing wheel having a different surface profile so as to provide the first and second processing surfaces. For example, the second processing step comprises resin bond polishing in which the second abrasive is bonded to the polishing wheel with a resin.

The surface of the super- hard material may have at least one linear dimension of at least 10 mm, 20 mm, 40 mm, 60 mm, 80 mm, 100 mm, 120 mm, or 140 mm. After the first processing step, the surface of the super-hard material may have a surface roughness Ra of no more than 5 μιη, 3 μιη, 1 μιη, or 500 nm. The first processing step may comprise a plurality of processing steps using sequentially smaller abrasive particles to achieve such a final surface roughness. Furthermore, after the second processing step, the surface of the super- hard material has a surface roughness Ra of no more than 20 nm, 15 nm, 10 nm, 5 nm, 3 nm, or 1 nm. Such low surface roughness values can be achieved by, for example, using a fine resin bond polishing technique after lapping.

It will be appreciated that certain embodiments of the present invention will be applicable when a low surface roughness is required but when some deviation from a flat or smoothly curved convex or concave target profile can be tolerated in the end application. It should also be appreciated that embodiments of the present invention will only be easy and cost effective to implement when the surface profile of the second processing step is relatively easy to modify to match changes in the surface profile of the first processing surface. This is the case when using resin bond polishing in which a resin is coated onto a carrier and abrasive particles are mounted in the resin coating. In this case, the resin can be molded to have a surface profile meeting the previously defined characteristics.

In addition to the above, it will also be appreciated that according to certain embodiments one can control the surface profile of the first processing step (e.g. the surface profile of the lapping wheel) in the final stages of the first processing step to generate a surface profile on the super-hard material such that after applying the two stage second processing step (e.g. a two stage polishing step) as described herein a desired flat, convex, or concave surface profile is achieved without undue deviations from a target flat or smoothly curved convex or concave target profile. The first processing step is usually adapted to remove a significant portion of surface material from a super-hard material to move towards a desired thickness, flatness (or curvature), and surface roughness. However, surface processing steps which are designed to remove a significant portion of surface material from a super-hard material are usually not optimized to achieve a very low surface roughness. For example, after the first processing step the surface of the super-hard material may have a surface roughness Ra of no more than 5 μιη, 3 μιη, 1 μιη, or 500 nm but will usually have a surface roughness of no less than 50 nm, 100 nm, 200 nm, 300 nm, or 500 nm. In contrast, after the second processing step the surface of the super- hard material may have a surface roughness Ra of no more than 20 nm, 15 nm, 10 nm, 5 nm, 3 nm, or 1 nm.

Certain embodiments of the present invention are particularly useful when processing relatively large areas of super-hard material. For example, the surface of the super- hard material may have at least one linear dimension of at least 10 mm, 20 mm, 40 mm, 60 mm, 80 mm, 100 mm, 120 mm, or 140 mm. Such large pieces of super- hard material are more likely to extend over portions of the first processing surface which are not perfectly fiat (or curved) and thus are more likely to deviate from a perfect target surface profile after completion of the first processing step and prior to polishing in a second processing step. As such, larger pieces of super-hard material will be more prone to requiring a long final polishing step to achieve a desired low surface roughness. Accordingly, embodiments of the present invention can achieve larger gains in terms of time and cost of processing for larger pieces of super-hard material.

Certain embodiments of the present invention are particularly useful when processing involves diamond materials due to the extreme hardness of such materials. For example, the super-hard material being processed may comprise synthetic diamond material such as a wafer of poly crystalline CVD diamond material. Furthermore, the abrasive used to achieve the surface processing in the first and second processing steps may be formed of particles of diamond material.

In practice as-grown super-hard material will usually be cut, lapped, and then polished to generate a target surface profile and surface roughness for a particular application. Cutting is usually achieved using a laser although other cutting techniques such as electron beam cutting can be used. Further still, if the super-hard material is electrically conductive it may be cut using electric discharge machining.

Lapping is then performed to generate an approximate surface profile. In a standard lapping process a super-hard material is mounted on a rotatable processing wheel. An abrasive slurry comprising super-hard abrasive particles within a carrier fluid is dripped onto the surface of the processing wheel from above. Generally, the abrasive slurry is dripped onto the processing wheel near a central region thereof and the abrasive slurry moves radial outwards across the processing wheel during rotation of the processing wheel in use. For a rough lapping process where a significant amount of material is to be removed from a surface of the super-hard material product, the super-hard abrasive particles may be relatively large in size, e.g. having a particle size of greater than 1 μιη. These abrasive particles are larger than pores within the surface of the processing wheel and thus roll between the surface of the processing wheel and the surface of the super-hard material within an interface region in order to cause surface micro-cracking of the super-hard material and removal of material from the surface of the super-hard material.

