CN117234046A - Exposure equipment and defocus compensation method - Google Patents

Abstract

The invention discloses an exposure device and a defocus compensation method, wherein the exposure device comprises: an optical path imaging device and an electron beam exposure machine; the optical path imaging device comprises a light source, a beam expanding collimation device, an iris diaphragm, a first projection device, a first reflection device, a second projection device, an optical amplifying assembly and an image sensor which are sequentially arranged; a mask plate is arranged between the iris diaphragm and the first projection device, and a grating array and a plurality of slit arrays with different widths are arranged on the mask plate. According to the invention, the mask plate with the grating array and the slit arrays with different widths is used, a plurality of slit marks with different widths are introduced into the light beam, and when the defocus amount is large, the image sensor can capture partial images with slits with different widths, and the partial images with slits with different widths contain more defocus information, so that the high-precision defocus detection of an expanded range is realized on the electron beam exposure machine.

Description

Exposure equipment and defocus compensation method

Technical Field

The invention relates to the technical field of semiconductor manufacturing, in particular to exposure equipment and a defocus compensation method.

Background

The electron beam exposure technology is a micro-nano processing technology for performing direct writing exposure on a silicon wafer coated with electron beam resist through focusing high-energy electron beams, is an extension application of the optical lithography technology, has higher resolution, plays an important role in the fields of chip processing, integrated circuit manufacturing and the like, and is limited by the focal depth of the silicon wafer while the electron beam exposure machine has higher resolution. The resolution of electron beam exposure is largely dependent on the size of the electron beam spot, which must be perfectly focused on the wafer surface in order to achieve the smallest electron beam spot for optimum exposure and resolution. However, due to the limitation of the silicon wafer manufacturing process, the surface of the silicon wafer cannot be completely smooth, and the height fluctuation always occurs locally. In addition, due to the influence of external environments such as temperature, pressure and the like and factors such as man-made and mechanical factors during the fixation of the silicon wafer, the silicon wafer can be inclined, warped and deformed, and the complex surface environment of the silicon wafer is determined by the series of uneven surfaces formed by the inclination, the warping, the deformation and the like. Because various defocusing factors are accumulated, the defocusing amount of the silicon wafer may exceed the range of the focal depth, and the exposure effect may be seriously affected.

Currently, the mainstream focusing and leveling system in the photoetching technology adopts oblique optical triangulation to detect the surface of a silicon wafer, namely, a detection beam is used for irradiating the surface of the silicon wafer, and a charge coupled device sensor or a complementary metal oxide semiconductor sensor is used for receiving reflected light by utilizing the characteristic of high reflectivity of the surface of the silicon wafer. The height information of the silicon wafer is loaded on the light beam, when the height of the silicon wafer changes relatively, the reflected light beam changes accordingly, and the actual height information of the silicon wafer is estimated by analyzing the position change of the reflected light on the sensor. In the existing focusing and leveling technology, the mask plate is only provided with a grating array, and due to the periodic structure of the grating array, the provided defocus information is limited, the detectable defocus range is smaller, and the use requirement of the electron beam exposure machine for realizing large-scale detection cannot be met.

Accordingly, the prior art has drawbacks and needs to be improved and developed.

Disclosure of Invention

The invention aims to solve the technical problems that the optical path imaging device and the defocus compensation method aim to solve the problems that the detectable defocus range of an electron beam exposure machine in the prior art is smaller and the use requirement of the electron beam exposure machine for realizing large-scale high-precision defocus detection cannot be met.

The technical scheme adopted for solving the technical problems is as follows:

an exposure apparatus, comprising: an optical path imaging device and an electron beam exposure machine; the optical path imaging device comprises a light source, a beam expanding collimation device, an iris diaphragm, a first projection device, a first reflection device, a second projection device, an optical amplifying assembly and an image sensor which are sequentially arranged; when the exposure equipment is in a detection state and the target silicon wafer is positioned between the first projection device and the second projection device, the light source, the beam expansion collimation device, the iris diaphragm, the first projection device, the first reflection device, the target silicon wafer, the second reflection device, the second projection device, the optical amplification assembly and the image sensor form a light path; a mask plate is arranged between the iris diaphragm and the first projection device, and a grating array and a plurality of slit arrays with different widths are arranged on the mask plate and used for introducing light beams in the light path into grating marks and a plurality of slit marks with different widths; the optical path imaging device is used for generating a target image with a grating mark and a slit mark;

the electron beam exposure machine comprises a movable displacement table and an exposure machine body; the movable displacement platform is positioned between the first reflecting device and the second reflecting device and is used for placing a target silicon wafer, and the exposure machine body is positioned above the movable displacement platform and is used for performing defocus compensation on the target silicon wafer and exposing the target silicon wafer according to a target image.

