JP2000195780A - Charged particle beam exposure device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the pattern dimension accuracy from deteriorating due to re-reflected electrons by executing the correcting exposure besides the usual pattern exposure. SOLUTION: A charged particle beam is irradiated to form a desired pattern on a sample, while the exposure region on the sample is divided like meshes and the additional exposure quantity to reflected particles from other small regions due to the beam irradiation is calculated every divided small region. A mask board 61 has correcting exposure masks 63 different in beam transmittance in corresponding sizes to the small regions, and the correcting exposure mask 63 of the board 61 is selected so as to correct the nonuniformity of the additional exposure quantity among the individual small regions, thereby executing the correcting exposure of each small region. Using a correcting exposure mask 63 formed so that the beam transmittance differs according to the additional exposure quantity at each small region, the correcting exposure is applied to a plurality of small regions en bloc.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure technique using a charged particle beam, and more particularly to a charged particle beam exposure apparatus for reducing a long-distance sensitizing effect due to particles reflected on a sample surface.

[0002]

2. Description of the Related Art In recent years, a pattern transfer apparatus for transferring an LSI (Large Scale Integrated Circuit) pattern onto a sample by an electron beam has been proposed. In this apparatus, an electron beam emitted from an electron gun is irradiated on a pattern transfer mask, and an electron beam having pattern information transmitted through or scattered by the mask is reduced by an objective lens and a reduction lens. Then, the pattern is transferred to the sample placed in the sample chamber by positioning with a deflector.

Since this type of pattern transfer apparatus has a higher throughput than other electron beam lithography apparatuses, it is becoming the mainstream of electron beam lithography. However, in the pattern transfer apparatus using an electron beam, there is a problem that the accuracy of the pattern dimension is deteriorated due to the “long-distance photosensitive action” described below.

When patterning a sample using an electron beam, as shown in FIG.
When 1 is incident on the sample (the substrate 83 coated with the resist 82), reflected electrons or secondary electrons resulting from this are emitted from the sample. These reflected electrons or secondary electrons are
It hits a structure above the sample, specifically the ceiling of the sample chamber or the lower surface of the objective lens 84, from which reflected electrons or secondary electrons are emitted again and return to the sample. The range in which such reflected electrons and secondary electrons (hereinafter collectively referred to as re-reflected electrons) 85 reach on the sample is almost equal to the distance between the sample and the structure above it.

[0005] For example, in the case of a general pattern transfer device, the range in which the re-reflected electrons reach extends from several mm to several tens of mm from the point of arrival of the primary beam. The re-reflected electrons become extra exposure compared to the original exposure, and give the resist an exposure amount more than expected. When the resist is given an exposure amount larger than expected, the size of the pattern transferred to the resist varies. Such an exposing action caused by the re-reflected electrons will be referred to herein as a long-distance photosensitive action.

[0006] The dimensional fluctuation of the pattern due to the long-distance photosensitive action is caused by repetitive multiple scattering of re-reflected electrons between the sample and the structure located above the sample, thereby increasing the area over a wider area, for example, two to three times the above assumption. May range.
Further, photoelectrons and photons generated by irradiating an object above the sample with photons such as X-rays generated when the primary beam is incident on the sample, or reflected electrons from the sample enter a structure above the sample. Photons, such as X-rays, generated from the light source also cause a long-distance sensitizing effect.

[0007] Such a long-distance photosensitive action causes deterioration in uniformity of the resist pattern dimension in the sample plane when the pattern density of the pattern exposed on the sample (resist on the sample) varies depending on the location. . As an example, let us consider a case where a large pattern 92 exists near a line and space (L / S) pattern 91 as shown in FIG. 9 (that is, the pattern density increases). When the large pattern 92 is transferred, re-reflected electrons are generated and reach the area of the L / S pattern 91. Then, the total irradiation amount in the L / S pattern portion is
The number of the large patterns 82 increases as compared with the case where they are not in the vicinity.

