JP4309541B2 - Aberration measurement method and aberration measurement mark - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an alignment apparatus mounted on an exposure apparatus. of The present invention relates to an aberration measurement method and an aberration measurement mark.
[0002]
[Prior art]
In the manufacture of semiconductor devices, alignment devices mounted on exposure apparatuses can be broadly classified into the following three methods. In the first method, the alignment mark is irradiated with a laser, and scattered light from the alignment mark is detected. The second method is to project an image of an alignment mark obtained through a magnifying optical system on the light receiving surface of a CCD camera and measure the position of the alignment mark using an image processing technique. The third method uses heterodyne interference. For example, the LSA sensor manufactured by Nikon as the first method, the FIA sensor manufactured by Nikon as the second method, the alignment sensor manufactured by Canon, and the third method are used. There are LIA sensors manufactured by Nikon and ATHENA manufactured by ASML.
[0003]
Judgment of the performance of the alignment apparatus required in actual manufacture of a semiconductor device has two stages. The first stage is whether or not a signal can be detected without exception, and high versatility is required. The second stage is measurement accuracy and accuracy stability. The first type alignment apparatus is excellent in the first stage. That is, the first type alignment apparatus is characterized by low signal detection leakage and high versatility. However, with the miniaturization of semiconductor devices, the first type of alignment apparatus is becoming difficult to say that the accuracy of the second stage is sufficient.
[0004]
For this reason, expectation for the alignment apparatus belonging to the second system is increasing. That is, as long as the alignment signal can be detected, higher alignment accuracy can be expected in the second method than in the first method. Therefore, in the future, it is considered that the alignment sensor of the second system is actively used by compensating for the low versatility by adjusting the design of the alignment mark and the cross-sectional structure.
[0005]
FIG. 9 is a schematic diagram of an alignment apparatus that is an object of the present invention belonging to the second system. In FIG. 9, reference numeral 1 denotes an illumination optical system, 2 denotes a halogen lamp that irradiates light to the illumination optical system 1, and 3 denotes an objective lens that passes light from the illumination optical system 2. Reference numeral 4 denotes a substrate, and 101 denotes an alignment mark formed on the substrate 4. Reference numeral 5 denotes an enlarged projection optical system that enlarges the reflected light of the alignment mark 101, and reference numeral 6 denotes a CCD camera that receives the image of the alignment mark 101 enlarged by the enlargement optical system 5. Reference numeral 81 denotes a reticle, and 82 denotes an electron optical system.
[0006]
As shown in FIG. 9, the light from the halogen lamp 2 is passed through the objective lens 3 through the illumination optical system 1 and incident on the alignment mark 101 formed on the substrate 4. The light reflected in the vicinity of the alignment mark 101 forms an image of the alignment mark on the light receiving surface of the CCD camera 6 by the magnifying projection optical system 5. Here, the alignment mark 101 is generally a mark as shown in FIG. This mark has a structure in which seven belt-like patterns having a width of 6 μm are arranged at intervals of 12 μm, and each belt-like pattern has a groove-like or convex cross-sectional structure. In addition, the alignment apparatus measures the position of the alignment mark from the edge signal of the 6 μm pattern of the alignment mark 101.
[0007]
In order to overcome the low versatility that is the disadvantage of the alignment system belonging to the second method, the detection performance with low step alignment marks is described in Reference 1 (SPIE vol.3051, P.836-845). An improved type alignment apparatus with an improved value has been reported. At present, a phase shift type alignment apparatus described in Document 1 has also been proposed.
[0008]
The alignment apparatus described in this document 1 can obtain a good signal even for an alignment mark having a small mark contrast and a small level difference. Specifically, as shown in FIG. 9, the illumination stop 8 is disposed between the lenses of the illumination optical system 1, and the 180 ° phase plate 9 is disposed between the lenses of the enlarged projection optical system. Therefore, only the light passing near the center of the lens is inverted by 180 °. As a result, alignment is possible even with respect to an alignment mark for which a good alignment signal has not been obtained conventionally.
[0009]
The illumination stop 8 and the phase plate 9 are shown in FIGS. 21 (a) and 21 (b). FIG. 21A shows a plan view of the illumination stop 8 and FIG. 21B shows a plan view of the phase plate 9. As shown in FIGS. 21A and 21B, the illumination stop 8 and the phase plate 9 have a similar shape. The transmission part 95 of the phase plate 9 corresponding to the light shielding part 92 of the illumination stop 8 is formed of a transparent film, and the phase of the transmission part 95 is shifted by 180 ° with respect to the transmission part 96 other than the transmission part 95.
[0010]
However, when the alignment is performed by the second method, the measurement value of the mark position is shifted due to a so-called manufacturing error such as an aberration of the magnifying optical system 5 and an optical axis shift. This measurement value deviation, that is, the amount by which the measurement position is deceived is considered to be a factor of the alignment offset. In addition, this alignment offset is expected to vary not only depending on the alignment sensor, but also depending on the mark design, cross-sectional structure, and the like. That is, the alignment offset includes not only the alignment offset between exposure apparatuses, that is, the alignment offset for each exposure apparatus, but also the alignment offset generated due to the difference in the structure of the alignment mark generated in the same exposure apparatus.
[0011]
In alignment using the same alignment mark, the difference in alignment offset for each alignment device is considered to be caused by the aberration of the optical system that projects the image of the alignment mark on the detector. Similarly, in the alignment using the same alignment apparatus, it is considered that the difference in alignment offset for each alignment mark having a different mark structure is caused by the aberration of the optical system.