Further, a grating array, a first slit array, a second slit array, a third slit array, a fourth slit array, a fifth slit array and a sixth slit array are arranged on the mask plate; the grating array is positioned at the center of the mask plate, the first slit array is positioned at a first position of the grating array, the second slit array is positioned at a second position of the grating array, the third slit array and the fourth slit array are arranged in parallel and are positioned at a third position of the grating array, and the third slit array is positioned between the grating array and the fourth slit array; the fifth slit array is arranged in parallel with the sixth slit array and is positioned in a fourth direction of the grating array, and the fifth slit array is positioned between the grating array and the sixth slit array.

Further, the first slit array comprises a first slit; the second slit array comprises a second slit; the third slit array comprises a third slit and a fourth slit, the third slit is positioned between the grating array and the fourth slit, and the width of the third slit is smaller than the width of the fourth slit; the fourth slit array comprises three fifth slits, a sixth slit and a seventh slit with sequentially increasing widths, wherein the fifth slit is positioned between the fourth slit and the sixth slit, and the sixth slit is positioned between the fifth slit and the seventh slit; the fifth slit array comprises an eighth slit and a ninth slit, the eighth slit is positioned between the grating array and the ninth slit, and the width of the eighth slit is smaller than that of the ninth slit; the sixth slit array comprises a tenth slit, an eleventh slit and a twelfth slit with sequentially increasing widths, wherein the tenth slit is positioned between the ninth slit and the eleventh slit, and the eleventh slit is positioned between the tenth slit and the eleventh slit.

The invention also provides a defocus compensation method based on the exposure equipment, which comprises the following steps:

detecting a silicon wafer to be detected by using an optical path imaging device to obtain a target image;

and processing the target image to obtain the defocus amount of the silicon wafer to be detected, and performing defocus compensation according to the defocus amount.

Further, the electron beam exposure machine further comprises a calculation control module; the target image is processed to obtain the defocus amount of the silicon wafer to be detected, and defocus compensation is carried out according to the defocus amount, and the method comprises the following steps:

the calculation control module performs image recognition on the target image to obtain a recognition result;

if the identification result is that the target image has the grating mark, the calculation control module processes the target image to obtain a first defocus amount of the silicon wafer to be detected, and the exposure machine body performs defocus compensation on the silicon wafer to be detected according to the first defocus amount;

if the identification result is that the target image does not have the grating mark, the calculation control module divides the target image to obtain a target slit mark image, processes the target slit mark image to obtain the final defocus amount of the silicon wafer to be detected, and the exposure machine body performs defocus compensation on the silicon wafer to be detected according to the final defocus amount.

Further, before the detecting the silicon wafer to be detected by the optical path imaging device to obtain the target image, the method further comprises:

detecting a reference silicon wafer by using an optical path imaging device to obtain a reference image and a plurality of non-reference images;

the calculation control module processes the plurality of non-reference images to obtain a calibration curve of the reference silicon wafer;

the reference silicon wafer is placed on the upper surface of the movable displacement table, the reference image is an image corresponding to the reference silicon wafer when the reference silicon wafer is located at the reference position of the movable displacement table, the reference position is a position where the defocus amount of the reference silicon wafer is zero, the non-reference image is an image corresponding to the reference silicon wafer when the reference silicon wafer is located at the non-reference position of the movable displacement table, the defocus amount of the reference silicon wafer is non-zero when the reference silicon wafer is located at the non-reference position of the movable displacement table, and the calibration curve comprises a corresponding relation between the defocus amount and the position of the movable displacement table.

Further, if the identification result is that the target image has a grating mark, the calculation control module processes the target image to obtain a first defocus amount of the silicon wafer to be measured, and the exposure machine body performs defocus compensation on the silicon wafer to be measured according to the first defocus amount, including:

