CN100570491C - The pattern image setting of step-by-step exposure machine and multiplying power error compensation method - Google Patents
The pattern image setting of step-by-step exposure machine and multiplying power error compensation method Download PDFInfo
- Publication number
- CN100570491C CN100570491C CNB2006100360820A CN200610036082A CN100570491C CN 100570491 C CN100570491 C CN 100570491C CN B2006100360820 A CNB2006100360820 A CN B2006100360820A CN 200610036082 A CN200610036082 A CN 200610036082A CN 100570491 C CN100570491 C CN 100570491C
- Authority
- CN
- China
- Prior art keywords
- pattern image
- error
- multiplying power
- alignment
- magnification error
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Exposure And Positioning Against Photoresist Photosensitive Materials (AREA)
- Exposure Of Semiconductors, Excluding Electron Or Ion Beam Exposure (AREA)
Abstract
The invention discloses a kind of pattern image setting and multiplying power error compensation method of step-by-step exposure machine, wherein the pattern image setting method is: (A) obtain pattern image allow alignment error with respect to the difference value of the magnification error on the two-dimensional directional of target exposure region; And (B) allow that according to this alignment error and this different difference decide the width value of pattern image, and make this pattern image length value this width value greater than twice; Multiplying power error compensation method is: (a) long axis direction of the long limit parallel direction of setting and pattern image, measure at the magnification error between target exposure region and pattern image on the long axis direction; Reach when (b) being not equal to zero as if this magnification error, the corresponding projection multiplying power of adjusting step-by-step exposure machine, it is zero making pattern image and the magnification error of target exposure region on long axis direction.Method of the present invention can improve the alignment accuracy and the exposure efficiency of step-by-step exposure machine.
Description
Technical field
The present invention relates to a kind of step printing machine technology, refer in particular to the pattern image setting and the multiplying power error compensation method of a kind of step-by-step exposure machine of the alignment accuracy that is used to improve step-by-step exposure machine and exposure rate.
Background technology
Exposure manufacture process is the some in the micro-photographing process (photo-Iithography), and micro-photographing process is widely used in for example semiconductor (Semiconductor), display panels (Liquid crystal display) and printed circuit board (PCB) industries such as (Printedcircuit board).Exposure method is utilization one exposure machine usually, on design transfer to the sensing optical activity substrate on the light shield.
Substrate also needs the multiprogramming through baking, removing photoresistance, etching or the like and so on except that through exposing, substrate is produced dwindle or distortion such as amplification, bending as size (scale), and this type of distortion has increased the degree of difficulty that carries out exposure manufacture process.
Processing procedure with multilayer (Multi-layer) printed circuit board (PCB) is an example, produce required multilayer circuit structure, needs through exposure manufacture process for several times.Yet, the manufacturing of each layer printed circuit board is except the process exposure manufacture process, still need pass through other different processing procedure, in the middle of may comprise hot processing procedure (Thermal process), for example cleaning, film forming, exposure, baking, and etching etc., this type of processing procedure all may cause printed circuit board (PCB) to produce change in size not of uniform size; Learn according to experimental data, the change in size of printed circuit board (PCB) about 250ppm (parts per million) between the 500ppm, and with the printed circuit board (PCB) of the above-mentioned hot processing procedure of a collection of process, the difference of plate and plate change in size each other also can reach the degree of 100ppm.
Therefore, when each layer printed circuit board was exposed, the size of the pattern image of its exposure on this layer printed circuit board must change according to the physical size of printed circuit board (PCB) and be adjusted.In order to satisfy the demand that to adjust this pattern image size, force the printed circuit board (PCB) dealer to abandon tradition gradually and can't adjust the contact exposure of pattern image size (contactexposure) equipment, change employing into and can adjust contactless (contact less) apparatus for projection exposure of pattern image size, as be widely used in the step-by-step exposure machine of semiconductor industry.Step-by-step exposure machine can utilize the projection multiplying power that changes projection lens's group, adjusts the size of this pattern image.
Consult Fig. 1, traditional step-by-step exposure machine 1 comprises the projection lens's group 11 that can adjust the projection multiplying power, be used for the graphic pattern projection on the light shield 12 is imaged on the sensing optical activity substrate 13, step-by-step exposure machine 1 also comprises the light shield microscope carrier 14 that is used for fixing light shield 12 and the relative position of projection lens's group 11, be used for fixing the baseplate carrier 15 of substrate 13 and the relative position of projection lens's group 11, and the control module 10 of each member 11,14,15 running of control.Have a plurality of pattern units 121 on the light shield 12, and also have a plurality of modular units 131 on the sensing optical activity substrate 13 with arranged with arranged.The size of modular unit 131 is generally the physical size of a chip (chip or die), and the pattern in the pattern unit 121 is wherein one deck circuit structure pattern of this chip usually, and the circuit structure through several layers of heaps can form complete chip.When desiring to expose, at first need respectively with light shield 12 and sensing optical activity substrate 13 is placed on the light shield microscope carrier 14 and baseplate carrier 15 on, utilize projection lens's group 11 that pattern unit 121 is imaged on the corresponding modular unit 131.
