JPH10135104A - Manufacturing method of charged particle ray or x-ray transferring mask - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method of manufacturing a mask for charged particle ray or X-ray transfer, in which a percentage of boundary regions not contributing to pattern formation can be reduced (or which can suppress increase in the area of the boundary regions). SOLUTION: In the method for manufacturing a mask for charged particle ray or X-ray transfer having columns Hs, corresponding to boundary regions of a pattern to be transferred onto a target, the pattern is a mask 10' which is made up of boundary regions 10b' having no pattern and a multiplicity of small regions 10a' divided by the boundary regions on a membrane 33'. A member of a 3-layer structure, that is, an Si layer 33, an intermediate layer 32 and an Si substrate 31 is prepared. The method includes steps of subjecting Si substrate 31 to an inductive coupled plasma etching, sidewall protective plasma etching or excessively low-temperature reactive ion etching process, to form a plurality of columns Hs having a wall face Hs' perpendicular or substantially perpendicular to the Si layer 33 and/or intermediate layer 32 at set positions corresponding to the boundary regions 10b', and also to form openings defined between the columns at set positions corresponding to small regions 10a'.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mask (mask and reticle) for transferring a predetermined pattern to a target (for example, a sensitive substrate) sensitive to X-rays or charged particle beams (electron beam, ion beam, etc.). And a method of manufacturing a mask in a broad sense, including:

[0002]

2. Description of the Related Art In recent years, the development of semiconductor integrated circuit technology has been remarkable, and the tendency of miniaturization and high integration of semiconductor elements has been remarkable. As a lithography apparatus for printing an integrated circuit pattern on a semiconductor wafer, a so-called optical stepper apparatus using light has been generally used.

[0003] However, as the miniaturization of circuit patterns progresses, there is a concern about the resolution limit of light, and electron beams, ion beams,
In recent years, studies and development of lithography apparatuses using X-rays have been actively conducted. Then, a charged particle beam reduction transfer apparatus or an X-ray reduction transfer apparatus that irradiates a mask having a predetermined pattern with an electron beam, an ion beam, or X-ray, and reduces and transfers a pattern in the irradiation range to a wafer by a projection optical system is described. Proposed.

Among the masks used in such a reduction transfer apparatus, for example, those shown in FIGS. 4 and 5 are known as masks for electron beams. Mask 21 of FIG.
Is a silicon mask substrate 22 provided with through holes 23, and the mask substrate 22 is formed with a thickness (for example, 50 μm) sufficient to absorb an electron beam.

[0005] The electron beam irradiated on the mask 21 passes through the through hole 2.
3, the electron beam EB having passed therethrough is focused on a resist surface of a sensitive substrate (eg, a silicon wafer coated with a resist) 25 by a pair of projection lenses 24a and 24b. The pattern corresponding to the shape is transferred. The mask 100 in FIG. 5 is obtained by forming a pattern of the scatterers 30a on the surface of a silicon mask substrate 20, and the mask substrate 20 is thinned to a thickness through which an electron beam can easily pass (membrane). .

When the mask 100 is irradiated with an electron beam,
Only the membrane 20 (the electron beam transmission region 20 of the membrane)
The electron beam EB2 (shown by a two-dot chain line in the figure) that also passed through the scatterer 30a has a greater degree of forward scattering than the electron beam EB1 (shown by a solid line in the figure) that passed through a). Therefore, if the aperture 7 is provided in the vicinity of the electron beam crossover image CO by the projection lens 5, a contrast according to the degree of scattering of the electron beams EB1 and EB2 on the sensitive substrate 110 can be obtained.

As shown in FIG. 4, a substrate having a through-hole pattern formed on a substrate is called a stencil mask. As shown in FIG. 5, a substrate having no through-hole and having a scatterer pattern formed on a membrane is scattered. It is called a transmission mask. The stencil mask (1) cannot form a donut-shaped pattern in which through holes are connected in a ring shape. (2) Since most of the electron beam is absorbed by the thick mask substrate 22, a large amount of heat is generated on the mask. Causing large thermal deformation (pattern distortion) of the mask,
And the like.

On the other hand, the scattering transmission mask is expected to solve the problems of the stencil mask. That is, in the scattering transmission mask, the membrane 20
5A can support the scatterer 30a.
As shown in the portion, a donut-like pattern in which the scatterers 30a are isolated in an island shape can be formed.

Further, since it is not necessary to completely block the electron beam with the scatterer 30a, the amount of the electron beam blocked at the scattering transmission mask portion is small. Then it is suppressed. As described above, in the stencil mask, the thickness of the substrate 22 is increased so that most of the electron beam is absorbed by the non-through hole portion of the substrate 22, so that a large amount of heat is generated in the mask. This causes large thermal deformation (pattern distortion) of the mask.

Therefore, in the case of a stencil mask, if the thickness of the substrate portion to be irradiated with the electron beam is reduced to form a membrane and a scattering stencil mask in which a through-hole pattern is formed in such a membrane, the above-mentioned problem (2) occurs. Can be solved. Examples of the material of the membrane used for each mask include Si ₃ N ₄ , Be, C (diamond), SiC, Al ₂ O ₃ , Al, Si, SiO _{2, and the} like, and the material of the scatterer used for the scattering transmission mask. For example, tungsten, Au, and the like have been studied.

However, Be has the disadvantage that its oxide is toxic, and Si ₃ N ₄ , C (diamond), SiC, Al ₂ O ₃ , Al, Si, SiO ₂
And the like, there is a disadvantage that the electron beam has a low transmittance due to a relatively short mean free path of the electron beam. As is clear from the above-described transfer principle of the scattering transmission mask, in order to increase the contrast of the pattern image projected on the sensitive substrate 110 and perform high-precision exposure, the absorption or scattering of the charged particle beam in the scatterer 30a must be reduced. It is preferable to increase the absorption and reduce the absorption and scattering of the charged particle beam in the membrane 20.

