Manufacturing process for electrical components and electrode assembly
The method forms a ultra-thin barrier layer by plasma nitriding and embedding a second electrode layer in ultra-thin trenches using a PCM sputtering device, addressing issues of trench opening reduction and metal diffusion, enhancing the performance and yield of semiconductor components.
Patent Information
- Application Number
- DE112012005736
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2012-01-24
- Filing Date
- 2012-09-25
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2032-09-25
AI Technical Summary
Existing methods for forming metal layers in ultra-thin trenches of semiconductor integrated circuits face challenges such as trench opening reduction, metal diffusion, and unevenness, which affect the performance and yield of electronic components.
A manufacturing method involving the formation of a first electrode formation layer, followed by plasma nitriding to create a ultra-thin barrier layer, and then embedding a second electrode formation layer without an additional seed layer, using a PCM sputtering device with a point-cusp magnetic field to control plasma density and ionization.
This method enables metal layer embedding in trenches as small as 15 nm without reducing the opening diameter, improving yield and reducing metal diffusion, resulting in a flat and uniform metal wiring layer with enhanced performance.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
Technical area
[0001] The invention relates to a manufacturing method for an electronic component and an electrode structure, and more particularly to a manufacturing method for an electronic component and an electrode structure, the method comprising a step of embedding a metal layer in a recess formed in a workpiece. State of the art
[0002] Conventional semiconductor integrated circuits have used a gate-first process in which a gate insulating layer and a gate electrode are first formed on a wafer surface, followed by etching. In recent years, element downsizing has advanced the thinning of a gate insulating layer in a MOSFET. When a SiO2 layer is used as the gate insulating layer, a tunnel current is generated if the layer thickness is 2 nm or less, which is a currently required value, and thus a gate leakage current increases. To counteract this, recent thought has been given to replacing a gate insulating layer material with a higher dielectric material having a relative permittivity higher than that of the SiO2 layer. Such a process can reduce the SiO2 equivalent thickness (EOT).Equivalent Oxide Thickness (EOT) can decrease even if the actual insulating layer thickness is increased. There is a need to further reduce the EOT required for a MOSFET with a gate length of 22 nm or less. To meet this demand, the actual insulating layer thickness must be increased by using a high dielectric material, and thereby the gate leakage current must be reduced. However, in the gate-first process, a source / drain formation step is performed after the gate formation. Thus, the gate insulating layer and the gate electrode are heated, and the heating of the insulating layer and the metal layer causes diffusion. This leads to problems such as mobility degradation and a shift in the operating voltage (Vt).To solve such problems, a gate-last process has been intensively studied and developed, in which source / drain formation is performed first and a gate insulating layer and a gate electrode are formed last. Since the gate region is formed last in this process, it is believed that the heating temperature in the gate region can be lowered and that mobility degradation and operating voltage (Vt) drift, which are problems with the gate-first process, can be suppressed. What should be achieved by the gate-last process is to form various metal layers on a mold having a 15 nm or smaller opening and a 30 nm or more depth (hereinafter referred to as a trench), and to control a material deposited on a sidewall or bottom of the trench to have a desired thickness.Since different metal layers are stacked, interdiffusion between the metal layers must also be suppressed.
[0003] In the gate-last method, examples of the method for forming the various metal layer materials include a CVD (chemical vapor deposition) method, an atomic layer adsorption deposition method, and a sputtering method. The CVD method has problems with film thickness controllability, in-plane uniformity, and reproducibility due to an incubation period during formation. The atomic layer adsorption deposition method has excellent film thickness controllability but also a cost problem because the growth time increases when forming a thick film and an expensive source material gas is used. In these methods, which utilize a chemical reaction of source material gases, a film can be uniformly formed not only on the bottom of the trench but also on its sidewall. On the other hand, an increase in the deposition thickness narrows the trench opening.As a means to solve these problems, a method has been proposed to form various metal layer materials by a sputtering method with excellent layer thickness controllability, in-plane uniformity, and reproducibility.
[0004] For example, JP 2004-506090 A discloses a device capable of forming a film on a sidewall of a trench by a sputtering process under a high pressure of 133,322 Pa (= 1 Torr) or more, similar to the CVD process. In this process, the directionality of sputter ions relative to a wafer surface is suppressed by sputtering under high pressure. Thus, film formation can also occur on the sidewall of the trench. JP 3 193 875 B2 discloses a technique and device for forming layered Ti and TiN barrier substrates using the sputtering process, then forming an Al seed layer to facilitate the migration of an Al layer, and then embedding Al by high-temperature migration. In this process, Al can be embedded in the trench, while Al diffusion is suppressed with the layered Ti and TiN barrier substrates.
[0005] WO 2011 / 081202 A1 discloses a manufacturing method for an electronic component for embedding a low melting point metal in a recess having excellent scattering power by forming barrier layers with sufficient barrier properties and sufficient wettability and preventing or reducing the aggregation of the low melting point metal deposited at high temperature.This method comprises the following steps: forming a first barrier layer of TiNx on a workpiece by plasma treatment while applying a first bias power to an electrode in contact with the workpiece; forming a second barrier layer of TiNx on the first barrier layer by plasma treatment while applying a second bias power to an electrode, the second bias power providing a lower ion injection energy than the first bias power; and embedding the low melting point metal on the second barrier layer.
[0006] JP 2011-153374 A discloses a manufacturing method for an electronic component for enabling the embedding of Al even in a very small trench having an opening diameter of 22 nm or less without reducing the opening diameter. This method comprises: a first step of sputtering while forming a magnetic field on a target surface by a magnet unit having a plurality of magnets arranged at lattice points of a polygonal lattice, the magnets being arranged such that adjacent magnets have opposite poles; the first step forms a barrier layer containing titanium nitride on a workpiece having a recess formed therein; and a second step of embedding a low-melting-point metal layer directly on the barrier layer under a temperature condition that allows the low-melting-point metal layer to flow.
[0007] JP 2007 - 295 004 A discloses a manufacturing method for an electronic component in which a titanium layer is formed in a recess by sputtering and this titanium layer is then nitrided by heating in a nitrogen atmosphere.
[0008] Finally, reference is made to the subsequently published US 8 835 296 B2, which discloses an electrode structure according to claim 14 and a manufacturing method for an electronic component, in which a titanium layer is formed in a recess by sputtering and nitrogen gas is introduced during the sputtering in order to nitride the titanium layer.
[0009] In addition, reference is made to the documents US 2003 / 0 024 478 A1 and JP 2002 - 198 441 A.
[0010] As described above, the trench opening diameter is reduced during layer formation into a recently very small pattern by stacking multiple metal layers. To address this, a metal layer formation technique is required that can minimize the reduction in the opening diameter even when stacking multiple metal layers. Furthermore, in Al embedding, it has been known that Al diffusion degrades the properties of a metal layer used for a gate electrode region. Thus, a formation technique for an ultra-thin barrier layer that suppresses Al diffusion is also required. Brief description of the invention
[0011] However, the above techniques each have the following problems.
