JP2004163160A - Joining measuring device and mask for split exposure - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a joining measuring device for accurately measuring the state of joining even if mutually overlapping areas of sub-fields (split pattern regions) are small, and to provide a mask for split exposure. <P>SOLUTION: This measuring device 10 is structured so as to measure the state of split pattern regions adjoining each other with an elongate overlap region interposed between them and joined with each other, among a plurality of split pattern regions transferred by split exposure to different positions of a substrate 11. The device 10 is equipped with means 13 to 18 for lighting two line-and-space marks formed in the overlap region, and calculation means 18 to 22 for detecting light L1 generated from the substrate 11 correspondently to the amount of phase shift of the two line-and-space marks to calculate joining shift length between adjoining split pattern regions based on the distributed state of intensity of the light. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a joint measuring device for measuring a joint state between a plurality of divided pattern areas transferred to different positions on a substrate by divided exposure in a manufacturing process of a semiconductor element or a liquid crystal display element, and a divided exposure mask. .
[0002]
[Prior art]
Conventionally, various exposure methods such as electron beam (EB) reduction exposure, EB proximity exposure, and extreme ultraviolet (EUV) exposure have been proposed in order to cope with pattern miniaturization accompanying high integration of semiconductor devices and the like. I have.
Furthermore, in these exposure methods, a technique is proposed in which one large pattern area (for example, one chip of a semiconductor element) is divided into a plurality of small pattern areas (hereinafter, referred to as “subfields”) and exposed while joining them. (For example, see Patent Document 1). According to this divisional exposure, the pattern of the mask can be accurately transferred to the substrate even when the mask itself has distortion.
[0003]
However, when the pattern of the mask is transferred to the substrate by the divided exposure, after the divided exposure, the joining state of adjacent subfields on the substrate may be measured to determine whether or not the joining state is correct. Will be needed.
Therefore, conventionally, a well-known double mark (such as a square shape (see FIG. 8) or a frame shape) is formed in an overlapping area between adjacent subfields, and the subfield is determined based on a shift amount of the center position of the double mark. We were measuring the state of joining. Incidentally, the outside of the double mark is formed with one subfield, and the inside is formed with the other subfield. FIGS. 8A and 8B are a plan view and a sectional view of a square double mark.
[0004]
[Patent Document 1]
JP 2000-124118 A
[0005]
[Problems to be solved by the invention]
However, the overlapping area between adjacent subfields is very small, and the double mark formed in the overlapping area is also very small (for example, 5 μm square). It was not possible to measure well.
[0006]
Enlarging the double mark by giving priority to the accuracy of the joint measurement corresponds to securing a large overlapping area between subfields, and other than the overlapping area in the subfield (transfer area of the original pattern). Is not preferred because it becomes smaller.
SUMMARY OF THE INVENTION It is an object of the present invention to provide a joint measurement device and a division exposure mask that can accurately measure a joint state even when an overlap region between subfields (divided pattern regions) is small.
[0007]
[Means for Solving the Problems]
The invention according to claim 1, wherein the first divided pattern region and the second divided pattern region which are adjacently transferred to the substrate via an elongated overlapping region are formed in the overlapping region, and the first divided pattern region is formed in the first divided pattern region. A first mark group in which a plurality of first line marks perpendicular to the longitudinal direction of the overlapping region formed by transferring the divided pattern regions are arranged at a constant pitch along the longitudinal direction; A plurality of second line marks perpendicular to the longitudinal direction, formed by transferring the divided pattern area, and a second mark group arranged at a constant pitch along the longitudinal direction. A joint measuring device for measuring a joint state between the first divided pattern region and the second divided pattern region, wherein the illumination unit illuminates the mark group; Detecting light, and based on the intensity distribution state of the light that changes according to the amount of phase shift in the longitudinal direction between the first mark group and the second mark group, the first divided pattern area and the second Calculating means for calculating an amount of displacement of joining with the two divided pattern areas.
[0008]
According to a second aspect of the present invention, in the joining measurement apparatus according to the first aspect, the illumination means is parallel to a plane including the longitudinal direction of the overlapping region and a normal direction of the substrate and the method is performed in parallel. The mark group is illuminated by irradiating light in a narrow wavelength band from a direction oblique to the line direction, and the calculating unit detects diffracted light generated from the mark group according to the phase shift amount. And calculating the amount of the displacement based on the angular distribution of the intensity of the diffracted light.
[0009]
The invention according to claim 3 is the splicing measurement device according to claim 1, wherein the illumination unit illuminates the mark group by irradiating light in a wide wavelength band, and the calculation unit includes A diffraction light generated from the mark group is detected according to the phase shift amount, and the joining shift amount is calculated based on a wavelength distribution state of the intensity of the diffraction light.
