JP3350416B2 - Pattern formation method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a pattern forming method.

[0002]

2. Description of the Related Art Optical lithography in a semiconductor manufacturing process has advantages such as process simplicity and low cost.
It has been widely used in device production. In recent years, miniaturization of elements of 0.25 μm or less has been achieved by shortening the wavelength (KrF excimer laser light source). For further miniaturization, ArF excimer laser light sources of shorter wavelengths and Levenson-type phase shift masks are being developed, and are expected as mass-production lithography tools compatible with the 0.15 μm rule. However, there are many challenges to achieve this, and the time involved in its development has become longer,
There is a concern that it will not be able to keep up with the speed of device miniaturization.

On the other hand, electron beam lithography, which is a first candidate for post-optical lithography, has been proved to be capable of processing down to 0.01 μm using a narrow beam. Although there seems to be no problem for the time being from the viewpoint of miniaturization, there is a problem in throughput as a device mass production tool. In other words, it takes time to draw fine patterns one by one in order. In order to shorten the writing time, several devices have been developed, such as a cell projection method for partially writing a repeated portion of a ULSI pattern collectively. However, even with the use of these apparatuses, they have not been able to catch up with the throughput of optical lithography.

As a method of increasing the throughput of electron beam lithography, pattern transfer to the same resist is performed by light exposure and electron beam exposure, and the electron beam lithography apparatus can process an hour by reducing the exposure area of electron beam exposure. Increasing the number of wafers, that is, mixing and matching the same layer of light and EB has been proposed. JP-A-4-1
No. 55812 discloses that in a lithography step of pattern formation, pattern transfer to the same resist is performed using light exposure and electron beam exposure using a phase shift mask. According to this, most of the patterns constituting the element are transferred using a phase shift mask, and the portion where the inconvenience occurs in the arrangement of the phase shifter is repaired by electron beam drawing, so that the electron beam drawing area is made as small as possible. As a result, the number of wafers that the electron beam lithography apparatus can process per hour is increased. However, in this method, although the drawing area is small, it is not possible to transfer a pattern having a resolution lower than the limit resolution of the phase shift mask, and cannot cope with future miniaturization of devices.

[0005] Further, since it takes a long time to manufacture a mask when a small number of devices of various types are manufactured, as a means for solving the problem, Japanese Patent Laid-Open No. 1-293616 discloses a unitary common to various semiconductors for the same resist. And a method of drawing a pattern unique to each semiconductor element with an electron beam. That is,
A mask for a portion common to various elements is prepared in advance, and electron beam writing is used only for a portion where another pattern is changed. In this method, since it is not necessary to make a mask for each type of mask, the period from element design to manufacturing can be shortened. However, in the case of this method, similarly to the above-described method, the method cannot be applied when a pattern having a resolution lower than the limit in light exposure exists in the functional block. Yes, to apply the electron beam exposure it takes really time for going drawn one by one turn of the pattern, it has become as difficult to use as a lithography system for drawing a fine pattern at a high speed.

[0006] As described above, in the mix-and-match of the same layer of light and EB, which has been conventionally carried out to improve the throughput, the resolution of electron beam exposure is not sufficiently utilized and the throughput is lower than that of the optical stepper. There were problems such as not reaching the same level.

In order to solve the above problem, the inventors of the present invention have proposed a lithography method for exposing the same photosensitive material with EB and light having both excellent resolution exceeding the light of electron beam exposure and throughput equivalent to that of an optical stepper. The system has already been proposed (Japanese Patent Application No. 9-46683).

FIGS. 22 and 23 are views showing the basic concept of the lithography system. FIG. 22 is a diagram showing a concept of a pattern forming method and a lithography system, and FIG. 23 is a diagram showing a planar arrangement of the lithography system. In these figures, 1 is a stepper (for example, a deep-UV stepper), 2 is a cell projection type electron beam exposure apparatus, 3 is a resist coating / developing apparatus, and 4 is each of the above-described apparatuses for processing a resist by an inline process. It is a device for transporting the space between the two under an environment where the atmosphere is controlled.

Next, the operation of the system configured as described above will be described. The wafer 5 coated with the resist by the resist coating / developing device 3 is moved by the device 4 to a stepper 1.
The reticle pattern is reduced on the entire surface of the wafer 5 and is sequentially exposed. When the exposure is completed, the wafer 5
Is transferred to the electron beam exposure apparatus 2 by the apparatus 4. After the alignment of the pattern to be subjected to the electron beam exposure on the light-exposed wafer is completed, each chip on the entire surface of the wafer 5 is sequentially exposed to the electron beam. At this time, in order to increase the throughput of electron beam exposure, the repetitive pattern is exposed by a cell projection method. Further, since the throughput of electron beam exposure is generally lower than the throughput of the stepper 1, the system is configured such that a plurality of electron beam apparatuses are arranged so that the processing capacity of the stepper is not limited by the processing capacity of the electron beam apparatus. And stepper 1
A plurality of electron beam exposure apparatuses 2 can process wafers 5 unloaded from the apparatus in parallel. This achieves both high resolution of electron beam exposure and high throughput of the stepper. After all the patterns have been drawn on all the chips on the entire surface of the wafer 5, the wafer 4 is returned to the coating / developing device 3 by the device 4 and developed to complete the pattern formation.
As a resist that can be used in such a system, a chemically amplified resist having high resolution and high sensitivity, for example, CGR248 (shipley) can be given. Since the performance of such a chemically amplified resist is easily deteriorated by various chemical substances in the air, the apparatus 4 may be installed and handled in an environment in which various apparatuses are installed and between various apparatuses under environmental control. I have to. This suppresses a change in pattern dimensions before and after exposure.

With such a lithography system, a device pattern including a fine pattern of the 0.1 μm rule can be formed at a high throughput.

FIG. 24 shows a chemically amplified negative resist CGR2 having a thickness of 0.65 μm using the above lithography system.
48 is an example of a fine pattern formed at 48. Deep-
A pattern of 0.25 μm or more was formed using a UV stepper, and a pattern of 0.25 μm or less was formed by electron beam exposure. The developing solution was a TMAH aqueous solution, and the developing condition was 0.14 N for 60 seconds. A pattern up to 0.1 μm was surely formed, and it was found that this lithography system had sufficient performance in terms of resolution.

FIG. 25 shows the result of a trial calculation of the throughput of the present lithography system. The exposure pattern used for the throughput estimation was 256 MD with the 0.15 μm rule.
This is the gate layer of the RAM. Throughput was estimated when 100 chips of this pattern were arranged on the entire surface of an 8-inch wafer and exposed. The sensitivity of the resist was 10 μC / cm ² . The electron beam lithography system used for the estimation was HL-80 manufactured by Hitachi.
0D. The performance of this device is described in [Reference 1] Nak
aya et al. , J. et al. Vac. Sci. Tec
hnol. , B8 (6), 1990, p1836. ,
[Reference 2] Y. Shoda et al. , J. et al. Vac.
Sci. Technol. , B9 (6), 1991, p
2940. [Reference 3] H .; Itoh et al. ,
J. Vac. Sci. Technol. , B10
(6), 1992, p 2799. Was referenced. In this estimation, one stepper and one electron beam exposure apparatus were used. Using cell projection (number of cells:
5) The throughput in the case of exposure only by electron beam exposure is 0.3 sheets / h, while the pattern of 0.25 μm rule or more is exposed by a Deep-UV stepper, and the pattern below it is subjected to cell projection. When exposed by an electron beam (number of cells: 5), a much higher value of 2.8 sheets / h is shown. Three electron beam exposure devices are arranged,
It is determined that if the wafers from the stepper can be processed in parallel, a sufficient throughput that can be used as a mass production tool can be secured.