[0012] The method described in JP 2004-506090 A, which involves sputtering at a high pressure of 133.322 Pa (= 1 Torr) or more, has the problem that, although deposition can occur on the sidewall of the trench, the trench opening is narrowed when the trench opening is reduced to 15 nm or less. Furthermore, the Al embedding method described in JP 3 193 875 B2 has the problem that the layered Ti and TiN barrier layer must be deposited thickly to suppress Al diffusion. Another problem is that the trench opening is narrowed because the Al seed layer is also formed on the layered Ti and TiN barrier layer to facilitate Al migration.
[0013] Meanwhile, in the method described in WO 2011 / 081202 A1, a first barrier layer and a second barrier layer must be formed so that agglomeration of a low-melting-point metal deposited at a high temperature can be prevented or reduced and a recess with high scattering power can be filled with the low-melting-point metal. This leads to the problem of increasing the thickness of the barrier layers. Meanwhile, the method described in JP 2011-153374 A enables Al to be embedded even in a very small trench with an opening diameter of 22 nm or less without reducing the opening diameter. However, in the method described in JP 2011-153374 A, the Al layer must be embedded under a temperature condition that allows the layer to flow.In the case of a very small trench with an opening diameter of 15 nm or less, there is a concern that unevenness on the surface of the Al layer may affect the performance of the electronic components.
[0014] The invention was made taking the above-described conventional problems into consideration. It is an object of the invention to provide a manufacturing method for an electronic component capable of solving the above-described problems, suppressing the reduction of the trench opening, and suppressing the diffusion of a metal layer embedded in a trench.
[0015] As a result of intensive studies to solve the above problem, the inventors have completed the invention by finding that a second electrode formation layer (e.g., an Al wiring layer) can be easily embedded without unevenness directly on an ultra-thin barrier layer (e.g., a TiAlN layer) even if no Al seed layer is additionally formed between the ultra-thin barrier layer (e.g., the TiAlN layer) and the second electrode formation layer (e.g., the Al wiring layer) by forming a first electrode formation layer (e.g., a TiAlN layer) in a recess (e.g., a trench) formed in a workpiece (first step), forming the ultra-thin barrier layer (e.g., the TiAlN layer) bythe TiAl layer) is plasma nitrided to form a nitriding layer (second step), and then the second electrode formation layer (e.g., the Al wiring layer) is formed on the ultra-thin barrier layer (e.g., the TiAlN layer) (third step).
[0016] To achieve the above object, more specifically, an embodiment of the invention is a manufacturing method for an electronic component having the features of claim 1.
[0017] According to the present invention, metal can be embedded even in a very small trench having an opening diameter of 15 nm or less while suppressing the reduction of an opening diameter by forming a first electrode formation layer (e.g., a TiAl layer) formed on a workpiece having a recess (e.g., a trench) formed therein (first step), by forming an ultra-thin barrier layer (e.g., a TiAl layer) by forming a nitride layer by plasma nitriding a surface of the first electrode formation layer (e.g., the TiAl layer) (second step), and by forming a second electrode formation layer (e.g., an Al wiring layer) on the ultra-thin barrier layer (e.g., the TiAl layer) (third step).Therefore, even when the electronic component manufacturing method of the present invention, which includes the step of embedding a metal layer, is applied to a wiring manufacturing method, a metal layer can be embedded while suppressing the reduction of an opening diameter of a very small trench having an opening diameter of 15 nm or less. Furthermore, in an electrode structure formed by the electronic component manufacturing method of the present invention, a metal wiring layer is formed flatly without unevenness on an ultra-thin barrier layer (e.g., a TiAlN layer) while suppressing the reduction of an opening diameter of a very small trench having an opening diameter of 15 nm or less.This results in an effect whereby the embedding capacity can be increased and the yield of a gate-load component can be improved.
[0018] According to a further embodiment of the invention, an electrode structure having the features of claim 14 is provided. According to a further embodiment of the invention, an electrode structure having the features of claim 15 is provided. Short description of the drawings Fig. 1 is a schematic representation of a PCM sputtering device according to the invention. Fig. 2 is a layout diagram of a magnetic mechanism provided in the device according to the invention. Fig. Figure 3A is an illustrative diagram of a particle transport process in low-pressure sputtering. Fig. 3B is an explanatory view of the shape of a sputtered layer formed in a trench by low-pressure sputtering. Fig. 3C is an explanatory view of the shape of the sputtered layer formed in the trench by low-pressure sputtering. Fig. 3D is an explanatory representation of a particle transport process in high pressure sputtering. Fig. 3E is an explanatory diagram of the shape of a sputtered layer formed in a trench by high-pressure sputtering. Fig. 3F is an explanatory view of the shape of the sputtered layer formed in the trench by high pressure sputtering. Fig. Figure 4 is a schematic diagram showing the trench size dependence in a gate-last process when a conventional CVD process is used for the formation technique. Fig. 5 is a schematic diagram showing the trench size dependency in the gate-last process when a PCM sputtering method according to an embodiment of the invention is used for the formation technique. Fig. 6 is a diagram showing the configuration of a semiconductor manufacturing apparatus according to the embodiment of the invention. Fig. Figure 7A is a flowchart showing a conventional Al embedding step. Fig. 7B is a flowchart showing an Al embedding step according to the embodiment of the invention. Fig. Figure 8A is a diagram showing a conventional layered structure. Fig. 8B is a diagram showing a layered structure according to the embodiment of the invention. Fig. Figure 9 is a table showing Al embedding properties when respective substrates are used. Fig. Figure 10A is a graphical representation showing a relationship between the type of substrate and an embedding rate. Fig. Figure 10B is a graphical representation showing a relationship between the type of substrate and reflectivity. Fig. Figure 10C is a graphical representation showing a relationship between the type of substrate and the resistivity value. Fig. Figure 11 is a graph showing an Al diffusion amount when plasma nitriding is performed under respective bias conditions. Fig. Figure 12 is a graph showing nitrogen concentration on TiAl when plasma nitriding is performed under respective bias conditions. Fig. Figure 13 is a graph showing an investigation result of an effective work function of an n-type MOSFET. Fig. 14 is a graph showing an investigation result of a leakage current in the n-type MOSFET. Fig. 15A is a diagram showing one of the steps of a manufacturing method for an electronic component according to an example of the invention. Fig. 15B is a diagram showing one of the steps of the manufacturing method of the electronic component according to the example of the invention. Fig. 15C is a diagram showing one of the steps of the manufacturing method of the electronic component according to the example of the invention. Fig. 15D is a diagram showing one of the steps of the manufacturing method for the electronic component according to the example of the invention. Fig. 15E is a diagram showing one of the steps of the manufacturing method of the electronic component according to the example of the invention. Fig. 15F is a diagram showing one of the steps of the manufacturing method of the electronic component according to the example of the invention. Description of embodiments- embodiment -
[0019] As a result of intensive studies to solve the above problem, the inventors have completed the invention by finding that a second electrode formation layer (e.g., an Al wiring layer) can be easily embedded without unevenness directly on an ultra-thin barrier layer (e.g., a TiAlN layer) even if no Al seed layer is additionally formed between the ultra-thin barrier layer (e.g., the TiAlN layer) and the second electrode formation layer (e.g., the Al wiring layer) by forming a first electrode formation layer (e.g., a TiAlN layer) in a recess (e.g., a trench) formed in a workpiece (first step), forming the ultra-thin barrier layer (e.g., the TiAlN layer) by coating the surface of the first electrode formation layer (e.g.,The TiAl layer is plasma nitrided to form a nitriding layer (second step), and then the second electrode formation layer (e.g., the Al wiring layer) is formed on the ultra-thin barrier layer (e.g., the TiAlN layer) (third step). Specifically, the inventors have discovered a manufacturing method for an electronic component that suppresses the shrinkage of a trench opening by forming the ultra-thin TiAlN barrier layer and embedding Al on the TiAlN layer, and that includes a step of embedding Al in the trench using a barrier layer capable of suppressing Al diffusion.
