Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
A FILTER FOR A DEPOSITION PROCESS, RELATED METHODS, AND PRODUCTS THEREOF
Document Type and Number:
WIPO Patent Application WO/2019/197512
Kind Code:
A1
Abstract:
An assembly for use in a physical vapour deposition process, the assembly comprising: a filter having a body including a plurality of apertures therethrough; and a shield configurable to selectively cover a subset of the plurality of apertures.

Inventors:
JAIN, Himanshu (Inorganic Chemistry Laboratory, University of OxfordSouth Parks Road, Oxford Oxfordshire OX1 3QR, OX1 3QR, GB)
EDWARDS, Peter (Inorganic Chemistry Laboratory, University of OxfordSouth Parks Road, Oxford Oxfordshire OX1 3QR, OX1 3QR, GB)
Application Number:
EP2019/059196
Publication Date:
October 17, 2019
Filing Date:
April 11, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
OXFORD UNIVERSITY INNOVATION LIMITED (Buxton Court, 3 West Way Botley, Oxford Oxfordshire OX2 0JB, OX2 0JB, GB)
International Classes:
C23C14/04; C23C14/08; C23C14/34; C23C14/56; H01J37/34
Foreign References:
US20140034489A12014-02-06
US20140374250A12014-12-25
US3352282A1967-11-14
US20130149868A12013-06-13
Attorney, Agent or Firm:
BARKER BRETTELL LLP (100 Hagley Road, Edgbaston Birmingham, West Midlands B16 8QQ, B16 8QQ, GB)
Download PDF:
Claims:
CLAIMS

1. An assembly for use in a physical vapour deposition process, the assembly comprising:

a filter having a body including a plurality of apertures therethrough; and a shield configurable to selectively cover a subset of the plurality of apertures.

2. An assembly according to claim 1, wherein the shield includes a cover portion and an opening, the cover portion and the opening being alignable with the subset of the plurality of apertures so as to selectively cover or uncover the subset of the plurality of apertures.

3. An assembly according to claim 1 or claim 2, wherein the shield includes a shroud for channelling flux particles through the uncovered subset of the plurality of apertures, optionally wherein the shroud extends from the shield towards a first face of the body of the filter or away from the body of the filter.

4. An assembly according to any preceding claim, wherein the filter includes at least one substrate holder for accommodating at least one substrate across a portion of at least one aperture, optionally wherein the or each substrate holder is provided on a distal face of the body.

5. An assembly according to claim 4, wherein the or each substrate holder is configured to support the at least one substrate substantially perpendicular to an axis of symmetry of the respective aperture.

6. An assembly according to any preceding claim, wherein the shield is rotatable relative to the filter for sequential covering and uncovering of subsets of the plurality of apertures.

7. An assembly according to any preceding claim, wherein the plurality of apertures are angularly-spaced about the body, and/or wherein the plurality of apertures are equally-spaced about the body, and/or wherein the apertures are arranged with rotational symmetry about a central axis of the body.

8. A method of implementing a physical vapour deposition process using a physical vapour deposition machine containing an assembly comprising a filter having a body including a plurality of apertures therethrough and a shield configurable to selectively cover at least one of the plurality of apertures, the method comprising: creating a vacuum or partial vacuum in the physical vapour deposition machine;

arranging the shield to cover a subset of the plurality of apertures;

carrying out a first physical vapour deposition process;

arranging the shield to cover a different subset of the plurality of apertures; carrying out a second physical vapour deposition process;

wherein the first and second physical vapour deposition processes are carried out without releasing the vacuum or partial vacuum.

9. A filter, e.g. a flux collimator, for use in a physical vapour deposition process where a flux containing flux particles of one or more chemical species is deposited on to a substrate, the filter comprising:

a body having an aperture therethrough;

the aperture being configured to allow passage of flux particles travelling in a direction perpendicular to the substrate, whilst restricting passage of flux particles travelling in directions non-perpendicular to the substrate.

10. A filter according to claim 9, wherein the aperture includes at least one angled sidewall.

11. A filter according to claim 9 or claim 10, wherein the aperture has a longitudinal axis that is angled relative to a plane of the body.

12. A filter according to any of claims 9 to 11, wherein the aperture has a main bore, optionally wherein the main bore has a substantially-constant cross-section.

13. A filter according to claim 12, wherein the aperture includes a counter-bore that is larger than the main bore.

14. A device structure comprising:

a substrate; a thin-film layer; and

an intermediate layer interposed between the substrate and the thin-film layer, the intermediate layer having a density distinct from that of the thin-film layer. 15. A device structure according to claim 14, wherein the intermediate layer is a subplantation layer

16. A device structure according to claim 14 or claim 15, wherein the substrate is formed of a first material system, the thin-film layer is formed of a second material system, and the intermediate layer is formed of a combination of the first material system and the second material system.

17. A device structure according to any of claims 14 to 16, wherein the intermediate layer comprises or consists essentially of a functional material.

18. A device structure according to any of claims 14 to 17, wherein the thin-film layer comprises, or consists essentially of, a functional material.

19. A device structure according to claim 17 or claim 18, wherein the functional material comprises, or consists essentially of, an oxide.