As described above in detail, according to such a lithography system, there is provided a mass production system after optical lithography having both an excellent resolution exceeding the light of electron beam exposure and a throughput equivalent to that of an optical stepper. However, the following problems have become apparent in the above-described system. That is, the backscattered electrons generated at the time of electron beam exposure fall onto the light exposure patterns approaching the EB exposure area, and these patterns are exposed with an overdose, that is, an overexposure amount. For this reason, it has been found that when development is performed with developing conditions determined for an appropriate exposure amount, a pattern which has been exposed to light with an exposure amount exceeding the EB exposure region cannot be finished to predetermined dimensions.

FIG. 26 schematically shows how the light pattern is affected by the backscattered electrons when the light pattern and the EB pattern are formed close to each other (within the range of influence of the backscattered electrons). FIG. In this figure, it is assumed that the proximity effect correction has been performed on the EB pattern so that all the patterns can be satisfactorily resolved.

In FIG. 26 (a), the light pattern 10 is set to 0.1.
23 μm line and space, and 0.2 mm from the end.
An EB pattern having a width of 0.15 μm is formed at a distance of 15 μm, and an EB pattern having a width of 0.15 μm is further formed at a distance of 1.35 μm. The drawing density of the EB pattern group 11 (the ratio of the EB irradiation region to be implanted per unit area) is 10 %. In FIG. 26B, the 0.23 μm line and space optical pattern group 10 and the 0.15 μm line and space EB pattern group 11 are separated by 0.15 μm, and the writing density of the EB writing area is 50%. is there.

FIG. 27 shows experimentally how much the size of the closest optical pattern changes depending on the writing density of the EB pattern when the two patterns are arranged close to each other as described above. The result. The resist is the same as that shown in FIG. As can be seen from the figure, when the drawing density is 50%, the dimensions fluctuate by almost 50%. It is expected that even if the drawing density is 10%, the variation will be about 10%.

Since the backscattered electron distribution is regarded as a Gaussian distribution, it becomes smaller as the distance from the EB irradiation area increases. In the case of an electron beam having an acceleration voltage of 50 kV, the radius of the electron beam is about 30 μm on a silicon wafer. . Therefore, the fluctuation of the light pattern decreases as the distance from the EB pattern increases, and the fluctuation hardly occurs at a distance of about 30 μm.

[0018]

As can be seen from the above examination results, in the method of forming a pattern by separately striking the same resist with light and EB, how to avoid the influence of the backscattered electrons depends on the resolution. This is a major problem in forming patterns with good dimensional controllability.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems. It is an object of the present invention to prevent backscattered electrons generated at the time of electron beam exposure from being applied to a light exposure pattern close to an EB exposure area. To provide a pattern forming method that has good resolution and dimensional controllability of a pattern, has excellent resolution exceeding the light of electron beam exposure, and has the same throughput as an optical stepper by compensating for the effects caused by falling. It is in.

[0020]

[Means for Solving the Problems]

(A1) In a pattern forming method according to the present invention, a desired pattern is transferred to a photosensitive material formed on a substrate to be processed by light exposure using a reticle mask and charged beam exposure, and at least resolution of the light exposure is performed. Pattern transfer below the limit is a pattern forming method performed by charged beam exposure,
The light exposure pattern is formed by performing a predetermined correction so as to suppress (avoid) the influence of the backscattered electrons generated during the charged beam exposure.

(A2) In the above A1, the predetermined correction is performed by using a reticle mask used for light exposure.

(A3) In the above-described A1, the predetermined correction is performed by changing a pattern size of a reticle mask used for light exposure.

(A4) In the above A1 to A3, the reticle mask is manufactured by a charged beam exposure apparatus using proximity effect correction of irradiation amount control, and the incident irradiation amount map at the time of manufacturing the reticle mask is modulated. The above-mentioned predetermined correction is performed.

(A5) In the above A4, the modulation of the incident irradiation amount map is performed by judging the presence or absence of the influence of backscattered electrons at the time of irradiation of the charged beam.

(A6) In the above A4, the modulation of the incident irradiation amount map is represented by: dimensional conversion coefficient of wafer resist × backscattered electron amount = reduction ratio of stepper × dimensional conversion coefficient of reticle resist × incident electron amount. It is characterized in that it is performed by obtaining the amount of incident electrons from a relational expression.

(A7) In the above A4, the step of generating the data of the incident irradiation amount map includes a step of obtaining an appropriate irradiation amount map when drawing a light exposure pattern on a reticle substrate, and a step of converting the charged beam exposure pattern to a wafer. A step of obtaining a stored energy distribution map due to backscattered electrons when drawing on, a step of converting the stored energy distribution map due to backscattered electrons into a light exposure pattern dimension variation map, and a light exposure pattern dimension variation map A step of enlarging the reciprocal multiple of the reduction ratio of the light exposure stepper, a step of converting the enlarged light exposure pattern size variation distribution map into a reticle exposure irradiation amount conversion value distribution map, and Subtracting a reticle exposure dose conversion value distribution map.

(A8) In the above-mentioned A1 to A3, the predetermined correction is performed by modulating a figure size at the time of manufacturing a reticle mask.

(A9) In the above-mentioned A8, the modulation of the figure size is performed by judging the presence or absence of the influence of backscattered electrons at the time of irradiation of the charged beam.

(A10) In the above A8, the modulation of the figure size is calculated from a relational expression expressed by a dimension conversion coefficient of a wafer resist × amount of backscattered electrons = a reduction ratio of a stepper × a dimension conversion amount of a figure size of a reticle mask. It is characterized in that it is performed by obtaining the size conversion amount of the figure size.

(A11) In A8, the step of generating the map data of the figure size includes a step of obtaining a stored energy distribution map by backscattered electrons when the charged beam exposure pattern is drawn on the wafer; Converting the stored energy distribution map into a light exposure pattern dimensional variation distribution map, enlarging the light exposure pattern dimensional variation distribution map to the reciprocal multiple of the reduction ratio of the light exposure stepper, and drawing on the reticle substrate. Subtracting the enlarged light exposure pattern dimensional variation distribution map from the light exposure pattern to be exposed.

(A12) In A1, the predetermined correction is performed by correcting device pattern data.

(A13) In the above A1 or A12,
The predetermined correction is performed by separating the pattern region of the light exposure and the pattern region of the charged beam exposure by a predetermined distance except for a pattern region connecting these two regions.

(A14) In the above-mentioned A13, the fixed distance is a distance corresponding to a scattering diameter of backscattered electrons generated at the time of irradiation of the charged beam.

(A15) In the above-mentioned A13, the predetermined distance is determined by judging the degree of influence of backscattered electrons generated during irradiation of the charged beam.