[0012]
Conventionally, such an alignment offset has been avoided by having a correction value in advance for each alignment sensor or each difference in the structure of the alignment mark. However, the alignment accuracy required today is becoming increasingly severe. In addition, cases in which exposure apparatuses are mixed are increasing. For this reason, it is difficult to reach the required alignment accuracy. Therefore, the importance of eliminating the alignment offset itself is increasing.
[0013]
The occurrence of an alignment offset due to the interaction of the alignment device's aberration, adjustment error, etc., and the cross-sectional shape of the alignment mark and its layer structure has been known in the past in document 2 (Jpn.J.Appl.Phys.Part1, No.12). (2), Vol. 36 (1997) pp. 7512-7516). In this document 2, attention is paid to the step of the alignment mark as a cause of occurrence of the alignment offset, and the height of the step and the symmetry of the alignment signal are investigated.
[0014]
Further, conventionally, after a lens is adjusted in advance as a single unit, this lens is incorporated into an alignment apparatus. Then, adjustment after incorporation is performed based on the symmetry of the alignment signal at the test mark. For this reason, it is difficult to perform high-precision adjustment only by adjusting the lens alone.
[0015]
[Problems to be solved by the invention]
Thus, in the conventional alignment apparatus, when the alignment is performed, the measurement value of the mark position is shifted due to the aberration of the magnifying optical system. This deviation in measured values does not only occur for each exposure apparatus. This deviation in the measured value occurs even in the same exposure apparatus due to the difference in the structure of the alignment mark.
[0016]
SUMMARY OF THE INVENTION An object of the present invention is to reduce an alignment offset that causes a difference between alignment devices or alignment marks. of An object of the present invention is to provide an aberration measurement method and an aberration measurement mark.
[0017]
[Means for Solving the Problems]
In order to achieve the above object, the present invention uses the following means.
[0019]
The aberration measuring method of the present invention includes a first mark portion formed on a base substrate and made of a large pattern having a large pattern width, and a second mark portion made of a small pattern having a pattern width narrower than the large pattern. The large pattern and the small pattern illuminate an aberration measurement mark arranged in parallel with each other in at least one direction through an illumination optical system, and an image of the aberration measurement mark is enlarged using an enlargement optical system. Projecting on the light receiving surface of the detector, detecting the positions of the large pattern and the small pattern of the aberration measurement mark, and measuring the aberration based on the positions of the large pattern and the small pattern.
[0020]
Preferably, at least one of the large pattern and the small pattern of the aberration measurement mark is composed of a plurality of fine patterns that cannot be observed with the resolution of an optical system used for alignment, and the fine patterns are densely packed to form the large pattern or It functions as the small pattern.
[0021]
Desirably, the measurement of the aberration is performed at the time of alignment of pattern exposure of an actual device, and the aberration measurement mark functions as an alignment mark. The aberration is measured by adjusting an alignment device. Further, the measurement of the aberration is performed at the time of alignment of pattern exposure of an actual device, the aberration measurement mark functions as an alignment mark, and the aberration is determined from a difference between the position of the large pattern and the position of the small pattern. A distortion amount of the measurement mark is calculated, and a correct position of the aberration measurement mark is calculated based on at least one of the position of the large pattern and the position of the small pattern and the distortion amount.
[0022]
The aberration measurement mark of the present invention has a first mark portion made of a large pattern having a large pattern width and a second mark portion made of a small pattern having a pattern width narrower than the large pattern. The patterns are arranged in the vicinity in parallel to each other in at least one direction, and the aberration is measured from a pattern image obtained by irradiating the pattern with light.
[0023]
Desirably, at least one of the large pattern and the small pattern is centered on the other. Further, the large pattern and the small pattern are formed in an arbitrary composition ratio, and a plurality of them are alternately formed. Further, at least one of the large pattern and the small pattern is composed of a plurality of fine patterns that cannot be observed with the resolution of the optical system used for alignment, and the fine patterns are densely functioning as the large pattern or the small pattern. . The large pattern and the small pattern are formed as alignment marks.
[0024]
In the present invention, the center of the coma aberration is detected by the optical system that detects the center of the aberration measurement mark and the center of the coma aberration with a detector and translates the decentered component of the coma aberration provided in the magnifying optical system based on the detection result. Is adjusted to align with the center of the aberration measurement mark. Thereby, the alignment offset which arises between exposure apparatuses or originates in the structure of the alignment mark can be reduced.
[0025]
In particular, a pattern in which a large pattern and a small pattern are formed side by side is used as the aberration measurement mark. When the interval between the patterns of the aberration measurement mark having this pattern is detected by the detector, the pattern interval is not constant due to the eccentricity of the coma aberration. This coma should be symmetric with respect to the center of the lens and tend to increase toward the outside, but if the distribution of coma is decentered, it will not be symmetric. This is one of the causes of offset depending on the structure of the device and the type of alignment device. Therefore, the magnitude of the coma aberration can be estimated by measuring the deviation of the pattern interval.
[0026]
In addition, when an aberration measurement mark having this pattern is used, the optical component that detects the principal ray is adjusted to be constant by the optical system that translates the decentered component of the coma aberration. The coma aberration can be symmetric with respect to the center, and the offset between alignment devices or alignment marks can be reduced.
[0027]
Further, by incorporating a plane-parallel plate into the magnifying optical system and moving the principal ray in parallel by adjusting the tilt, there is no change in the shape of the aberration distribution associated with the adjustment, so that simple and quick adjustment is possible.