[0010]
The invention according to claim 4 is a division exposure mask in which a plurality of division pattern areas transferred to different positions on a substrate by division exposure are two-dimensionally arranged, wherein the division pattern area has an elongated shape in a predetermined direction. And two outer edge regions arranged opposite to each other in a direction perpendicular to the predetermined direction, and a plurality of first line marks perpendicular to the predetermined direction are provided in one of the two outer edge regions in the predetermined direction. There is provided a first mark group for joint measurement arranged at a constant pitch along the other, and a plurality of second line marks perpendicular to the predetermined direction are provided in the other of the two outer edge regions in the predetermined direction. A second mark group for joint measurement arranged at the pitch along the first mark group is provided, and the first mark group and the second mark group have a predetermined phase shift amount. .
[0011]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(1st Embodiment)
The first embodiment of the present invention corresponds to claims 1, 2, and 4.
As shown in FIG. 1A, a splicing measurement device 10 according to the first embodiment includes an inspection stage 12 that supports a wafer 11 and an illumination optical system that emits illumination light L1 for the wafer 11 on the inspection stage 12 ( 13 to 17), an image forming optical system (18 to 20) for forming an image of the wafer 11 illuminated by the illumination light L1, a CCD image pickup device 21, and a signal processing device 22.
[0012]
The joint measurement device 10 is a device that measures a joint state of adjacent subfields among a plurality of subfields (described later) transferred to different positions on the wafer 11 by the divided exposure. The subfield corresponds to a “divided pattern area” in the claims. The wafer 11 corresponds to a “substrate”.
Before specifically describing the joint measurement apparatus 10, the wafer 11 to be measured and the joint measurement mark will be described with reference to FIGS.
[0013]
The wafer 11 is in a state after the division exposure and development of the resist film formed on the uppermost layer and before the etching of a predetermined material film immediately adjacent to the resist film. Incidentally, the material film immediately below the resist film is actually processed via the resist film when the result of the joint measurement (described later) by the joint measuring device 10 is good.
[0014]
As shown in FIG. 2A, a plurality of chips 23 are two-dimensionally arranged on the wafer 11. As shown in FIG. 2B, one chip 23 is divided into a plurality of subfields 24 (12 in FIG. 2). The size and shape of one subfield 24 are as shown in FIG.
That is, in each chip 23 (FIG. 2B) of the wafer 11, the plurality of subfields 24 are connected via the overlapping region 25 (hatched portion). The size Xs, Ys of the subfield 24 is, for example, about 100 μm, and the width Lo of the overlapping region 25 is, for example, about 5 μm.
[0015]
The overlapping area 25 between two adjacent subfields 24 has an elongated shape as shown by a bold frame 25a in FIG. 2B. Hereinafter, the elongated overlapping region 25 in the thick line frame 25a is simply referred to as “overlap region 25a”.
Here, in the case of subfields 24 adjacent to each other in the X direction, the longitudinal direction of the overlapping region 25a coincides with the Y direction. Similarly, in the case of subfields 24 adjacent in the Y direction, the longitudinal direction of the overlap region 25a coincides with the X direction. The longitudinal direction of the overlap region 25a corresponds to a measurement direction of a joining deviation amount Δ described later. The overlapping area 25a corresponds to an “overlap area” in the claims.
[0016]
Then, a joint measurement mark 26 (see FIGS. 3B and 3C) is formed in an overlapping area 25a between adjacent subfields 24. FIG. 3B is an enlarged view of the joint measurement mark 26 formed in the portion of the dotted frame 25b (the overlapping region 25a elongated in the Y direction) in FIG. 2B. FIG. 3C is a cross-sectional view of FIG.
[0017]
Although illustration and description are omitted, a joining measurement mark having the same configuration as that obtained by rotating the joining measurement mark 26 by 90 degrees is formed in the overlapping region 25a elongated in the X direction.
With reference to FIGS. 3B and 3C, the joining measurement mark 26 will be described. The joining measurement mark 26 has a plurality of line marks 26A perpendicular to the longitudinal direction (Y) of the overlapping area 25a, and has a constant pitch P along the longitudinal direction (Y). ₁ Similarly, a plurality of line marks 26B perpendicular to the longitudinal direction (Y) have a constant pitch P along the longitudinal direction (Y). ₁ , And a second mark group (26B) arranged in a row.
[0018]
The two mark groups (26A) and (26B) have a pitch P ₁ Are equal line and space marks, and the line widths Lm of the respective line portions (line marks 26A and 26B) are also equal to each other. The line marks 26A and 26B are linear concave portions formed in the uppermost resist film of the wafer 11.
The phase shift amount φ between the two mark groups (26A) and (26B) includes a component of a predetermined design value (a phase shift amount φo in the later-described divided exposure mask 27 in FIG. 5) and a divided exposure value. And the component of the joint displacement amount Δ in the longitudinal direction (Y) that has occurred at some time.
[0019]
Further, one of the two mark groups (26A) and (26B) constituting the joint measurement mark 26 is formed together with one subfield 24 of the overlapping area 25a where the joint measurement mark 26 is formed. , And the other of the mark groups (26A) and (26B) are formed together with the other subfield 24 of the overlapping area 25a. That is, the two mark groups (26A) and (26B) represent the reference positions of two adjacent subfields 24, respectively.