[0020] With reference to the drawings, an embodiment of the invention will now be described in detail below.
[0021] Fig. 1 is a schematic diagram of a PCM sputtering apparatus 100 according to this embodiment. The PCM sputtering apparatus 100 includes a target electrode 402, a chamber 201, and a lower electrode 301 disposed facing the target electrode 402. A shield 403 is disposed below the target electrode 402 with a gap therebetween. Furthermore, a target electrode insulator 404 is attached to a side wall of the chamber 201 to support the target electrode 402. A magnet mechanism 405 is disposed above the target electrode 402, away from the target electrode 402. The magnet mechanism 405 includes a plurality of magnet pieces 406, a magnet support plate 407, and a magnetic field adjustment magnet body 408.
[0022] It should be noted that in this description, PCM (Point-Cusp Magnetic Field) is used to form a cusp magnetic field closed by four adjacent magnets using a design as described later. Fig. 2 stands.
[0023] The chamber 201 provided on the atmosphere side of the target electrode 402 includes an upper wall (ceiling wall) 202, a cylindrical side wall 203, and a bottom wall 204. The target electrode 402 is provided above the chamber 201 and is electrically insulated from the chamber 201 by the target electrode insulator 404. While main parts of the target electrode 402 are made of non-magnetic metals such as Al, SUS, and Cu, a target material required for film formation on a substrate 306 is provided on the lower pressure side of the target electrode 402. Fig. 1 omits the representation of the target material.
[0024] A high-frequency power source for the target electrode 102 supplies a high-frequency current to the target electrode 402 via a tuning box for the target electrode 101. Furthermore, a direct current source 103 is connected to the tuning box for the target electrode 101 to apply a direct voltage. Within the chamber 201, a plurality of gas injection ports 409 are provided for introducing a process gas such as argon into the chamber 201. The chamber 201 is evacuated by a vacuum evacuation pump 410 via a gas outlet port 205. Furthermore, the chamber 201 is also provided with a diaphragm pressure transducer (DG) 430 for measuring the pressure within the chamber 201, and the gas outlet port 205 is provided with a variable orifice 431 for adjusting an outlet amount according to the pressure.
[0025] The lower electrode 301 includes a stage holder 302 and a lower electrode insulator 303. The stage holder 302 is fixed to the bottom wall 204 via the lower electrode insulator 303. The stage holder 302 and the chamber 201 are electrically insulated from each other by the lower electrode insulator 303. The substrate 306 is placed on a top surface of the stage holder 302. Furthermore, a temperature control mechanism 412 may be provided on the lower electrode 301 or the stage holder 302 to control the substrate 306 to a temperature required for film formation. A lower electrode high-frequency power source 305 supplies high-frequency power to the lower electrode 301 via a lower electrode tuning box 304.
[0026] Fig. Fig. 2 is a component diagram of the magnet mechanism 405 provided in the device according to this embodiment. Referring to Fig. 2, the shape of the magnetic mechanism 405 is described in detail. Fig. 2 is a plan view of the magnetic mechanism 405 as seen from the side of the target electrode 402.
[0027] As in Fig. As shown in Figure 2, the plurality of magnet pieces 406 and the annular magnetic field adjusting magnet body 408 are supported and provided in the disk-shaped magnet carrier 407. The magnetic field adjusting magnet body 408 is arranged to cover a part of an outer periphery of an area on the side of the target electrode 402 where the plurality of magnet pieces 406 are arranged. Reference numeral 403a in Fig. 2 represents an inner diameter of the shield 403, and the number of small circles each represents a shape of the magnet pieces 406. Furthermore, the magnet pieces 406 have the same shape and the same magnetic flux density. Furthermore, the letters N and S represent magnetic poles of the magnet pieces 406 as viewed from the target electrode 402 side.
[0028] The magnet pieces 406 are arranged in a grid (in the X-axis direction and Y-axis direction) at approximately equal pitches (5 to 100 mm). Furthermore, the adjacent magnet pieces 406 have poles opposite to each other when viewed from the target electrode 402 side. Meanwhile, in a square formed by any four magnet pieces 406 arranged along the X-axis direction and Y-axis direction, the magnet pieces 406 adjacent to each other along the diagonal direction have the same pole when viewed from the target electrode 402 side. With such an arrangement, a point-cusp magnetic field (PCM) 411 is formed using any four adjacent magnet pieces 406.
[0029] The height of the magnet pieces 406 is typically more than 2 mm, and their cross-section is square or circular. The diameter, height, and material of the magnet pieces 406 can be adjusted appropriately by a process application. When high-frequency current is applied to the target electrode 402, plasma is generated through a capacitive coupling mechanism. The plasma is subject to the action of the closed point-cusp magnetic field 411.
[0030] The magnetic field adjustment magnet body 408 extends to partially overlap the outer periphery of the area on the target electrode 402 side where the magnet pieces 406 are arranged. This can suppress (control) the magnetic field strength in a gap between the target electrode 402 and the shield 403. The magnetic field adjustment magnet body 408 can be made of a material capable of controlling the magnetic field strength in the gap between the target electrode 402 and the shield 403. A material with high magnetic permeability, such as SUS 430, is preferable. Magnetic field adjustment can be achieved by adjusting the area where the magnet pieces 406 and the magnetic field adjustment magnet body 408 overlap with each other in the magnet mechanism 405.Specifically, adjusting the area where the magnet pieces 406 and the magnetic field adjusting magnet body 408 overlap with each other allows a magnetic field required for sputtering the target electrode 402 to be applied to the outermost periphery of the target electrode 402. Thus, the magnetic field strength in the gap between the target electrode 402 and the shield 403 can be adjusted.
[0031] Fig. Figure 3A is an illustrative diagram of a particle transport process in low-pressure sputtering. As in Fig. 3A, in low pressure sputtering, sputter particles do not collide and scatter before reaching the substrate 306 from the target electrode 402. Fig. 3B is an explanatory diagram of the shape of a sputtered layer 452 formed in a substrate end portion by low-pressure sputtering in a trench 453. Meanwhile, Fig. 3C is an explanatory view of the shape of the sputter layer 452 formed in a substrate center portion by low-pressure sputtering in the trench 453. As can be seen from the Fig. 3B and Fig. As can be seen in Figure 3C, during low-pressure sputtering, the shape of the sputter layer 452 is distorted in the substrate end portion compared to the substrate center portion. Note that reference numeral 454 denotes a base substrate.
