JP7488147B2 - Hard mask and method for manufacturing the same - Google Patents

Description

The present invention relates to a hard mask that is formed on the surface of a processing object to limit the processing area when the processing object is subjected to a specific processing, and a method for manufacturing the hard mask.

For example, in the manufacturing process of a semiconductor device, there is a process of performing a dry etching process on a substrate or a predetermined thin film (e.g., _SiO2 film) formed on the substrate surface, and at this time, a hard mask, for example, is provided on the surface of the processing object to limit the processing range of the dry etching. As this type of hard mask, it is generally known to use a laminated film of a first metal film such as a TiN film and a second metal film such as a W film (tungsten-containing film) (see, for example, Patent Document 1). In this case, a TiN film and a tungsten-containing film are sequentially formed on the surface of the processing object by a sputtering method, and then a predetermined opening is patterned by, for example, a lithography technique.

Such hard masks are required to have a density that is resistant to dry etching, as well as good adhesion to the object to be processed (in other words, low stress on each film that constitutes the hard mask), and to be able to pattern openings with high accuracy in a later process. However, as in the above conventional example, if the tungsten-containing film is formed by sputtering a tungsten-containing metal target, it has been found that this can lead to deterioration of the processed shape of the opening when patterning the opening in a later process. Therefore, the inventors of the present application have conducted extensive research and have discovered that the deterioration of the processed shape is due to the large crystal grains (for example, an average grain size of 50 nm or more) of the tungsten-containing film formed by the sputtering method.

JP 2014-78579 A

The present invention was made based on the above findings, and aims to provide a hard mask having a tungsten-containing film that has a density that exhibits dry etching resistance, does not impair the function of good adhesion to the processing object, and enables patterning of openings without causing deterioration of the processed shape in a later process, and a method for manufacturing the hard mask.

In order to solve the above problems, the hard mask of the present invention is formed on the surface of a processing object to limit the processing area when the processing object is subjected to a predetermined processing, and is characterized in that it has an underlayer and a tungsten-containing film laminated on this underlayer, and the underlayer has a surface roughness Ra of 0.35 nm or less on the surface on which the tungsten-containing film is laminated .

In the present invention, it is preferable that the underlayer is composed of at least one selected from a Ti film, a TiN film, a WN film, and a TiWN film.

In order to solve the above problem, the method for manufacturing the hard mask of the present invention includes a first step of forming a TiN film as an underlayer on the surface of the processing object, and a second step of forming a tungsten-containing film on the surface of the underlayer, and the first step is characterized in that a rare gas and a nitrogen-containing gas are introduced into a processing chamber in a vacuum atmosphere in which a titanium target and a processing object are placed, power is applied to the target to form a TiN film by reactive sputtering, and the flow rate ratio of the nitrogen-containing gas to the rare gas is set to 3.9 or less. In this case, a rare gas is introduced into a processing chamber in a vacuum atmosphere in which a tungsten target and a processing object are placed, power is applied to the target to form a tungsten-containing film by sputtering, and the power applied to the target is preferably set within the range of 4 kW to 8 kW.

Here, when the hard mask is composed of a laminated film of a TiN film as an underlayer formed by reactive sputtering using a Ti target and nitrogen as a reactive gas, and a W film formed by sputtering using a W target, in order to pattern the opening without causing deterioration of the processed shape in the subsequent process, the above findings suggest that it is necessary to microcrystallize (or more preferably, to make the W film non-crystallized). To achieve this, it is considered that the W film laminated thereon can be microcrystallized by adjusting the surface roughness Ra of the TiN film within a predetermined range, suppressing as much as possible the movement of sputter particles that fly off from the W target and adhere to the TiN film surface in the early stage of W film formation, and increasing the number of nuclei generated by the aggregation of the sputter particles.

In the present invention, the flow rate ratio of the nitrogen-containing gas to the rare gas in the first step is set to 3.9 or less, thereby forming a TiN film with a surface roughness Ra of 0.35 nm or less on the surface of the object to be processed. It has been confirmed that when a W film is laminated on this, the W film is more microcrystalline than that of the conventional example. As a result, since the film configuration of the hard mask is not changed, the function of having a density that exhibits dry etching resistance and good adhesion to the object to be processed is equivalent to that of the conventional example described above, and since the W film is microcrystalline, it is possible to pattern the opening without causing deterioration of the processed shape in the subsequent process.

