GB2534192B

GB2534192B - Surface Processing Apparatus and Method

Info

Publication number: GB2534192B
Application number: GB1500732.1A
Authority: GB
Inventors: Xu Cigang; Sundaram Ravi; Watson Ray
Original assignee: Oxford Instruments Nanotechnology Tools Ltd
Current assignee: Oxford Instruments Nanotechnology Tools Ltd
Priority date: 2015-01-16
Filing date: 2015-01-16
Publication date: 2019-10-23
Anticipated expiration: 2035-01-16
Also published as: GB2534192A; GB201500732D0

Description

Surface Processing Apparatus and Method

Field of the Invention

The present invention relates to surface processing of a substrate, in particular to a surface processing apparatus used in various Chemical vapour deposition (CVD) processes conducted on a substrate.

Background

Thin film deposition on substrates is well known. There are a number of processes by which deposition can be enacted, such as by low pressure Chemical vapour deposition (LPCVD) through use of a (partial) vacuum, and plasma enhanced Chemical vapour deposition (PECVD) through use of a plasma source.

Such processes typically operate at substrate temperatures between 200°C and 800°C. However, higher temperature LPCVD is also well known, typically as an epitaxial growth process in the range 800°C to 1300°C.

The apparatus used for such processes can be broadly divided into two types: those used in hot wall processes and those used in cold wall processes. In cold wall process, there is a heated substrate in a processing chamber that is usually substantially cooler. The chamber is often in a parallel plate configuration, where the substrate sits on a heated table, facing a plate with an array of holes called a “showerhead”, through which reactant gases are distributed.

Each of the table and showerhead can form electrodes by being electrically conductive and connected to a suitable circuit. When one or both of the table and showerhead are electrodes, then the cold wall processing format is highly suitable for PECVD.

Altematively, a high-density plasma (HDP) source, either close-coupled or positioned remotely in the gas stream flowing to the chamber, such as an induction-coupled plasma (ICP), can be operatively connected to the processing chamber, making a high-density plasma Chemical vapour deposition (HDPCVD) or induction-coupled plasma Chemical vapour deposition (ICPCVD) configuration, with the substrate table either grounded or electrically biased. ln a cold wall processing format, heat can also be applied to the walls of the chamber or to the electrodes, to prevent condensation of vapours and to reduce cooling of the substrate on the heated electrode. Any such heating will usually be limited to a maximum of about 400°C, and more usually up to a maximum of 200°C, to make construction with the current standard engineering materiais and vacuum seals possible.

Sometimes it is desirable to heat a substrate to a temperature in the range of 800°C to 1200°C to enable LPCVD processes to create the material phases of interest for extremely thin layer deposition (including deposition of a single atomic layer on a substrate), or annealing of thin metal films to form metal dots for growing nanostructures on, such as carbon nanotubes. This can also require other processes to be carried out (such as pre and post plasma treatments).

Further, operating at such high temperatures (e.g. those above 800°C) means that the chamber components used for normal cold wall processing are vulnerable to thermal damage. It is therefore highly desirable to protect the chamber from the heat emitted from the heated table. ln order to avoid thermal damage, it is known to use radiation shields to manage the propagation of heat through a chamber. These also reduce thermal loss from the heated table, which in turn reduces the power required to maintain the table at the desired temperature.

The use of radiation shields is known in PECVD assemblies, but only for secondary purposes such as: limiting heat transport from the underside of the table as is disclosed in US 8,753,447 B2 and US 7926440 B1; protecting specific regions of the process chamber as is disclosed in US 6,106,625 and US 8778813 B2, or in lamp-heated systems, to manage the wafer temperature uniformity and protect the wafer back from deposition as is disclosed in US 5493987.

Further, in PECVD and HDPCVD chambers, having a heat shield between the heated table and a top electrode or plasma source prevents the plasma from reaching the substrate. There is also a prejudice against using such shields in heated deposition systems, because temperature cycling of chamber components held above the substrate can promote particle shedding from the shield, leading to contamination of the substrate.

It is also known to use a plate with holes of a suitable size located between the plasma source and the substrate in HDPCVD, as this modifies the distribution of plasma communicated to the substrate. Plasma etching can be carried out through a use of a similar arrangement, for example as disclosed in EP 1889946 B1. The plates used would also offer some heat shielding if a heated table were used, but would only partially prevent contamination of the chamber by any material evaporating from the substrate, which is understandably undesirable. ln an ideal configuration, a heat shield would be placed as close as possible to the heated table holding the substrate so that the maximum amount of thermal radiation is intercepted. Also, having small holes or no holes in the plate are desirable for minimising both thermal radiation from the table and chamber contamination by material evaporating from the substrate.

However, locating the shield in such close proximity to the heated table with such small holes creates an obstacle in the path of any gas used as a deposition médium, which will lead to a lack of uniformity in the film deposited on the substrate, especially in a plasma enhanced process, and a difficulty in installing the substrate in the correct position due to a lack of access between the shield and table restricting access to the upper surface of the table. As such this configuration will likely result in a badly located substrate upon which only a non-uniform pattern is able to be deposited. A further disadvantage is that forcing any plasma in a PECVD or HDPCVD process to follow such a path through small holes or round the shield will likely quench the species to be deposited from the plasma, inhibiting the processing of the substrate.

There is therefore a need for an arrangement that overcomes these disadvantages.

Summary of Invention

According to a first aspect of the invention, there is provided a surface processing apparatus for use in the surface processing of a substrate by at least one process selected from cold wall Chemical vapour deposition (CVD), plasma enhanced CVD (PECVD) low pressure CVD (LPCVD), Metalorganic CVD (MOCVD), High-density plasma CVD (HDPCVD), remote plasma CVD and/or atomic layer deposition, the surface processing apparatus comprising: a heatable mount with a surface on which a substrate is mountable; and a shield assembly including a shield for restricting the flow of heat away from the mount, the shield having one or more apertures for the transmission of material through the shield in use to a substrate on the mount, and a shield support that supports the shield, wherein the shield support is operable to adjust the distance between the shield and the surface of the mount.