20. A device structure according to any of claims 17 to 19, wherein the functional material comprises a binary, tertiary, or ternary materials system. 21. A device structure according to any of claims 17 to 20, wherein the functional material comprises one or more of indium, gallium, zinc, silicon, tin and/or oxygen.

22. A device structure according to any of claims 14 to 21, wherein the intermediate layer has a thickness of up to or at least 100 A, up to or at least 200 A, up to or at least 300 A, up to or at least 400 A, up to or at least 500 A, up to or at least 800 A, or up to or at least 1000 A.

23. A device structure according to any of claims 14 to 22, wherein the thin-film layer has a thickness of up to or at least 500 A, and/or up to or at least lOpm.

24. A method of producing a device structure having a substrate, a functional thin- film layer, and a subplantation layer interposed between the substrate and the thin- film layer, the subplantation layer having a density distinct from that of the thin-film layer, the method comprising:

a single physical vapour deposition step in which a flux containing flux particles of one or more chemical species is deposited on the substrate, wherein flux particles travelling in a direction perpendicular to the substrate are preferentially allowed to impinge on the substrate, whilst flux particles travelling in directions nonperpendicular to the substrate are restricted from impinging on the substrate, thereby causing formation of the subplantation layer on the substrate.

25. A physical deposition apparatus comprising:

a physical vapour deposition machine; and

an assembly according to any of claims 1 to 7 and/or a filter according to any of claims 9 to 13.

Description:
A FILTER FOR A DEPOSITION PROCESS, RELATED METHODS, AND

PRODUCTS THEREOF

The present invention relates to a filter for a physical vapour deposition process, for example a sputtering process, and use of such a filter. The present invention also relates to a filter in the form of a flux collimator, a method of using such a flux collimator in a physical vapour deposition process, and a device structure, which may comprise a thin-film formed through such a method.

Metal oxide thin-films composed of indium oxide (In 2 0 3 , also known as indium sesquioxide), tin oxide (Sn0 2 ), zinc oxide (ZnO), or their mixtures, are at the heart of important current electronic or display technologies, both at the micro-scale 'device' level, as well as at the macro-scale 'gadget' level. An instance of a micro-scale application is use of the ternary material system (In, Ga, Zn, O) for fashioning the channel of the latest generation of thin-film transistors; whilst an instance of a macroscale application are thin-films of the (In, Sn, O) system which are, due to their transparency and conductivity, widely used to enable a 'touchscreen' human-machine interface for any number of electronic gadgets available commercially.

Wide commercial success for the applications mentioned above has been, in the main, limited to applications in which the thin-film is disposed upon a flat and rigid substrate. This is because the longevity of electrical or optical performance of a thin- film is underwritten first by its mechanical integrity. This situation may be considered alongside the wide range of identified potential applications which may be serviced by or which likely will emerge upon the availability of thin-films that can withstand at least moderate bending (at least once, or repeated). In the design and manufacture of devices comprising thin films deposited on flexible substrates, film adhesion and/or film integrity considerations may be important, e.g. nearly as important, as important, or more important than electrical or optical considerations.

The difference in the coefficients of thermal expansion between the substrate and the film, the physical arrangement of atoms at the two surfaces facing the common interface, and the chemistry possible between the same atoms during and postdeposition, are among the intrinsic - as opposed to preparative protocol dependent - factors upon which depends substrate-film adhesion. To aid the adhesion, it is not uncommon to deposit first a (thin) transition layer upon the substrate before deposition of the thin-film proper. Evidently, the preparative steps required to produce such mediate layers introduce process and chemistry complexity, adding cost towards the development and day-to-day performance of deposition protocols.

It is an object of the present invention to alleviate or overcome one or more of the issues noted above.

According to a first aspect, there is provided an assembly for use in a physical vapour deposition process, the assembly comprising:

a filter having a body including a plurality of apertures therethrough; and a shield configurable to selectively cover a subset of the plurality of apertures.

The shield may include a cover portion and an opening, the cover portion and the opening being alignable with the subset of the plurality of apertures so as to selectively cover or uncover the subset of the plurality of apertures.

The shield may include a shroud for channelling flux particles through the uncovered subset of the plurality of apertures.

The shroud may extend from the shield towards a first face of the body of the filter or away from the body of the filter.

The filter may include at least one substrate holder for accommodating at least one substrate across a portion or at least one aperture.

The or each substrate holder may be provided on a distal face of the body.

The or each substrate holder may be configured to support the at least one substrate substantially perpendicular to an axis of symmetry of the respective aperture.

A top view outline of each aperture may have a radial balance and/or symmetry.

The filter and/or assembly may have an axis of rotational symmetry. The shield may be rotatable relative to the filter for sequential covering and uncovering of subsets of the plurality of apertures.

The plurality of apertures may be angularly- spaced about the body.

The plurality of apertures may be equally-spaced about the body.

The apertures may be arranged with rotational symmetry about a central axis of the body.

