Conformal smooth titanium nitride layer and method for forming the same

Abstract

The disclosed technology generally relates to forming a thin film containing titanium nitride (TiN), and more specifically to forming a thin film containing (TiN) by a cyclic vapor deposition process. In one embodiment, a method for forming a thin film containing TiN comprises the steps of forming a first portion of the thin film by exposing a semiconductor substrate to one or more first cyclic vapor deposition cycles, each comprising exposure to a first Ti precursor and exposure to a first N precursor, and forming a second portion of the thin film by exposing the semiconductor substrate to one or more second cyclic vapor deposition cycles, each comprising exposure to a second Ti precursor and exposure to a second N precursor, wherein the exposure to one or both of the first Ti precursor and the first N precursor during one or more first cyclic vapor deposition cycles is performed at a different pressure compared to the corresponding exposure to one or both of the second Ti precursor and the second N precursor during one or more second cyclic vapor deposition cycles. The embodiment also relates to a semiconductor structure including a thin film and a method for forming the same.

Description

Technology Field

[0001] The disclosed technology generally relates to forming a titanium nitride layer, and more specifically to an conformal, smooth titanium nitride layer. Background Technology

[0002] Titanium nitride (TiN) has been widely used in the fabrication of various structures for integrated circuits (ICs). For example, TiN has been used as a diffusion barrier, for various electrodes, and for metallized structures. The widespread use of TiN in IC manufacturing may be due to its structural, thermal, and electrical properties. As the dimensions of various IC structures shrink, TiN is formed in features with increasingly smaller dimensions and complex topologies. For example, as technology nodes scale beyond the 10 nm node, a TiN layer capable of conformally lining high aspect ratio trenches and vias with dimensions as small as a few nanometers is required, for instance, as a diffusion barrier. Although technologies such as physical vapor deposition (PVD) and chemical vapor deposition (CVD) have been used for decades to form TiN in the IC industry, their use may eventually be limited as the need for conformality of TiN films to be deposited in smaller trenches or vias increases. On the other hand, while Atomic Layer Deposition (ALD) has been proven for conformal deposition of TiN films, some electrical properties (e.g., conductivity) and physical properties (e.g., surface roughness) of the film may be inferior compared to TiN films formed using other methods, such as Physical Vapor Deposition (PVD). Therefore, there is a need for an ALD method to form TiN-based films with excellent surface smoothness and step coverage, while also possessing electrical and physical properties that are comparable to or superior to those of TiN films formed by PVD and CVD for use in IC fabrication. means of solving the problem

[0003] In one embodiment, a method for forming a thin film comprising titanium nitride (TiN) by a cyclic vapor deposition process comprises the step of forming a first portion of the thin film on a semiconductor substrate by exposing the semiconductor substrate to one or more first cyclic vapor deposition cycles, each comprising exposure to a first Ti precursor and exposure to a first N precursor. The method further comprises the step of forming a second portion of the thin film on the first portion of the thin film by exposing the semiconductor substrate to one or more second cyclic vapor deposition cycles, each comprising exposure to a second Ti precursor and exposure to a second N precursor. During one or more second ALD cycles, exposure to one or both of the Ti precursor and the N precursor is performed at a higher pressure than the corresponding exposure to one or both of the Ti precursor and the N precursor during one or more first ALD cycles.

[0004] In another embodiment, a method for forming a thin film comprising titanium nitride (TiN) by a cyclic vapor deposition process comprises the step of providing a semiconductor substrate comprising a trench or via having an aspect ratio greater than 1. The method further comprises the step of forming a thin film within a trench or via by exposing the semiconductor substrate to one or more first cyclic vapor deposition cycles, each comprising exposure to a first Ti precursor and exposure to a first N precursor, to form a first portion of the thin film within the trench or via. The method further comprises the step of forming a second portion of the thin film on the first portion of the thin film by exposing the semiconductor substrate to one or more second cyclic vapor deposition cycles, each comprising exposure to a second Ti precursor and exposure to a second N precursor. Exposure to one or both of the first Ti precursor and the first N precursor during one or more first cyclic vapor deposition cycles is carried out at a different pressure compared to the corresponding exposure to one or both of the second Ti precursor and the second N precursor during one or more second cyclic vapor deposition cycles.

[0005] In another embodiment, the semiconductor structure comprises a semiconductor substrate comprising a non-metallic sidewall surface within a trench or via having an aspect ratio greater than 5. The semiconductor structure further comprises a thin film comprising TiN conformally coating the non-metallic sidewall surface, wherein the ratio of the thickness of the thin film formed in the lower 25% of the trench or via height to the upper 25% of the trench or via height exceeds 0.9. Brief explanation of the drawing

[0006] FIGS. 1a to 1d schematically illustrate the nucleation and growth mechanisms of thin films under different growth modes. Figure 2 is a cross-sectional transmission electron microscope image of a TiN layer grown on an oxide-coated silicon substrate by thermal atomic layer deposition. FIG. 3a is a flowchart schematically illustrating an atomic layer deposition method for forming a TiN layer by exposing a substrate to multiple cycles with different corresponding precursor exposure pressures according to an embodiment. FIG. 3b schematically illustrates a cross-sectional view of a semiconductor structure comprising a TiN layer formed by an atomic layer deposition method in which a substrate is exposed to multiple cycles with different corresponding precursor exposure pressures according to an embodiment. FIG. 4 schematically illustrates pressure tracking of different cycles of an atomic layer deposition method in which a substrate is exposed to multiple cycles with different corresponding precursor exposure pressures according to an embodiment. FIG. 5 schematically illustrates a cross-sectional view of a via lined with a TiN layer having different thicknesses in different parts of the via. FIG. 6 is a graph showing the experimentally measured surface roughness and step coverage trends as a function of thickness in a TiN layer formed by an atomic layer deposition method in which a substrate is exposed to multiple cycles with different corresponding precursor exposure pressures according to an embodiment. Figure 7a is a cross-sectional transmission electron microscope image of a high aspect ratio via lined with a TiN layer formed by an atomic layer deposition method in which the substrate is exposed to an ALD cycle performed at the same precursor exposure pressure. Figure 7b is a cross-sectional transmission electron microscope image of the upper region of the high aspect ratio via shown in Figure 7a. Figure 7c is a cross-sectional transmission electron microscope image of the lower region of the high aspect ratio via shown in Figure 7a. FIG. 8a is a cross-sectional transmission electron microscope image of a TiN layer formed in the upper region of a high aspect ratio via similar to that shown in FIG. 7a by an atomic layer deposition method in which the substrate is exposed to multiple cycles with different corresponding precursor exposure pressures according to an embodiment. Figure 8b is a cross-sectional transmission electron microscope image of a TiN layer formed in the lower region of a high aspect ratio via and trench shown in Figure 8a. Figure 9 is a graph showing a statistical comparison of the measured step coverage between a TiN layer formed by atomic layer deposition at a single exposure pressure and a TiN layer formed by atomic layer deposition at multiple exposure pressures according to an embodiment. FIG. 10 schematically illustrates a cross-sectional view of a via lined with a TiN layer formed by an atomic layer deposition method in which a substrate is exposed to multiple cycles with different corresponding precursor exposure pressures according to an embodiment. Specific details for implementing the invention

[0007] As previously mentioned, the integrated circuit (IC) industry requires smooth conformal TiN films having excellent electrical and physical properties and methods for forming such films. To address these and other requirements, the present invention discloses a smooth conformal thin film comprising TiN that exhibits conformal properties of a film deposited by a cyclic vapor deposition method, while also possessing electrical and physical properties superior to or equivalent to those of a TiN film formed by conventional physical vapor deposition (PVD) and chemical vapor deposition (CVD) methods, and a cyclic vapor deposition method for forming such a thin film. Specifically, a method for forming a thin film comprising titanium nitride (TiN) comprises the step of forming a first portion of a thin film on a semiconductor substrate by exposing the semiconductor substrate to one or more first cyclic vapor deposition cycles, each comprising exposure to a first Ti precursor and exposure to a first N precursor. The present method further comprises the step of forming a second portion of a thin film on a first portion of a thin film by exposing a semiconductor substrate to one or more second cyclic vapor deposition cycles, each comprising exposure to a second Ti precursor and exposure to a second N precursor. Exposure to one or both of the second Ti precursor and the second N precursor during one or more second cyclic vapor deposition cycles is different from the corresponding exposure to one or both of the first Ti precursor and the first N precursor during one or more first cyclic vapor deposition cycles. The cyclic vapor deposition process disclosed herein is sometimes referred to as atomic layer deposition (ALD). However, the cyclic vapor deposition process is not limited to an atomic layer deposition process. For example, the precursors may partially or substantially saturate the reaction surface in the various embodiments described herein.