The natural assumption is that a shield with one or more apertures large enough not to disrupt gas flow or to quench deposition materiais from a plasma is not sufficiently effective in diminishing the passage of heat away from the mount. However, we have found that surprisingly, the shielding effect offered is sufficient when the distance between the shield and the substrate is able to be adjusted so that an appropriate balance between all the factors can be achieved.

As such, the apparatus advantageously allows constraint on the passage of heat away from the mount whilst also allowing uniform deposition onto a substrate in a single implementation and suitable access for loading the substrate (usually a wafer) onto the surface of the mount.

The support may be operable to adjust the distance between the shield and the surface of the mount in a number of ways, such as using a lifting mechanism that is able to adjust the distance between shield and the surface of the mount to increase the height of the shield relative to the height of the mount. To enable the apparatus to be kept isolated from externai contaminants and to preserve the low pressure processing environment, a seal may be used, such as a sliding seal around a slidable ram or column. Typically however, the shield support includes at least one lifting mechanism actuable to adjust the height of the shield, each lifting mechanism having bellows adapted to form a seal around the lifting mechanism.

The reason for using a bellows instead of a sliding seal is that mechanical actuation is able to be provided to the support with a seal that is resilient to degradation, such as vacuum bellows by enclosing the moving parts and expanding to hold the part and contracting to release the part. If a sliding seal were to be used instead, there would likely be degradation in performance over time caused by particles from the deposition system becoming entrapped in the sliding seal, which is avoided when using bellows.

It may be desirable to be able to make minor adjustments to the position (e.g. height or lateral position) of the shield relative to the surface of the mount to ensure there is a suitable distance between the substrate and the shield and that the alignment of the one or more apertures of the shield and the substrate is correct. This can be achieved by physically adjusting the position of the shield on the support, though typically, at least one of the at least one lifting mechanisms has a wedge mechanism actuable to adjust the position of at least a part of the shield, allowing repeatable and precise movement of a moveable part of the shield, which reduces the likelihood of error when making such an adjustment.

Advantageously, a single actuation may set both the distance between the heat shield assembly and the substrate, and the degree of alignment of the one of more apertures in the shield (e.g. by adjusting the degree of alignment of one or more apertures in a first plate of the shield with one or more apertures in a second plate of the shield), through the use of a wedge mechanism.

The one or more apertures may take up any proportion of the shield as long as the shield is able to maintain its structurai integrity. Typically, each aperture provides a transparent area in a surface of the shield, and wherein the typical total transparent area in said surface provided by the one or more apertures is between 0% and 50% of the area of the surface.

We have found that having a total transparent area of less than 10% of the area of the surface of the shield viewed from the heated mount reduces the deposition rate in PECVD too much, and having a transparent area greater than 50% of the area of the surface of the shield allows too much heat to radiate through the shield. By having the total transparent area adjustable, this assists in striking the balance between a plasma treatment rate and radiation shielding. For example, when a PECVD process is used, having a total transparent area in the range 10% to 30% is preferred, when using CVD processes, which are limited by surface reactions, a smaller total transparent area can be used, preferably in the range 0% to 10%.

The shield may be any shape and/or size. For example, the area of shield in a plane parallel to the surface of the mount may be smaller than the area of the surface of the mount (i.e. may be contained within the edges of the surface). However, typically the shield extends at least to the edge of the surface on which the substrate is mountable, and may extend further (so as to have an area in a plane parallel to the surface equal to or large than the area of the surface).

When the area of the shield in a plane parallel to the surface is at least as large as the area of the surface of the mount, the transparent area (i.e. the area of the one or more apertures) can extend as close to the edge of the surface of the mount as possible (and in the case where the shield has a larger area in said plane than the surface, up to the edge and beyond it if desired). This allows the use of substrates of similar size to the surface and also to an aperture shape or pattern that maximises the use of the space available, providing the greatest opportunity for uniform deposition onto any substrate possible. If a substrate to be processed occupies only part of the surface of the heated mount, then preferably, the one or more apertures in the shield only occupy a corresponding region of the shield to a region on the surface of the mount in which the substrate is positioned, preferably with the one or more apertures extending laterally relative to the substrate by up to about 10mm beyond the edge of the substrate.

As the distance between the shield and the surface on which the substrate is mountable is adjustable, the distance between the shield and a deposition surface (i.e. the surface of a substrate onto which material is deposited in use) of substrate mounted on the surface in use can be set by a user or by some other means. A number of factors may determine the minimum and maximum distances between the shield and the deposition surface of the substrate. Such factors may be the range of movement of the support, the space required for mounting the substrate on the surface or the effect a particular distance has on the shielding effect or uniformity of deposition. Typically, the minimum distance between the shield and a deposition surface of a substrate when mounted on the surface in use is at least 1.0mm, preferably between 5.0mm and 12.0mm.

The minimum distance between the shield and a deposition surface may of course be the minimum distance from any point on the deposition surface to the shield (e.g. the base of the shield), or may be an average (such as the mean or some other average) distance from the deposition surface to the shield. The minimum distance of 1mm is given as a practical distance that ensures the shield does not touch the substrate. Maintaining a separation between the shield and deposition surface of the substrate can also mitigate the effect on the uniformity of deposition of gas-jetting at the one or more apertures where thermally driven reactions at the surface are the rate limiting step. This is because gas jets normally require a separation distance approximately equivalent to the separation of the centres of each aperture, as this allows the gas jets to overlap at the substrate due to divergence of the jet. ln a PECVD deposition process, the gap between the shield and a deposition surface of the substrate should be at least 5.0mm, and preferably between 5.0mm and 12.0mm, which gives optimum uniformity of material deposited on the deposition surface. ln a CVD process the same gap should be at least 1.0mm, and preferably between 5.0mm and 10.0mm. To avoid contamination of the substrate and to maintain the radiation shielding effect, the shield should not touch the substrate.

The shield may be made of a high melting point metal, alloy or dielectric, such as molybdenum, nickel, Inconel (RTM), alumina, aluminium nitride or boron nitride, or graphite or Silicon. This allows the shield to withstand the high temperatures producible by the mount.