(B1) The pattern forming method according to the present invention is a method for forming a desired pattern on a substrate by irradiating the substrate with a charged beam. It is characterized in that a pattern size is changed by obtaining a relationship with a dimension variation amount and changing an irradiation amount.

(B2) In the pattern forming method according to the present invention, a desired pattern transfer to a photosensitive material formed on a substrate to be processed is performed by light exposure using a reticle mask and charged beam exposure, and at least light exposure. In the pattern forming method in which the pattern transfer below the resolution limit of is performed by the charged beam exposure, a correction is used for the light exposure in which the light exposure pattern is corrected so as to suppress (avoid) the influence of the backscattered electrons generated during the charged beam exposure. The pattern size of the photomask is changed by changing the pattern size of the mask, and beforehand obtaining the relationship between the amount of change in the amount of irradiation of the charged beam used for drawing the photomask and the amount of change in the pattern size. It is characterized by making it.

(B3) In B1 or B2, the relationship between the variation in the irradiation amount of the charged beam and the variation in the pattern dimension is set to a desired value by changing the amount of blurring of the charged beam (beam edge resolution). It is characterized by the following.

(B4) In the pattern forming method according to any one of B1 to B3, an irradiation amount map is created in advance.

(B5) In the above B4, the step of preparing the irradiation amount map includes a step of obtaining an appropriate irradiation amount map when drawing the light exposure pattern on the photomask substrate, and the step of forming the charged beam exposure pattern on the wafer. Obtaining a stored energy distribution map due to backscattered electrons at the time of drawing; converting the stored energy distribution map due to backscattered electrons into a light exposure pattern size variation distribution map; A step of enlarging the reciprocal multiple of the reduction ratio of the exposure stepper, a step of converting the enlarged light exposure pattern dimensional variation distribution map into a photomask exposure dose conversion value distribution map, and Subtracting the mask exposure irradiation amount conversion value distribution map.

(B6) In the above B3, the amount of blur of the charged beam is changed by changing the focal position in the optical system for converging the charged beam.

(C1) In the pattern forming method according to the present invention, a desired pattern transfer to a photosensitive material formed on a substrate to be processed is performed by light exposure using a reticle mask and charged beam exposure, and at least light exposure In the pattern forming method in which the pattern transfer below the resolution limit of is performed by charged beam exposure, a desired pattern is formed using a high-contrast photosensitive organic photosensitive material having sensitivity to light and a charged beam. .

(C2) In C1, the photosensitive organic photosensitive material of the PH jump mechanism is used as the photosensitive organic photosensitive material.

The high-contrast photosensitive organic photosensitive material is, for example, as shown in FIG. 21, when the amount of irradiation on the photosensitive organic photosensitive material reaches a certain level or more, the residual film ratio sharply increases. Photosensitive material that decreases.

According to each of the above-mentioned inventions, it is possible to form a pattern with high resolution and high throughput by controlling the influence of backscattered electrons generated at the time of charged beam exposure.

[0045]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, an embodiment (first embodiment) according to invention A of the present application will be described.

FIGS. 1 and 2 are flow charts for explaining a pattern forming method according to a first specific example of the first embodiment. Since the lithography system used to carry out the pattern forming method is the same as that shown in FIGS. 22 and 23 described in the section of the related art, a detailed description will be omitted.

Next, the pattern forming method shown in FIGS. 1 and 2 will be described. Here, in order to simplify the description, the description will be made using the pattern shown in FIG.

First, the CAD data of the device pattern (a) is set at the boundary of the resolution dimension value which is looser than the limit resolution of the optical stepper used in the lithography system (this is referred to as
(Referred to as boundary dimension values), CAD data of the reticle for light exposure, and CAD data for electron beam exposure (b).
As a result, as shown in FIG. 3A, it is assumed that the light beam is divided into a light pattern (a thick part) and an EB pattern (a thin part).

Next, drawing data, that is, reticle drawing data (c) and EB drawing data (d) are created from the respective CAD data. When creating EB drawing data, proximity effect correction is performed so that all EB drawing patterns can be formed with good dimensions. Note that the proximity effect correction is performed by irradiation amount control. FIG. 3B,
FIG. 3C shows an irradiation amount map calculated at this time.
The energy distribution accumulated in the resist when each of the patterns in FIG. 3A is exposed is shown by the cross section AA in FIG.
It shows the case where it is seen by. The profile in FIG. 3C is the sum of the energies of the incident electrons and the backscattered electrons. Since both exposures are performed on the same resist, the actual stored energy distribution in the resist is as shown in FIG. The light pattern close to the EB pattern is affected by the backscattered electrons, and the stored energy is increased.
Therefore, in the batch development, the light pattern is formed differently from the desired pattern. The effect of the backscattered electrons on the resist dimensions can be easily estimated if the graph of FIG. 27 is obtained in advance.

It is determined whether or not the influence of the backscattered electrons on the light pattern can be neglected based on whether or not the dimensional variation thus obtained is equal to or less than an allowable value. For example, when it is determined that there is no influence, the separated optical pattern data is converted to create reticle drawing data, and the reticle is formed using, for example, an EB exposure apparatus for reticle drawing. Thereafter, a light pattern is formed by light exposure using the reticle, and an EB pattern is formed using the separately prepared EB drawing data, whereby a desired pattern can be formed.

On the other hand, when the dimensional variation of the light pattern due to the backscattered electrons exceeds the allowable value, the reticle manufacturing method is changed so that the light pattern has a desired pattern size under the influence of the backscattered electrons. The method will be described below. That is, the dimensional change caused by the backscattered electrons is absorbed by changing the reticle size in advance.

FIG. 3 (e) shows the distribution of incident electron irradiation amount on the section AA of the drawing data for producing the reticle by the EB exposure apparatus. However, in the reticle drawing data created based on this map, the backscattered electrons received during EB pattern exposure (the distribution is shown in FIG. 3 (f). Occurs, so E is calculated from the dose distribution shown in this map.
A reticle pattern drawing correction dose data is created by subtracting the dose received during the B exposure (e). However, in the manufacture of a reticle, it is necessary to determine the relationship between the amount of electron beam irradiation and the dimensions of the resist used on the reticle substrate due to differences such as different resists used and different backscattered electron coefficients of the substrate. FIG. 4A is an example of this, and is a diagram showing the relationship between the amount of electron irradiation and the resist dimensions in a resist used when manufacturing a reticle.

Since the dimension on the reticle is transferred onto the wafer by multiplying it by the reduction ratio of the stepper, conversely, in order to correct the dimensional variation on the wafer due to backscattered electrons with the reticle, the reciprocal of the reduction ratio is required. The size must be increased by the amount. Therefore, taking these factors into consideration, the dimension conversion coefficient of the wafer resist (nm / μC / cm ² )
× backscattered electron quantity (μC / cm ² ) = reduction ratio of stepper × dimensional conversion coefficient of reticle resist (nm / μC / c)
m ² ) × incident electron quantity (μC / cm ² ), the incident electron quantity is obtained, and this can be reduced from the irradiation quantity shown in FIG. 3 (e) obtained without considering the effect of backscattered electrons. It will be good. That is, FIG.
If the pattern is formed with the irradiation amount indicated by the dotted line in FIG. 2B, the reticle can be formed with a size in which the dimensional variation of the light pattern due to the backscattered electrons is considered in advance.

