JP2011186312A - Photomask and method for manufacturing semiconductor device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a photomask which suppresses the variance of an optical image due to defocusing with respect to a periodic edge pattern. <P>SOLUTION: The photomask has a periodic pattern 21A which is a pattern formed by repeatedly disposing a light shielding part 13A on a transparent substrate at a fixed pitch and is to be transferred onto a wafer by exposure with oblique incident illumination; and an SRAF part 210C which is a pattern formed by repeatedly disposing a light shielding part 130B in the periphery of the periodic pattern 21A at the same pitch as the light shielding part 13A and is to be not transferred onto a wafer by exposure with oblique incident illumination. A combination of transmittance, a phase, and a pattern width of a transmission part 120C transmitting exposure light, out of the SRAF part 210C, is adjusted so that the intensity of zero order diffracted light and the intensity of first order diffracted light which are radiated onto the wafer via the SRFA part 210C are equal to each other. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a photomask and a method for manufacturing a semiconductor device.

  In recent years, with the miniaturization of semiconductor devices, the distance between adjacent wirings is approaching the resolution limit of photolithography. In order to form a fine periodic pattern, two-beam interference is required, and the angle at which the exposure light is incident on the photomask is severely limited. For this reason, it is difficult to stably form patterns other than the periodic pattern. In addition, optical image variation (decrease in the depth of focus) due to defocusing at the time of exposing a pattern arranged at the end of the periodic pattern has become significant.

  Conventionally, variation of optical image due to defocusing by using multiple periodic patterns (SRAF: Sub-Resolution Assist Features (Pattern)) around the edge of the periodic pattern as the main pattern. However, this measure has become insufficient with the recent miniaturization of patterns.

  The photomask described in Patent Document 1 has a gate pattern that is exposed and transferred onto a transfer target surface, and a translucent pattern provided along the gate pattern. In addition, a transparent region is provided between the gate pattern and the translucent pattern. The translucent pattern has a phase difference between the light transmitted through the translucent pattern and the light transmitted through the transparent region within a predetermined range, and the transmissivity is the distance between the translucent pattern and the gate pattern. Regardless of the setting, it is set in a range that is not transferred to the transfer target surface.

  In addition, the photomask described in Patent Document 2 includes at least a semi-transparent area and a transparent area with respect to exposure light, and a phase difference of light passing through the translucent area and the transparent area is approximately 180 °. . A transparent auxiliary pattern having the same phase difference of transmitted light as that of the transparent region and a transparent pattern is arranged on the main pattern to be projected.

  However, in the former and the latter prior arts, the target of the optical proximity effect is high-order diffracted light having a third or higher order. For this reason, the optical proximity effect with respect to the periodic end pattern arranged at the periodic end of the periodic pattern cannot be controlled, and the periodic end pattern has a problem that variation in optical images due to defocusing becomes large.

JP 2003-302739 A JP-A-6-301192

  The present invention provides a photomask and a method for manufacturing a semiconductor device that suppress variation in an optical image due to defocusing with respect to a periodic end pattern arranged at a periodic end of the periodic pattern.

  According to one aspect of the present invention, the pattern is formed by repeatedly arranging the first light-shielding film on the transparent substrate at a constant pitch, and the pattern is transferred onto the substrate to be exposed by exposure with oblique incidence illumination. A pattern formed by repeatedly arranging the main pattern and the second light-shielding film around the main pattern at the same pitch as the pitch, and the pattern is transferred onto the exposed substrate by exposure with oblique incidence illumination. And the auxiliary pattern so that the diffracted light intensity of the zeroth-order diffracted light and the diffracted light intensity of the first-order diffracted light irradiated onto the substrate to be exposed through the auxiliary pattern are equal to each other. Combination of the transmittance of the transmissive part of the exposure light in which the second light shielding film is not disposed, the phase of the transmissive part, and the pattern width of the transmissive part. Photomask is provided, characterized in that it is an integer.

  According to another aspect of the present invention, the pattern is formed by repeatedly arranging the first light-shielding film on the transparent substrate at a constant pitch, and the pattern is transferred onto the exposed substrate by exposure with oblique incidence illumination. The pattern is formed by repeatedly arranging the main pattern and the second light-shielding film on the periphery of the main pattern at the same pitch as the pitch, and is transferred onto the exposed substrate by exposure with oblique incidence illumination. The second light-shielding film is arranged in both a direction in which the first light-shielding film is arranged and a direction perpendicular to the direction. None, and the auxiliary pattern so that the diffracted light intensity of the zeroth-order diffracted light and the diffracted light intensity of the first-order diffracted light irradiated onto the substrate to be exposed through the auxiliary pattern are equal Photomask wherein the pattern size of the transmission portion of the exposure light of which the second light-shielding film is not disposed is determined is provided.

