JP5332776B2 - Method for manufacturing transfer mask - Google Patents

Description

  The present invention relates to a method for manufacturing a transfer mask used for exposure of a charged particle beam such as an electron beam or an ion beam, and a transfer mask.

  In recent years, miniaturization of LSIs and the like has progressed rapidly, and development of lithography techniques for forming further fine circuit patterns of these elements has been promoted. In particular, in pattern formation with a line width of 65 nm or less, the exposure method using a conventional ArF excimer laser as an exposure light source has reached the resolution limit. As an alternative lithography technique, an immersion lithography method in which the space between the lens and the wafer to be exposed is filled with a medium having a refractive index higher than that of air to improve the effective resolution has become the mainstream.

  However, when the immersion lithography method is used, the effective resolution can be improved, but the depth of focus becomes low, so that it is difficult to form a high aspect ratio hole pattern. In addition, the resolution limit may be reached for a fine pattern of 45 nm node or less by the immersion lithography method. One of the methods for solving this problem is charged particle beam lithography.

  The charged particle beam lithography is a technique that uses a charged particle beam such as an electron beam or an ion beam as an exposure light source instead of an excimer laser such as ArF or KrF that has been conventionally used.

  In charged particle beam lithography, a charged particle beam serving as an exposure light source is irradiated onto a transfer mask on which a desired charged particle beam transmission hole pattern is formed, and a resist on the wafer is exposed to form a fine pattern. This method can be expected to improve the depth of focus and the resolution compared to the conventional exposure method using an excimer laser.

  Conventionally, SOI (Silicon On Insulator) wafers are often used for manufacturing transfer masks used for charged particle beam exposure. As shown in FIG. 11, in the SOI wafer (14), an etching stopper layer 12 made of a silicon oxide film is formed on a support layer 11 made of single crystal silicon, and the single crystal silicon is made on the etching stopper layer 12. It has a three-layer structure in which a thin film layer 13 is formed.

  An example of the structure of the transfer mask 17 using an SOI wafer is shown in FIG. An opening 15 for transmitting charged particle beams is formed in the support layer 11, and a charged particle beam transmitting hole 16 for forming charged particle beams in a fine pattern is formed in the thin film layer 13. The thin film layer 13 is a single-layer self-supporting film (hereinafter referred to as a membrane). The thickness of the membrane varies depending on the acceleration voltage of the charged particle beam used for the exposure and the exposure method. The resist is exposed by the charged particle beam that has passed through.

  Usually, in manufacturing a transfer mask, a charged particle beam transmission hole and an opening are separately formed, and therefore an etching stopper layer is required when forming the charged particle beam transmission hole and the opening. An advantage of using an SOI wafer for manufacturing a transfer mask is that a thin film layer can be used as a membrane and a silicon oxide film can be used as an etching stopper layer. When a thin film layer and an etching stopper layer are formed on a silicon wafer by a CVD (Chemical Vapor Deposition) method or a PVD (Physical Vapor Deposition) method, precise control of the thickness of each layer and management of film defects are required. The number of processes and manufacturing cost for mask manufacturing increase.

  On the other hand, by using an SOI wafer with a high degree of technical perfection as a transfer mask manufacturing wafer, it is possible to reduce the number of steps and costs involved in manufacturing the transfer mask. For this reason, SOI wafers are often used as substrates for manufacturing transfer masks for charged particle beam exposure.

  The use of an SOI wafer on which a thin film layer or an etching stopper layer has been formed in advance facilitates the manufacture of the transfer mask itself, while the entire SOI wafer is warped due to the compressive stress of the etching stopper layer. There is. In general, the stress of a thin film is divided into tensile stress and compressive stress. The tensile stress acts in the direction in which the film itself contracts, and the substrate on which the thin film having the tensile stress is warped in a concave shape on the thin film side. Since the compressive stress exerts a force in the direction in which the thin film itself extends, the substrate after film formation warps convexly toward the thin film side.