The shield may be curved, domed or planar. Typically, the shield comprises at least one plate, wherein the one or more apertures are formed through the plate(s). Using a plate as a shield keeps the shield simple and ensures the one or more apertures are only formed in a single plane giving a more predictable deposition pattern.

The plate may be supported by the support and possibly by another support (structure) as well. Typically however, the plate is rigid, self-supporting and resistant to distortion when heated. This allows as few supporting elements as possible (such as the lifting structures) to be used to keep the amount of components and amount of materiais used to a minimum. For example, when the plate is rigid, self-supporting and resistant to distortion, only one lifting element can be used (although more than one can be used if wanted). Of course, some other form of support could also be used.

An example plate that could be used as (part of) the shield, is a 0.1 mm to 3.0mm thick molybdenum sheet as this is sufficiently rigid, self-supporting and resistant to distortion when heated. Further supports may be provided so that one or more thinner plates may be used, as each thinner plate will usually require support around its edge.

Altematively, the shield may comprise at least two plates, one above the other, wherein the one or more apertures are formed though each plate. This allows the shielding effect to be amplified whilst minimising the quenching effect by using as little material for the shield as possible. Of course, more plates can be used and/or may comprise other components.

As well as the distance between the shield and the surface on which the substrate is mounted being adjustable, the distance between each plate may be adjustable. Typically though, each plate is axially fixed relative to each other plate. This is advantageous as the number of moving parts in the shield support is kept to a minimum.

The gap between the plates (i.e. the distance between the two plates of each pair of adjacent plates) may be 0.1 mm or more, such as between 0.1 mm and 7mm, and preferably between 1.0mm and 3.0mm. This allows gas and/or plasma to pass between the plates without the passage being too restricted, while ensuring that separate plates do not touch each other.

The plates may each be fixed in position laterally with respect to each other. Typically however, at least one plate is moveable laterally and relative to the other plate(s). This allows the transparent area of the shield as a whole to be adjusted by laterally translating or laterally rotating at least one plate relate to the other plate(s). This is achieved as the degree of alignment (i.e. overlap) between the apertures in the laterally moveable plate(s) with the apertures in the other plate(s) will change on lateral movement of the moveable plate(s), modifying the transparent area of the shield as a whole.

The total transparent area may be adjustable. A higher transparency is desirable for PECVD processes, often conducted at lower temperature, and a lower transparency is desirable for CVD processes which are conducted at higher temperature where radiation shielding and contamination protection are more important.

Each aperture may be the same shape as each other aperture, or may be a different shape from one or more other apertures. Further, the size of each aperture may be the same as each other aperture or may be different from one or more other apertures. Nonetheless, there may be a particular ratio between the (average) size of each aperture and thickness of the plate in which the aperture is located. The ratio of a diameter of each aperture to the thickness of the plate may be in the range of 0.3 to 100.0.

Whilst other ratios are possible, typically each aperture in each plate has a diameter of 0.5 to 20 times the thickness of the respective plate. This ratio allows a suitable amount of charged plasma species to be transmitted to the substrate, as the higher the aspect ratio of the aperture diameter to the plate thickness, the higher is the fraction of charged species transmitted to a substrate.

When an aperture is circular, an example aperture size would be 2.50mm for a 0.17mm thick plate. Of course, other shapes, such as squares, triangles, pentagons, hexagons or other polygons can be used with a comparable dimension used instead of a diameter. The apertures are described as two dimensional shapes as, even though each aperture is three dimensional, due to the thickness of the plate in which they are located relative to the width of the apertures, each aperture is effectively two dimensional. A wide range of diameters is possible for each aperture. Typically though, each aperture in each plate has a diameter of 1.0mm to 5.0mm. This assists in ensuring adequate process rate of the deposition materiais on a substrate, especially when combined with the shield having a distance from the deposition surface of the substrate of 5.0mm to 12.0mm. For example, a large aperture diameter, such as a diameter of between about 1.0mm and 5.0mm, for each aperture assists in maintaining the deposition rate in a PECVD process.

Typically, each plate has a plurality of apertures provided in an array. Preferably, each plate has an array provided with apertures arranged in rows and columns, or in a hexagonal array so that each aperture is equidistant from its six nearest neighbours. A separation between nearest neighbours of about 3.25mm is typical, giving the array a pitch of about 3.25mm. Accordingly, each plate may have a plurality of apertures provided in an array, wherein the array has a pitch of approximately 3.25mm An altemative is that a distribution of apertures may be provided that compensates for non-uniformity in the deposition.

By the “pitch” of an array, we mean the distance from the centre of each element of the array to the centre of each adjacent element of the array.

We have found that having a plurality of apertures in each plate provided in an array with a pitch of approximately 3.25 mm assists in ensuring optimum uniformity of deposition materiais on the deposition surface of the substrate. This is especially the case when the apertures have a diameter in the range given above and the distance between the plate closest to the substrate in use and the deposition surface of the substrate complies with the minimums set above. For most uniform deposition, the minimum distance between the substrate and the shield should be greater than the pitch between apertures. This is because the uniformity of deposition is, at least in part, determined by the pitch in the array of apertures and the distance between the shield and the substrate, especially in the case of PECVD.

When plasma deposition processes are used, electrodes are needed in order to establish a plasma. Whilst electrodes may be placed anywhere around the environment in which the plasma is created, typically, at least a part of the shield is electrically conductive. This enables the shield to be an electrode suitable for use in establishing a plasma and to be connected to ground.

The whole shield may be electrically conductive, or altematively, a part of the shield (such as a plate) with the greatest distance from the surface of the mount is electrically conductive. This may be when a PECVD process is used that uses an RF-driven (Radio Frequency) top electrode. Altematively if the substrate table is an RF-driven electrode or is connected to a DC power supply, then preferably the lowest part of the shield is electrically conductive, to form the counter-electrode. The shield or at least a part of it (preferably at least any electrically conductive part) is resistant to the plasma chemistry employed to avoid the shield deteriorating.