In the scattering stencil mask,
It is preferable to reduce the absorption of charged particle beams in the membrane that causes pattern distortion. However, the above-mentioned materials of the membrane are all amorphous single crystals or polycrystals, and the absorption or scattering of charged particle beams cannot be sufficiently suppressed unless they are considerably thin. Therefore, for example, in a scattering transmission mask, it has been considered to set the thickness of the membrane 20 to be as thin as about 10 nm. However, the energy of the scatterer 30a increases the temperature of the membrane and causes distortion in the mask (pattern distortion). And the transfer accuracy may be degraded.

In particular, as shown in part A of FIG. 5A, it is considered that the heat of the scatterer 30a hardly escapes at a portion where the scatterer 30a is isolated in an island shape, and the temperature rise is sharp. Also,
If the membrane 20 is too thin, the strength is greatly reduced, and the scatterer 30a cannot be supported by the membrane 20. That is, in order to suppress the absorption and scattering of the charged particle beam in the membrane and to prevent the transfer accuracy from deteriorating due to the pattern distortion due to the temperature rise of the membrane,
A structure that reduces the thickness of the membrane and that thermally and strongly holds the thinned membrane is required.

Therefore, as a scattering transmission mask or a stencil mask for a charged particle beam, “a large number of small areas each having a pattern to be transferred to a sensitive substrate on a membrane are divided by a boundary area where the pattern does not exist, A mask (a mask having a structure for thermally and strength-holding a membrane) in which a support is provided at a portion corresponding to the boundary region is used.

For example, as a scattering transmission mask for an electron beam reduction transfer apparatus, a large number of small areas 100a each having a pattern to be transferred to a sensitive substrate (eg, a wafer coated with a resist) are boundary areas (no pattern exists). A region in which a lattice-shaped support X is provided at a portion corresponding to the boundary region is used, and a scattering transmission mask 100 as an example is shown in FIG.
6 is shown.

The mask 100 shown in FIGS. 5 and 6 has an electron beam scatterer 30a formed in each of the plurality of small areas 100a on the upper surface of the membrane 20 through which the electron beam is transmitted. A lattice-shaped support X is provided at a portion corresponding to the lattice-shaped boundary region 100b. FIG. 5A is a plan view of a pattern (an example) formed in the small region 100a of the mask, and FIG. 5 is a layout diagram between the small region 100a of the mask and the sensitive substrate 110.
(B) shows each of them.

When a small area 100a of the mask is irradiated with an electron beam at a time, as shown in FIG.
0 (only the electron beam transmission region 20a of the membrane) passes through the scatterer 30 rather than the electron beam EB1 (shown by a solid line in the figure).
The degree of forward scattering of the electron beam EB2 (shown by a two-dot chain line in the figure) that also passed through a becomes large. In the electron beam reduction transfer device (one example), the original pattern of the mask 100 is transferred to the sensitive substrate 11.
A pair of projection lenses 5 and 6 for reducing and projecting to 0 are provided, and a scattering aperture 7 having a through hole 7a for passing only an electron beam passing near the crossover CO by the projection lens 5 is provided.

The above-described electron beam transmitting region 20 of the membrane
a and the scattering of the electron beam in the scatterer 30a,
The electron beam E that has also passed through the scatterer 30a rather than the electron beam EB1 that has passed only through the membrane 20 (electron beam transmission region 20a)
B2 has a higher ratio of being blocked by the scattering aperture 7. Therefore, the resist on the sensitive substrate 110 is exposed to the electron beam transmitting region 20a (FIG. 5) which is a gap between the scatterers 30a.
Exposure is performed in a pattern corresponding to (a blank portion in FIG. 3A).

In some cases, the resist may be exposed corresponding to the pattern of the scatterer 30a by providing a through hole around the center of the scattering aperture 7 and providing a through hole therearound. Further, an electron beam absorber may be provided instead of the scatterer 30a. FIG. 6 shows that “a large number of small regions each having a pattern to be transferred to the sensitive substrate on the membrane are divided by a boundary region where the pattern does not exist, and a grid-like column is provided at a portion corresponding to the boundary region. Scattering mask 1 which is an example of a mask (a mask having a structure for thermally and strongly retaining a membrane)
00, the pattern of each small area 100a is
10 is a schematic view showing the state of transfer (the transfer is performed similarly when a scattering stencil mask is used).

Each of the small areas 100a is one of the sensitive substrates 110.
A pattern to be transferred to an area 110a for a chip (semiconductor of one chip) is provided with each divided partial pattern, and each divided partial pattern is transferred to the sensitive substrate 110 (divided transfer). The external shape of the sensitive substrate 110 is as shown in FIG. 6B, and FIG. 6B shows a part of the sensitive substrate 110 (Va portion in FIG. 6B) in an enlarged manner.

In FIG. 6, the z-axis is taken in parallel with the optical axis AX of the electron beam optical system, and the x-axis and y-axis are taken in parallel with the arrangement direction of the small areas 100a. The electron beam is scanned stepwise in the y-axis direction while continuously moving the mask 100 and the sensitive substrate 110 in the x-axis direction as indicated by arrows Fm and Fw, and the pattern of the small region 100a in a row is formed. Are sequentially transferred, and after the pattern transfer of that row is completed, the next row of the small area 100a adjacent in the x-axis direction is scanned with an electron beam, and thereafter, the transfer (divided transfer) is similarly performed for each small area 100a. repeat.

The transfer order of the scanning order and the sensitive substrate 110 small regions 100a in this case, each of the arrows A _m, A
_As shown by _w . The reason why the continuous movement direction of the mask 100 and the sensitive substrate 110 is opposite is that the x-axis and the y-axis directions of the mask 100 and the sensitive substrate 110 are reversed by the pair of projection lenses.

The ratio of the continuous moving speed of the mask 100 and the sensitive substrate 110 is made to match the pattern reduction ratio from the mask 100 to the sensitive substrate 110. For example, if the pattern reduction rate is 1
In the case of / 4, the moving speed of the mask 100 is
Set to 4 times the moving speed of 0. With such a setting, the positional relationship between the mask 100 and the sensitive substrate 110 in the x-axis direction is kept constant. CO is a crossover by the optical system.