  According to another aspect of the present invention, the pattern is formed by repeatedly arranging the first light-shielding film on the transparent substrate at a constant pitch, and the pattern is transferred onto the exposed substrate by exposure with oblique incidence illumination. The pattern is formed by repeatedly arranging the main pattern and the second light-shielding film on the periphery of the main pattern at the same pitch as the pitch, and is transferred onto the exposed substrate by exposure with oblique incidence illumination. A pattern transfer step of irradiating exposure light with the oblique incident illumination to the photomask having an auxiliary pattern that is not applied, and transferring the pattern onto the substrate to be exposed, the photomask including the auxiliary pattern The diffracted light intensity of the 0th-order diffracted light and the diffracted light intensity of the 1st-order diffracted light irradiated onto the substrate to be exposed via the The combination of the transmittance of the transmissive part of the exposure light in which the second light shielding film is not arranged in the auxiliary pattern, the phase of the transmissive part, and the pattern width of the transmissive part is adjusted. When the exposure light is irradiated by the oblique incidence illumination and the pattern is transferred onto the substrate to be exposed, the diffracted light intensity of the zeroth-order diffracted light and the diffracted light intensity of the first-order diffracted light are equal to the auxiliary pattern. A method for manufacturing a semiconductor device is provided, wherein exposure light is irradiated.

  According to the present invention, it is possible to suppress variations in the optical image due to defocusing with respect to the periodic end pattern arranged at the periodic end of the periodic pattern.

FIG. 1 is a view for explaining two-beam interference used in the exposure method according to the first embodiment. FIG. 2 is a diagram for explaining multibeam interference. FIG. 3 is a diagram illustrating an example of a cross-sectional configuration of the photomask according to the first embodiment. FIG. 4 is a diagram for explaining the transmittance of the photomask according to the first embodiment. FIG. 5 is a diagram for explaining the relationship between the pattern size of the SRAF portion and the diffracted light intensity. FIG. 6 is a diagram showing the image intensity when exposed using a photomask. FIG. 7 is a top view illustrating a configuration example of the photomask according to the second embodiment. FIG. 8 is a diagram for explaining the positional relationship between the primary light and the pupil when the SRAF unit is configured by a lattice pattern. FIG. 9 is a diagram for explaining the relationship between the pattern dimension and the diffracted light intensity when the SRAF portion is in a lattice shape.

  Exemplary embodiments of a photomask and a semiconductor device manufacturing method will be explained below in detail with reference to the accompanying drawings. Note that the present invention is not limited to these embodiments.

(First embodiment)
FIG. 1 is a diagram for explaining two-beam interference used in the exposure method according to the first embodiment, and FIG. 2 is a diagram for explaining multi-beam interference. In the present embodiment, variations in the optical image due to defocusing are suppressed with respect to the periodic end pattern located at the periodic end of the periodic pattern (main pattern) close to the resolution limit. In this embodiment, in order to form a fine pattern close to the resolution limit, dipole illumination (bipolar illumination) 1X with two-beam interference that is oblique incidence illumination is used.

  In the present embodiment, a case where dipole illumination 1X is used as a light source for two-beam interference will be described. However, as a light source for two-beam interference, fourth illumination (quadrupole illumination) or annular illumination may be used. Good.

  The dipole illumination 1X has two light sources 2X, and the emitted light emitted from these two points is sent to the photomask 4X side through the filter 3X. Thereby, the emitted light from the light source 2X is obliquely incident on the photomask 4X, and the 0th-order diffracted light (0th-order light L0x) of the emitted light and the first-order diffracted light (first-order light L1x) of the emitted light are on the lens 5X side. Sent to. In the lens 5X, the zero-order light L0x and the primary light L1x are condensed and irradiated onto the wafer (on the substrate to be exposed) 6X. As a result, the zero-order light L0x and the first-order light L1x interfere with each other, and the wafer 6X is subjected to pattern exposure by the interference light.

  As described above, in the two-beam interference, the exposure light (0th-order light L0x) incident on the photomask 4X and the diffracted light (first-order light L1x) due to the periodic pattern on the photomask 4X are condensed on the wafer 6X. An optical image (interference fringe) is formed.

  On the other hand, as shown in FIG. 2, in the case of multi-beam interference, circular illumination (monopolar illumination) 1Y that is direct incidence illumination is used. The circular illumination 1Y is illumination that exposes a resist applied on a substrate such as the wafer 6Y by multi-beam interference.