Here, for the sake of simplicity, the operation has been described with a single pattern. However, in accordance with this concept, dimensional correction is performed on all the light exposure patterns, and reticle drawing data is prepared to produce a reticle.

The step (e) of creating the corrected dose data used in the reticle fabrication described above can be performed by the steps shown in FIG. That is, a step S1 for obtaining an appropriate irradiation dose map when drawing a light exposure pattern on a reticle substrate, and a step S2 for obtaining a stored energy distribution map by backscattered electrons when drawing an electron beam exposure pattern on a wafer. A step S3 of converting the stored energy distribution map due to the backscattered electrons into a light exposure pattern dimension variation distribution map, and a step S4 of enlarging the light exposure pattern dimension variation distribution map to a reciprocal multiple of the reduction ratio of the light exposure stepper. Converting the enlarged light exposure pattern size variation amount distribution map into a reticle exposure irradiation amount conversion value distribution map, and subtracting the irradiation amount conversion value distribution map obtained in step S5 from the appropriate irradiation amount map obtained in step S1. Desirably, the method comprises the step S6. A desired pattern can be formed by exposing a light pattern by light exposure using the reticle and exposing the EB pattern using EB drawing data prepared separately.

FIGS. 5 and 6 are flowcharts for explaining a pattern forming method according to the second specific example of the first embodiment. Since the lithography system used to carry out the pattern forming method is the same as that shown in FIGS. 22 and 23 described in the section of the related art, a detailed description will be omitted.

Next, the pattern forming method shown in FIGS. 5 and 6 will be described. Here, in order to simplify the description, the description will be made using the pattern shown in FIG.

First, the device pattern CAD data is defined as (a) at the boundary of the resolution dimension value which is looser than the limit resolution of the optical stepper used in the lithography system (this is referred to as
(Referred to as boundary dimension values), CAD data of the reticle for light exposure, and CAD data for electron beam exposure (b).
As a result, as shown in FIG. 3A, it is assumed that the light beam is divided into a light pattern (a thick part) and an EB pattern (a thin part).

Next, drawing data, that is, reticle drawing data (c) and EB drawing data (d) are created from the respective CAD data. When creating EB drawing data, proximity effect correction is performed so that all EB drawing patterns can be formed with good dimensions. Note that the proximity effect correction is performed by irradiation amount control. FIG. 3B,
FIG. 3C shows an irradiation amount map calculated at this time.
The energy distribution accumulated in the resist when each of the patterns in FIG. 3A is exposed is shown by the cross section AA in FIG.
It shows the case where it is seen by. The profile in FIG. 3C is the sum of the energies of the incident electrons and the backscattered electrons. Since both exposures are performed on the same resist, the actual stored energy distribution in the resist is as shown in FIG. The light pattern close to the EB pattern is affected by the backscattered electrons, and the stored energy is increased.
Therefore, in the batch development, the light pattern is formed differently from the desired pattern. The effect of the backscattered electrons on the resist dimensions can be easily estimated if the graph of FIG. 27 is obtained in advance.

It is determined whether or not the influence of the backscattered electrons on the light pattern can be neglected based on whether or not the dimensional variation thus obtained is equal to or less than an allowable value. For example, when it is determined that there is no influence, the separated optical pattern data is converted to create reticle drawing data, and the reticle is formed using, for example, an EB exposure apparatus for reticle drawing. Thereafter, a light pattern is formed by light exposure using the reticle, and an EB pattern is formed using the separately prepared EB drawing data, whereby a desired pattern can be formed.

On the other hand, when the dimensional variation of the light pattern due to the backscattered electrons exceeds the allowable value, the reticle manufacturing method is changed so that the light pattern has a desired pattern size under the influence of the backscattered electrons. The method will be described below. That is, the dimensional change caused by the backscattered electrons is absorbed by changing the reticle size in advance.

FIG. 3E shows the distribution of incident electron irradiation amount on the section AA of the drawing data for producing the reticle by the EB exposure apparatus. However, in the reticle drawing data created based on this map, the backscattered electrons received during EB pattern exposure (the distribution is shown in FIG. 3 (f). Occurs, so E is calculated from the figure size shown in this map.
The corrected reticle pattern drawing data is created by subtracting the amount of the dimensional change received during the B exposure (e). However, in the manufacture of a reticle, it is necessary to determine the relationship between the amount of electron beam irradiation and the dimensions of the resist used on the reticle substrate due to differences such as different resists used and different backscattered electron coefficients of the substrate. FIG. 4A is an example of this, and is a diagram showing the relationship between the amount of electron irradiation and the resist dimensions in a resist used when manufacturing a reticle.

Since the dimension on the reticle is transferred onto the wafer by multiplying the reduction rate of the stepper, on the contrary, in order to correct the dimensional change on the wafer due to the backscattered electrons with the reticle, the reciprocal of the reduction rate is required. The size must be increased by the amount. Therefore, taking these factors into consideration, the dimension conversion coefficient of the wafer resist (nm / μC / cm ² )
The amount of backscattered electrons (μC / cm ² ) = the reduction ratio of the stepper × the amount of size conversion of the figure size of the reticle mask (nm). It is only necessary to reduce the figure size when it is not considered. A reticle can be formed with a size in which a dimensional change of a light pattern due to backscattered electrons is considered in advance.

Here, for the sake of simplicity, the operation has been described with a single pattern. However, in accordance with this concept, dimensional correction is performed on all light exposure patterns, and reticle drawing data is prepared to produce a reticle.

The step (e) of creating the corrected reticle pattern drawing data used in the reticle fabrication described above can be performed by the steps shown in FIG. That is,
Step S1 of obtaining a stored energy distribution map due to backscattered electrons when an electron beam exposure pattern is drawn on a wafer
1 and a step S of converting a stored energy distribution map due to backscattered electrons into a light exposure pattern size variation distribution map.
12, a step S13 of obtaining a dimensional variation distribution map obtained by enlarging the light exposure pattern dimensional variation distribution map to the reciprocal multiple of the reduction ratio of the light exposure stepper, and a step S13 from the light exposure pattern data to be drawn on the reticle substrate. And a step S14 of subtracting the obtained dimensional variation distribution map after the enlargement.

A desired pattern can be formed by exposing a light pattern by light exposure using the reticle and exposing the EB pattern using EB drawing data prepared separately.

FIG. 7 is a flowchart for explaining a pattern forming method according to a third specific example of the first embodiment. Since the lithography system used to carry out the pattern forming method is the same as that shown in FIGS. 22 and 23 described in the section of the related art, a detailed description will be omitted.

Next, the pattern forming method shown in FIG. 7 will be described. Here, for the sake of simplicity, FIG.
This will be described using the pattern shown in FIG.

First, the device pattern CAD data is defined as (a) at the boundary of the resolution dimension value which is looser than the limit resolution of the optical stepper used in the lithography system (this is referred to as
(Referred to as boundary dimension values), CAD data of the reticle for light exposure, and CAD data for electron beam exposure (b).