When performing the transfer (split transfer) in the above procedure,
Simply projecting the pattern of the small area 100a in a row in the y-axis direction onto the sensitive substrate 110 with a pair of projection lenses as it is, the sensitive substrate 110 corresponding to each small area 100a.
Between the transfer regions 110b of the
A gap corresponding to b is generated. Therefore, each small area 100
It is necessary to correct the pattern transfer position by deflecting the electron beam EB that has passed through a in the y-axis direction by an amount corresponding to the width Ly of the boundary region 100b.

In the x-axis direction, in addition to moving the mask 100 and the sensitive substrate 110 at a constant speed corresponding to the pattern reduction ratio, the transfer of the small area 100a in one row is completed and the small area 100a in the next row is completed. At the time of transfer, it is necessary to deflect the electron beam EB in the x-axis direction by the width Lx of the boundary region 100b and correct the pattern transfer position so that no gap in the x-axis direction is generated between the transferred regions 110b. There is.

Next, among masks used in the reduction transfer apparatus, for example, a mask shown in FIG. 7 (one example) is known as an X-ray transfer mask. This mask 11 is X
A support 13 is provided on one side of a mask substrate (membrane) 12 which is thinned so as to transmit rays, and an X-ray absorber 14 and a shielding film 15 are provided on the other side. The support columns 13 are provided as a structure for thermally and strength-holding the membrane, whereby the mask substrate (membrane) is provided.
12 is divided into a plurality of rectangular small areas 12a.

In pattern transfer using such an X-ray transfer mask 11, X-rays are scanned stepwise with respect to each small area 12a, and a pattern corresponding to the arrangement of the absorber 14 in each small area 12a is formed. Sensitive substrate 17 using an optical system (not shown)
Are sequentially reduced and transferred. According to the transfer method using a mask as described above, the thinned mask substrate (membrane) is firmly supported by the columns, so that deflection or thermal distortion of the mask substrate due to charged particle beam irradiation or X-ray irradiation can be reduced. Can be suppressed.

[0028]

The pillars of the mask can be formed, for example, by selective etching using a single crystal silicon (silicon wafer) etchant. Here, the pillars 13 and X are formed by etching using an etching rate difference between the (100) plane and the (111) plane of single crystal silicon with respect to an etching solution.
Therefore, the support 13, the wall surfaces 13a, Xa of X are (11
1) The surface is a wall surface inclined by 54.7 degrees with respect to the plane ((100) plane) of the mask substrates (membrane) 12 and 20 (see FIGS. 5 and 7).

When the X-ray transfer mask 11 is irradiated with divergent X-rays, the inclination of the wall surface 13a does not cause any particular problem because the inclination of the wall surface 13a functions as an escape for the divergence of the X-ray. However, the irradiation of the mask is not limited to divergent X-rays, but there are also charged particle beams and synchrotron radiation X-rays with a very small opening angle. As long as it is incident on
There is no need to escape the support wall.

Rather, the inclination of the wall surfaces of the columns 13 and X causes an increase in the width of the columns 13 and X as compared with the case where the wall surface is formed perpendicular to the surfaces of the substrates (membrane) 12 and 20. Causes the problem that the proportion of the boundary area not involved in the pattern formation increases and the size of the masks 11 and 100 increases. As the size of the mask increases, the field of the irradiation optical system for the mask expands,
There are adverse effects such as an increase in the movable range of the mask stage.

In the case of performing the division transfer in a charged particle beam reduction transfer apparatus or the like, in order to join patterns of each small area of the mask on the sensitive substrate, a charged particle beam or the like that has passed through the mask is used as a support. An operation of deflecting by an amount corresponding to the width is required. For this reason, when the width of the columns 13 and X increases, the amount of deflection also needs to be increased, which causes a problem that deflection distortion increases and transfer accuracy deteriorates.

Further, the conventional membrane material (amorphous material or metal) constituting a mask (scattering transmission mask or scattering stencil mask) has a low electron beam transmittance, so that a predetermined transmittance cannot be obtained unless the thickness is considerably reduced. There was a problem that it could not be obtained. The amorphous material (for example, Si ₃ N ₄ , Al ₂ O ₃ , SiO ₂ ) has a problem that the thermal conductivity is poor and the membrane is easily thermally deformed (particularly when the thickness is small). .

Further, when the thickness of the conventional scatterer material (eg, gold or tungsten) constituting the scattered transmission mask is reduced to such an extent that high-precision processing (pattern formation) is possible, energy loss of electron beams is reduced. Therefore, there is a problem that the temperature rise due to the temperature increase becomes a cause of thermal deformation (pattern distortion) of the membrane. The present invention has been made in view of such a problem, and has a mask in which the proportion occupied by a boundary region not involved in pattern formation has been reduced (suppression of enlargement), a mask having a good thermal conductivity of a membrane, a scatterer, In the absorber,
Provided is a method for manufacturing a charged particle beam transfer mask or an X-ray transfer mask that can manufacture at least one of a mask having a small temperature rise due to energy loss of a charged particle beam or X-ray. The purpose is to:

[0034]

Therefore, the present invention firstly provides a mask in which a large number of small areas each having a pattern to be transferred to a target on a membrane are divided by a boundary area where the pattern does not exist. A method of manufacturing a mask for charged particle beam transfer or X-ray transfer provided with a support at a portion corresponding to the boundary region,
a step of preparing a member having a three-stage structure of i-layer / intermediate-layer / Si substrate; and forming the pattern to be transferred at each set location of the small area on the Si layer, whereby Providing a small region and a boundary region; forming a resist pattern layer having an opening corresponding to each of the small regions on the Si substrate; and using the resist pattern layer as an etching mask, By performing inductively coupled plasma etching, side wall protection plasma etching, or cryogenic reactive ion etching on a plurality of pillars having wall surfaces perpendicular or substantially perpendicular to the Si layer and / or the intermediate layer, the pillars correspond to the boundary region. Forming an opening between the columns corresponding to each of the small regions, and forming the opening in the opening. Removing the exposed intermediate layer by etching to expose the membrane of the Si layer in each opening, thereby producing a charged particle beam transfer mask or an X-ray transfer mask. (Claim 1).