  The circular illumination 1Y has one light source 2Y, and the emitted light emitted from the light source 2Y is sent to the photomask 4Y side through the filter 3Y. Thereby, the light emitted from the light source 2Y is incident on the photomask 4Y, and the 0th order light to the Nth order diffracted light (Nth order light LNy) (N is a natural number of 2 or more) of the emitted light is on the lens 5Y side. Sent to. FIG. 2 shows a case where the 0th-order light L0y of the emitted light, the primary light L1y of the emitted light, the secondary light L2y of the emitted light, and the tertiary light L3y of the emitted light are sent to the lens 5Y side. In the lens 5Y, the zero-order light L0y to the third-order light L3y are condensed and irradiated onto the wafer 6Y. As a result, the zero-order light L0y to the third-order light L3y interfere with each other, and the wafer 6Y is subjected to pattern exposure by the interference light.

  In the present embodiment, the optical proximity effect on the periodic end pattern arranged at the periodic end of the periodic pattern is controlled by using the dipole illumination 1X that is oblique incidence illumination. The periodic pattern is a pattern in which line patterns having a predetermined line width and space patterns having a predetermined space width are alternately formed.

  FIG. 3 is a diagram illustrating an example of a cross-sectional configuration of the photomask according to the first embodiment. As shown in FIG. 3A, on the glass substrate (transparent substrate) 11 of the photomask 4X, a periodic pattern 21A in which line patterns are arranged at a predetermined period (pattern pitch), and the periodic pattern 21A An SRAF portion 21B in which SRAF patterns are arranged at a predetermined cycle is formed in a region adjacent to the end portion. The periodic pattern 21A is a main pattern transferred by exposure, and the SRAF portion 21B is an auxiliary pattern that is arranged at the same pitch as the periodic pattern 21A and is not transferred by exposure.

  The periodic pattern 21A is a pattern formed by repeatedly arranging the light shielding portions 13A as the first light shielding film on the glass substrate 11 at a constant pitch, and the pattern is transferred onto the wafer 6X by exposure with oblique incidence illumination. Pattern. Specifically, in the periodic pattern 21A, a light shielding portion 13A having a width La indicating a line pattern on the mask pattern and a transmitting portion 14A having a width Sa indicating a space pattern on the mask pattern have a constant pitch (La + Sa = P). The pattern is arranged alternately and repeatedly.

  The SRAF portion 21B is a pattern formed by repeatedly arranging the light shielding portion 13B, which is the second light shielding film, around the periodic pattern 21A at the same pitch as the periodic pattern 21A, and is exposed by oblique incidence illumination. Thus, the pattern is not transferred onto the wafer 6X. Specifically, the SRAF unit 21B has a constant pitch (Lb + Sb = P) between the light shielding unit 13B having a width Lb indicating a line pattern on the mask pattern and the transmission unit 14B having a width Sb indicating a space pattern on the mask pattern. The pattern is arranged alternately and repeatedly. Thus, the periodic pattern 21A and the SRAF portion 21B are formed with the same pitch P (La + Sa = Lb + Sb).

  Further, a transmission part 12B is formed between the SRAF part 21B and the glass substrate 11. In the present embodiment, the diffracted light intensity of the 0th-order light L0x irradiated to the wafer 6X and the primary light are adjusted by adjusting the transmittance of the transmissive portion 12B or a later-described transmissive portion 12C to a predetermined transmittance. Balance the diffracted light intensity of L1x. As described above, the two-beam interference can suppress variations due to defocusing of the optical image due to the balance of the light intensities of the zero-order light L0x and the first-order light L1x.

  Note that the SRAF pattern may be configured as the SRAF portion 21C as in the photomask 4Z shown in FIG. The SRAF unit 21C has a configuration in which a transmission unit 12C with adjusted transmittance is disposed at the position of the transmission unit 14B. In the case of the photomask 4Z, the light intensity of the 0th-order light L0x and the primary light L1x is adjusted to a desired magnitude by the transmission part 12C. Therefore, the transmission part 12B becomes unnecessary.

  By the way, when the SRAF unit 21B is arranged in an area adjacent to the end of the periodic pattern 21A arranged at a predetermined cycle, the SRAF unit 21B is arranged at the same cycle as the periodic pattern 21A. In order to prevent the SRAF portion 21B from being transferred onto the wafer, the light shielding portion 13B is made thicker than the light shielding portion 13A of the periodic pattern 21A, thereby reducing the overall optical image intensity.

  However, when the optical characteristic of the light shielding part is a pattern with a constant period, the diffraction efficiency is determined by the size of the light shielding part (or transmission part). For this reason, when the dimension of the light-shielding part (or transmission part) that minimizes the variation due to defocus in the periodic pattern area on the photomask is selected, the balance between the zero-order light L0x and the primary light L1x in the SRAF area is It will surely collapse compared to the periodic pattern. This is because the pattern dimensions are different between the light shielding part of the periodic pattern and the light shielding part of the SRAF region.

  Therefore, in the present embodiment, the photomask is configured so that the balance between the 0th-order light L0x and the primary light L1x is maintained even in the region where the SRAF is provided. For example, a transmission part 12B with adjusted transmittance is formed below the SRAF part 21B. In this way, the transmittance of the region where the SRAF parts 21B and 21C are arranged is adjusted as compared with the periodic pattern 21A.

