JP3574729B2 - Lens aberration measurement method - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method for measuring a lens aberration of a projection exposure apparatus used for manufacturing a semiconductor device or the like.
[0002]
[Prior art]
In a lithography process for manufacturing a semiconductor device or the like, a reduction projection exposure apparatus is mainly used. Miniaturization of a semiconductor device has been achieved by improving the performance of the reduction projection exposure apparatus. However, in order to further reduce the size of the pattern, it is necessary to reduce the aberration of the reduction projection lens. Conventionally, the aberration of a lens mounted on a reduction projection exposure apparatus has been evaluated using information obtained from the shape of a formed resist pattern. FIG. 9 shows an example of a change in resist cross-sectional shape depending on the presence or absence of coma of the lens. When there is no aberration, as shown in FIG. 9A, the inclination angles of the side surfaces on both sides of the resist pattern 91 are substantially equal, while when there is aberration, as shown in FIG. The inclination angles of the side surfaces on both sides of the resist pattern 92 are different, and in this case, the inclination angles of the right side surfaces are gentle. Normally, the difference between the inclination angles on both sides or the difference between the inclination widths 93 and 94 is used for empirical evaluation. However, in this evaluation method, the obtained value differs depending on the reflectance of the substrate, the resist film thickness, and the characteristics of the resist, and the amount of aberration cannot be measured accurately.
[0003]
[Problems to be solved by the invention]
To reduce lens aberrations, it is necessary to measure, correct, and select various aberrations of the lens with high accuracy. An object of the present invention is to solve the above problems and to provide a means for accurately measuring the amount of aberration of a lens mounted on a reduction projection exposure apparatus.
[0004]
[Means for Solving the Problems]
In order to achieve the above object, according to the present invention, the light intensity of a second light intensity peak, that is, a so-called sub-peak, which is generated when a pattern is formed by using a translucent phase shift mask, is centered on the main pattern by the amount of aberration of the lens. The lens aberration is quantitatively determined by utilizing the difference between the opposing regions.
[0005]
Further, by arranging the measurement pattern over the entire surface of the mask, the lens aberration in the exposure area can be evaluated by one-shot exposure.
[0006]
Further, by arranging the light-shielding pattern, it is possible to perform exposure at a fine pitch, and it is possible to reduce the influence of the resist coating variation in the wafer, the influence of the substrate, and the like, and to perform evaluation.
[0007]
BEST MODE FOR CARRYING OUT THE INVENTION
The principle of the present invention will be described with reference to FIG. FIG. 1A is a plan view of a translucent phase shift mask, and FIG. 1B is a cross-sectional view of the translucent phase shift mask. 1 is a glass substrate, 2 is a translucent phase shift film, and 3 is a main pattern formed in a transparent region. As the translucent phase shift film 2, a CrON film was used. The transmissivity of the translucent phase shift film 2 for exposure light was 6%. Here, a CrON film is used as the translucent film, but the present invention is not limited to this. An ordinary translucent phase shift mask may be used as long as the phase of light passing through the translucent portion and the transparent portion is substantially inverted, such as a multilayer film of a transparent film such as CrO, CrN, MoSiO, MoSiON, or SiO2. Structure is good. As shown in FIG. 1 (c), the amplitude distribution of the light passing through the mask is such that the light passing through the main pattern 3, which is a light transmitting portion, has a positive sign, while the light transmitted through the translucent phase shift film 2 has a positive sign. The phase of the transmitted light is inverted and has a negative sign. When this light is projected on the wafer through a lens, the phase is inverted at the boundary between the main pattern 3 which is a light transmitting portion and the translucent phase shift film 2 as shown in FIG. The light intensity becomes almost zero. Therefore, the spread of light intensity is suppressed, and a fine pattern with high contrast can be formed. However, in addition to the light intensity 4 passing through the main pattern 3, a second light intensity peak 5 or a so-called sub-peak occurs. Normally, when this mask is used, the exposure amount of the resist is set so that the second light intensity peak is not transferred. In the present invention, the subpeak which is not normally transferred to the wafer is transferred to the resist on the wafer, and the aberration of the lens is evaluated.