Next, drawing data, that is, reticle drawing data (c) and EB drawing data (d) are created from the respective CAD data. When creating EB drawing data, proximity effect correction is performed so that all EB drawing patterns can be formed with good dimensions. Note that the proximity effect correction is performed by irradiation amount control. It is examined from the EB drawing data how much the light pattern is affected by the backscattered electrons. A map 3D of the dimensional variation is obtained using the graph of FIG. 27, and it is determined whether or not the influence of the backscattered electrons on the light pattern can be ignored based on whether the variation is equal to or less than an allowable value. .

In the case where there is an influence, in the first and second specific examples, the problem was solved by correcting the size of the reticle. However, in this example, a simpler method without correcting the reticle size will be described.

That is, since the range of the backscattered electrons on the wafer is at most about 30 μm in radius, as shown in FIG. 8, in the region where the influence of the backscattered electrons exceeds the allowable value, the light exposure pattern is The device pattern data is re-created so as to be separated from the EB exposure pattern by a certain distance d, that is, a back scattering diameter or more. Using the improved data, it is confirmed again that the light pattern is not affected by the backscattered electrons at the time of EB exposure through the data division flow. However, if there is a connection line between both patterns, it is inevitably affected by the backscattered electrons, so that measures should be taken so that there is no problem even if a dimensional change occurs. Thereafter, the reticle drawing data is created by converting the improved light pattern data, and the reticle is created using, for example, an EB exposure apparatus for reticle drawing. Thereafter, a light pattern is exposed by light exposure using the reticle, and the EB pattern is exposed using EB drawing data prepared separately, thereby forming a desired pattern.

FIGS. 9 to 14 are views for explaining a specific example relating to the manufacture of a device using the lithography system of the present embodiment. Here, a description will be given of an example of processing a gate electrode of a MOSFET.

FIG. 9 is a diagram showing a basic structure of a semiconductor device manufactured by the method described in the first to third specific examples. Referring to FIG. 1, a polysilicon region 103 and a resist 104 are formed on a semiconductor substrate 101. Here, LOCOS (Local Oxida) is a selective oxidation method using Si ₃ N ₄ as an oxidation mask.
Although a step due to the STI (Shell) is shown in FIG.
Similarly, the present invention can be applied to a step in, for example, “ow trench isolation”. The resist 104 is
A resist sensitive to p-UV light and an electron beam, for example, V
N-HS, and its thickness is about 500 nm.

FIG. 10 is a plan view of a pattern of a gate electrode exposed to an electron beam for a fine pattern. Since the depth of focus is as deep as several μm or more in the exposure with the electron beam, the exposure tolerance for the step is incomparable to the exposure with the normal Deep-UV light. Therefore, it is possible to perform patterning with high precision without the resist being cut even at the stepped portion generated between the element region and the element isolation region.

Hereinafter, a process for forming a gate electrode will be described with reference to FIGS.

FIG. 11 is a perspective view of the state of FIG. 9 as viewed obliquely from the left hand side. On a semiconductor substrate 101, a gate oxide film region, an element isolation oxide film region 102, and a polysilicon region 103 ( 200 nm), resist region 104
(500 nm) are sequentially stacked. Here, the resist is negative, and only the gate is exposed by electron beam exposure, and the others are exposed by Deep-UV light using the reticle mask described in the first to third specific examples to form the entire pattern.

First, pattern exposure using Deep-UV light is performed. Next, a thin line portion is patterned using an electron beam exposure apparatus. FIG. 12 shows a state in which a latent image has been formed in the resist by light exposure and a state in which a latent image has been formed by electron beam exposure. Deep
-By using the reticle mask described in the first to third specific examples for pattern exposure with UV light, the same resist is irradiated with light and EB without considering backscattered electrons at the time of electron beam exposure. Patterns can be formed separately.

Next, development is performed to obtain the state shown in FIG.
Here, a TMAH aqueous solution, 0.27
I use the specified one. Thereafter, an RIE process is performed based on the resist pattern, and a shape of the gate electrode as shown in FIG. 14 can be obtained.

As described above, according to this embodiment, it is possible to form a pattern by separately striking the same resist with light and EB without considering backscattered electrons at the time of forming an EB exposure pattern. it can. As a result, for light exposure, by taking charge of the formation of a pattern looser than the critical resolution, reticle fabrication is simplified and the exposure process latitude is increased. In addition, for electron beam exposure, only the formation of a pattern having a size equal to or smaller than the boundary dimension value significantly reduces the electron beam exposure time.

Next, an embodiment (second embodiment) according to invention B of the present application will be described.
Will be described.

FIGS. 15 and 16 are flowcharts for explaining a pattern forming method according to the second embodiment.
Since the lithography system used to carry out the pattern forming method is the same as that shown in FIGS. 22 and 23 described in the section of the related art, a detailed description will be omitted.

Next, the pattern forming method shown in FIGS. 15 and 16 will be described. Here, in order to simplify the explanation, the explanation will be made using the pattern shown in FIG.

First, the device pattern CAD data is defined as (a) at the boundary of a resolution dimension value that is looser than the limit resolution of the optical stepper used in the lithography system (this is referred to as
(Referred to as boundary dimension values), CAD data of a photomask for light exposure, and CAD data for electron beam exposure (b). As a result, as shown in FIG. 17A, it is assumed that the light beam is divided into a light pattern (a thick part) and an EB pattern (a thin part).

Next, drawing data, that is, photomask drawing data (c) and EB drawing data (d) are created from the respective CAD data. When creating EB drawing data, proximity effect correction is performed so that all EB drawing patterns can be formed with good dimensions. Note that the proximity effect correction is performed by irradiation amount control. FIG.
(B) and (c) show the dose distribution map calculated at this time, and show the energy distribution accumulated in the resist when the pattern of FIG. FIG. The profile in FIG. 17C is the sum of the energies of the incident electrons and the backscattered electrons. Since both exposures are performed on the same resist, the distribution of stored energy in the actual resist is shown in FIG.
(D). The light pattern close to the EB pattern is affected by the backscattered electrons, and the stored energy is increased. Therefore, in the batch development, the light pattern is formed differently from the desired pattern. The effect of the backscattered electrons on the resist dimensions can be easily estimated if the graph of FIG. 27 is obtained in advance.

It is determined whether or not the influence of the backscattered electrons on the light pattern can be ignored based on whether or not the dimensional variation thus obtained is equal to or less than an allowable value. For example, when it is determined that there is no influence, the separated optical pattern data is converted to create photomask drawing data, and a photomask is formed using an EB exposure apparatus for drawing a photomask. Thereafter, a light pattern is formed by light exposure using this photomask, and E is formed using EB drawing data prepared separately.
A desired pattern can be formed by forming each of the B patterns.

On the other hand, when the dimensional variation of the light pattern due to the backscattered electrons exceeds the allowable value, the photomask manufacturing method is changed so that the light pattern has a desired pattern size under the influence of the backscattered electrons. The method will be described below. In short, the dimensional change caused by the backscattered electrons is absorbed by changing the photomask dimensions in advance.