If the identification result is that the target image has a grating mark, the calculation control module divides the target image to obtain a first target grating image, and processes the first target grating image to obtain a first digital grating corresponding to the first target grating image, wherein the first digital grating is stored in a matrix form;

the calculation control module converts the first target grating image into a matrix form to obtain a first grating image matrix, multiplies the first grating image matrix and the first digital grating to obtain a first grating superposition matrix, and processes the first grating superposition matrix to obtain a first grating superposition image;

the calculation control module carries out spline interpolation processing on the first grating superimposed image to obtain a first smooth image, fits the first smooth image to obtain a first fitting curve, and obtains the center coordinate of the first grating superimposed image according to the first fitting curve;

the calculation control module obtains a corresponding first reference grating image in a reference image according to a first target grating image, processes the first reference grating image to obtain a first reference grating superimposed image, processes the first reference grating superimposed image to obtain a central coordinate of the first reference grating superimposed image, compares the central coordinate of the first grating superimposed image with the central coordinate of the first reference grating superimposed image to obtain a first central position variation of the silicon wafer to be measured, and obtains a first defocus amount of the silicon wafer to be measured according to the first central position variation;

And the exposure machine body adjusts the current value of the objective lens coil in the exposure machine body according to the first defocus amount so as to perform defocus compensation on the silicon wafer to be detected.

Further, if the identification result is that the target image does not have the grating mark, the calculation control module segments the target image to obtain a target slit mark image, processes the target slit mark image to obtain a final defocus amount of the silicon wafer to be measured, and the exposure machine body performs defocus compensation on the silicon wafer to be measured according to the final defocus amount, including:

if the identification result is that the target image does not have the grating mark, the calculation control module divides the target image to obtain a target slit mark image, and the size of the target slit mark image is smaller than that of the target image;

the calculation control module determines a target slit region in the target slit mark image, acquires a gray value of each row in the target slit region, and determines the center coordinate of the target slit region according to the gray value of each row in the target slit region;

the calculation control module determines a corresponding target reference slit region in the reference image according to the target slit region, processes the target reference slit region to obtain a center coordinate of the target reference slit region, and compares the center coordinate of the target slit region with the center coordinate of the target reference slit region to obtain a second center position variation of the silicon wafer to be detected;

The calculation control module obtains the final defocus amount of the silicon wafer to be measured according to the second central position variation;

and the exposure machine body adjusts the current value of the objective lens coil according to the final defocus amount so as to perform defocus compensation on the silicon wafer to be detected.

Further, the calculating control module obtains a final defocus amount of the silicon wafer to be measured according to the second central position variation, including:

the calculation control module obtains a second defocus amount of the silicon wafer to be measured according to the second central position variation, and determines a target position according to the second defocus amount and the calibration curve;

the calculation control module controls the movable displacement platform to move to a target position;

the calculation control module acquires a second target image, wherein the second target image is a target image corresponding to the silicon wafer to be detected, generated by the optical path imaging device, when the movable displacement platform moves to a target position, and the second target image is provided with a grating mark;

the calculation control module divides the second target image to obtain a second target grating image, and processes the second target grating image to obtain a second digital grating corresponding to the second target grating image, wherein the second digital grating is stored in a matrix form;

The calculation control module converts the second target grating image into a matrix form to obtain a second grating image matrix, multiplies the second grating image matrix and the second digital grating to obtain a second grating superposition matrix, and processes the second grating superposition matrix to obtain a second grating superposition image;

the calculation control module carries out spline interpolation processing on the second grating superimposed image to obtain a second smooth image, fits the second smooth image to obtain a second fitting curve, and obtains the center coordinates of the second grating superimposed image according to the second fitting curve;

the calculation control module obtains a corresponding second reference grating image in the reference image according to the second target grating image, processes the second reference grating image to obtain a second reference grating superimposed image, processes the second reference grating superimposed image to obtain a center coordinate of the second reference grating superimposed image, compares the center coordinate of the second grating superimposed image with the center coordinate of the second reference grating superimposed image to obtain a second center position variation of the silicon wafer to be measured, and obtains a final defocus amount of the silicon wafer to be measured according to the second center position variation.

The present invention also provides a computer-readable storage medium storing a computer program executable for implementing the steps of the defocus compensation method as described above.

The invention discloses an exposure device and a defocus compensation method, wherein the exposure device comprises: an optical path imaging device and an electron beam exposure machine; the optical path imaging device comprises a light source, a beam expanding collimation device, an iris diaphragm, a first projection device, a first reflection device, a second projection device, an optical amplifying assembly and an image sensor which are sequentially arranged; a mask plate is arranged between the iris diaphragm and the first projection device, and a grating array and a plurality of slit arrays with different widths are arranged on the mask plate. According to the invention, the mask plate with the grating array and the slit arrays with different widths is used, a plurality of slit marks with different widths are introduced into the light beam, when the defocus amount is large, the image sensor can generate partial images with slits with different widths, and the partial images with the slits with different widths contain more defocus information, so that the high-precision defocus detection of an expanded range is realized on the electron beam exposure machine.

