JP2005202170A - Exposure mask and exposure method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To efficiently form a circumferential structure, when forming a three-dimensional structure, using a large tone pattern. <P>SOLUTION: An exposure mask used for an aligner comprises: a first mask region M1 for obtaining transmittance of illumination light, corresponding to a ratio by continuously arranging a plurality of pattern blocks composed of a pair of a light-shielding pattern for shielding illumination light emitted from the aligner and a transmitting pattern for transmitting the illumination light and gradually changing the ratio of the light-shielding pattern to the transmission pattern; and a second mask region M2 composed of an all light-shielding pattern P0 for shielding the illumination light and an all transmitting pattern P100 for transmitting the illumination light and obtaining the light-shielding part and the transmission part, corresponding to the shape of each pattern. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an exposure mask and an exposure method for forming a three-dimensional shape such as an optical lens array by exposure.

  It is used in the manufacture of semiconductors and liquid crystal devices as one of the manufacturing methods for minute optical components such as microlens arrays used in video device application products such as CCD (Charge Coupled Device) and LCD (Liquid Crystal Display). There is a method of applying photolithography technology.

  That is, by giving a desired exposure dose distribution to the photoresist, which is a photosensitive material, the photoresist is processed three-dimensionally, and by using the photoresist as a mask, silicon, a glass substrate, etc. are processed three-dimensionally. ing.

  The photomask used in this lithography process is realized by multiple exposure using a plurality of masks as shown in FIG. An exposure method according to this technique will be described in one dimension with reference to FIG. The final exposure dose distribution is D (X) in FIG.

  First, the exposure amount E [1] is given to the region <1> with the mask (1) in FIG. Next, the exposure amount E [2] is given to the region <2> with the mask (2). At this time, the total exposure amount D1 of the region <1> is E [1] + E [2]. Further, the mask (3), the mask (4),..., The mask (n) (not shown) are sequentially exposed with the exposure amounts E [3], E [4],. The final exposure dose D [i] is D [i] = E [i] + E [i + 1] +... + E [n], and a desired discrete exposure dose distribution is obtained. In this case, the number n of masks corresponds to the exposure position resolution. For example, when n = 10, an exposure step of 10 gradations is obtained.

  Recently, in addition to the multiple exposure method using a plurality of masks, a single exposure using a so-called gray-tone mask having a continuous distribution in the transmittance of the light shielding film as shown in Patent Document 1 can be performed. A method for obtaining a desired exposure dose distribution has also been developed. This conceptual diagram is shown in FIG.

  In the above-described two methods, the former technique based on multiple mask multiple exposure is multiple multiple exposure and multistage exposure in terms of time, so that a stepped shape remains in the obtained integrated exposure amount distribution. Further, the number of exposure gradations to be obtained is the number of masks, that is, the number of exposures, and there is a problem that a practically sufficient number of gradations cannot be obtained because it is about 10 steps. Moreover, the complexity of the exposure process and the mask cost proportional to the number of masks occur, and various problems arise.

  Although the one-time exposure method using the latter gray-tone mask can obtain an almost continuous exposure amount distribution, it is generally very difficult to create such a gray-tone mask, and a special film material, Since a special film forming technique is required, the mask cost becomes very high. Further, the special film material is worried about change with time with respect to heat, and there is also a concern about performance stability (thermal stability) during use.

  As a measure for solving these problems, the inventors of the present invention configure a mask with a binary mask pattern having a resolution limit pitch or less, and change the aperture size or the dot-remaining pattern size to each spatially. A technology that enables three-dimensional structure formation has been proposed (Japanese Patent Application No. 2003-18439).

  In designing such a mask, the 0th-order light intensity of a single two-dimensional array pattern having a certain pitch and a certain opening size is theoretically calculated, and based on this, the opening size for obtaining a predetermined light amount at a predetermined position is determined. A design technique is adopted in which the opening size is changed depending on the location so that a desired resist residual film thickness can be obtained at a predetermined position.

  By the way, in an actual microlens array, not only the lens portion but also a structure (column or the like) for guiding the flow of the injected resin to the lens peripheral portion, an alignment mark for assembly, and the like must be formed. Further, the lens portion is not formed on the basis of the upper surface (surface) of the substrate, but is more often formed with a predetermined depth reference (hereinafter referred to as a lens portion base level) from the substrate surface. Conventionally, when forming these structures, as shown in FIG. 11, by using a plurality of masks to selectively change the exposure area and changing the exposure amount of each exposure mask, desired micro-optics can be obtained. A device structure was formed.