It has been found that, with a positive resist, the dimension of the line pattern becomes smaller than expected as the irradiation dose increases. Therefore, if a long-distance photosensitive action is present, even if the patterns have the same dimensions, the dimensions will differ if the pattern density in the vicinity differs.
In addition, this long-distance sensitizing action has an effect of several mm to several tens.
It extends to a range of up to 0 mm, and as can be seen from the description of the generation mechanism described above, this is different from the so-called proximity effect which has conventionally been a problem in electron beam exposure.

Conventionally, in order to reduce the amount of re-reflected electrons, FIG.
As shown in FIG. 5, an antireflection plate 87 made of a light element having a small atomic number such as beryllium (Be) or carbon (C) was placed above the sample. This is, for example, a 1 mm thick copper (Cu) plate having a thickness of 5 mm and a diameter of 150 mm.
The structure is such that a m plate made of Be is attached. The reason for using such a light element is that the amount of reflected electrons (and secondary electrons) is relatively smaller than that of a heavy element such as iron (Fe) commonly used as a material for a sample chamber or an objective lens. It is because it becomes less. But actually, such a B
Even if an antireflection plate made of a light element such as e or C is used, the effect of preventing re-reflected electrons is not sufficient.

[0010]

As described above, in the conventional charged particle beam exposure apparatus, there is a problem that a pattern dimension variation occurs in a sample surface due to a long-distance photosensitive action by re-reflected electrons. Even if an anti-reflection plate made of a light element such as Be or C is used to reduce the amount of re-reflected electrons, the effect of preventing re-reflected electrons is not sufficient.

The present invention has been made in view of the above circumstances, and an object thereof is to prevent deterioration of pattern dimensional accuracy due to re-reflected electrons (in particular, deterioration of uniformity in a substrate surface). It is an object of the present invention to provide a charged particle beam exposure apparatus which can improve the pattern dimensional accuracy.

[0012]

(Structure) In order to solve the above problem, the present invention employs the following structure.

According to the present invention, in a charged particle beam exposure apparatus, means for exposing a desired pattern on a sample by irradiating a charged particle beam, an exposure area on the sample is divided into meshes, and each of the divided areas is divided into meshes. A mask substrate having a plurality of correction exposure masks each having a size corresponding to the small region and a different beam transmittance, and a means for calculating an additional exposure amount due to reflected particles accompanying beam irradiation of another small region for each small region. ,
Means for selecting a correction exposure mask on the mask substrate and correcting and exposing each of the small areas so as to correct the non-uniformity of the additional exposure amount between each of the small areas.

According to the present invention, in a charged particle beam exposure apparatus, a means for exposing a desired pattern on a sample by irradiating a charged particle beam, an exposure area on the sample is divided into meshes, and each of the divided areas is divided into meshes. Means for calculating the additional exposure amount due to the reflected particles accompanying the beam irradiation of other small regions for each small region, and correction using a correction exposure mask so as to correct the non-uniformity of the additional exposure amount between each small region Means for exposing, wherein the correction exposure mask is formed such that the beam transmittance is different for each area corresponding to each small area on the sample, according to the additional exposure amount in each small area. It is characterized by having.

(Operation) As described above, when the pattern density is not uniform in the sample surface, the long-distance photosensitive operation is not uniform in the sample surface. As a result, the uniformity of the pattern size in the sample surface also deteriorates. As one of the methods for suppressing this deterioration, there is a method of performing correction exposure so as to cancel out unevenness of the exposure amount due to the long-distance photosensitive action. That is, the additional exposure is performed at a place where the long-distance photosensitive action is small, so that the additional exposure amount corresponding to the long-distance photosensitive action is made uniform in the substrate surface.

In the present invention, in order to perform this correction exposure,
A feature is that a plurality of correction exposure masks corresponding to various correction exposure amounts are used by gradually changing the pattern density or the aperture ratio.