The present invention also provides a mask that is divided into a plurality of small areas each having a pattern to be transferred to a target on a membrane and separated by a boundary area where the pattern does not exist. A method of manufacturing a mask for charged particle beam transfer or X-ray transfer provided with a support at a portion to be formed, a step of preparing a member having a three-stage structure of Si layer / intermediate layer / Si substrate;
Forming a resist pattern layer having an opening corresponding to each set point of the small area on the substrate, inductively coupled plasma etching on the Si substrate using the resist pattern layer as an etching mask, sidewall protection plasma; By performing etching or cryogenic low-temperature reactive ion etching, a plurality of pillars having wall surfaces orthogonal or substantially orthogonal to the Si layer and / or the intermediate layer are formed corresponding to the set locations of the boundary region,
A step of forming openings between the columns corresponding to each set point of the small area, and removing the intermediate layer exposed in the openings by etching, and forming the S in each opening.
exposing the membrane of the i-layer, and providing the large number of small regions and boundary regions by forming the pattern to be transferred at each set location of the small region on the membrane. A method for manufacturing a charged particle beam transfer mask or an X-ray transfer mask (claim 2). "

The third aspect of the present invention is a method according to claim 1 or 2, wherein the intermediate layer is an etching stopper and is formed of a metal or an alloy. provide. Further, the present invention fourthly provides a mask in which a large number of small regions each having a pattern to be transferred to a target on a membrane are divided by a boundary region where the pattern does not exist, and a portion corresponding to the boundary region is provided. In a method for manufacturing a charged particle beam transfer or X-ray transfer mask provided with columns, a Si single crystal having a (110) plane on both surfaces and an n-type (111) plane on an orientation flat is provided. A step of preparing a wafer and, on one surface of the Si wafer,
Forming a boron-doped Si single crystal layer, and forming the pattern to be transferred at each set location of the small region on the boron-doped Si single crystal layer, thereby forming the plurality of small regions and the boundary region. Providing an etching mask layer having an opening corresponding to each of the small regions on the other surface of the Si wafer; and forming the etching mask layer on the other surface exposed at the opening of the etching mask layer. By performing anisotropic etching, a membrane made of a boron-doped Si single crystal layer is formed corresponding to each of the small regions, and a plurality of pillars made of Si single crystal and having wall surfaces orthogonal or substantially orthogonal to the membrane are formed. Forming a mask corresponding to the boundary region. Other provides a method (claim 4) "to produce a mask for X-ray transfer.

Also, the present invention is a fifth aspect of the present invention which is a mask in which a large number of small areas each having a pattern to be transferred to a target on a membrane are divided by a boundary area where the pattern does not exist. In a method of manufacturing a charged particle beam transfer or X-ray transfer mask in which a support is provided at a portion where
10) A step of preparing a Si wafer having an n-type (111) plane in an orientation flat with a plane, and forming a boron-doped S on one surface of the Si wafer.
forming an i-single-crystal layer, and forming, on the other surface of the Si wafer, an etching mask layer each having an opening corresponding to each set point of the small area;
By performing anisotropic etching on the other surface exposed at the opening of the etching mask layer, a membrane made of a boron-doped Si single crystal layer is formed corresponding to each set location of the small region, and Si Forming a plurality of pillars made of a single crystal and having a wall surface orthogonal or substantially orthogonal to the membrane in correspondence with the setting location of the boundary area, and transferring the support to the setting location of the small area on the membrane; And a method of manufacturing a charged particle beam transfer mask or an X-ray transfer mask (Claim 5).

The present invention also provides a sixth aspect of the present invention, wherein the pattern to be transferred is formed by forming a scatterer or absorber of a charged particle beam or an absorber of an X-ray at each set location in the small area. A method for manufacturing a scattered transmission mask for charged particle beam transfer or a mask for X-ray transfer according to claims 1 to 5 (claim 6). " Also,
Seventh, the present invention provides a method of manufacturing the scatterer or absorber as atomic number 1.
A manufacturing method according to claim 6 (claim 7), wherein the manufacturing method is formed by using metal elements 4 to 47.

In the eighth aspect of the present invention, the pattern to be transferred is formed by forming a through-hole pattern at each set location of the small area. Method for manufacturing a stencil mask for charged particle beam transfer (Claim 8) "is provided.

[0040]

BEST MODE FOR CARRYING OUT THE INVENTION A large number of small areas each having a pattern to be transferred to a target on a membrane are masks divided by a boundary area where the pattern does not exist, and a support is provided at a portion corresponding to the boundary area. According to the present invention, which is a method for manufacturing the provided charged particle beam transfer mask or X-ray transfer mask, it is possible to form a support having a wall surface perpendicular or substantially perpendicular to the membrane.

Therefore, according to the present invention, it is possible to manufacture a mask in which the ratio occupied by the boundary region not involved in the pattern formation is reduced (the size is suppressed) (claims 1 to 8). That is, according to the present invention, since the width of the column is smaller than that in the case where the wall surface of the column is formed to be inclined with respect to the membrane, the ratio of the boundary region not involved in pattern formation on the mask is also smaller. In addition, the size of the mask can be suppressed.

Therefore, there is no adverse effect such as an increase in the field of the optical system for irradiating the mask and an increase in the movable range of the mask stage, which accompany the enlargement of the mask. Further, there is no adverse effect that transfer accuracy is deteriorated due to an increase in deflection distortion due to an increase in the size of the mask when performing the division transfer. According to the present invention, the ratio of the boundary area not involved in pattern formation on the mask is reduced.
A mask for transferring a chip pattern with a higher degree of integration onto a wafer can be manufactured.

According to the present invention, the support having a wall surface perpendicular or substantially perpendicular to the membrane is made of Si layer / intermediate layer / Si
Inductively coupled plasma etching, sidewall protection plasma etching or cryogenic reactive ion etching on a Si substrate which is a component of a member having a three-stage structure of the substrate, or having a (110) plane of Si single crystal, It can be formed by performing anisotropic etching on the surface ((110) plane) of a Si wafer having an n-type (111) plane in an orientation flat.