[0032] On the other hand, Fig. 3D an explanatory diagram of a particle transport process in high pressure sputtering using the Fig. 1 shown PCM sputtering device 100 of this embodiment. As in Fig. As shown in Figure 3D, during high-pressure sputtering, sputter particles 450 are distributed in a container by colliding and scattering before reaching the substrate 306 from the target electrode 402, resulting in a sheath acceleration 451 above the substrate 306. Thus, the sputter particles enter evenly. Fig. 3E is an explanatory diagram of the shape of a sputtering layer 452 formed in the substrate end portion by high-pressure sputtering in a trench 453. Meanwhile, Fig. 3F is an explanatory view of the shape of the sputter layer 452 formed in the substrate center portion by high-pressure sputtering in the trench 453. As can be seen from the Fig. 3E and Fig. 3F, the high-pressure sputtering according to this embodiment allows a sputter layer 452 having a symmetrical coverage shape to be deposited in the trenches 453 over the entire substrate, and it can also suppress the film formation on the sidewall.
[0033] Fig. 4 is an explanatory diagram of a gate-load formation technique for layering various materials using a conventional CVD method in very small trench openings having opening diameters of 32 nm and 15 nm, respectively. A very small trench structure 601 has a pre-formed insulating base layer 602. A high-dielectric insulating layer 603 is formed on the insulating base layer 602. On the high-dielectric insulating layer 603, a metal nitride layer 604, a metal nitride layer 605, a metal nitride layer 606, and a metal layer 607 are formed for controlling an operating voltage, as well as a barrier layer 608 and an Al seed layer 609 for embedding. When such various materials are formed by the CVD method, the layers can be uniformly formed not only at the bottom of the trench but also on the sidewall.On the other hand, however, it is understood that increasing the thickness of the layers to be formed narrows the trench opening. Therefore, the opening of the very small trench, which has a 15 nm opening diameter, will be closed unless the thickness of each layer is reduced. Therefore, even if the thickness of a barrier base layer needs to be increased to enhance the barrier properties, sufficient layer thickness cannot be ensured.
[0034] Meanwhile, Fig. 5 is an explanatory view of a gate-load formation technique for stacking different materials using the Fig. 1 according to this embodiment. The insulating base layer 602 has been previously formed at the bottom of the very small trench structure 601. The high dielectric insulating layer (gate insulating layer) 603 is formed on the insulating base layer 602. A metal nitride layer 701, a metal nitride layer 702, and a metal nitride layer 703 are formed thereon as a third electrode formation layer for controlling an operating voltage. Furthermore, a metal layer (first electrode formation layer) 704 and a barrier layer (nitride layer) 705 are formed thereon for embedding. Since the device of this embodiment can suppress the formation of a sputter layer on the sidewall, compared to the device shown in Fig. 4, various materials can be layered while suppressing a reduction in the trench opening. This allows even film formation in a very small 15 nm pattern with the same film thickness as the 32 nm trench. Accordingly, even if the trench size is reduced, the various materials can be layered without changing their optimized film thicknesses. Moreover, the barrier layer formed using the method of this embodiment is achieved only by plasma nitriding a single-layer titanium-aluminum film. Thus, the number of layers in the stacking structure can be reduced.
[0035] The above point is further discussed with reference to the Fig. 1 to 3. In this embodiment, an ultra-thin barrier layer is formed by using any four adjacent magnet pieces 406 in Fig. 2 the point-cusp magnetic field 411 is formed and in the chamber 201 a high plasma density (e.g. 1 × 10 11 / volume unit). In the case of the Fig. In the PCM sputtering apparatus 100 shown in FIG. 1, the plasma density for implementing the method of this embodiment is increased by increasing the pressure in the chamber 201, the power of the power source for the target electrode 102, or the frequency of the power source for the target electrode 102, and decreased by decreasing the pressure in the chamber 201, the power of the target electrode 402 (cathode), or the RF frequency of the target electrode 402 (cathode). Therefore, compared with a normal sputtering method, the plasma formed in the chamber 201 has a high ionization rate by increasing the frequency of the power source for the target electrode 102 (to, for example, 40 MHz to 60 MHz), by increasing the power of the power source for the target electrode 102 (to, for example, 3000 W), and by increasing the pressure in the chamber 201 (to, for example, 10 Pa). Thus, a TiAl layer can be formed uniformly on the bottom and side wall of a step of a trench.In addition, by using the in . Fig. In the PCM sputtering apparatus 100 shown in Figure 1, the surface of the TiAl layer formed in the trench formed in the workpiece is plasma nitrided, forming a nitride layer. Since 60 MHz is used, the plasma formed in the chamber 201 has a high ionization rate even under high pressure. Since nitrogen ion energy is suppressed, an ultra-thin TiAlN barrier layer can thus be formed. Furthermore, a second electrode formation layer (e.g., an Al wiring layer) can be directly embedded on the ultra-thin TiAlN barrier layer without any unevenness.
[0036] Fig. 6 shows a semiconductor manufacturing apparatus 500 used in an electronic component manufacturing method including a step of embedding a metal layer according to this embodiment. The semiconductor manufacturing apparatus 500 includes: a chamber 501 for forming a titanium-aluminum layer as a first step; a chamber 502 for plasma nitriding the titanium-aluminum layer as a second step; a chamber 503 for embedding Al as a third step; and chambers 504 and 505 for forming metal layers attached to enable layer formation of various other metal materials.The apparatus further includes a transfer chamber 506 equipped with a vacuum transfer unit capable of transferring the substrate to the chambers 501 to 505 without exposing the substrate to the atmosphere; and a loadlock chamber 507 for transferring the substrate from the atmosphere to the vacuum. Note that the PCM sputtering apparatus 100 of the embodiment shown in FIG. Fig. 1 can be used as the chambers 501 to 505. The use of the semiconductor manufacturing apparatus 500 of this embodiment enables continuous processing without exposing the substrate to the atmosphere. Thus, impurities such as moisture, carbon, and oxygen can be prevented from adsorbing on an interface. Accordingly, the substrate can be transferred to the next step without changing the properties of the layers formed in each apparatus. Note that the semiconductor manufacturing apparatus 500 has a controller (not shown) including an arithmetic processor such as a CPU, and performs predetermined processing on the substrate to be processed by outputting command signals to the respective processing devices 501 to 507 according to a predetermined program.It should be noted that each of the chambers 501 to 507 has a control device (not shown) such as a PLC (Programmable Logic Controller) for controlling a device such as a flow controller and a suction pump according to the command signals output from the controller.
[0037] Fig. 7A is a flowchart of an Al embedding step according to the conventional technique (Patent Document 2 described above). Fig. 7B is a flowchart of an Al embedding step according to this embodiment. In the conventional Al embedding step, a layered barrier layer formation step 810 is performed to suppress Al diffusion. Then, a barrier layer formation step 811 is performed to prevent Al diffusion on the layered barrier layer. After that, an Al seed layer formation step 812 is performed to facilitate Al migration. Then, a high-temperature Al embedding step 813 is performed.