According to a second aspect, there is provided a method of implementing a physical vapour deposition process using a physical vapour deposition machine containing an assembly comprising a filter having a body including a plurality of apertures therethrough and a shield configurable to selectively cover at least one of the plurality of apertures, the method comprising:

creating a vacuum or partial vacuum in the physical vapour deposition machine;

arranging the shield to cover a subset of the plurality of apertures;

carrying out a first physical vapour deposition process;

arranging the shield to cover a different subset of the plurality of apertures; carrying out a second physical vapour deposition process;

wherein the first and second physical vapour deposition processes are carried out without releasing the vacuum or partial vacuum.

It is therefore possible to implement a high-throughput deposition process as the process conditions can be changed without release of the vacuum within the machine. Thus, multiple experiments on any given material system may be carried out in a shortened space of time.

Further physical vapour deposition processes may be implemented, and the processing conditions may be changed between each physical vapour deposition process. A total of three, four, five, or any other number of processes may be carried out without release of the vacuum or partial vacuum, optionally with a corresponding number of changes in process conditions. According to a third aspect, there is provided a filter, e.g. a flux collimator, for use in a physical vapour deposition process where a flux containing flux particles of one or more chemical species is deposited on to a substrate, the filter comprising:

a body having an aperture therethrough;

the aperture being configured to allow passage of flux particles travelling in a direction substantially perpendicular to the substrate, whilst restricting passage of flux particles travelling in directions substantially non-perpendicular to the substrate.

The aperture may include at least one angled sidewall.

The aperture may have a longitudinal axis that is angled relative to a plane of the body.

The aperture may have a main bore, and the main bore may have a substantially- constant cross-section.

The aperture may include a counter-bore that is larger than the main bore.

The filter may include a plurality of apertures which may be arranged in pairs or groups.

The plurality of apertures may be angularly- spaced about the body.

The plurality of apertures may be equally-spaced about the body.

The apertures may be arranged with rotational symmetry about a central axis of the body.

According to a fourth aspect, there is provided a device structure comprising:

a substrate;

a thin-film layer; and

an intermediate layer interposed between the substrate and the thin-film layer, the intermediate layer having a density distinct from that of the thin-film layer.

The intermediate layer may be a subplantation layer. The substrate may be formed of a first material system, the thin- film layer may be formed of a second material system, and the intermediate layer may be formed of a combination of the first material system and the second material system.

The intermediate layer may comprise or consist essentially of a functional material.

The thin-film layer may comprise or consist essentially of a functional material.

The functional material of the intermediate layer and the thin-film layer may be the same functional material.

The functional material may comprise, or consist essentially of, an oxide such as a transparent conducting oxide.

The functional material may comprise a binary, tertiary, or ternary materials system. The functional material may comprise one or more of indium, gallium, zinc, tin, silicon, and/or oxygen.

The functional material may comprise a materials system comprising or consisting essentially of: In, Si, Zn and O; In, Ga, Zn and O; In, Zn and O; In, Ga and O; In, Sn and O; In and O; Sn and O; or Zn and O.

The intermediate layer, e.g. subplantation layer, may have a thickness of up to or at least 100 A, up to or at least 200 A, up to or at least 300 A, up to or at least 400 A, up to or at least 500 A, up to or at least 800 A, and/or up to or at least 1000 A.

The thin-film layer may have a thickness of up to or at least 500 A, and/or up to or at least 10 pm.

The substrate may be flexible.

The device structure may be at least partially transparent.

According to a fifth aspect, there is provided a method of producing a device structure having a substrate, a functional thin-film layer, and a subplantation layer interposed between the substrate and the thin- film layer, the subplantation layer having a density distinct from that of the thin- film layer, the method comprising:

a single physical vapour deposition step in which a flux containing flux particles of one or more chemical species is deposited on the substrate, wherein flux particles travelling in a direction substantially perpendicular to the substrate are preferentially allowed to impinge on the substrate, whilst flux particles travelling in directions substantially non-perpendicular to the substrate are restricted from impinging on the substrate, thereby causing formation of the subplantation layer on the substrate.

According to a sixth aspect, there is provided a physical deposition apparatus comprising:

a physical vapour deposition machine; and

an assembly according to the first aspect and/or a filter according to the third aspect.

The physical vapour deposition machine may be a sputtering machine.

According to a seventh aspect, there is provided an implement for use in a film deposition process, comprising:

at least one aperture.

The implement may comprise a means to accommodate at least one substrate across a portion of the at least one aperture. The substrate may be accommodated on a distal side of the implement.

Each substrate may be accommodated in an individually controllable orientation.

The at least one substrate may be disposable substantially perpendicular to an axis of symmetry through the corresponding aperture.

A top view outline of the aperture may have radial balance and/or symmetry.

The implement may possess an axis of rotational symmetry. The implement may comprise a means to block passage of a portion of a given flux towards one or more apertures.

The implement may comprise a means to permit passage of a portion of a given flux towards one or more apertures.

The implement may comprise a means to block travel, through at least one aperture, of a portion of the incoming flux.

The implement may comprise a means to permit travel, through at least one aperture, of only a portion of the incoming flux.