[0020]
Therefore, for example, when the joining state of the subfields 24 adjacent in the X direction is shifted in the Y direction, the phase shift between the two mark groups (26A) and (26B) is made according to the joining shift amount Δ. The amount φ changes and differs from a design value (a phase shift amount φo described later).
[0021]
Further, the wafer 11 and the joint measurement mark 26 are produced through a development process after the division exposure using the exposure apparatus 30 shown in FIG. 4 and the division exposure mask 27 shown in FIG.
As shown in FIG. 4A, the exposure apparatus 30 includes a wafer stage 31 that supports the unexposed wafer 11, a mask stage 32 that supports the divided exposure mask 27, a projection lens 33, and a stage control device. And an illumination light source (not shown). On the unexposed wafer 11, a resist film is formed on the uppermost layer.
[0022]
The exposure apparatus 30 uses a charged particle beam such as an electron beam (EB) or an ion beam or extreme ultraviolet (EUV) light to transfer the pattern of the division exposure mask 27 to the wafer 11 via a projection lens 33. Device. At the time of the division exposure, the wafer stage 31 and the mask stage 32 are controlled synchronously by the stage controller 34 (described later).
[0023]
As shown in FIG. 5A, a plurality of subfields 28 necessary for forming one chip 23 (FIG. 2B) of the wafer 11 are two-dimensionally arranged on the divided exposure mask 27. (12 in the figure). The arrangement pitches of the subfields 28 formed on the division exposure mask 27 in the XY directions are Xp and Yp. The arrangement pitches Xp and Yp are used as the amount of movement of the mask stage 32 during the divided exposure.
[0024]
As shown in FIG. 5B, four mark groups (35A), (35B), (36A), and (36A) for joint measurement are provided in an outer edge region 28a (hatched portion) of each subfield 28. (36B) is formed. FIG. 5B is an enlarged view of a portion of the circular frame 27a in FIG. 5A. Each of the mark groups (35A), (35B), (36A), and (36B) includes a plurality of line marks 35A, 35B, 36A, and 36B.
[0025]
The mark groups (35A) and (35B) opposed to each other in the X direction are the same as the mark groups (36A) and (36B) opposed to each other in the Y direction when one of them is rotated by 90 degrees. Configuration. For this reason, the description of the mark groups (36A) and (36B) arranged to face each other in the Y direction is omitted.
With reference to FIG. 5B, the joining measurement mark groups (35A) and (35B) that are opposed to each other in the X direction will be specifically described. In one mark group (35A), a plurality of line marks 35A perpendicular to the longitudinal direction (Y) of the outer edge region 28a have a constant pitch P along the longitudinal direction (Y). ₂ Are arranged. Similarly, the other mark group (35B) includes a plurality of line marks 35B perpendicular to the longitudinal direction (Y) having a constant pitch P along the longitudinal direction (Y). ₂ Are arranged.
[0026]
The two mark groups (35A) and (35B) have a pitch P ₂ Are equal line and space marks, and the line widths of the respective line portions (line marks 35A, 35B) are also equal to each other. Further, the two mark groups (35A) and (35B) have a phase shift amount φo of a predetermined design value.
Then, the divided exposure using the exposure apparatus 30 (FIG. 4A) having the above configuration and the divided exposure mask 27 (FIG. 5) is performed as follows. For example, subfields 28 (four in FIG. 5A) arranged in the X direction of the division exposure mask 27 are sequentially transferred to different X positions in the chip 23 (FIG. 2B) of the wafer 11. The case will be described.
[0027]
{Circle around (1)} First, of a plurality of sub-fields 28 of the divided exposure mask 27, one sub-field 28 located on the optical axis 33a (FIG. 4A) of the projection lens 33 is illuminated, and its pattern (4 (Including three mark groups (35A), (35B), (36A), and (36B)) on a predetermined area of the wafer 11 on the optical axis 33a.
Thus, a pattern including the mark groups (26A) and (26B) corresponding to the mark groups (35A) and (35B) of the subfield 28 is transferred to a predetermined area of the wafer 11, and the subfield 24 is formed. .
[0028]
{Circle around (2)} Next, the stage controller 34 of the exposure apparatus 30 moves both the wafer stage 31 and the mask stage 32 stepwise in the X direction.
The movement amount of the wafer stage 31 is equal to the difference (Xs-Lo) between the size Xs of the subfield 24 on the wafer 11 side shown in FIG. 2B and the width Lo of the overlapping area 25a to be secured. The amount of movement of the mask stage 32 is equal to the arrangement pitch Xp of the subfields 28 on the side of the divided exposure mask 27 shown in FIG.
[0029]
By moving the mask stage 32 stepwise by "Xp", the subfield 28 next to the divided exposure mask 27 is positioned on the optical axis 33a (see FIG. 4B). In FIG. 4B, in order to distinguish the subfields 28 positioned on the optical axis 33a before and after the step movement, the one before the movement is the subfield 28 (1), and the one after the movement is the subfield 28 (2). ). Hereinafter, the same reference numerals are used in the description.
