Method of Deposition

Abstract

In a method for sputter depositing an additive-containing aluminium nitride film containing an additive element like Sc or Y, a first layer of the additive-containing aluminium nitride film is deposited onto a substrate disposed within a chamber by pulsed DC reactive sputtering. A second layer of the additive-containing aluminium nitride film is deposited onto the first layer by pulsed DC reactive sputtering. The second layer has the same composition as the first layer. A gas or gaseous mixture is introduced into the chamber when depositing the first layer. A gaseous mixture comprising nitrogen gas and an inert gas is introduced into the chamber when depositing the second layer. The percentage of nitrogen gas in the flow rate (in sccm) when depositing the first layer is greater than that when depositing the second layer.

Claims

1. A method for sputter depositing an additive-containing aluminium nitride film containing an additive element selected from Sc or Y, the method comprising the steps of: depositing a first layer of the additive-containing aluminium nitride film onto a substrate disposed within a chamber by pulsed DC reactive sputtering; and depositing a second layer of the additive-containing aluminium nitride film onto the first layer by pulsed DC reactive sputtering, the second layer having the same composition as the first layer; wherein: the step of depositing the first layer comprises introducing a gas or gaseous mixture into the chamber at a flow rate (in sccm), and 87-100% of the flow rate (in sccm) is a flow of nitrogen gas; the step of depositing the second layer comprises introducing a gaseous mixture into the chamber at a flow rate (in sccm), the gaseous mixture comprising nitrogen gas and an inert gas; and the percentage of the nitrogen gas in the flow rate (in sccm) used during the step of depositing the first layer is greater than a percentage of the nitrogen gas in the flow rate (in sccm) used during the step of depositing the second layer.

2. A method according to claim 1, wherein the additive element is scandium.

3. A method according to claim 1, wherein the additive element is present in an amount in the range 0.5 At % to 40 At %.

4. The method according to claim 1, wherein 90-100% of the flow rate (in sccm) used during the step of depositing the first layer is a flow of nitrogen gas.

5. The method according to claim 4, wherein the flow rate (in sccm) used during the step of depositing the first layer consists essentially of a flow of nitrogen gas only.

6. The method according to claim 1, wherein the flow of nitrogen gas used during the step of depositing the first layer is in the range 50 to 500 sccm.

7. The method according to claim 1, wherein the gas or gaseous mixture used during the step of depositing the first layer comprises nitrogen gas and an inert gas.

8. The method according to claim 1, wherein the chamber has a pressure in the range 2-6 mTorr during the step of depositing the first layer.

9. The method according to claim 1, wherein the chamber has a pressure in the range 1.5-7.5 mTorr during the step of depositing the second layer.

10. The method according to claim 1, wherein the first layer has a thickness of less than 70 nm.

11. The method according to claim 1, wherein the additive-containing aluminium nitride film has a thickness of 0.3 m or greater; 0.6 m or greater; or about 1 m.

12. The method according to claim 1, wherein the additive-containing aluminium nitride film has a thickness of 2 m or less.

13. The method according to claim 1, wherein the step of depositing the first layer is performed with an electrical bias power applied to the substrate.

14. The method according to claim 13, wherein the step of depositing the second layer is performed with no electrical bias power applied to the substrate or with an electrical bias power applied to the substrate that is lower than the electrical bias power applied during the step of depositing the first layer.

15. The method according to claim 1 further comprising the step of etching a surface of the substrate prior to the step of depositing the first layer so that the first layer is deposited onto the etched surface of the substrate.

16. The method according to claim 1 wherein the substrate is a silicon substrate.

17. The method according to claim 1 wherein the substrate comprises a metallic layer, such as a molybdenum layer, onto which the first layer of the additive-containing aluminium nitride film is deposited.

18. The method according to claim 17 further comprising the step of depositing the metallic layer onto a substrate precursor.

19. An additive-containing aluminium nitride film produced by the method according to claim 1.

20. An additive-containing aluminium nitride film containing an additive element selected from Sc or Y in an amount in the range 8 At % to 40 At %; and having a defect density of less than 50 defects per 100 m.sup.2.

21. A piezoelectric device comprising an additive-containing aluminium nitride film according to claim 20.

22. The piezoelectric device according to claim 21, in which the piezoelectric device is a bulk acoustic wave (BAW) device.