According to an eighth aspect, there is provided a method to deposit a film, comprising the steps of:

providing an implement comprising at least one aperture and a first means to accommodate at least one substrate across at least a portion of each aperture, and a second means to permit travel, through at least one aperture, of a portion of an incoming flux;

accommodating, via a first means, at least one substrate across at least a portion of at least one aperture;

obtaining a set of process conditions;

obtaining an incoming flux created on the basis of the process conditions; obtaining deposition of a filtered flux, in the form of a film, on at least a portion of at least one substrate;

wherein the filtered flux is the portion of the incoming flux permitted to travel to the substrate by the second means.

The filtered flux may be substantially orthogonal to the substrate.

According to a ninth aspect, there is provided a method to fabricate a subplantation layer, comprising the steps of:

providing an implement comprising an aperture and a first means to accommodate a substrate across the aperture, and a second means to permit travel, through the aperture, of a portion of an incoming flux;

accommodating, via a first means, a substrate across the aperture; obtaining a set of process conditions;

obtaining an incoming flux created on the basis of the process conditions; obtaining deposition of a filtered flux, in the form of a film, on the substrate; wherein the filtered flux is substantially orthogonal to the substrate.

Unless mutually exclusive, or specifically described otherwise, any aspect may include any features described in relation to any other aspect, without limitation.

Non-limiting embodiments will now be described in detail with reference to the Figures, in which:

Figure 1 is a cube-plot motif representation corresponding to the designed experiment of Table 1, wherein, HJ27, HJ11, HJ23, HJ3, HJ7, HJ35, HJ15, HJ31, and HJ19 refer to individual thin-films;

Figure 2 is a plot of X-ray diffraction (XRD) data of thin-films obtained from the designed experiment of Table 1;

Figures 3a to 3d are Kiessig Fringes Profiles (KFPs) of thin-films referred to in Table 1;

Figures 4a and 4b show the structure of single layer thin-films and bilayer thin-films, respectively;

Figures 5a and 5b show thicknesses and densities, respectively, of the single layer samples corresponding to the back face of the cube-plot of Figure 1;

Figures 6a to 6d shows thicknesses (Figures 6a and 6c) and densities (Figures 6b and 6d) of the thin-film layer (TFL) and subplantation layer (SPL) respectively of the bilayer samples corresponding to the centre-point and front face of the cube-plot of Figure 1;

Figure 7a shows variation of length scales (roughness and thickness) and density of the SPL with respect to ratio of the same length scales respectively; Figure 7b shows variation of the difference between the thickness and the roughness of the SPL, and of SPL density relative to the substrate material (CEXG, Corning Eagle Glass XG), with respect to ratio of the same length scales respectively. RE power (RFP);

Figure 8 shows roughness of the TFL of the single layer samples corresponding to the back face of the cube-plot of Figure 1;

Figures 9a and 9b show roughness of the TFL and SPL, respectively, of the bilayer samples corresponding to the centre-point and front face of the cube- plot of Figure 1;

Figure 10a shows the electron number density for the thin-films having nominal composition InioSiiZnisC^;

Figure 10b shows the electron number density of the bilayer samples relative to that of the crystalline phase of In 2 SioZn306;

Figure 10c shows the relative (mass) density of the bilayer samples relative to that of the crystalline phase of In 2 SioZn306;

Figures 11a and lib show an assembly comprising a sputter flux collimator (SFC) and shield in perspective and plan views, respectively;

Figures 12a and 12b show the SFC of Figure l la in reverse plan view and cross-sectional view, respectively;

Figure 13 shows the shield of Figure 1 la in plan view; and

Figure 14 depicts the assembly of Figure l la in situ within a sputtering machine.

In RF sputtering, the RF Power (RFP), Process Gas Pressure (PGP), Oxygen Percent (OP) (in the process gas), temperature of the substrate, and substrate type (single crystal or amorphous) are five important process conditions via which it is possible to control the microstructure of a thin- film that forms on the substrate. In the presently- described embodiment, thin-films were deposited at room-temperature (with no external heating of the substrate) on to (amorphous) glass substrates; fixing thus the last two of the five aforementioned process conditions. The material from which the substrate is comprised in the present embodiment is Corning Eagle XG Glass.

A two- level three factor full factorial with centre-point experiment design was employed to efficiently investigate/discover the region of the available process conditions space where a subplantation layer is obtainable. The sputtering kit / machine used was model PVD75 by Kurt J. Lesker Company. The experiment design is formally listed in terms of Coded Units (CUs) in Table 1, embedded below. The cube-plot motif as shown in Figure 1, naturally showcases the structure of the experiment design, and will be used for further discussion. All samples produced in this study were produced via a radio frequency (RF) sputter deposition that lasted for the same duration (25 minutes).

Run Standard Process conditions (CL’) Sample Single or Bi

order order RI P PC IP OP identifier Layer

1 5 - - + HJ3 BL

2 9* 0 0 0 Hi? BL

3 2 + Hi l l BL

4 4 + + HJ 15 SL

5 7 - + + HJ 19 SL

6 6 + - + HJ23 BL

7 I - - HJ27 BL

8 g + + + HJ31 SL

9 3 - + HJ35 SL

Table 1 : The two-level three- factor designed experiment employed for fabricating thin- films of material InioSiiZnisC^ (see also Figure 1). To frustrate aliasing, the“standard order” was randomized to obtain a“run order”. The substrate was CEXG (with known surface roughness 5-10 A). The run marked * is the center-point run. See Table 2 for the interconversion of the process parameters, namely RFP, PGP, and OP, between the CU and RWU.