[0008] By exposing the substrate to Ti and / or N precursors at a relatively low pressure, e.g., less than 3 Torr, during the deposition of the first part of the thin film, the initial film growth can proceed substantially in a layer-by-layer growth mode, which advantageously results in a lower average grain size and lower surface roughness compared to a comparable TiN film deposited by exposing the substrate to Ti and / or N precursors at a higher pressure, e.g., exceeding 3 Torr or 5 Torr. On the other hand, by exposing the substrate to Ti and / or N precursors at a relatively high pressure, e.g., exceeding 3 Torr, during the deposition of the second part of the thin film, the later part of the film growth advantageously results in higher conformality or step coverage compared to a comparable TiN film deposited by exposing the substrate to Ti and / or N precursors at a relatively low pressure, e.g., less than 3 Torr or less than 1 Torr.

[0009] In addition, since the first part of the TiN film grows in a layer-by-layer mode, the second part of the film can continue to grow in a layer-by-layer mode using the first part as a template, compared to a comparable film grown starting with exposure to Ti and / or N precursors at a relatively higher pressure.

[0010] As a final result, a thin film comprising a first portion and a second portion, deposited by depositing at two different corresponding exposure pressures for one or both of the Ti precursor and the N precursor according to the method disclosed herein when deposited on a specific surface, e.g., a non-metallic surface, advantageously has a combination of superior surface roughness and conformability compared to a thin film layer formed on the same surface using a single pressure. Alternatively or additionally, due to the partially improved smoothness and conformability, the thin film has a relatively low electrical resistivity compared to a TiN layer formed by some conventional methods.

[0011] As described herein, compounds referred to by constituent elements without specific stoichiometric ratios should be understood to include all possible non-zero concentrations of each element unless explicitly limited. For example, titanium nitride (TiN) is a compound of the general formula Ti, comprising TiN, Ti3N4, Ti4N3, Ti6N5, Ti2N, and TiN2. x It should be understood to encompass all possible stoichiometric and non-stoichiometric compositions of titanium nitride that can be expressed as N (where x>0), as well as other non-stoichiometric compositions of Ti and N.

[0012] As mentioned above, titanium nitride (TiN) plays an important role in the manufacturing of integrated circuits (ICs). While techniques such as physical vapor deposition (PVD) and chemical vapor deposition (CVD) have been used to deposit TiN in the IC industry, there is an increasing need for deposition methods to form TiN-based films with high conformability without significant degradation of electrical and physical properties.

[0013] Furthermore, while plasma-enhanced processes such as plasma-enhanced atomic layer deposition (PE-ALD) can be effective for forming conformal films on surfaces with relatively low aspect ratios, these processes may not be effective for depositing films inside vias and cavities with relatively high aspect ratios. Without being limited to theory, one possible reason for this is that the plasma or its active component cannot reach the deeper parts of high aspect ratio vias in some situations. In such situations, different parts of the via are exposed to different amounts of plasma or its active component, which can cause undesirable structural effects of non-uniform deposition, such as a thicker film being deposited near the opening of the via compared to the deeper parts [sometimes called cusping or keyhole formation]. For this reason, thermal ALD may be more advantageous because it does not depend on the ability of the plasma or its active component to reach the surface portion being deposited.

[0014] However, while thermal ALD technology may be suitable for forming relatively conformable TiN films on topography, particularly on topography with a relatively high aspect ratio (e.g., greater than 1:1), the inventors have discovered that TiN films formed by thermal ALD may be inferior to TiN films formed by PVD or CVD in some aspects, such as film roughness and electrical resistivity. In this regard, the inventors have discovered that some electrical and / or physical properties of ALD-grown TiN-based films may be affected by the growth mode. Specifically, the inventors have discovered that while it may be desirable to grow TiN-based films in a two-dimensional layer-by-layer growth mode in ALD, such a layer-by-layer growth mode may not be easily achieved in some situations. The inventors have also discovered that growing TiN-based films in a layer-by-layer growth mode by ALD poses particular problems in IC manufacturing for forming TiN-based films on non-metallic surfaces, such as insulating surfaces like oxide and nitride surfaces, or semiconductor surfaces like doped silicon surfaces and undoped silicon surfaces. With reference to FIGS. 1a through 1d, the inventors have recognized that, as described herein without being bound by any particular theory, the extent to which TiN-based films can be grown in a layer-by-layer growth mode may depend on an initial growth mode that varies depending on the surface type.

[0015] FIG. 1a schematically illustrates the nucleation of a TiN layer, and FIG. 1b through 1d illustrate different growth modes of a TiN layer on different surfaces. Referring to FIG. 1a, when precursor molecules (104) reach the surface of a substrate (100), they are physically adsorbed thereon. Some of the adsorbed molecules (104) may diffuse along the surface of the substrate (100) until they reach a position energetically favorable for chemisorption. Surface diffusion is influenced, among other things, by the substrate temperature, the substrate material, and the kinetic energy of the adsorbed molecules. When the size of the nucleus formed by the chemisorbed molecules exceeds a certain size (sometimes referred to as the "critical size") determined by the balance between volume free energy and surface energy, the nucleus becomes energetically stable and may begin to grow in size. The layer (108) of the stable nucleus thus formed continues to grow by incorporating additional precursor molecules (104). As schematically illustrated in FIG. 1b through 1d, subsequent film growth can be classified according to different growth modes.

[0016] FIG. 1b schematically illustrates a three-dimensional island growth mode [sometimes referred to as the Volmer-Weber growth mode] that forms a layer (112) of three-dimensional islands. Without being bound by any theory, the island growth mode may be dominant when the net surface free energy associated with the three-dimensional islands is positive, which indicates that the deposited atoms are bonded more strongly to each other than to the substrate. It will be understood that the energy of the ALD growth of the TiN layer may encourage the island growth mode, for example, when a metallic TiN layer is deposited on the surface of some semiconductor and / or insulating material.

[0017] FIG. 1c illustrates a layer-by-layer growth mode [sometimes referred to as the Frank-van der Merwe growth mode] that results in the formation of a relatively smooth two-dimensional layer (116). Without being bound by any theory, the layer-by-layer growth mode may prevail when the deposited atoms are bonded more strongly to the substrate than to each other so that a stable two-dimensional layer (116) is energetically preferred. The layer-by-layer growth mode may be maintained when the binding energy between layers continuously decreases from the first single layer of the TiN layer to a bulk-crystalline value.

[0018] FIGS. 1b and 1c show two different possible growth modes of a thin film, but it can be seen that in some situations, an intermediate growth mode between the layer-by-layer growth mode and the three-dimensional growth mode is possible. FIG. 1d illustrates an example of an intermediate growth mode known as the Stranski-Krastanov (SK) growth mode. Without being bound by any theory, SK growth can occur in thin film growth that starts in a layer-by-layer mode. When layer-by-layer growth becomes undesirable after the formation of one or more single layers, the island growth mode begins to predominate over the layer-by-layer growth mode, creating a thin film structure (120) in which three-dimensional islands are formed on a two-dimensional initial layer. The SK growth mode can occur as a strain relaxation mechanism (strain-induced roughening).