[0028]
Further, by using the aberration measurement mark as an alignment mark, it becomes possible to align with the aberration measurement. Therefore, at the same time as measuring the position of the alignment mark, the magnitude of the coma aberration of the alignment apparatus can be measured. Thereby, the alignment which correct | amended the alignment offset previously is realizable.
[0029]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings.
[0030]
(First embodiment)
With reference to FIG. 1 thru | or FIG. 3, the adjustment method of the alignment apparatus which concerns on 1st Embodiment of this invention is demonstrated.
[0031]
FIG. 1 is a schematic diagram showing the overall configuration of the alignment apparatus according to the present embodiment. In FIG. 1, 1 denotes an illumination optical system, 2 denotes a halogen lamp that irradiates light to the illumination optical system 1, and 3 denotes an objective lens that passes light from the illumination optical system 2. Reference numeral 4 denotes a substrate, 5 denotes an enlargement projection optical system for enlarging reflected light from the substrate 4, and 6 denotes a CCD camera for receiving an image of the substrate 4 enlarged by the enlargement optical system 5. 8 indicates an illumination stop disposed between the lenses of the illumination optical system 1, 7 indicates a parallel plane plate disposed between the magnifying optical system 5 and the CCD camera 6, and 7 'indicates the inclination of the parallel plane plate 7. An inclination adjustment mechanism to be adjusted is shown.
[0032]
As shown in FIG. 1, the light from the halogen lamp 2 is passed through the objective lens 3 through the illumination optical system 1 to illuminate the aberration measurement mark formed on the substrate 4. The light reflected in the vicinity of the aberration measurement mark forms an image of the aberration measurement mark on the light receiving surface of the CCD camera 6 via the magnification projection optical system 5.
[0033]
Between the magnifying optical system 5 and the CCD camera 6, a parallel plane plate 7 capable of adjusting the tilt is disposed. By adjusting the inclination of the plane parallel plate 7, the principal ray between the plane parallel plate 7 and the substrate 4 can be moved in parallel. An inclination adjusting mechanism 7 ′ is connected to the parallel plane plate 7, and the inclination of the parallel plane plate 7 can be adjusted by the inclination adjusting mechanism 7 ′. As described above, the function of translating the aberration distribution is realized by incorporating the plane parallel plate 7 in the magnifying optical system and adjusting the inclination of the plane parallel plate 7.
[0034]
The reason why the aberration distribution is translated by adjusting the inclination of the plane parallel plate 7 will be described below.
[0035]
The principal ray can also be translated in parallel by shifting the position of an element constituting the magnifying optical system 5, such as the lens, in the lateral direction with respect to the optical axis or changing the tilt of the lens. However, in this case, since the aberration distribution is also changed, the magnitude of the aberration itself is also changed. Therefore, in such an adjustment, after the principal ray is translated, an adjustment for reducing the magnitude of the aberration itself must also be performed.
[0036]
On the other hand, the plane parallel plate 7 incorporated in the magnifying optical system 5 can translate the principal ray according to the inclination and thickness thereof, like the elements constituting the magnifying optical system 5. At this time, it is possible to translate the aberration distribution without changing the shape of the aberration distribution.
[0037]
In this way, incorporating the plane-parallel plate 7 in the magnifying optical system 5 and translating the principal ray with the inclination thereof does not change the shape of the aberration distribution accompanying the adjustment, so that simple and quick adjustment is possible. .
[0038]
Further, an illumination stop 8 for reducing coherency is disposed between the lenses of the illumination optical system 1. By narrowing the illumination light with the illumination stop 8, the influence of coma aberration of the magnifying optical system 5 can be emphasized.
[0039]
The inclination of the plane parallel plate 7 is adjusted based on the aberration measurement mark image detected by the CCD camera 6 and the distribution of coma aberration. As a result, the optical axis of the magnifying optical system 5 is adjusted to translate the decentered component of coma aberration, and the alignment apparatus is adjusted. The operation and principle will be specifically described below.
[0040]
A general optical system has an aberration called coma aberration. This aberration is described in Reference 3 (Max Born and Emil Wolf, “Principles of Optics,” 6th edition, (Pergamon Press)), where the wavefront aberration Φ on the pupil plane of the optical system is expressed by the equation (1) It is shown that there is.
[0041]
Φ = Fy ₀ ρ ^Three cosθ (1)
F in equation (1) is a coefficient representing the magnitude of aberration, (p, θ) is polar coordinates on the pupil plane, y ₀ Is the distance from the center of the optical axis on the substrate 4. In this equation, Fy which is the substantial magnitude of coma on the substrate 4 ₀ Is the distance y from the optical axis center ₀ It shows that it increases in proportion to.
[0042]
FIG. 2 shows a point image in an optical system with coma aberration. As shown in FIG. 2, in an optical system with coma aberration, a point image has a shape in which a comet has a tail. This indicates that when the pattern is Fourier-expanded, the position of the image is shifted as the high frequency component. That is, the position of the image is shifted depending on the mark shape. Accordingly, the low-frequency component forms a small image near the point image, the high-frequency component is greatly shifted in the y-axis direction, and the size of the pattern image is also increased.
[0043]
Next, FIG. 3 shows the relationship between the distribution of coma aberration and the aberration measurement mark. In FIG. 3, 31 indicates an aberration measurement mark, and 32 indicates the center of the aberration measurement mark 31. Reference numeral 33 denotes a coma aberration distribution, and 34 denotes the center of the coma aberration distribution 33.