FIG. 2 is a schematic cross-sectional view showing a hard mask according to an embodiment of the present invention. FIG. 2 is a diagram for illustrating a sputtering apparatus configured as a cluster tool for carrying out the hard mask manufacturing method of the present embodiment. FIG. 3 is a diagram for explaining the film formation chamber shown in FIG. 2 . 1A is an FIB-SIM image showing the results of invention experiment 1 for confirming the effects of the present invention, and FIG. 1B is an FIB-SIM image showing the results of comparison experiment 1. 1A and 1B are AFM images showing the results of an inventive experiment 2 confirming the effects of the present invention, and FIG. 1C is an AFM image showing the results of a comparative experiment 2. 1A and 1B are diagrams showing the results of an inventive experiment 2 for confirming the effects of the present invention, and FIG. 1C is a diagram showing the results of a comparative experiment 2. 1A and 1B are AFM images showing the results of an inventive experiment 3 for confirming the effects of the present invention, and FIG. 1C is an AFM image showing the results of a comparative experiment 3. 1A and 1B are SEM images showing the results of an inventive experiment 3 confirming the effects of the present invention, and FIG. 1C is an SEM image showing the results of a comparative experiment 3. 11 is a graph showing experimental results confirming the effects of the present invention. 11 is a graph showing experimental results confirming the effects of the present invention.

The following describes, with reference to the drawings, an embodiment of a hard mask HM that is formed on the surface of the processing object Sw to limit the etching processing area when the processing object Sw is subjected to dry etching processing, and a method for manufacturing the same.

Referring to FIG. 1, the hard mask HM has an underlayer Ly1 formed on the surface of the processing object Sw, and a tungsten-containing film Ly2 laminated on the underlayer Ly1. As the processing object Sw, a substrate such as a silicon wafer or a substrate having a predetermined thin film (e.g., an insulating film such as a SiO ₂ film (TEOS film) or a metal film such as an Al film) formed on the surface of the substrate can be used. As the underlayer Ly1, a layer having high adhesion to the processing object Sw is used, and for example, one selected from a Ti film, a TiN film, a WN film, and a TiWN film can be used, and preferably, a TiN film can be used. As the tungsten-containing film Ly2, a layer having a density (e.g., 19 g/cm ³ or more) that exhibits dry etching resistance and a high selectivity (e.g., 200 or more) to the SiO ₂ film (TEOS) is used, and for example, one selected from a W film and a WN film can be used, and preferably, a W film can be used. The thickness of the underlayer Ly1 can be set within a range of, for example, 1 to 30 nm, and the thickness of the tungsten-containing film Ly2 can be set within a range of, for example, 100 to 300 nm.

An opening Op having a predetermined contour is patterned in the hard mask HM using known lithography techniques, and the portion of the workpiece Sw exposed at the bottom of this opening Op is dry etched to form the desired etched shape.

The tungsten-containing film Ly2 is generally formed by sputtering using a W target. However, when the tungsten-containing film Ly2 is formed by sputtering, the processed shape of the opening Op may deteriorate. Through intensive research, the inventors have discovered that this is due to the large crystal grains of the tungsten-containing film Ly2 formed by sputtering. Based on this discovery, it is necessary to microcrystallize (more preferably, to non-crystallize) the tungsten-containing film Ly2.

Therefore, in this embodiment, the underlayer Ly1 formed on the surface of the processing object Sw has a surface roughness Ra of 0.35 nm or less, and when the tungsten-containing film Ly2 is formed on the surface of the underlayer Ly1 by sputtering, the tungsten-containing film Ly2 is confirmed to be microcrystalline compared to the conventional example. As a result, since the film configuration of the hard mask HM is not changed, the function of having a density that exhibits dry etching resistance and good adhesion to the processing object Sw is equivalent to that of the conventional example, and since the tungsten-containing film Ly2 is microcrystalline, the opening Op can be patterned and formed without causing deterioration of the processed shape in the subsequent process. Below, an embodiment of the manufacturing method of the hard mask HM will be described using an example in which the underlayer Ly1 is a TiN film and the tungsten-containing film Ly2 is a W film. Note that the opening Op can be formed by using known lithography techniques, etc., and detailed description will be omitted here.