  The etching stopper layer made of a silicon oxide film that causes warping of the SOI wafer is a compressive stress, and this compressive stress generates a convex warped deformation on the etching stopper layer side. The amount of warpage varies somewhat depending on the SOI wafer manufacturing method, but is generally determined by the stress and film thickness of the etching stopper layer. The silicon oxide film constituting the etching stopper layer is a thermal oxide film formed by heat treatment, and its stress is a compressive stress of about 300 MPa.

  The amount of warpage of the entire SOI wafer differs depending on the film thickness and diameter of each layer constituting the SOI wafer in addition to the stress of the etching stopper layer. For example, the film thickness of the thin film layer is 2 μm, and the film thickness of the etching stopper layer is In an 8-inch SOI wafer having a thickness of 1 μm and a support layer thickness of 725 μm, it is expected that a warp in the form of swelling about 85 μm on the thin film layer side of the entire SOI wafer will occur.

  When such a large warp occurs in the SOI wafer, the position accuracy of the charged particle beam transmission hole formed in the membrane is adversely affected. For example, the transfer mask is installed in the exposure machine in a state where it is installed in an electrostatic chuck type mask holder. However, if a transfer mask having a large warp is installed in the mask holder, the entire transfer mask is deformed. This deformation does not have reproducibility and regularity, and it is impossible to take avoidance measures such as determining the position of the charged particle beam transmission hole by predicting the deformation.

  Further, the warpage of the SOI wafer has an adverse effect on the production of the transfer mask. For example, a charged particle beam transmission hole formed on a membrane is formed by plasma etching using a resist pattern formed by charged particle beam drawing as an etching mask. When a SOI wafer is installed in a charged particle beam drawing machine, Since the SOI wafer itself undergoes non-reproducible deformation as in the case of the exposure machine, the positional accuracy of the resist pattern of the charged particle beam transmission hole formed on the membrane also deteriorates.

  For this reason, in order to manufacture a transfer mask with high transfer accuracy and to prevent deterioration of the transfer accuracy of the mask installed in the exposure machine, it is necessary to manufacture the transfer mask using an SOI wafer with less warping deformation. It becomes.

  As a method for reducing the warpage of the SOI wafer, for example, a method has been proposed in which a silicon oxide film previously formed on the back surface of the support layer of the SOI wafer is used as the warp adjustment layer (see, for example, Patent Document 1). Although this method is effective, when removing the etching stopper layer made of the silicon oxide film, it is necessary to take measures so that the silicon oxide film used as the warp adjusting layer is not etched at the same time. This method also leaves an insulating silicon oxide film on the back surface of the transfer mask. Therefore, when the charged particle beam is irradiated on the back surface of the transfer mask during actual charged particle beam drawing, the transfer mask itself is charged up. However, there is a problem that the charged particle beam is deflected and transfer accuracy is greatly deteriorated.

JP 2002-151385 A

  An object of the present invention is to obtain a transfer mask having excellent transfer accuracy, which can greatly reduce a fatal defect of the transfer mask, which is a warp caused by the compressive stress of an etching stopper layer.

According to a first aspect of the present invention, there is provided a transfer mask manufacturing method comprising: forming an etching stopper layer made of silicon oxide by a thermal oxidation method in a region from a surface to a predetermined depth in a membrane forming region in a support layer made of single crystal silicon. Polishing the etching stopper layer; forming a thin film layer on the surface of the support layer on which the etching stopper layer is formed; and the support layer on which the etching stopper layer and the thin film layer are formed. Of these, the step of removing the support layer in the membrane forming region from the back surface to form an opening, the step of removing the etching stopper layer through the opening of the support layer, and the opening in the thin film layer And a step of forming a charged particle beam transmission hole communicating therewith.

According to a second aspect of the present invention, there is provided a transfer mask manufacturing method comprising: preparing a support layer made of single-crystal silicon; and subjecting the entire surface of the support layer to a region excluding a membrane formation region and the back surface of the support layer by thermal oxidation. A step of forming a mask for patterning by a patterning method, a step of forming an etching stopper layer made of silicon oxide by a thermal oxidation method in a region from the surface to a predetermined depth in the membrane forming region of the support layer, and the etching stopper layer A step of forming a thin film layer on the surface of the support layer on which the etching stopper layer is formed, and the membrane formation region of the support layer on which the etching stopper layer and the thin film layer are formed. Removing the support layer from the back surface to form an opening, and through the opening of the support layer Wherein for the step of removing the etching stopper layer, and forming a charged particle beam transmission hole communicating with the opening in the thin film layer, characterized by characterized by at least Te.