[0008]
FIG. 10 shows a main configuration example of a reduction projection exposure apparatus that projects a pattern on a resist. Light emitted from a light source 101 irradiates a mask 106 via a fly-eye lens 102, condenser lenses 103 and 105, and a mirror 104. In some cases, the mask 106 is provided with a pellicle 107 for preventing pattern transfer failure due to foreign matter adhesion. The mask pattern drawn on the mask 106 is projected via a projection lens 108 onto a wafer 109 as a sample substrate. The mask 106 is placed on a mask stage 118 controlled by the mask position control means 117, and the center of the mask 106 and the optical axis of the projection lens 108 are accurately aligned. The wafer 109 is vacuum-adsorbed on the sample stage 110. The sample stage 110 is mounted on a Z stage 111 movable in the optical axis direction of the projection lens 108, that is, in the Z direction, and further mounted on an XY stage 112. The Z stage 111 and the XY stage 112 are driven by the respective driving units 113 and 114 in accordance with control commands from the main control system 119, and can be moved to desired exposure positions. The position is accurately monitored by the laser length measuring device 115 as the position of the mirror 116 fixed to the Z stage 111. The projection lens 108 has various aberrations. However, at present, there is no method for accurately measuring the aberration of the projection lens mounted on the reduction projection exposure apparatus, which is a problem. When a translucent phase shift mask is used as the mask 106 and a pattern is transferred to a resist on the wafer 109, the aberration of the lens 108 changes the second light intensity peak 5 described above. FIGS. 1D and 1E are examples. FIG. 1D shows the light intensity distribution when there is no aberration of the lens. The second light intensity peak 5 occurs at the same light intensity around the main pattern. FIG. 1E shows a light intensity distribution when the lens has a coma aberration. There is a difference between the light intensities of the sub-peaks 7 and 8 facing each other with the main pattern as the center. In the present invention, using this phenomenon, the lens aberration is quantified, and the aberration is evaluated with high accuracy.
[0009]
A first embodiment of the present invention will be described with reference to FIGS. 2 to 5 and FIGS. FIG. 2 shows the relationship between the coma aberration, one of the lens aberrations, and the light intensity of the second light intensity peaks 7 and 8 of the light intensity distribution shown in FIG. Here, a stepper having an exposure wavelength λ = 0.248 μm and a numerical aperture NA = 0.55 of the lens was used for pattern transfer. The transmittance of the translucent film is 6% when the transmittance of the transparent portion is 100%, the phase difference between the transparent portion and the translucent portion is 180 °, the design dimension W of the light transmitting portion corresponding to the main pattern 3 to be transferred. Using a mask of 0.40 μm square (2 μm square on the mask because the magnification of the projection exposure optical system is １ ／). The light intensity at the second light intensity peak 7 increases substantially in proportion to the increase in the amount of coma. On the other hand, the light intensity of the second light intensity peak 8 located at a position facing the center of the main pattern decreases as the coma aberration amount increases.
[0010]
FIG. 3 shows the ratio of the light intensity between the second light intensity peak 8 and the second light intensity peak 7. It can be seen that the light intensity ratio increases in proportion to the coma. Therefore, by obtaining the exposure ratio at which the second light intensity peaks 7 and 8 are respectively transferred, the coma aberration can be obtained using FIG.
[0011]
FIGS. 4 and 12 show a cross-sectional shape and a planar shape of the resist pattern actually tested. A resist 10 is applied on a substrate 9 as shown in FIG. 4A and exposed and developed in a usual process to form a mask pattern on the resist. Here, the difference from the related art is that the amount of exposure is adjusted so that the second light intensity peaks 7 and 8 are transferred to the resist when a pattern is formed. As shown in FIG. 4B, in addition to the pattern 11 formed by the light that has passed through the main pattern, the second light intensity peak 7 is used to determine an exposure amount E1 at which the resist starts to decrease in film thickness. Under this condition, the second light intensity peak 8 is not transferred to the position 12 opposite to the pattern 11 as the center. Next, as shown in FIG. 4C, an exposure amount E2 at which the resist starts to be reduced in film thickness is determined by the second light intensity peak 8 at a position 12 opposed to the resist film reduced portion 14 with the pattern 11 as a center. The coma aberration amount can be obtained by converting the exposure amount ratio between the exposure amount E2 and the exposure amount E1 into the light intensity ratio of the sub-peak and applying the result to FIG. Actual measurement results were E1 = 76.8 mJ / square cm and E2 = 48.0 mJ / square cm. The light intensity ratio in this case was E2 / E1 = 76.8 / 48.0 = 1.6, and the coma aberration amount was about 0.11 × λ. Alternatively, a method may be used in which the second light intensity peaks 7 and 8 are simultaneously transferred to the resist, and the reduced amount of each resist film is converted into light intensity using a sensitivity characteristic curve of the resist.