  That is, the lens base portion is first exposed, then the alignment mark portion is exposed, and then the resin base level is exposed. Thereafter, exposure is repeated with a plurality of masks to form a lens portion.

JP 2002-189280 A

  However, even if the lens portion is collectively exposed with a gray tone mask, an exposure process using a plurality of masks eventually results in forming a peripheral structure. For this reason, it leads to complication of the exposure process and an increase in process time, and there arises a problem that the merit of gray tone mask exposure is lowered.

  In addition, there is a difference in the height of the structure other than the lens part in the target structure, and when the lens part is formed to a predetermined depth (offset depth), the predetermined depth is dug in an offset manner with a gray-tone mask having low transmittance. Must be. Therefore, there arises a problem of process processing time that the exposure time increases and a problem of damage to the exposure apparatus due to long exposure.

  The present invention has been made to solve such problems. That is, according to the present invention, in an exposure mask used in an exposure apparatus, a plurality of pattern blocks each composed of a pair of a light-shielding pattern that blocks illumination light emitted from the exposure apparatus and a transmission pattern that transmits illumination light are continuous. A first mask region having a constant pitch of the continuous pattern blocks and a ratio of the light shielding pattern to the transmission pattern gradually changing to obtain a transmittance of illumination light according to the ratio. And a second mask region that includes a total light shielding pattern that shields illumination light and a total transmission pattern that transmits illumination light, and that obtains a light shielding portion and a light transmission portion corresponding to the shape of each pattern.

  In the present invention, first, a mask is constituted by a binary pattern (second mask region) including a 100% transmission pattern for base level exposure and a 100% light shielding pattern in the mask. The base level molding pattern is exposed to give an exposure amount necessary for forming the base portion. Further, a gray tone pattern (first mask region) for forming a three-dimensional shape and a peripheral step is arranged at another location in the same mask, and this gray tone pattern and the above binary pattern are overlaid. By performing the combined exposure, the base level, the three-dimensional shape portion, and the peripheral step portion as a whole can be exposed and formed using a single mask (two mask regions).

  Therefore, according to the present invention as described above, since the base level of the three-dimensional shape portion is formed by exposure in the entire transmission pattern portion, the exposure time can be minimized. Further, since the three-dimensional shape portion and the peripheral step portion are formed by batch exposure using a gray tone pattern, the number of exposure processes can be reduced.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic plan view for explaining an exposure mask (hereinafter simply referred to as “mask”) according to the present embodiment. This mask mainly has a configuration in which a first mask region M1 and a second mask region M2 are disposed adjacent to the same base material.

  The first mask region M1 is a region provided with a gray tone pattern for forming a three-dimensional shape object (lens in this embodiment) by exposure, and includes a pattern block PB that matches the shape of the exposure target. Yes. The pattern block PB will be described later.

  On the other hand, the second mask region M2 is a binary pattern including a total light shielding pattern P0 that shields 100% of the illumination light from the exposure apparatus and a total transmission pattern P100 that transmits 100% of the illumination light. From these patterns, a light shielding part and a light transmitting part corresponding to the shape of the pattern can be obtained.

  Here, the pattern block PB of the first mask region M1 constituting the mask according to the present embodiment will be described based on the schematic diagram of FIG. That is, in the reduction projection exposure apparatus S used for transfer, the pattern surface of the mask M and the surface of the wafer W are in a conjugate (imaging) relationship, and usually the pattern on the lower surface of the mask M is imaged on the surface of the wafer W. Transfer the pattern.

  However, when the exposure wavelength (λ), the constituent mask pattern pitch (P), the numerical aperture (NA) of the exposure apparatus, and the coherence factor (σ) representing the secondary light source size are given, an image is formed on the surface of the wafer W. The minimum pitch (Pmin) that can be obtained is expressed by Equation 1. Here, Pmin represents a dimension on the wafer.

