JP4019491B2 - Exposure method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention provides a fine resist pattern from light transmitted through a photomask by using an exposure apparatus in a manufacturing process of a highly integrated semiconductor device, etc., a first high-resolution exposure using a phase shift method, and a second and subsequent normal exposure. And an exposure method suitable for forming a high-resolution image on a wafer.
[0002]
[Prior art]
A photomask used in the manufacturing process of a semiconductor device has a structure in which a light shielding film is formed on a glass substrate. In the pattern formation process of a semiconductor element, the pattern of the light shielding film (light shielding pattern) on the photomask is Projection exposure is performed on the photoresist applied to the wafer surface. The light-shielding pattern on the photomask is obtained by converting the designed CAD data into data for the drawing device, and patterning the light-shielding film faithfully based on this, and how accurate this photomask pattern is on the wafer. It is important in the semiconductor photolithography process whether to transfer well.
[0003]
In the photolithography process in the manufacturing process of a fine semiconductor device, a high resolution exceeding the resolution limit determined by the wavelength of light is required because it is necessary to form a pattern near the exposure wavelength. Therefore, in recent years, a phase shift method has been used as a photolithography technique capable of forming a fine pattern having an exposure wavelength or less.
[0004]
The principle of the spatial frequency modulation type phase shift method will be described below.
FIG. 6 is a diagram showing the principle of this phase shift method in comparison with the conventional method, and FIG. 7 is a diagram showing the Fourier spectrum of the phase shift method and the conventional method.
Here, it is assumed that the light intensity transmittance period takes a value near the resolution limit in a one-dimensional periodic pattern with 1 / ν ₀ . Since the light transmission target is a mask pattern close to the resolution limit, considering only the sinusoidal fundamental frequency component, the amplitude of the photomask transmission light can be approximated as follows.
[0005]
[Expression 1]
Conventional mask:
T (x) = | cos2πν ₀ x | (1-1)
Phase shift mask:
T (x) = cos2πν ₀ x (1-2)
[0006]
In the conventional method shown in FIG. 6 (a), the zero-order diffracted light in which the light transmitted through the photomask travels straight along the optical axis and the angle of θ (sin θ = ν ₀ λ) with respect to the optical axis are ± 1 next time. The light is separated from the folded light and enters the projection lens. On the other hand, as shown in FIG. 6B, in the phase shift method, the light transmitted through the phase shift mask is separated into ± first-order diffracted light having an angle θ / 2 with respect to the optical axis, and enters the projection lens. . In either case, only the diffracted light transmitted through the inside of the projection lens contributes to image formation.
The Fourier spectrum formed on the pupil plane of the projection lens is expressed as follows from the Fourier transforms of equations (1-1) and (1-2).
[0007]
[Expression 2]
Conventional mask:
F (ν) = (4 / π) {δ (ν) / 2 + [δ (ν + ν ₀ ) + δ (ν−ν ₀ )] / 2+...} (2-1)
Phase shift mask:
F ([nu) = (1/2) _{[δ (ν + ν 0/2} ) + δ (ν-ν 0/2) ] ... (2-2)
[0008]
As shown in FIG. 7, in the conventional transmission mask without considering the phase, [nu = 0, whereas there is a spectral component ± [nu _0, the phase shift mask, ν = ± ν _0/2 depends only on To do. That is, the basic spectrum obtained from the phase shift mask is located at a half position of the spectrum obtained from the transmission mask. This is equivalent to halving the spatial frequency of the mask pattern. The projection lens of the stepper acts as a low-pass filter that transmits only a spatial frequency component smaller than the natural frequency νc (= NA / λ). Considering the case where the spatial frequency ν _{0 of the} fine pattern to be transferred is νc <ν ₀ ≦ 2νc, in the case of a transmission type mask, the spectral component of ± ν ₀ does not pass, so that the contrast of the image cannot be obtained. On the other hand, to form a pattern image on the image plane because the phase shift mask transmits a basic spectrum ν = ± ν _0/2. This is a feature of the spatial frequency modulation type phase shift mask that has the greatest effect of improving the resolution.
[0009]
As shown in FIG. 6B, the basic layout structure of the spatial frequency modulation type phase shift mask uses two phases of 0 degrees and 180 degrees, and shifters are used so that the mask openings are alternately in opposite phases. Place. As shown in FIG. 6B, there are shifters in which the shifter is laminated with a film different from the light shielding film, and those in which the glass surface is etched to give the function of the shifter.