With reference to FIG. 2, SM is a sputtering apparatus configured with a cluster tool that implements the manufacturing method of the hard mask HM of this embodiment. The sputtering apparatus SM includes a transfer chamber Tc in which a transfer robot R is disposed, and deposition chambers Pc1, Pc2 and a load lock chamber Lc that are disposed so as to surround the transfer chamber Tc. The transfer chamber Tc, the load lock chamber Lc, and the deposition chambers Pc1, Pc2 are each connected to an exhaust pipe leading to a vacuum pump unit (sometimes not shown), so that each chamber can be evacuated. In addition, a vent gas line (not shown) is further connected to the load lock chamber Lc, so that the load lock chamber Lc can be vented to atmospheric pressure. As the transfer robot R, for example, a known so-called frog leg type having a robot arm 10a and a robot hand 10b provided at the tip of the robot arm 10a to hold a substrate Sw can be used, and the substrate Sw can be transported to a predetermined position in each chamber by the transfer robot R. The transfer chamber Tc, the load lock chamber Lc, and the deposition chambers Pc1 and Pc2 are connected via a gate valve Iv, so that each chamber can be isolated from the others. The deposition chamber Pc1 for depositing the TiN film Ly1 and the deposition chamber Pc2 for depositing the W film Ly2 have the same structure except for the material of the target 3, so the deposition chamber Pc1 will be described below as a representative.

Referring also to FIG. 3, the deposition chamber Pc1 is defined by a vacuum chamber 1. An exhaust pipe 11 leading to a vacuum pump unit Pu consisting of a turbo molecular pump, a rotary pump, or the like is connected to the bottom wall of the vacuum chamber 1, so that the deposition chamber Pc1 can be evacuated to a vacuum. A gas pipe 13 with a mass flow controller 12 is connected to the side wall of the vacuum chamber 1, so that a rare gas (e.g., argon gas) and nitrogen gas can be introduced into the deposition chamber Pc1 in a vacuum atmosphere at predetermined flow rates. In the following, terms indicating directions such as "up" and "down" are explained based on the installation position shown in FIG. 3.

A stage 2 is disposed at the bottom of the vacuum chamber 1 via an insulator _I1 , so that the substrate Sw can be positioned and held with its film-forming surface facing up. The stage 2 incorporates a heater 21 as a heating means, and by passing electricity through the heater 21 from a power source (not shown), the substrate Sw can be heated to a predetermined temperature during film formation.

A cathode unit C is attached to the ceiling of the vacuum chamber 1. The cathode unit C has a Ti target 3 arranged facing the substrate Sw, and a magnet unit 4 arranged above the target 3. The target 3 has a shape (circular in plan view) corresponding to the contour of the substrate Sw, and is attached to the lower surface of a backing plate 31 attached to the vacuum chamber 1 via an insulator _I2 . The target 3 is connected to an output from a sputtering power source E, such as a DC power source or a high-frequency power source with a predetermined frequency (for example, 150 KHz to 13.56 MHz), depending on the material of the target 3, so that a predetermined power can be input to the target 2 during film formation. As the magnet unit 4, a known structure that generates a magnetic field in the space below the sputtering surface 3a of the target 3, captures electrons ionized below the sputtering surface 3a during sputtering, and efficiently ionizes sputtered particles scattered from the target 3 can be used, so a detailed description will be omitted here.

The sputtering apparatus SM has a known control means Ct equipped with a microcomputer, sequencer, etc., which is not specifically shown, and the control means Ct is used to comprehensively control the operation of the sputtering power supply E, the mass flow controller 12, the vacuum pump unit Pu, the magnet unit 4, etc.