Because projection lens's group 11 of step-by-step exposure machine 1 is subject to the eyeglass caliber size, has a maximum exposure scope (workingfield), can't look like traditional contact exposure equipment, light shield 1 is directly abutted on the substrate 13, in the single exposure process, all pattern units 121 on the light shield 12 all are projected on the respective modules unit 131 on the substrate 13, but need be according to the maximum exposure scope of projection lens's group 11, the pattern area 122 of decision maximum feasible, make the size of this pattern area 122 by projection lens's group 11 formed pattern images 123 less than the maximum exposure scope, make pattern image 123 be projected in default target exposure region 132 on the substrate 13 by projection lens's group 11.
Consult Fig. 2, for pattern image 123 can be coincided (overlay) exactly on target exposure region 132, can on pattern image 123 and target exposure region 132, set the reference point of a plurality of mutual correspondences respectively, whether the position of correction reference point overlaps by measuring also, as judging whether pattern image 123 accurately coincides in the foundation of target exposure region 132.Generally speaking, these reference points are so-called alignment mark (alignment mark).At when operation, the coordinate of the actual alignment mark 133 on the first measurement target exposure region 132, and contrast with the coordinate of the default alignment mark 124 that is stored in the pattern image 123 in the control module 10 in advance.Actual alignment mark 133 is an example with three corners of target exposure region 132 or pattern image 123 respectively with default alignment mark 124 in the drawings, in practical application, this alignment mark 133,124 is not limited to the corner, can be cross, square, rectangle, or any feature that obviously can be used as identification, be not limited in the inside of pattern image 123 or target exposure region 132, can be positioned at the outside of pattern image 123 or target exposure region 132 yet.Generally speaking, certainly existing either large or small position deviation between actual alignment mark 133 and the default alignment mark 124, go out three kinds of alignment errors (alignment error) that can directly be compensated according to these position deviation definables, be respectively skew (shift) error, rotation (rotate) error and multiplying power (scale) error by step-by-step exposure machine 1 corresponding mechanism.
The offset error (see figure 3) is meant the position generation displacement of the position of target exposure region 132 with respect to pattern image 123, the rotation error (see figure 4) is meant that target exposure region 132 has an anglec of rotation with respect to pattern image 123, and the magnification error (see figure 5) is meant the multiplying power variation that the size of target exposure region 132 is amplified or dwindled with respect to the size generation of pattern image 123.Offset error and rotation error are to drive substrate 13 translations or rotation is compensated by baseplate carrier 15, and magnification error is to be amplified or the projection multiplying power of dwindling pattern image 123 is compensated by projection lens's group 11.In addition, what mention is that true origin herein is the geometric center that is defined in pattern image 123 or target exposure region 132.
For the user, the characteristic of the size meeting remote effect product of alignment error, therefore, when formulating the design criteria (design rule) of product, can set one simultaneously and allow alignment error Er, if pattern image 123 is allowed alignment error Er with respect to the alignment error of target exposure region 132 less than this, promptly can not impact this product.
In addition, when practical application, can choose an integer adjacent pattern unit 121 usually, avoid choosing the non-integer pattern unit 121 of (referring to non-) for complete one as pattern area 122.Because incomplete pattern unit 121 formed pattern images, need imperfect pattern unit 121 formed pattern images with another correspondence, using the mode of sewing up (stitch) is made up, to form a pattern image with complete circuit, and the mode of this stitching comprises independently exposure process of secondary, its pattern image that exposes for the first time also may cause unnecessary alignment error with the pattern image of exposure for the second time, this alignment error easily causes the bonding error of circuit structure, and then influences circuit characteristics such as resistance, electric capacity.
Consult Fig. 6, for tradition is used for the exposure method process flow diagram of step-by-step exposure machine 1, this exposure method is the coordinate of three actual alignment marks 133 on the measurement target exposure region 13 in step 21 earlier; In step 22, go out the offset error and the rotation error of X and Y direction according to the coordinate Calculation of alignment mark 133; Reach the magnification error that in step 23, goes out X and Y direction according to the coordinate Calculation of alignment mark 133; In step 24, adjust the projection multiplying power of projection lens's group again, and the position of adjusting light shield 12 or substrate 13 in step 25 is to eliminate the offset error and the rotation error of X and Y direction, to eliminate alignment error according to the mean value of the magnification error of X and Y direction; In step 26, target exposure region 132 is exposed at last.