The intermediate layer is used as an etching stopper for inductively coupled plasma etching, side wall protection plasma etching or cryogenic reactive ion etching. In the anisotropic etching, a boron-doped Si single crystal layer serves as an etching stopper. The intermediate layer can be made of a metal or an alloy (claim 3). When a mask is manufactured by forming the intermediate layer of a metal or an alloy, heat conduction is improved as compared with a mask manufactured by a conventional SOI wafer. The charge in the SiO ₂ layer of the SOI mask when using a charged particle beam is used. There is an effect that the problem of up can be avoided.

According to the present invention, there is provided a mask in which a plurality of small areas each having a pattern to be transferred to a target on a membrane that transmits a charged particle beam or an X-ray are divided by a boundary area where the pattern does not exist, A method for manufacturing a charged particle beam transfer mask or an X-ray transfer mask in which a scatterer or absorber of a charged particle beam or an X-ray absorber is formed in each of the plurality of small regions (claim 6). Applied.

The present invention also provides a stencil mask for charged particle beam transfer in which a plurality of small areas each having a through-hole pattern to be transferred to a target on a membrane are separated by a boundary area where the through-hole pattern does not exist. (Claim 8). The scatterer or absorber of the mask according to the present invention is preferably made of a metal element having an atomic number of 14 to 47 (claim 7).

According to the manufacturing method according to the seventh aspect of the present invention, it is possible to manufacture a mask in which the temperature rise in the scatterer or the absorber due to the energy loss of the charged particle beam or the X-ray is small. Therefore, according to the manufacturing method of the seventh aspect, the thickness of the scatterer or the absorber is set to 20.
A scatterer of about 0 nm (in the case of Ag having an atomic number of 47) to about 1 μm (in the case of Ti having an atomic number of 14), which is not so thin that a temperature rise is a problem and not so hot that high-precision processing becomes difficult Or it can be an absorber.

The membrane according to the present invention can be constituted by a Si layer or a B-doped silicon single crystal layer formed on a (110) silicon single crystal plane.
In the case of using the B-doped silicon single crystal layer, it can also be used as an etching stopper when forming a membrane by anisotropic etching. Further, when the membrane is composed of the B-doped silicon single crystal layer, the thermal conductivity becomes better than that of a conventional membrane material (a mask having a good thermal conductivity of the membrane can be manufactured).

As described above, according to the present invention (claims 1 to 8), a mask in which the proportion occupied by a boundary region not involved in pattern formation is reduced (upsizing is suppressed) and a mask having a good thermal conductivity of a membrane And / or a mask in which the temperature rise due to the energy loss of the charged particle beam or the X-ray in the scatterer or the absorber is small. Hereinafter, the present invention will be described in more detail with reference to examples, but the present invention is not limited to these examples.

[0050]

[Embodiment 1] A mask manufacturing method according to the present embodiment will be described by the following steps (see FIGS. 1 and 2). The mask manufactured according to the present embodiment is a mask in which a large number of small regions each having a pattern to be transferred to a target on a membrane are divided by a boundary region where the pattern does not exist, and corresponds to the boundary region. This is a scattering transmission mask for charged particle beam transfer, in which a support is provided at a portion to be transferred.

Step 1. On the two surfaces F, a Si single crystal (11
The Si wafer 1 having the (0) plane and the n-type (111) plane in the orientation flat OF is prepared. Step 2. On one surface of the Si wafer 1, a B (boron) -doped Si single crystal layer 2 'is used as an epitaxial layer.
Alternatively, the heat diffusion layer is formed to a thickness of 50 nm. Step 3. Si ₃ N _{4 on} the back surface (another surface) of the Si wafer
The layer and the SiO ₂ layer are each formed to a thickness of 1 μm and laminated, and a photoresist is applied thereon.

Step 4. Add orient to the photoresist
2mm x 11mm length parallel to the AT flat
Photo by forming two rows of square pattern at 2.3mm pitch
A resist pattern is used. Step 5. Etching the photoresist pattern
And the SiO_Two Etch layers and etch further
SiO_TwoUsing the layer as an etching mask,_Three
N_FourEtch the layer.

Step 6. The (110) plane F of the Si single crystal is anisotropically etched in a KOH solution using the etched Si ₃ N ₄ layer E as an etching mask, thereby forming a membrane (B-doped Si single crystal epitaxial layer or thermal diffusion layer). 2) and support (made of Si single crystal) H
(See FIGS. 1A and 1B).

The anisotropic etching is performed on (1) of the Si wafer.
10) A positive voltage is applied to the surface F, and a negative voltage is applied to the Ti electrode while a current is flowing. When the anisotropic etching of the (110) plane F of the Si single crystal progresses and reaches the P-type epitaxial layer or thermal diffusion layer (B-doped Si single crystal layer) 2 ′, a voltage is applied to this layer 2 ′. Is not applied, so that the etching does not proceed.

At this time, since the current stops flowing, it is understood that the etching has been completed.

The wall surface H ′ of the support H is made of Si single crystal (11
1) It becomes a plane. Step 7. A 200 nm thick Cr layer (scattering layer or absorption layer) 3 is formed on the membrane 2 by sputtering, a resist is applied thereon, and a resist pattern is formed by EB lithography. Here, if the Cr layer 3 is coated on the entire surface of the membrane 2, a large stress may be generated and the membrane 2 may be destroyed. Therefore, there is no necessity for the coating on the membrane facing the support H or on the membrane around the mask. The portion N is masked to prevent Cr from being sputtered.

Step 8. Using the resist pattern as an etching mask, the Cr layer 3 is patterned by RIE etching to form a patterned Cr layer 3a and a charged particle beam transmitting region 2a (formation of the small region 10a and the boundary region 10b). Here, the pattern of the Cr layer 3a and the charged particle beam transmission region 2a perpendicular to the membrane 2 is formed by RIE etching.