[0038] However, the Al embedding step using the apparatus of this embodiment can achieve full embedding performance even after performing a titanium-aluminum film formation step 815 as the first step, a titanium aluminum nitride film formation step 816 as the second step, and then directly performing an Al embedding formation step 817 without using seed aluminum. To deposit a single-layer TiAl film in the first step 815, the film formation is performed by using a metal alloy target of TiAl as the target, setting the substrate temperature to 30°C, the RF target power of TiAl to 1500 W, and the DC voltage to 430 V, and also by using Ar as the inactive gas, setting an Ar injection rate of 100 sccm, and adjusting the pressure in the chamber to 10 Pa with an automatic regulator.In order to control the shape of the layers to be formed, the layer formation is further carried out by setting the RF power for the substrate electrode to 50 W.
[0039] Next, in the second step 816, plasma nitriding is performed to form the titanium aluminum nitride layer by setting the substrate temperature to 30°C, the RF target power of Ti to 3000 W, the DC voltage to 0 V, the injection rate of nitrogen as the reactive gas to 100 sccm, and the pressure in the chamber to 10 Pa with the automatic regulator. By setting the DC voltage to 0 V, a high-density, low-energy nitrogen plasma is generated. Thus, only a surface layer of the titanium aluminum layer is plasma nitrided to form a TiAlN barrier layer. At this time, the frequency of the high-frequency power source for forming a high-density plasma using a point-cusp magnetic field under the pressure described above is preferably between 10 MHz and 100 MHz, more preferably between 40 MHz and 60 MHz.
[0040] Next, for Al deposition, in the third step 817, film formation is performed by setting the substrate temperature to 30 to 450°C, the RF target power of Al to 3000 W, and the DC voltage to 215 V. Furthermore, by using Ar as the inactive gas, setting an Ar injection rate of 100 sccm, and adjusting the pressure in the chamber to 10 Pa with the automatic regulator. To increase the film thickness at the bottom of the trench, film formation is also performed by setting the RF power for the substrate electrode to 200 W. At this time, the frequency of the high-frequency power source for forming a high-density plasma using the point-cusp magnetic field under the pressure described above is preferably between 10 MHz and 100 MHz, more preferably between 40 MHz and 60 MHz.
[0041] Fig. 8A is an explanatory diagram showing a conventional structure (Patent Document 2). Fig. 8B is an explanatory diagram showing a structure of this embodiment. While the conventional structure includes a Ti or TiN base layer 6 on a TiAlN layer 5, the structure of this embodiment includes an Al layer 7 on a TiAlN layer 5 because the TiAlN layer 5 is formed by plasma nitriding on a surface of a TiAl layer 4 without using a Ti or TiN base layer.
[0042] It should be noted that the above-described structure of Patent Document 3 is achieved by depositing a first titanium nitride layer 3 (first barrier layer) on a gate insulating layer 2, and then depositing a second TiN base layer (second barrier layer) on the first titanium nitride layer 3. Thus, the barrier layers have an increased thickness. On the other hand, the structure of Patent Document 4 is achieved by stacking a titanium nitride layer 3 and an Al layer 7 on a gate insulating layer 2. However, the Al layer 7 must be deposited under a temperature condition that causes the Al layer 7 to flow. Therefore, in the case of a very small trench with an opening diameter of 15 nm or less, unevenness on the surface of the Al layer will affect the performance of the electronic components.
[0043] Fig. Figure 9 shows results obtained by investigating the Al embedding properties after using the methods described in the Fig. 1 and Fig. 6, various substrates were formed according to this embodiment. Each of the substrates shown in Fig. 9 is provided with a ratio expressed by X / Y. The denominator in (Y) represents the number of trenches examined, and the numerator (X) represents the number of trenches fully embedded with Al in the third step. For example, 7 / 7 in the photographs in case (c) of the TiAlN underlayer (RT) of this embodiment represents full embedding with seven trenches fully embedded with aluminum out of seven examined trenches. On the other hand, 1 / 7 in the photograph in case (b) of the layered TiAlN / Ti underlayer (RT) represents a low embedding rate with one trench fully embedded with aluminum out of seven examined trenches. The Al embedding characteristics were evaluated by SEM (secondary electron microscopy) analysis.
[0044] (a) in Fig. Figure 9 is a result (referred to as TiAl underlayer) obtained by Al embedding as a third step after the 10 nm thick TiAl layer had been deposited in the first step. (b) in Fig. Figure 9 is a result (referred to as a layered TiAlN / Ti base) obtained by forming the titanium aluminum nitride layer as a second step after depositing the 10 nm thick TiAl layer in the first step, then forming a Ti layer as well, and then embedding the Al layer as a third step. (c) in Fig. Figure 9 is a result (referred to as TiAlN substrate according to this embodiment) obtained by performing the titanium aluminum nitride layer formation step as the second step after depositing the 10 nm thick TiAl layer in the first step, and then performing Al embedding as the third step. (d) in Fig. Figure 9 is a result (referred to as a layered TiAlN / Ti base) obtained by performing the titanium aluminum nitride film formation step as the second step after depositing a 10 nm thick TiAl film in the first step, then further forming a TiN film, and then performing Al embedding as the third step. Furthermore, (a) to (d) are evaluated by setting the Al embedding temperature as the third step to RT (room temperature), 250°C, and 450°C (note that the result for 250°C is omitted in (d).
[0045] In (a) in Fig. 9, although the generation rate of gaps (hereinafter referred to simply as gaps) improves when the Al embedding temperature is increased from room temperature to 250°C and 450°C, many gaps are observed. In (b) in Fig. 9, the gap generation rate improves when the Al embedding temperature is increased from room temperature to 250°C, and no gaps are observed when the temperature is further increased to 450°C. On the other hand, in (c), Fig. 9, according to this embodiment, the void generation rate is low even at room temperature, and no voids are observed when the Al embedding temperature is increased to 250°C and 450°C. It is believed that this is because, in the TiAl base or the layered TiAlN / Ti base, Al embedding causes Ti and Al to react with each other, facilitating alloy formation and suppressing Al migration. (d) in Fig. Figure 9 shows that although Al embedding can be achieved even at room temperature, the flatness is inferior compared to the case using the TiAlN underlayer according to this embodiment. Therefore, the results show that the use of the single-layer TiAlN barrier layer of this embodiment suppresses alloying and facilitates Al migration, thereby achieving better surface flatness.
[0046] In addition, surface aggregation of Al is observed in each of (a) to (c). It is known that large aggregation deteriorates Al reflectivity and increases resistivity. Furthermore, when planarizing by CMP in the next step, uneven Al thickness due to aggregation causes a problem in element processing. When the TiAlN layer of this embodiment is used, and especially when the Al embedding temperature is 250°C, complete Al embedding can occur, and Al aggregation can be suppressed compared to the other base layers. This demonstrates that using the TiAlN layer as the base layer facilitates Al migration, thereby achieving a sufficient embedding rate without increasing the Al embedding temperature.