If exposure using such a correction exposure mask is performed under the same optical conditions as normal exposure, the correction pattern itself is transferred, and exposure unevenness occurs. In order to prevent exposure unevenness, the focus is shifted by an objective lens or the like, and the correction pattern is blurred and transferred. By performing the transfer by blurring in this manner, uniform exposure can be achieved without unevenness without resolution of the correction pattern itself. The degree of the blur may be set, for example, such that adjacent patterns overlap each other at 50% of each beam profile. In this way, the connecting portions of the adjacent patterns are smoothly connected. When the beam size is changed by shifting the focus, the beam size may be kept unchanged by appropriately adjusting a reduction lens or the like.

As described above, according to the present invention, non-uniformity in the amount of re-reflected electrons is eliminated by performing correction exposure in addition to normal pattern exposure instead of reducing the amount of re-reflected electrons. To make the sum of the amount of re-reflected electrons and the corrected exposure uniform in all small areas), and to suppress variations in pattern dimensions in the sample plane that depend on the pattern density due to long-range photosensitivity. Becomes possible.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Before describing the embodiments, the basic principle of the present invention will be described with reference to FIG.

In the present invention, the procedure for correcting the dimensional fluctuation in the substrate due to the long-distance photosensitive action is as follows. First, a quantitative evaluation of the long-distance photosensitive action in the substrate surface is performed. The exposure (drawing) area in the substrate is a small area m
(i) Divide into (i = 1 to n). The size of this small area is set to such a size that approximation that the long-distance photosensitive action is constant in the area can be applied. The range in which the dimensional variation due to the long-distance photosensitive action cannot be ignored differs depending on the configuration of the apparatus or the resist system. Usually, the radius often ranges to several cm. In such a case, if the size of the small area is set to about 1 mm square, the above approximation is sufficiently established.

Next, for each small area m (i), the average dose <D (i)> within the area is determined. This can be obtained from the pattern data and the dose data.
Next, the degree of the long-distance photosensitive action from one small area m (i) to another small area m (j) is determined by the converted dose Dr (i →
j). Here, the converted dose is a dose obtained by converting a long-distance photosensitive action into a dose of a primary beam (an incident beam used for pattern exposure). Where D
Let r (i → j) be Dr (i → j) = <D (i)> · f (xi−xj) (formula 1). Here, f (xi-xj) represents the converted dose of the long-distance photosensitive action per unit dose, with the distance from the incident point of the incident beam as a variable. The two-dimensional distribution of the converted dose f (xi-xj) is obtained in advance by another experiment. Where the distance between two small areas |
xi-xj | is the distance between the centers of the small areas. Using the above equation, the sum Dr (j) of the long-distance photosensitive action from each area to a certain area m (j) is an integral from i = 1 to n of (Equation 1), and thus Dr (j) ) = Σ _i Dr (i → j) = Σ _i <D (i)> · f (xi−xj) (Equation 2) In this way, the degree of the long-distance photosensitive action for all the small areas is quantitatively determined. In order to correct the non-uniformity of the dimensional fluctuation caused by the long-distance photosensitive action,
For all the small areas m (j), Dr (j) + C (j) = K; where K = (constant) (Equation 3) The correction dose amount C (j) for each small area And exposure may be performed. Here, for simplicity, Max
[Dr (j)] = K.

In the present invention, the correction dose is adjusted by changing the density or the aperture ratio of the pattern on the correction exposure mask. As the correction pattern, a hole pattern as shown in FIG. 2A or an L / S pattern as shown in FIG. 2B can be used. It is preferable that these patterns have a uniform pattern density.

Further, in actual exposure, it is necessary to blur and expose the pattern so that the correction pattern itself is not transferred. In order to transfer the pattern by blurring, the focus may be shifted by changing the intensity of the objective lens. Alternatively, there is a method in which a minute pattern smaller than the resolution limit of the electron optical system or the resist system is used as the correction pattern. This is particularly effective in the case of a reduction optical system. For example, if the reduction ratio of the optical system is 1/20,
0.4 .mu.m on the mask corresponds to 0.02 .mu.m on the sample surface, which is less than the resolution limit and is not resolved, and can be regarded as a substantially uniform dose distribution. Note that the correction of the long-distance photosensitive action may be appropriately combined with the proximity effect correction.