The substrate Sw before film formation is placed in the load lock chamber Lc of the sputtering device SM, and the load lock chamber Lc is evacuated to a predetermined vacuum level. The substrate Sw is then transported from the load lock chamber Lc to the film formation chamber Pc1 in a vacuum atmosphere via the transport chamber Tc by the transport robot R, and the substrate Sw is transferred to the stage 2. Argon gas and nitrogen gas are introduced into the film formation chamber Pc1 as sputtering gas at flow rates of 15 to 50 sccm and 20 to 200 sccm, respectively (at this time, the pressure in the film formation chamber Pc1 becomes 0.1 to 30 Pa) to maintain the so-called transition region (region where the metal mode transitions to the compound mode) in reactive sputtering. In addition, DC power of, for example, 5 kW to 50 kW is input from the sputtering power source E to the target 3. During sputtering, the substrate Sw is heated to a predetermined temperature (100 to 300°C) by the heater 21. This creates a plasma atmosphere in the vacuum chamber 1, and the titanium target 3 is reactively sputtered. At this time, by setting the flow rate ratio of the nitrogen-containing gas to the rare gas to 3.9 or less, the surface roughness Ra of the TiN film Ly1 formed on the surface of the substrate Sw can be set to 0.35 nm or less.

After the formation of the TiN film Ly1 is completed, the substrate Sw is transported to the film formation chamber Pc2 and handed over to the stage 2. Argon gas is introduced into the film formation chamber Pc2 at a flow rate of 100 to 200 sccm (at this time, the pressure in the film formation chamber Pc2 is 0.1 to 30 Pa). In addition, 4 kW to 8 kW of DC power with a negative potential is input from the sputtering power supply E to the target 3. This forms a plasma atmosphere in the vacuum chamber 1, and the tungsten target 3 is sputtered. During sputtering, the substrate Sw is heated to a predetermined temperature (room temperature (20°C) to 350°C) by the heater 21. Sputter particles scattered from the target 3 by sputtering adhere to and deposit on the surface of the TiN film Ly1, so that the W film Ly2 is formed (laminated) on the TiN film Ly1. It has been confirmed that the W film Ly2 is microcrystalline (or amorphous) compared to that of the conventional example.

After the W film Ly2 is formed, the substrate Sw is transported to the load lock chamber Lc, which is then vented to atmospheric pressure, and the substrate Sw is then retrieved. Then, in a post-process, an opening Op is patterned in the W film Ly2 and the TiN film Ly1 using known lithography techniques or the like.

As described above, by setting the flow rate ratio of the nitrogen-containing gas to the rare gas introduced during the formation of the TiN film Ly1 to 3.9 or less, it is possible to form a TiN film Ly1 having a surface roughness Ra of 0.35 nm or less. Since the surface of the formed TiN film Ly1 has low lubricity, when the W film Ly2 is formed on the surface of the TiN film Ly1 by sputtering, the movement of sputtered particles on the surface of the TiN film Ly1 is suppressed. As a result, the number of nuclei generated by the aggregation of sputtered particles at the beginning of the formation of the W film Ly2 increases, and the W film Ly2 becomes more microcrystalline than that of the conventional example. Therefore, even if the opening Op is patterned in a later process, the processed shape of the opening Op does not deteriorate.

Next, in order to confirm the above-mentioned effects, the following experiments were carried out using the sputtering apparatus SM. First, in invention experiment 1, the processing object was a silicon wafer with a diameter of 300 mm on the surface of which a SiO ₂ (TEOS) film was formed to a thickness of 100 nm (hereinafter referred to as "substrate Sw"). The substrate Sw was held on stage 2 of film formation chamber Pc1 (stage temperature was 300° C.), argon gas and nitrogen gas were introduced into film formation chamber Pc1 at flow rates of 15 sccm and 58 sccm, respectively (the flow rate ratio of nitrogen gas to argon gas was 3.9 at this time), and 14 kW of DC power was applied to Ti target 3 to form a TiN film Ly1 to a thickness of 10 nm. The substrate Sw on which the TiN film Ly1 was formed was held on the stage 2 of the film formation chamber Pc2 (stage temperature: 100° C.), argon gas was introduced into the film formation chamber Pc2 at a flow rate of 100 sccm, and 4 kW of DC power was applied to the W target 3 to form a W film Ly2 to a thickness of 100 nm. The results of observing the surface of the W film Ly2 with a focused ion beam-scanning ion microscope (FIB-SIM: Focused Ion Beam-Scanning Ion Microscope) are shown in FIG. 4(a). According to this, it was confirmed that the W film Ly2 was microcrystallized compared to that in Comparative Experiment 1 described later (see FIG. 4(b)). This is also clear from the fact that the resistivity of the W film Ly2 was 20.2 μΩcm, which is one order of magnitude higher than the resistivity (9.8 μΩcm) in Comparative Experiment 1 described later.