It is difficult for the mount to instantly brought to the desired temperature, especially when the temperatures used are sufficiently high to allow CVD to be achieved. Therefore typically, the temperature of the mount is rampable up or down by 3 to 300 degrees centigrade (°C) per minute. By “rampable”, we mean able to be increased (or decreased) by a specific amount over a specific period. Preferably, a rate of change of temperature of at least 30°C per minute is achievable. This allows different process temperatures to be used sequentially without excessive heating and cooling times.

Further, the assembly is then suitable as a multi-functional apparatus capable of simple thermal annealing in a controlled atmosphere or in a vacuum; of thermally driven LPCVD processes, using gases or the vapour from liquid or solid precursors, such as Metalorganic Chemical Vapour Deposition (MOCVD) processes; and of PECVD or HDPCVD processes using gases, liquid or solid precursors, including remote and close coupled plasmas. Atomic layer deposition processes, including plasma enhanced processes, can also then be performed in the same environment.

Typically, the assembly may further comprise a chamber within which the mount and shield assembly are located. This allows any CVD process to be contained limiting the risk of contamination of the materiais used in the CVD process, or externai contamination.

The chamber may have a controller adapted to control the flow of one or more gases or vapours through the chamber. This allows a process to be tailored to a specific type or use.

For example, the controller may be adapted to control the flow of: hydrogen, for reducing surface oxides prior to layer formation; one or more gases for a plasma cleaning process, such as oxygen to remove carbon or a fluorine-bearing gas such as tetrafluoromethane (CF4) or sulphur hexafluoride (SF6); one or more carbon-bearing gases such as methane or acetylene for graphene or carbon nanotube growth; both a boron-bearing compound such as diborane and a nitrogen-bearing compound such as ammonia for hexagonal boron nitride growth; molybdenum and sulphur-bearing gases such as molybdenum chloride (e.g. M0CI5) , molybdenum hexacarbonyl (Mo(CO)6), hydrogen sulphide (H2S), Di-lsopropyl Sulphide (C6Hi4 S) for molybdenum disulphide growth; other precursors for transition metal dichalcogenide formation combining one or more of molybdenum, tungsten, niobium or tantalum with one or more of sulphur, selenium and tellurium; silane or dilute silane for silicon-containing films; an oxidiser such as oxygen or nitrous oxide (N2O) to form oxide compounds; a nitrogen bearing gas such as nitrogen or ammonia to form nitride compounds; and/or trimethyl or triethyl gallium and ammonia for gallium nitride MOCVD.

Other gases and vapours can be added according to the desired deposition process. It can be beneficiai to have separate gas feeds for different gases, especially to introduce high molecular weight compounds downstream from a high-density plasma source, to prevent excessive dissociation of such compounds distant from the intended deposition region. Different gas feeds can also be beneficiai for keeping gases separate that have a strong tendency to react with each other.

Typically, the apparatus further comprises a plasma source operatively connected to the chamber. This allows plasma to be introduced into/produced in the chamber.

As noted above, the shape, size and distribution/position of the one or more apertures may have an effect on the distribution of deposition materiais on the substrate in use. Typically, the physical form of the one or more apertures and/or the distribution of the one or more apertures is adapted to provide a predetermined processing pattern upon the deposition surface of the substrate. This allows reliable and repeatable distribution of deposition materiais on a substrate in use.

Typically, the surface processing apparatus is for use in the surface processing of a substrate by at least two processes selected from cold wall Chemical vapour deposition (CVD), plasma enhanced CVD (PECVD) low pressure CVD (LPCVD), Metalorganic CVD (MOCVD), High-density plasma CVD (HDPCVD), remote plasma CVD and/or atomic layer deposition in a single process sequence. This allows the same apparatus to be at least dual purpose, which makes it more useful, and allows a greater range of processes to be carried out whilst limiting the increase in the amount of equipment needed to perform those processes.

According to a second aspect, there is provided a method of surface processing for surface processing of a substrate by at least one process selected from cold wall Chemical vapour deposition (CVD), plasma enhanced CVD (PECVD) low pressure CVD (LPCVD), Metalorganic CVD (MOCVD), High-density plasma CVD (HDPCVD), remote plasma CVD and/or atomic layer deposition, the method comprising the steps: loading a substrate onto a surface of a heatable mount in a processing chamber; setting the distance between the surface and a shield adjacent to the surface to a predetermined distance; heating the substrate; performing one of the surface processing processes on the substrate by introducing gases and/or vapours into the chamber; reducing the temperature of the substrate; adjusting the distance between the surface and the shield to a distance that allows the substrate to be transferred; and unloading the substrate from the mount.

This allows constraint on the passage of heat away from the mount whilst also allowing uniform deposition onto a substrate in a single implementation and good access for loading the substrate (usually a wafer) onto the surface of the mount.

Typically, at least two processes selected from cold wall CVD, PECVD, LPCVD, MOCVD, HDPCVD, remote plasma CVD and/or atomic layer deposition are performed on the substrate in a single process sequence. Of course, each of the at least two processes may be performed at different points in the method depending on when it would be most appropriate to perform it taking into account mount temperature, shield position and total transparent area of the shield.

Typically, the shield may have one or more closable apertures, the method may further comprise the step of closing the apertures before heating the substrate. Closing the apertures before heating the substrate allows the environment around the mount to be protected from excess heat and from any material evaporating form the substrate during the heating process.

Also typically, the method may further comprise the step of opening the apertures after reducing the temperature of the substrate. This allows a more direct path between the substrate and the surrounding environment to be used for any post processing that may be desired.

Additionally, the chamber may be operatively connected to a plasma source, and the method may further comprise the steps of performing a plasma treatment process, such as hydrogen treatment, after loading the substrate onto the surface of the mount, and optionally performing post-processing plasma processing before unloading the substrate. This allows any contaminants to be removed from the various components reducing the risk of cross contamination of a substrate or component between processes. A number of mechanisms can be used to load the substrate onto the surface of the mount, such as robotic wafer loading system, or manually. Typically, the substrate is loaded onto the surface using a load lock mechanism. The substrate can also be unloaded using a load lock mechanism.