Consult Fig. 1, when target exposure region 132 expose finish after, utilize baseplate carrier 15 to drive substrates 13 and move, pattern image 123 is exposed corresponding to the next target exposure region of substrate 13 and repeat above-mentioned steps, promptly with alleged displacement and repetition (step; Repeat) mode is carried out complete exposure to substrate 13.
Consult Fig. 7, yet in the processing procedure of printed circuit board (PCB), because the dimensional variation that causes target exposure region 132 through overheated processing procedure is often for anisotropy (anisotropic).That is to say, because the material behavior of printed circuit board (PCB) own, be different at X with degree on the Y direction through making dimensional variation that target exposure region 132 produces behind the overheated processing procedure.Because the dimensional variation of target exposure region 132 directly has influence on the magnification error of target exposure region 132 with respect to pattern image 123, thereby cause the magnification error of X and Y direction also inequality.
Because projection lens's group of step-by-step exposure machine can't be compensated respectively with regard to the magnification error of X and Y direction, traditional multiplying power error compensation mode can be adjusted the projection multiplying power of projection lens's group according to the mean value of the magnification error of X and Y direction, to compensate above-mentioned anisotropic dimensional variation.Yet such compensation result that compensation way produced still can cause the pattern image 123 can't be exactly in alignment with target exposure region 132 as shown in Figure 8.This pattern image 123 after the compensation among the figure still exists one to aim at error E rx with respect to target exposure region 132 at directions X, and the Y direction still exists one to aim at error E ry.
For instance, as shown in Figure 7, the length of not making the directions X of the preceding pattern image 123 of multiplying power compensation is Ix, the length of Y direction is Iy, and the dimensional variation of the directions X of target exposure region 132 is Δ X (Δ X is the length ratio of target exposure region 132 with respect to the directions X of pattern image 123), the dimensional variation of Y direction is Δ Y (Δ Y is the length ratio of this target exposure region 132 with respect to the Y direction of pattern image 123), the alignment error Erx that can calculate directions X is the dimensional variation Δ X (Erx=Ix Δ X) that the length Ix of directions X is multiplied by directions X, in like manner, the alignment error Ery of Y direction is the dimensional variation Δ Y (Ery=Iy Δ Y) that the length Iy of Y direction is multiplied by the Y direction.
When carrying out the compensation of magnification error, projection lens's group can be adjusted the projection multiplying power according to the mean value ((Δ X+ Δ Y)/2) of the magnification error of X and Y direction, therefore, under the prerequisite of not considering offset error and rotation error, as shown in Figure 8, pattern image 123 after the adjustment multiplying power, the alignment error Erx of its directions X is dimensional variation Δ X (Erx=Ix (Δ X-((Δ X+ Δ Y)/2)))=Ix ((Δ X-Δ Y)/2) that the length Ix of directions X is multiplied by directions X; In like manner, the alignment error Ery of its Y direction is dimensional variation Δ Y (Ery=Iy (Δ Y-((Δ X+ Δ Y)/2)))=Iy ((Δ Y-Δ X)/2) that the length Iy of Y direction is multiplied by the Y direction.Therefore, can learn, when step-by-step exposure machine is handled anisotropic distortion in traditional multiplying power error compensation mode, pattern image 123 exposure region 132 that can't aim at the mark exactly, and still exist magnification error therebetween.
In order above-mentioned magnification error will to be reduced to minimum, improving the alignment accuracy of pattern image with respect to this target exposure region, the U.S. the 6th, 580 has proposed a kind of deformation compensation method of projection exposure system therefor in No. 494 patent cases.
Utilize the pattern image 123 among Fig. 8 to illustrate, this pattern image 123 is the matrix of 3x3, alignment error corresponding on directions X is Erx, the 1x3 matrix of a hurdle (column) is as desiring the projected pattern image in the middle of supposing to choose in this 3x3 matrix, can find clearly that its alignment error on directions X is that Erx ' and Erx ' are less than Erx.That is to say that this pattern image 123 is to become positively related with the length of the directions X of this pattern image 123 with respect to the magnification error of the directions X of this target exposure region 132.In like manner, this pattern image 123 also is to become positively related with the Y direction length of this pattern image 123 with respect to the magnification error of the Y direction of this target exposure region 132.
The feature of this deformation-compensated method promptly is to utilize above-mentioned magnification error to become positively related notion with the length of pattern image, with the X and the Y direction length of the pattern area on the light shield, includes the weight factor when calculating this projection lens's group and adjust the projection multiplying power in.If during the equal in length of the X of pattern area and Y direction, the calculating of projection multiplying power with generally be as good as, but when if the length of the X of pattern area and Y direction is unequal, the calculating of the projection multiplying power of projection lens's group just can be included the influence of the length of X and Y direction in.