  Note that the photomasks 4X and 4Z used in the present embodiment may be binary masks or halftone masks. When the photomasks 4X and 4Z are binary masks, the light shielding portions 13A and 13B are made of a light shielding film such as chromium in order to block the exposure light in the light shielding portions 13A and 13B. Further, when the photomasks 4X and 4Z are halftone masks, the light shielding portions 13A and 13B are made of a chromium oxide film or the like in order to block a part of the exposure light and allow the light shielding portions 13A and 13B to transmit a part thereof. It consists of.

  FIG. 4 is a diagram for explaining the transmittance of the photomask according to the first embodiment. FIG. 4 shows a top view of the photomask. Photomasks 40Z and 41Z shown in FIG. 4 have the same configuration as the photomask 4Z shown in FIG. Note that the photomasks 40Z and 41Z may have the same configuration as the photomask 4X shown in FIG. The photomask 40Z is a first example of the photomask according to this embodiment, and the photomask 41Z is a second example of the photomask according to this embodiment.

  SRAF portions 210C and 211C of the photomasks 40Z and 41Z correspond to the SRAF portion 21C shown in FIG. The light shielding units 130B and 131B of the SRAF units 210C and 211C correspond to the light shielding unit 13B, and the transmission units 120C and 121C correspond to the transmission unit 12C. Here, the case where the light shielding part 13A and the light shielding parts 130B and 131B are halftone parts will be described.

  The array portion that is the periodic pattern 21A is exposed so that the diffracted light intensity of the zero-order light L0x and the diffracted light intensity of the first-order light L1x are equal. The SRAF units 210C and 211C are configured with a pitch P like the periodic pattern 21A and the periodic patterns 21B and 21C. The transmissive portion 120C of the photomask 40Z has a pattern width of 31% of the pitch P, and the transmissive portion 121C of the photomask 41Z has a pattern width of 39% of the pitch P.

  In the photomask 40Z, the transmittances of the light shielding unit 130B and the transmission unit 120C are adjusted so that the diffracted light intensity of the 0th-order light L0x and the diffracted light intensity of the primary light L1x are equal. In the photomask 41Z, the transmittance of the transmission part 121C is adjusted so that the diffracted light intensity of the 0th-order light L0x and the diffracted light intensity of the first-order light L1x are equal.

  FIG. 5 is a diagram for explaining the relationship between the pattern size of the SRAF portion and the diffracted light intensity. 5A shows the relationship between the pattern size of the SRAF unit 210C having a predetermined phase and the diffracted light intensity, and FIG. 5B shows the pattern size of the SRAF unit 211C having a predetermined phase. And the diffracted light intensity. 5A and 5B, the horizontal axis represents the mask value (pattern width of the transmission part), and the vertical axis represents the diffracted light intensity at the SRAF parts 210C and 211C. 5A and 5B, the diffracted light intensity of the array portion 21A is not shown, and the pattern width (mask value) of the transmission portion 14A of the array portion 21A is indicated by an arrow.

  FIG. 5A shows a diffracted light intensity curve D0 of the 0th-order light L0x and a diffracted light intensity curve D1 of the first-order light L1x in the SRAF unit 210C. The diffracted light intensity curves D0 and D1 are used to calculate information on the structure of the dipole illumination 1X, information on the structure of the lens 5X, the pitch P, the transmittance, the phase, etc. of the SRAF unit 210C as exposure conditions. It is a characteristic calculated by inputting.

  Therefore, the diffracted light intensity curves D0 and D1 have characteristics corresponding to the transmittance of the SRAF unit 210C. In the present embodiment, of the diffracted light intensity curves D0 and D1, the diffracted light intensity of the 0th-order light L0x and the diffracted light intensity of the first-order light L1x with respect to the pattern width of the transmission part 120C (31% of the pitch P). The transmittances of the light shielding part 130B and the transmissive part 120C are calculated so as to be equal to each other. In other words, the diffracted light intensity curve D0 of the 0th-order light L0x and the diffracted light intensity curve D1 of the 1st-order light L1x overlap with each other with the mask value of the SRAF unit 210C illustrated by the arrow (pattern width of the transmitting part 120C). The transmittances of the light shielding unit 130B and the transmission unit 120C are calculated in advance.

  For example, when the transmittance of the light shielding part 13A of the periodic pattern 21A is 6%, a material whose transmittance of the light shielding part 130B of the SRAF part 210C is 12% is used for the light shielding part 130B. Further, the transmittance of the transmissive portion 14A of the periodic pattern 21A is 100%, whereas the transmittance of the transmissive portion 120C of the SRAF portion 210C is set to 50%. Specifically, for the SRAF unit 210C having the transmission unit 120C, the transmittance of the light shielding unit 130B and the transmission unit 120C so that the diffracted light intensities of the 0th-order light L0x and the first-order light L1x are equal to each other is determined. The light shielding unit 130B and the transmission unit 120C are prepared using a member having the determined transmittance.