23. The piezoelectric device according to 22 comprising a first and a second electrode, with the additive-containing aluminium nitride film being disposed between the first and second electrodes.

Description

DESCRIPTION OF THE DRAWINGS

[0043] Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0044] FIG. 1 is an SEM image showing Al.sub.80Sc.sub.20N defects;

[0045] FIG. 2 is an SEM image show an Al.sub.80Sc.sub.20N defect;

[0046] FIG. 3 is a TEM cross-sectional image of Al.sub.80Sc.sub.20N defects;

[0047] FIG. 4 is an illustration of a first layer in an AlScN film comprising a point defect;

[0048] FIG. 5 is an illustration of a first layer in an AlScN film, which is free from point defects;

[0049] FIG. 6 is an SEM image of an Al.sub.80Sc.sub.20N film comprising a 17 nm thick first layer having a tensile stress;

[0050] FIG. 7 is an SEM image of an Al.sub.80Sc.sub.20N film comprising a 17 nm thick first layer having a compressive stress; and

[0051] FIG. 8 is a flow chart exemplifying a method of the present invention.

DETAILED DESCRIPTION OF THE DISCLOSURE

[0052] The inventors have discovered an advantageous process for sputter depositing an additive-containing aluminium nitride film. The method can help to improve crystallinity and texture, and reduce crystallite defects in additive-containing aluminium nitride films. The additive-containing aluminium nitride film contains an additive element, such as scandium (Sc) or yttrium (Y). The results presented below are in relation to aluminium scandium nitride (Al.sub.1xSc.sub.xN). However, the method is generally applicable to aluminium yttrium nitride (Al.sub.1xY.sub.xN).

[0053] The film is deposited by reactive sputtering such as pulsed DC reactive sputtering. General details concerning apparatus which can be used or readily adapted for use in the present invention are described in the Applicant's European Patent applications EP2871259 and EP3153603, the entire contents of which are hereby incorporated by reference.

[0054] The apparatus comprises a substrate disposed in a chamber. The apparatus further comprises a target. The target is a composite target formed from aluminium and the additive element. The composition of the target can determine the amount of additive element contained in the sputter deposited film. The use of multiple targets is possible but is likely to be less economically attractive. Pulsed DC sputtering comprises applying pulses of DC power to the target during the deposition process.

[0055] In a first step, a first layer of the additive-containing aluminium nitride film is sputter deposited from the target onto a substrate that is disposed in a chamber. The first layer is deposited by pulsed DC reactive sputtering. The first layer can be a seed layer. During the deposition of the first layer, a gas or gaseous mixture comprising nitrogen, and optionally an inert gas, such as argon, is introduced into the chamber. The flow rate of nitrogen gas (in sccm) during the first step is 87-100% of the total gas flow rate (in sccm) during the first step. Optionally, the flow rate of the nitrogen gas (in sccm) during the first step is 90-100%, 95-100%, 98-100% or about 100% of the total gas flow rate (in sccm) during the first step. Preferably, the gas or gaseous mixture consists only of nitrogen gas. The first layer typically has a thickness of less than about 70 nm, less than 60 nm, less than 50 nm, preferably less than 20 nm. In some embodiments, the first layer has a thickness of about 17 nm.

[0056] In a second step, a second layer of the additive-containing aluminium nitride film is subsequently deposited onto the first layer, e.g. onto an initial seed layer. The deposition of the second layer can be a bulk deposition. The second layer is deposited by pulsed DC reactive sputtering. During the deposition of the second layer, a second gaseous mixture comprising nitrogen and an inert gas, such as argon, is introduced into the chamber. Other inert gases, such as xenon and krypton, can be contemplated, although are less preferable due to their higher cost. The proportion of nitrogen gas in the second gas mixture is typically less than the proportion of nitrogen gas in the first gas or gaseous mixture. In one embodiment, the flow rate of the nitrogen gas during the deposition of the second layer is 83 sccm, and the flow rate of argon gas is 17 sccm. That is, the flow rate of nitrogen gas during the second step is about 83% of the total flow rate (in sccm).

[0057] Typical deposition parameters for the experiments on a silicon substrate are shown in Table 1.