[0010]
FIG. 8 is a diagram showing the principle of so-called Levenson type phase shift using a negative resist.
In the photomask shown in FIG. 8, the chrome openings provided at the pattern formation locations are alternately arranged with a 180 ° phase shifter (glass etching groove) and without a 0 ° phase groove. Therefore, in this phase shift method, a phase difference of light transmitted through the pattern itself is used, and a negative resist in which the pattern transfer portion remains undissolved in the developer is essential. Practical negative resists are difficult to obtain and the pattern layout itself needs to be a pattern that produces a phase difference. Therefore, simple lines and spaces are effective, but isolated patterns have no effect of phase shift. Therefore, the phase shift method based on the premise that this negative resist is used has the drawbacks that there are many restrictions on the resist material and pattern formation that are the prerequisite.
[0011]
Therefore, in recent years, a plurality of exposures including a phase shift exposure using a mask in which a light transmitted through both sides of a fine pattern to be formed has a phase difference and a normal exposure for removing unnecessary patterns generated thereby are being studied. (Hereinafter referred to as phase shift multiple exposure method). This technique is discussed in the literature of "Automatic Generation of Phase-Shifting Mask Patterns Using Shifter-Edge Lines, MICRO-AND-NANO-ENGINEERING '97" by Sohka et al.
FIG. 9 is a diagram showing the principle of Levenson type phase shift using a positive resist as the phase shift multiple exposure method.
The phase shift mask shown in FIG. 9 (FIG. 9A) has openings provided on both sides of a fine pattern (chrome pattern) with a 180-degree phase shifter (glass etching groove) and no 0-degree phase groove. Are alternately arranged. Since a positive resist is used, the periphery is covered with chrome. In the second exposure (normal exposure), the surrounding wiring portion (not shown) is resolved using the photomask shown in FIG. 9B while shielding the fine pattern.
[0012]
[Problems to be solved by the invention]
In general, in this phase shift multiple exposure method, a latent image pattern (hereinafter referred to as a transfer pattern) having a desired pattern line width is formed in the first high-resolution exposure, and already formed in the second normal exposure. The transferred pattern is covered with a light shielding portion of the mask pattern to prevent further exposure.
However, in practice, this transfer pattern often cannot be completely shielded by the second normal exposure pattern, and there is a problem of leak exposure in which the transfer pattern is exposed to some extent even by the second normal exposure.
[0013]
In such a case, conventionally, the mask pattern is changed so that the light can be completely shielded by the second normal exposure. However, the second normal exposure often serves to resolve the peripheral pattern connected to the fine pattern, and if the fine pattern is covered with a large area, the peripheral pattern cannot be partially resolved. Sometimes. In addition, it is conceivable to change the pattern arrangement itself, but in this case, both the first and second exposure patterns must be changed, resulting in a significant pattern correction.
[0014]
The present invention has been made in view of the above circumstances, and its object is to form a fine pattern by multiple exposure of the first high-resolution exposure using the phase shift method or the like and the second and subsequent normal exposures. It is an object of the present invention to provide an exposure method for resolving a desired fine pattern width while reducing variations.
[0015]
[Means for Solving the Problems]
The present inventor has intensively studied to solve the above-described problems of the prior art. As a result, as a phase shift multiple exposure method in which only the part with strict line width control is formed by phase shift and the other part is exposed with a normal mask, both sides of the fine line width direction of the part with strict line width control are different. In the first high-resolution exposure, in which the fine pattern is transferred onto the photoresist layer by the transmitted light of the phase difference, the latent image pattern of the photoresist is formed to be thicker than before, so that the second and subsequent normal exposure patterns The present inventors have found an exposure condition that can obtain a desired pattern line width without changing the above. In addition, it has been found that such a high-resolution exposure condition change is also effective as a method for extending the process margin.