  FIG. 5B shows a diffracted light intensity curve E0 of the 0th-order light L0x and a diffracted light intensity curve E1 of the first-order light L1x in the SRAF unit 210C. The diffracted light intensity curves E0 and E1 are used to calculate information on the structure of the dipole illumination 1X, information on the structure of the lens 5X, the pitch P, the transmittance, the phase, and the like of the SRAF unit 211C as exposure conditions. Calculated by inputting.

  Further, the diffracted light intensity curves E0 and E1 have characteristics corresponding to the transmittance of the SRAF portion 211C. In the present embodiment, of the diffracted light intensity curves E0 and E1, the diffracted light intensity of the 0th-order light L0x and the diffracted light intensity of the first-order light L1x with respect to the pattern width of the transmission part 121C (39% of the pitch P). The transmittance of the transmissive part 121C is calculated so as to be equal to each other. In other words, the diffracted light intensity curve E0 of the zeroth-order light L0x and the diffracted light intensity curve E1 of the first-order light L1x overlap with each other with the mask value of the SRAF portion 211C illustrated by the arrow (pattern width of the transmissive portion 121C). The transmittance of the transmission part 121C is calculated in advance.

  For example, when the transmittance of the light shielding unit 13A of the periodic pattern 21A is 6%, the transmittance of the light shielding unit 131B of the SRAF unit 211C is also set to 6%. Further, the transmittance of the transmissive portion 14A of the periodic pattern 21A is 100%, whereas the transmittance of the transmissive portion 121C of the SRAF portion 211C is set to 70%. Specifically, the transmittance of the transmissive part 121C is determined such that the diffracted light intensities of the zero-order light L0x and the first-order light L1x are equal to the SRAF part 211C having the transmissive part 121C. The transmission part 121C is produced using the member.

  As described above, in this embodiment, when the SRAF units 210C and 211C are exposed, the SRAF units 210C and 211C are not transferred, and the diffraction efficiencies of the 0th-order light L0x and the first-order light L1x on the wafer 6X are high. The transmittance and pattern width of the SRAF units 210C and 211C are determined so as to be equal.

  The relationship between the pattern dimensions of the SRAF units 210C and 211C and the diffracted light intensity shown in FIG. 5 is an example when the SRAF units 210C and 211C have a predetermined phase. Therefore, the diffracted light intensity curves D0 and D1 shown in FIG. 5 indicate curves corresponding to the phases of the SRAF units 210C and 211C. Therefore, when the photomasks 40Z and 41Z are manufactured, the diffracted light intensities of the zero-order light L0x and the first-order light L1x may be made equal by adjusting the transmittance and pattern width of the SRAF units 210C and 211C. Then, the diffracted light intensities of the zero-order light L0x and the first-order light L1x may be made equal by adjusting the phases of the SRAF units 210C and 211C. Further, the diffracted light intensities of the zero-order light L0x and the first-order light L1x may be made equal by adjusting the transmittance, phase, and pattern width of the SRAF units 210C and 211C in combination.

  In this case, in the photomask 40Z, the transmittance, phase, and pattern of the light shielding unit 130B with respect to the exposure light are set such that the diffraction efficiency of the exposure light in the SRAF unit 210C is equal between the 0th order light L0x and the first order light L1x. A combination of the width, the transmittance of the transmission part 120C with respect to the exposure light, the phase, and the pattern width is determined. Further, in the photomask 41Z, the transmittance, phase, and pattern width of the transmissive part 121C with respect to the exposure light so that the diffraction efficiency of the exposure light in the SRAF part 211C is equal between the 0th order light L0x and the primary light L1x. And the combination is determined.

  The photomasks 40Z and 41Z are manufactured so that the pitch P satisfies, for example, k1 value ≦ 0.28. In other words, the photomasks 40Z and 41Z are manufactured so that the value of the pitch P of the photomasks 40Z and 41Z is equal to or less than the value calculated by 0.56 × (exposure light wavelength) ÷ (numerical aperture).

  Next, the result of simulating an optical image using the photomask 41Z having the diffracted light intensity characteristic shown in FIG. 5B will be described. FIG. 6 is a diagram showing the image intensity when exposed using a photomask.

  The horizontal axis in FIG. 6 indicates the relative position of the mask pattern (SRAF portion, periodic pattern), and the vertical axis indicates the image intensity. 6A shows the image intensity of the conventional photomask, and FIG. 6B shows the image intensity of the photomask 41Z. Also, the image intensity characteristics Y1 and Y3 shown in FIG. 6 are the image intensities when exposed at the first focus value, and the image intensity characteristics Y2 and Y4 are the image intensities when exposed at the second focus value. is there. The second focus value is, for example, a best focus value, and the first focus value is a focus value when there is a predetermined defocus from the best focus.