Use of a load lock mechanism is advantageous as it enables a substrate to be loaded into position while the wafer table is maintained close to its temperature of use. Parts of the table may be damaged if exposed to air while at elevated temperature.

The method may be executed using a surface processing apparatus comprising any combination of the features described above.

Brief description of figures

An example of a surface processing apparatus according to the invention will now be described with reference to the accompanying drawings, in which:

Figure 1 shows a sectional view of an example surface processing assembly;

Figure 2 shows a partial perspective view of an example surface processing assembly;

Figure 3 shows a process used to deposit material onto the surface of a substrate;

Figure 4 shows a graph of the relationship found between the shield to substrate distance against uniformity of deposition for an example surface processing assembly;

Figure 5 shows a Raman spectrum taken following use of an example surface processing apparatus to deposit material onto a substrate; and

Figure 6 shows a second Raman spectrum taken following use of another example surface processing apparatus to deposit material onto a substrate.

Detailed description

The apparatus described herein is used in processing a surface of a substrate in at least one processes selected from cold wall CVD, PECVD, LPCVD, MOCVD, HDPCVD, remote plasma CVD and/or atomic layer deposition. In particular, the apparatus is able to be used to control the uniformity and rate of deposition onto the substrate whilst also protecting other components used in the CVD, PECVD, LPCVD, MOCVD, HDPCVD, remote plasma CVD and/or atomic layer deposition single process sequence from the heat emitted from a mount by restricting the flow of heat away from the mount, and also allowing sufficient access before and after the processing to load and unload the substrate from the mount.

The apparatus is shown in Figure 1. This shows a lower section of a vacuum chamber 1. At least a partial vacuum is able to be formed inside the chamber, so the chamber 1 is a sealable enclosure.

The chamber 1 has a chamber wall 10 with an inlet 12 through which a gas, vapour or plasma may be supplied into the chamber from an externai source (not shown). There may also be an outlet (or the inlet may double as an outlet) to allow gases, vapour or plasma to be removed from the chamber. The inlet may be viewed as a plasma source as it can provide the raw materiais for forming a plasma. There may also be multiple inlets and/or outlets.

The chamber wall 10 surrounds a table or mount 2. The mount has at least an upper surface 20 on which a substrate (not shown) may be mounted. The mount is also able to be heated, which is achieved by the mount containing a heating element or some other heating means. Parts of the mount are able to be cooled by means of the passage of coolant through the mount. This is normally supplied from an externai source (not shown).

The mount 2 is heatable to at least 800°C. Because of this, the mount is not in direct contact with any inner surface of the chamber 1, but instead is held separate from (i.e. out of contact with) the chamber to avoid damaging the chamber.

So, instead of being placed against a wall of the chamber, the mount 2 is held above an internai base 14 of the chamber 1 by supports (not shown). The mount also has Service connections 22, such as those that control the heating of the mount, connected to it. These Service connections pass through the chamber wall 10 to a controller 24 that allows the temperature of the mount to be externally controlled. A shield assembly 4 is held above the mount 2 by two supports 3. The supports 3 are two pillars 36 that extend from the internai base of the chamber 1 to the same height above the internai base of the chamber or just above the upper surface 20 of the mount 2. Each pillar has a top 30 with an inwardly facing lateral projection 38 to which the shield assembly is attached.

Each support 3 is constructed of standard engineering materiais, such as stainless Steel, which has suitable strength and heat resistance properties, and is compatible with the gases used for the processes. Each support offers a degree of thermal isolation between the walls of the chamber and the shield, and in processes where plasma is used, each support offers an electrically conducting path to the chamber walls to enable a plasma excitation current to pass along each support.

The shield assembly can comprise one or more plates. By increasing the number of plates, greater radiation shielding is provided, but also greater attenuation of gases excited by plasma above the shielding. When there are multiple plates, it is preferable for the apertures through the shield assembly to be adjustable.

The shield assembly comprises three axially separated plates (a lower plate 4C, middle plate 4B and upper plate 4A), each of which is attached to the supports 3. As can be seen in Figure 2, the lower plate 4C and upper plate 4A are each attached to the supports 3 by posts 34, secured to the posts by wire binding and middle plate 4B is attached to a wedge mechanism 6. The plates may even each be formed as a single piece with at least part of one or more of the supports, or may be connected to the supports by another suitable means.

Each pillar 36 also contains a lifting mechanism 32 in the form of a column 320 to which the top 30 of the respective pillar is attached. The column is able to be raised and lowered, which in turn allows the shield assembly 4 to be raised and lowered.

To allow the column of each lifting mechanism 32 to be raised, the column 320 extends through the chamber wall 10. To ensure that the interior of the chamber 1 is isolated from the outside environment (as well as the reverse), the part of each column that is outside of the chamber is held in a Container 322, and a seal is formed around the respective column by a set of vacuum bellows 5.

The lifting mechanism is used to raise the shield to increase the distance to the surface to allow a substrate to be loaded onto the upper surface of the mount. Once a substrate is loaded, the lifting mechanism is then used to lower the shield to a particular distance from the upper surface of the substrate onto which material is to be deposited upon.

The column 320 is raised and lowered by a pneumatic piston. Other types of lifting mechanism such as stepper motors which provide continuously variable position control are also possible. Actuation is commanded by the control system (i.e. the controller 24) for the whole tool.

Instead of using a column, each support could be made of a number of concentric cylinders allowing each support to be telescopic which are driven from outside of the chamber and have an internai support structure if need be.

Figure 2 shows that the shield assembly 4 has the same shape as the mount 2, is circular and has three axially separated plates (an upper plate 4A, a middle plate 4B and a lower plate 4C) one above another. The plates have approximately the same diameter as the mount 2 above which they are located. This also shows the separation between the upper surface 20 of the mount and the lower plate 4C of the shield assembly 4 to be greater than the axial separation of the plates.

Each plate has multiple perforations or apertures 40 (only visible on the upper plate in Figure 2) arranged in a regular pattern in the centre of the respective plates 4A, 4B, 4C. The apertures each provide a transparent portion in the respective plate which gas, vapour or plasma is able to pass through and, in this case, have the same distribution pattern in each plate.