By the method of the present embodiment having the above steps, a scattered transmission mask (see FIG. 2) for reduced transfer of a charged particle beam is manufactured. Step 7 and step 8 are performed in step 2
And step 3 may be performed. FIG. 1 (c)
FIG. 4 is a cross-sectional view of a main part of the charged particle beam reduction transfer mask manufactured by the method of the present embodiment.

The column H has a width (short side, x direction in FIG. 2) b of 3
00 μm, length (long side, y direction in FIG. 2) L is 11 mm,
The height h (z direction in FIG. 2) is 2 mm, and the arrangement pitch p of the columns.
Is 2.3 mm. The wall surface H ′ of the support H is the (111) plane of the Si single crystal, and is a wall perpendicular to the membrane 2 ((110) plane). In the scattering transmission mask for charged particle beam reduction transfer manufactured by the method of the present embodiment,
The charged particle beam transmittance and the thermal conductivity of the membrane (B-doped Si single crystal layer) 2 are improved as compared with the case where the conventional membrane material is used, and the thickness of the membrane 2 is set to a relatively large value of 50 nm. The desired transmittance is obtained.

As a result, the thickness of the membrane 2 can be made larger than that of the conventional membrane material (about 10 nm), and the heat-resistant deformation resistance and strength can be increased. In the scattering transmission mask manufactured by the method of the present embodiment for transfer of reduced charged particle beam, the scatterer 3a or the absorber 3
a is constituted by Cr of atomic number 24 contained in the metal elements of atomic numbers 14 to 47.

Therefore, the thickness of the scatterer 3a or absorber 3a is set to 200 nm (in the case of Ag having an atomic number of 47) to 1 μm.
m (in the case of Ti of atomic number 14) of 200 nm, which is not so thin that the temperature rise becomes a problem (the temperature rise due to the energy loss of the charged particle beam is small), and high-precision processing is difficult It can be a scatterer or absorber that is not so hot.

In this embodiment, a method for manufacturing a scattered transmission mask for charged particle beam reduction transfer has been described. However, an X-ray reduction transfer mask can be manufactured by the same process. According to the present embodiment, a mask in which the ratio occupied by the boundary region not involved in the pattern formation is reduced (the size is reduced),
It is possible to manufacture a mask having a good charged particle beam transmittance and thermal conductivity of a membrane, and a mask having a small temperature rise due to energy loss of a charged particle beam in a scatterer or an absorber.

[0063]

[Embodiment 2] The mask manufacturing method of the present embodiment will be described by the following steps (see FIG. 3). The mask manufactured according to the present embodiment is a mask in which a large number of small regions each having a pattern to be transferred to a target on a membrane are divided by a boundary region where the pattern does not exist, and corresponds to the boundary region. This is a charged particle beam transfer mask in which a support is provided at a portion to be charged.

Step 1. Si layer (10 μm thickness) 33 / intermediate layer (300 nm thick Au layer) 32 / Si substrate (50 thickness)
A member having a three-stage structure of (0 μm) 31 is prepared (FIG. 3).
(A)). Step 2. On the Si substrate 31, a resist pattern layer R having an opening corresponding to each set portion of the small area is formed (FIG. 3B).

Step 3. Inductively coupled plasma etching (using a mixed gas of SF ₆ / CHF ₃ ) is performed on the Si substrate 31 using the resist pattern layer R as an etching mask, so that the Si layer 33 and the intermediate layer 32 are orthogonal or substantially orthogonal to the wall surface. Are formed so as to correspond to each set point of the boundary area, and an opening (1 mm × 1) between the columns corresponding to each set point of the small area is formed.
mm).

Here, since the selectivity between the resist layer R and the Si substrate 31 is 20 or more, it is possible to form the pillar Hs having the wall surface Hs ′ perpendicular or substantially perpendicular to the Si layer 33 and the intermediate layer 32 ( FIG. 3 (c)). Step 4. Intermediate layer (Au layer) exposed at the opening
32 and the resist layer R are removed by dry etching,
The membrane 33 'of the Si layer 33 is exposed at each opening (FIG. 3D).

Step 5. The pattern to be transferred is formed on each of the set locations of the small area on the membrane 33 '(FIG. 3E). Here, when manufacturing a scattering transmission mask, the pattern to be transferred is formed by forming a scatterer or absorber (Cr) of a charged particle beam at each set location of the small area (implementation). (Similar to steps 7 and 8 of Example 1).

In the case of manufacturing a scattering stencil mask, a through-hole pattern is formed at each set location in the small area by using a fine processing process (for example, ECR plasma etching or magnetron plasma etching). The pattern to be transferred is formed. By the method of the present embodiment including the above steps, a charged particle beam reduction transfer mask is manufactured.

The manufactured mask has a column width b ′ of 10
0 μm, length 1 mm, height h ′ is 500 μm, and arrangement pitch p ′ of the pillars is 1.1 mm. In the charged particle beam reduction transfer mask manufactured by the method of the present embodiment,
The scatterer or absorber 3a 'is made of Cr having an atomic number of 24 included in metal elements having atomic numbers of 14 to 47.

Therefore, the thickness of the scatterer or absorber 3a 'is set to 200 nm (in the case of Ag having an atomic number of 47) to 1 μm.
(In the case of Ti of atomic number 14)
A scatterer or absorber having a thickness of 00 nm, which is not so thin that the temperature rise becomes a problem (the temperature rise due to the energy loss of the charged particle beam is small) and not so hot that high-precision processing becomes difficult. .

In the present embodiment, the pattern to be transferred is formed in the last step, but after preparing a member having a three-stage structure of Si layer / intermediate layer / Si substrate, the small area on the Si layer is prepared. The pattern to be transferred may be formed at the set position. In this embodiment, a method for manufacturing a charged particle beam reduced transfer mask has been described. However, an X-ray reduced transfer mask can be manufactured by the same process.

According to the present embodiment, a mask in which the proportion occupied by a boundary region not involved in pattern formation is reduced (larger size is suppressed), and a mask in which a temperature rise due to energy loss of a charged particle beam in a scatterer or an absorber is small. , Can be manufactured.