Method for producing a transfer mask according to the invention of claim 3 is the invention of claim 1 or 2, after the step of forming a thin layer on the support layer surface, and further comprising a step of polishing the thin film layer To do.

According to a fourth aspect of the present invention, there is provided a transfer mask manufacturing method according to any one of the first to third aspects, wherein the thin film layer is one of a single crystal silicon layer and a polycrystalline silicon layer. And

  According to the present invention, it is possible to manufacture a transfer mask in which an etching stopper layer made of a silicon oxide film having a large compressive stress does not remain. Therefore, a fatal defect of the transfer mask called warpage caused by the compressive stress of the etching stopper layer is eliminated. A transfer mask that can be greatly reduced and has excellent transfer accuracy can be obtained.

  Therefore, when the transfer mask manufactured by the transfer mask manufacturing method of the present invention is used, patterns of semiconductor devices and the like can be manufactured with a high yield.

It is a schematic sectional drawing which shows the process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows the other process of the manufacturing method of the transfer mask which concerns on Embodiment 1 of this invention. It is a schematic sectional drawing which shows an example of the conventional SOI wafer. It is a schematic sectional drawing which shows an example of the conventional transfer mask.

  Hereinafter, a method for manufacturing a transfer mask according to the first embodiment of the present invention will be described in detail with reference to the drawings. 1 to 10 are schematic cross-sectional views showing the steps of the transfer mask manufacturing method according to Embodiment 1 of the present invention.

  First, as shown in FIG. 1, a support layer (substrate) 21 made of single crystal silicon is prepared. In addition, let the surface which forms the thin film layer in the support layer 21 be a surface, and let the surface on the opposite side to the said surface be a back surface. Here, in order to accurately process the thin film layer 24 to be performed later, it is desirable to use the support layer 21 having as little warp as possible.

  Next, a thermal oxidation mask 22 is formed on the entire surface of the support layer 21 excluding the membrane formation region A and the entire back surface of the support layer 21 using a known patterning method (see FIG. 2).

  Next, by performing a thermal oxidation process on the support layer 21 on which the thermal oxidation mask 22 is formed, in the membrane formation region A in the support layer 21, an etching made of silicon oxide in a region from the surface to a predetermined depth. A stopper layer 23 is formed. Thereafter, the unnecessary thermal oxidation mask 22 is removed (see FIG. 3).

  Here, when the etching stopper layer 23 made of silicon oxide is formed on the surface of the support layer 21 made of single crystal silicon by a thermal oxidation method, the surface portion of the etching stopper layer 23 usually rises from the surface of the support layer 21. Thereby, distribution may arise in the film thickness of the thin film layer formed later. Therefore, when it is necessary to make the thickness of the thin film layer as uniform as possible, the surface portion of the etching stopper layer 23 may be polished after the step of forming the etching stopper layer 23 on the surface of the support layer 21. Desirable (see FIG. 4).

  Next, a thin film layer 24 is formed by CVD or the like on the surface of the support layer 21 on which the etching stopper layer 23 is formed (see FIG. 5).

  Here, as a material for forming the thin film layer 24, for example, since a processing technique in a conventional transfer mask manufactured using an SOI wafer can be applied, single crystal silicon can be preferably used. However, when it is difficult to form a single crystal silicon layer on the surface of the support layer 21, polycrystalline silicon can be preferably used instead.

  Further, the film thickness of the thin film layer 24 formed on the surface of the support layer 21 needs to be larger than a desired film thickness when the thin film layer 24 is polished later.

  Further, in the step of forming the thin film layer 24 on the surface of the support layer 21, the stress of the thin film layer 24 is controlled so that the warp of the substrate is as close to zero as possible when the thin film layer 24 later becomes a membrane. It is desirable.