Drawings

FIG. 1 is a schematic view showing a structure of a preferred embodiment of an exposure apparatus according to the present invention;

FIG. 2 is a schematic view of a mask plate according to a preferred embodiment of the present invention;

FIG. 3 is a flow chart of a preferred embodiment of the defocus compensation method of the present invention;

FIG. 4 is a schematic view of a displaced moire pattern according to the present invention;

FIG. 5 is a schematic illustration of gray scale distribution of a target slit region in the present invention;

FIG. 6 is a schematic diagram showing a gray scale distribution of a partial image corresponding to a third slit array according to the present invention;

FIG. 7 is a schematic diagram of error results of multiple defocus detection performed on the same wafer in the present invention;

reference numerals illustrate:

1. a light source; 2. a beam expanding collimation device; 3. an iris diaphragm; 4. a mask plate; 5. a first projection device; 6. a first reflecting device; 7. a target silicon wafer; 8. a second reflecting device; 9. a second projection device; 10. an optical amplifying assembly; 11. an image sensor; 12. a calculation control module; 13. an exposure machine body; 20. a movable displacement stage; 21. an objective table; 22: a piezoelectric displacement stage; 30. a grating array; 40. a first slit array; 50. a second slit array; 60. a third slit array; 61. third slit, 62, fourth slit; 70. a fourth slit array; 71. a fifth slit; 72. a sixth slit; 73. a seventh slit; 80. a fifth slit array; 81. an eighth slit; 82. a ninth slit; 90. a sixth slit array; 91. a tenth slit; 92. an eleventh slit; 93. a twelfth slit.

Detailed Description

The invention provides an optical path imaging device and a defocus compensation method, which are used for making the purposes, technical schemes and effects of the invention clearer and more definite, and the invention is further described in detail below by referring to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

In the existing focal plane detection method, gratings are generally used as projection marks, moire fringes are formed by using other gratings with different periods in front of an image sensor, and the actual defocus amount of a silicon wafer is reversely pushed by analyzing the change of the moire fringes. The method needs to be provided with a plurality of gratings, and has complex light path and higher assembly difficulty. Meanwhile, due to the periodic structure of the grating array, the measuring range of the method is smaller and is only in the micrometer scale. In particular, for special shaped silicon wafers, such as stepped silicon wafers, the focal plane changes to millimeter or even centimeter level in the whole exposure process, and the prior art cannot perform high-precision detection well. In addition, even for a flat silicon wafer, the surface of the silicon wafer cannot be completely flat due to the problem of the manufacturing process, local fluctuation is generated, and the electron beam exposure machine needs to fix the silicon wafer by using conductive adhesive in the exposure process to solve the charge accumulation effect, which also introduces additional defocus amount of the silicon wafer.

The invention adopts a mode of combining the slit mark and the grating mark to carry out defocusing compensation, adopts a pixel coding mode to generate the digital grating corresponding to the grating, combines the grating and the digital grating to generate moire fringes, reduces the number of the gratings, optimizes the light path structure and is easy to assemble. The high-precision detection of the defocusing amount is realized, and meanwhile, as the slit marks are arranged around the grating marks, the detection range is enlarged, and the detection range is increased from micron level to centimeter level.

The technical scheme of the invention provides exposure equipment, as shown in fig. 1, which comprises the following steps: an optical path imaging device and an electron beam exposure machine; the light path imaging device comprises a light source 1, a beam expansion collimation device 2, an iris diaphragm 3, a first projection device 5, a first reflection device 6, a second reflection device 8, a second projection device 9, an optical amplification assembly 10 and an image sensor 11 which are sequentially arranged; when the exposure apparatus is in a detection state and the target silicon wafer 7 is positioned between the first projection device 5 and the second projection device 9, the light source 1, the beam expanding and collimating device 2, the iris 3, the first projection device 5, the first reflection device 6, the target silicon wafer 7, the second reflection device 8, the second projection device 9, the optical amplifying assembly 10 and the image sensor 11 form an optical path; a mask plate 4 is arranged between the iris diaphragm 3 and the first projection device 5, and a grating array and a plurality of slit arrays with different widths are arranged on the mask plate 4 and used for introducing light beams in the light path into a grating mark and a plurality of slit marks with different widths; the optical path imaging device is used for generating a target image with a grating mark and a slit mark;

The electron beam exposure machine comprises a movable displacement table 20 and an exposure machine body 13; the movable displacement stage 20 is located between the first reflecting device 6 and the second reflecting device 8 and is used for placing a target silicon wafer 7, and the exposure machine body 13 is located above the movable displacement stage 20 and is used for performing defocus compensation on the target silicon wafer 7 and exposing the target silicon wafer according to a target image.