FIG. 17E shows the distribution of incident electron irradiation amount at the section AA of the drawing data for producing a photomask by the EB exposure apparatus. However, in the photomask drawing data created based on this map, the size is determined by the backscattered electrons received at the time of EB pattern exposure (the distribution is shown in FIG. 17 (f), which is obtained at the time of the dose correction calculation). Since a fluctuation occurs, a photomask pattern drawing dose map is created by subtracting the dose received during the EB exposure from the dose distribution shown in this map (e). However, in the production of a photomask, it is necessary to determine the relationship between the amount of electron beam irradiation and the size of the resist used on the photomask substrate, due to differences in the resist used, the reflection electron coefficient of the substrate, and the like. FIG.
(A) is an example, and is a diagram showing a relationship between an electron irradiation amount and a resist dimension in a resist used at the time of manufacturing a photomask.

Since the dimension on the photomask is transferred onto the wafer by multiplying it by the reduction ratio of the stepper, conversely, in order to correct the dimensional variation on the wafer due to backscattered electrons with the photomask, the reduction ratio is required. The size must be increased by the reciprocal of. Therefore, taking these factors into consideration, the dimension conversion coefficient of the wafer resist (nm / μC / cm ² )
× backscattered electron quantity (μC / cm ² ) = reduction ratio of stepper × dimension conversion coefficient of photomask resist (nm / μC)
/ Cm ² ) × incident electron amount (μC / cm ² ). The incident electron amount is obtained from the relational expression of FIG. 17 (e) obtained without considering the effect of backscattered electrons. What we need to do is reduce it. That is, FIG.
8 (b), if the pattern is formed with the irradiation amount indicated by the solid line,
A photomask can be formed with a dimension in which a dimensional variation of a light pattern due to backscattered electrons is considered in advance.

The problem here is the size that must be changed by the photomask. Generally, the reduction ratio of a stepper is 1/4 or 1/5,
The pattern size of the photomask becomes four times or five times. That is, the size that must be changed is four times or five times the size on the wafer. In the present invention, the dimensions of the photomask are changed by changing the irradiation amount at the time of drawing the photomask. However, as shown in FIG. In some cases, the required amount cannot be achieved. The lower limit of the dose is determined by whether or not the resist pattern is resolved, and the upper limit is determined by the fact that the pattern size does not change even if the dose is increased. Therefore, as the core of the present invention,
The amount of dimensional change can be increased by increasing the amount of beam blur in photomask writing.

FIG. 18 (c) shows the relationship between the irradiation amount and the resist size when the beam blur (beam edge resolution) is 30 nm, 50 nm, and 100 nm. FIG. 18A shows a case where the blur amount is 30 nm. It can be seen that, as the blur amount increases, the dimensional change increases even with the same change in the irradiation amount, and as a result, the changeable range increases. As a result, the range that can be corrected is greatly expanded, and the pattern size of the photomask can be changed only by the irradiation amount modulation. That is, by setting the blur amount to 100 nm, a dimensional change amount of twice or more can be secured. However, if the beam is excessively blurred, a problem arises in the dimensional controllability of a drawing apparatus having a poor irradiation amount accuracy. Therefore, it is desirable to set the minimum necessary blur amount.

Here, for the sake of simplicity, the operation has been described using a single pattern. However, in accordance with this concept, dimension correction is performed for all light exposure patterns, and photomask drawing data is created to create a photomask.

The step (e) of creating the corrected dose data used in the fabrication of the photomask described above can be performed by the steps shown in FIG. That is, a step S1 for obtaining an appropriate irradiation dose map when drawing a light exposure pattern on a photomask substrate, and a step S2 for obtaining a stored energy distribution map due to backscattered electrons when drawing an electron beam exposure pattern on a wafer. A step S3 of converting the stored energy distribution map by the backscattered electrons into a light exposure pattern dimension variation distribution map, and a step S4 of enlarging the light exposure pattern dimension variation distribution map to a reciprocal multiple of the reduction ratio of the light exposure stepper. Converting the enlarged light exposure pattern size variation amount distribution map into a photomask exposure dose conversion value distribution map, and a dose conversion value distribution map obtained in step S5 from the appropriate dose map obtained in step S1. And a step S6 of subtracting A desired pattern can be formed by exposing a light pattern by light exposure using this photomask and exposing each of the EB patterns using EB drawing data prepared separately.

As described above, according to this example, it is possible to change the pattern size of the photomask only by modulating the irradiation dose, and to obtain an excellent resolution exceeding the light of electron beam exposure and a throughput equivalent to that of the optical stepper. A combined pattern forming method can be provided easily and with high precision.

FIGS. 9 to 14 are views for explaining a specific example relating to the manufacture of a device using the lithography system of the present embodiment. Here, a description will be given of an example of processing a gate electrode of a MOSFET.

FIG. 9 is a diagram showing a basic structure of a semiconductor device manufactured by the method described in this embodiment. Referring to FIG. 1, a polysilicon region 103 and a resist 104 are formed on a semiconductor substrate 101. Here, a step due to LOCOS, which is a selective oxidation method using Si ₃ N ₄ as an oxidation mask, is shown, but the present invention can be similarly applied to a step in another element isolation method such as STI. The resist 104 is a resist sensitive to Deep-UV light and an electron beam, for example, VN-HS,
Its thickness is about 500 nm.

FIG. 10 is a plan view of a pattern of a gate electrode exposed to an electron beam for a fine pattern. Since the depth of focus is as deep as several μm or more in the exposure with the electron beam, the exposure tolerance for the step is incomparable to the exposure with the normal Deep-UV light. Therefore, it is possible to perform patterning with high precision without the resist being cut even at the stepped portion generated between the element region and the element isolation region.

Hereinafter, a process of forming a gate electrode will be described with reference to FIGS.

FIG. 11 is a perspective view of the state of FIG. 9 as viewed obliquely from the left hand side. On the semiconductor substrate 101, a gate oxide film region, an element isolation oxide film region 102, and a polysilicon region 103 ( 200 nm), resist region 104
(500 nm) are sequentially stacked. Here, the resist is negative, and only the gate is exposed by electron beam exposure, and the others are exposed by Deep-UV light using the photomask described above to form an entire pattern.

First, pattern exposure using Deep-UV light is performed. Next, a thin line portion is patterned using an electron beam exposure apparatus. FIG. 12 shows a state in which a latent image has been formed in the resist by light exposure and a state in which a latent image has been formed by electron beam exposure. Deep
-By using the photomask described above for pattern exposure with UV light, the same resist can be divided into light and EB to form a pattern without considering backscattered electrons during electron beam exposure. Can be.

Next, development is performed to obtain the state shown in FIG.
Here, a TMAH aqueous solution, 0.27
I use the specified one. Thereafter, an RIE process is performed based on the resist pattern, and a shape of the gate electrode as shown in FIG. 14 can be obtained.

As described above in detail, according to the present embodiment, it is possible to easily and accurately correct the influence of the backscattered electrons at the time of forming the EB exposure pattern using the photomask pattern dimensions. Therefore, for light exposure,
By taking charge of the formation of a pattern that is looser than the limit resolution,
Photomask fabrication is simplified and the exposure process latitude is increased. In addition, for electron beam exposure, only the formation of a pattern having a size equal to or smaller than the boundary dimension value significantly reduces the electron beam exposure time.

Next, an embodiment (third embodiment) according to invention C of the present application will be described.
Will be described.