[0030]
Further, by moving the wafer stage 31 stepwise by "Xs-Lo", the next transfer area 29 (see FIG. 4B) of the wafer 11 is positioned on the optical axis 33a. However, in the transfer area 29, the front end portion 29a in the movement direction (X) of the wafer stage 31 is located at the rear end portion 24a of the subfield 24 formed in the above (1) (the portion where the mark group (26B) has already been transferred). ).
[0031]
(3) After the wafer stage 31 and the mask stage 32 are step-moved as in (2), the subfield 28 (2) of the division exposure mask 27 is illuminated, and its pattern (four mark groups ( 35A), (35B), (36A), and (36B)) are transferred to the transfer area 29 of the wafer 11.
As a result, a pattern including the mark groups (26A) and (26B) corresponding to the mark groups (35A) and (35B) of the subfield 28 (2) is transferred to the transfer area 29 of the wafer 11, and the subfield 24 is transferred. Is formed. Specifically, the mark group (26A) is transferred to the front end portion 29a of the transfer area 29, and the mark group (26B) is transferred to the rear end portion (see FIG. 4C).
[0032]
As a result, at the front end portion 29a of the transfer area 29, the mark group (26A) obtained by the current transfer and the mark group (26B) obtained by the previous transfer overlap each other, as shown in FIGS. 3B and 3C. The joint measurement mark 26 is formed. However, the pattern of the concavo-convex pattern is actually obtained after the development. The front end portion 29a of the transfer region 29 corresponds to the “overlap region 25a” described above.
[0033]
Incidentally, the two sub-fields 28 (1) and (2) of the divided exposure mask 27 are adjacent to each other via the overlapping area 25a (FIG. 2B) when the pattern is transferred to the wafer 11. A mark group (35B) is located in the subfield 28 (1) corresponding to the overlapping area 25a on the wafer 11, and a mark group (35B) is located in the subfield 28 (2) corresponding to the overlapping area 25a. 35A) is provided.
[0034]
(4) Next, the wafer stage 31 and the mask stage 32 are moved stepwise in the same manner as in the above (2), and then in the subfield 28 next to the divided exposure mask 27 in the same manner as in the above (3). Perform pattern transfer.
As described above, the division exposure using the exposure apparatus 30 and the division exposure mask 27 involves pattern transfer of the subfield 28 of the division exposure mask 27 (the above (1), (3),...) And the wafer stage 31 and The step movement (the above (2), (4),...) Of the mask stage 32 is alternately repeated.
[0035]
That is, the plurality of sub-fields 28 formed on the division exposure mask 27 are sequentially transferred to different X positions in the chip 23 (FIG. 2B) of the wafer 11 while being connected via the elongated overlapping region 25a. Will be done.
Then, by developing the wafer 11 after the above-described divided exposure, a joining measurement mark 26 (FIGS. 3B and 3C) is formed in an overlapping area 25a of adjacent subfields 24 of the wafer 11. It will appear as an uneven pattern. In this state, the wafer 11 is placed on the inspection stage 12 of the joint measuring apparatus 10 (FIG. 1) of the first embodiment, and the joint state of the adjacent subfields 24 is measured.
[0036]
Next, a specific configuration of the splicing measurement device 10 will be described.
Although not shown, the inspection stage 12 holds a holder that holds the wafer 11 in a horizontal state, an XY drive unit that drives the holder in the horizontal direction (XY directions), and drives the holder in the rotation direction (θ direction). And a .theta. Drive unit.
The illumination optical system (13 to 17) includes a laser light source 13 arranged eccentrically with respect to the optical axis O1, a collimator lens 14, an illumination field stop 15, and an illumination relay lens 16 arranged sequentially along the optical axis O1. And a half mirror 17. The half mirror 17 has the reflection / transmission surface 17a inclined at an angle of about 45 ° with respect to the optical axis O1, and is also arranged on the optical axis O2 of the imaging optical system (18 to 20). The optical axis O1 of the illumination optical system (13 to 17) is perpendicular to the optical axis O2 of the imaging optical system (18 to 20).
[0037]
The laser light source 13 emits light in a narrow wavelength band (for example, monochromatic light). The collimating lens 14 collimates the light from the laser light source 13. The illumination field stop 15 is an optical element that limits the field of view of the splicing measurement device 10, and has one elongated slit 15a as shown in FIG. The longitudinal direction of the slit 15a coincides with the eccentric direction of the laser light source 13. The illumination relay lens 16 collimates the light from the slit 15 a of the field stop 15.
[0038]
In the illumination optical system (13 to 17), the light emitted from the laser light source 13 uniformly illuminates the illumination field stop 15 from the oblique direction via the collimator lens 14. The light passing through the slit 15a of the illumination field stop 15 is guided to the half mirror 17 via the illumination relay lens 16, and is reflected by the reflection / transmission surface 17a (illumination light L1). To 20) on the optical axis O2.
[0039]
The imaging optical system (18 to 20) includes a first objective lens 18, a second objective lens 19, and a relay lens 20 arranged in order along the optical axis O2. The half mirror 17 of the illumination optical system (13 to 17) is disposed between the first objective lens 18 and the second objective lens 19.