Specifically, the light source 1 is configured to emit light, the beam expansion collimator 2 is configured to perform beam expansion collimation on the light source 1 to obtain a first light beam, the variable diaphragm 3 is configured to perform diameter adjustment on the first light beam to obtain a second light beam, the mask plate 4 is provided with a grating array and a plurality of slit arrays with different widths, the second light beam is used to introduce a grating mark and a plurality of slit marks with different widths to obtain a third light beam, the first projector 5 is configured to focus the third light beam to obtain a fourth light beam, the first reflector 6 is configured to perform optical path adjustment on the fourth light beam so that the fourth light beam is incident on a target silicon wafer, the second reflector 8 is configured to perform optical path adjustment on the fourth light beam passing through the target silicon wafer so that the fourth light beam passing through the target silicon wafer is incident on the second projector 9, the second projector 9 is configured to focus the fourth light beam passing through the target silicon wafer to obtain a fifth light beam, the optical amplifier 10 is configured to focus the fifth light beam, the fifth projector is configured to perform optical amplifier enhancement processing on the fifth light beam, and the fifth light beam is used to generate a slit image sensor 11.

The beam expansion collimating device 2 is a beam expansion collimating lens, the first projection device 5 is a projection lens, the first reflection device 6 is a reflection mirror, the second projection device 9 is a projection lens, and the second reflection device 8 is a reflection mirror. The projection unit is composed of a light source 1, a beam expansion collimation device 2, an iris 3, a mask plate 4, a first projection device 5 and a first reflection device 6. The detection unit is constituted by a second reflecting device 8, a second projecting device 9, an optical amplifying assembly 10 and an image sensor 11. The projection unit emits detection light to the surface of the target silicon wafer, the target silicon wafer reflects the detection light and then receives the detection light by the detection unit, and the reflected light has the height information of the surface height of the target silicon wafer. I.e. the fifth beam has height information of the target silicon wafer surface height. The light source 1 may be an LED or a laser light source. The movable displacement stage 20 includes a stage 21 and a piezoelectric displacement stage 22.

In one embodiment of the invention, a relay lens group is added between the first projection device 5 and the first reflection device 6 to optimize the optical path.

In an embodiment of the invention a relay lens group is added between the second projection device 9 and the second reflection device 8 to optimize the light path.

In one embodiment of the invention, the first reflecting device 6 is arranged in front of the first projection device 5 in the light path.

In one embodiment of the invention, the second projection device 9 is arranged in front of the second reflection device 8 in the light path.

In one embodiment of the invention, the first reflecting means 6 and the second reflecting means 8 are symmetrical with respect to the movable displacement stage 20.

According to the invention, the exposure machine body is used for exposing the target silicon wafer, and defocusing compensation is carried out before exposure according to the target image, so that a better electron beam spot can be obtained, and the resolution and quality of exposure are effectively improved.

In one embodiment of the present invention, the method for manufacturing a mask plate includes:

spin coating photoresist and conductive adhesive on quartz glass, and performing electron beam direct writing exposure to generate a first quartz glass sheet, wherein the first quartz glass sheet is provided with a preset pattern;

developing and fixing the first quartz glass sheet to obtain a second quartz glass sheet;

and coating and cleaning the second quartz glass sheet to obtain a mask.

Specifically, the material selected for the coating film is metal with high reflectivity. The preset pattern is a hollowed pattern. The part of the preset pattern, which is coated with the film and has metal, has light-proof property, and the part of the preset pattern, which does not have metal, has light-transmitting property. The portion of the preset pattern having the light transmission property is finally displayed as a bright stripe, and the portion of the preset pattern having the light non-transmission property is finally displayed as a dark stripe. The preset pattern is a combination of a slit array pattern and a grating array pattern. After the light passes through the mask plate, the slit marks corresponding to the slit array patterns and the grating marks corresponding to the grating array patterns are carried. The slit mark and the grating mark are combined to serve as the characteristic mark, and the characteristic mark can be carried when light passes through the mask plate, so that the follow-up defocusing direction identification and defocusing amount calculation are facilitated.

In one embodiment, the plating film is silver (Ag).

In one embodiment, the quartz glass and the coating are preceded by an adhesion material for joining the quartz glass and the coating metal.

In one embodiment, a first thickness of titanium (Ti) is plated onto the quartz glass followed by a second thickness of silver (Ag).

In one embodiment, the first thickness and the second thickness are adjustable according to photoresist thickness.