FIG. 19 is a flowchart for explaining a pattern forming method according to the third embodiment. Since the lithography system used to carry out the pattern forming method is the same as that shown in FIGS. 22 and 23 described in the section of the related art, a detailed description will be omitted.

It should be particularly noted in this embodiment that a high contrast resist, for example, a resist of a PH jump mechanism is used. The high contrast resist is, specifically, SEPR-4202P manufactured by Shin-Etsu Chemical Co., Ltd.
B.

Next, the pattern forming method of this embodiment will be described. Here, to simplify the explanation, FIG.
This will be described using the pattern shown by.

First, the device pattern CAD data is defined as (a) at the boundary of the resolution dimension value which is looser than the limit resolution of the optical stepper used in the lithography system (this is referred to as
(Boundary dimension value), reticle CAD data for light exposure, and CAD data for electron beam exposure (b). As a result, as shown in FIG. 20A, it is assumed that the light beam is divided into a light pattern (a thick part) and an EB pattern (a thin part).

Next, drawing data, that is, reticle drawing data (c) and EB drawing data (d) are created from the respective CAD data. When creating EB drawing data, proximity effect correction is performed so that all EB drawing patterns can be formed with good dimensions. Note that the proximity effect correction is performed by irradiation amount control.

FIGS. 20 (b) and 20 (c) show the distribution of the energy accumulated in the resist when the pattern of FIG. 20 (a) is exposed, respectively, in the irradiation amount map calculated at this time. 2) shows a cross section AA.
The profile in FIG. 20C is the sum of the energies of the incident electrons and the backscattered electrons. Since both exposures are performed on the same resist, the distribution of stored energy in the actual resist is as shown in FIG. The light pattern close to the EB pattern is affected by the backscattered electrons, and the stored energy is increased.

As shown in FIG. 21, a resist patterned by a mechanism called PH jump has a higher γ value (γ value: the remaining resist film thickness is 0% and 50% of the initial film thickness, respectively) as compared with the conventional resist. % Of the dose). In the light exposure, the non-irradiated portion of the pattern has a storage energy of about 0.3 times that of the irradiated portion due to leakage of the irradiated light, and the irradiation amount required for forming an optimal pattern of the light exposure is 10 times.
about 3 mJ / cm ² is accumulated energy quantity of the non-irradiated portion with respect mJ / cm ^2. Further, the amount of stored energy by which the light pattern adjacent to the EB pattern is raised by the influence of the backscattered electrons of the EB irradiation is about 5 mJ / cm ² when the drawing density of the EB pattern is 50%.
8 mJ / c for the non-irradiated part after light exposure and EB exposure
Approximately m ² of energy is stored. Accordingly, in the conventional resist, the change in the residual film ratio is remarkable, that is, the dimensional fluctuation is remarkable. On the other hand, in the case of a high contrast resist, the residual film ratio does not change, that is, the dimensional fluctuation does not occur.

As described above, since the influence of the backscattered electrons on the light pattern can be ignored by using the high-contrast resist, the separated light pattern data can be converted without considering the influence of the backscattered electrons to form the reticle. The drawing data is prepared, and a reticle is prepared using, for example, an EB exposure apparatus for reticle drawing. Then, a light pattern is prepared by light exposure using the reticle.
Each of the EB patterns is exposed using the drawing data,
A desired pattern can be formed.

According to the present embodiment, it is possible to form a pattern by separately striking the same resist with light and EB without considering backscattered electrons at the time of forming an EB exposure pattern. As a result, for light exposure, by taking charge of formation of a pattern that is looser than the critical resolution, reticle fabrication is simplified and the exposure process latitude is increased. In addition, for electron beam exposure, only the formation of a pattern having a size equal to or smaller than the boundary dimension value significantly reduces the electron beam exposure time.

The embodiments of the present invention have been described above, but the present invention is not limited to these embodiments. For example, the present invention is applicable to all patterning other than the gate electrode of a MOSFET, for example, an element region, a contact hole, a metal wiring layer, and the like. Also, MOSFET
In addition, the present invention can be applied to, for example, patterning of a fine region of a bipolar transistor.

In addition, the present invention can be implemented with various modifications without departing from the spirit thereof.

[0115]

According to the present invention, it is possible to form a pattern with high resolution and high throughput by controlling the influence of backscattered electrons generated at the time of charged beam exposure.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a pattern forming method according to a first specific example of the first embodiment of the present invention.

FIG. 2 is a flowchart showing a method for creating reticle drawing correction dose data according to a first specific example of the first embodiment of the present invention.

FIG. 3 is a diagram for explaining the operation of the pattern forming method according to the first embodiment of the present invention.

FIG. 4 is a diagram for explaining the operation of the pattern forming method according to the first embodiment of the present invention.

FIG. 5 is a flowchart showing a pattern forming method according to a second specific example of the first embodiment of the present invention.

FIG. 6 is a flowchart showing a method for creating corrected reticle drawing data according to a second specific example of the first embodiment of the present invention.

FIG. 7 is a flowchart showing a pattern forming method according to a third specific example of the first embodiment of the present invention.

FIG. 8 is a view showing an example of a pattern arrangement in a third specific example of the first embodiment of the present invention.

FIG. 9 is a sectional view showing a configuration example of a semiconductor device manufactured by the pattern forming method according to the first and second embodiments of the present invention.

FIG. 10 is a plan view showing a configuration example of a semiconductor device manufactured by the pattern forming method according to the first and second embodiments of the present invention.

FIG. 11 is a perspective view showing a manufacturing process of a semiconductor device manufactured by the pattern forming methods of the first and second embodiments of the present invention.

FIG. 12 is a perspective view showing a manufacturing process of a semiconductor device manufactured by the pattern forming methods of the first and second embodiments of the present invention.

FIG. 13 is a perspective view showing a manufacturing process of a semiconductor device manufactured by the pattern forming methods of the first and second embodiments of the present invention.

FIG. 14 is a perspective view showing a manufacturing process of a semiconductor device manufactured by the pattern forming methods of the first and second embodiments of the present invention.

FIG. 15 is a flowchart showing a pattern forming method according to a second embodiment of the present invention.

FIG. 16 is a flowchart illustrating a method of creating photomask drawing correction dose data according to the second embodiment of the present invention.

FIG. 17 is a view for explaining the operation of the pattern forming method according to the second embodiment of the present invention.

FIG. 18 is a view for explaining the operation of the pattern forming method according to the second embodiment of the present invention.

FIG. 19 is a flowchart showing a pattern forming method according to a third embodiment of the present invention.

FIG. 20 is a view for explaining the operation of the pattern forming method according to the third embodiment of the present invention.

FIG. 21 is a view for explaining the operation of the pattern forming method according to the third embodiment of the present invention.

FIG. 22 is a view showing the concept of a lithography system according to the present invention and the related art.

FIG. 23 is a diagram showing an example of a planar arrangement of a lithography system according to the present invention and the related art.

FIG. 24A is a micrograph showing an example of a resist pattern formed by a conventional pattern forming method, and FIG. 24B is a diagram showing a corresponding pattern.

FIG. 25 is a diagram showing the results of trial calculation of differences in throughput due to differences in lithography systems.