  The conventional photomask here is different from the photomask 41 </ b> Z in that the transmittance of the SRAF unit 30 is made to be the same as the transmittance of the periodic pattern 300. As shown in FIG. 6A, in the case of a conventional photomask, the image intensity corresponding to the line pattern at the end of the periodic pattern 300 is large between the first focus value and the second focus value. There is a great difference. In other words, the image intensity varies greatly depending on the focus. As described above, in the conventional photomask, the focus depth is reduced and the focus margin is reduced due to defocusing.

  On the other hand, as shown in FIG. 6B, in the case of the photomask 41Z of the present embodiment, the line pattern at the end of the periodic pattern 21A (the periodic end pattern arranged at the periodic end of the periodic pattern 21A). ) Is small between the first focus value and the second focus value. In other words, the image intensity characteristic due to focus is flattened, and variation in image intensity due to focus is suppressed. In addition, the image intensity of the SRAF pattern arranged at the endmost portion closest to the periodic pattern 21A in the SRAF portion 211C is small. Therefore, the SRAF portion 211C is difficult to be transferred onto the wafer 6X. As described above, in the photomask 41Z, it is possible to increase the focus margin without decreasing the depth of focus even if defocusing occurs.

  The photomask 4X and the photomask 4Z are produced for each layer of the wafer process, for example. Then, a semiconductor device (semiconductor integrated circuit) is manufactured using the manufactured photomasks 4X and 4Z. Specifically, the resist-coated wafer 6X is exposed using the photomask 4X or the photomask 4Z, and then the wafer 6X is developed to form a resist pattern on the wafer 6X. Then, the wafer 6X is etched using the resist pattern as a mask. As a result, an actual pattern corresponding to the mask pattern (periodic pattern 21A, etc.) of the photomasks 4X and 4Z is formed on the wafer 6X. When manufacturing a semiconductor device, the above-described production of the photomask 4X and the photomask 4Z, exposure processing, development processing, etching processing, and the like are repeated for each layer.

  As described above, according to the first embodiment, the transmittances of the SRAF units 210C and 211C are balanced so that the diffracted light intensities of the zero-order light L0x and the first-order light L1x are the same. On the other hand, the focus characteristic can be improved. Thereby, it is possible to suppress variation in the optical image due to defocusing with respect to the periodic end pattern, and as a result, it is possible to reduce the dimensional change of the resist pattern formed after exposure.

  Further, since the photomasks 40Z and 41Z are fabricated so that the pitch P of the photomasks 40Z and 41Z is equal to or less than the value calculated by 0.56 × (exposure light wavelength) ÷ (numerical aperture), the photomasks 40Z and 41Z are formed after exposure. Even if the resist pattern to be formed is a fine resist pattern, the dimensional change of the resist pattern can be reduced.

(Second Embodiment)
Next, a second embodiment of the present invention will be described with reference to FIGS. In the second embodiment, the SRAF part is configured by a lattice pattern, so that the SRAF part is not transferred and the diffracted light intensities of the zero-order light L0x and the first-order light L1x in the SRAF part are made equal.

  FIG. 7 is a top view illustrating a configuration example of the photomask according to the second embodiment. Among the constituent elements in FIG. 7, constituent elements that achieve the same functions as those of the photomasks 4X and 4Z of the first embodiment shown in FIG. 3 are given the same numbers, and redundant descriptions are omitted.

  The photomask 42Z of the present embodiment has a periodic pattern 21A and an SRAF unit 212D as mask patterns. The SRAF part 212D of the photomask 42Z corresponds to the SRAF parts 21B and 21C shown in FIG. The light shielding part 132D of the SRAF part 212D corresponds to the light shielding part 13B, and the transmission part 122D corresponds to the transmission parts 12B and 12C.

  In the photomask 42Z, the SRAF portion 212D is configured with a pattern layout having a shape different from that of the periodic pattern 21A. Specifically, the SRAF unit 212D is configured such that the light shielding unit 132D of the SRAF unit 212D is configured by a grid pattern, and the gap portion (island region) of the grid pattern is the transmission unit 122D. For example, the light shielding part 132D is arranged in both the direction in which the light shielding part 13A is arranged (horizontal direction in FIG. 7) and the direction perpendicular to the direction (vertical direction in FIG. 7) (longitudinal direction of the light shielding part 13A). It has a lattice shape. Further, in the SRAF unit 212D, the region of the transmission unit 122D is determined so that the zero-order light L0x and the first-order light L1x irradiated onto the wafer 6X via the SRAF unit 212D are equal. This realizes the SRAF unit 212D to be SRAF and balances the diffracted light intensities of the zero-order light L0x and the first-order light L1x of the SRAF unit 212D.