The apertures 40 are all circular and have approximately the same diameter.

The apertures 40 each provide a transparent area (from above) in the respective plate in which they are located. If only one plate were to be used for the shield, the apertures in that plate would cumulatively provide a total transparent area in the plate and therefore in the shield. However, as there are three plates in the example shown in Figure 2, the total transparent area of the shield is proportional to the degree of alignment of the apertures in the respective plates.

Due to the distribution of the apertures 40 on each plate 4A, 4B and 4C, each plate has a non-transparent rim 42 around its edge. Gas, vapour and plasma are therefore not able to pass through the respective plate at that point. This will cause gas, vapour or plasma passing through the shield assembly 4 to be concentrated over the centre of the mount and kept away from the edge of the mount.

Figure 2 shows a wedge mechanism 6 that forms part of the top 30 of the support 3 to which part of the shield apparatus 4 is connected. The wedge mechanism is attached to one of the three plates, which, in this case, is the middle plate 4B, and is able to move the middle plate 4B laterally relative to the lower plate 4C and upper plate 4A. Either a lateral movement or rotation is possible (e.g. through use of a fulcrum); although a lateral movement is preferred.

The wedge mechanism 6 includes a slot 62 in the side of one of the supports 3 with which a bearing 64, such as a roller, interacts. The bearing forms part of an arm 66 that is slidably engaged with the top 30 of the support for lateral movement, and to which at least one plate is attached (in this case the middle plate 4B).

The slot 62 has a flat surface inclined at an angle relative to the plates and the support 3. This means that, as a result of cooperation between the bearing 64 and the slot 62, as the top 30 is lowered (by lowering the column 320 to which it is attached) from an initial position, the arm moves laterally relative to the support and radially outwardly relative to the upper plate 4A and the lower plate 4C. As the top is retumed to its initial position, the arm is again moved laterally relative to the support and radially inwardly relative to the upper plate 4A and the lower plate 4C due to the cooperation between the bearing and the slot. The movement of the arm moves the plate to which it is attached (the middle plate 4B). The effect of the movement of the plate is detailed below. ln an initial position, the lower plate 4C, middle plate 4B and upper plate 4A have all their apertures fully aligned for maximum transparency.

As detailed above, the wedge mechanism 6 is able to move the middle plate 4B to a second position. If this movement is rotation, then in the second position, the apertures 40 in the plates 4A and 4C are rotated out of alignment with the apertures in the middle plate 4B. This reduces the total transparent area of the shield, as at least part of each aperture in the upper plate will be located over a non-transparent part of the lower plate (i.e. not over an aperture) upon rotation. When moved laterally, the same effect will occur, as the alignment between the apertures 40 in the upper plate 4A and the apertures in the middle plate 4B is broken.

Of course, the amount by which the wedge mechanism moves or rotates the upper plate 4A will determine how much of a reduction in the transparent area of the shield occurs. The process of rotating or laterally moving at least one of the plates is of course also reversible.

This change in transparent area can be varied by adjusting the amount of movement or rotation caused by the wedge mechanism. Indeed, when an appropriate aperture distribution is used, it is possible to shift the apertures 40 in the upper plate 4A a sufficient amount to ensure there is no overlap of the apertures in the upper plate with the apertures in the middle plate 4B. This will mean that the shield has no transparent area at all.

For example, when the apertures each have a diameter of 2.0mm and are distributed in an array with 2.5mm between each aperture (i.e. an array pitch of 4.5mm), when the upper plate is moved laterally by more than 2.0mm and less than 3.5mm, each aperture in the upper plate will no longer have any overlap with any aperture in the lower plate. Of course, the upper plate may be moveable by less than 2.0mm and more than 3.5mm meaning that the apertures in each plate will overlap.

Movement of the wedge mechanism causing at least one of the plates to move laterally relative to the other plate(s) can be independent of, or synchronous with, the raising and lowering of the shield. To be able to cause lateral movement of one of the plates independently of the plates being raised and lowered, this will usually mean that the slot 62 with which the bearing 64 interacts is able to move along at least part of the length of the support independently of the top 30 and column 320.

When the lateral movement of at least one plate is synchronous with the raising and lowering of the shield as in the arrangement shown in Figure 2, this means when the column 320 is lowered (lowering the upper plate 4A and lower plate 4C due to their connection to the column 320 by the post 34 and lateral projection 38 of the top 30, and lowering the middle plate 4B due its connection to the column by the wedge mechanism 6), the wedge mechanism is actuated, which causes the middle plate 4B to move horizontally (i.e. laterally) relative to the lower plate 4C and the upper plate 4A. This causes the apertures in three plates to become at least partially un-aligned.

The actuation of the wedge mechanism makes it is possible to control the alignment the apertures in each plate. For example, when the movement of the wedge mechanism is synchronous with the movement of the column, in a raised position, the holes are in a first alignment (for example, there is at least some direct view through the shields for PECVD processing), and in a lowered position the apertures in each plate are in a second alignment (for example, no direct transmission, for maximum heat shielding and minimum chamber contamination from the substrate).

Altematively, the degree of alignment can be controlled continuously (and independently) to provide all possibilities from full alignment to complete misalignment of the apertures in the plates at all positions of the plates. This is when the actuation of the wedge mechanism is independent of the movement of the column.

At least a part of the shield assembly is electrically conductive. This is usually the upper side of the upper most plate.

Figure 3 shows an example process where the apparatus described can be used to limit undesirable chamber contamination from material evaporating from the substrate when heated. It is well known that tools that are used to form graphene on copper become contaminated with copper that evaporates from the substrate.

Contamination can be limited by first loading the sample using a load lock mechanism, S101, with the mount at a temperature below that which would cause significant evaporation (e.g. below 800°C, preferably below 600°C). At this stage the apertures are at least partially aligned so there is a direct view through the shield.