[0047] Fig. Figure 10A is a graph showing embedding rates when different substrates are used. Fig. Figure 10B is a graph showing the reflectivity of the Al layer when the different substrates are used. Fig. Figure 10C is a graph showing the specific resistance values of the Al layer when using different substrates. It should be noted that the Fig. 10A to 10C show the results obtained when the Al embedding temperature was set to RT (room temperature, ie, about 20° to 30°C), 250°C, and 450°C, respectively. The vertical axis in Fig. 10A represents the embedding rate, the vertical axis in Fig. 10B represents the reflectivity, and the vertical axis in Fig. 10C represents the specific resistance value. The horizontal axes in the Fig. 10A to 10C represent the types of substrate. More specifically, A represents the type of substrate in each of the horizontal axes in the Fig. 10A to 10C the case of (in Fig. 9 (d)) layered TiAlN / TiN substrate, B represents the case of the (in Fig. 9 (b)) layered TiAlN / Ti substrate, and C represents the case of the (in Fig. 9 (a)). In addition, D, E and F all represent the (in Fig. 9 (c) according to this embodiment, which is obtained by nitriding the TiAl layer. However, D represents the case where only the surface of the TiAl layer is nitride, E represents the case where approximately half of the TiAl layer in its depth direction is nitride, and F represents the case where most of the TiAl layer in its depth direction is nitride.
[0048] As from Fig. As can be seen from Figure 10A, the embedding rate improves when Al embedding is performed at room temperature when the TiAlN base layer according to this embodiment is used. Furthermore, even at room temperature, the same embedding rate is achieved as when the layered TiAlN / TiN base layer is used, and complete embedding can be achieved. Fig. As can be seen from Figure 10B, even when using the TiAlN base layer according to this embodiment, the reflectivity deteriorates as the Al embedding temperature increases. However, the reflectivity can be improved compared to the case where the layered TiAlN / TiN base layer is used. To form an Al layer with high reflectivity, it is preferable that the TiAl surface is plasma nitrided to form a TiAlN base layer, and that the layer formation is carried out by setting the Al embedding temperature to 250°C or less. As can be seen from Fig. 10C, the specific resistance value deteriorates with increasing Al embedding temperature when the layered TiAlN / TiN base or the TiAl base is used. However, Fig. 10C also shows that the resistivity is constant and does not depend on the Al embedding temperature when the TiAlN underlayer according to this embodiment is used. Therefore, the TiAlN layer according to this embodiment must be used to embed the Al layer with a good embedding rate, high reflectivity, and low resistivity.
[0049] Fig. Figure 11 is a graph showing a result obtained by using SIMS (Secondary Ion Mass Spectroscopy) to examine the amount of Al that diffused into the TiAl layer, the second step of forming a titanium aluminum nitride layer after depositing the TiAl layer 10 nm thick in the first step, and then the third step of Al embedding at 450°C. Fig. 11, the vertical axis represents the secondary ion intensity, and the horizontal axis represents a depth from the Al surface. In Fig. 11, ▼ represents a case without plasma nitriding, ■ represents a case that plasma nitriding is performed without applying a bias power (0 W) to the lower electrode 301, ▲ represents a case that plasma nitriding is performed by applying a bias power of 50 W to the lower electrode 301, and • represents a case that plasma nitriding is performed by applying a bias power of 200 W to the lower electrode 301. Fig. 11 shows that Al in the TiAl layer not subjected to plasma nitriding diffuses into the TiAl layer, while Al diffusion into the TiAl layer can be suppressed when the TiAl layer is subjected to plasma nitriding of this embodiment. Fig. 11 also shows that regardless of whether the bias power is applied to the lower electrode 301 or not, the Al diffusion into the TiAl layer can be suppressed when the TiAl layer is subjected to the plasma nitriding of this embodiment.
[0050] Fig. Next, Figure 12 is a graph showing a result obtained by measuring a nitrogen concentration in the TiAlN layer by XPS (X-ray photoelectron spectroscopy) in the TiAlN layer formation step as a second step. Fig. 12, the vertical axis represents N2 concentration, and the horizontal axis represents depth. Depth is obtained by converting it into a rate when a silicon oxide layer is etched with Ar. The measurement direction is from the side of the TiAlN layer (surface side). The interface between the surface oxide layer and the TiAlN layer is defined as approximately 1 nm from the surface of the surface oxide layer. Fig. 12 ■ represents a nitrogen concentration distribution without nitriding. ◆ represents a nitrogen concentration distribution when nitriding is carried out with a bias power of 50 W, ▲ represents a nitrogen concentration distribution when nitriding is carried out with a bias power of 100 W, • represents a nitrogen concentration distribution when nitriding is carried out with a bias power of 200 W, □ represents a nitrogen concentration distribution when nitriding is carried out with a bias power of 400 W, and Δ represents a nitrogen concentration distribution when nitriding is carried out with a bias power of 800 W. Fig. Figure 12 shows that the nitrogen concentration in the TiAlN layer increases with increasing power applied to the substrate, and that the TiAl layer is nitrided more deeply. Based on the fact that the Ti content increases more than the Al content with increasing bias power, it can also be speculated that the TiAlN layer is formed with a higher TiN layer concentration than the AlN layer concentration, which allows for the suppression of Al diffusion.
[0051] Fig. Next, Fig. 13 is a graph showing effective work function values of an element formed by using the TiAlN layer formed in this embodiment and forming an Al layer on the TiAlN layer. Fig. 13, the vertical axis represents an effective work function (eWF), and the horizontal axis represents a bias power during nitriding of the TiAl layer. It should be noted that Fig. 13 shows a result obtained when the TiAl layer is not nitride at the point where the bias power is 0 W. In Fig. 13, ◆ represents an effective work function value obtained without heat treatment after element formation, ■ represents an effective work function value obtained when heat treatment is performed at 400°C for 15 minutes, and A represents an effective work function value obtained when heat treatment is performed at 450°C for 15 minutes. The result shows that when the TiAlN layer is used as the underlayer, even after heat treatment at 450°C for 15 minutes, the effective work function is 4.2 eV or less, which is suitable for an n-type MOSFET.
[0052] Fig. Fig. 14 is a graph showing the leakage current characteristics of the heat-treated member after the Al layer was formed on the TiAlN layer formed using this embodiment. Fig. 14, the vertical axis represents a leakage current (Jg), and the horizontal axis represents a temperature during post-deposition heat treatment (PDA). In Fig. 14, ◆ represents a leakage current value obtained without nitriding (ie, when the Al layer is formed on the TiAl layer instead of the TiAlN layer), ■ represents a leakage current value obtained when nitriding is performed with a power of 50 W applied to the substrate, and A represents a leakage current value obtained when nitriding is performed with a power of 200 W applied to the substrate. Fig. Figure 14 shows that the leakage current in the element using the TiAlN nitride layer improves compared to the element without nitride when subjected to high-temperature heat treatment at 450°C for 15 minutes. This leads to the speculation that the use of the TiAlN layer as shown in Fig. 11, can suppress Al diffusion and thus improve leakage current. The above results show that the use of a metal alloy nitride layer formed using this embodiment can achieve an effective work function suitable for the n-type MOSFET and can improve leakage current.
[0053] This embodiment includes the first step of forming the first electrode formation layer (e.g., TiAl), the second step of forming a nitride layer by plasma nitriding the first electrode formation layer, and the third step of embedding the second electrode formation layer (e.g., Al) on the nitride layer. Thus, low-resistance Al can be directly embedded without unevenness, and even on the workpiece in which very small trenches with an opening diameter of 15 nm or less are formed, the reduction of the opening diameter is suppressed.