[0040]
Then, the first objective lens 18 of the imaging optical system (18 to 20) receives and condenses the illumination light L1 from the half mirror 17 of the illumination optical system (13 to 17). Thus, the wafer 11 on the inspection stage 12 is illuminated by the illumination light L1 transmitted through the first objective lens 18 from a direction oblique to the normal direction of the wafer 11. The normal direction of the wafer 11 is parallel to the normal direction of the holder of the inspection stage 12.
[0041]
The above-described illumination optical systems (13 to 17) and the first objective lens 18 correspond to “illumination means” in the claims. The image forming optical system (18 to 20), the CCD image pickup device 21, and the signal processing device 22 correspond to "calculating means" in the claims.
Here, the angle at which the illumination light L1 is incident on the wafer 11 (the angle with respect to the normal direction of the wafer 11) is determined according to the eccentricity α of the laser light source 13. The eccentricity α of the laser light source 13 (that is, the incident angle of the illumination light L1) is used for the joint measurement formed in the overlapping region 25a (FIGS. 3B and 3C) of the adjacent subfields 24 of the wafer 11. Pitch P of mark 26 ₁ It is preferable to determine in consideration of the following.
[0042]
The range in which the illumination light L1 is incident on the wafer 11 corresponds to the point hatched area 10a in FIGS. 6A and 6B, and has an elongated shape. The point hatched area 10a is an area corresponding to the slit 15a of the illumination field stop 15 having a conjugate positional relationship with the wafer 11, and corresponds to a projected image of the slit 15a. Hereinafter, the point hatched area 10a is referred to as “illumination area 10a”.
[0043]
Further, the illumination area 10a on the wafer 11 has an elongated shape in the Y direction. This is because the longitudinal direction of the slit 15a of the illumination field stop 15 is parallel to the Z direction (the direction of the optical axis O2), and the light passing through the slit 15a is deflected by 90 ° by the half mirror 17.
The incident direction of the illumination light L1 and the longitudinal direction (Y direction) of the illumination area 10a have the following relationship. That is, the incident direction of the illumination light L1 is parallel to a plane including the longitudinal direction of the illumination area 10a and the normal direction of the wafer 11. The reason is that the longitudinal direction of the slit 15a and the eccentric direction of the laser light source 13 are matched.
[0044]
Further, the length Y of the illumination area 10a ₁₀ Is the width Y corresponding to one chip 23 of the wafer 11 ₂₃ It is set longer. Also, the width X of the illumination area 10a ₁₀ Is set to be smaller than the width Lo (FIG. 3B) of the overlapping region 25a between the subfields 24 of the wafer 11.
Therefore, according to the illumination light L1 (illumination area 10a), only the joint measurement mark 26 formed in the overlapping area 25a of the subfields 24 of the wafer 11 can be illuminated from an oblique direction (FIG. 6 (c)). FIG. 6A shows a state in which three overlapping regions 25a (joining measurement marks 26) are simultaneously illuminated.
[0045]
From the joint measurement mark 26 in the illumination area 10a of the wafer 11 irradiated with the illumination light L1, the diffracted light according to the phase shift φ between the two mark groups (26A) and (26B). L2 occurs. That is, the diffracted light L2 is generated with an angular distribution of intensity according to the phase shift amount φ. This diffracted light L2 is guided to the first objective lens 18.
[0046]
The first objective lens 18 (FIG. 1) collimates the diffracted light L2 generated from the wafer 11. The diffracted light L2 collimated by the first objective lens 18 passes through the half mirror 17 and enters the second objective lens 19. The second objective lens 19 focuses the diffracted light L2 from the half mirror 17. The relay lens 20 re-images the diffracted light L2 from the second objective lens 19 on the imaging surface of the CCD imaging device 21.
[0047]
As described above, the diffracted light L2 from the wafer 11 is primary-imaged through the first objective lens 18, the half mirror 17, and the second objective lens 19, and the light from the second objective lens 19 is Is re-imaged on the image pickup surface of the CCD image pickup device 21.
Here, as described above, the intensity of the diffracted light L2 generated from the wafer 11 depends on the phase shift amount φ of the mark groups (26A) and (26B) of the joining measurement marks 26 in the illumination area 10a. Has an angular distribution. Then, the angular distribution of the intensity of the diffracted light L2 is converted into a light / dark distribution on the imaging surface of the CCD imaging device 21. That is, on the imaging surface of the CCD imaging device 21, a diffraction image (image by the diffracted light L2) having a light and dark distribution corresponding to the phase shift amount φ of the mark groups (26A) and (26B) is formed.
[0048]
The CCD image sensor 21 is an area sensor in which a plurality of pixels are two-dimensionally arranged, captures the above-described diffraction image (light / dark distribution according to the phase shift amount φ), and outputs an image signal to the signal processing device 22. . The image signal represents a distribution relating to the luminance value of each pixel on the imaging surface of the CCD imaging device 21 (that is, a light / dark distribution according to the phase shift amount φ).