  This equation 1 is interpreted at a level where the diffracted light of the lowest order (± 1) is blocked or not blocked by the NA aperture of the projection lens. For example, when λ = 365 nm, NA = 0.5, and σ = 0.5, Pmin = 487 nm Become.

  That is, since the diffracted light does not reach the surface of the wafer W in the minute pitch pattern of Pmin or less, interference between the diffracted lights, that is, image formation of the mask pattern cannot occur. However, the 0th order light reaches the wafer W. At this time, the 0th-order light intensity decreases as the light-shielding band width increases at the same pitch, and the 0th-order light intensity increases as the pitch increases at a pitch of Pmin or less for the same light-shielding portion dimensions.

Specifically, in a one-dimensional line and space pattern, if the transmission area ratio in the unit repeating pattern is R (≦ 1), the light intensity reaching the wafer surface is normalized by the light intensity on the mask. Then, R ² is obtained. For example, the 0th-order light intensity at 1: 1 line and space is 0.25. Similarly, the 0th-order light intensity in the case of a 1: 1 two-dimensional square hole array is 0.0625.

  The mask M according to the present embodiment is provided with a set of a transmission pattern and a light shielding pattern in succession using this principle, thereby forming a pattern block PB. A basic mask design method related to this is described in a previous application (Japanese Patent Application No. 2003-18439) by the present inventor. FIG. 3 shows a pattern block PB in a mask for forming a concave cylindrical lens array as an example.

  Here, the overall level difference including the lens peripheral part, which is the subject of the present invention, will be described with reference to FIG. The resist to be used is a positive type and is applied to a glass substrate. For simplicity of explanation, the resist residual film shape after exposure and development is transferred at an etching selection ratio of 1: 1. Further, it is assumed as an example that a high refractive index resin is applied on a glass substrate subjected to this shape processing to form a one-dimensional concave cylindrical lens array panel.

  4A is a schematic plan view showing a target shape, FIG. 4B is a cross-sectional view taken along line A-A ′ in FIG. 4A, and FIG. 4B is a cross-sectional view taken along line B-B ′ in FIG. In this example, there is a lens portion (one-dimensional cylindrical lens array) that is a three-dimensional shape in the center, and there are a lens portion base level and a resin layer base level around it. In addition, there are resin injection columns at the four corners of the resin layer base level, and alignment mark portions are provided on both the left and right sides.

The step shape of each part is defined as follows based on the resist surface of the unexposed part as follows.
The base level of the lens part is depth D0.
The level difference of the lens part itself is DL.
The peripheral step portion 1 corresponds to the base level of the resin layer and has a depth D1.
The peripheral step 2 corresponds to the height of the alignment mark and has a depth D2.

  That is, the flat level difference level has three levels D0, D1, and D2 (D1> D2> D0), and a system in which a level difference level DL that continuously changes in the lens portion is added thereto.

  First, in order to simplify the explanation, it is assumed that the resist characteristics are simple as in the prior application (Japanese Patent Application No. 2003-18439) by the present inventor. It is assumed that the residual film characteristic of the resist in this embodiment can be approximated by the following formula 2.

  Here, Z is the depth amount from the unexposed portion (film reduction amount after development), dose is the given exposure amount, and A and B are approximation coefficients of ln (dose) versus Z. In order to simplify the explanation, FIG. 5A shows a structure in which the resin injection columns of the A-A ′ cross section in FIG. 4B and the B-B ′ cross section in FIG. This target shape is Z = f (X).

  The exposure dose distribution (E (x)) required to form the structure of FIG. 5A is expressed by the following equation (Equation 3) using the above-described target shape Z = f (X) and Equation 2. The conceptual diagram is shown in FIG.

  Here, it is assumed that the minimum transmittance portion of the first mask region, which is a gray tone mask, is configured with a mask pattern manufacturing limit. This is defined as a pitch dimension P and a space dimension Smin. Here, P is assumed to satisfy the non-imaging condition of Equation 1. Further, an upper limit exposure amount considering the damage of the exposure apparatus due to long exposure is set to Emax.

  Therefore, the exposure amount (dose (Lp)) given to the lens base level portion when the gray tone pattern is exposed is expressed by Equation 4.

  Further, since the depth D0 is necessary for forming the lens portion, the exposure amount (Ebin) to be given when the exposure is performed with the binary pattern of the second mask region is expressed by Equation 5.