Hereinafter, the present invention will be described in detail with reference to the illustrated embodiments.

(First Embodiment) A method of manufacturing a stencil-type correction exposure mask (aperture) as one of the correction exposure masks used in the exposure apparatus of the present invention will be described. To manufacture such a stencil type mask, an etching method, a lift-off method, or the like can be applied. Here, an outline of a manufacturing process in which an etching method is applied to an SOI (Silicon On Insulator) substrate will be described.

The structure of the SOI substrate is a 25 μm thick Si
A thin film layer, a 2 μm thick Si oxide film layer, and a 600 μm thick Si substrate (the plane orientation of the main surface of the substrate is (100)). First, a 100 nm thick S
An i-nitride film is formed using the LPCVD method. Next, a 1 μm thick resist is applied to the surface of the substrate. Next, a correction pattern is formed on the resist using photolithography or electron beam lithography.

Next, an aperture (opening) is formed in the Si thin film layer by dry etching using the patterned resist as a mask. For dry etching, reactive ion etching (RIE) in which a reactive gas such as a fluorocarbon (SF ₆ , BCl ₃ ) is appropriately combined.
Method. Next, a resist is applied to the back surface of the substrate,
Using a double-sided aligner, the pattern for the back surface opening is exposed by aligning the position with the pattern on the front side. Next, using the resist as a mask, the RIE method is used (the reaction gas is
Oxygen and Freon) and pattern the Si nitride film layer on the back surface.

Next, using the Si nitride film layer as a mask, anisotropic etching using a KOH aqueous solution is performed to form an opening from the back surface of the substrate. This KOH etching is stopped at the Si oxide film layer. Next, the silicon oxide film layer used as the KOH etching stop layer is removed by wet etching with an aqueous solution of ammonium fluoride to penetrate the aperture portion. Further, the Si nitride film is removed by a chemical dry etching (CDE) method using CF ₄ or oxygen radicals.

Next, a Ti film having a thickness of 100 nm and a Pt film having a thickness of 200 nm are formed as conductive layers on both surfaces of the aperture plate by RF sputtering. Ti,
The Pt film may be formed by a vacuum deposition method. FIG. 3 shows an example of the correction mask substrate thus manufactured.
In this example, aperture groups having different aperture ratios are arranged in an array on a substrate.

In FIG. 3, reference numeral 31 denotes a silicon substrate (mask substrate), 31 denotes an opening on the back surface of the substrate, 32 denotes a silicon membrane, and 33 denotes a group of apertures for correction exposure.

Next, the correction exposure pattern will be described. When a certain LSI pattern was exposed to a positive resist having a sensitivity of 24 μC / cm ² , the dimensional change due to the long-distance photosensitive action was ± 15 nm from the average value. That is, the dimensional difference between the place where the long-distance photosensitive action was the largest and the place where the photosensitive action was the smallest was 30 nm. From the two-dimensional distribution of the converted dose of the long-distance photosensitive action obtained in another experiment and the pattern data and the exposure data, the converted dose of the long-distance photosensitive action on the sample surface when this pattern was exposed. The ability to calculate the distribution was described above.

When the exposure area is divided into small areas of 1 mm square and the converted dose distribution is obtained, the difference in the converted dose between the small area having the largest long-distance exposure and the smallest area is 3 μC / cm ^2. Met. From now on, 1μC /
The dimensional difference per cm ² is 10 nm / μC / cm
It becomes ² . In this case, the type of the correction exposure amount is 0.1
0.12 μC / cm ² from ² μC / cm ² to 3 μC / cm ²
If 25 types are prepared in cm ² steps, the dimensional change due to the long-distance photosensitizing effect can be theoretically reduced to 30 nm / 25 =
It can be suppressed to 1.2 nm.