In one embodiment, the first thickness is 3nm and the second thickness is 50nm.

In one embodiment of the present invention, as shown in fig. 2, the mask plate is provided with a grating array 30 and a first slit array 40, a second slit array 50, a third slit array 60, a fourth slit array 70, a fifth slit array 80, and a sixth slit array 90; the grating array 30 is located at the center of the mask plate, the first slit array 40 is located at a first position of the grating array, the second slit array 50 is located at a second position of the grating array, the third slit array 60 and the fourth slit array 70 are arranged in parallel, and are located at a third position of the grating array 30, and the third slit array 60 is located between the grating array 30 and the fourth slit array 70; the fifth slit array 80 is disposed parallel to the sixth slit array 90 and is positioned in a fourth orientation of the grating array 30, the fifth slit array 80 being positioned between the grating array 30 and the sixth slit array 90.

Specifically, after passing through the mask plate, the light carries a grating mark and a slit mark, and the defocusing direction and the defocusing position are positioned according to the grating mark and the slit mark. The first slit array 40 is located above the grating array 30, the second slit array 50 is located below the grating array, the third slit array 60 and the fourth slit array 70 are located to the left of the grating array, and the fifth slit array 80 and the sixth slit array 90 are located to the right of the grating array. The grating array 30, the first slit array 40 and the second slit array 50 form a central region of the mask plate, and the third slit array 60, the fourth slit array 70, the fifth slit array 80 and the sixth slit array 90 form an expansion region of the mask plate. The invention can enlarge the measuring range of the defocus amount by setting the expansion area.

In one embodiment of the invention, the grating period of the grating array 30 is on the order of microns, with a 50% duty cycle.

In one embodiment of the present invention, the first slit array 40 and the second slit array 50 are symmetrically disposed about the grating array 30; the third slit array 60 and the fifth slit array 80 are symmetrically disposed about the grating array 30; the fourth slit array 70 and the sixth slit array 90 are symmetrically disposed about the grating array 30.

In one embodiment of the present invention, the expanded region of the mask includes a plurality of slit arrays having different widths.

Specifically, the invention expands the measuring range of the defocus amount by arranging a plurality of slit arrays with different widths on the mask plate.

In one embodiment of the present invention, as shown in FIG. 2, the first slot array 40 includes a first slot therein; the second slit array 50 includes a second slit therein; the third slit array 60 includes a third slit 61 and a fourth slit 62, the third slit 61 is located between the grating array 30 and the fourth slit array 70, and the width of the third slit 61 is smaller than the width of the fourth slit 62; the fourth slit array 70 includes three fifth slits 71, sixth slits 72 and seventh slits 73 with sequentially increasing widths, the fifth slits 71 are located between the fourth slits 62 and the sixth slits 72, and the sixth slits 72 are located between the fifth slits 71 and the seventh slits 73; the fifth slit array 80 includes an eighth slit 81 and a ninth slit 82 therein, the eighth slit 81 being located between the grating array 30 and the ninth slit 82, and a width of the eighth slit 81 being smaller than a width of the ninth slit 82; the sixth slit array 90 includes three slits 91, an eleventh slit 92 and a twelfth slit 93 with sequentially increasing widths, the tenth slit 91 being located between the ninth slit 82 and the eleventh slit 92, and the eleventh slit 92 being located between the tenth slit 91 and the twelfth slit 93.

Specifically, the invention can rapidly distinguish the defocusing direction and position the defocusing position by arranging different slit arrays and sequentially increasing the slit widths inside the slit arrays.

The invention also provides a defocus compensation method based on the exposure equipment, as shown in fig. 3, which comprises the following implementation steps:

s100, detecting a silicon wafer to be detected by using an optical path imaging device to obtain a target image;

specifically, the silicon wafer to be measured is placed on the movable displacement table, and the optical path imaging device generates a target image corresponding to the silicon wafer to be measured.

And S200, processing the target image to obtain the defocus amount of the silicon wafer to be detected, and performing defocus compensation according to the defocus amount.

Specifically, the defocus amount of the silicon wafer to be detected can be obtained by processing the target graph, and the exposure equipment performs defocus compensation on the silicon wafer to be detected according to the defocus amount.