[0012]
The main pattern is not limited to a square, but may be a polygon such as a hexagon or an octagon. By using a polygon pattern, the direction of coma aberration can be measured simultaneously with the amount of coma aberration. FIG. 11 is a plan view of a mask when an octagonal pattern is used. Reference numeral 122 denotes a translucent phase shift region, and reference numeral 123 denotes a main pattern formed by a transparent region. Here, the design dimension W of one side of the main pattern 123 is not particularly limited, but W = a · λ / NA (where the numerical aperture of the projection exposure optical system is NA, the exposure wavelength is λ, and a ≧ 0.4). It is desirable to set conditions. Here, a stepper having an exposure wavelength λ = 0.248 μm and a numerical aperture NA = 0.55 of the lens was used for pattern transfer. The transmissivity of the translucent film is 6% when the transmissivity of the transparent portion is 100%, the phase difference between the transparent portion and the translucent portion is 180 °, one side design dimension of the light transmissive portion corresponding to the main pattern 123 to be transferred. A mask having a W of 0.50 μm (2.5 μm square on the mask because the magnification of the projection exposure optical system is ５) was used. FIG. 12 shows the plan shape of the resist pattern actually tested. As shown in the figure, exposure and development are performed in a normal process to form a mask pattern on a resist. What is different from the related art is that the amount of exposure is adjusted so that the second light intensity peak is transferred to the resist when a pattern is formed. As shown in FIG. 12A, in addition to the pattern 131 formed by the light that has passed through the main pattern, an exposure amount E1 at which the resist starts to be reduced in thickness by the second light intensity peak is obtained. Under this condition, the second light intensity peak is not transferred to 132, which is a position facing the pattern 131 as a center. Next, as shown in FIG. 12B, an exposure amount E2 at which the resist starts to be thinned by the second light intensity peak at a position 132 facing the resist thinned portion 134 around the pattern 131 is obtained. The coma aberration amount can be obtained by converting the exposure amount ratio between the exposure amount E2 and the exposure amount E1 into the light intensity ratio of the sub-peak and comparing the calculated value with the calculated value. Actual measurement results were E1 = 76.8 mJ / square cm and E2 = 48.0 mJ / square cm. The light intensity ratio in this case was E2 / E1 = 76.8 / 48.0 = 1.6, and the coma aberration amount was about 0.11 × λ. In addition, the regular octagon made it possible to determine the direction of coma aberration with high accuracy.
[0013]
Coma is distributed in the plane of one shot of the reduction projection exposure apparatus. Therefore, it is desirable that the entire surface within a shot can be measured at once. FIG. 5 shows a measurement pattern arranged in an effective exposure area of a mask for measuring the amount of coma of a lens. FIG. 5A is a plan view of the mask, and FIG. 5B is an enlarged view of one of the evaluation patterns. Reference numeral 52 denotes a translucent phase shift area, which is arranged in an effective exposure area of the mask. Reference numeral 53 denotes a main pattern formed of a transparent area, and reference numeral 56 denotes an area outside the effective exposure area of the mask. Here, the design dimension W of the main pattern 3 is not particularly limited, but under the condition of W = a · λ / NA (where the numerical aperture of the projection exposure optical system is NA, the exposure wavelength is λ, and a ≧ 0.4). It is desirable to do. Here, the pattern is 0.40 μm square. A stepper having an exposure wavelength λ = 0.248 μm and a numerical aperture NA of the lens = 0.55 was used. A photomask was used in which the transmittance of the translucent portion of the mask was 6% and the phase difference between the transparent portion and the translucent portion was 180 °. If the arrangement pitch P of the main pattern is less than 0.78 μm, the accuracy of measurement is reduced due to the influence of an adjacent pattern, so the arrangement pitch of the main pattern is set to 0.78 μm or more. The arrangement pitch P of the main pattern is represented by P = b · λ / NA. Here, the numerical aperture of the projection exposure optical system is NA, the exposure wavelength is λ, and b ≧ 1.72. Using this mask, the amount of coma aberration on the entire surface of the lens could be measured. Also, by changing the exposure time in the wafer and transferring the pattern in a step-and-repeat manner, the presence or absence of sub-peak pattern transfer is recognized as a pattern defect using a normal pattern defect inspection device, so that the exposure amount of sub-peak transfer can be reduced. As a result, the sub-peak light intensity ratio could be efficiently obtained. The pattern defect inspection apparatus compares a reference pattern with a transfer pattern of an inspection pattern and recognizes a portion different from the reference pattern as a defect.