  Further, the portion having the largest exposure amount in the region of FIG. 5C is the resin layer (depth D1). The exposure amount (dose (D1)) necessary for forming this is as shown in Equation 6. Become.

  Here, when the minimum production limit of the space pattern of the gray tone pattern is Smin, and when the maximum production limit of the space pattern calculated from the minimum line size of the mask pattern and the pitch P is Smax, the following formula 7 Satisfy the relationship. Here, p, Smax, and Smin are converted dimensions on the wafer.

  This is a condition for forming the exposure amount distribution of FIG. 5D with a gray tone pattern. P and Emax are initially finely adjusted to satisfy this. Alternatively, the resist material used is changed to one with high contrast.

  As the gray-tone pattern pattern arrangement, the exposure distribution E (x) of Equation 3 described above is converted into a transmittance distribution by Equation 8 below.

  Finally, the space pattern dimension at each position is obtained from Equation 9 below.

  The structure shown in FIG. 5A is obtained by exposing a gray-tone pattern for forming a step structure including the lens portion designed in this way and a binary pattern for obtaining an offset depth with exposure amounts of Emax and Ebin, respectively. Can be formed.

  In the present embodiment, exposure is performed by superimposing the same area using the gray tone pattern, which is the first mask area designed as described above, and the binary pattern, which is the second mask area. You may make it adjust the exposure amount of exposure by a binary pattern according to the exposure amount by exposure by (2). That is, for example, after the lens portion is exposed with the gray tone pattern that is the first mask region, the exposure amount in the second mask region that is the binary pattern is adjusted, and the resin layer base level D1 is set to the second mask region. It is possible to control by the exposure amount of the area. Thereby, for example, the resin injection state is controlled, and the process can be optimized.

  Here, a method of manufacturing a microlens array using the mask according to the present embodiment will be described with reference to the schematic diagram of FIG. First, a photoresist (hereinafter simply referred to as “resist”), which is a photosensitive material, is applied onto a substrate made of, for example, a 6-inch diameter quartz glass wafer. The coating thickness is, for example, about 10 μm (see FIG. 6 (a)).

  Next, i-line is irradiated from a stepper which is one of exposure apparatuses, and the resist is exposed through the mask of this embodiment. At this time, an alignment mark required in a later process is also formed at the same time. After the exposure, the resist is developed so that the three-dimensional shape set by the mask can be transferred to the resist (see FIG. 6B).

  Next, the substrate is dry-etched through this resist. Thereby, the three-dimensional shape of the resist is transferred to the substrate. Thereafter, a resin having a high refractive index is applied to the quartz substrate onto which the three-dimensional shape has been transferred by spin coating or the like. Thus, a lens array having a positive power made of resin corresponding to the three-dimensional shape of the substrate is formed (see FIG. 6C).

  Such a microlens array is applied to an apparatus as shown in FIGS. 7 and 8 is a liquid crystal projector, and includes a TFT (Thin Film Transistor) formed on a quartz substrate and a liquid crystal layer formed on the TFT, and each pixel unit is driven by driving the TFT. This controls the orientation of the liquid crystal layer.

  In the microlens array ML formed by the mask of the present embodiment, each lens L is constituted by a resin layer corresponding to each pixel of such a liquid crystal projector. The microlens array ML and the peripheral structure can be formed by two exposures using the mask of this embodiment, and the mask itself is a binary mask composed of a combination of a light-shielding pattern and a light-transmitting pattern, and is easy to manufacture. Therefore, the microlens array ML applied to the liquid crystal projector can be provided at low cost, and the cost of the liquid crystal projector can be reduced.

  Further, as described above, the shape of the formed lens L can be freely set according to the ratio and arrangement of the light shielding pattern and the light transmitting pattern, and an accurate lens can be set by setting the exposure amount of the mask taking advantage of the development characteristics of the resist. The shape can be reproduced, and a highly accurate lens L can be provided without causing unnecessary seams at the boundaries of the individual lenses L.

  In the microlens array manufacturing method described above, an example is used in which a substrate is processed by dry etching after forming a three-dimensional photoresist shape. However, a mother die is produced by electroforming the photoresist. It is also possible to mass-produce resin or the like by the stamper method. In addition, the lens array formed by the mask of the present embodiment can be applied to a CCD or other liquid crystal display device, a semiconductor laser, a light receiving element, or an optical communication device other than a liquid crystal projector. Even the manufacture of shapes is applicable.