In one implementation, the electron beam exposure machine further comprises a calculation control module; the step S200 specifically includes:

step S210, the calculation control module performs image recognition on the target image to obtain a recognition result;

step S220, if the identification result is that the target image has a grating mark, the calculation control module processes the target image to obtain a first defocus amount of the silicon wafer to be detected, and the exposure machine body performs defocus compensation on the silicon wafer to be detected according to the first defocus amount;

And step S230, if the identification result is that the target image does not have the grating mark, the calculation control module divides the target image to obtain a target slit mark image, processes the target slit mark image to obtain the final defocus amount of the silicon wafer to be tested, and the exposure machine body performs defocus compensation on the silicon wafer to be tested according to the final defocus amount.

Specifically, image recognition is carried out on a target image, if the recognition result is that a grating mark is arranged in the target image, the defocus amount is not large at the moment, the target image is processed to obtain a first defocus amount of the silicon wafer to be detected, and defocus compensation is carried out on the silicon wafer to be detected according to the first defocus amount. If the identification result is that the target image does not have the grating mark, the defocus amount is larger at the moment, the target image is segmented to obtain a target slit mark image, the target slit mark image is processed to obtain the final defocus amount of the silicon wafer to be detected, and the exposure machine body performs defocus compensation on the silicon wafer to be detected according to the final defocus amount. According to the invention, whether the target image contains the grating mark is judged, so that the defocus amount is judged, and the defocus amounts are respectively processed according to the defocus amounts with different degrees, so that the defocus amount detection range is enlarged, and the defocus compensation can be accurately performed.

In one implementation, before the step S100, the method further includes:

detecting a reference silicon wafer by using an optical path imaging device to obtain a reference image and a plurality of non-reference images;

the calculation control module processes the plurality of non-reference images to obtain a calibration curve of the reference silicon wafer;

the reference silicon wafer is placed on the upper surface of the movable displacement table, the reference image is an image corresponding to the reference silicon wafer when the reference silicon wafer is located at the reference position of the movable displacement table, the reference position is a position where the defocus amount of the reference silicon wafer is zero, the non-reference image is an image corresponding to the reference silicon wafer when the reference silicon wafer is located at the non-reference position of the movable displacement table, the defocus amount of the reference silicon wafer is non-zero when the reference silicon wafer is located at the non-reference position of the movable displacement table, and the calibration curve comprises a corresponding relation between the defocus amount and the position of the movable displacement table.

Specifically, a reference silicon wafer is placed on the movable displacement table, and a reference image corresponding to the reference silicon wafer generated by the optical path imaging device is acquired. And moving the movable displacement table in the Z-axis direction to obtain a non-reference image of the reference image. And generating a calibration curve according to the non-reference image and the position of the movable displacement table. The invention can support the follow-up accurate defocusing compensation of the silicon wafer to be tested by generating the calibration curve.

In one implementation, the step S220 specifically includes:

if the identification result is that the target image has a grating mark, the calculation control module divides the target image to obtain a first target grating image, and processes the first target grating image to obtain a first digital grating corresponding to the first target grating image, wherein the first digital grating is stored in a matrix form;

the calculation control module converts the first target grating image into a matrix form to obtain a first grating image matrix, multiplies the first grating image matrix and the first digital grating to obtain a first grating superposition matrix, and processes the first grating superposition matrix to obtain a first grating superposition image;

the calculation control module carries out spline interpolation processing on the first grating superimposed image to obtain a first smooth image, fits the first smooth image to obtain a first fitting curve, and obtains the center coordinate of the first grating superimposed image according to the first fitting curve;

the calculation control module obtains a corresponding first reference grating image in a reference image according to a first target grating image, processes the first reference grating image to obtain a first reference grating superimposed image, processes the first reference grating superimposed image to obtain a central coordinate of the first reference grating superimposed image, compares the central coordinate of the first grating superimposed image with the central coordinate of the first reference grating superimposed image to obtain a first central position variation of the silicon wafer to be measured, and obtains a first defocus amount of the silicon wafer to be measured according to the first central position variation;

And the exposure machine body adjusts the current value of the objective lens coil in the exposure machine body according to the first defocus amount so as to perform defocus compensation on the silicon wafer to be detected.

Specifically, when the light beam reaches the image sensor, the image sensor generates a target image with a feature mark. The feature marks include grating marks and slit marks. The target image is a gray image with resolution of m×n, the lateral gray distribution is quasi-sinusoidal, the amplitude is maximum a, the minimum b, and b is non-zero.

In one implementation, the target image is an 8-bit gray scale image, the amplitude maximum a is less than 255, and the target image is a raster image with good contrast.

When the target image is provided with the grating marks, the grating marks are segmented and extracted, and a first target grating image containing the grating marks is obtained. The size of the first target grating image is smaller than that of the target image, and the first target grating image comprises at least one complete grating period. According to the invention, the first target grating image is used for analysis, and the size of the first target grating image is smaller than that of the target image, so that the first target grating image can be rapidly processed to perform defocusing compensation on the silicon wafer to be tested, and the overall defocusing compensation efficiency is effectively improved.