[0016]
The exposure method of the present invention comprises a high-resolution exposure for transferring a pattern having a severe line width controllability to a photoresist layer using a phase shift pattern, and a photoresist layer portion that has already been pattern-transferred by the high-resolution exposure. The photoresist is subjected to multiple exposures including normal exposure for transferring a pattern having a relatively loose line width controllability to the photoresist layer without using a phase shift pattern while protecting the mask pattern with a light shielding portion. This exposure method forms a latent image pattern. In the high-resolution exposure, an exposure condition (for example, an exposure amount) in which the line width of the portion where the line width controllability is severe in the latent image pattern transferred to the photoresist layer by the high-resolution exposure becomes thicker than a desired line width. ) Is used. In the normal exposure, an end portion that defines the line width is exposed at a portion where the line width controllability of the latent image pattern formed by the high resolution exposure is thicker than a desired line width. The line width is reduced, and a desired line width is obtained at a place where the line width controllability is severe after the normal exposure. Further, in the normal exposure, the Makusu pattern having the light shielding portion of the size that can be exposed by overlapping the photoresist layer portions that are already pattern transfer by the high resolution exposure by light leaking from the light blocking part Use.
[0017]
On the other hand, to thicker than the high-resolution exposure Thus the photoresist layer desired line width the line width of the strict point the line width controllability in the latent image pattern transferred to, even by changing the mask pattern width I can deal with it. The high-resolution exposure in this case, preferably, extracts the line width controllability strict position of the pattern from the already designed element shape pattern, with respect to extracted the pattern was thickened over a resized to at least the line width direction The mask pattern is used.
[0018]
The exposure method of the present invention is an exposure method in which a latent image pattern of a photoresist is formed by overlapping a plurality of exposures, and the first exposure is performed by transferring the latent image transferred to the photoresist layer. the line width of the pattern is exposed as made thicker than desired line width, the exposure of the second and subsequent latent image pattern of the photoresist already formed by extending the light-shielding area than the mask pattern used at the time of previous exposure , And the exposure amount is set to be larger than that in the previous exposure, whereby the line width of the latent image pattern formed by the first exposure is larger than the desired line width. , And the line width is gradually reduced to obtain a desired line width .
[0019]
In such an exposure method of the present invention, for example, in accordance with the pattern size of the light-shielding portion used for normal exposure, the exposure conditions (for example, exposure amount) at the first high-resolution exposure are changed, and the photoresist after the exposure is changed. The line width of the latent image pattern is set to be thick. Then, during the second and subsequent normal exposures, the latent image pattern of the photoresist is further exposed (leak exposure) using light leaking from the light shielding portion, thereby adjusting to a desired line width. In this method, not only the pattern change is unnecessary, but also the focus margin is increased and stable compared to the conventional method in which most of the pattern transfer is completed by increasing the exposure amount at the first high-resolution exposure. Pattern formation is possible.
On the other hand, in the method of changing the mask pattern used for the first high-resolution exposure, for example, a desired pattern line width can be obtained by increasing the fine pattern width on the photomask.
[0020]
DETAILED DESCRIPTION OF THE INVENTION
In the present invention, first, a fine pattern is transferred by high-resolution exposure using a phase shift method or the like, and then pattern transfer is performed so that unnecessary peripheral resist portions are removed after development in the second and subsequent normal exposures. The present invention relates to a multiple exposure method. Therefore, the type of fine pattern is not limited, but a typical example of a pattern desired to be below the resolution limit in the current semiconductor device is a gate pattern of a MOS transistor.
[0021]
Hereinafter, an example of a MOS transistor structure and a pattern will be described, and then an embodiment of a pattern generation method according to the present invention will be described in detail with reference to the drawings, taking as an example ultra-high resolution exposure of the gate pattern.
[0022]
FIG. 1 is a cross-sectional structure diagram of a general MOS integrated circuit.
In FIG. 1, reference numeral 100 denotes a MOS integrated circuit, 101 denotes a semiconductor substrate such as a p-type silicon wafer, 102 denotes a source / drain impurity region into which n-type impurities or the like are introduced at a high concentration, and 104 denotes formation of a parasitic transistor. Therefore, a channel stopper into which a p-type impurity or the like is introduced at a high concentration, LOCOS 106 as an element isolation layer, 108 a gate oxide film, and 110 a gate electrode made of n-type polysilicon or the like.
In such a MOS integrated circuit, the highest line width controllability is required when the gate electrode 110 is processed. That is, since the gate pattern width of the transistor (generally, the gate length Lg) determines the transistor characteristics (driving capability such as the gate threshold voltage Vth and the mutual conductance gm), and the line width variation directly affects the characteristic variation. Control of the line width of the gate electrode pattern is most important in forming a MOS transistor.
[0023]
FIG. 2 is a pattern plan view showing a pattern example that can be used in the phase shift multiple exposure method. FIG. 2 shows a state in which a pattern used for the first high-resolution exposure and a pattern used for the second normal exposure are overlapped.