[0044]
As shown in FIG. 3, the distribution of coma aberration ideally has a concentric distribution with respect to the center of the optical axis, as can be seen from Equation (1). However, in the actual magnifying optical system of the alignment apparatus, the center 32 (the center of the optical system) of the aberration measurement mark 31 and the center 34 of the coma aberration distribution 33 do not always coincide with each other. Rather, it is more common that both are eccentric.
[0045]
Therefore, by adjusting the inclination of the plane-parallel plate 7 incorporated in the magnifying optical system 5 and moving the optical axis in parallel, the center 32 of the aberration measurement mark and the center 34 of the coma aberration distribution are matched. Thereby, the symmetry of the coma aberration in the optical system is improved, and the alignment offset causing a difference for each alignment apparatus or for each alignment mark is reduced.
[0046]
According to this embodiment, the parallel flat plate 7 is incorporated in the magnifying optical system 5, the inclination is adjusted, and the main optical axis is moved in parallel, so that aberration correction without changing the shape of the aberration distribution is possible. Therefore, the alignment apparatus can be easily and quickly adjusted.
[0047]
(Second Embodiment)
With reference to FIG. 4 thru | or FIG. 7, the adjustment method of the alignment apparatus which concerns on 2nd Embodiment of this invention is demonstrated. The present embodiment is an embodiment showing a specific aberration measurement method, and is performed using the same alignment apparatus as in the first embodiment. In the following embodiments, detailed description of portions common to the first embodiment will be omitted.
[0048]
The feature point of this embodiment is that the alignment device has a function of translating the principal ray of the magnifying optical system, and an aberration measurement mark for measuring the aberration of the magnifying optical system is arranged on the substrate, and this aberration measurement is performed. The principal ray is translated according to the measured value from the mark, and the center of the optical axis of the projection lens and the center of the aberration distribution are adjusted to substantially coincide.
[0049]
A general optical system has wavefront aberration and blurs the detected image of the alignment mark or shifts the position of the image. Therefore, if an attempt is made to measure the position of the alignment mark based on the detected image of the alignment mark, an offset occurs between the actual position. This wavefront aberration can be divided into five types: spherical aberration, astigmatism, field curvature, coma aberration, and distortion, as shown in Document 2 described above. Among these aberrations, coma and distortion are aberrations in which the pattern image is shifted in the lateral direction.
[0050]
Distortion is not an important aberration particularly in alignment because all patterns are shifted equally without affecting the pattern size. However, the coma aberration is a particularly important aberration for the alignment apparatus because a difference occurs in the shift amount depending on the size and structure of the pattern.
[0051]
As can be seen from FIG. 2, the coma aberration has a difference in the shift amount depending on the size of the pattern. Therefore, the magnitude of coma aberration can be estimated by comparing the positions of the images of two types of patterns having different sizes. If this feature is used and two types of patterns having different sizes are used, these patterns can be used as aberration measurement marks.
[0052]
As shown in FIG. 3, the distribution of coma aberration is generally a concentric distribution. It is difficult to reduce the magnitude of the coma aberration itself, and it is not practical, but it is relatively easy to adjust the center position of the coma aberration distribution.
[0053]
Therefore, the alignment apparatus itself has a function of adjusting the aberration of the magnifying optical system, in particular, a function of translating the principal ray, and the amount of movement is adjusted according to the measurement value from the aberration measurement mark. Thereby, the center of the coma aberration distribution and the center of the magnifying optical system are matched. By this adjustment, an alignment apparatus that reduces the alignment offset that causes a difference between alignment apparatuses or alignment marks is realized.
[0054]
A plan view of a wafer on which specific aberration measurement marks are formed is shown in FIG. In FIG. 4A, 40a is a wafer and 40b is an aberration measurement mark. Here, although one aberration measurement mark 40b is shown in the vicinity of the center of the wafer 40a, it is not limited to that position, and the number thereof is not limited.
[0055]
FIG. 4B shows an enlarged view of the aberration measurement mark 40b. In FIG. 4B, 41 is a large pattern and 42 is a small pattern. In the mark 40b, the first mark portion formed by the large pattern 41 and the second mark portion formed by the small pattern 42 are arranged in the vicinity in parallel with each other in the vertical direction of the drawing. The first and second mark portions are formed symmetrically and alternately arranged. The large pattern 41 has a thick pattern width, and the small pattern 42 has a pattern width narrower than the pattern width of the large pattern 41. The large pattern 41 is a line pattern having a width of 6 μm, for example, and the small pattern 42 is a line pattern having a width of 1 μm, for example. Further, the large pattern 41 is arranged so that the pattern width of the large pattern 41 is larger than the pattern width of the small pattern 42 in a certain direction. Specifically, the large patterns 41 and the small patterns 42 are arranged at a cycle of 24 μm so that the line length directions are parallel to each other.
[0056]
FIG. 5 shows a pattern obtained by observing the aberration measurement mark shown in FIG. 4B with an alignment apparatus having coma aberration.
[0057]
In an optical system with coma aberration, the transfer position shifts depending on the pattern size. This deviation amount tends to increase as the pattern becomes smaller. For this reason, when the mark shown in FIG. 4B is observed with an alignment apparatus having coma aberration, as shown in FIG. 5, it is observed that the 1 μm pattern is shifted from the 6 μm pattern.
[0058]
At this time, if the distance (L1 and L2) between the 1 μm pattern and the 6 μm pattern on both sides is measured, the positional deviation amount δx = (L1−L2) / 2 of the 1 μm pattern can be calculated. The measurement is performed in a state where the diaphragm of the illumination optical system is made smaller than usual and the illumination coherency is set to about 0.1. Since this positional deviation amount δx is substantially proportional to the magnitude of coma aberration, the direction and magnitude of coma aberration can be estimated from the value of δx.