[0014]
A second embodiment will be described with reference to FIG. In the first embodiment, the substrate was moved in substantially the same step as the exposure region of one shot, and the exposure was repeated while changing the exposure amount, and the symmetry of the second light intensity was evaluated. In this case, an error in the film thickness of the substrate or the resist as the material to be processed, an error in the focal position caused by the apparatus, and the like cause a decrease in measurement accuracy. However, when the substrate moving step is smaller than the exposure shot size, the translucent portion is double-exposed, so that the intensity ratio of the light intensity peak cannot be obtained accurately.
[0015]
The second embodiment relates to a mask structure and an exposure method for avoiding double exposure. FIG. 6 shows the pattern arrangement of the mask. A light-shielding film, here a Cr film, was arranged so that the substrate could be moved at a pitch that was ultimately smaller than the exposure shot size and exposed. FIG. 6A is a plan view of the pattern, and FIG. 6B is an enlarged view of the pattern. 62 is a translucent phase shift section, 63 is a transparent section, and 64 is a light shielding section. Exposure conditions, transmittance of the translucent portion, phase difference, and the like are the same as in the first embodiment. In this mask structure, a light-shielding portion is provided around the translucent portion. Even if the pattern is transferred in a very small step and step and repeat, double exposure does not occur. However, desirably, the width of the translucent portion between the transparent portion 63 and the light shielding portion 64 needs to be ≧ 0.4 μm. If the width of the translucent portion is <0.4 μm, the sub-peak may not be transferred due to the effect of the light-shielding pattern, and the measurement accuracy is reduced. When the transfer optical systems are different, the arrangement position W1 of the light-shielding pattern is represented by W1 = c · λ / NA. Here, the numerical aperture of the projection exposure optical system is NA, the exposure wavelength is λ, and c ≧ 0.89. Further, the pitch of the step and repeat needs to be a step larger than the size of the translucent portion so that the translucent portion is not subjected to multiple exposure. The mask pattern arrangement can be applied to any arrangement other than that shown in FIG. 6 as long as it can prevent multiple exposure. For example, as shown in FIG. 7, a structure in which the light shielding portion is arranged in the step and repeat direction may be adopted. 72 is a translucent phase shift section, 73 is a transparent section, and 74 is a light shielding section. By setting the width W1 of the translucent portion to 0.4 μm or more, the second light intensity peak value can be obtained without being affected by the light shielding portion 74, and the present invention can be applied. The width W1 of the translucent portion is represented by W1 = c · λ / NA. Here, the numerical aperture of the projection exposure optical system is NA, the exposure wavelength is λ, and c ≧ 0.89. The light-shielding pattern 74 is disposed at a position facing at least the direction of the exposure step with the main pattern 73 as a center. The dimension W2 of one side of the light shielding pattern 74 needs to be set according to the number of steps and repeats and the pitch. The dimension W2 of one side of the light-shielding pattern 74 is represented by W2 = d · λ / NA. However, the numerical aperture of the projection exposure optical system is NA, the exposure wavelength is λ, and d ≧ 2.66.