[0054] Since sputtering is used to form the first electrode formation layer in this embodiment, the effect is achieved that an alloy composition of the first electrode formation layer to be formed at the bottom of the recess can be easily controlled by changing it toward a desired target composition. When the first electrode formation layer is formed by sputtering using, for example, a TiAl alloy target, excellent controllability of the TiAl alloy composition is achieved. In the case of using a Ti 50 Al 50 -Target, for example, a first electrode formation layer is formed at the bottom of the recess, which contains a Ti 50 Al 50 -composition. Alternatively, in the case of using a Ti 70 Al 30-Targets at the bottom of the recess, a first electrode formation layer is formed, which comprises a Ti 70 Al 30 composition than the target composition. Therefore, the effect is achieved that the TiAl alloy composition of the first electrode formation layer can be controlled as desired by changing the ratio of the TiAl target composition.
[0055] Since plasma treatment is used to nitride the first electrode formation layer in this embodiment, an effect is achieved that the surface of the first electrode formation layer formed in the workpiece at the bottom and sidewall of the recess can be nitrided at a low temperature. In the generally used CVD method, nitriding is performed by using gas (NH3 or the like) to nitride the metal alloy. However, this method requires raising the substrate temperature to a gas decomposition reaction temperature. Thus, raising the element temperature leads to interdiffusion between the insulating layer and the first electrode formation layer. As a result, desired device characteristics cannot be achieved.
[0056] In this embodiment, the use of the magnetic mechanism capable of forming a point-cusp magnetic field achieves in-plane uniformity at the strength of the magnetic field formed on the target surface, and ion particles in the plasma are allowed to perpendicularly penetrate into the bottom of the recess in the workpiece. As a result, the effect of uniformly plasma nitriding the surface of the first electrode formation layer is achieved. Furthermore, since a high-density plasma can be generated by the point-cusp magnetic field, a nitride film layer having a desired plasma nitrogen concentration can be formed in a short time. In addition, film formation can be performed at the bottom of the recess in the workpiece while suppressing adhesion of sputter particles to the sidewall of the recess in the workpiece.Since the point-cusp magnetic field is also used for embedding the second electrode formation layer (Al), the effect is achieved that the layer formation at the bottom of the recess in the workpiece can be carried out efficiently and that an embedding defect rate, ie, the yield, can be improved.
[0057] In this embodiment, the bias power is applied to the substrate holder during nitriding of the first electrode formation layer. Thus, ions can be attracted to the bottom of the recess in the workpiece, and nitriding can be carried out evenly. Furthermore, by increasing or decreasing the bias power, the surface nitride layer thickness of the first electrode formation layer can be controlled. This achieves the effect of suppressing the diffusion of the second electrode formation layer into the lower electrode layer in the gate-load structure.
[0058] In this embodiment, the first electrode formation layer is nitride in a high-pressure atmosphere (e.g., 1 to 200 Pa). Thus, ionization can be facilitated and plasma particle energy can be suppressed. Accordingly, only the surface of the first electrode formation layer can be plasma nitrided. As a result, the effect of further suppressing the diffusion of the second electrode formation layer into the lower electrode layer is achieved.
[0059] In this embodiment, the second electrode formation layer is formed at a low temperature (room temperature to 250°C). Thus, the effect of the second electrode formation layer being excellent in surface flatness, low in resistivity, and high in reflectivity is achieved.
[0060] In this embodiment, the respective steps are performed without exposure to the atmosphere. This prevents impurities from adsorbing at the interfaces between the respective layers. As a result, the effect of forming a layered structure in the gate-load device with desired electrical properties is achieved. - First example -
[0061] A first example of the invention will be described in detail with reference to the drawings. Fig. 15D and Fig. 15E are representations showing a first step in which, using the Fig. 1 and Fig. 6, a TiAl layer is formed in a trench structure, a second step in which a TiAlN layer is formed on a TiAl surface layer by plasma nitriding, and then a third step of Al embedding. First, as shown in Fig. 15D, a TiAl layer 905 is deposited in a trench structure 901 and 902 by sputtering. A metal alloy target made of TiAl is used as the target, and Ar is used as the sputtering gas. Next, the TiAl layer 905 is plasma nitrided using a Ti metal target and a nitrogen gas as the gas to form a nitrogen plasma, thereby converting the TiAl layer 905 into a TiAlN layer. It should be noted that, although the entire TiAl layer 905 is converted into the TiAlN layer by plasma nitriding in this example, only a part (e.g., a surface) of the TiAl layer 905 may be converted. Next, as shown in Fig. As shown in Figure 15E, an Al layer 906 is embedded on the TiAlN layer by sputtering. An Al metal target is used as the target, and Ar is used as the sputtering gas.
[0062] The substrate temperature can be adjusted within a range of 25°C to 500°C. The target power can be adjusted within a range of 100 W to 5000 W. The sputtering gas pressure can be adjusted within a range of 1 Pa to 200 Pa, the Ar gas flow rate can be adjusted within a range of 10 sccm to 500 sccm, and the nitrogen gas flow rate can be adjusted within a range of 1 sccm to 100 sccm.
[0063] The specific deposition conditions in this example are described below. For the deposition of the TiAl layer 905 in the first step, a metal alloy target made of TiAl is used, the substrate temperature is set to 30°C, the RF power for the Ti target is set to 1500 W, the DC voltage is set to 430 V, Ar is used as the inactive gas, the Ar injection rate is set to 100 sccm, and the pressure in the chamber is adjusted to 10 Pa by an automatic controller. Furthermore, to control the shape of the layers to be formed, the deposition is carried out by setting the RF power for the substrate electrode to 50 W. Through the above formation step, the TiAl layer 905 is formed to have a layer thickness of 5 nm to 10 nm.Next, in the second step for forming the TiAl layer by plasma nitriding the TiAl layer 905, plasma nitriding is performed by setting the substrate temperature to 30°C, the RF power for the Ti target to 3000 W, the DC voltage to 0 V, the feed rate of nitrogen as the reactive gas to 100 sccm, and the pressure in the chamber to 10 Pa by the automatic controller. Then, in the third step for deposition of the Al layer 906, the film formation is performed by setting the substrate temperature to 400°C, the RF power for the Al target to 3000 W, and the DC voltage to 215 V, and also by using Ar as the inactive gas, and setting the Ar feed rate to 100 sccm and the pressure in the chamber to 10 Pa by the automatic controller. To increase the layer thickness at the bottom of the trench, the layer formation is also carried out by setting the RF power for the substrate electrode to 200 W. - Second example -
[0064] The second example is one used in the gate-last method.
[0065] The second example of the invention is described below with reference to the drawings. Fig. 15A to 15F are diagrams showing steps of a method for manufacturing a semiconductor device as a second example of the invention. In this example, metal gate electrodes that realize appropriate effective work functions are formed in a region for forming an n-type MOSFET as a first region and a region for forming a p-type MOSFET as a second region by performing the first step of depositing a TiAl layer, the second step of forming a TiAl barrier layer by plasma nitriding, and the third step of Al embedding, as in the first example.