The samples were produced by sputtering a target having nominal composition InioSiiZni onto a substrate of Corning ® Eagle XG ® (CEXG), an example of an alkaline earth boro-aluminosilicate glass. Thus, the thin-films produced were of the ternary (In, Si, Zn, O) system. A PANalytical X’Pert Pro X-ray diffractometer was used to obtain the X-Ray Diffraction (XRD) profiles of the samples (see Figure 2). A Rigaku Model SmartLab X-ray diffractometer was used to obtain the Kiessig Fringes Profiles (KFPs) via small angle X-ray Reflectometry (XRR), and the profiles were analysed via the Rigaku GlobalFit 2.0 Integrated Thin-film Analysis Software, the results of which can be seen in Figure 3.

A Fourier Transform (FT) (shown at Figure 3(c) and Figure 3(d) respectively) was taken of the KFPs (shown at Figure 3(a) and Figure 3(b)). Quite unexpectedly - since all samples had been obtained via a single deposition step - two peaks were observed in the FT of the KFP of certain samples. The two peaks imply the existence of three interfaces, and therefore of two layers; that is, the samples in question had a bilayer structure. The nine samples listed in Table 1 were thus segregated into two groups corresponding to a single layer structure (Figures 3(a) and 3(c)) and a bilayer structure (Figures 3(b) and 3(d)).

Guided by the FTs, modelling of the KFPs of the samples based upon the structures shown in Figure 4(a) (single layer samples) and Figure 4(b) (bilayer samples) was performed to fit respectively the data shown in Figures 3(a) and 3(b). Analysis of the results obtained from the modelling as at Figure 5 for the single layer samples, and at Figure 6 enabled us to identify one of the layers in the bilayer samples as a subplantation layer. Considering now the straightforward case of the single- layer samples (with reference to Table 1), the layer thickness obtained independently via refinement of the model at Figure 4(a) to fit the KFPs at Figure 3(a) are as per Figure 5(a), and are in agreement with the first-principles experimental prediction of the FT at Figure 3(c). The simultaneous increase in thickness and density from HJ35 to HJ15 (at fixed high PGP and low OP) is an as-expected outcome of the increase in RFP. The decrease in thickness from HJ19 to HJ31 (at fixed high PGP and high OP) is more interesting and indicates an interaction between the RFP and the OP process conditions. Indeed, linear modelling of the results reveals an RFP: OP interaction term with a magnitude comparable to the main-effect individually of RFP or of OP. Without wishing to be bound by any theory, it is postulated that sputtering, of the forming film on the substrate, may be enhanced due to increased O 2 content in the deposition chamber leading to a decreased (final) thickness. The signature of this effect is an increase in surface roughness from HJ19 to HJ31 (see Figure 8). The increase in density from HJ15 to HJ31 (at fixed high PGP and high RFP) is also consistent with this hypothesis, since more oxygen is incorporated in the forming film.

RWU

Rpp PGP OP

(Watt) (mTorr) (%)

- 25 2 0

CU 0 87.5 6 10

+ 150 10 20

Table 2: Interconversion between the CU and RWU of process conditions; with reference to Table 1.

Considering next the case of the bilayer samples, a progressive emergence of a subplantation layer is observed as one proceeds, via the centre-point, from the high PGP face (back face) to the low PGP face (front face) of the experiment design shown in Figure 1. With reference also to Table 2, the three values of PGP spanned approximately an order of magnitude, with the lowest being the minimum pressure necessary to maintain conditions conducive to the striking and sustenance of a plasma within the deposition chamber, and the highest being at the upper limit of the pressure which would ensure arrival at the substrate of adequate sputter flux.

The layer thickness obtained independently via refinement of the model at Figure 4(b) to fit the KFPs at Figure 3(b) are as per Figure 6(c) (SPL) and Figure 6(a) (TFL), and are in agreement with the first-principles prediction of the FT at Figure 3(d). The refinement procedure also allows estimation of the density of each of the two layers, namely, the SPL and TFL, and the same are listed at Figure 6(d) (SPL) and Figure 6(b) (TFL) for the bilayer samples. The above ideas can be applied to the TFL of the bilayer samples (with reference to Table 1), with the caveat that all bilayer samples are produced at lower (0 or -, refer Table 2) PGP. The increase in thickness from HJ27 to HJ11 (at fixed low PGP and low OP), and in HJ3 to HJ23 (at fixed low PGP and high OP), is an as-expected outcome of the increase in RFP. The respective simultaneous decrease in density for the two previous cases is likely due to the decrease in densification (of matter) due to rapid film growth at high RFP; this should manifest as a change in morphology between the two samples. The thickness and density values for HJ7 (centre-point sample) are consistent with the above ideas.