In FIG. 2, reference numeral 2 denotes a gate electrode pattern, 2a denotes a light shielding pattern portion that is superimposed on an actual gate portion of the gate electrode pattern, 3 denotes a fine pattern that becomes an actual gate portion, and 4 denotes a 180-degree phase shifter. Reference numeral 6 denotes an element region pattern although it is not a pattern directly used for forming the gate electrode.
The fine pattern 3 and the 180-degree phase shifter 4 constitute a pattern used for the first high-resolution exposure, and the gate electrode pattern 2 including the light shielding pattern portion 2a constitutes a pattern used for the second normal exposure.
[0024]
First Embodiment In the conventional phase shift multiple exposure method, the exposure amount is set so that a pattern having a predetermined width is obtained in the first high-resolution exposure for the photoresist layer applied on the wafer. I went up a lot. In the second exposure, the basic idea is that the light exposure pattern part (the light shielding part of the mask pattern) that protects the latent image pattern of the photoresist after the first exposure prevents the exposure from proceeding too much. there were.
[0025]
In contrast, in the phase shift multiple exposure method of the present embodiment, the exposure amount in the first high-resolution exposure is set relatively low. For this reason, the line width of the latent image pattern of the photoresist formed in the first high-resolution exposure is larger than the desired pattern line width.
In the next second normal exposure, for example, by setting the exposure amount higher than the first exposure, the latent image pattern of the photoresist after the first exposure is further exposed even in the second exposure. That is, the second normal exposure in this example is different from the conventional one, and the pattern light after the second normal exposure is positively used by leaking light leaking from the edge of the light shielding pattern portion 2a in the gate electrode pattern 2 in particular. A desired pattern line width is obtained after developing the width.
[0026]
In order to examine the usefulness of the exposure method according to the present embodiment, an actual device was used to transfer the pattern at the time of forming the gate electrode to the photoresist layer on the wafer, and the line width of the fine gate portion was measured.
In this experiment, a photomask for so-called Levenson phase shift was used. As shown in FIG. 9, in this photomask, etching grooves were formed in the quartz glass substrate every other opening of the chromium pattern formed in the quartz glass substrate, and this was used as a 180-degree phase shifter. As a sample, a wafer before forming a gate electrode of a device including a DRAM portion and a logic portion was prepared, and an underlay type antireflection film was first formed with a predetermined thickness. On top of this, a chemically amplified positive resist was applied with a thickness of 0.72 μm, heat-treated, and then exposed. As an exposure apparatus, a KrF excimer stepper (NA: 0.50) was used. In this exposure, after preparing samples with various exposure amounts and depths of focus, the width of the fine pattern part with a design center of 0.16 μm is measured for each of the DRAM part and the logic part using an ordinary length measuring machine. did.
[0027]
FIG. 3 is a graph showing the results of this experiment. In FIG. 3A, only the high-resolution exposure using the phase shift pattern is performed on the premise of the conventional way of thinking that the normal exposure using the second and subsequent light-shielding patterns does not contribute to the line width resolution. It is. On the other hand, in FIG. 3B, after the first high-resolution exposure is performed at a lower exposure amount than in the exposure of FIG. 3A, the second exposure is performed at a higher exposure amount compared to the first exposure. This is a case where normal exposure is performed. In each figure, the horizontal axis of the graph represents the depth of focus, and the vertical axis represents the fine pattern line width measured after development under the same conditions. FIG. 4 is a diagram schematically showing a planar pattern of the logic part and the DRAM part.
[0028]
As a result of performing exposure with various exposure amounts in this experiment, the optimum exposure amount when performing only the high-resolution exposure of FIG. 3A was 44 mJ. In addition, the optimum exposure amount in the case of performing the normal exposure in FIG. 3B was 24 mJ. This is because light leaking from the chrome mask portion of the light shielding pattern portion at the time of the second exposure affects the latent image formation of the photoresist.
[0029]
In order to actively use this leak exposure, the exposure amount at the first high-resolution exposure is lowered (24, 26 mJ), and the exposure amount at the second normal exposure is higher than that at the first exposure (44 mJ). Then (FIG. 3B), it was found that the line width variation due to defocusing can be effectively suppressed as compared with the case where a high-resolution pattern is formed at a stretch with a high exposure amount from the beginning (FIG. 3A). That is, in FIG. 3B, the focus margin that cannot be secured, which is about 0.50 μm in FIG. 3A, is enlarged to 0.7 μm.