[0049]
When the signal processing device 22 captures the image signal output from the CCD image pickup device 21, the signal processing device 22 detects, based on the brightness distribution (that is, the light / dark distribution according to the phase shift amount φ), the displacement of the connection between the subfields 24 of the wafer 11. The quantity Δ is calculated, and the joining state is measured based on the calculation result.
The calculation of the joining shift amount Δ is performed by (1) calculating the phase shift amount φ of the mark groups (26A) and (26B) of the joining measurement marks 26 from the luminance distribution of the image signal by a known simulation, and (2) This is performed by comparing the obtained positional deviation amount φ with a predetermined design value (a phase deviation amount φo in the divided exposure mask 27 in FIG. 5). The procedure (2) is performed by, for example, subtracting the phase shift amount φo of the design value from the position shift amount φ calculated.
[0050]
Further, instead of the procedures (1) and (2), the joining deviation amount Δ may be calculated by the following method. That is, a library is prepared in the signal processing device 22 in advance, and the correlation between the information on the luminance distribution (brightness / darkness distribution according to the phase shift amount φ) of the image signal output from the CCD image sensor 21 and the storage information of the library is provided. By performing the calculation, the joining deviation amount Δ may be calculated.
[0051]
The joint displacement amount Δ calculated by the signal processing device 22 is calculated using the joint measurement mark 26 in which the line marks 26A and 26B are arranged along the Y direction as shown in FIG. At the time of the division exposure using the apparatus 30 and the division exposure mask 27, a movement error when the wafer stage 31 and the mask stage 32 are step-moved in the Y direction is considered to be a main cause.
[0052]
For this reason, the stitching deviation Δ in the Y direction calculated by the signal processing device 22 (the measurement result of the stitching measurement device 10) is fed back to the stage controller 34 of the exposure device 30 to perform the stitching at the next divided exposure. By finely adjusting the movement amount (Yp) of the mask stage 32 in the Y direction based on the shift amount Δ, highly accurate step movement in the Y direction is realized.
[0053]
Similarly, in order to calculate the displacement Δ in the X direction between the subfields 24 of the wafer 11, first, the wafer 11 is rotated by 90 ° together with the holder of the inspection stage 12. Then, the joint measurement mark in the overlapping area 25a which is elongated in the X direction is illuminated by the illumination light L1 (illumination area 10a), and the joint in the X direction is obtained based on the angular distribution of the intensity of the obtained diffracted light L2. The shift amount Δ is calculated.
[0054]
The displacement amount Δ in the X direction is mainly caused by a movement error when the wafer stage 31 and the mask stage 32 are step-moved in the X direction during the divided exposure using the exposure apparatus 30 and the divided exposure mask 27. Conceivable. The measurement result is fed back to the exposure apparatus 30, and the amount of movement (Xp) of the mask stage 32 in the X direction is finely adjusted, so that a highly accurate step movement in the X direction can be realized in the next divided exposure.
[0055]
As described above, in the joint measuring apparatus 10 of the first embodiment, the joint measuring mark 26 (line and line) formed in the overlapping region 25a of the adjacent subfields 24 of the wafer 11 along its longitudinal direction. Since the (space mark) is used, even if the overlapping area 25a is very small (for example, 5 μm in width), the joining state between adjacent subfields 24 can be accurately measured.
[0056]
Further, in the splicing measurement device 10, as shown in FIG. ₁₀ Is the width Y of one chip 23 ₂₃ In order to simultaneously illuminate all the overlapping regions 25a (joining measurement marks 26) existing in one direction in the chip 23, the connection state of the subfields 24 in one chip 23 is set. The overall trend can be measured at high speed, and the throughput is improved.
[0057]
Further, the joint measurement device 10 illuminates the joint measurement mark 26 in the overlap region 25a from an oblique direction using light in a narrow wavelength band (for example, monochromatic light), and thereby measures the intensity of the diffracted light L2 generated thereby. Since the joining state is measured based on the angle distribution, a measurement result averaged in the bright region 10a can be obtained. For this reason, the apparent SN improves.
[0058]
In addition, in the splicing measurement apparatus 10, the overlapping area 25a between the adjacent subfields 24 of the wafer 11 can be reduced while maintaining the accuracy of the splicing measurement at a high level. A large pattern transfer area) can be secured, and good division exposure can be performed.
Further, in the first embodiment, for the joint measurement by the joint measurement device 10, a line and space mark (26) is formed in the overlapping region 25a of the subfields 24 of the wafer 11, so that the joint measurement is performed. The line-and-space mark obtained later and after the etching process can be used as wiring to connect patterns between adjacent subfields 24.
[0059]
(Modification of First Embodiment)
In the first embodiment described above, the pitch P of the joint measurement mark 26 formed in the overlap region 25a of the wafer 11 is ₁ In consideration of the above, the eccentric amount α of the laser light source 13 (that is, the incident angle of the illumination light L1) is determined, but the present invention is not limited to this.