Considering now the SPL of the bilayer samples (with reference to Table 1), the SPL layer was modelled as a mixture of CEXG and Ini 0 SiiZni 5 O32 (50% by mass each). The density values as obtained were not inconsistent with the known densities of the CEXG (2.38 g/cm 3 ), Ih 2 q3 (7.18 g/cm 3 ), and ZnO (5.61 g/cm 3 ). The opposing direction of change in thickness and density from HJ27 to HJ11 is as expected due to the increase in RFP which leads to an enhanced sputter flux impinging on to the substrate, causing in turn a greater penetration of the sputter flux into the CEXG substrate, leading thereby to a thicker but less dense SPL.

Figure 7 gives support for the creation of the SPL consistent with the aforementioned ideas. The experimentally-observed negative correlation between the roughness-to- thickness ratio and the relative density (with respect to density of CEXG) of the SPL - across thin- films deposited at different process conditions (refer to Table 1) - lends credence to it being a subplantation layer. The results may be understood as follows: the relative density is lowest at the highest roughness-to-thickness ratio; this situation corresponds to the very beginnings of incorporation of (higher mass) In and Zn species within the substrate, that is, in place of and via removal (by being sputtered away) of the (lower mass) original constituents of the substrate (glass); it will be recognized that this situation corresponds to the beginnings of the formation of a subplantation layer.

With increasing In and Zn incorporation, the relative density continues to increase; and simultaneously, due to the continual bombardment by a collimated flux of equal mass In and Zn species, the thickness (of the subplantation layer) increases at progressively lower roughness disadvantage. Experimental observations support this picture: the roughness of the SPL decreases monotonically with decreasing roughness- to-thickness ratio. At the highest value of relative density, the roughness-to-thickness ratio tends to zero, indicating the creation of a smooth interface. The described embodiments therefore show an ability to create a near zero-roughness interface/transition between a substrate and thin-film for an industrially- important physical vapour deposition process, in a manner that simultaneously decreases also the coefficient of thermal expansion differential between them without introduction of any extraneous chemistry. Evidently, this may present a method to create precisely controlled interfaces between a substrate and the thin-film atop it.

PGP MPL Collision Omni (CO,

(CU) (cm) estimated

- 1.72 14

0 0.58 41

+ 0.35 69

Table 3 : The MFL is the distance between Argon atoms at the pressure indicated (see table 2 for conversion between CU and RWU) and is taken as a surrogate for the distance travelled by a given assemblage of sputtered flux before it undergoes a collision. The change in momentum of the assemblage in question as a result of each collision will be a function of its own mass and the mass of the atom with which it collides, as also the initial momenta of the two species. The CC then is the number of collisions that a sputtered assemblage will undergo over a distance equal to the throw-distance (distance between the sputter target and the substrate holder; = 23.7 cm for the presently employed sputtering sputtering kit / machine), and is estimated by dividing the throw-distance by the MFL of gas atoms in the deposition chamber during sputtering. The MFL is estimated assuming that the process pressure in the deposition chamber is due entirely to Argon atoms via the relation,

MFL = RT / V2nd 2 N A P, where, R = 62.3639 m 3 mTorr/mol K, is the ideal gas constant, T is the absolute temperature in Kelvin (= 298.15 K), d is the relevant atomic diameter (= 142 pm for an Argon atom), N A is the Avogadro constant (= 6.022 xl0 23 /mol), and P is the presssure in mTorr. It is evident that an explicit collimation arrangement is essential to reduce the angular spread of the sputter flux arriving at the substrate even at the lowest pressures that are yet amenable for striking a plasma for sputtering. Absent such precaution, the condition of normal incidence of the sputter flux on to the substrate is not met, which would preclude the development of the SPL.

Ballistic travel of material flux sputtered from the target to the substrate is practically impossible given the process conditions inherent to sputtering. As such, a sputter flux collimator (SFC) has been developed and used, which enables a collimation arrangement to select a substantially unidirectional component of the sputter flux from among the sputter flux available close to the substrate. The Mean Free Lengths (MFLs) listed in Table 3 informed the design of the SFC. The SFC 10 is shown as part of an assembly 100 in Figures l la and l lb, in plan view in Figure l2a, and as a cross-sectional view along line A-A of Figure l2a in Figure l2b. The assembly 100 shown in Figures l la and l lb also comprises a shield 12, which is further shown in Figure 13. The assembly 100 is shown in situ within a sputtering machine 1000 in Figure 14.

The SFC 10 of the depicted embodiment has a planar, annular body 14 with a plurality of apertures 16 therethrough. The apertures 16 extend from a front face 18 of the body 14 to a back face 20 of the body 14. The apertures 16 are arranged in radially- aligned pairs around the body 14. In the depicted embodiment, a total of eighteen pairs of apertures 16 are equally-spaced circumferentially about the centre of the body 14, the centres of each pair being 20° separated from adjacent pairs - giving the body 18-fold rotational symmetry around a central axis.

The aforementioned embodiment of the SFC 10 therefore enables the conduct of a two- level three factor full factorial with centre-point experiment design, if four samples are to be produced for each process condition. This is since a two-level three factor full factorial requires a total of 2 3 (= 8) experimental runs. The center point requires one (= 1) additional experimental run. This makes a total of 9 (= 8 + 1) runs. Four adjacent apertures arranged in a 2-above-2 can then be utilized for any given experimental run, that is, for any given set of process conditions (RFP, PGP, OP), yielding thus 4 samples per process conditions set.