[0066] As in Fig. 15A, a trench structure 901 and 902 is formed in the region for forming the n-type MOSFET and the region for forming the p-type MOSFET, respectively. As shown in Fig. 15B, a metal nitride layer 903 and a metal nitride layer 904 are formed using the sputtering apparatus according to the invention so that they cover the inner sides of the respective trench structures. As shown in Fig. Then, as shown in Figure 15C, the metal nitride layer 903 and the metal nitride layer 904 forming the bottom of the trench structure 901 in the n-type MOSFET formation region are removed by lithography and etching. In this example, the metal nitride layer 903 is wet-etched using a mixed solution of sulfuric acid, hydrogen peroxide solution, and water, and the metal nitride layer 904 is removed by Ar plasma etching.
[0067] As in Fig. 15D, the substrate is then passed through the Fig. 6 is transferred to a chamber 501 that carries out the sputtering method according to the invention to deposit a TiAl layer in the first step, which is one of the steps of the invention, thereby forming a metal alloy layer (TiAl layer) 905 to cover the trench structure. As shown in Fig. 15E, the substrate is then transferred to a chamber 502 to perform the second step of plasma nitriding the TiAl layer, and the metal alloy layer (TiAl layer) 905 is converted into a metal alloy nitride layer (TiAlN layer) by plasma nitriding. The substrate is further transferred to a chamber 503 to perform the third step of Al embedding, and a metal layer (Al layer) 906 is formed by Al embedding. Planarization is then performed using a CMP technique to form the Fig. 15F shown structure.
[0068] It is found that when the substrate temperature is set to 300°C to 400°C in the step of forming the Al metal layer, the metal alloy nitride layer 905 formed from the metal alloy layer 905 diffuses into the metal nitride layer 900 in the trench structure 901, which is the region for forming the n-type MOSFET, and thus an effective work function suitable for the n-type MOSFET can be achieved. Meanwhile, it is found that in the trench structure 902, which is the region for forming the p-type MOSFET, an effective work function suitable for the p-type MOSFET can be maintained because the metal nitride layer 903 and the metal nitride layer derived from the metal alloy layer 905 suppress Al diffusion, and thus an effective work function required for a gate-last process can be achieved.Furthermore, it is expected that wiring error is improved compared to the conventional technique. This demonstrates that the TiAlN barrier layer formed using the device of the present invention has good barrier properties against Al diffusion. As a result of measuring the effective work function, EOT, and leakage current characteristics of the fabricated device, it is confirmed that using the Al embedding method of the present invention can achieve effective work functions (4.4 eV or less in the n-type MOSFET and 4.6 eV or more in the p-type MOSFET) suitable for the respective MOSFETs without increasing the EOT.
Claims
[1] Manufacturing process for electronic component, comprising: a first step (815) of forming a first electrode formation layer (4) in a recess formed in a workpiece, the recess having an opening diameter of 15 nm or less; a second step (816) in which a nitriding layer (5) is formed by plasma nitriding a surface of the first electrode formation layer (4); and a third step (817) in which a second electrode formation layer (7) is formed on the nitriding layer (5), wherein the first electrode formation layer (4) is formed in the first step (815) by sputtering while a magnetic field is formed on a target surface by a magnet unit having a plurality of magnets arranged at lattice points of a polygonal lattice, which magnets are arranged such that adjacent magnets have opposite poles, the nitriding layer (5) is formed in the second step (816) by nitriding the surface of the first electrode formation layer (4) while a magnetic field is formed on the target surface by the magnet unit, and the second electrode formation layer (7) is formed by sputtering in the third step (817) while a magnetic field is formed on the target surface by the magnet unit. [2] The electronic component manufacturing method according to claim 1, wherein the second step (816) is performed in an atmosphere of 1 Pa to 200 Pa. [3] A manufacturing method for an electronic component according to claim 1, further comprising: a fourth step in which a gate insulating layer (2) is formed between the workpiece and the first electrode forming layer (4) before the first step (815). [4] A manufacturing method for an electronic component according to claim 3, further comprising: a fifth step in which, after the fourth step and before the first step (815), a third electrode formation layer (3) is formed between the first electrode formation layer (4) and the gate insulating layer (2). [5] The electronic component manufacturing method according to claim 1, wherein the first electrode formation layer (4) contains Ti and Al, and the second electrode formation layer (7) contains Al. [6] The electronic component manufacturing method according to claim 1, wherein the first step to the third step (815, 816, 817) are carried out after the workpiece is transferred into a vacuum. [7] The electronic component manufacturing method according to claim 1, wherein in the first step (815), deposition is carried out on the bottom of the recess by forming plasma with high density and high ionization rate. [8] The electronic component manufacturing method according to claim 1, wherein in the second step (816), the nitriding layer (5) whose thickness is ultra-thin is formed by forming plasma with suppressed nitrogen ion energy. [9] The electronic component manufacturing method according to claim 8, wherein the plasma with the suppressed nitrogen ion energy is formed in the second step (816) by setting a DC voltage applied to the target to 0 V. [10] The electronic component manufacturing method according to claim 1, wherein the second electrode formation layer (7) having a flat surface is formed in the third step (817) by setting the workpiece to a low temperature. [11] The electronic component manufacturing method according to claim 10, wherein the low temperature is in a range from room temperature to 250°C. [12] The manufacturing method of electronic component according to claim 4, wherein the third electrode formation layer (7) is a layer for controlling an operating voltage of the electronic component. [13] The electronic component manufacturing method according to claim 1, wherein the nitriding layer (5) is a barrier layer for suppressing diffusion from the second electrode formation layer (7). [14] An electrode structure comprising a gate insulating layer (2), a first electrode forming layer (4), a nitriding layer (5) and a second electrode forming layer (7) which are stacked on a workpiece having a recess formed therein in the specified order, wherein the first electrode formation layer (4), the nitriding layer (5) and the second electrode formation layer (7) are layers formed by the manufacturing method for an electronic component according to claim 1, and the recess has an opening diameter of 15 nm or less. [15] An electrode assembly manufactured by a manufacturing method for an electronic component, comprising: Forming a first electrode formation layer (4) in a recess formed in a workpiece, the recess having an opening diameter of 15 nm or less; performing plasma nitriding directly on the surface of the first electrode formation layer (4); and Forming a second electrode formation layer (7) which is free of Ti and directly contacts the electrode formation layer (4) which has been subjected to plasma nitriding, wherein the first electrode formation layer (4) is a layer formed by sputtering while forming a magnetic field on a target surface by a magnet unit having a plurality of magnets arranged at lattice points of a polygonal lattice, which magnets are arranged such that adjacent magnets have opposite poles, Plasma nitriding takes place while a magnetic field is created on the target surface by the magnet unit, and the second electrode forming layer (7) is a layer having a flat surface formed by sputtering while a magnetic field is formed on the target surface by the magnet unit.
Citation Information
Patent Citations
Method for forming dual metal gate of semiconductor element
JP2002198441A
Method and Apparatus for Performing High Pressure Physical Vapor Deposition
JP2004506090A
Method of fabricating semiconductor device
JP2007295004A
Method for manufacturing electronic component including procedure of filling metallic film
JP2011153374A
Method and apparatus for low thermal metal fill and planarization of contacts, vias and trenches in semiconductor wafers
JP3193875B2