For example, while changing the amount of eccentricity α of the laser light source 13, a light-dark pattern of a diffraction image on the imaging surface of the CCD image sensor 21 may be observed, and the eccentric amount α may be set so that the light-dark pattern becomes clearest. . Such a setting may be performed manually by the judgment of the operator or automatically by the judgment of the signal processing device 22. In the case of performing this automatically, it is preferable to set the eccentricity α so that the amplitude of the light and dark pattern of the diffraction image is maximized.
[0060]
Further, instead of changing the eccentricity α of the laser light source 13, even when the wavelength λ of light in a narrow wavelength band (for example, monochromatic light) emitted from the laser light source 13 is continuously changed, similarly, It is possible to select a setting that makes the light and dark pattern of the diffraction image clearest.
Further, each time the amount of eccentricity α (or wavelength λ) of the laser light source 13 is changed, an image signal corresponding to the light and dark pattern of the diffraction image on the imaging surface of the CCD image sensor 21 is taken in, and the amount of eccentricity α (or wavelength λ) is obtained. The splicing measurement may be performed based on a number of different image signals. In this case, since the amount of information increases, the accuracy of the joint measurement is improved.
[0061]
In the above-described first embodiment, the joint measurement is performed by simultaneously illuminating all the overlapping regions 25 a (joint measurement marks 26) existing in one direction in the chip 23 of the wafer 11. The joint measurement may be performed for each overlapping region 25a. Such a measurement is realized by shortening the slit 15a of the illumination field stop 15.
[0062]
Furthermore, in the first embodiment described above, when switching between the joining measurement in the X direction and the joining measurement in the Y direction, the wafer 11 is rotated by 90 ° together with the holder of the inspection stage 12. The light source 13 may be rotated by 90 °. However, a state in which the longitudinal direction of the slit 15a and the eccentric direction of the laser light source 13 must be maintained.
[0063]
(2nd Embodiment)
The second embodiment of the present invention corresponds to claims 1, 3, and 4.
The joint measuring device 50 according to the second embodiment includes an incandescent lamp 51 in place of the laser light source 13 of the above-described joint measuring device 10 (FIG. 1), a spectroscope 52 in place of the CCD image sensor 21, and a signal A signal processing device 53 is provided instead of the processing device 22.
[0064]
The incandescent lamp 51 is arranged on the optical axis O1, and emits light in a wide wavelength band (for example, white light). Therefore, according to the illumination light L1 (illumination area 10a) of FIG. 6, only the joint measurement mark 26 formed in the overlapping area 25a of the subfields 24 of the wafer 11 can be illuminated from the vertical direction.
Thereafter, from the joint measurement mark 26 in the illumination area 10a of the wafer 11, the diffracted light L2 is generated according to the phase shift amount φ (FIG. 6B) between the two mark groups (26A) and (26B). appear. That is, the diffracted light L2 is generated with a wavelength distribution of intensity corresponding to the phase shift amount φ. After passing through the imaging optical system (18 to 20), the light enters the spectroscope 52.
[0065]
The spectroscope 52 detects the wavelength distribution of the intensity of the diffracted light L2, and outputs the detection result (spectral signal) to the signal processing device 53. The signal processing device 53 calculates a joining shift amount Δ between the subfields 24 of the wafer 11 based on the spectrum signal output from the spectroscope 52, and measures the joining state based on the calculation result.
[0066]
The joint displacement Δ is calculated by (3) the phase shift φ of the mark groups (26A) and (26B) of the joint measurement marks 26 from the spectrum signal by a known simulation, and (4) the obtained position. This is performed by comparing the shift amount φ with a predetermined design value (the phase shift amount φo in the divided exposure mask 27 in FIG. 5). The procedure of (4) is performed, for example, by subtracting the phase shift amount φo of the design value from the position shift amount φ calculated.
[0067]
Further, instead of the procedures (3) and (4), the joining deviation amount Δ may be calculated by the following method. That is, a library is prepared in the signal processing device 53 in advance, and a correlation operation is performed between information on a spectral signal (wavelength distribution corresponding to the phase shift amount φ) output from the spectroscope 52 and information stored in the library. May be used to calculate the joint displacement amount Δ.
[0068]
As described above, also in the joint measuring apparatus 50 of the second embodiment, the joint measuring mark 26 (line and line) formed in the overlapping region 25a of the adjacent subfields 24 of the wafer 11 along its longitudinal direction. Since the (space mark) is used, even if the overlapping area 25a is very small (for example, 5 μm in width), the joining state between adjacent subfields 24 can be accurately measured.
[0069]
Also, the joint measurement device 50 can measure the overall tendency of the connection state of the subfields 24 in one chip 23 at high speed similarly to the above-described joint measurement device 10, thereby improving the throughput. Since a large area (transfer area of the original pattern) other than the overlapping area 25a in the subfield 24 can be ensured, good division exposure can be performed.
[0070]
Further, the joint measurement device 50 illuminates the joint measurement mark 26 in the overlapping region 25a from the vertical direction using light in a wide wavelength band (for example, white light), and the wavelength of the intensity of the diffracted light L2 generated thereby. Since the joining state is measured based on the distribution, simple and inexpensive measurement is possible.