[0019] In addition to the interaction between the deposit and the substrate, other factors such as substrate temperature, reactor pressure, and deposition rate can significantly influence the nucleation and initial growth processes, which in turn affect the final nanostructure or microstructure of the resulting thin film. Deposition conditions that enhance surface diffusion, such as relatively high substrate temperatures, relatively low pressure, and / or low deposition rates, can promote growth in a layer-by-layer mode. Thus, as disclosed herein, by enhancing surface diffusion during the deposition of an initial portion of the TiN film, for example by lowering the pressure and growth rate, the initial film growth according to the embodiments can substantially proceed in a layer-by-layer growth mode.

[0020] It has been found that when TiN is grown by ALD on various surfaces of interest in IC manufacturing, such as dielectric and semiconductor surfaces, the ALD growth can be initiated in a three-dimensional island growth mode or an SK growth mode. For example, in some situations, the ALD growth of TiN on substrate surfaces including doped and undoped Si, SiO2, Si3N4, and other high-K or low-K materials can proceed in an island growth mode or an SK growth mode. The inventors have discovered that, as illustrated in FIG. 2, due in part to the initial growth mode of the island or SK growth mode, the subsequent growth of TiN by ALD often results in undesirable film morphology for various coatings of ultrathin conformal TiN on structures with high aspect ratios.

[0021] Figure 2 is a cross-sectional transmission electron microscope image of a TiN layer grown by thermal ALD on a Si substrate coated with natural oxide. After growing the initial film in a 3D island or SK growth mode, the ALD growth of TiN is often characterized by the competitive growth of adjacent crystals with different orientations, resulting in V-shaped particles that are close to the nucleation layer in some situations and finish in a column shape at higher film thicknesses. As shown in Figure 2, the resulting film morphology includes a facetted column top that generates significant surface roughness and a column boundary having a lower density compared to the particles. It will be understood that the column boundary may have significantly worse diffusion barrier properties compared to the particles themselves and may act as the path with the least resistance to the transport of undesirable contaminants through the TiN layer.

[0022] The inventors have discovered that when forming an initial portion of a TiN layer on a non-metallic surface by exposing a substrate to Ti and / or N precursors at a relatively low pressure, e.g., less than 1 tor, the initial three-dimensional or SK growth mode may be suppressed and the layer-by-layer growth mode may be promoted during the initial stage of TiN deposition, e.g., the nucleation stage. Among other reasons, this may be particularly because it takes longer for the local diffusion of adsorbed Ti and N precursor molecules to diffuse and wet the substrate surface, particularly the non-metallic surface, at a relatively low contact angle. The TiN layer grown at a relatively low exposure pressure produces a layer that uniformly covers a large area of ​​the non-metallic surface without substantially forming islands, causing the initial stage of growth to tend to further encourage the layer-by-layer growth mode, whereas on a substrate surface where ALD TiN typically encourages the three-dimensional island or SK growth mode described above. Accordingly, by initiating ALD of TiN by exposing the substrate to Ti and / or N precursors at a relatively low precursor exposure pressure, e.g., less than 3 Torr, the generated initial layer can be grown in a layer-by-layer mode, e.g., during the nucleation stage. Subsequent bulk growth stages, which can be carried out by exposing the substrate to Ti and / or N precursors at a relatively high precursor exposure pressure, e.g., exceeding 3 Torr, can continue in a layer-by-layer mode. By using the method according to the embodiment, some of the disadvantages of conventional ALD of TiN (which may generally relate to column-type growth as described above following an initial growth characterized by an island or SK growth mode) can be avoided when a TiN layer is formed directly by ALD on an inorganic layer comprising some semiconductor and / or insulating materials, particularly Si, SiO2 and / or Si3N4.

[0023] FIG. 3a is a flowchart schematically illustrating an atomic layer deposition method (300) for forming a TiN layer by exposing a substrate to a plurality of cycles with different corresponding precursor exposure pressures according to an embodiment. The resulting film may have at least two regions formed at corresponding different exposure pressures. FIG. 3b schematically illustrates a cross-sectional view of a semiconductor structure (350) including a TiN layer formed by an atomic layer deposition method in which a substrate is exposed to a plurality of cycles with different corresponding precursor exposure pressures according to the method illustrated in FIG. 3a. Referring to FIG. 3a, the method (300) includes the step (310) of providing a substrate having a non-metallic surface to a reaction chamber configured for ALD, e.g., thermal ALD. The method (300) further includes the step (320) of forming a first portion of a thin film on a substrate by exposing the semiconductor substrate to one or more first ALD cycles, each including exposure to a first Ti precursor and a first N precursor at a first individual exposure pressure. The method (300) further comprises a subsequent step, such as a bulk deposition step, comprising forming a second portion of a thin film on a first portion of a thin film (330) by exposing a semiconductor substrate to one or more second ALD cycles, each comprising exposure to a second Ti precursor and a second N precursor at a second individual exposure pressure. Exposure to one or both of the Ti precursor and the N precursor during one or more second ALD cycles is performed at a higher pressure than the corresponding exposure to one or both of the Ti precursor and the N precursor during one or more first ALD cycles.

[0024] Referring to FIG. 3b, this is a cross-sectional view of a semiconductor thin film structure (350) comprising a substrate (360) having a non-metallic surface, e.g., a dielectric and / or semiconductor surface. First, a first portion (370) of a thin film containing TiN is formed on the substrate (360), and a second portion (380) of the thin film is formed on the first portion (370). The first portion and the second portion (370, 380) are formed by the atomic layer deposition method shown in FIG. 3a, which exposes the substrate (360) to different corresponding precursor exposure pressures for first and second cycles. Since the first part (370) can be grown in a layer-by-layer growth mode during an initial stage, for example, the nucleation stage as described above, at least the first part (370) or both the first part and the second part (370, 380) may not have adjacent crystals with different orientations characterized by columnar growth of substantially V-shaped crystals and relatively high (e.g., 10% of the thickness) surface roughness. The resulting TiN layer has superior properties including one or more of relatively high conformality or step coverage, lower surface roughness, smaller average grain size, higher electrical conductivity and / or barrier properties compared to similar thin film layers formed at a single pressure during the nucleation and bulk deposition stages.

[0025] As described throughout the present invention and specification, it will be understood that the semiconductor substrate on which the TiN thin film according to the embodiment is formed may be made of a variety of substrates, including but not limited to: a group IV elemental material (e.g., Si, Ge, C, or Sn) or an alloy of a group IV material (e.g., SiGe, SiGeC, SiC, SiSn, SiSnC, GeSn, etc.); a group III to V compound semiconductor material (e.g., GaAs, GaN, InAs, etc.) or an alloy formed of a group III to V material; or a doped semiconductor substrate that may be made of a group II to VI semiconductor material (CdSe, CdS, ZnSe, etc.) or an alloy formed of a group II to VI material.

[0026] According to specific embodiments, the substrate may also be implemented as a semiconductor on an insulator, such as a silicon on an insulator (SOI) substrate. An SOI substrate generally comprises a silicon-insulator-silicon structure isolated from a supporting substrate using an insulator layer, such as a SiO2 layer, in which the various structures described above are embedded. Additionally, it will be seen that the various structures described herein may be formed at least partially on an epitaxial layer formed on or near a surface region.

[0027] Additionally, the substrate may include topological features such as vias, cavities, holes, or trenches having one or more semiconductor or dielectric surfaces, and various structures formed thereon, such as diffusion regions, isolation regions, electrodes, vias, and lines, to name a few, and any structure including a TiN layer according to an embodiment may be formed thereon. Accordingly, the non-metallic surface on which the TiN layer according to an embodiment is formed may include a semiconductor surface, for example, a doped or undoped Si surface, and / or a dielectric surface, for example, an interlayer dielectric (ILD) surface, a mask or hard mask surface, or a gate dielectric surface (which may include inorganic insulators, oxides, nitrides, high K dielectrics, low K dielectrics, or carbon, to name a few dielectric materials).

[0028] As described throughout this Appendix and the specification, the term "reactor chamber" refers to any reaction chamber, including a single-wafer processing reaction chamber or a batch wafer processing reaction chamber, configured to be suitable for atomic layer deposition (ALD). In a thermal ALD reactor, the substrate may be placed on a suitable substrate holder, such as a susceptor or a carrier boat. The substrate may be heated directly by conduction through a heated susceptor, or indirectly by radiation from a radiation source, such as a lamp, or by convection through heated chamber walls.