In Comparative Experiment 1 for the above-mentioned Invention Experiment 1, a W film Ly2 was formed directly on the surface of the substrate Sw without forming a TiN film Ly1. The results of observing the surface of the formed W film Ly2 using FIB-SIM are shown in FIG. 4(b). It was confirmed that the crystal grains of the W film Ly2 were larger than those in the above-mentioned Invention Experiment 1. The resistivity value of this W film Ly2 was measured and found to be 9.8 μΩcm.

In invention experiment 2, the substrate Sw was held on the stage 2 of the film-forming chamber Pc1, argon gas was introduced into the film-forming chamber Pc1 at a flow rate of 45 sccm, and 4 kW DC power was applied to the titanium target 3 to form a Ti film Ly1 of 30 nm as a base layer. The surface of this Ti film Ly1 was observed by an atomic force microscope (AFM: Atomic Force Microscopy) (AFM image (convexo-concave image)) as shown in FIG. 5(a). The surface roughness Ra of this Ti film Ly1 was 0.31 nm. A W film Ly2 of 100 nm was formed on the surface of this Ti film Ly1 in the same manner as in invention experiment 1, and the average grain size of this W film Ly2 was measured by a laser diffraction grain size distribution measuring device, as shown in FIG. 6(a). The average grain size of the W film Ly2 was 31 nm. In addition, when the specific resistance value of the W film Ly2 was measured, it was equivalent to that of invention experiment 1 (20 to 23 μΩcm). These measurement results showed that the W film Ly2 was microcrystalline.

A TiN film Ly1 was formed in the same manner as in the above invention experiment 1, except that the thickness of the TiN film Ly1 was 30 nm. The result of observing the surface by AFM (AFM image) is shown in FIG. 5(b). The surface roughness Ra of this TiN film Ly1 was 0.35 nm. A W film Ly2 was formed to a thickness of 100 nm on the surface of this TiN film Ly1 in the same manner as in the above invention experiment 1, and the result of measuring the average grain size of this W film Ly2 is shown in FIG. 6(b). The average grain size of the W film Ly2 was 32 nm. The specific resistance value of the W film Ly2 was measured and was equivalent to that of the above invention experiment 1 (20 to 23 μΩcm). These measurement results showed that the W film Ly2 was microcrystallized.

In Comparative Experiment 2 for the above-mentioned invention experiment 2, the underlayer Ly1 was a laminated film of a WSi film 5 nm and a TiN film 30 nm. The deposition conditions for the WSi film were as follows: target 3 was made of WSi, argon flow rate was 100 sccm, pressure was 0.7 Pa, DC power was 4 kW, and substrate temperature was 300°C. The result of observing the surface of the laminated film Ly1 by AFM (AFM image) is shown in FIG. 5(c). The surface roughness Ra of this laminated film Ly1 was 0.43 nm. As in the above-mentioned invention experiment 1, a W film Ly2 was deposited to a thickness of 100 nm, and the result of measuring the average grain size of this W film Ly2 is shown in FIG. 6(c). The average grain size was 59 nm. In addition, the specific resistance value of the W film Ly2 was measured and found to be 11 μΩcm, which is the same as that of the above-mentioned comparative experiment 1. These measurement results show that if the surface roughness Ra of the underlayer Ly1 is greater than 0.35 nm, the W film Ly2 will not be microcrystallized.