(Modification)
In the first and second embodiments described above, the joint measurement is performed based on the diffracted light L2 generated from the joint measurement mark 26 in the overlapping area 25a of the subfields 24 of the wafer 11, but the joint measurement Since the pattern derived from the structure also appears in the scattered light generated from the mark 26, the joining measurement may be performed based on the scattered light. In this case, it is conceivable to use an optical CD length measuring device as a scattered light detector.
[0071]
Further, in the first and second embodiments described above, as shown in the divisional exposure mask 27 shown in FIG. ₂ The pitch P as shown in FIG. 3B is formed on the wafer 11 by using two mark groups (35A) and (35B) whose line widths are equal to each other. ₁ And two line groups (26A) and (26B) having the same line width Lm are formed, and the joint measurement is performed using these marks. The same measurement is performed even when the line widths of the mark groups are different. be able to.
[0072]
【The invention's effect】
As described above, according to the present invention, the joining state can be accurately measured even if the overlapping area between the subfields (divided pattern areas) is small.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating an overall configuration of a splicing measurement device 10 according to a first embodiment.
FIG. 2 is a diagram illustrating chips 23 (a) on a wafer 11 and a plurality of subfields 24 (b).
FIGS. 3A and 3B are diagrams showing configurations (b) and (c) of one subfield 24 (a) and a joining measurement mark 26. FIG.
FIGS. 4A and 4B are schematic views (a) of the exposure apparatus 30 and FIGS.
FIG. 5 is a diagram showing a plurality of subfields (a) on a division exposure mask 27 and a configuration (b) such as mark groups (35A) and (35B).
FIG. 6 is a diagram illustrating an illumination area 10a and an incident direction of illumination light L1 by the joint measurement device 10 according to the first embodiment.
FIG. 7 is a diagram illustrating an overall configuration of a joining measurement apparatus 50 according to a second embodiment.
FIG. 8 is a schematic diagram illustrating a conventional double mark.
[Explanation of symbols]
10,50 splicing measuring device
11 Wafer
12 Inspection stage
13 Laser light source
14 Collimating lens
15 Illumination field stop
16 Lighting relay lens
17 Half mirror
18 First objective lens
19 Second objective lens
20 relay lens
21 CCD image sensor
22,53 signal processor
23 chips
24, 28 subfield
25, 25a overlapping area
26 Splice Measurement Mark
26A, 26B, 35A, 35B, 36A, 36B Line mark
27 Division exposure mask
28a Outer edge area
29 Transcription area
30 Exposure equipment
31 Wafer stage
32 mask stage
33 Projection lens
34 Stage control device
51 Incandescent light bulb
52 Spectrometer

Claims (4)

The first divided pattern area is formed by being transferred to the overlapping area of the first divided pattern area and the second divided pattern area adjacently transferred to the substrate via the elongated overlapping area, and the first divided pattern area is transferred and formed. A first mark group in which a plurality of first line marks perpendicular to the longitudinal direction of the overlap region are arranged at a constant pitch along the longitudinal direction, and the second divided pattern region are transferred and formed. A plurality of second line marks perpendicular to the longitudinal direction, and a second mark group arranged at a constant pitch along the longitudinal direction. A joint measurement device for measuring a joint state with the second divided pattern region,
Lighting means for illuminating the mark group;
Detecting light generated from the mark group, based on the intensity distribution state of the light that changes according to the amount of phase shift between the first mark group and the second mark group in the longitudinal direction; A stitching measurement device, comprising: a calculating unit that calculates a stitching shift amount between the split pattern region and the second split pattern region. The splicing measurement device according to claim 1,
The illuminating unit is configured to irradiate light in a narrow wavelength band from a direction oblique to the normal direction and parallel to a plane including the longitudinal direction of the overlapping region and a normal direction of the substrate, and Illuminate the mark group,
The calculating means detects diffracted light generated from the mark group according to the phase shift amount, and calculates the stitching shift amount based on an angular distribution state of the intensity of the diffracted light. Combination measuring device. The splicing measurement device according to claim 1,
The illumination means illuminates the mark group by irradiating light in a wide wavelength band,
The calculating means detects diffracted light generated from the mark group according to the phase shift amount, and calculates the joining shift amount based on the wavelength distribution state of the intensity of the diffracted light. Combination measuring device. A divided exposure mask in which a plurality of divided pattern areas transferred to different positions on the substrate by the divided exposure are two-dimensionally arranged,
The divided pattern region has an elongated shape in a predetermined direction and includes two outer edge regions arranged to face each other in a direction perpendicular to the predetermined direction,
In one of the two outer edge regions, a first mark group for joint measurement in which a plurality of first line marks perpendicular to the predetermined direction are arranged at a constant pitch along the predetermined direction is provided.
In the other of the two outer edge regions, a second mark group for joint measurement in which a plurality of second line marks perpendicular to the predetermined direction are arranged at the pitch along the predetermined direction is provided,
A division exposure mask, wherein the first mark group and the second mark group have a predetermined phase shift amount.

Images