In the case of FIG. 3B, it was found that the line width did not change so as to cause a problem even if the exposure amount was slightly changed. Although the line width tends to be slightly reduced on the DRAM side, this can be dealt with by changing the design center.
[0030]
As a result, the exposure amount of ± 5% is obtained by the method of this embodiment in which a desired gate line width is obtained by reducing the exposure amount at the first high-resolution exposure and increasing the exposure amount at the second normal exposure. Even if the amount varies, a resist pattern for forming a gate of 0.16 μm can be formed with a variation of ± 10% with a large focus margin of ± 0.35 μm.
[0031]
In the above description, the case where the gate pattern is formed by the second exposure is illustrated, but the exposure method of the present embodiment can be applied not only to the gate pattern but also to other fine patterns, and in the second and subsequent normal exposures. There is no limit to the number of times. When performing multiple normal exposures, it is preferable to use a mask pattern having a light-shielding portion whose line width is larger than that of the mask pattern used at the previous exposure, and to increase the exposure amount from the previous exposure. Then, the focus margin can be expanded as described above. Note that the initial exposure is not limited to the one using the phase shift method in the sense that the exposure amount is increased while sequentially expanding the light-shielding region in this way by multiple times of exposure.
[0032]
Second Embodiment This embodiment is a case where a mask pattern used for high-resolution exposure is changed instead of changing the exposure amount in the first embodiment.
[0033]
FIG. 5 is a pattern plan view in each generation process of this high-resolution exposure pattern generation example.
In this method, a fine pattern is extracted from input CAD data. First, the gate pattern 10 and the element region pattern 12 are read out, these element shape patterns are overlapped with each other (FIG. 5A), and a graphic product process (AND process) is performed (FIG. 5B). The fine gate portion 14 is extracted (FIG. 5C).
Next, the extracted fine gate portion 14 is resized twice. Specifically, as shown in FIG. 5D, the gate width Wgo of the fine gate portion 14 is increased by a predetermined length on both sides in the gate width direction in order to obtain an alignment margin at the time of multiple exposure. Further, as shown in FIG. 5E, the fine gate portion 14a after being expanded in the gate width direction is now resized on both sides in the gate length ("line width" in the present invention) direction. Thereby, the final fine pattern 14b in which the original gate length Lgo is expanded to Lg is obtained. The resized gate length Lg is one of the parameters that most affects the line width for resolving the gate pattern.
Finally, as shown in FIG. 5F, the generation of the high resolution pattern is completed by disposing phase shifters of 180 degrees and 0 degrees on both sides of the fine pattern 14b. The phase shifter arrangement is performed so that the phase difference is 180 degrees on both sides of each fine pattern when there are a plurality of fine patterns.
[0034]
The first exposure is performed using the high-resolution exposure pattern in which the width of the fine pattern (for example, the gate length) is expanded in this way. In the second and subsequent normal exposures, exposure is performed by covering the latent image pattern of the photoresist formed by the first exposure with a light shielding pattern portion.
As a result, as in the first embodiment, a desired pattern line width can be obtained by positively using light leaking from the light shielding pattern portion during the second exposure. In addition, the focus margin can be expanded as compared with the case where the exposure amount at the first exposure is increased and most of the pattern is resolved by one exposure.
[0035]
【The invention's effect】
According to the exposure method of the present invention, for example, when forming a fine latent image pattern of a photoresist by multiple exposure of a first high-resolution exposure using a phase shift method or the like and a second or subsequent normal exposure. The variation in desired fine pattern width can be reduced. In addition, the focus margin can be increased.
[Brief description of the drawings]
FIG. 1 is a cross-sectional structure diagram of a general MOS integrated circuit showing an example of a device having a fine pattern to which the exposure method of the present invention can be applied.
FIG. 2 is a pattern plan view showing a pattern example that can be used in the phase shift multiple exposure method according to the embodiment of the present invention.
FIG. 3 is a graph showing the results of an experiment conducted for examining the usefulness of the multiple exposure method according to the first embodiment of the present invention. (A) is a case where only high-resolution exposure using a phase shift pattern is performed on the premise of the conventional way of thinking that normal exposure using the second and subsequent light-shielding patterns does not contribute to line width resolution. . (B) shows the case where the first high-resolution exposure is performed at a lower exposure amount than that in the exposure of (a), and then the second normal exposure is performed at a higher exposure amount as compared with the first exposure time. is there.