[0029] Generally, in an ALD process, reactants or precursors, such as oxidation and reduction reactants, are alternately introduced into a reaction chamber in which a substrate is placed. The introduction of one or more reactants or precursors may also be alternated with a purge and / or pump-out process to remove excess reactants or precursors from the reaction chamber. Reactants may be introduced into the reaction chamber under conditions over a period suitable for ensuring that the surface of the substrate is at least partially saturated with the precursors or reactants and / or reaction products of the reactants. Subsequently, excess or residual precursors or reactants may be purged and / or pumped out of the reaction chamber to be removed from the substrate. The pump-out process may be performed by a suitable vacuum pumping process, and the purge step may be performed by introducing a non-reactive or inert gas, such as nitrogen or a noble gas, into the reaction chamber. In the example described below, with respect to the layer formed by thermal ALD, there are generally two categories of precursors or reactants: nitrogen (N) precursors and titanium (Ti) precursors.

[0030] Hereinafter, with reference to FIG. 4, an exemplary embodiment of a method (300) (Fig. 3a) for forming a thin film comprising at least two regions of TiN, formed by exposing a substrate to a plurality of cycles with a corresponding precursor exposure pressure different according to the embodiment by ALD, e.g., thermal ALD, is described.

[0031] Atomic layer deposition of TiN by exposing the substrate to multiple cycles at different corresponding precursor exposure pressures

[0032] Referring again to FIG. 3a, a substrate (substrate (360) of FIG. 3b) including a non-metal surface is provided (310) to a reaction chamber, and then the method (300) is carried out so that the semiconductor substrate is exposed to one or more first ALD cycles to form a first portion of a thin film on the non-metal surface, and then the semiconductor substrate is exposed to one or more second ALD cycles to form a second portion of a thin film. Below, the exposure pressure applied during the first and second ALD cycles is schematically described.

[0033] FIG. 4 schematically illustrates pressure traces corresponding to the exposure of a substrate to Ti and N precursors during a first cycle (400A) or step, e.g., a nucleation step, to form a first portion (370) (Fig. 3b) of a thin film according to various embodiments, and also during a second cycle (400B) or step, e.g., a bulk growth step, to form a second portion (380) (Fig. 3b) of a thin film. Referring to FIG. 4, the first portion of the thin film is formed by exposing a semiconductor substrate to one or more first ALD cycles (400A), each comprising one or more exposures (404) or exposure pulses to a partial pressure of a first Ti precursor and one or more exposures (408) or exposure pulses to a partial pressure of a first N precursor. A second portion of the thin film is formed by exposing a semiconductor substrate to one or more second ALD cycles (400B), each comprising one or more exposures (412) or exposure pulses to a partial pressure of a second Ti precursor and one or more exposures (416) or exposure pulses to a partial pressure of a second N precursor.

[0034] As schematically illustrated, each of the exposure to the first Ti precursor (404), the exposure to the first N precursor (408), the exposure to the second Ti precursor (412), and the exposure to the second N precursor (416) may have different partial pressure regimes, including corresponding partial pressure rising regimes (404A, 408A, 412A, and 416A), major exposure regimes (404B, 408B, 412B, and 416B), and partial pressure lowering regimes (404C, 408C, 412C, and 416C). Each partial pressure rising regime (404A, 408A, 412A, and 416A) may correspond, for example, to each precursor introduced into the reaction chamber. Each of the primary exposure regimes (404B, 408B, 412B, 416B) may correspond to a relatively constant amount of each precursor within the reaction chamber. A relatively constant amount of each precursor may be maintained, for example, using a pressure transducer or a throttle valve. Each of the partial pressure reduction regimes (404C, 408C, 412C, 416C) may correspond, for example, to a regime when individual precursors are purged or pumped out of the reaction chamber.

[0035] Referring still to FIG. 4, it will be understood that in some runs, the precursor may be pumped out and / or purge out after each exposure. In some runs where the precursor may be pumped out without purging, the reaction chamber pressure may be substantially represented by the partial pressure of each precursor, and the pressure trace of the exposures (404, 408, 412 and 416) may represent the reaction chamber pressure or the precursor partial pressure during substantially individual exposures. In some runs where the precursor is not pumped out but purged with an inert gas, the reaction chamber pressure may be represented by the total reaction chamber pressure (404P, 408P, 412P and 416P) corresponding to the exposures (404, 408, 412 and 416), where the total reaction chamber pressure originates from the mixture of each precursor and the inert gas.

[0036] In practice, a combination of pumping and purging may be used for higher throughput and improved membrane quality. In such execution, the substrate may receive partial pressures of the first Ti precursor, the first N precursor, the second Ti precursor, and the second N precursor while measuring the total pressure (404P, 408P, 412P, and 416P), including during purging and pumping. In some embodiments, the total chamber pressure may be maintained relatively constant throughout a given precursor exposure or exposure pulse while adjusting the pumping intensity using a pressure transducer and replacing the removed precursor with an inert gas. In such execution, one or more first ALD (400A) cycles to form the first part ((370) in FIG. 3b) may each include one or more exposures (404) to a partial pressure of the first Ti precursor (the measured parameter may be the total reaction chamber pressure (404P)), and one or more exposures (408) to a partial pressure of the first N precursor (the measured parameter may be the total reaction chamber pressure (408P)). Similarly, one or more second ALD cycles (400B) for forming the second part ((380) in FIG. 3b) may each include one or more exposures (412) to a partial pressure of the second Ti precursor (the measured parameter may be the total pressure (412P)), and one or more exposures (416) to a partial pressure of the second N precursor (the measured parameter may be the total pressure (416P)).

[0037] According to various embodiments, during exposure to the precursor, the measured total reaction chamber pressure may be proportional to the partial pressure of the precursor. Thus, the total pressures (412P and 416P), which are each higher relative total pressures (404P and 408P), correspond to the higher partial pressures of the second Ti precursor and the second N precursor, respectively, relative to the partial pressures of the first Ti precursor and the first N precursor, respectively. However, the embodiments are not limited to this, and in other embodiments, the total pressures (412P and 416P), which are each higher relative total pressures (404P and 408P), may correspond to the same or lower partial pressures of the second Ti precursor and the second N precursor, respectively, relative to the partial pressures of the first Ti precursor and the first N precursor, respectively.

[0038] Referring again to the method (300) illustrated in FIG. 3a, during one or more second ALD cycles in the later (e.g., bulk deposition step), one or both of the exposure pressures of the second Ti precursor and the second N precursor are higher than the corresponding one or both of the exposure pressures of the first Ti precursor and the first N precursor during one or more first ALD cycles in the earlier (e.g., nucleation step). In some embodiments, the exposure pressure may be a partial pressure of the precursor or the total pressure of the reaction chamber. Thus, in various embodiments, referring to FIG. 4, one or both of the exposure (412) to the second Ti precursor and the exposure (416) to the second N precursor may be performed at a higher partial pressure and / or a higher total reaction chamber pressure compared to the corresponding one or both of the exposure (404) to the first Ti precursor and the exposure (408) to the first N precursor, respectively.

[0039] Referring still to FIG. 4, in various embodiments, the corresponding partial pressure or total pressure between the corresponding exposures to the Ti and N precursors during the first cycle and the second cycle (400A, 400B) may be the corresponding partial pressure or total pressure during any one of the partial pressure rising regime (404A, 408A, 412A and 416A), the main exposure regime (404B, 408B, 412B and 416B), and the partial pressure falling regime (404C, 408C, 412C and 416C). For example, exposure (412, 416) to one or both of the second Ti precursor and the second N precursor during the respective major exposure regimes (412B and 416B) during the second ALD cycle (400B) may be performed at a higher total pressure or partial pressure compared to exposure (404, 408) to one or both of the first Ti precursor and the first N precursor during the respective major exposure regimes (404B and 408B) during the first ALD cycle (400A). In various other embodiments, the corresponding partial pressure or total pressure between the corresponding exposures to the Ti and N precursors during the first cycle and the second cycle (400A and 400B) may be the corresponding average, mean, or peak partial pressure or total pressure during the exposures (404, 408, 412, and 416).