[0059]
By aligning the center of the coma aberration distribution measured in this way with the center of the aberration measurement mark (center of the optical system) using the function of translating the chief ray, for each alignment device or each alignment mark An alignment apparatus that reduces an alignment offset in which a difference occurs is realized.
[0060]
The same effect can be obtained by using marks as shown in FIGS. 6 and 7 instead of FIG.
[0061]
FIG. 6A shows a mark in which a plurality of large patterns 41 exist, and a plurality of small patterns 42 are arranged between the large patterns 41. FIG. 6B shows marks in which one large pattern 41 and one small pattern 42 are arranged. FIG. 7A shows a mark in which two large patterns 41 and two small patterns 42 are arranged. FIG. 7B shows a mark in which two large patterns 41 are arranged with a small pattern 42 in between. FIG. 7C shows a mark in which two large patterns 41 are arranged with two small patterns 42 in between.
[0062]
As described above, the marks shown in FIGS. 6 and 7 are not formed by alternately arranging the large pattern 41 and the small pattern 42 one by one, but if each pattern is arranged, the amount of deviation is measured. And can measure aberrations. In particular, in the pattern shown in FIG. 6B, only one large pattern 41 and one small pattern 42 are arranged, but for example, aberration measurement can be sufficiently performed in the initial adjustment.
[0063]
As described above, according to this embodiment, the coherency is reduced by the illumination stop 8, and the projection lens of the projection lens is measured from an image obtained by measuring the intervals between marks in which thick patterns and thin patterns are alternately arranged at equal intervals. A coma aberration or a deviation from the optical axis center with respect to the coma aberration is calculated. Thereby, it becomes possible to measure aberration. Further, this measurement result can be fed back to the inclination adjustment of the parallel flat plate.
[0064]
(Third embodiment)
FIG. 8 shows a configuration of an aberration measurement mark used in the alignment apparatus according to the third embodiment of the present invention. In FIG. 8, 41 indicates a large pattern and 42 indicates a small pattern. The present embodiment relates to a modification of the second embodiment. In the aberration measurement mark used for adjustment of the alignment apparatus, the first mark portion formed by the large pattern 41 and the second mark portion formed by the small pattern 42 are formed symmetrically and alternately arranged in two directions. It is characterized by being. This aberration measurement mark is a mark in which the large pattern 41 and the small pattern 42 shown in FIG. 4B are overlapped in two orthogonal directions. This aberration measurement mark is also formed at any position on the wafer 40a, as in FIG. The advantage of performing aberration measurement using such a mark will be described below.
[0065]
Usually, the parallel movement of the principal ray of the optical system is performed by moving the elements (lens, mirror, flat plate, etc.) constituting the optical system in parallel or adjusting the tilt. These elements are normally supported at three points for reasons such as stress applied to the elements and positioning. With such three-point support, it is impossible to independently adjust the parallel movement of the principal ray in two orthogonal directions. That is, whenever the x direction is adjusted, y ₀ The direction will shift. Therefore, the measurement mark for x direction and y ₀ If the direction measurement marks are different from each other, it is necessary to switch each mark for each adjustment in each direction, and the efficiency is remarkably deteriorated.
[0066]
Therefore, if the aberration measurement mark shown in FIG. 8 is used, the coma aberration distribution can be measured simultaneously in two orthogonal directions. Therefore, it is possible to improve the adjustment efficiency and the adjustment speed for the element supported at three points.
[0067]
In this embodiment, the aberration measurement marks in which the patterns are orthogonal to each other are used. However, if the marks are overlapped in two directions, they need not be orthogonal. Further, the aberration measurement mark need not be a line pattern, and may be a pattern in which patterns having different dimensions are alternately arranged in an arbitrary configuration ratio in one or more arbitrary directions. Of course, the same effect can be obtained even if an optical system constituted by elements other than those supported at three points is used.
[0068]
(Fourth embodiment)
FIG. 9 is a diagram showing an overall configuration of an alignment apparatus that is a subject of the present invention. The present embodiment is characterized in that a mark capable of directly measuring aberration is formed on the alignment mark. In FIG. 9, reference numeral 1 denotes an illumination optical system, 2 denotes a halogen lamp that irradiates light to the illumination optical system 1, and 3 denotes an objective lens that passes light from the illumination optical system 2. Reference numeral 4 denotes a substrate, and 101 denotes an alignment mark formed on the substrate 4. Reference numeral 5 denotes an enlarged projection optical system that enlarges the reflected light of the alignment mark 101, and reference numeral 6 denotes a CCD camera that receives the image of the alignment mark 101 enlarged by the enlargement optical system 5. Reference numeral 8 denotes an illumination stop disposed between the lenses of the illumination optical system 1, and 9 denotes a 180 ° phase plate disposed between the lenses of the magnification projection optical system. Reference numeral 81 denotes a reticle, and 82 denotes an electron optical system.