FIG. 26 is a diagram schematically showing the influence of backscattered electrons received from an electron beam exposure pattern in which a light exposure pattern is close.

FIG. 27 is a diagram showing a result of experimentally examining the influence of backscattered electrons received from an electron beam exposure pattern in which a light exposure pattern is close.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Deep-UV stepper 2 ... Cell projection type electron beam exposure apparatus 3 ... Resist coating / developing apparatus 4 ... Transport apparatus 5 ... Wafer

──────────────────────────────────────────────────続 き Continuation of front page (72) Inventor Waka Sugihara 33, Shinisogo-cho, Isogo-ku, Yokohama-shi, Kanagawa Prefecture Toshiba Production Technology Research Institute Co., Ltd. (56) References JP-A-2-87616 (JP, A) JP-A-59-117214 (JP, A) JP-A-58-42229 (JP, A) JP-A-5-217871 (JP, A) JP-A-10-90878 (JP, A) (58) Int.Cl. ⁷ , DB name) H01L 21/027

Claims (18)

(57) [Claims] 1. A desired pattern transfer to a photosensitive material formed on a substrate to be processed is performed by light exposure using a reticle mask and a charged beam exposure, and at least a pattern transfer below a resolution limit of the light exposure is charged. In the pattern forming method performed by beam exposure, the light exposure pattern is formed by performing a predetermined correction so as to suppress dimensional fluctuations received by backscattered electrons generated at the time of charged beam exposure ,
The predetermined correction is performed on a reticle mask used for light exposure.
A pattern forming method, wherein the method is performed by changing a pattern dimension . 2. The reticle mask is manufactured by a charged beam exposure apparatus using proximity effect correction of irradiation amount control, and the predetermined correction is performed by modulating an incident irradiation amount map at the time of manufacturing the reticle mask. The pattern forming method according to claim 1 , wherein: 3. The pattern forming method according to claim 2 , wherein the modulation of the incident irradiation amount map is performed by judging whether or not there is an influence of backscattered electrons during irradiation of the charged beam. 4. The modulation of the incident dose map is based on a relational expression expressed by the following formula: dimensional conversion coefficient of wafer resist × backscattered electron quantity = reduction ratio of light exposure apparatus × dimensional conversion coefficient of reticle resist × incident electron quantity. 3. The pattern forming method according to claim 2 , wherein the method is performed by obtaining an amount of incident electrons. 5. The method according to claim 1, wherein the step of generating the data of the incident irradiation amount map includes a step of obtaining an appropriate irradiation amount map when drawing a light exposure pattern on a reticle substrate, and a step of generating a charged beam exposure pattern on a wafer. Obtaining a stored energy distribution map due to backscattered electrons, converting the stored energy distribution map due to backscattered electrons into a light exposure pattern size variation distribution map, and converting the light exposure pattern size variation distribution map into a light exposure device . A step of enlarging to a reciprocal multiple of the reduction ratio, a step of converting the enlarged light exposure pattern size variation amount distribution map into a reticle exposure irradiation amount conversion value distribution map, and converting the reticle exposure irradiation amount from the appropriate irradiation amount map. 3. The pattern forming method according to claim 2 , comprising a step of subtracting a value distribution map. 6. The pattern forming method according to claim 1 , wherein the predetermined correction is performed by modulating a figure size at the time of manufacturing a reticle mask. 7. The pattern forming method according to claim 6 , wherein the modulation of the figure size is performed by judging the presence or absence of the influence of backscattered electrons at the time of irradiation of the charged beam. 8. The modulation of the figure size is obtained from a relational expression represented by a dimension conversion coefficient of a wafer resist × amount of backscattered electrons = a reduction ratio of a light exposure apparatus × a size conversion amount of a figure size of a reticle mask. 7. The pattern forming method according to claim 6 , wherein the method is performed by obtaining a dimension conversion amount. 9. The step of generating the map data of the figure size includes a step of obtaining a stored energy distribution map by backscattered electrons when a charged beam exposure pattern is drawn on a wafer, and a stored energy distribution map by backscattered electrons. From the light exposure pattern dimension variation distribution map, the step of enlarging the light exposure pattern dimension variation distribution map to the reciprocal multiple of the reduction ratio of the light exposure apparatus , and the step of converting the light exposure pattern to be drawn on the reticle substrate. 7. The pattern forming method according to claim 6 , further comprising a step of subtracting the light exposure pattern size variation distribution map after the enlargement. 10. A desired pattern transfer to a photosensitive material formed on a substrate to be processed is performed by light exposure using a reticle mask and charged beam exposure, and at least the pattern transfer below the resolution limit of light exposure is charged. in the pattern forming method of performing beam exposure, rows subjected to patterning a predetermined correction to the light exposure pattern to suppress the dimensional fluctuation experienced by backscattered electrons generated at the time of the charged particle beam exposure
Correct the device pattern data by the predetermined correction.
A pattern forming method characterized by performing the following . Wherein said predetermined correction, claims a pattern area of the optical exposure with a charged beam exposure pattern area, except the pattern region connecting these two regions, and carrying out by separating certain distance 11. The pattern forming method according to item 10 . 12. The pattern forming method according to claim 11 , wherein the fixed distance is a distance corresponding to a scattering diameter of backscattered electrons generated at the time of irradiation with a charged beam. 13. The pattern forming method according to claim 11 , wherein the predetermined distance is determined by judging a degree of influence of backscattered electrons generated at the time of irradiation of the charged beam. 14. A desired pattern transfer to a photosensitive material formed on a substrate to be processed is performed by light exposure using a reticle mask and a charged beam exposure, and at least the pattern transfer below the resolution limit of the light exposure is charged. In the pattern forming method performed by beam exposure, correction is performed by changing the pattern size of a photomask used for light exposure, so that the light exposure pattern is suppressed from being affected by backscattered electrons generated during charged beam exposure. A pattern forming method, comprising: obtaining a relationship between a variation amount of an irradiation amount of a charged beam used for mask drawing and a variation amount of a pattern dimension; and changing the irradiation amount to change a pattern dimension of the photomask. 15. The method of claim 1, characterized in that a desired value by varying the amount of blurring of the relationship charged beam irradiation amount variation amount and pattern size change amount of the charged beam
5. The pattern forming method according to 4 . 16. A pattern forming method according to claim 14, wherein an irradiation amount map is created in advance. 17. The step of creating the dose map,
A step of obtaining an appropriate dose map when drawing a light exposure pattern on a photomask substrate; a step of obtaining a stored energy distribution map by backscattered electrons when drawing a charged beam exposure pattern on a wafer; Converting the stored energy distribution map into a light exposure pattern size variation amount distribution map, enlarging the light exposure pattern size variation amount distribution map to a reciprocal multiple of the reduction ratio of the light exposure device , and expanding the light exposure. claims, wherein the step of converting the pattern dimension variation distribution map on a photomask exposure dose converted value distribution map, in that it consists of a step of the the appropriate dose map subtracting the photomask exposure irradiation equivalence distribution map Item 17. The pattern forming method according to Item 16 . 18. The pattern forming method according to claim 15 , wherein the amount of blur of the charged beam is changed by changing a focal position in an optical system that converges the charged beam.