[0040] Referring still to FIG. 4, in the illustrated embodiment, the total pressure and / or partial pressure during exposure to the first Ti precursor (404) and exposure to the first N precursor (408) is different, and the total pressure and / or partial pressure during exposure to the second Ti precursor (412) and exposure to the second N precursor (416) is different. However, the embodiment is not limited thereto, and in some embodiments, the total pressure and / or partial pressure during exposure to the first Ti precursor (404) and exposure to the first N precursor (408) may be kept constant and / or, or the total pressure and / or partial pressure during exposure to the second Ti precursor (412) and exposure to the second N precursor (416) may be kept constant.

[0041] Referring still to FIG. 4, the total pressure during exposure (404) to the first Ti precursor and exposure (408) to the first N precursor, which may be the same or different, may be a pressure within a range limited by any of the values, such as 0.01 to 0.2 Torr, 0.2 to 0.4 Torr, 0.4 to 0.6 Torr, 0.6 to 0.8 Torr, 0.8 to 1.0 Torr, 1.0 to 1.5 Torr, 1.5 to 2.0 Torr, 2.0 to 2.5 Torr, 2.5 to 3.0 Torr, or any of these values. The total pressure during each of the exposure (412) to the second Ti precursor, which may be the same or different, and the exposure (416) to the second N precursor may be a pressure within a range limited by any of the values, such as 3.0 to 4.0 Torr, 4.0 to 5.0 Torr, 5.0 to 6.0 Torr, 6.0 to 7.0 Torr, 7.0 to 8.0 Torr, 8.0 to 9.0 Torr, 9.0 to 10.0 Torr, 10.0 to 11.0 Torr, 11.0 to 12.0 Torr. The ratio of the total pressure (measured in Torr units) of the reaction chamber during exposure to the second Ti precursor (412) and exposure to the first Ti precursor (404) may be a range limited by 2 to 5, 5 to 10, 10 to 20, 20 to 50, 50 to 100, or any of these values. Similarly, the ratio of the total pressure of the reaction chamber during exposure to the second N precursor (416) and exposure to the first N precursor (408) may be a range limited by 2 to 5, 5 to 10, 10 to 20, 20 to 50, 50 to 100, or any of these values. In each of the exposures (404, 408, 412 and 416), each Ti or N precursor may constitute a percentage of the total amount of gas molecules in the reaction chamber, ranging from 1 to 2%, 2 to 5%, 5 to 10%, 10 to 20%, 20 to 50%, 50 to 100%, or any of these values.

[0042] Referring still to FIG. 4, according to various embodiments, the total pressure or partial pressure during exposure (404) to the first Ti precursor and exposure (408) to the first N precursor, along with the flow rates of each precursor and inert gas and the pumping intensity of the reaction chamber, is controlled so that the deposition rate during the first cycle (400A) or step is a value within a range limited by 0.10 to 0.20 Å / cycle, 0.20 to 0.30 Å / cycle, 0.30 to 0.40 Å / cycle, 0.40 to 0.50 Å / cycle, 0.50 to 0.60 Å / cycle, or any of these values ​​per cycle including exposure (404) to the first Ti precursor and exposure (408) to the first N precursor. Along with the flow rates of each precursor and inert gas and the pumping intensity of the reaction chamber, the total pressure or partial pressure during the exposure (412) to the second Ti precursor and the exposure (416) to the second N precursor is controlled so that the deposition rate during the second cycle (400B) or step is within a range limited by 0.20 to 0.30 Å / cycle, 0.30 to 0.40 Å / cycle, 0.40 to 0.50 Å / cycle, 0.50 to 0.60 Å / cycle, 0.60 to 0.70 Å / cycle, 0.60 to 0.70 Å / cycle, 0.70 to 0.80 Å / cycle, or any of these values ​​per cycle including the exposure (404) to the first Ti precursor and the exposure (408) to the first N precursor. The ratio of the deposition rate per cycle during the second cycle (400B) to the deposition rate per cycle during the first cycle (400A) may be a ratio in a range limited by 1 to 1.5, 1.5 to 2.0, 2.5 to 3.0, or any of these values.

[0043] The inventors have discovered that various technical advantages of the TiN thin film disclosed herein can be realized when forming a first portion (370) (Fig. 3b) of a thin film containing TiN (320) (Fig. 3a) and forming a second portion (380) (Fig. 3b) (330) (Fig. 3a), respectively, includes exposing a semiconductor substrate to a range of values ​​defined by 1 to 25 cycles, 26 to 50 cycles, 50 to 100 cycles, 100 to 200 cycles, 200 to 300 cycles, 300 to 400 cycles, 400 to 500 cycles, 500 to 600 cycles, or any of these values ​​of the first cycle (400A) (Fig. 4) and the second cycle (400B) (Fig. 4). According to various embodiments, the ratio of the number of second cycles to the number of first cycles may be greater than a ratio within a range limited by 1, 2, 5, or 10, or any of these values, or may be less than a ratio within a range limited by 1, 0.5, 0.1, or 0.1, or any of these values. The total thickness of the thin film containing TiN, comprising the first part (370) (Fig. 3b) and the second part (380) (Fig. 3b), may not exceed about 25 nm, 20 nm, 15 nm, 10 nm, 7 nm, 4 nm, or 2 nm, or may have a combined stack thickness having a value within a range limited by any of these values. The thickness ratio between the first part (370) (Fig. 3b) and the second part (380) (Fig. 3b) may be a ratio in a range limited by any of the values, such as about 1:20 to 1:10, 1:10 to 1:5, 1:5 to 1:2, 1:2 to 1:1, 1:1 to 2:1, 2:1 to 5:1, 5:1 to 10:1, 10:1 to 20:1, or any of these values.You will see that in some embodiments, for example, where higher conformality may be more important than lower film roughness, the first part (370) (Fig. 3b) may be relatively thinner, whereas in other embodiments, for example, where lower film roughness may be more important than higher conformality, the second part (380) (Fig. 3b) may be relatively thinner.

[0044] Referring still to FIG. 4, each of the substrate exposure (404) to the first Ti precursor and the substrate exposure (412) to the second Ti precursor causes the surface of the substrate to be substantially or partially saturated with the first Ti precursor or the second Ti precursor, respectively. After each of the substrate exposure (404) to the first Ti precursor and the substrate exposure (412) to the second Ti precursor, any excess or residual first and / or second Ti precursor or reaction products thereof that are not adsorbed or chemically adsorbed on the substrate surface may be pumped out and / or purge out.

[0045] Similarly, each of the substrate exposure (408) to the first N precursor and the substrate exposure (416) to the second N precursor causes the substrate to be substantially or partially saturated with the first N precursor or the second N precursor, respectively. After each of the substrate exposure (408) to the first N precursor and the substrate exposure (416) to the second N precursor, any excess or residual first and / or second N precursor or their reaction products that are not adsorbed or chemically adsorbed on the substrate surface may be pumped out and / or purge out. Applying the substrate to one or more exposures to the first Ti precursor and one or more exposures to the first N precursor may form approximately a single layer or less of TiN per cycle. Similarly, applying the substrate to one or more exposures to the second Ti precursor and one or more exposures to the second N precursor may form approximately a single layer or less of TiN per cycle.

[0046] In some embodiments, exposure to a first Ti precursor (404), exposure to a first N precursor (408), exposure to a second Ti precursor (412) and / or exposure to a second N precursor (416) may be performed sequentially multiple times before introducing another precursor. For example, advantageously, under some circumstances, exposing the substrate to the Ti precursor and / or N precursor more than once may result in a higher level of surface saturation, for example, when substantial steric hindrance effects are present.