When the substrate is being loaded, the distance between the shield and upper surface of the mount is sufficiently large to allow access to the upper surface of the mount, for example 9mm. Optionally, a plasma treatment step is performed such as a hydrogen plasma treatment S102 while the apertures are open. The shield is then lowered to a minimum distance from the deposition surface of the substrate, S103. ln this example, the wedge mechanism is actuated synchronously with the lowering of the shield, so the apertures are closed by lateral movement of the plates relative to each other.

The mount is then ramped up in temperature to the temperature suitable for conducting CVD, S104, and a CVD process is conducted, S105.

The substrate is then allowed to cool, S106, either through active cooling (e.g. raising the gas pressure in the chamber), or through passive cooling (such as allowing the substrate to cool naturally when the temperature of the mount is reduced).

Once cool (e.g. at a temperature of 800°C, preferably 600°C), optionally, the apertures in the shield are opened, S107, for example, by at least partially realigning the apertures in each plate. When the apertures are opened, a post-process plasma processing is performed, S108. The substrate is then unloaded, S109, after the shield assembly is raised by the lifting mechanism to provide suitable access to the upper surface of the mount.

The following are example processes conducted using the apparatus described above: PECVD processes

Two stainless Steel plates, one above the other, and with apertures aligned, were used in a standard cold wall PECVD procedure. A Silicon dioxide (SiO2) PECVD process was used. The process conditions used were: a temperature of 300°C, a rate of flow into the chamber of 320sccm (standard cubic centimetres per minute, equivalent to approximately 0.0053 litres/second) of SiH4(5%)/N2 (i.e. a mix of 5% Silane and 95% Nitrogen), a 1000sccm (approximately 0.0167ltr/s) flow of N2O and a 480sccm (approximately 0.0080ltr/s) flow of N2, all at a pressure of 1Torr (approximately 133.3Pa), with a 100W RF (Radio Frequency) input to generate the plasma, over a period of 5 minutes.

Two tests were performed of depositing material onto a standard 200mm diameter Silicon wafer. The first test used a shield assembly with a 254mm diameter stainless Steel plate with a thickness of 0.3mm and a uniform distribution of 3mm diameter apertures supported 7.5mm above the mount, with another 254mm diameter stainless Steel plate with a thickness of 0.3mm and a uniform distribution of 3mm diameter apertures located 1.5mm above the lower plate. The deposition rate loss was 76.6% compared to using no shield, and the deposition uniformity using the shield was ±9.1%.

The second test used a shield assembly with a 254mm diameter stainless Steel plate with a thickness of 0.3mm and a uniform distribution of 3mm diameter apertures located 5.8mm above the mount, and then another 254mm diameter stainless Steel plate with a thickness of 0.3mm and a uniform distribution of 3mm diameter apertures located 5.8mm above the lower plate. The deposition rate loss was 81.7% compared to using no shield, with a deposition uniformity when using the shield assembly of ±9.1%

The non-uniformity of deposition on the substrate becomes much higher at a substrate to lower shield gap. This was due to a tendency to print the shield aperture pattern on the substrate.

This can be relationship is exemplified in Figure 4, which shows that the variation in uniformity of deposition generally decreases as the distance between a single plate and the substrate is increased from 1mm to 10mm for a single plate with a thickness of 2mm and apertures, each with a diameter of 4.5mm, arranged in an array with 9mm pitch.

This demonstrates that deposition in a PECVD assembly is not prevented fully when using a shield assembly.

On the other hand, as a reduced plasma density reaches the substrate under the plates, due to quenching, this may favour the growth of materiais that are sensitive to plasma effect, including few-monolayer materiais such as graphene.

Deposition of Graphene on a substrate A two layer shield assembly similar to the one described above was used in a specialised cold wall PECVD arrangement with a high temperature mount. It was known that without the shield assembly, the surface of the showerhead used to emit gases into the chamber would begin to degrade when the mount was at a temperature above approximately 1100°C, in spite of significant cooling applied to the rear of the showerhead assembly, and good thermal conduction paths between the showerhead face and the cooling.

When the shield assembly was fitted, it was found that it was possible to raise the temperature of the mount to approximately 1150°C, which is sufficient to melt copper foils (which have a melting point of approximately 1085°C) resting on a molybdenum carrier plate on the mount, without damaging the showerhead.

The power required to heat the mount in this arrangement to 1100-1200°C was around 5.5kW (in a vacuum) without the shield assembly, compared to around 4.5kW with the shield assembly fitted. When the distance between the mount and the shield assembly is at the lower end of the range of 1 to 10mm, less power is needed (i.e. the smaller the distance between the mount and the shield assembly, the smaller the amount of power is that is required to heat the mount), as there is reduced heat loss from the mount. For the same reason, when the apertures were closed, also, less power is needed.

Using this hardware, with a gap of 9mm between the substrate and the shield, we were able to form graphene on nickel foils. The foils were placed on a molybdenum carrier plate and placed on the mount using a load lock mechanism. The table was heated above 1000°C in a vacuum, and a mixture of methane and hydrogen gas admitted with a pressure below 1Torr (approximately 133.3Pa). The gas flow was stopped and the mount cooled to below 600°C before unloading the carrier. The foils were analysed then by Raman spectroscopy and showed the characteristic signature of graphene, which can be seen in Figure 5.

Deposition of hexagonal boron nitride on a substrate

The same assembly as used to deposit graphene on nickel was used to deposit few-layer hexagonal boron nitride, another ‘two dimensional’ (2D) material onto nickel foils.

Nickel foils were placed on a boron nitride carrier plate and introduced to a PECVD chamber using a load lock. The mount was heated to above 1000°C and the foils were annealed in hydrogen for 10 minutes. A mixture of diborane and ammonia gas in flow ratio 1:9 at a pressure of 150mTorr (approximately 20.0Pa) was then introduced for 10 minutes. The mount was then cooled to 800°C and the carrier unloaded. The distance of the shield assembly from the carrier plate was about 9mm with the apertures were closed.

The foils were analysed using Raman spectroscopy, the result of which is shown in Figure 6. This shows the characteristic peak of hexagonal boron nitride.