[0073]

[Embodiment 3] A mask manufacturing method of the present embodiment will be described by the following steps. The mask manufactured according to the present embodiment is a mask in which a large number of small regions each having a pattern to be transferred to a target on a membrane are divided by a boundary region where the pattern does not exist, and corresponds to the boundary region. This is a charged particle beam transfer mask in which a support is provided at a portion to be charged.

Step 1. Si layer (2 μm thickness) / Intermediate layer (Al layer with a thickness of 100 nm) / Si substrate (1 mm thickness)
A member having a three-stage structure is prepared. Step 2. On the Si substrate, a resist pattern layer having an opening corresponding to each set location of the small area is formed. Step 3. Using the resist pattern layer as an etching mask, the Si substrate is subjected to cryogenic low-temperature reactive ion etching (substrate temperature -130 ° C., using 10 mTorr of SF ₆ gas), so as to be orthogonal or substantially perpendicular to the Si layer and the intermediate layer. A plurality of struts having orthogonal wall surfaces are formed so as to correspond to the respective setting locations in the boundary area, and openings (1 mm) between the struts corresponding to the respective setting locations in the small area are provided.
× 1 mm).

Step 4. The intermediate layer (Al layer) exposed in the openings is removed by dry etching to expose the membrane of the Si layer in each opening. Step 5. The pattern to be transferred is formed in each of the small areas on the membrane. Here, when manufacturing a scattering transmission mask, a scatterer or absorber (C
By forming r), the pattern to be transferred is formed (similar to steps 7 and 8 in Example 1).

In the case of manufacturing a scattering stencil mask, a through-hole pattern is formed at each set location in the small area by using a fine processing process (for example, ECR plasma etching or magnetron plasma etching). The pattern to be transferred is formed. By the method of the present embodiment including the above steps, a charged particle beam reduction transfer mask is manufactured.

The manufactured mask has a column width of 300 μm.
m, length 1mm, height 1mm, arrangement pitch of pillars
1.3 mm. In the charged particle beam reduction transfer mask manufactured by the method of the present embodiment, the scatterer or the absorber is made of Cr having the atomic number 24 included in the metal elements having the atomic numbers 14 to 47.

Therefore, the thickness of the scatterer or absorber is set to 2
200n in the range of about 00 nm (for Ag with atomic number 47) to about 1 μm (for Ti with atomic number 14)
m, a scatterer or absorber that is not so thin that the temperature rise is problematic (the temperature rise due to the energy loss of the charged particle beam is small) and not so hot that high-precision processing becomes difficult. .

In this embodiment, the pattern to be transferred is formed in the last step. However, after preparing a member having a three-stage structure of Si layer / intermediate layer / Si substrate, the small area on the Si layer is prepared. The pattern to be transferred may be formed at the set position. In this embodiment, a method for manufacturing a charged particle beam reduced transfer mask has been described. However, an X-ray reduced transfer mask can be manufactured by the same process.

According to the present embodiment, a mask in which the proportion occupied by the boundary region not involved in the pattern formation is reduced (larger size is suppressed), and a mask in which the temperature rise due to the energy loss of the charged particle beam in the scatterer or absorber is small. , Can be manufactured.

[0081]

As described above, according to the present invention,
Temperature rise due to energy loss of charged particle beam or X-ray in a mask in which the proportion occupied by a boundary region not involved in pattern formation has been reduced (the size has been suppressed), a mask having a good thermal conductivity of a membrane, or a scatterer or an absorber. And at least one of the smaller masks can be manufactured.

[Brief description of the drawings]

FIG. 1 is a schematic side view showing a part of a process of manufacturing a charged particle beam reduction transfer mask of Example 1.

FIG. 2 is a schematic perspective view showing a charged particle beam reduction transfer mask of Example 1.

FIG. 3 is a schematic side view showing a part of a process of manufacturing a charged particle beam reduction transfer mask of Example 2.

FIG. 4 is a view schematically showing a stencil mask 21 of a general electron beam reduction transfer mask and a transfer principle using the stencil mask 21.

FIG. 5 shows a wall surface Xa inclined with respect to the membrane 20.
Conventional scattering transmission mask 100 provided with columns X (lattice-like) having
3A is a plan view illustrating a small region 100a of the mask, and FIG. 3B is a diagram illustrating the layout.

FIG. 6 is a diagram schematically showing a state in which a pattern of each small region 100a is transferred to a sensitive substrate 110 using the scattering transmission mask 100 of FIG.

FIG. 7 shows a wall 13 inclined with respect to the membrane 12.
FIG. 1A is a side view showing a conventional X-ray transfer mask provided with a support 13 having a, and FIG. 2B is a diagram schematically showing divided transfer of a pattern using the X-ray transfer mask.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Si wafer 2 ... Membrane 2a, 2a '... Charged particle beam transmission area 2' ... B-doped Si single crystal layer 3 ... Scatterer or absorber 3a, 3a '... Patterning Scattered body or absorber 5 ... Projection lens 6 ... Projection lens 7 ... Aperture 7a ... Through hole 10 ... Scattering transmission mask for charged particle beam reduction transfer 10a, 10a '... Sensitive Small area 10b, 10b 'with pattern to be transferred to substrate 110 Boundary area 11 X-ray transfer mask 12 Mask substrate (membrane) 12a To be transferred to sensitive substrate 17 Small area with pattern 13 ... Wall surface 13a inclined with respect to membrane
13a ... wall surface of the pillar inclined with respect to the membrane 14 ... patterned X-ray absorber 15 ... X-ray shielding film 17 ... sensitive substrate 20 ... mask substrate (membrane) 20a: electron beam transmission area 21: stencil mask (mask for electron beam reduction transfer) 22: mask substrate 23: through hole 24: projection lens (24a, 24b) 25: sensitive substrate 30: scattering Body or absorber 30a Patterned scatterer or absorber 31 Si substrate 32 Intermediate layer 33 Si layer 33 'Si membrane 100 Scattering transmission mask (mask for electron beam reduction transfer) 100a, a small area having a pattern to be transferred to the sensitive substrate 100b, a boundary area 110, a sensitive substrate 110a, one chip of the sensitive substrate 110 Sensitive substrate 110 corresponding to the (one semiconductor) worth of area 110b · · subregion 100a
A: Island-shaped scatterer E: Patterned Si ₃ N ₄ layer (etching mask) F: (110) crystal plane of Si single crystal H, H _S: Membrane Columns having vertical wall surfaces H ′, H _S ′ ··· Wall surfaces of vertical columns with respect to the membrane N ・ ・ ・ Part where Cr is not sputtered (masking part) OF ・ Orientation flat R ・ ・ ・ Pattern Resist layer X X: Support having a wall inclined with respect to membrane Xa: Wall of support inclined with respect to membrane AX: Optical axis of electron beam optical system b: Width of support H ... height p of ... strut H _S 'width h of ... strut H _S' length b of the arrangement pitch L ... strut H height p ... strut H strut H '· arrangement pitch CO .. more crossover image of ... strut H _S