[0069]
As shown in FIG. 9, the light from the halogen lamp 2 is passed through the objective lens 3 through the illumination optical system 1 and incident on the alignment mark 101 formed on the substrate 4. The light reflected in the vicinity of the alignment mark 101 forms an image of the alignment mark on the light receiving surface of the CCD camera 6 by the magnifying projection optical system 5. An illumination stop 8 is disposed between the lenses of the illumination optical system 1, and a 180 ° phase plate 9 is disposed between the lenses of the magnification projection optical system. Therefore, only the light passing near the center of the lens is inverted by 180 °. As a result, alignment is possible even with respect to an alignment mark for which a good alignment signal has not been obtained conventionally. The illumination stop 8 and the phase plate 9 are shown in FIGS. 21 (a) and 21 (b). FIG. 21A shows a plan view of the illumination stop 8 and FIG. 21B shows a plan view of the phase plate 9. As shown in FIGS. 21A and 21B, the illumination stop 8 and the phase plate 9 have a similar shape. The transmission part 95 of the phase plate 9 corresponding to the light shielding part 92 of the illumination stop 8 is formed of a transparent film, and the phase of the transmission part 95 is shifted by 180 ° with respect to the transmission part 96 other than the transmission part 95.
[0070]
In general alignment apparatuses, residual aberrations already present in the design stage, aberrations caused by adjustment errors that occur in the stage of assembling the optical system, optical axis shift between the enlarged projection optical system 5 and the pupil, the illumination optical system 1 It is inevitable that an error factor such as an optical axis shift between the illuminating diaphragm 8 and the illuminance unevenness of the halogen lamp 2 is included. Among these error factors, the aberration of the enlargement projection optical system 5, particularly coma, shifts the pattern image depending on the size of each pattern constituting the alignment mark and the controller. Therefore, in order to improve the overlay accuracy, it is important to perform processing for this coma aberration.
[0071]
A wafer 40a on which the alignment mark 101 according to the present embodiment is formed is shown in FIG. In FIG. 10, reference numeral 40a denotes a wafer, and 111 denotes a block on the wafer 40a. Reference numerals 101 a and 101 b denote alignment marks formed in the block 111.
[0072]
As shown in FIG. 10A, the wafer 40a is divided into regions (referred to as actual device pattern transfer blocks 111) where device patterns can be transferred at once. In this block 111, an actual device pattern (not shown) and alignment marks 101a and 101b for adjusting the alignment in the x and y directions are formed.
[0073]
Both of the alignment marks 101a and 101b have the same shape and are formed in different directions. Of these, FIG. 10B is an enlarged view of the alignment mark 101a for x-direction alignment adjustment. As shown in FIG. 10B, the alignment mark 101a has a structure in which the band-like pattern (large pattern 41) constituting the conventional alignment mark shown in FIG. The large pattern 41 and the small pattern 42 are arranged alternately. Similar to the conventional alignment mark shown in FIG. 20, the cross-sectional structure of the alignment mark shown in FIG. 10B may be either a groove shape or a convex shape.
[0074]
FIG. 11 shows a pattern when the alignment mark shown in FIG. 10B is observed with an alignment apparatus having coma aberration. In an optical system with coma aberration, the transfer position shifts depending on the pattern size. The amount of deviation depends on the pattern size. For this reason, when the alignment mark shown in FIG. 10B is observed with an alignment apparatus having coma aberration, as shown in FIG. 11, the 1 μm small pattern 42 is shifted as if the 6 μm large pattern 41 is shifted. Observed. At this time, if the distance (L) between the 1 μm pattern 42 and the 6 μm pattern 41 is measured, the relative positional deviation amount δx = 12 μm−L of the 1 μm pattern 42 can be calculated. This relative positional shift amount δx is considered to be substantially proportional to the magnitude of the coma aberration. For this reason, the direction and magnitude of the coma aberration can be estimated from the value of the positional deviation amount δx.
[0075]
As described above, when the alignment mark 101a as shown in FIG. 10B is used, the position of the alignment mark 101a can be measured and at the same time the coma aberration of the alignment apparatus can be measured. That is, it is possible to estimate the amount of alignment mark position distortion due to coma during alignment. Therefore, it is possible to realize alignment in which the amount of distortion is corrected in advance.
[0076]
In this embodiment, the alignment mark 101a as shown in FIG. 10B is used, but the present invention is not limited to this. For example, an alignment mark 101 composed of a pair of large patterns 41 and small patterns 42 as shown in FIG. 12A, or a plurality of large patterns 41 and a plurality of small patterns as shown in FIG. Of course, the alignment mark 101 may be constituted by the pattern 42.
[0077]
(Fifth embodiment)
FIG. 13 shows the configuration of an alignment mark 101 according to the fifth embodiment of the present invention. The present embodiment relates to a modification of the fourth embodiment, and is characterized in that two large patterns 41 are formed side by side at symmetrical positions with one small pattern 42 interposed therebetween.
[0078]
FIG. 14 shows a pattern obtained by observing the alignment mark 101 shown in FIG. 13 with an alignment apparatus having coma aberration. As in the fourth embodiment, as shown in FIG. 14, it is observed that the 1 μm small pattern 42 is shifted from the 6 μm large pattern 41. At this time, by measuring the distances L1 and L2 between the 1 μm pattern 42 and the 6 μm patterns 41 on both sides thereof, the relative displacement amount δx = (L1−L2) / 2 of the 1 μm pattern 42 can be calculated. The advantages will be described below.
[0079]
A general alignment optical system has an enlargement magnification error and distortion. This enlargement magnification error does not cause a measurement error during alignment. Moreover, although distortion becomes a measurement error, its magnitude is always a constant value. Therefore, by placing a fixed value offset on the measurement result, it is possible to cope with the distortion problem, so that an alignment error does not occur.
[0080]
However, when the relative positional deviation amount δx is calculated from the distance measurement between the large pattern 41 and the small pattern 42, the alignment mark shown in the fourth embodiment deviates from the designed value of 12 μm even when there is no coma. . In this case, it is necessary to grasp in advance the magnification error and distortion. Therefore, the positional deviation amount δx is calculated as in this embodiment. The positional deviation amount δx calculated in this way is not affected by the magnification error or distortion. Therefore, alignment that is not affected by the magnification error or distortion can be realized.