[0033] (the difference in terms of dose) / since it is (resist ^{sensitivity) = 3μC / cm 2 / 24μC} / cm 2 = 0.125, the type of correction exposure mask pattern density (open area ratio) 12.5 25 types are prepared from 0.5% to 0.5% in 0.5% steps. Reduction ratio of exposure equipment is 1/20
Then, the size of one mask is 1 mm □ × 20 = 2
0 mm □. These 25 types of masks are arranged on a mask substrate. FIG. 4 shows a cross-sectional view of an example of such a mask substrate. In the figure, reference numeral 41 denotes a mask substrate, and reference numeral 43 denotes a correction exposure mask. In this example, 25 types of mask groups are arranged in a 5 × 5 matrix. The pattern of the mask for this correction exposure is an L / S pattern or a hole-shaped pattern.

Using the mask substrate for correction exposure manufactured in this manner, a small area of 1 mm square is sequentially corrected and exposed while selecting a mask for correction exposure so as to satisfy the condition of (Equation 3). In addition, the dimensional fluctuation due to the long-distance photosensitive action could be corrected to an average value of ± 2 nm or less.
In this correction exposure, the focus is shifted by adjusting the objective lens, and the L / L of the correction exposure pattern is adjusted.
S and hole-shaped patterns were not resolved so that uniform exposure was achieved. The degree of defocus may be such that adjacent patterns overlap at 50% of their beam profiles. When the beam size is changed by shifting the focal point, the beam size may be kept unchanged by appropriately adjusting a reduction lens or the like.

(Second Embodiment) Next, as another example of the correction exposure mask used in the exposure apparatus of the present invention, a method of manufacturing a scattering type correction mask will be described.

First, the Si of the main surface whose plane orientation is (110)
On both sides of the substrate, 300 nm of Si was formed by LPCVD.
A nitride film is formed. Next, a 10 mm thick Cr layer and a 50 mm W layer are sequentially formed on the surface of the substrate by sputtering. Next, a resist is applied to the back surface of the substrate, and after baking, 25 patterns each having a width of 4 mm and a length of 100 mm are formed in parallel at a pitch of 6 mm on the back surface of the substrate using an exposure apparatus. However, the long side direction of each pattern is parallel to the direction of the plane direction (211). After pattern exposure, the resist is developed and post-baked, and then RIE equipment (etching gas is CF ₄ and oxygen)
Is used to pattern the Si nitride film layer using the resist as a mask.

Next, back etching of the Si substrate is performed using a 20 wt%, 90 ° C. KOH aqueous solution and using the patterned Si nitride film as a mask. Under the above etching conditions, the silicon nitride film layer on the front side is etched in about 3 hours and 30 minutes. The shape of the opening is such that the side wall is vertical in the long side direction, and the side wall is approximately 35
It has a ° taper. Through this back etching step, a (parallelogram) window of a Si nitride film having a width of 4 mm and a length of about 95 mm can be formed.

Next, a resist is applied to the front side of the substrate,
After further baking, patterning is performed using an exposure apparatus. Next, after the resist is developed and further baked, an RIE device (etching gas is, for example, SF ₆ , C
Using HF ₃ ), the resist pattern is transferred to the W layer. In this etching step, the Cr layer below the W layer becomes an etching stop layer. FIG. 5 shows an example of the mask substrate thus manufactured. In the figure, reference numeral 51 denotes a mask substrate, and 53 denotes a correction exposure mask. In this example, by sequentially exposing the correction exposure mask formed in a band shape, the correction exposure is completed for the entire chip.

FIG. 6 shows an example of a scattering type mask. In the figure, 61 is a silicon substrate having a plane orientation (110), 61a is an opening on the back surface of the substrate, 62 is a membrane of a silicon nitride film, 6
Reference numeral 3 denotes a transfer pattern using tungsten.