Claims (8)

[Claims] 1. A mask in which a large number of small areas each having a pattern to be transferred to a target on a membrane are divided by a boundary area where the pattern does not exist.
In a method of manufacturing a charged particle beam transfer or X-ray transfer mask having a column provided at a portion corresponding to the boundary region, a step of preparing a member having a three-stage structure of Si layer / intermediate layer / Si substrate Forming the pattern to be transferred at each set location of the small area on the Si layer to provide the large number of small areas and boundary areas; and forming the small areas on the Si substrate. Forming resist pattern layers each having an opening corresponding to a region; and performing inductively coupled plasma etching, sidewall protection plasma etching, or cryogenic reactive ion etching on the Si substrate using the resist pattern layer as an etching mask. Thereby, a plurality of pillars having wall surfaces perpendicular or substantially perpendicular to the Si layer and / or the intermediate layer are connected to the boundary. Forming the openings corresponding to the respective regions, and forming openings between the columns corresponding to the respective small regions; removing the intermediate layer exposed in the openings by etching; A method for manufacturing a charged particle beam transfer mask or an X-ray transfer mask, comprising: exposing a membrane of a layer. 2. A mask in which a large number of small areas each having a pattern to be transferred to a target on a membrane are divided by a boundary area where the pattern does not exist,
In a method of manufacturing a charged particle beam transfer or X-ray transfer mask having a column provided at a portion corresponding to the boundary region, a step of preparing a member having a three-stage structure of Si layer / intermediate layer / Si substrate Forming a resist pattern layer having an opening corresponding to each of the setting portions of the small area on the Si substrate; and inductively coupled plasma etching on the Si substrate using the resist pattern layer as an etching mask. A plurality of pillars having wall surfaces perpendicular or substantially perpendicular to the Si layer and / or the intermediate layer are formed in correspondence with the set locations of the boundary region by performing sidewall protection plasma etching or cryogenic reactive ion etching. Forming an opening between the columns corresponding to each of the setting locations of the small area; and forming the opening in the opening. Removing the exposed intermediate layer by etching to expose the membrane of the Si layer in each opening; and at each set location of the small region on the membrane,
Providing the plurality of small areas and boundary areas by forming the patterns to be transferred, respectively. A method for manufacturing a charged particle beam transfer mask or an X-ray transfer mask . 3. The method according to claim 1, wherein the intermediate layer is an etching stopper, and is formed of a metal or an alloy. 4. A mask in which a large number of small areas each having a pattern to be transferred to a target on a membrane are divided by a boundary area where the pattern does not exist,
A method for manufacturing a charged particle beam transfer or X-ray transfer mask in which a support is provided at a portion corresponding to the boundary region, wherein a silicon single crystal (110) plane is provided on both surfaces, and the n-type is oriented flat. Si with (111) plane
A step of preparing a wafer; a step of forming a boron-doped Si single crystal layer on one surface of the Si wafer; and the step of transferring to each set location of the small region on the boron-doped Si single crystal layer. Forming a plurality of small regions and boundary regions by forming respective patterns; and forming an etching mask layer having an opening corresponding to each of the small regions on the other surface of the Si wafer. By performing anisotropic etching on the other surface exposed at the opening of the etching mask layer, a membrane made of a boron-doped Si single crystal layer is formed corresponding to each of the small regions, and a Si single crystal is formed. A plurality of struts having wall surfaces perpendicular or substantially perpendicular to the membrane are formed corresponding to the boundary region. And a method for manufacturing a charged particle beam transfer mask or X-ray transfer mask. 5. A mask in which a plurality of small regions each having a pattern to be transferred to a target on a membrane are divided by a boundary region where the pattern does not exist,
A method for manufacturing a charged particle beam transfer or X-ray transfer mask in which a support is provided at a portion corresponding to the boundary region, wherein a silicon single crystal (110) plane is provided on both surfaces, and the n-type is oriented flat. Si with (111) plane
A step of preparing a wafer; a step of forming a boron-doped Si single crystal layer on one surface of the Si wafer; and an opening corresponding to each set location of the small region on the other surface of the Si wafer. Forming an etching mask layer having each, and performing anisotropic etching on the other surface exposed at the opening of the etching mask layer, thereby forming a membrane made of a boron-doped Si single crystal layer into each of the small regions. A step of forming a plurality of struts made of Si single crystal and having a wall surface orthogonal or substantially perpendicular to the membrane and corresponding to the set points of the boundary region, Forming each of the patterns to be transferred at a set position of the small area. Electron particle beam transfer mask or X
A method for manufacturing a line transfer mask. 6. The pattern to be transferred is formed by forming a scatterer or absorber of a charged particle beam or an absorber of an X-ray at each set location of the small area. Item 6. A method for producing a scattered transmission mask for charged particle beam transfer or an X-ray transfer mask according to Item 1 to 5. 7. The scatterer or the absorber having an atomic number of 14
7. The manufacturing method according to claim 6, wherein said metal element is formed of a metal element of any one of (a) to (47). 8. The stencil for charged particle beam transfer according to claim 1, wherein the pattern to be transferred is formed by forming a through-hole pattern at each set location of the small area. A method of manufacturing a mask.