[0016]
Using this mask, the aberration of the projection lens was measured. The exposure step was determined so that the translucent portions did not overlap twice. FIG. 8 shows a sectional view and a plan view of the pattern formed on the resist. Here, 85 is a substrate, 86 is a resist, 81, 82, 83 and 84 are patterns to which the transparent pattern 3 is transferred, and 81, 82, 83 and 84 are portions where the resist is reduced in film thickness by the second light intensity 7. , 81, 82, 83, and 84 indicate portions where the resist is reduced in film thickness by the second light intensity 8. In the exposure, step exposure was performed while increasing the exposure amount in the order of exposure steps 1, 2, 3, and 4. In the exposure step 2, the sub-peak 82 began to be transferred to the left of the main pattern 82, and in the exposure step 4, the sub-peak was transferred to the right of the main pattern 84 at 84. As a result of an actual experiment, the exposure amount of the exposure step 2 and the exposure amount step 4 was 48.0 mJ / square cm, and the exposure amount E2 of the exposure step 4 was 76.8 mJ. / Square cm. Therefore, the light intensity ratio is E1 / E2 = 76.8 / 48.0 = 1.6. When the coma aberration amount is obtained from this value using FIG. 3, it is easily understood that the coma aberration amount is about 0.11 × λ. Further, since the exposure step was reduced by arranging the light-shielding pattern, variations in the resist film thickness and the influence of the substrate were reduced, so that quantitative coma aberration could be evaluated with high accuracy. Further, even when the main pattern was changed to a polygon and a circle, the same effect as in the first embodiment was obtained.
[0017]
Although the dimensions and coefficients of the main pattern and the light-shielding pattern are limited as described above, the optimal values of the size and shape of the main pattern and the light-shielding pattern differ depending on the transmittance of the translucent region. For example, when the transmittance changes, the light intensity passing through the translucent region changes. For example, when the transmittance is changed to 4%, the light intensity passing through the translucent region becomes small. As a result, the main pattern size, the light shielding pattern arrangement position, and the like are changed. However, if each is optimized, the lens aberration measurement can be performed with almost no problem. Therefore, it is necessary to optimize the light shielding pattern position and the like in accordance with the transmissivity of the translucent region and the main pattern. The shapes of the main pattern and the light-shielding pattern are not limited to rectangles and hole patterns. It is applicable when sub-peaks such as a line pattern, a cross-shaped pattern, a polygonal pattern such as a hexagon and an octagon, and a circular pattern are generated around the main pattern. However, the hole pattern shown in this embodiment can be evaluated regardless of the direction of the aberration of the lens, and is practical. The transmissivity of the translucent region is not limited to this embodiment, but can be applied by using a coefficient suitable for the transmissivity. Although the light-shielding region is provided with the light-shielding film, it is not limited to this. Means such as a method of arranging transparent patterns equal to or less than the resolution limit at a predetermined pitch so that the light intensity becomes 0 on the wafer is not limited. However, since the preparation of the mask is simple, it is practical to arrange a light-shielding film as shown in this embodiment. Further, the structure and material of the mask are not limited to the materials used in this embodiment. That is, in the present invention, the structure of the mask used includes a transparent region, a translucent region, and a light-shielding region, and the phase difference of light passing through the transparent region and the translucent region is 180 °, and the periphery of the main pattern to be projected is If the pattern is arranged such that the second light intensity is transferred, the object can be achieved. Further, when the lenses were selected and corrected for the reduction projection exposure apparatus using this measurement method, the individual difference of the lenses could be reduced to half of the conventional one. Furthermore, as a result of forming a pattern of an ultra LSI using a reduced projection exposure apparatus that has performed a lens selection using the present measurement method, it is possible to reduce the pattern displacement caused by the coma of the lens to one third of the conventional pattern. A higher density pattern arrangement was realized. Further, the failure rate of the VLSI product was reduced to 2/3.
[0018]
【The invention's effect】
By applying the present invention, various aberrations of a lens mounted on a reduction projection exposure apparatus can be measured with high accuracy. As a result, the lens can be corrected and selected, and the individual difference of the lens can be made smaller than before. In particular, the pattern displacement caused by the coma of the lens can be reduced as compared with the related art, and as a result, a higher-density pattern arrangement can be realized, and the manufacture of the VLSI can be realized using optical lithography. Further, the defect rate of the VLSI product can be reduced, which is industrially advantageous.