For instance, as shown in Figure 7, the directions X length of not making this preceding pattern image 123 of multiplying power compensation is Ix, Y direction length is Iy, and the dimensional variation of the directions X of target exposure region 132 is Δ X (Δ X is the length ratio of target exposure region 132 with respect to the directions X of pattern image 123), the dimensional variation of Y direction is Δ Y (Δ Y is the length ratio of target exposure region 132 with respect to the Y direction of pattern image 123), the alignment error Erx that can calculate directions X is the dimensional variation Δ X (Erx=Ix Δ X) that the length Ix of directions X is multiplied by directions X, in like manner, the alignment error Ery of Y direction is the dimensional variation Δ Y (Ery=Iy Δ Y) that the length Iy of Y direction is multiplied by the Y direction.
When carrying out the compensation of magnification error, the adjustment of projection multiplying power no longer is that traditional mean value ((Δ X+ Δ Y)/2) according to X and Y direction magnification error adjusts, but adjust according to ((Ix Δ X+Iy Δ Y)/(Ix+Iy)), meaning promptly, include the length of X and Y direction in calculating as the projection multiplying power of pattern image 123, therefore, under the prerequisite of not considering offset error and rotation error, as shown in Figure 8, pattern image 123 after the adjustment multiplying power, the alignment error Erx of its directions X is dimensional variation Δ X (Erx=Ix (Δ X-((Ix Δ X+Iy Δ Y)/(Ix+Iy))))=IxIy (Δ X-Δ Y)/(Ix+Iy) that the length Ix of directions X is multiplied by directions X; In like manner, the alignment error Ery of its Y direction is dimensional variation Δ Y (Ery=Iy (Δ Y-((Ix Δ X+Iy Δ Y)/(Ix+Iy))))=IxIy (Δ Y-Δ X)/(Ix+Iy) that the length Iy of Y direction is multiplied by the Y direction.
Under the order of magnitude of only considering Erx and Ery, (Δ X-Δ Y) or (Δ Y-Δ X) all can be represented with Δ S, and compare with traditional multiplying power compensation way, and the ratio of actual Ix of substitution and Iy, as shown in the table:
Under the situation of the length ratio Ix/Iy=1 that can learn in the directions X of pattern image and Y direction, through the U.S. the 6th, 580, magnification error after the multiplying power error compensation method compensation that No. 494 patent cases are proposed is identical with magnification error after traditional multiplying power compensation way compensates, but if under the situation of length ratio Ix/Iy ≠ 1 of the directions X of pattern image and Y direction, the U.S. the 6th, 580, the magnification error of the more traditional multiplying power mode of the magnification error of No. 494 patent cases (getting the maximal value of directions X magnification error Erx and Y direction magnification error Ery) is little, has effectively reduced magnification error really.
Yet the U.S. the 6th, 580, the shortcoming of the multiplying power compensation method that proposes in No. 494 patent cases, be that this compensation method is based on traditional multiplying power error compensation method, and utilize at least three alignment marks on the measurement target exposure region, with the X that calculates this target exposure region respectively and the magnification error of Y direction, since the length of more having introduced this pattern image as weight factor to calculate the projection multiplying power, though therefore effectively reduced magnification error, but also make the more traditional multiplying power error compensation mode of calculating of projection multiplying power, seem more tediously long and complicated, so that exposure rate reduces.
Summary of the invention
The objective of the invention is at the deficiencies in the prior art, a kind of pattern image setting and multiplying power error compensation method that improves the step-by-step exposure machine of alignment accuracy and exposure efficiency is provided.
Step-by-step exposure machine pattern image setting method of the present invention, be that pattern image is projected on the substrate, substrate has a plurality of modular units that are the old arrangement of square, and what the pattern image setting method comprised the following steps: that (A) obtain pattern image allows that alignment error and pattern image are with respect to the target exposure region difference value of magnification error on two-dimensional directional respectively; And (B) according to allowing that alignment error and difference value decide the width value of this pattern image, and make the width value of the length value of pattern image greater than twice.
The multiplying power error compensation method of step-by-step exposure machine of the present invention, step-by-step exposure machine is used for the projective patterns image to the target exposure region, the width value of this pattern image is that correspondence is allowed alignment error and difference value, the length value of pattern image is greater than the width value of twice, and the long axis direction of the long limit parallel direction of setting and pattern image, multiplying power error compensation method comprises the following step: (a) measure at the magnification error between target exposure region and pattern image on the long axis direction; Reach when (b) being not equal to zero as if this magnification error, the corresponding projection multiplying power of adjusting step-by-step exposure machine, it is zero making pattern image and the magnification error of target exposure region on long axis direction.
Pattern image setting of the present invention and multiplying power error compensation method, by setting the specific dimensions of pattern image earlier, make magnification error be controlled in indirectly and allow within the alignment error, improved the alignment accuracy of step-by-step exposure machine the time in the face of anisotropic base plate deformation; And be different from traditional multiplying power compensation method, need take into account the complicated calculations of X and Y direction magnification error simultaneously, the present invention only utilizes the coordinate of two actual alignment marks 333 to carry out the compensation of Y direction multiplying power and calculates, to simplify the precision that multiplying power compensated and improved the multiplying power error compensation of Y direction, the exposure rate of step-by-step exposure machine can effectively be improved.