Those skilled in the art of experiment designs will recognize that the presently disclosed SFC admits a very wide variety of experimental designs, for example, full factorial designs, half factorial designs, quarter factorial designs. Those skilled in the art will realize that the presently disclosed SFC can be used to implement experimental designs according to, for example, Fisher, or Rao, or Bose, or Cox, or Srivastava, or Taguchi, or Box, or Shrikhande. Those skilled in the art will also realize that such designs correspond to different levels of resolution, with full factorial designs having the highest possible resolution.

Each aperture 16 is 11 mm wide with a substantially square cross-section, although with rounded corners - commonly referred to as a squircle. The body 14 is 13 mm deep and the apertures 16 are directed through the body 14 at an angle of 10° offset from a central axis of the body 14. The aforementioned offset angle is exactly the same at the angle at which is oriented the plane of the sputter target / sputter gun, in the described embodiment.

The back / distal face 20 includes a 13 mm wide countersunk hole 22 around each aperture 16, which serves as a means to accommodate one of more substrate. Circular cut-outs 24 are provided at each corner of the countersunk hole 22, to aid for instance, the use of a tweezer to dispose or remove the substrate.

Although shown with a total of eighteen pairs of apertures 16 and therefore thirty-two total apertures 16, a greater or lesser number of apertures 16 may be included within other embodiments of SFC. Also, the apertures may be grouped in subject to the design constraint that the incoming sputter flux remain substantially uniform across their combined cross section. The apertures may be provided at different spacing, even or otherwise, about the body. The body 14 may be solid rather than annular. Each of these aspects may be changed based on the desired characteristics of the SFC, the experiments for which it is intended, and/or the physical vapour deposition, e.g. sputtering, machine in which it is to be used.

In this example embodiment, the SFC 10 is designed to allow for the placement of a substrate substantially perpendicular to the collimated sputter flux. In this example embodiment, the direction of choice is the direction perpendicular to the plane of the target, and it is in this direction that the sputter flux at-target is maximal (assuming that a Lambertian distribution of sputter flux emerges from the target). Of course, the directional distribution of sputter flux is likely to change as it makes its way towards the substrate, yet the perpendicular direction may be a natural direction to choose, given symmetry, and given that maximal flux is initially available in this direction.

In addition to the aforementioned collimation arrangement, the design of a given SFC may take into account, for example, requirements imposed by modalities that are to be used to characterize the samples, the size of the vacuum chamber of the sputtering kit, the size of the target that is sputtered, considerations of uniformity of thickness of the resulting film, and statistical design of experiment considerations, to fix the size of each substrate that is to be accommodated. The shield 12 comprises a plate 26 with a slot 28 that extends radially outwards from a substantially central part of the plate 26. In this example embodiment, the slot 28 has a constant angular width. The slot 28 is sized such that, when aligned with a pair of apertures 16 in the SFC 10, a path is provided through the slot 28 and the apertures 16. By rotating the SFC 10 relative to the shield 12, different apertures 16 can be covered or uncovered, providing paths through the different pairs of apertures 16, depending on the relative orientation of the SFC 10 and shield 12. The shield 12 also includes a shroud 30 that acts to block incoming flux particles travelling towards non- desired pair of apertures 16 of the SFC 10. The shroud 30 extends vertically around three sides of the slot 28, with only the radially outermost edge of the slot 28 being open - no apertures 16 are positioned in this direction so leakage of the flux particles may be less of a concern.

Although described as a slot, an opening of any shape may be used, and the remainder of the plate may be considered as a cover portion.

By using the shield 12 in conjunction with the SFC 10, high throughput studies of fixed materials systems can be achieved under different process conditions, without the need to release the vacuum between deposition cycles or experimental runs. It may therefore be possible, for example, to identify optimum manufacturing conditions at a much faster rate than is possible with known systems.

The above is achievable by first lining up the slot 28 in the shield 12 with a first set of apertures 16, before beginning a deposition step, for example by sputtering, resulting in a layer of target material being deposited on a substrate located on the back / distal face 20 of SFC 10. As the shield 12 can be rotated within the machine without releasing of the vacuum, the slot 28 in the shield 12 can be realigned with a different set of apertures 16 in order to carry out a further deposition step on the substrate. This can be repeated any number of times, and, optionally, the process conditions within the machine can be altered each time. Thus, higher throughput can be achieved and with decrease of cost. We have demonstrated a decrease in monetary cost by a factor of 3x, and a decrease in time required by a factor of lOx, over traditional methods (for the costing framework in existence within the facilities accessed by us). A large number of samples may be produced in a time- and cost-efficient manner. In such a case, the substrate may be held stationary, whilst the shield 12 is rotated. As mentioned above, the assembly including the SFC 10 comprises a means to accommodate a substrate such that the substrate is normal to the arriving collimated sputter flux. Given the“physical vapour deposition” nature of sputtering, a large- enough deviation from this condition will washout the normal- incidence conditions important/necessary for creation of an SPL. It is noted that, in its bringing together the aforementioned design considerations, the SFC 10 does not compromise the ability to create and sustain a plasma within the vacuum chamber as per the original design of the sputtering kit.