4 is a diagram schematically illustrating a planar pattern of a device used in the experiment of FIG. 3, in which (a) shows a logic portion and (b) shows a DRAM portion.
FIG. 5 is a pattern diagram illustrating a generation process of a high-resolution exposure pattern used in a phase shift multiple exposure method according to a second embodiment of the present invention, taking a gate pattern as an example.
FIG. 6 is a diagram showing the principle of a phase shift method in comparison with a conventional method.
FIG. 7 is a diagram showing Fourier spectra of a phase shift method and a conventional method.
FIG. 8 is a diagram showing the principle of so-called Levenson-type phase shift using a negative resist.
FIG. 9 is a diagram illustrating the principle of Levenson-type phase shift using a positive resist as a phase shift multiple exposure method.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Mask pattern for 1st time and 2nd exposure, 2 ... Gate electrode pattern, 2a ... Light-shielding pattern part, 3 ... Fine pattern part, 4 ... 180 degree phase shifter, 6 ... Element area pattern, 10 ... Gate pattern, 12 ... element region pattern, 14 ... fine gate part, 14a, 14b ... fine gate part after size change (fine pattern).

Claims (7)

High-resolution exposure that uses a phase shift pattern to transfer the pattern of areas with severe line width controllability to the photoresist layer, and the portion of the photoresist layer that has already been transferred by the high-resolution exposure is protected by the light shielding part of the mask pattern On the other hand, a latent image pattern of the photoresist is formed by multiple exposures including normal exposure in which a pattern at a relatively loose line width controllability to the photoresist layer is transferred without using a phase shift pattern. An exposure method comprising:
In the high-resolution exposure, using an exposure condition in which the line width of the portion where the line width controllability is severe in the latent image pattern transferred to the photoresist layer by the high-resolution exposure becomes thicker than a desired line width,
In the normal exposure, the portion defining the line width is exposed and exposed at the portion where the line width controllability of the latent image pattern formed by the high resolution exposure is thicker than a desired line width. An exposure method in which a line width is reduced and a desired line width is obtained at a location where the line width controllability is severe after the normal exposure. The exposure method according to claim 1, wherein in the high-resolution exposure, an exposure amount is adjusted as the exposure condition. In the normal exposure, claims the use of the mask pattern having the light shielding portion of the size that can be exposed by overlapping the photoresist layer portions that are already pattern transfer by the high resolution exposure by light leaking from the light blocking part Item 2. The exposure method according to Item 1. High-resolution exposure that uses a phase shift pattern to transfer the pattern of areas with severe line width controllability to the photoresist layer, and the portion of the photoresist layer that has already been transferred by the high-resolution exposure is protected by the light shielding part of the mask pattern However, exposure to form a photoresist latent image pattern by multiple exposures including normal exposure for transferring a pattern having a relatively loose line width controllability to the photoresist layer without using a phase shift pattern A method,
In the high-resolution exposure, using a mask pattern in which the line width of the portion where the line width controllability is severe in the latent image pattern transferred to the photoresist layer by the high-resolution exposure becomes thicker than a desired line width,
In the normal exposure, the portion defining the line width is exposed and exposed at the portion where the line width controllability of the latent image pattern formed by the high resolution exposure is thicker than a desired line width. An exposure method in which a line width is reduced and a desired line width is obtained at a location where the line width controllability is severe after the normal exposure. The high resolution exposure, the line extracting width controllability strict position of the pattern from the already designed element shape pattern, using said mask pattern thickened over resizing at least in the line width direction with respect to extracted the pattern The exposure method according to claim 4. In the normal exposure, claims the use of the mask pattern having the light shielding portion of the size that can be exposed by overlapping the photoresist layer portions that are already pattern transfer by the high resolution exposure by light leaking from the light blocking part Item 5. The exposure method according to Item 4. An exposure method for forming a latent image pattern of a photoresist by performing multiple exposures,
The first exposure is performed such that the line width of the latent image pattern transferred to the photoresist is thicker than a desired line width,
Exposure of second and subsequent, by protecting the latent image pattern of the photoresist already formed by extending the light-shielding area than the mask pattern used at the time of previous exposure, and is more than the previous exposure amount of exposure , Exposing an end portion defining the line width of the latent image pattern formed by the first exposure so that the line width is thicker than a desired line width, and gradually reducing the line width to obtain a desired line width. Exposure method to obtain .

Images