TABLE-US-00001 TABLE 1 Typical process parameters for deposition of AlScN by two-step process. AlScN first step AlScN second step (seed layer) (bulk deposition) Film thickness <20 nm ~980 nm N.sub.2 flow (sccm) 150 83-85 Ar flow (sccm) 0 17 Platen RF bias power (Watts) >250 0-100 Chamber pressure (mTorr) 4 3

[0058] 1 m Al.sub.80Sc.sub.20N films were sputter deposited onto a silicon substrate using a single target using the methods described above. Table 2 shows how varying the proportion of nitrogen gas during the deposition of the first layer (i.e. initial seed layer) affects the defect density (per 100 m.sup.2) in as deposited 1 m Al.sub.80Sc.sub.20N films. Defect density was determined using a scanning electron microscope (SEM) image at a magnification of 6,000 times. Table 3 shows how varying the proportion of nitrogen gas during the deposition of the first layer (i.e. initial seed layer) affects the texture of the as deposited Al.sub.80Sc.sub.20N film. X-ray diffraction (XRD) full width half maximum (FWHM) measurements were used to determine the texture (or crystallinity) of samples at the centre, mid-radius and the edge of the substrate. A lower FWHM value corresponds to a more crystalline film. Prior to the deposition process, the silicon substrates were subjected to a 2 minute degas step at 350 C. The example listed in the final row of Tables 2 and 3 included the additional step of subjecting the substrate to a 7.5 nm low bias etch step at 350 C. prior to the sputter deposition process. An SE-LTX module, which is commercially available from SPTS Technologies Limited, is suitable for performing for the pre-treatment degassing and etch steps. The 1 m Al.sub.80Sc.sub.20N films (shown in Tables 2 and 3) comprise a 17 nm thick compressive first layer (e.g. initial seed layer), and a 983 nm second layer (e.g. bulk layer). The first layer was produced using a 300 W RF bias on the platen. The second layer was produced using a RF bias power that was selected in order to achieve a zero stress across the entirety of the film. That is, a zero stress across both the first and second layers. Typically the bias applied to the platen during the second step is less than the bias applied to the platen during the first step. During the deposition of the second layer, the flow rate of nitrogen gas was 83 sccm, and the flow rate of argon gas was 17 sccm.

TABLE-US-00002 TABLE 2 Defect density of 1 m Al.sub.80Sc.sub.20N films on Si substrate at different percentage nitrogen flows. % N.sub.2 flow (sccm) during deposition Defect density per 100 sq. m of first layer Edge Mid radius Centre 78 25 >250 >500 83 (Standard) 10 30 >250 94 5 20 >100 100 1 15 70 100 (Degas + etch) 0 0 5

TABLE-US-00003 TABLE 3 XRD FWHM measurements of 1 m Al.sub.80Sc.sub.20N films on Si substrate at different percentage nitrogen flows. % N.sub.2 flow (sccm) during deposition Texture (0002) FWHM (Deg) of first layer Edge Mid radius Centre 78 1.73 1.86 1.81 83 (Standard) 1.77 1.74 1.60 94 1.70 1.68 1.56 100 1.63 1.61 1.58 100 (Degas + etch) 1.64 1.60 1.54

[0059] Table 2 shows that as the proportion (i.e. percentage flow) of nitrogen gas increases, the defect density at the edge, mid radius and centre of the substrate decreases. Table 3 shows that as the proportion of nitrogen gas during the first (seed) step increases, the texture (0002) FWHM value tends to decrease at the edge, mid radius and centre of the substrate. These effects are most pronounced when the gas used during the deposition of the first (seed) layer consists only of nitrogen gas.

[0060] Without wishing to be bound by any theory or conjecture, it is believed that crystallographic defects are induced by point defects, such as atom misalignment, misplacement or vacancy. The majority of crystal defects in AlScN films are believed to originate from the surface of the substrate material on which the AlScN film is grown. The resulting defect propagates throughout the film and is observable at the film surface. FIG. 3 shows a TEM cross-section of crystal defects 30 in an Al.sub.80Sc.sub.20N film. The defect 30 propagates through the film. These defects are particularly prominent in additive-containing aluminium nitride films where the additive element (e.g. Sc or Y) is in atomic concentrations of more than about 8 At %. For AlScN films, it is believed that the AlScN grains can be either nitrogen or aluminium (scandium) ended. Without wishing to be bound by any theory or conjecture, it is believed that if a nitrogen layer is deposited as the initial atomic layer, a crystallographic defect will form when an atom of another type is also incorporated into the initial atomic layer of nitrogen. FIG. 4 shows an Al/Sc atom 40 that is incorporated into an initial atomic layer of nitrogen to form a point defect. This defect can propagate throughout the AlScN film. Again without wishing to be bound by any theory or conjecture, it is believed that increasing the proportion of nitrogen gas content during the deposition of the first layer favours the deposition of a seed layer which is substantially free from point defects. For example, the initial atomic layer 50 may substantially consist of nitrogen only (as shown in FIG. 5).