One problem with the aforementioned lapping configuration is that the lapping process can be difficult to control in order to achieve high rates of material processing without causing undue damage to the surface of the super-hard material being processed. Furthermore, another problem with the aforementioned lapping configuration is that the lapping process can be difficult to obtain uniform processing across large areas of super-hard material.

Figures 9 and 10 illustrate a lapping apparatus configured to solve the aforementioned problems. The apparatus comprises a rotatable processing wheel 2 on which a super-hard material product to be processed is mounted. The mounting configuration may comprise a carrier substrate 4 mounted to the super-hard material 6 such that a loading force is applied to the super-hard material 6 via the carrier substrate 4. For example, the super-hard material 6 may be bonded to the carrier substrate 4. Alternatively, the super-hard material may be retained in a free-standing configuration which is not bonded to a carrier substrate.

The super-hard material 6 is adhered to a carrier substrate 4 and arranged such that a surface of the super-hard material is in contact with a surface of the processing wheel 2 with an interface region disposed between the surface of the super-hard material 6 and the surface of the processing wheel 2. A weight 8 is provided on the carrier substrate 4 such that the super- hard material 6 is pressed against the surface of the processing wheel 2 with a suitable loading force. In an alternative configuration a pneumatic arrangement can be utilized to apply the loading force in place of the weight 8.

As in the previously described arrangement, the carrier substrate 4 (if present) and the super- hard material product 6 can be mounted on the processing wheel within a constraining ring 10 which constrains a location of the super-hard material product 6 over the processing wheel 2. The constraining ring 10 may comprise a number of slots for allowing the passage of abrasive fluid therethrough although it is possible to utilize a constraining ring which does not comprise slots. The constraining ring 10 has an internal diameter which is larger than the diameter of the super-hard material product 6 and carrier substrate 4. Furthermore, both the constraining ring 10 and the super-hard material product 6 are mounted so as to rotate on the surface of the processing wheel 2 driven by rotation of the processing wheel. In the illustrated configuration the constraining ring 10 is rotatable mounted on the processing wheel 2 via constraining arm 12. In certain configurations the constraining ring 10 and/or the super-hard material product 6 are rotatably driven independently of the processing wheel 2 and this can be desirable to provide a controlled rotation of the constraining ring 10 and/or the super-hard material product 6 relative to the processing wheel 2. In this case, the constraining arm 12 may comprise driven wheels for rotating the constraining ring 10 and/or the super-hard material product 6. Alternatively, a rotating force may be applied from directly above the super-hard material product 6, e.g. via an upper surface of the super-hard material product 6, the carrier substrate 4, the weight 8, and/or via a pneumatic loading configuration if present.

The apparatus of Figures 9 and 10 is thus configured to provide an under- feed arrangement for the abrasive slurry. Abrasive slurry is fed upwards through a rotational post 20 as illustrated by arrow 22. The processing plate 2 is adapted to provide a plurality of feed ports 24 disposed in the surface thereof such that in use an abrasive slurry is fed through the feed ports 24 onto the surface of the processing wheel from underneath the processing wheel as illustrated by arrows 26. The abrasive particles then move radial outwards from the feed ports 24 across the surface of the processing wheel 2 and roll through the interface region between the super-hard material product 6 and the processing wheel 2 in order to cause surface micro-cracking of the super-hard material product and removal of material from the surface of the super-hard material product.

The plurality of feed ports 24 can be radially distributed across the surface of the processing wheel such that at least a portion of the abrasive slurry is fed directly from the feed ports into the interface region between the surface of the processing wheel and the surface of the super- hard material product being processed.

Surprisingly, it has been found that higher rates of material processing can be achieved in a much more controllable manner using a lapping configuration in which the surface of the processing wheel has one or more feed ports disposed therein and the abrasive slurry is fed through the feed ports during processing of the super-hard material product onto the surface of the processing wheel from underneath the processing wheel rather than dripped onto the surface of the processing wheel from above as is done in a more standard lapping configuration. A better surface finish is also achieved, especially for large polycrystalline CVD diamond wafers when compared with a top feed approach. While not being bound by theory, it is believed that the under-feed configuration is advantage over the top-feed configuration for the following reasons.