Since the pattern of this mask is supported by a membrane such as a Si nitride film, the pattern area can be made relatively large. For example, in the above production example,
If the reduction ratio of the exposure apparatus is 1/4, 1 mm x 25 mm
Area can be collectively exposed. In this example, since there are 25 correction masks, a 25 mm square area can be subjected to correction exposure without replacing the mask substrate. Also in this case, if the correction exposure amount is determined in 1 mm square units, sufficient correction accuracy can be maintained. Scatterer layer (W
The correction pattern to be formed on the (layer) may be a hole shape or an L / S pattern as in the case of the first embodiment.

When exposing the correction pattern, so-called blur exposure is performed by adjusting the objective lens system and shifting the focus. As for the degree of defocus, if the edge profiles overlap at a height of 50% in the beam current distribution of the adjacent patterns, the joints between the patterns are also noticeable without resolution. Exposure can be performed at a substantially uniform irradiation density without causing the irradiation. Correction exposure is performed with an appropriate size by appropriately adjusting the reduction lens.

(Third Embodiment) An example of an electron beam exposure apparatus using a correction exposure mask is schematically shown in FIG.
The electron beam emitted from the electron source 10 passes through a condenser lens system 11 and is formed by an aperture 12. Further, blanking can be appropriately performed by the blanking deflector 13. The beam formed by the aperture 12
The light is appropriately deflected by the beam deflector 14 and is irradiated on the exposure mask 21 through the projection lens system 15. Exposure mask 2
As 1, a mask for pattern transfer and a mask for correction exposure are replaceable, and one of the masks is set on a movable mask stage 23 driven by a mask stage drive system 22.

The electron beam having the pattern information passing through the exposure mask (in this case, the pattern transfer mask) 21 is reduced by the reduction lens 16 and the objective lens 18. The electron beam scattered by the mask 21 has an aperture 1
Blocked at 7. This is effective for improving the contrast of the pattern. Then, the electron beam is positioned by the beam deflector 19 and the sample 24 such as a wafer is
An image is radiated thereon, whereby the pattern is transferred.
The sample 24 is set on a movable sample stage 26 driven by a sample stage drive system. The operation of both the movable stages 23 and 26 for the mask and the sample is performed by the laser interferometer 2.
7 and is fed back to the beam control system.

When a normal LSI pattern is transferred using this exposure apparatus, unnecessary exposure is added due to the long-distance photosensitive action of the re-reflected electrons as described above. Therefore, correction exposure is performed using the correction exposure mask described in detail in the first embodiment or the second embodiment. This correction exposure mask may be formed on the same substrate as the LSI pattern transfer mask, or may be formed on another substrate. In the latter case, the mask may be replaced with an LSI pattern transfer mask by a mechanism for replacing the mask substrate, or both may be arranged side by side on a transfer mask stage.

The correction exposure mask has several LSs.
In some cases, it can be used for correction of an I transfer mask for general purposes. In such a case, if the mask for correction exposure is always mounted and only the mask for LSI transfer is replaced, the trouble of replacing the mask for correction exposure can be saved. In this case, a correction exposure is performed by selecting a mask having an optimum pattern density from general-purpose correction exposure masks according to the pattern density of the LS1 pattern.

As another application, it is also possible to adjust the correction exposure amount by appropriately adjusting the exposure time.

When the pattern is transferred using the apparatus of this embodiment, the correction exposure is performed in addition to the normal pattern exposure, so that the sum of the amount of re-reflected electrons and the correction exposure is made uniform in all the small areas. Can be done. That is, it is possible to eliminate the non-uniformity of the amount of re-reflected electrons, and it is possible to suppress the variation of the pattern dimension in the sample surface depending on the pattern density due to the long-distance photosensitive action.

The present invention is not limited to the above embodiments. In the embodiment, a plurality of exposure masks having different beam transmittances are prepared in advance, and an exposure mask corresponding to each small area to be corrected and exposed is selected.
A plurality of small areas can be collectively exposed by using a correction exposure mask formed such that the beam transmittance varies depending on the amount of additional exposure in each small area. In the embodiment, the electron beam exposure apparatus has been described as an example, but the present invention can be applied to an ion beam exposure apparatus. In addition, various modifications can be made without departing from the scope of the present invention.