[0061] It is preferable to use a nitrogen-rich or nitrogen-only atmosphere during the deposition of the first layer (i.e. initial seed layer). This reduces the number and density of defects in the additive-containing aluminium nitride film, and can improve the electromechanical coupling efficiency of the film. This permits a higher concentration of additive element to be present in the films, whilst maintaining acceptable levels of defect density and texture. Maintaining acceptable levels of defect density and texture in additive-containing aluminium nitride films having a high concentration of additive element (e.g. >8 At %) is not readily achievable using known methods, for example, where the proportion of nitrogen gas in the deposition of the first step is less than about 83-87%.

[0062] The Al.sub.80Sc.sub.20N films shown in Tables 2 and Tables 3 were prepared by depositing a compressive initial seed layer, followed by a bulk deposition so that the overall stress in the Al.sub.80Sc.sub.20N film is zero. The stress in the deposited film can be controlled by varying the substrate bias power. Table 4 shows how XRD FWHM measurements of a 1 m Al.sub.80Sc.sub.20N film varies if the first layer has a tensile or compressive stress. The 1 m Al.sub.80Sc.sub.20N films of Table 4 were formed by introducing only nitrogen gas into the chamber during the deposition of the first layer. The first layer had a thickness of 17 nm, and the second (bulk) layer had a thickness of 983 nm. FIGS. 6 and 7 show SEM images of the Al.sub.80Sc.sub.20N surface for films deposited with a tensile and compressive first layer respectively. Al.sub.80Sc.sub.20N films deposited with a compressive first layer exhibit a lower defect density, and an improved texture compared to Al.sub.80Sc.sub.20N films deposited with a tensile first layer.

TABLE-US-00004 TABLE 4 FWHM texture of 1 m Al.sub.80Sc.sub.20N films on Si substrate with varying stress in first layer. Centre texture FWHM Edge texture FWHM First layer stress (Deg) (Deg) Tensile 1.66 1.77 Compressive 1.58 1.63

[0063] The effect of substrate material and surface condition was investigated. 1 m Al.sub.80Sc.sub.20N films were deposited onto a molybdenum (Mo) coated substrate using a single composite target. Other metallic materials can be used as the coating material instead of Mo. The Mo-coated substrate was prepared according to the method shown in FIG. 8. A substrate precursor was initially degassed (step 802). A Mo coating was deposited onto the degassed substrate precursor in a Mo deposition module (step 806). The Mo coating was etched using a low bias etch treatment (step 808). The substrate was subsequently transferred to a sputter deposition module for a two-step AlScN deposition process to be performed (steps 810 and 812). The two-step AlScN deposition process comprises i) depositing a first layer (step 810) onto the Mo-coated surface of the substrate in a nitrogen-rich or nitrogen-only atmosphere, followed by ii) depositing a second layer (i.e. bulk deposition) onto the first layer (step 812). The process conditions used in steps 810 and 812 can be the same as or different to those described above in relation to other embodiments of the invention.

[0064] Tables 5 and 6 show how the defect density and texture varies by varying the proportion of nitrogen gas during the deposition of the first layer when depositing a 1 m Al.sub.80Sc.sub.20N film onto a molybdenum (Mo) coated substrate using the method of FIG. 8.