Using a top feed approach all abrasive particles entering an interface region between the surface of the processing wheel and the surface of the super-hard material must do so from an edge of the super-hard material. It has been found that this can lead to edge chipping, edge rounding, and/or groove formation across the surface of the super-hard material being processed. In contrast, if the abrasive slurry is under- fed then at least a portion of the slurry can be introduced directly under the super-hard material being processed in the interface region. As such, this abrasive material moves from a central region of the super- hard material to an edge region rather than from an edge region to a central region. It has been found that such a modified lapping technique reduces edge chipping, edge rounding, and grooving in the super-hard material being processed and thus can lead to a better surface finish. In addition, regardless of the direct under- feed to interface region configuration, it is also believed that generally an under-feed configuration allows the abrasive slurry to be introduced onto the surface of the processing wheel at a more controllable rate and optionally in a continuous stream. In addition to the above, it is also possible using the modified lapping process to achieve more uniform processing across a large surface of a super-hard material such as a polycrystalline CVD diamond wafer. As previously indicated, standard lapping techniques involve dripping a suspension of diamond grit onto the lapping wheel from above. However, using such a technique requires grit to move into a peripheral region of the interface between the lapping wheel and a polycrystalline CVD diamond wafer and then propagate across the interface region in order to process the surface of the polycrystalline CVD diamond wafer. The grit particles are broken down as they hit the peripheral region of the wafer and during propagation under the wafer. This can result in differential processing of peripheral and central regions of the wafer with central regions being processed by smaller particles of grit than peripheral regions. This problem is particular to processing of super-hard materials, such as diamond wafers, as other materials do not cause the diamond grit to be broken down into smaller particles. As previously described, in order to solve this problem the lapping apparatus has been modified to feed the suspension of diamond grit from an underside of the lapping wheel through holes in the lapping wheel at locations which result in the grit being fed directly into the interface region between the wafer and the lapping wheel. As such, using this arrangement it is possible to avoid differential processing of peripheral and central regions of the wafer.

After lapping, the super-hard material is polished and this is particular desirable when a low surface roughness is required. As described previously, polishing is a two-body surface processing technique in which abrasive material is fixed in a polishing wheel such as via resin bonding. A resin bond polishing wheel comprises a carrier substrate on which a layer of resin is disposed. Abrasive particles are set into the resin of the processing surface of the wheel. In accordance with embodiments of the present invention, two or more polishing wheels are provided and the layer of resin in each wheel is molded such that its exposed processing surface has a surface profile meeting the previously defined characteristics.

Embodiments of the present invention may be applied to a range of super- hard materials including a range of diamond materials, cubic boron nitride materials (cBN), sapphire, and composites comprising the aforementioned materials. Diamond materials include chemical vapour deposited (CVD) single crystal and polycrystalline synthetic diamond materials of a variety of grades, high pressure high temperature (HPHT) synthetic diamond materials of a variety of grades, natural diamond material, and diamond composite materials such as polycrystalline diamond which includes a metal binder phase (PCD) or silicon cemented diamond (ScD) which includes a silicon/silicon carbide binder phase. Certain embodiments are particular useful for processing synthetic diamond materials such as polycrystalline CVD synthetic diamond materials and particularly to such materials over relatively large surface areas.

Utilizing the above described surface processing techniques it is possible to generate super- hard material products which have a surface profile indicative of the surface processing methodology used to generate the super-hard material products. In particular, the super-hard material products will have a surface profile which has a low surface roughness but which has a surface profile which deviates by a small but significant amount from a perfectly flat or precisely curved profile. Accordingly, a super-hard material product may be formed comprising: a Vickers hardness of no less than 2000 kg/mm ² ; a largest linear dimension of at least 10 mm, 20 mm, 40 mm, 60 mm, 80 mm, 100 mm, 120 mm, or 140 mm; a surface profile with a root mean square deviation from a smooth flat, convex, or concave surface profile, said root mean square deviation being at least 1 μιη, 3 μιη, 5 μιη, 7 μιη, or 10 μιη and no more than 50 μιη, 40 μιη, 30 μιη, or 20 μιη, and a surface roughness Ra of no more than 20 nm, 15 nm, 10 nm, 5 nm, 3 nm, or 1 nm.

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 appending claims.