  Further, when it is necessary to make the film thickness of the thin film layer 24 formed later as uniform as possible, it is desirable to polish the thin film layer 24 (see FIG. 6).

  Next, a photoresist or the like is applied to the back surface of the support layer 21 with a spinner or the like to form a photosensitive layer, and patterning processing such as pattern exposure and development is performed to form an etching mask 25 (see FIG. 7).

  Next, the support layer 21 is etched by dry etching or the like using the etching mask 25 until the etching stopper layer 23 formed in the membrane formation region A is reached, so that the opening 26 is formed. Thereafter, the etching stopper layer 23 and the etching mask 25 remaining in the membrane formation region A are removed, and the membrane formation region A is made into a membrane composed of the thin film layer 24 (see FIG. 8).

  Next, an electron beam resist or the like is applied onto the membrane formed of the thin film layer 24 with a spinner or the like to form a photosensitive layer, and patterning processing such as electron beam drawing and development is performed to form an etching mask 27 ( (See FIG. 9).

  Next, the charged particle beam transmission hole 28 is formed in the membrane by etching using the etching mask 27 by dry etching or the like. Thereafter, the remaining etching mask 27 is removed to obtain a transfer mask 29 (see FIG. 10).

  In the first embodiment, the charged particle beam transmission hole 28 is formed in the membrane made of the thin film layer 24 after the opening 26 is formed in the support layer 21. The opening 26 may be formed in the support layer 21 after the particle beam transmission hole 28 is formed.

  According to the above-described transfer mask manufacturing method according to the first embodiment of the present invention, unlike a conventional transfer mask manufactured using an SOI wafer, an etching stopper layer made of a silicon oxide film having a large compressive stress does not remain. A transfer mask can be manufactured. That is, according to the above-described transfer mask manufacturing method according to the first embodiment of the present invention, warpage, which is a fatal defect of the transfer mask, is greatly reduced as compared with a conventional transfer mask manufactured using an SOI wafer. A transfer mask having excellent transfer accuracy can be obtained. As a result, when the transfer mask manufactured by the transfer mask manufacturing method according to Embodiment 1 of the present invention is used, patterns of semiconductor devices and the like can be manufactured with a high yield.

Example 1
Embodiment 1 of the transfer mask manufacturing method of the present invention will be described below.

  First, a support layer (substrate) 21 having a diameter of 200 mm and made of single crystal silicon having a thickness of 725 μm was prepared. Note that a surface of the support layer 21 on which the thin film layer is formed is a front surface, and a surface opposite to the surface is a back surface.

  Here, the amount of warpage of the substrate was measured to be +2.0 μm. The sign of the amount of warpage is defined as + (plus) when it swells on the front surface side of the support layer 21 and-(minus) when it swells on the back surface side. Further, in order to exclude the influence of the substrate's own weight, ½ of the difference between the warpage amount measured with the front surface side of the support layer 21 as the upper surface and the warpage amount measured with the back surface side of the support layer 21 as the upper surface is The amount of warpage was set (see FIG. 1).

  Next, a silicon nitride film having a thickness of 500 nm was formed on both the front and back surfaces of the support layer 21 by a CVD method. Thereafter, the entire surface of the back surface of the support layer 21 and the membrane formation region A are removed by processing the silicon nitride film formed on the surface of the support layer 21 using photolithography and dry etching using a mixed gas plasma of fluorocarbon type. A thermal oxidation mask 22 made of a silicon nitride film was formed on the surface (see FIG. 2).

  Next, the surface of the support layer 21 in the membrane formation region A is made of silicon oxide by a thermal oxidation method in which the support layer 21 on which the thermal oxidation mask 22 is formed is heated to 1000 ° C. in an atmosphere containing oxygen. An etching stopper layer 23 was formed. Thereafter, the unnecessary thermal oxidation mask 22 was removed by immersing the substrate in a phosphoric acid solution (see FIG. 3).

  Next, by polishing the surface portion of the etching stopper layer 23, the surface of the etching stopper layer 23 was made flat with respect to the surface of the support layer 21. As a result, the film thickness of the etching stopper layer 23 was 1.0 μm (see FIG. 4).