[0081]
The present invention is not limited to the above embodiment. Although the alignment mark 101 as shown in FIG. 13 is used, other configurations may be used, for example, a configuration in which a plurality of small patterns 42 as shown in FIG. Of course, the aberration measurement mark 31 or the aberration measurement mark 31 having a configuration in which the large pattern 41 and the small pattern 42 that do not have a configuration ratio of 1: 1 as shown in FIG.
[0082]
(Sixth embodiment)
16 to 17 show the configuration of an alignment mark according to the sixth embodiment of the present invention. This embodiment relates to a modification of the fourth embodiment, and fine patterns that cannot be separated with the resolution of the alignment optical system are densely packed. As the patterns are densely packed in this way, it is characterized by substantially functioning as a large pattern 41 ′.
[0083]
FIG. 16A shows an alignment mark 31 that functions as a large pattern 41 ′ in which thin strip-like patterns are densely packed. The reason why such a dense pattern functions as the large pattern 41 ′ will be described below.
[0084]
In many cases, the maximum dimension and the minimum dimension of the pattern constituting the alignment mark used in the actual manufacture of a semiconductor device are limited due to restrictions on the manufacturing process. For this reason, there is a case where a large pattern having a width of 6 μm cannot be used. In such a case, the alignment mark must be formed using a pattern having a dimension that is within the constraints of the manufacturing process. Therefore, a pattern that is finer than the resolution of the alignment optical system is concentrated to form a large pattern. When the dense pattern is measured by the alignment optical system, it is determined as one large pattern having a large width. Therefore, the measurement result of the alignment optical system shows that the dense pattern is not different from the large pattern at all, and such a dense pattern is substantially the same as the large pattern 41 used in the fourth and fifth embodiments. To work.
[0085]
As described above, according to this embodiment, alignment can be performed in substantially the same manner as the combination of the large pattern 41 and the small pattern 42 even when there is a restriction on the pattern size in the manufacturing process.
[0086]
The present invention is not limited to the above embodiment. Although the alignment mark 31 as shown in FIG. 16A is used, other configurations may be used. For example, as shown in FIG. 16 (b), the large pattern 41 is formed by densely gathering the strip-like patterns obtained by further dividing the thin strip-like pattern constituting the large pattern 41 ′ shown in FIG. 16 (a) in the longitudinal direction. The alignment mark 31 functioning as “may be used. In other words, any pattern in which fine patterns are denser than the resolution of the alignment optical system may be used. Also, as shown in FIG. Alternatively, the small pattern 42 ′ may be the alignment mark 31 that functions as the small pattern 42 ′ in which thin band-like patterns are densely packed. Furthermore, as shown in FIG. '' Each line may be divided in the longitudinal direction of the line pattern. In this case, each divided fine pattern is an alignment sensor. Not be the following fine pattern resolution.
[0087]
(Seventh embodiment)
FIG. 18 is a flowchart of mark position measurement in alignment according to the seventh embodiment of the present invention. In the present embodiment, alignment signal processing will be described with reference to the flowchart shown in FIG.
[0088]
First, similarly to the alignment performed in normal exposure, the alignment mark is irradiated with light, and the scattered light from the alignment mark is received by the CCD camera to obtain an alignment signal waveform (181).
[0089]
Next, the position of the alignment mark is measured from the alignment signal waveform in the same manner as the processing normally performed (182). In parallel with the measurement of the mark position, the distance between the large pattern 41 and the small pattern 42 is measured (183), and the amount of the alignment mark measurement position is calculated based on the measured distance (184). ).
[0090]
Next, a correct alignment mark position is calculated based on the obtained mark position and the amount of distortion, and the position is corrected (185).
[0091]
In the mark position measurement in such alignment, an example of the deduction amount calculation step (184) will be described with reference to FIG.
[0092]
In an alignment apparatus using a generally used halogen lamp as a light source and a numerical aperture of 0.3, the pattern size and the positional deviation amount are in an inversely proportional relationship. Accordingly, as shown in FIG. 19, when the horizontal axis represents the pattern position and the vertical axis represents the reciprocal of the pattern size, the position where the straight line passing through the large pattern position a and small pattern position b intersects the horizontal axis, that is, the pattern size. The position at which the reciprocal of is substantially 0 is the actual position c of the alignment mark. Therefore, if alignment is performed with reference to this position c, the amount of distortion generated for each alignment mark can be corrected simultaneously with normal position measurement.
[0093]
It is also possible to arrange a plane parallel plate 7 shown in FIG. 1 in the optical path of light received by the CCD camera 6. In this case, the inclination of the straight line passing through the position a and the position b shown in FIG. 19 can be increased by correcting the aberration according to the obtained distortion amount. As a result, the relative positional deviation amount δx can be reduced, the amount of distortion can be estimated more accurately, and coma can be corrected.
[0094]
The present invention is not limited to the above embodiment. For example, the pattern widths of the large pattern 41 and the small pattern 42 are 6 μm and 1 μm, respectively, but the present invention is not limited to this combination. The optimum combination of dimensions differs depending on the numerical aperture of the alignment optical system, the illumination coherency, the wavelength range of the light source, and the like, and is thus determined by an experiment using the target alignment apparatus. Further, the structure of the alignment mark is not limited to a cross-sectional structure such as a convex shape, a concave shape, and an embedded shape as in this embodiment, and any structure can be used as long as it can be detected by the alignment sensor.