Description of drawings
Below in conjunction with the embodiment in the accompanying drawing the present invention is described in further detail, but does not constitute any limitation of the invention.
Fig. 1 is the structural representation that traditional multiplying power error compensation method is applied in step-by-step exposure machine;
Fig. 2 is coincide a synoptic diagram when the target exposure region of traditional pattern image, and dotted line is represented pattern image among the figure, and solid line is represented the target exposure region;
Fig. 3 is the offset error synoptic diagram of pattern image and target exposure region;
Fig. 4 is the rotation error synoptic diagram of pattern image and target exposure region;
Fig. 5 is the magnification error synoptic diagram of pattern image and target exposure region;
Fig. 6 is the process flow diagram of traditional exposure method;
Fig. 7 is a kind of anisotropic base plate deformation synoptic diagram;
Fig. 8 adopts traditional multiplying power error compensation method to anisotropic base plate deformation caused among Fig. 7 magnification error synoptic diagram;
Fig. 9 is that method of the present invention is used for the step-by-step exposure machine synoptic diagram;
Figure 10 is the magnification error synoptic diagram that method of the present invention produces in the anisotropy distortion;
Figure 11 is a method correction magnification error synoptic diagram of the present invention
Figure 12 is a pattern image setting method flow diagram of the present invention;
Figure 13 is an exposure program process flow diagram of the present invention;
Figure 14 is a multiplying power error compensation method process flow diagram of the present invention.
Embodiment
Consult shown in Figure 9, be used for the preferred embodiment of step-by-step exposure machine 3 for pattern image setting of the present invention and multiplying power error compensation method, step-by-step exposure machine 3 comprises the projection lens's group 31 that can adjust the projection multiplying power, be used for the graphic pattern projection on the light shield 32 is imaged on the sensing optical activity substrate 33, step-by-step exposure machine 3 more comprises the light shield microscope carrier 34 that is used for fixing light shield 32 and the relative position of projection lens's group 31, be used for fixing the baseplate carrier 35 of substrate 33 and the relative position of projection lens's group 31, and the control module 30 of each member 31,34,35 running of control.Have a plurality of pattern units 321 on the light shield 32, and also have a plurality of modular units 331 on the sensing optical activity substrate 33 with arranged with arranged.The size of modular unit 331 is generally the physical size of a chip (chip or die), and the pattern in the pattern unit 321 is wherein one deck circuit structure pattern of chip usually, by heaping several layers circuit structure, can form complete chip.In the time of need exposing, at first need respectively light shield 32 and sensing optical activity substrate 33 to be arranged on light shield microscope carrier 34 and the baseplate carrier 35, pattern unit 321 is imaged on the corresponding modular unit 331 by projection lens's group 31.The magnification error method of present embodiment and the establishing method of pattern image are to be incorporated in the control module 30 by software or the hardware such as integrated circuit (IC), the convenient method of carrying out present embodiment by control module 30.
Because projection lens's group 31 of step-by-step exposure machine 3 is subjected to the restriction of eyeglass caliber size, has a maximum exposure scope, need maximum exposure scope according to projection lens's group 31, determine the pattern area 322 of a maximum feasible, make this pattern area 322 by projection lens's group 31 formed pattern image 323 sizes less than this maximum exposure scope, so that pattern image 323 can be projected in substrate 33 default target exposure regions 332 by projection lens's group 31.
Mention that in background technology pattern image 323 is directly proportional with the length of this pattern image 323 with respect to the magnification error of target exposure region 332.Therefore, pattern image 323 is directly proportional in the length of directions X with this pattern image 323 with respect to the magnification error of target exposure region 332 at directions X.In like manner, pattern image 323 also is directly proportional in the length of Y direction with this pattern image 323 with respect to the magnification error of target exposure region 332 in the Y direction.
Be that the true origin of present embodiment is the geometric center that is defined in this pattern image 123 or target exposure region 132 what this will mention earlier.
Therefore, as Figure 10, suppose that not making the preceding pattern image 323 of compensation is Ix in the length of directions X, the length of Y direction is Iy.And target exposure region 332 is Δ X (Δ X is that target exposure region 332 is with respect to the length ratio of pattern image 123 at directions X) in the dimensional variation of directions X, and target exposure region 332 is Δ Y (Δ Y is that target exposure region 332 is with respect to the length ratio of pattern image 323 in the Y direction) in the dimensional variation of Y direction.Therefore, can calculate target exposure region 332 is the dimensional variation Δ X (Erx=Ix Δ X) that the length Ix of directions X is multiplied by directions X with respect to the magnification error Erx of the directions X of pattern image 323; In like manner, target exposure region 332 is the dimensional variation Δ Y (Ery=Iy Δ Y) that the length Iy of Y direction is multiplied by the Y direction with respect to pattern image 323 at the magnification error Ery of Y direction.