A low-roughness substrate may be chosen, for example, Corning Eagle XG Glass. The substrate may be prepared and/or treated prior to deposition, e.g. to reduce surface roughness and/or to clean the substrate. This may enhance substrate-film adhesion.

In the described experiments, the materials system used was the ternary system (In, Si, Zn, O). This material system results in the formation of a transparent conducting oxide (TCO) on the substrate. Other materials systems can be used, including those that result in other TCOs or other functional materials for use, for example, in electronic or optoelectronic devices. The materials systems may be binary, tertiary, or ternary systems, e.g. (Sn, Zn, In, O) or (Ga, Zn, In O), or related systems. Functional materials suitable for use in transistors or transparent electrodes, for example within flexible touchscreens or other electronic applications, may be of particular interest. The substrate may be flexible. The provision of a defined, controlled, subplantation layer may allow for reliable, robust adhesion of the deposited functional material on/to the substrate. However, the invention is not limited to flexible substrates, and device structures deposited on substantially rigid substrates may also benefit from the application of a subplantation layer.

In the assembly shown in Figure 14, assembly 100 comprising the SFC 10 and shield 12 are positioned within the vacuum chamber of a sputtering machine 1000. The sputtering machine 1000 includes outlets 1002 to a vacuum pump, along with inlets 1004 for process gases, which may be delivered to a gas ring 1006, and argon gas. More than one sputter gun is available, allowing thus, for the use of one or more sputter targets 1008. In the view shown in Figure 14, two sputter guns are shown. A greater number of sputter guns may be available, depending on the set-up and machine being utilised. Two targets 1008 are provided, each of which is connected to a magnetron 1010.

With reference to Table 3 and Figure l2b, the cross-sectional dimensions and the depth of the aperture through the SFC 10 are chosen such that they are shorter than the mean free length (MFL) in the range of process pressure sought to be investigated.

Those skilled in the art will understand that the sputter flux is hyperthermal. By “hyperthermal”, it is meant that the flux particles have a kinetic energy greater than the average temperature within the sputtering chamber, or the average temperature of the target, or the any other bulk average temperature of any component within the sputtering chamber.

Figure 10(a) gives the number of electrons per volume (that is, the electron number density) for the InioSiiZnisC^ thin-films at issue. Two outstanding features are recognisable. Firstly, the electron number density as indicated may be compared to the electron number density for the crystalline phase of the composition I^SioZnaOe; this last composition crystallizes as a hexagonal bravais lattice with a unit cell volume = 413.43A 3 and with three formula units per unit-cell, and has a density of 6.29 g/cm 3 . Given the atomic numbers of In, Zn, and O (46, 30, and 8 respectively), one obtains the number of electrons per unit cell at 708, and an electron number density of = 1.7125 eVA 3 .

In the first instance, considering only the four bilayer samples in the front face (HJ27 and HJ3 at low RFP, and HJ11 and HJ23 at high RFP) - with the caveat that for the bilayer samples the electron density as mentioned is an average value across the combined TFL and the SPL, and not the electron density in any individual layer - the electron number density is compared with the (mass) density of the TFL of the bilayer samples. In order to make the aforementioned densities comparable (as they have different dimensions), they are normalised with respect to the corresponding values of the crystalline phase of a (very) close chemical composition In 2 SioZn306, and the results are shown in Figure 10(b) and Figure 10(c) respectively. A comparison of Figures 10(b) and 10(c) indicates that the differential in scattering cross-section of the deposition chamber process gas for the different atoms sputtered from the target material is unlikely to be able to account for the observed divergence in the numbers. So for instance, while the (mass) density of all samples are clustered around the value for In 2 SioZn 3 0 6 (a maximum change of about +6% is observed at HJ3), the electron number density of all bilayer samples shows a much more significant change of about -27% (at both the high RFP bilayer samples, viz., HJ23 and HJ11) and of about -61% (at both the low RFP bilayer samples, viz., HJ3 and HJ27).

Since the (mass) density of the four bilayer samples at the front face of the cube-plot of Figure 1 are within a few percent of each other, it can be reasonably expected for their morphologies too to be similar. The observed changes in electron number densities between the low RFP (HJ3 and HJ27) and high RFP (HJ23 and HJ11) samples likely has contribution from the microstructure; neutron scattering experiments to measure the radial distribution function(s) are presently being contemplated, first for the samples at the low PGP/front face to measure the amorphous structure between the TFLs deposited at low RFP (HJ3 and HJ27) and at high RFP (HJ23, HJ11).

The lowered density of the TFL of the samples at the back face is not unexpected as they are deposited at high PGP and consequently the incoming sputter flux undergoes a much greater loss of kinetic energy during traversal from the target to the substrate (refer to Table 3), which results in increased porosity in the forming film structure. The density change accruing thereof due to such a mundane“morphological” effect is expected to be much greater than the more interesting density change, if any, accruing due to a change in atom-atom distance.

While the example embodiments have been described with reference to sputtering, it will be appreciated that the present disclosure may also be applied to other deposition techniques, in particular physical deposition techniques for thin-film deposition such as evaporation deposition.