  In the case of the SRAF units 210C and 211C described in the first embodiment, the primary light transmitted through the photomasks 4X and 4Z is divided into two directions and sent to the lens 5X side. Of the branched primary light, one primary light is sent onto the pupil of the lens 5X, and the other primary light is sent outside the pupil of the lens 5X. Thereby, the primary light transmitted on the pupil of the lens 5X among the primary light is irradiated on the wafer 6X, and the primary light transmitted outside the pupil of the lens 5X is not irradiated on the wafer 6X. Accordingly, half of the primary light transmitted through the photomasks 4X and 4Z is irradiated onto the wafer 6X.

  On the other hand, when the SRAF unit 212D is configured in a lattice pattern as in the present embodiment, the primary light transmitted through the photomask 42Z is divided into four directions and sent to the lens 5X side. FIG. 8 is a diagram for explaining the positional relationship between the primary light and the pupil when the SRAF unit is configured by a lattice pattern. FIG. 8 shows the positional relationship between the primary light and the pupil when viewed from the upper surface side of the lens 5X.

  When the SRAF unit 212D is configured with a lattice pattern, the primary light transmitted through the SRAF unit 212D of the photomask 42Z is divided into four directions and sent to the lens 5X side. And each primary light is sent to the position 32A-32D in the surface where the lens 5X is arrange | positioned. Of the positions 32A to 32D, the position 32A is a position in the pupil 33 of the lens 5X, and the positions 32B to 32D are positions outside the pupil 33 of the lens 5X. Therefore, one of the branched primary lights is sent onto the pupil of the lens 5X, and the other three primary lights are sent out of the pupil of the lens 5X. Thereby, the primary light transmitted on the pupil of the lens 5X among the primary light is irradiated on the wafer 6X, and the primary light transmitted outside the pupil of the lens 5X is not irradiated on the wafer 6X. Therefore, 1/4 of the primary light transmitted through the photomask 42Z is irradiated onto the wafer 6.

  Further, the 0th-order light L0x is sent to the position 31 in the pupil 33 and irradiated onto the wafer 6. Therefore, in the present embodiment, the SRAF unit is configured so that the ¼ primary light of the primary light L1x transmitted through the SRAF unit 212D and the 0th order light L0x transmitted through the SRAF unit 212D have the same diffracted light intensity. A 212D mask pattern is configured. As a result, the wafer 6X is exposed with 2/5 light transmitted through the SRAF unit 212D. In the case of the SRAF units 210C and 211C described in the first embodiment, the wafer 6X is exposed with 2/3 of the light transmitted through the SRAF units 210C and 211C. As described above, in the present embodiment, the intensity of the light applied to the wafer 6X from the SRAF unit 212D is reduced and becomes smaller than that of the SRAF units 210C and 211C, so that the pattern of the SRAF unit 212D is transferred onto the wafer. It will not be done. As a result, the SRAF unit 212D is changed to SRAF.

  FIG. 9 is a diagram for explaining the relationship between the pattern dimension and the diffracted light intensity when the SRAF portion is in a lattice shape. FIG. 9 shows the relationship between the pattern size of the SRAF unit 212D having a predetermined phase and the diffracted light intensity. In FIG. 9, the horizontal axis represents the mask value (pattern width of the transmissive portion), and the vertical axis represents the diffracted light intensity at the SRAF portion 212D. In FIG. 9, the diffracted light intensity of the array portion 21A is not shown, and the pattern width (mask value) of the transmission portion 14A of the array portion 21A is indicated by an arrow.

  Zero-order light L0x and primary light when exposure light is irradiated with dipole illumination 1X on a lattice-like pattern formed by arranging line patterns in the vertical and horizontal directions at the same pitch P as the periodic pattern 21A L1x diffracted light intensity characteristics are calculated in advance. FIG. 9 shows the diffracted light intensity curve F0 of the zeroth-order light L0x and the diffracted light intensity curve F1 of the first-order light L1x in the SRAF unit 212D as the calculated diffracted light intensity characteristics.

  In the present embodiment, the mask value at which the diffracted light intensity curve F0 and the diffracted light intensity curve F1 of the primary light overlap is determined as the pattern size of the transmission part 122D. The diffracted light intensity curves F0 and F1 shown in FIG. 9 input information regarding the structure of the dipole illumination 1X, information regarding the structure of the lens 5X, the pattern pitch of the SRAF unit 210C, and the like as exposure conditions to a simulation apparatus that calculates the diffracted light intensity. It is calculated by doing. The transmissive part 122D of the SRAF part 212D has a square shape, for example, and the pattern dimension of one side thereof is, for example, 82% of the pitch P.