At this moment, if only considering the magnification error of selected Y direction compensates, the projection multiplying power of its compensation is Δ Y, as shown in figure 11, the magnification error of its Y can be eliminated (Ery=Iy (Δ Y-Δ Y)=0) in theory fully, and the magnification error of directions X then is multiplied by the difference value (Erx=Ix (Δ X-Δ Y)) of the dimensional variation of X and Y direction for the length of directions X.In like manner, if only considering the magnification error of selected directions X compensates, the projection multiplying power of its compensation is Δ X, the magnification error of its X can be eliminated (Erx=Ix (Δ X-Δ X)=0) in theory fully, and the magnification error of Y direction then is multiplied by the difference value (Ery=Iy (Δ Y-Δ X)) of the dimensional variation of X and Y direction for the length of Y direction.Suppose only to consider the order of magnitude of Erx and Ery, (Δ X-Δ Y) or (Δ Y-Δ X) can be represented with Δ S.Therefore, compensate as if selected the magnification error to the Y direction, then the magnification error of directions X is Erx=Ix Δ S, and if selected the magnification error to directions X compensates, then the magnification error of Y direction is Ery=Iy Δ S.Wherein Δ S is that institute provides in the testing of materials report that calculates voluntarily before carrying out exposure manufacture process or provided by the material supplier, when doing the multiplying power compensation, Δ S can be considered as a constant actual.Can find that thus when only wherein a direction was done the compensation of magnification error to X or Y direction, the magnification error on the other direction was only relevant with the length of this other direction and be directly proportional.
In order to reach the purpose that allows magnification error reduce as far as possible, in the present embodiment, suppose the length Iy of the length Ix of directions X less than the Y direction, therefore, if the magnification error of selected Y direction compensates, the magnification error of this Y direction can be eliminated in fact fully, and still the magnification error of existence (Erx=Ix Δ S) can be less on the directions X.
With traditional multiplying power error compensation method, the 6th, 580, No. 494 patent cases of the U.S., and the present invention makes comparisons, and is as shown in the table:
In addition, can learn by last table, at the length ratio Ix/Iy of the directions X of pattern image and Y direction greater than 1/2, be that the length of Y direction is during less than the directions X length of twice, magnification error by gained after the still more traditional multiplying power error compensation method compensation of magnification error of present embodiment compensation back gained is big, but at the length ratio Ix/Iy ratio of the directions X of pattern image and Y direction less than 1/2, be the length of Y direction during greater than the directions X length of twice, the magnification error of the more traditional multiplying power error compensation method compensation of the magnification error back gained of present embodiment compensation back gained is little.
Therefore, present embodiment, under the condition of Y direction length greater than the directions X length of twice of setting pattern image, selected the magnification error to the Y direction compensates, and the magnification error after the more traditional multiplying power compensation way compensation of resulting magnification error (Erx=Ix Δ S) meeting is little.
And, because present embodiment only compensates the magnification error of Y direction, need not measure and compensate the magnification error of directions X, therefore when the practical operation step-by-step exposure machine, be used for measuring and calculate the number of the alignment mark of magnification error, can be kept to along two alignment marks of Y direction configuration from three alignment marks that tradition is used.
Further, the user also can set one according to product performance and allow alignment error Er, and utilizes known difference value Δ S, tries to achieve the width value Ix of pattern image at directions X, promptly according to above-mentioned formula (Erx=Ix Δ S), can push away:
Ix＝Er/ΔS。
Be noted that for convenience of explanation, above-mentioned defined Er is the magnification error of directions X that fixes on the geometric center gained of pattern image according to true origin, therefore, Ix in fact only is the positive directions X length of pattern image, and figure in fact
The length Ws of case image directions X should be (comprising positive directions X and negative directions X) Ix of twice, is:
Ws＝2xIx＝2xEr/ΔS。
For instance, need be less than 2.5 μ m if the user formulates pattern image according to product performance with respect to the alignment error of allowing of target exposure region, and Δ S calculates with employed material of printed circuit board (PCB) now and process technique thereof, is example with 100ppm, can get according to above-mentioned formula:
In other words Ws=2xEr/ Δ S=2x2.5 μ m/100ppm=0.5m=50mm, is allowing that the directions X length of pattern image needs less than 50mm under the condition of alignment error need less than 2.5um.
It should be noted, for convenience of explanation, set the Width that directions X is a pattern image 323 (referring to short-axis direction) in the present embodiment, and the Y direction is the length direction (referring to long axis direction) of pattern image 323, those skilled in the art all understands, the width of pattern image 323, length direction also can be decided to be Y direction, directions X respectively, not limited by present embodiment.