In invention experiment 3, a TiN film Ly1 was formed to a thickness of 40 nm in the same manner as in invention experiment 1, except that the flow rates of argon gas and nitrogen gas were 15 sccm and 58 sccm, respectively (the flow rate ratio of nitrogen gas to argon gas was 3.9). The result of observing the surface by AFM (AFM image) is shown in FIG. 7(a). The surface roughness Ra of this TiN film Ly1 was 0.35 nm. The SEM image of the TiN film Ly1 formed is shown in FIG. 8(a). A W film Ly2 was formed to a thickness of 100 nm on the surface of the TiN film Ly1 formed in this way in the same manner as in invention experiment 1, and the result of observing the surface by AFM (AFM image) is shown in FIG. 7(b). The surface roughness Ra was 0.81 nm. The SEM image of the W film Ly2 formed is shown in FIG. 8(b). In addition, the flow rate of argon gas was fixed at 15 sccm, and only the flow rate of nitrogen gas was changed to 48 sccm and 150 sccm to form the TiN film Ly1 (the flow rate ratio of nitrogen gas to argon gas was 3.2 and 10), and the surface roughness Ra of each was measured, as shown in Figure 9, and was 0.33 nm and 0.47 nm. According to the present invention experiment 3, it was found that the surface roughness Ra of the TiN film Ly1 can be reduced to 0.35 nm or less by setting the flow rate ratio of nitrogen gas to argon gas to 3.9 or less.

Figure 7(c) shows an AFM image and Figure 8(c) shows an SEM image of a W film Ly2 formed directly on the surface of the substrate Sw to a thickness of 100 nm, as in the above Comparative Experiment 1. This shows that the crystal grains of the W film Ly2 are large, as in the above Comparative Experiment 1.

Although the embodiments of the present invention have been described above, various modifications are possible without departing from the scope of the technical concept of the present invention. In the above embodiments and experiments, a case where a microcrystalline W film Ly2 is formed compared to the conventional example has been described as an example, but the present invention can also be applied to the case where an amorphous W film is formed.

In the above embodiment and experiment, the flow rate ratio of the nitrogen-containing gas to the rare gas introduced during deposition of the TiN film serving as the underlayer Ly1 is set to 3.9 or less, thereby making the surface roughness Ra of the underlayer Ly1 0.35 nm or less. However, the surface roughness Ra of the underlayer Ly1 may be made 0.35 nm or less by appropriately setting deposition conditions other than the flow rate ratio.

In the above embodiment, the DC power input during deposition of the W film Ly2 is set within the range of 4 kW to 8 kW, but in order to deposit the W film Ly2 at a high deposition rate, it is possible to set the input power to 6 kW or more, more preferably 7 kW or more. When the input power is set to such a high value, it has been found that the W film has compressive stress, as shown in FIG. 10. In this case, by using a Ti film, TiN film, or WN film having tensile stress as the underlayer Ly1, the underlayer Ly1 can advantageously also function as a buffer film that relieves the stress of the W film Ly2.

HM...hard mask, Ly1...TiN film (underlayer), Ly2...W film (tungsten-containing film), Sw...substrate (object to be processed).

Claims (4)

A hard mask is formed on a surface of a processing object to limit a processing area when the processing object is subjected to a predetermined processing, the hard mask having an underlayer and a tungsten-containing film laminated on the underlayer,
A hard mask, wherein the underlayer has a surface roughness Ra of 0.35 nm or less on the surface on which the tungsten-containing film is deposited . The hard mask according to claim 1, characterized in that the underlayer is composed of at least one selected from the group consisting of a Ti film, a TiN film, a WN film, and a TiWN film. 3. A method for manufacturing the hard mask according to claim 2, comprising: a first step of forming a TiN film as an underlayer on a surface of an object to be processed; and a second step of forming a tungsten-containing film on the surface of the underlayer,
The first step of the method for manufacturing a hard mask is characterized in that a rare gas and a nitrogen-containing gas are introduced into a vacuum atmosphere processing chamber in which a titanium target and a processing object are placed, power is applied to the target to form a TiN film by reactive sputtering, and the flow rate ratio of the nitrogen-containing gas to the rare gas is set to 3.9 or less. The method for manufacturing a hard mask according to claim 3, characterized in that the second step includes introducing a rare gas into a vacuum atmosphere processing chamber in which a tungsten target and a processing object are placed, applying power to the target to form a tungsten-containing film by sputtering, and setting the applied power to the target within the range of 4 kW to 8 kW.

Images