  Thus, in the photomask 42Z, the dimension of the transmission part 122D is adjusted so that the diffracted light intensity of the 0th-order light L0x and the diffracted light intensity of the first-order light L1x are equal. In other words, the dimension of the transmission part 122D is adjusted so that the diffracted light intensity curve F0 of the 0th-order light L0x and the diffracted light intensity curve F1 of the first-order light L1x overlap. Then, based on the dimensions of the transmission part 122D, the pattern layout of the SRAF part 212D is created, and the photomask 42Z is produced.

  As described above, according to the second embodiment, since the pattern layout shape of the SRAF unit 212D is made to be a lattice, the diffracted light intensities of the zero-order light L0x and the first-order light L1x are balanced. The focus characteristic can be improved with respect to the periodic end pattern. Thereby, it is possible to suppress variation in the optical image due to defocusing with respect to the periodic end pattern, and as a result, it is possible to reduce the dimensional change of the resist pattern formed after exposure.

  1X dipole illumination, 4X, 4Z, 40Z, 41Z, 42Z photomask, 12B, 12C, 14A, 14B, 120C, 121C, 122D transmission part, 13A, 13B, 130B, 131B, 132D light shielding part, 21A periodic pattern, 21B, 21C, 210C, 211C, 212D SRAF part, L0x 0th order light, L1x 1st order light.

Claims (5)

A main pattern which is a pattern formed by repeatedly arranging the first light-shielding film on the transparent substrate at a constant pitch, and which is transferred onto the substrate to be exposed by exposure with oblique incidence illumination;
A pattern formed by repeatedly arranging the second light-shielding film around the main pattern at the same pitch as the pitch, and the pattern is not transferred onto the substrate to be exposed by exposure with oblique incidence illumination. With patterns,
Have
The second light-shielding film is arranged in the auxiliary pattern so that the diffracted light intensity of the 0th-order diffracted light and the diffracted light intensity of the 1st-order diffracted light irradiated onto the substrate to be exposed through the auxiliary pattern are equal. A combination of the transmittance of the transmissive part of the exposure light that is not exposed, the phase of the transmissive part, and the pattern width of the transmissive part is adjusted.   The second light-shielding film is arranged in the auxiliary pattern so that the diffracted light intensity of the 0th-order diffracted light and the diffracted light intensity of the 1st-order diffracted light irradiated onto the substrate to be exposed through the auxiliary pattern are equal. The combination of the transmittance of the non-exposure light transmitting portion, the phase of the transmitting portion, the pattern width of the transmitting portion, the transmittance of the first light shielding film, and the phase of the first light shielding film The photomask according to claim 1, wherein the photomask is adjusted.   3. The photomask according to claim 1, wherein a dimension of the pitch is equal to or less than a value calculated by 0.56 × exposure light wavelength ÷ numerical aperture. A main pattern which is a pattern formed by repeatedly arranging the first light-shielding film on the transparent substrate at a constant pitch, and which is transferred onto the substrate to be exposed by exposure with oblique incidence illumination;
A pattern formed by repeatedly arranging the second light-shielding film around the main pattern at the same pitch as the pitch, and the pattern is not transferred onto the substrate to be exposed by exposure with oblique incidence illumination. With patterns,
Have
The second light-shielding film is arranged in both a direction in which the first light-shielding film is arranged and a direction perpendicular to the direction to form a lattice shape, and the substrate to be exposed through the auxiliary pattern The pattern size of the transmission part of the exposure light in which the second light-shielding film is not arranged in the auxiliary pattern so that the diffracted light intensity of the 0th-order diffracted light and the diffracted light intensity of the first-order diffracted light irradiated on the top are equal. A photomask characterized by having been determined. A main pattern which is a pattern formed by repeatedly arranging the first light-shielding film on the transparent substrate at a constant pitch, and which is transferred onto the substrate to be exposed by exposure with oblique incidence illumination;
A pattern formed by repeatedly arranging the second light-shielding film around the main pattern at the same pitch as the pitch, and the pattern is not transferred onto the substrate to be exposed by exposure with oblique incidence illumination. With patterns,
A pattern transfer step of irradiating exposure light with the oblique incident illumination to the photomask having a pattern transfer onto the exposed substrate,
The photomask has the second of the auxiliary patterns so that the diffracted light intensity of the zeroth-order diffracted light and the diffracted light intensity of the first-order diffracted light irradiated onto the substrate to be exposed through the auxiliary pattern are equal. The combination of the transmittance of the transmissive part of the exposure light in which no light shielding film is arranged, the phase of the transmissive part, and the pattern width of the transmissive part is adjusted,
When the exposure light is irradiated by the oblique incidence illumination and the pattern is transferred onto the substrate to be exposed, the diffracted light intensity of the zeroth-order diffracted light and the diffracted light intensity of the first-order diffracted light are equal to the auxiliary pattern. A method for manufacturing a semiconductor device, wherein exposure light is irradiated.

Images