[Brief description of the drawings]
FIG. 1 is an explanatory diagram of the principle of the present invention.
FIG. 2 is an explanatory diagram of the principle of the present invention.
FIG. 3 is an explanatory diagram of the principle of the present invention.
FIG. 4 is an explanatory diagram of a main embodiment of the present invention.
FIG. 5 is a plan view of a main embodiment of the present invention.
FIG. 6 is a plan view of a second embodiment of the present invention.
FIG. 7 is an explanatory view of a second embodiment of the present invention.
FIG. 8 is an explanatory view of a second embodiment of the present invention.
FIG. 9 is an explanatory diagram of a conventional technique.
FIG. 10 is an explanatory diagram of the principle of the present invention.
FIG. 11 is an explanatory view of a main embodiment of the present invention.
FIG. 12 is an explanatory view of a main embodiment of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Glass substrate, 2, 52, 62, 72, 122 ... Translucent phase shift film, 3, 53, 63, 73, 123 ... Transparent main pattern, 4 ... Light intensity peak of transparent main pattern, 5 ... Transparent main pattern , A light intensity peak of the transparent main pattern when there is a coma of the lens, a first subpeak when there is a coma of the lens, a second subpeak when there is a coma of the lens ..., 9, 85, 90 ... substrate, 10, 86, 13 ... resist, 11, 81, 82, 83, 84, 131 ... pattern transferred with transparent main pattern, 12, 81, 82, 83, 84, 132 ... Resist light-reduced portions due to second light intensity peak, so-called sub-peak 8, 14, 81, 82, 83, 84, 134... Resist film-reduced portions due to second light intensity peak, so-called sub-peak 7, 56, 76 ... effective exposure area Outer, 64,74 ... light shielding pattern, 106 ... mask, 108 ... projection lens, 109 ... wafer.

Claims (5)

A translucent area is provided in the translucent area and the translucent area for the exposure light, has a circular opening arranged two-dimensionally at a predetermined pitch, and the light transmitted through the translucent area and the light A step of preparing a photomask in which the phase of the light transmitted through the opening is inverted with respect to the phase of the light,
A step of preparing a substrate having a resist film formed thereon,
Using a projection exposure apparatus, transferring a pattern of the photomask to the resist film such that a second light intensity peak generated near the opening is transferred to the resist film;
A lens aberration measurement method, comprising: measuring lens aberration of the projection exposure apparatus using information transferred to the resist film based on the second light intensity peak. A light is provided in the translucent area and the translucent area with respect to the exposure light, has a polygonal opening two-dimensionally arranged at a predetermined pitch, and transmits light through the translucent area. A step of preparing a photomask whose phase is inverted with respect to the light transmitted through the opening,
A step of preparing a substrate having a resist film formed thereon,
Using a projection exposure apparatus, transferring a pattern of the photomask to the resist film such that a second light intensity peak generated near the opening is transferred to the resist film;
A lens aberration measurement method, comprising: measuring lens aberration of the projection exposure apparatus using information transferred to the resist film based on the second light intensity peak. A translucent area for the exposure light, a plurality of openings provided in the translucent area, and a light-blocking area provided at a predetermined distance from the opening between the plurality of openings. A step of preparing a photomask in which the light passing through the translucent region and the light passing through the opening have phases inverted from each other,
A step of preparing a substrate having a resist film formed thereon,
Using a projection exposure apparatus, transferring a pattern of the photomask to the resist film so that a second light intensity peak generated near the opening is transferred to the resist film;
A lens aberration measurement method, comprising: measuring lens aberration of the projection exposure apparatus using information transferred to the resist film based on the second light intensity peak. The predetermined distance W is not less than W = c · λ / NA (however, the numerical aperture of the projection optical system of the projection exposure apparatus is NA, the exposure wavelength is λ, and c ≧ 0.89). The method for measuring lens aberration according to claim 3 . The predetermined distance W is not less than W = d · λ / NA (however, the numerical aperture of the projection optical system of the projection exposure apparatus is NA, the exposure wavelength is λ, and d ≧ 2.66). The method for measuring lens aberration according to claim 3 .

Images