  Further, in this embodiment, for the sake of simplicity, the mask pattern design method has been described using a method using a theoretical value of the aperture ratio. However, this was proposed in a previous application (Japanese Patent Application No. 2003-281489) by the present inventor. As described above, the present invention is also effective in a method using the residual film characteristics obtained by exposing the calibration mask. In this case, the calibration mask is exposed with Emax in the same manner as the actual exposure described above, and the entire wafer is exposed with an exposure amount Ebin in a binary pattern or maskless continuously 100%. The sample wafer is developed to obtain correlation data between the pattern size and the amount of film reduction. You may make it perform the pattern design of a gray tone pattern using this correlation data. In the above embodiment, the lens portion is exposed with the gray tone pattern that is the first mask region, and the peripheral step portion is exposed with the binary pattern that is the second mask region. You may comprise so that the gray tone pattern which exposes a part of peripheral step part other than an area | region lens part may be included.

  The present invention can be applied to a three-dimensional object such as a switch, a relay, or a sensor using MEMS (Micro Electro Mechanical System). Furthermore, the present invention can be applied to formation of an arbitrary shape of the base shape of a substrate in semiconductor manufacturing or the like.

It is a schematic plan view explaining the exposure mask which concerns on this embodiment. It is a schematic diagram explaining a pattern block. It is a schematic diagram which shows the pattern block in the mask for forming a concave cylindrical lens array. It is a schematic diagram explaining the whole level | step difference level. It is a schematic diagram explaining the exposure amount, (a) is a target formation shape, (b) is an exposure amount distribution E (X) for forming the shape of (a), and (c) is based on the second mask region. The exposure amount distribution, (d) is a diagram showing the exposure amount distribution by the first mask region. It is a schematic diagram explaining the manufacturing process of a micro lens array. It is a schematic diagram (the 1) explaining the example applied to a liquid crystal projector. It is a schematic diagram (the 2) explaining the example applied to a liquid crystal projector. It is a schematic diagram (the 1) explaining a prior art example. It is a schematic diagram (the 2) explaining a prior art example. It is a schematic diagram (the 3) explaining a prior art example.

Explanation of symbols

  M1 ... first mask area, M2 ... second mask area, P0 ... total light shielding pattern, P100 ... total transmission pattern

Claims (5)

In an exposure mask used in an exposure apparatus,
A plurality of pattern blocks each composed of a pair of a light-shielding pattern that blocks illumination light emitted from the exposure apparatus and a transmission pattern that transmits the illumination light are continuously arranged, and the pitch of the continuous pattern blocks Is a first mask region that obtains the transmittance of the illumination light according to the ratio by gradually changing the ratio of the light shielding pattern and the transmission pattern,
A light-shielding pattern that shields the illumination light and a transmission pattern that transmits the illumination light, the light-shielding part corresponding to the shape of each pattern, and a second mask region for obtaining a light-transmitting part. Mask for exposure. The exposure mask according to claim 1, wherein the first mask region and the second mask region are provided on the same base material. A plurality of pattern blocks each composed of a pair of a light shielding pattern that blocks illumination light emitted from the exposure apparatus and a transmission pattern that transmits the illumination light are continuously arranged, and the pitch of the continuous pattern blocks is A first mask region that obtains the transmittance of the illumination light according to the ratio by gradually changing the ratio between the light shielding pattern and the transmission pattern, and the light shielding pattern that shields the illumination light and the illumination Using an exposure mask that includes a total transmission pattern that transmits light, and includes a light shielding portion corresponding to the shape of each pattern and a second mask region that obtains the light transmission portion,
An exposure method comprising: performing exposure with the first mask region and exposure with the second mask region by superimposing them on the same region. 4. The exposure method according to claim 3, wherein a base height of the step structure and the lens portion structure formed by the first mask region is adjusted by adjusting an exposure amount in exposure by the second mask region. 5. . The exposure mask according to claim 1, wherein the first mask includes a pattern block corresponding to a target shape to be formed by exposure, and a pattern block for forming a peripheral step portion of the target shape.

Images