[0047] Referring still to FIG. 4, it will be understood that the relative order of exposure to the first Ti precursor and the first N precursor can be selected depending on the competitive situation. In some implementations, the first Ti precursor may advantageously be the first precursor to which the substrate surface is exposed. For example, one or more direct exposures of the Si surface to the first Ti precursor can form one or more single layers of TiSi and prevent the formation of SiN, which may be advantageous for lowering the contact resistance between the underlying Si and the TiN layer formed thereon. However, in some other implementations, the first N precursor may advantageously be the first precursor to which the substrate is exposed. For example, by directly exposing the Si surface to the first N precursor, one or more single layers of SiN can be intentionally formed, which may be advantageous for improving the barrier properties of the stack.

[0048] In various embodiments, it will be seen that the frequency and number of repetitions of the substrate exposure to the first Ti reactant and / or the first N precursor in each of the first cycle (408A), and to the second Ti reactant and / or the second N precursor in each of the second cycle (408B), may be changed to obtain a desired thickness and stoichiometry based on various considerations, including sensitivity to steric hindrance effects of the precursor.

[0049] According to various embodiments, non-limiting examples of first and second Ti precursors, which may be the same or different for forming the first and second portions of the TiN layer according to the embodiments, include titanium tetrachloride (TiCl4), tetrakis(dimethylamino)titanium (TDMAT), or tetrakis(diethylamino)titanium (TDEAT). Having the same precursor for the first and second portions of TiN may be advantageous, for example, in terms of lower cost and / or easier process design. However, having different precursors for the first and second portions of TiN may be advantageous, for example, in terms of different deposition characteristics or film quality.

[0050] According to various embodiments, non-limiting examples of first and second N precursors, which may be the same or different for forming the first and second portions of the TiN layer according to the embodiments, include ammonia (NH3), hydrazine (N2H4), or monomethylhydrazine (CH3(NH)NH2, "MMH"). Having the same precursor for the first and second portions of TiN may be advantageous, for example, in terms of lower cost and / or easier process design. However, having different precursors for the first and second portions of TiN may be advantageous, for example, in terms of different deposition characteristics or film quality.

[0051] According to various embodiments, non-limiting examples of inert gases for purging may include nitrogen N2, or noble gases such as Ar or He.

[0052] When one or both of the first and second portions (370, 380) (Fig. 3b) of the thin film containing TiN are formed at a substrate temperature of 350°C to 800°C, 450°C to 750°C, 500°C to 700°C, 550°C to 650°C, or any of these values, for example, about 600°C, the various technical advantages and benefits described herein may be realized. Maintaining the same temperature during the growth of the first and second portions (370, 380) may be advantageous for easy process control and throughput, as temperature adjustment during the process can take a long time.

[0053] In various embodiments, the exposure time or pulse time of each of the first and second Ti precursors and the first and second N precursors may be a duration of about 0.1 to 1 second, 1 to 10 seconds, 10 to 30 seconds, 30 to 60 seconds, or any of these values.

[0054] Advantageously, when a TiN layer is formed using an atomic layer deposition method in which the substrate is exposed for multiple cycles at different corresponding precursor exposure pressures according to various embodiments, one or both of the surface roughness and electrical resistivity may be substantially reduced to that of a conventional TiN film including a TiN film formed using a different ALD process with a single pressure setpoint. As deposited, a thin film containing TiN formed according to the method described herein and having the thickness and the thickness ratio between the first part and the second part (370, 380) (Fig. 3b) may have a root mean square (RMS) surface roughness of a range of values ​​limited by 3%, 4%, 5%, 6%, 7%, 8%, and 9%, or any one of these values, based on the average thickness of the thin film. Alternatively, in the deposited state, a thin film containing TiN having the aforementioned thickness and a thickness ratio between the first part and the second part (370, 380) (Fig. 3b) may have an RMS surface roughness value less than a value in a range limited by 2.5 nm, 2 nm, 1.5 nm, 1.0 nm, 0.5 nm, or any of these values.

[0055] In a deposited state, a thin film comprising TiN formed according to the method described herein and having the thickness and the ratio of the first part and the second part (370, 380) (Fig. 3b) as described above may have an electrical resistivity of less than 70 μΩ-cm, 70 to 100 μΩ-cm, 100 to 130 μΩ-cm, 130 to 160 μΩ-cm, 160 to 190 μΩ-cm, 190 to 220 μΩ-cm, 220 to 250 μΩ-cm, 250 to 280 μΩ-cm, 280 to 310 μΩ-cm, or greater than 310 μΩ-cm, or a range of values ​​limited by any of these values ​​(e.g., less than about 200 μΩ-cm).

[0056] In addition to reduced surface roughness and electrical resistivity, a thin film containing TiN formed according to the method disclosed herein has high conformality when deposited on a structure with a high aspect ratio. In the context of a structure with a high aspect ratio, one measure of conformality is referred to herein as step coverage. A structure with a high aspect ratio may be, for example, vias, holes, trenches, cavities, or similar structures. As an exemplary example, FIG. 5 schematically illustrates a semiconductor structure (500) on which an exemplary structure with a high aspect ratio (516) is formed to illustrate some exemplary metrics for defining and / or measuring the conformality of a thin film formed on a structure with a high aspect ratio. The illustrated structure with a high aspect ratio (516) is lined with a TiN layer (512) having different thicknesses in different parts. As described herein, a structure with a high aspect ratio has an aspect ratio greater than 1, for example, a ratio in which the depth or height (H) in the opening region of the structure with a high aspect ratio (516) is divided by the width (W). In the illustrated example, the high aspect ratio structure (516) is a via formed through a dielectric layer (508), for example an intermetallic dielectric (ILD) layer, formed on a semiconductor substrate (504) so ​​that the bottom surface of the high aspect ratio structure (516) exposes the semiconductor (504) placed underneath. The TiN layer (512) may coat different surfaces of the high aspect ratio structure (516) with different thicknesses. As described herein, one metric for defining or measuring the conformality of a thin film formed with a high aspect ratio is called step coverage. Step coverage may be defined as the ratio between the thin film thickness in the lower or bottom region of the high aspect ratio structure and the thin film thickness in the upper or top region of the high aspect ratio structure. The upper or top region may be a region of the high aspect ratio structure at a relatively small depth of, for example, 0 to 10% or 0 to 25% of H measured from the top of an opening.The lower or bottom area may be an area of ​​a high aspect ratio structure at a relatively large depth of, for example, 90 to 100% or 75 to 100% of H measured from the top of an opening. In some high aspect ratio structures, step coverage may be defined or measured by the ratio of the thickness of the film (512A) formed on the lower surface of the high aspect ratio structure to the thickness of the film (512C) formed on the upper or top sidewall surface. However, it will be understood that some high aspect ratio structures may not have a well-defined bottom surface or a bottom surface with a small radius of curvature. In such structures, step coverage may be more consistently defined or measured by the ratio of the thickness of the film (512B) formed on the lower or bottom sidewall surface of the high aspect ratio structure to the thickness of the film (512C) formed on the upper or top sidewall surface.

[0057] As described above, a thin film containing TiN formed according to the method disclosed herein results in reduced surface roughness and electrical resistivity, while providing high conformability in structures with a high aspect ratio. According to various embodiments, structures with a high aspect ratio having an aspect ratio exceeding a range of values ​​defined by any of these values, such as 1, 2, 5, 10, 20, 50, 100, 200, or any of these values, can be conformally coated with a TiN film according to the embodiment with a step coverage defined herein exceeding 70%, 80%, 90%, 95%, or having a range of values ​​defined by any of these values.