Therefore, before 3 pairs of substrates 31 of stepping exposure machine expose, can carry out pattern image setting as Figure 12 earlier.
At first, in step 41, control module 30 is obtained the difference value Δ S that allows alignment error Er and magnification error earlier.
Secondly, in step 42, control module 30 will be according to allowing that the difference value Δ S of alignment error Er and magnification error calculates the width value Ws of pattern image, refers to twice allowed the difference value Δ S of alignment error Er divided by magnification error.
At last, in step 43, decide the length value of pattern image, make length value greater than more than two times of width value Ws according to the length of width value Ws and modular unit 331.
This moment, control module 30 is according to pattern image 323 correspondences of having set selected pattern area 322 on light shield 32, and substrate 33 separated into the target exposure region 332 of a plurality of correspondences, and further target exposure region 332 is stored in the control module 30 with the relevant coordinate data of pattern area 322.And for carrying out the multiplying power compensation to the Y direction, as Figure 10, each target exposure region 332 is provided with two actual alignment marks 333 that are provided with along the Y direction in the present embodiment, and more stores the coordinate of two corresponding default alignment marks 324 in the control module 30.So, as Figure 13, stepping exposure machine 3 can begin exposure program.At first, step 51 is selected a pattern area 322 of light shield 32, and determines a target exposure region 332 corresponding with this pattern area 322, and obtains the coordinate of the alignment mark 333 on the target exposure region 332.Secondly, in step 52, control module 30 is calculated the magnification error of offset error, rotation error and Y direction according to the coordinate figure of alignment mark 333, adjust light shield 32, the position of substrate 33 and the projection multiplying power of projection lens's group 31 with correspondence, to carry out the compensation of offset error, rotation error and magnification error.After finishing compensation, in step 53, pattern area 322 is thrown projective patterns image 323 to target exposure region 332 by projection lens's group 31, exposes.After 332 exposures of target exposure region were finished, execution in step 54 judged whether that other unexposed target exposure region 332 exists.
If step 54 is judged as sometimes, in selecting unexposed target exposure region 332 one of execution in step 55, control module 30 utilizes baseplate carrier 35 to drive substrates 33 running fixes again, and rebound step 51, the repeated exposure flow process.So, utilize displacement and the mode that repeats can allow substrate 33 complete exposures.At this, because offset error and rotation error compensate non-improvement content of the present invention, and are known by those of ordinary skill in the art, so do not describe in detail herein.
For allowing the easier quilt of multiplying power error compensation method of present embodiment understand, the multiplying power error compensation program implementation flow process of each target exposure region 332 in the present embodiment will be described hereinafter.
As Figure 14, at first in step 521, control module 30 utilizes the coordinate of the measured actual alignment mark 333 of previous steps 51 to calculate both distances in the Y direction.
Secondly, in step 522, control module 30 compares the distance of 333 of the actual alignment marks distance with two default alignment marks 324 of its storage, with the magnification error of decision Y direction.The ratio that also is distance and the distance of default alignment mark 324 of actual alignment mark 333 is magnification error.
Then, in step 523, control module 30 judges whether magnification error is zero, is meant whether projected pattern image 323 and target exposure region 332 are existed magnification error on the Y direction.If step 523 is judged as and is, represent pattern image 323 and target exposure region 332 on the Y direction, not to have the magnification error that needs compensation, so finish the multiplying power error compensation operation, begin exposure and continue step 53.When if step 523 is judged as not, execution in step 524.
In step 524, control module 30 is pressed the projection multiplying power of the next corresponding projection lens of the magnification error group 31 of Y direction, 333 the Y directions of two default alignment marks that make the pattern image 323 that will throw are apart from the distance that is same as the actual alignment mark 333 of target exposure region 332, and then make the length of projected pattern image 323 on the Y direction consistent with the physical length of the Y direction of target exposure region 332, to carry out the multiplying power error compensation of Y direction.Step 524 also finishes the multiplying power compensation program after finishing.
Pattern image setting of the present invention and multiplying power error compensation method, allow the projection multiplying power of projection lens's group 31 only adjust at Y direction magnification error according to target exposure region 332, make the pattern image 323 that is projected be equal to the Y direction length of this target exposure region 332 in the length of Y direction, the magnification error that is the Y direction can be eliminated in fact fully, and the magnification error of directions X is also allowed alignment error Er less than what the user set, reach really improve aim at precision high with simplify the effect that compensation is calculated.