[0049]

As described above in detail, according to the present invention, by performing correction exposure in addition to normal pattern exposure,
Easily degrades in-plane uniformity of dimensional accuracy caused by so-called re-reflection particles, in which reflected or secondary electrons from the sample due to the incident beam hit the structure above the sample and return to the sample again. Can be corrected.

[Brief description of the drawings]

FIG. 1 is a schematic diagram for explaining a basic principle of the present invention and showing a procedure for obtaining an additional exposure amount by re-reflected electrons for each small area obtained by dividing an exposure area into a mesh shape.

FIG. 2 is a conceptual diagram of a correction exposure mask for explaining the basic principle of the present invention.

FIG. 3 is a view showing an example of a mask substrate for correction exposure according to the first embodiment.

FIG. 4 is a diagram showing an example of a correction exposure mask corresponding to 25 types of aperture ratios.

FIG. 5 is a diagram showing an example of a correction exposure mask according to a second embodiment.

FIG. 6 is a diagram showing an example of a scattering mask.

FIG. 7 is a diagram showing a basic configuration of an electron beam exposure apparatus according to a third embodiment.

FIG. 8 is a view for explaining a distance photosensitive action by re-reflected electrons.

FIG. 9 is a diagram showing an example of a pattern arrangement in which a long-distance photosensitive action occurs.

FIG. 10 is a diagram showing an example of a conventional backscattered electron prevention plate.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Electron gun 11 ... Condenser lens system 12, 17 ... Aperture 13 ... Blanking deflector 14, 19 ... Beam deflector 15 ... Projection lens system 16 ... Reduction lens 18 ... Objective lens 21 ... Transfer mask 22 ... Movable mask Stage 23: Mask stage drive system 24: Sample 25: Movable sample stage 26: Sample stage drive system 27: Laser interferometer 31, 41, 51, 61 Silicon substrate 31a, 61a Opening 32, 62 ... Membrane 33, 43, 53, 63: Mask for correction exposure

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] November 11, 1999 (1999.11.
11)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0013

[Correction method] Change

[Correction contents]

According to the present invention, in a charged particle beam exposure apparatus, means for exposing a desired pattern on a sample by irradiating a charged particle beam, an exposure area on the sample is divided into meshes, and each of the divided areas is divided into meshes. In each small area, the reflected particles accompanying the beam irradiation of other small areas hit the structure above the sample.
Means for calculating the additional exposure amount by the re-reflected electrons returning again, and a mask substrate having a plurality of correction exposure masks having different beam transmittances in sizes corresponding to the small areas,
Means for selecting a correction exposure mask on the mask substrate and correcting and exposing each of the small areas so as to correct the non-uniformity of the additional exposure amount between each of the small areas.

[Procedure amendment 3]

[Document name to be amended] Statement

[Correction target item name] 0014

[Correction method] Deleted

Claims (2)

[Claims] 1. A means for exposing a desired pattern on a sample by irradiating a charged particle beam, an exposure area on the sample being divided into meshes, and a beam of another small area for each of the divided small areas. Means for calculating the amount of additional exposure due to reflective particles associated with the irradiation, a mask substrate having a plurality of correction exposure masks having sizes corresponding to the small regions and different beam transmittances, and an additional amount of exposure between each of the small regions Means for selecting a correction exposure mask on the mask substrate and correcting and exposing each small area so as to correct the non-uniformity of the charged particle beam exposure apparatus. 2. A means for exposing a desired pattern on a sample by irradiating a charged particle beam, an exposure region on the sample being divided into meshes, and a beam of another small region for each of the divided small regions. Means for calculating the amount of additional exposure by the reflective particles associated with the irradiation, and means for performing correction exposure using a mask for correction exposure to correct the non-uniformity of the amount of additional exposure between each small area, The charged particle beam, wherein the correction exposure mask is formed so that a beam transmittance differs according to an additional exposure amount in each small region for each region corresponding to each small region on the sample. Exposure equipment.