[0058] Physical characterization of TiN formed by exposing a substrate to multiple cycles at different corresponding precursor exposure pressures

[0059] FIG. 6 is a graph showing the experimentally measured root mean square (RMS) surface roughness trend (604) and step coverage trend (608) as a function of the number of first cycles of exposure to Ti and N precursors at a relatively low chamber pressure of 0.5 Torr among a total of 600 combined first cycles (e.g., nucleation step) and second cycles (e.g., bulk deposition step). The second cycles of exposure to Ti and N precursors were performed at a relatively high chamber pressure of 5 Torr. Each experimental data point in FIG. 6 was taken from a TiN film grown on a natural SiO2-coated Si substrate and a TiN film grown in vias formed on SiO2 having an aspect ratio of about 40:1 for surface roughness measurements. The measured deposition rates for the first and second cycles were 0.28 Å / cycle and 0.38 Å / cycle, respectively. Experimental data were measured for four different TiN films grown with 0 first cycles (0 Å) / 600 second cycles (228 Å), 50 first cycles (14 Å) / 550 second cycles (209 Å), 200 first cycles (56 Å) / 400 second cycles (152 Å), and 600 first cycles (168 Å) / 0 second cycles (0 Å). The total thicknesses of the four TiN films were approximately 228 Å, 223 Å, 208 Å, and 168 Å, respectively. As previously mentioned, the measured surface roughness values ​​of the TiN films decrease as the relative number of first cycles involving exposure at relatively lower pressures increases. Unbound by any theory, this may be because lower growth rates tend to allow more surface diffusion, which tends to reduce surface roughness and promote layer-by-layer growth.The measured surface roughness values ​​for thin films grown with 0 first cycles / 600 second cycles, 50 first cycles / 550 second cycles, and 200 first cycles / 400 second cycles were approximately 21 Å, 17.5 Å, and 12.5 Å, respectively, corresponding to about 9%, 8%, and 6% based on the total thickness of the individual TiN films. Additionally, as discussed above, the measured step coverage values ​​of the TiN films were higher in the thin films grown with 0 first cycles / 600 second cycles compared to the films grown with 600 first cycles / 0 second cycles. Without being bound by any theory, this may be because higher pressure tends to cause more precursors to reach the bottom of a higher aspect ratio, which tends to improve step coverage. However, surprisingly, the inventors found that increasing the number of first cycles up to about 50 times (8% of the total number of cycles) actually improves step coverage. Accordingly, according to some embodiments, the step of forming a first portion of a TiN film comprises alternately exposing a semiconductor substrate to 1 to 50 cycles, each including exposure to a first Ti precursor and exposure to a first N precursor, at a relatively low exposure pressure of about 3 tor.

[0060] FIGS. 7a through 9 illustrate additional experimental comparisons between TiN films grown by exposing a substrate to a cycle at the same precursor exposure pressure and TiN films grown by exposing a substrate to multiple cycles at different corresponding precursor exposure pressures according to the present embodiment. FIG. 7a is a cross-sectional transmission electron microscope image of a high aspect ratio via lined with a TiN layer formed by an atomic layer deposition method in which the substrate is exposed to an ALD cycle at the same precursor exposure pressure corresponding to the second cycle. FIGS. 7b and 7c are transmission electron microscope (TEM) images of a TiN film grown using only the second cycle of exposure to Ti and N precursors at a relatively high chamber pressure of 5 Torr. The TEM images are of a via formed in SiO2 having an aspect ratio of about 40:1, taken in the upper (Fig. 7b) and lower (Fig. 7c) regions of the via. In contrast, FIGS. 8a and 8b are transmission electron microscope (TEM) images of a TiN film grown using a combination of first and second cycles of exposure to Ti and N precursors at a relatively low chamber pressure (0.5 Torr) and a relatively high chamber pressure (5 Torr), depending on the embodiment. The TEM images are of vias formed in SiO2 having an aspect ratio of about 40:1, taken from the upper (Fig. 8a) and lower (Fig. 8b) regions of the via. FIG. 9 is a graph illustrating an experimental and statistical comparison between the step coverage (904) measured from the TEM microscope images shown in FIGS. 7a through 7c and the step coverage (908) measured from the TEM microscope images shown in FIG. 8a and 8b. The data points in FIG. 9 represent ratios taken from different locations within the lower region of the via and different locations within the upper region of the via. Although not easily visible in the TEM image, the statistical comparison in Fig. 9 clearly shows the higher intermediate step coverage of 93% for the TiN film deposited according to the embodiment and 87% for the TiN film deposited using a single exposure pressure.Furthermore, the statistical spread of the step coverage measured for the TiN film deposited according to the embodiment is substantially smaller than that for the TiN film deposited using a single exposure pressure, which indicates that the film roughness is significantly higher in the latter case.

[0061] use

[0062] Thin films comprising TiN formed using different exposure pressures according to various embodiments disclosed herein may be used in various applications, particularly those having structures and / or non-metallic surfaces with a relatively high aspect ratio, where the substrate can benefit from various advantageous properties of the TiN layer disclosed herein. Exemplary applications include depositing vias, holes, trenches, cavities, or similar structures having an aspect ratio (e.g., a ratio defined as depth divided by top width) exceeding a range of values ​​defined by 1, 2, 5, 10, 20, 50, 100, 200, or any of these values.

[0063] For example, FIG. 10 illustrates an application related to a diffusion barrier for a contact structure (e.g., source or drain connection) formed on an active semiconductor substrate region that can be doped to a high concentration. A portion of a semiconductor device (1000) is illustrated, comprising a substrate (1004) on which an interlayer or intermetallic dielectric (ILD) layer is formed, the dielectric layer (1008), which is a dielectric material such as oxide or nitride. To form connections for various regions of the substrate (1004), including various doped regions, e.g. source and drain regions, vias or trenches may be formed through the dielectric layer (1008). The vias or trenches may expose the dielectric sidewalls of the vias as well as an exposed bottom surface, which may include various non-metallic surfaces, e.g., a substrate surface (e.g., a silicon substrate surface). The bottom surface and side of the via may be conformally coated with a first portion of a TiN layer (corresponding to the first portion (370) in FIG. 3b) formed according to various embodiments described herein, followed by a second portion (corresponding to the second portion (380) in FIG. 3b). According to various embodiments disclosed herein, the first conformal portion may first be formed directly on the inner surface of the via, and then a second conformal TiN layer may be formed. The lined via may then be filled with a metal, for example, W, Al, or Cu, to form a connection plug (1016). For example, the via may be filled with tungsten by CVD using, for example, WF6.

[0064] The barrier layer (1012) formed according to the embodiment may be advantageous for various reasons. Specifically, due to the conformal characteristics of the barrier layer (1012) formed by ALD, the tendency for pinching off during the subsequent metal filling process may be substantially reduced. Additionally, as described above, the barrier layer (1012) can effectively prevent material transport across it, such as the external diffusion of dopants (B, P) from the substrate (1004) as well as the internal diffusion of reactants, etchants, and metals (e.g., F, Cl, W, or Cu) from the connection plug forming process. The barrier effect may be enhanced by reduced surface roughness and increased step coverage. Furthermore, as described above, the layer-by-layer growth mode may reduce the overall contact resistance of the barrier layer (1012). Moreover, due to the reduced surface roughness, a relatively thinner barrier layer (1012) may be formed while still achieving the desired barrier function, thereby further reducing the contact resistance.

[0065] Other uses of the TiN layer formed according to various embodiments disclosed herein include a recessed substrate (e.g., a buried electrode or line), an electrode (e.g., a DRAM capacitor electrode or a gate electrode), a metallization barrier for a higher metal level (e.g., a barrier for a via / trench for a Cu connection / line), a vertical rod electrode with a high aspect ratio, or a conductive structure formed in vias and through-silicon vias (TSVs) for three-dimensional memory.

[0066] Although the present invention has been described herein with reference to specific embodiments, these embodiments are not intended to limit the invention and are presented for illustrative purposes only. It will be apparent to those skilled in the art that modifications and improvements may be made without departing from the spirit and scope of the invention.

[0067] Such simple modifications and improvements to the various embodiments disclosed herein are within the scope of the disclosed technology, and the specific scope of the disclosed technology will be further limited by the appended claims.

[0068] From the foregoing, it will be understood that any feature of any one of the embodiments may be combined with or replaced with any other feature of any other embodiment.