[0095]
As described above in detail, according to the present invention, since it is possible to adjust the center deviation of the coma aberration distribution, it is possible to reduce the alignment offset that is different for each alignment device or each alignment mark. .
[0096]
Further, according to another aspect of the present invention, the size of the coma aberration can be estimated from the positional deviation of the pattern due to the decentering of the coma aberration by using the alignment mark formed by arranging the large and small patterns side by side. The offset can be reduced.
[0097]
In addition, the present invention can be variously modified and implemented without departing from the scope of the invention.
[0098]
【The invention's effect】
As described above, according to the present invention, an alignment apparatus that can reduce the alignment offset that causes a difference between alignment apparatuses or alignment marks. of An aberration measurement method and an aberration measurement mark can be provided.
[Brief description of the drawings]
FIG. 1 is a diagram showing an overall configuration of an alignment apparatus according to a first embodiment of the present invention.
FIG. 2 is a diagram showing a point image in an optical system with coma aberration.
FIG. 3 is a diagram showing the relationship between coma aberration distribution and aberration measurement marks;
FIG. 4 (a) is a plan view of a wafer on which aberration measurement marks according to a second embodiment of the present invention are formed. FIG. 4B is an enlarged view of the aberration measurement mark shown in FIG.
FIG. 5 is a diagram showing a pattern when an aberration measurement mark is observed with an alignment apparatus.
FIGS. 6A and 6B are views showing modifications of the aberration measurement mark according to the second embodiment.
FIGS. 7A to 7C are views showing modifications of the aberration measurement mark according to the second embodiment.
FIG. 8 is a diagram showing a configuration of an aberration measurement mark used in an alignment apparatus according to a third embodiment of the present invention.
FIG. 9 is a diagram showing an overall configuration of an alignment apparatus that is a subject of the present invention.
FIG. 10A is a view showing a wafer on which an alignment mark according to a fourth embodiment of the present invention is formed. FIG. 10B is an enlarged view of the alignment mark shown in FIG.
FIG. 11 is a diagram showing a pattern when an aberration measurement mark is observed with an alignment apparatus.
FIGS. 12A and 12B are views showing modifications of the aberration measurement mark according to the fourth embodiment. FIGS.
FIG. 13 is a diagram showing a configuration of an alignment mark according to a fifth embodiment of the present invention.
FIG. 14 is a diagram showing a pattern when an alignment mark according to a fifth embodiment of the present invention is observed with an alignment apparatus having coma aberration.
FIGS. 15A and 15B are views showing a modification of the alignment mark according to the fifth embodiment. FIGS.
FIG. 16A is a diagram showing a configuration of an alignment mark according to a sixth embodiment of the present invention. FIG. 16B and FIG. 16C are views showing a modification of the alignment mark according to the embodiment.
FIG. 17 is a view showing a modification of the alignment mark according to the sixth embodiment of the present invention.
FIG. 18 is a view showing a flowchart of mark position measurement in alignment according to the seventh embodiment of the present invention.
FIG. 19 is a diagram showing a process of calculating the amount of fraud according to the seventh embodiment of the present invention.
FIG. 20 is a diagram showing a configuration of a conventional alignment mark.
FIG. 21A is a diagram showing a configuration of an illumination stop used in a conventional alignment apparatus. FIG. 21B is a diagram showing a configuration of a phase plate used in a conventional alignment apparatus.
[Explanation of symbols]
1. Illumination optical system,
2 ... Halogen lamp,
3 ... Objective lens,
4 ... substrate
5 ... Enlarged projection optical system,
6 ... CCD camera,
7 ... Parallel plane plate,
7 '... tilt adjustment mechanism,
8 ... Lighting diaphragm,
9 ... Phase plate,
31, 40b, 101, 101a, 101b ... aberration measurement mark (alignment mark),
32 ... Center of aberration measurement mark,
33 ... coma aberration distribution,
34 ... Center of coma aberration distribution,
40a ... wafer,
41, 41 ′, 41 ″, 41 ′ ″ ... large pattern,
42, 42 ', 42''... small pattern,
81 ... reticle,
82: Electro-optical system,
111 ... Block.

Claims (2)

A first mark portion formed on a base substrate and made of a large pattern with a large pattern width, and a second mark portion made of a small pattern with a pattern width narrower than the large pattern, the large pattern and the small pattern being Illuminate the aberration measurement marks arranged in parallel with each other in at least one direction via the illumination optical system,
An image of the aberration measurement mark is projected onto the light receiving surface of the detector using an enlargement optical system,
Detecting the position of the large pattern and the small pattern of the aberration measurement mark;
Aberration is measured based on the position of the large pattern and the small pattern ,
At least one of the large pattern and the small pattern of the aberration measurement mark is composed of a plurality of fine patterns that cannot be observed with the resolution of the optical system used for alignment, and the fine patterns are densely packed to form the large pattern or the small pattern. Aberration measurement method characterized by functioning as : A first mark portion made of a large pattern having a large pattern width and a second mark portion made of a small pattern having a narrower pattern width than the large pattern, wherein the large pattern and the small pattern are mutually in at least one direction; It is arranged in the vicinity in parallel, the aberration is measured from the pattern image obtained by irradiating the pattern with light,
  At least one of the large pattern and the small pattern is composed of a plurality of fine patterns that cannot be observed with the resolution of the optical system used for alignment, and the fine patterns are densely functioning as the large pattern or the small pattern. Characteristic aberration measurement mark.

Images