TABLE-US-00005 TABLE 5 Defect density of 1 m Al.sub.80Sc.sub.20N films on Mo substrate at different percentage nitrogen flows. % N.sub.2 flow (sccm) Defect density per 100 sq. m in seed step Pre-treatment Edge Mid radius Centre 83 (Standard) Degas and etch 15 >50 >150 100 Degas and etch 2 10 25 100 Degas only, >100 >1000 >1000 no etch

TABLE-US-00006 TABLE 6 FWHM texture of 1 m Al.sub.80Sc.sub.20N films on Si substrate at different percentage nitrogen flows. % N.sub.2 flow (sccm) Texture (0002) FWHM (Deg) in seed step Pre-treatment Edge Mid radius Centre 83 (Standard) Degas and etch 1.69 1.91 1.79 100 Degas and etch 1.65 1.77 1.70

[0065] Tables 5 and 6 show that the defect density for a 1 m Al.sub.80Sc.sub.20N on a Mo-coated substrate can be reduced, and the texture can be improved by using only nitrogen gas in the gaseous atmosphere during the deposition of the first layer. Additionally, conditioning the surface of the substrate by mild etching before depositing AlScN can also help to suppress the formation of crystallographic defects and can improve the texture and crystallinity of the resultant AlScN film.

[0066] The effect of the thickness of the first layer on defect density, crystallinity and texture was investigated. 1 m Al.sub.80Sc.sub.20N films were prepared on a Mo-coated substrate using only nitrogen gas during the deposition of the first layer. The Mo-substrates were prepared according to the method of FIG. 8. The thickness of the first layer was varied, and the texture of the resultant film was measured using XRD FWHM measurements. The results are shown in Table 7. The thinner first layer resulted in a more textured (i.e. improved texture) Al.sub.80Sc.sub.20N film. This effect is more significant at the edge of the substrate. A suitable thickness of the first layer is typically less than 70 nm, less than 60 nm, less than 50 nm, less than 30 nm, less than 25 nm, or less than 20 nm.

TABLE-US-00007 TABLE 7 Influence of first layer thickness on Al.sub.80Sc.sub.20N texture. N.sub.2 Seed Texture (0002) FWHM (Deg) Thickness (nm) Edge Mid radius Centre 17 1.70 1.71 1.65 50 1.83 1.79 1.66

[0067] In particular, the combination of depositing a thin first layer in a nitrogen-rich or nitrogen-only atmosphere has been found to significantly reduce defect density, and improve crystallinity and texture. These beneficial effects are observed even at high concentrations of additive element concentrations. Therefore, methods of the present invention are particularly suited for depositing additive-containing aluminium nitride films having high concentrations of the additive element whilst maintaining acceptable levels of defect density, crystallinity and texture.

[0068] The methods described above can be used to deposit additive-containing aluminium nitride films, such as Al.sub.1xSc.sub.xN, with varying concentrations of additive element. 1 m Al.sub.1xSc.sub.xN films were deposited with 0 At %, 9 At %, 15 At % and 20 At % on a bare silicon substrate. The additive-containing aluminium nitride films were deposited from a single, composite target. The amount of additive material in the deposited film was determined by the composition of the target. The first layer was deposited with an RF bias power of 200-350 W applied to the substrate. Only nitrogen gas was introduced into the chamber during the deposition of the first layer. That is, the flow rate during the step of depositing the first layer consisted of a flow of nitrogen gas. The thickness of the first layer was about 20 nm. The texture at the edge and centre of the as deposited film was measured, and the results are shown in Table 8.

TABLE-US-00008 TABLE 8 FWHM texture of 1 m Al.sub.1-xSc.sub.xN films on Si substrate with varying seed compositions Composition % N.sub.2 flow of X in (sccm) in Texture (0002) FWHM (Deg) Al.sub.1-xSc.sub.xN first step Edge Centre 0.15 83 1.88 1.95 100 1.76 1.78 0.09 83 1.66 1.69 100 1.61 1.62 0 83 1.56 1.55 100 1.50 1.50

[0069] The present inventors found that depositing a first layer (e.g. an initial seed layer) of about 20 nm, wherein 100% of the flow rate (in sccm) is a flow of nitrogen gas (N.sub.2), improved the texture of the as deposited films at a range of additive element concentrations. Improvements were observed for all additive element concentrations. This is particularly advantageous for higher additive element concentrations, where known prior methods result in unacceptable levels of defects and poor texture. The present method permits acceptable levels of texture and defect density to be achieved for additive element concentrations above 8 At %, 9 At %, 10 At %, 15 At %, 20 At %, and 25 At %.