[0069] Unless the context clearly requires otherwise, throughout the Detailed Description and claims, words such as “include,” “including,” etc., should be understood in an inclusive sense—that is, “includes but not limited to these”—rather than in an exclusive or exhaustive sense. The word “coupled” as used herein generally refers to two or more elements that may be directly connected or connected through one or more intermediate elements. Likewise, the word “connected” as used herein generally refers to two or more elements that may be directly connected or connected through one or more intermediate elements. Additionally, “in this,” “above,” “below,” and words of similar meaning, when used herein, refer to the entirety of this, rather than a specific part of this. Where the context permits, words in the above Detailed Description using the singular or plural may each include the plural or singular, respectively. With respect to a list of two or more items, the word “or” includes all of the following interpretations: any item in the list, all items in the list, and any combination of items in the list.

[0070] Furthermore, conditional language used herein, e.g., “can,” “able to,” “e.g.,” “e.g.,” “e.g.,” “e.g.,” “e.g.,” is generally intended to convey the meaning that a particular embodiment includes a particular feature, element, and / or state while another embodiment does not include them, unless specifically otherwise stated or otherwise understood within the context in which they are used. Accordingly, such conditional language is not generally intended to imply that a feature, element, and / or state is required in any one or more embodiments, or to imply whether such feature, element, and / or state must be included in or performed in any particular embodiment.

[0071] Although specific embodiments have been described, these embodiments are presented merely as examples and are not intended to limit the scope of the disclosure. In practice, the novel apparatus, method, and system described herein may be implemented in various other forms; furthermore, various omissions, substitutions, and modifications may be made to the forms of the method and system described herein without departing from the spirit of the disclosure. For example, while features are presented in a given arrangement, other embodiments may perform similar functions with different components and / or sensor topologies, and some features may be deleted, moved, added, subdivided, combined, and / or modified. Each of these features may be implemented in various different ways. Any suitable combination of elements and actions of the various embodiments described above may be combined to provide additional embodiments. The various features and processes described above may be implemented independently of each other or may be combined in various ways. All possible combinations and sub-combinations of the features of the disclosure are intended to be within the scope of the disclosure.

Claims

Claim 1 A method for forming a thin film containing titanium nitride (TiN) by a cyclic vapor deposition process, comprising: a step of forming a first portion of a thin film on a semiconductor substrate by exposing the semiconductor substrate to one or more first cyclic vapor deposition cycles, each including exposure to a first Ti precursor and exposure to a first N precursor; and a step of forming a second portion of a thin film perpendicularly on the first portion of the thin film by exposing the semiconductor substrate to one or more second cyclic vapor deposition cycles, each including exposure to a second Ti precursor and exposure to a second N precursor, wherein the exposure to one or both of the second Ti precursor and the second N precursor during the one or more second cyclic vapor deposition cycles is performed at a higher pressure than the corresponding exposure to one or both of the first Ti precursor and the first N precursor during the one or more first cyclic vapor deposition cycles. Claim 2 A method according to claim 1, wherein exposure to one or both of the first Ti precursor and the first N precursor during one or more first cyclic vapor deposition cycles is carried out at a reactor pressure of less than 1 Torr, and exposure to one or both of the second Ti precursor and the second N precursor during one or more second cyclic vapor deposition cycles is carried out at a reactor pressure greater than 5 Torr. Claim 3 A method according to claim 1, wherein exposure to the second Ti precursor and the second N precursor, respectively, during one or more second cyclic vapor deposition cycles is performed at a higher pressure compared to the corresponding exposure to the first Ti precursor and the first N precursor, respectively, during one or more first cyclic vapor deposition cycles. Claim 4 A method according to claim 1, wherein the step of forming a first portion of a thin film comprises depositing at a first deposition rate of less than 0.3 Å per first cyclic vapor deposition cycle, which includes exposure of a semiconductor substrate to a first Ti precursor and exposure to a first N precursor, and the step of forming a second portion of a thin film comprises depositing at a second deposition rate greater than 0.3 Å per second cyclic vapor deposition cycle, which includes exposure of a semiconductor substrate to a second Ti precursor and exposure to a second N precursor. Claim 5 A method according to claim 1, wherein the root mean square surface roughness of the thin film is less than 8% of the thin film thickness. Claim 6 A method according to claim 1, wherein the step of forming the first and second portions of the thin film is performed by thermal cyclic vapor deposition. Claim 7 A method according to claim 1, wherein the step of forming a first portion of a thin film comprises directly exposing one or both of a semiconductor surface and an insulator surface to one or more first cyclic vapor deposition cycles. Claim 8 A method according to claim 1, wherein the step of forming one or both of the first and second parts of a thin film comprises growing in a layer-by-layer growth mode. Claim 9 A method according to claim 1, wherein the step of forming a first portion of a thin film comprises alternately exposing a semiconductor substrate to 1 to 50 first cyclic vapor deposition cycles. Claim 10 A method according to claim 1, wherein the step of forming the first and second portions of the thin film is formed at a temperature of 400°C to 600°C. Claim 11 A method according to claim 1, wherein a semiconductor substrate comprises a trench or via having an inner surface including a non-metallic sidewall surface within the trench or via having an aspect ratio greater than 1, and the step of forming a thin film comprises conformally lining the inner surface, and wherein the ratio of the thickness of the thin film formed at the lower 25% of the trench or via height and the upper 25% of the trench or via height exceeds 0.

9. Claim 12 A method according to claim 1, wherein the electrical resistivity of the thin film is less than 200 μΩ-cm. Claim 13 In claim 1, the method wherein the first Ti precursor is identical to the second Ti precursor, and the first N precursor is identical to the second N precursor. Claim 14 A step of providing a semiconductor substrate including a trench or via having an aspect ratio greater than 1; A method for forming a thin film containing titanium nitride (TiN) by a cyclic vapor deposition process, comprising the step of forming a thin film in a trench or via by exposing a semiconductor substrate to one or more first cyclic vapor deposition cycles, each including exposure to a first Ti precursor and exposure to a first N precursor, and then forming a thin film in a trench or via by exposing the semiconductor substrate to one or more second cyclic vapor deposition cycles, each including exposure to a second Ti precursor and exposure to a second N precursor, thereby forming a second thin film perpendicular to the first thin film. Claim 15 A method according to claim 14, wherein exposure to one or both of the second Ti precursor and the second N precursor during one or more second cyclic vapor deposition cycles is performed at a higher pressure compared to the corresponding exposure to one or both of the first Ti precursor and the first N precursor during one or more first cyclic vapor deposition cycles. Claim 16 A method according to claim 15, wherein exposure to one or both of the second Ti precursor and the second N precursor during one or more second cyclic vapor deposition cycles is performed at a pressure at least five times higher than the corresponding exposure to one or both of the first Ti precursor and the first N precursor during one or more first cyclic vapor deposition cycles. Claim 17 A method according to claim 14, wherein the step of forming a first portion comprises alternately exposing a semiconductor substrate to 1 to 50 first cyclic vapor deposition cycles. Claim 18 A method according to claim 14, wherein the step of forming a first part comprises exposing a semiconductor substrate to a first first cyclic vapor deposition cycle, and the step of forming a second part comprises exposing the semiconductor substrate to a second second cyclic vapor deposition cycle that is more than twice the first number of the first cyclic vapor deposition cycles. Claim 19 In claim 14, a method wherein the root mean square surface roughness of the thin film is less than 8% of the thin film thickness. Claim 20 A method according to claim 14, wherein the step of forming a thin film containing TiN is formed by thermal cyclic vapor deposition at a temperature of 400°C to 600°C. Claim 21 A method according to claim 14, wherein the trench or via has an inner surface including an aspect ratio exceeding 5 and a non-metallic sidewall surface, the step of forming a thin film comprises conformally lining the inner surface, and the ratio of the thickness of the thin film formed at the lower 25% of the trench or via height to the upper 25% of the trench or via height exceeds 0.

9. Claim 22 In claim 14, the method in which the first Ti precursor is identical to the second Ti precursor and the first N precursor is identical to the second N precursor. Claim 23 delete Claim 24 delete Claim 25 delete Claim 26 delete Claim 27 delete Claim 28 delete Claim 29 delete

Images