Claims (4)
1, a kind ofly is used to improve the alignment accuracy of step-by-step exposure machine and the method for exposure rate, this method comprises pattern image setting and multiplying power error compensation, it is characterized in that described pattern image setting comprises following steps: (A) obtain pattern image allow alignment error with respect to the difference value of the magnification error of target exposure region on two-dimensional directional; And (B) allow that according to this alignment error and this difference value decide the width value of pattern image, and make the width value of the length value of pattern image greater than twice; Described multiplying power error compensation comprises following steps: (a) long axis direction of the long limit parallel direction of setting and pattern image, measure at the magnification error between target exposure region and pattern image on the long axis direction; Reach when (b) being not equal to zero as if this magnification error, the corresponding projection multiplying power of adjusting step-by-step exposure machine, it is zero making pattern image and the magnification error of target exposure region on long axis direction.
2, according to claim 1ly be used to improve the alignment accuracy of step-by-step exposure machine and the method for exposure rate, it is characterized in that in the step (B) of described pattern image setting that width value is to allow that with twice alignment error is divided by the difference value of pattern image with respect to the magnification error of target exposure region on two-dimensional directional.
3, according to claim 1ly be used to improve the alignment accuracy of step-by-step exposure machine and the method for exposure rate, it is characterized in that in the step (a) of described multiplying power error compensation, be to be provided with two alignment marks, promptly get magnification error by the distance value of 2 alignment marks on the measurement target exposure region between 2 alignment marks default of the numerical value that distance obtained on the long axis direction divided by pattern image along long axis direction.
4, according to claim 1ly be used to improve the alignment accuracy of step-by-step exposure machine and the method for exposure rate, it is characterized in that in the step (b) of described multiplying power error compensation, is that correspondence is adjusted step-by-step exposure machine projection multiplying power and made the length value of pattern image consistent with the length of target exposure region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100360820A CN100570491C (en) | 2006-06-26 | 2006-06-26 | The pattern image setting of step-by-step exposure machine and multiplying power error compensation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100360820A CN100570491C (en) | 2006-06-26 | 2006-06-26 | The pattern image setting of step-by-step exposure machine and multiplying power error compensation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1869817A CN1869817A (en) | 2006-11-29 |
| CN100570491C true CN100570491C (en) | 2009-12-16 |
Family
ID=37443511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100360820A Expired - Fee Related CN100570491C (en) | 2006-06-26 | 2006-06-26 | The pattern image setting of step-by-step exposure machine and multiplying power error compensation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100570491C (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102543785B (en) * | 2010-12-08 | 2016-01-13 | 无锡华润上华科技有限公司 | Alignment specification verification method |
| JP6399739B2 (en) * | 2013-09-27 | 2018-10-03 | キヤノン株式会社 | Exposure apparatus, exposure method, and device manufacturing method |
| CN106227002B (en) * | 2016-09-21 | 2018-01-23 | 中山新诺科技股份有限公司 | A kind of method for the efficiency for improving adjustment splicing and multiplying power size |
| CN112965344A (en) * | 2021-02-08 | 2021-06-15 | 上海度宁科技有限公司 | Photoetching system and exposure compensation method thereof |
-
2006
- 2006-06-26 CN CNB2006100360820A patent/CN100570491C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1869817A (en) | 2006-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101885385B1 (en) | Method and apparatus for performing pattern alignment | |
| CN100480620C (en) | System and method for manufacturing printed circuit boards employing non-uniformly modified images | |
| CN101957567A (en) | The position correction of the alignment head in the bull alignment system | |
| JP5496041B2 (en) | Displacement calculation method, drawing data correction method, drawing method, and drawing apparatus | |
| CN100570491C (en) | The pattern image setting of step-by-step exposure machine and multiplying power error compensation method | |
| JP5264406B2 (en) | Exposure apparatus, exposure method, and device manufacturing method | |
| KR20220143743A (en) | Systems and methods for correcting overlay errors in lithography processes | |
| KR20120106536A (en) | Drawing data correction apparatus and drawing apparatus | |
| JP2000182934A (en) | Position aligning method for semiconductor wafer | |
| JP2010103476A (en) | Alignment unit and exposure apparatus | |
| JP3617046B2 (en) | Exposure method | |
| CN115220308A (en) | Processing system, processing method, measuring apparatus, substrate processing apparatus, and article manufacturing method | |
| Chang et al. | Analysis of pattern distortion by panel deformation and addressing it by using extremely large exposure field fine-resolution lithography | |
| Chang et al. | Extremely Large Exposure Field With Fine Resolution Lithography Technology To Enable Next Generation Panel Level Advanced Packaging | |
| CN110275399B (en) | Method for measuring position relation error of alignment camera of laser direct imaging equipment | |
| JP7336201B2 (en) | Forming methods, patterning systems, and methods of manufacturing articles | |
| JP2022027019A (en) | Measurement method, exposure apparatus, and manufacturing method of article | |
| JP2005114847A (en) | Projection exposure apparatus | |
| JPH11195591A (en) | Alignment method | |
| JPH0982597A (en) | Exposure method | |
| JP2018151577A (en) | Exposure apparatus, exposure method, program, determination method, and article manufacturing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20091216 Termination date: 20190626 |