There is an absence of significant peaks visible in the 1200 to 1300 cm'1 (cmA-1) range in Figure 6 from the cubic phase (c-BN), carbonated phase (Bx CXNX) and BN soot. This means that there is little unwanted co-deposition of the cubic phase, carbon contaminated phase orfrom amorphous Boron-Nitride (BN) soot, showing that the process can be well controlled with minimal contamination from unwanted phases and from carbon.

Claims

1. A surface Processing apparatus for use in the surface Processing of a substrate by at least one process selected from cold wall Chemical vapour deposition (CVD), plasma enhanced CVD (PECVD) low pressure CVD (LPCVD), Metalorganic CVD (MOCVD), High-density plasma CVD (HDPCVD), remote plasma CVD and/or atomic layer deposition, the surface Processing apparatus comprising: a heatable mount with a surface on which a substrate is mountable; and a shield assembly including a shield for restricting the flow of heat away from the mount, the shield having one or more apertures for the transmission of material through the shield in use to a substrate on the mount, and a shield support that supports the shield, wherein the shield support is operable to adjust the distance between the shield and the surface of the mount.

2. The surface Processing apparatus according to claim 1, wherein the shield support includes at least one lifting mechanism actuable to adjust the height of the shield, each lifting mechanism having bellows adapted to form a seal around the lifting mechanism.

3. The surface Processing apparatus according to claim 2, wherein at least one of the at least one lifting mechanisms has a wedge mechanism actuable to adjust the position of at least a part of the shield.

4. The surface Processing apparatus according to any one of claims 1, 2 or 3, wherein each aperture provides a transparent area in a surface of the shield, and wherein the total transparent area in said surface provided by the one or more apertures is greater than 0% and up to 50% of the area of the surface.

5. The surface Processing apparatus according to any one of the preceding claims, wherein the shield extends at least to the edge of the surface on which the substrate is mountable.

6. The surface Processing apparatus according to any one of the preceding claims, wherein the minimum distance between the shield and a deposition surface of a substrate when mounted on the surface in use is at least 1.0mm.

7. The surface processing apparatus according to claim 6, wherein the minimum distance between the shield and the deposition surface of the substrate when mounted on the surface in use is between 5.0mm and 12.0mm.

8. The surface processing apparatus according to any one of the preceding claims, wherein the shield comprises at least one plate, wherein the one or more apertures are formed through the plate(s).

9. The surface processing apparatus according to any one of the preceding claims, wherein the shield comprises at least two plates, one above the other, wherein the one or more apertures are formed though each plate.

10. The surface processing apparatus according to claim 8, wherein each plate is axially fixed relative to each other plate.

11. The surface processing apparatus according to any one of claims 8 or 9, wherein at least one plate is moveable laterally and relative to the other plate(s).

12. The surface processing apparatus according to any one of claims 7 to 10, wherein each aperture in each plate has a diameter of 0.3 to 100.0 times the thickness of the respective plate.

13. The surface Processing apparatus according to claim 11, wherein each aperture in each plate has a diameter of 1.0mm to 5.0mm.

14. The surface Processing apparatus according to any one of the preceding claims, wherein at least a part of the shield is electrically conductive.

15. The surface Processing apparatus according to any one of the preceding claims, further comprising means for varying the temperature of the mount.

16. The surface Processing apparatus according to claim 15, wherein the means for varying the temperature of the mount is configured to ramp the temperature up or down by at least 30 degrees centigrade (°C) per minute.

17. The surface Processing apparatus according to any one of the preceding claims, further comprising a chamber within which the mount and shield assembly are located.

18. The surface Processing apparatus according to claim 15, wherein the chamber has a controller adapted to control the flow of one or more gases or vapours through the chamber.

19. The surface Processing apparatus according to claim 15 or claim 16, further comprising a plasma source operatively connected to the chamber.

20. The surface Processing apparatus according to claim 17, wherein the physical form of the one or more apertures and/or the distribution of the one or more apertures is adapted to provide a predetermined Processing pattern upon the deposition surface of the substrate.

21. The surface Processing apparatus according to any one of the preceding claims, wherein the surface Processing apparatus is for use in the surface processing of a substrate by at least two processes selected from cold wall Chemical vapour deposition (CVD), plasma enhanced CVD (PECVD) low pressure CVD (LPCVD), Metal organic CVD (MOCVD), High-density plasma CVD (HDPCVD), remote plasma CVD and atomic layer deposition in a single process sequence.

22. A method of surface processing for surface processing of a substrate by at least one process selected from cold wall Chemical vapour deposition (CVD), plasma enhanced CVD (PECVD) low pressure CVD (LPCVD), Metalorganic CVD (MOCVD), High-density plasma CVD (HDPCVD), remote plasma CVD and/or atomic layer deposition , the method comprising the steps: loading a substrate onto a surface of a heatable mount in a processing chamber; setting the distance between the surface and a shield adjacent to the surface to a predetermined distance; heating the substrate; performing one of the surface processing processes on the substrate by introducing gases and/or vapours into the chamber; reducing the temperature of the substrate; adjusting the distance between the surface and the shield to a distance that allows the substrate to be transferred; and unloading the substrate from the mount.

23. The method according to claim 22, wherein at least two processes selected from cold wall CVD, PECVD, LPCVD, MOCVD, HDPCVD, remote plasma CVD and atomic layer deposition are performed on the substrate in a single process sequence.

24. The method according to claim 22 or 23, wherein the shield has one or more closable apertures, the method further comprising the step of closing the apertures before heating the substrate.

25. The method according to claim 24, further comprising the step of opening the apertures after reducing the temperature of the substrate.

26. The method according to any one of claims 22 to 25, wherein the chamber is operatively connected to a plasma source, the method further comprising the steps of performing a plasma treatment process after loading the substrate onto the surface of the mount.

27. The method according to claim 26, wherein the plasma treatment process is hydrogen cleaning.

28. The method according to claim 26 or claim 27, wherein post-processing plasma Processing is performed before unloading the substrate.

29. The method according to any one of claims 22 to 28, wherein the substrate is loaded onto the surface using a load lock mechanism.

30. The method according to any one of claims 22 to 29, wherein the method is executed using a surface Processing apparatus according to any one of claims 1 to 21.