  Next, a thin film layer 24 made of polycrystalline silicon was formed by CVD on the surface of the support layer 21 on which the etching stopper layer 23 was formed (see FIG. 5).

  Next, by polishing the surface portion of the thin film layer 24, the surface of the thin film layer 24 was made flat. As a result, the film thickness of the thin film layer 24 became 2.0 micrometers (refer FIG. 6).

  Next, a photoresist was applied to the back surface of the support layer 21 with a spinner to form a photosensitive layer having a thickness of 50 μm, and patterning processes such as pattern exposure and development were performed to form an etching mask 25 (see FIG. 7). .

  Next, the support layer 21 is etched by dry etching using a fluorocarbon-based mixed gas plasma using the etching mask 25 until the etching stopper layer 23 formed in the membrane formation region A is reached. Formed. Thereafter, the etching stopper layer 23 and the etching mask 25 remaining in the membrane formation region A were removed by a hydrofluoric acid solution and a resist stripping solution, respectively, so that the membrane formation region A was a membrane composed of the thin film layer 24 (see FIG. 8). ).

  Next, an electron beam resist is applied onto the membrane made of the thin film layer 24 with a spinner to form a photosensitive layer having a thickness of 1.0 μm, and patterning processing such as electron beam drawing and development is performed to form an etching mask 27. (See FIG. 9).

  Next, charged particle beam transmission holes 28 were formed in the membrane by dry etching using a fluorocarbon-based mixed gas plasma using the etching mask 27. Then, the transfer mask 29 was obtained by removing the remaining etching mask 27 (see FIG. 10).

  The amount of warpage of the transfer mask 29 produced by the above method was −1.5 μm, and the surface of the transfer mask 29 was very flat. In addition, when the transfer mask 29 was used to perform electron beam exposure and the pattern transfer accuracy was measured, the transfer pattern was compared with the case where a transfer mask having a conventional structure manufactured using an SOI wafer provided with the same pattern was used. It has been confirmed that the positional accuracy of is significantly improved.

DESCRIPTION OF SYMBOLS 21 Support layer 22 Thermal oxidation mask 23 Etching stopper layer 24 Thin film layer 25 Etching mask 26 Opening part 27 Etching mask 28 Charged particle beam transmission hole 29 Transfer mask A A membrane formation area

Claims (4)

A step of forming an etching stopper layer made of silicon oxide in a region from the surface to a predetermined depth in the membrane forming region in the support layer made of single crystal silicon by a thermal oxidation method;
Polishing the etching stopper layer;
Forming a thin film layer on the surface of the support layer on which the etching stopper layer is formed;
Removing the support layer in the membrane formation region from the back surface of the support layer on which the etching stopper layer and the thin film layer are formed, and forming an opening;
Removing the etching stopper layer through the opening of the support layer;
Forming a charged particle beam transmission hole communicating with the opening in the thin film layer;
A method for manufacturing a transfer mask, comprising: Preparing a support layer made of single crystal silicon;
Forming a mask for thermal oxidation by a patterning method on the entire surface of the support layer excluding the membrane formation region and the entire back surface of the support layer;
Forming an etching stopper layer made of silicon oxide by a thermal oxidation method in a region from the surface to a predetermined depth in the membrane formation region in the support layer;
Polishing the etching stopper layer;
Forming a thin film layer on the surface of the support layer on which the etching stopper layer is formed;
Removing the support layer in the membrane formation region from the back surface of the support layer on which the etching stopper layer and the thin film layer are formed, and forming an opening;
Removing the etching stopper layer through the opening of the support layer;
Forming a charged particle beam transmission hole communicating with the opening in the thin film layer;
A method for manufacturing a transfer mask, comprising: Wherein after the step of forming a thin layer on the support layer surface, method of manufacturing a transfer mask according to claim 1 or 2, characterized in that it comprises a step of polishing the thin film layer. The method for manufacturing a transfer mask according to any one of claims 1 to 3 , wherein the thin film layer is one of a single crystal silicon layer and a polycrystalline silicon layer .

Images