And carrying out pixel coding on the first target grating image to obtain a corresponding first digital grating. And storing the first target grating image in a matrix form to obtain a first target grating matrix. Wherein, in the first digital grating, only two numbers of 0 and 1 are adopted, 1 represents bright stripes, and 0 represents dark stripes. The first digital grating has a grating period of 50% and includes a portion with a number of 1 and a portion with a number of 0. The gray scale of the original optical grating is maintained at the part of the first digital grating which is 1; the portion of the first digital grating that is 0 clears the gray scale distribution of the original optical grating. Multiplying the first digital grating by the first target grating to obtain a first grating superposition matrix. According to the invention, the target grating matrix is obtained by combining the digital grating and the grating array, the first grating superposition array is processed to obtain the first grating superposition image, and the first grating superposition image is an image containing moire fringes, so that the complexity of an optical path is reduced, and the situation of higher assembly difficulty caused by assembling a plurality of gratings is avoided.

As shown in fig. 4, fig. 4 is a schematic view of the displaced moire fringes. The moire fringes can amplify the tiny displacement, and the amplification relation is expressed as follows: Wherein->The variation of moire fringes is that p is the grating period of an optical grating array, p' is the grating period of a digital grating, theta is the included angle between incident light and the vertical line on the surface of the silicon wafer, and +.>Is the defocus amount of the silicon wafer. The invention can accurately perform defocusing compensation by judging the micro displacement by using the moire fringes.

And processing the first grating superimposed image, updating all pixels with 0 in the first grating superimposed image to b to obtain a reinforced grating superimposed image, and performing low-pass filtering on the reinforced grating superimposed image to obtain an initial smooth image. Since the defocus amount is only related to the x-position of the initial smoothed image. And performing cubic spline interpolation processing on the x-axis direction of the initial smooth image to obtain a first smooth curve. And carrying out fitting treatment on the first smooth curve to obtain a first fitting curve. And obtaining the coordinate x of the maximum value and the gray value g corresponding to the maximum value according to the first fitting curve. And obtaining coordinates x1 and x2 with g/2 gray values at two sides of the distance maximum value, and taking (x1+x2)/2 as the center coordinate of the first grating superimposed image.

Obtaining a corresponding first reference grating image in a reference image according to a first target grating image, processing the first reference grating image to obtain a first reference grating superimposed image, processing the first reference grating superimposed image to obtain a central coordinate of the first reference grating superimposed image, comparing the central coordinate of the first grating superimposed image with the central coordinate of the first reference grating superimposed image to obtain a first central position change amount of the silicon wafer to be detected, and obtaining a first defocus amount of the silicon wafer to be detected according to the first central position change amount. According to the invention, the center coordinates of the first grating superimposed image and the center coordinates of the first reference grating superimposed image are compared, so that defocus compensation is further carried out, and more accurate defocus compensation can be realized.

In one implementation, the step S230 specifically includes:

if the identification result is that the target image does not have the grating mark, the calculation control module divides the target image to obtain a target slit mark image, and the size of the target slit mark image is smaller than that of the target image;

the calculation control module determines a target slit region in the target slit mark image, acquires a gray value of each row in the target slit region, and determines the center coordinate of the target slit region according to the gray value of each row in the target slit region;

the calculation control module determines a corresponding target reference slit region in the reference image according to the target slit region, processes the target reference slit region to obtain a center coordinate of the target reference slit region, and compares the center coordinate of the target slit region with the center coordinate of the target reference slit region to obtain a second center position variation of the silicon wafer to be detected;

the calculation control module obtains the final defocus amount of the silicon wafer to be measured according to the second central position variation;

and the exposure machine body adjusts the current value of the objective lens coil according to the final defocus amount so as to perform defocus compensation on the silicon wafer to be detected.

Specifically, when the target image does not have the grating mark, only the slit mark is formed at the moment, and the defocus amount of the silicon wafer to be measured is large. And dividing the target image to obtain a target slit mark image, wherein the size of the target slit mark image is smaller than that of the target image. The target slit mark image is obtained by dividing the target image, and the size of the target slit mark image is smaller than that of the target image. The image processing efficiency can be quickened, and the defocusing adjusting efficiency can be improved.

In one implementation, the target slit mark image is a partial image corresponding to a slit mark located in the central region. By capturing the target slit mark image, an average gray scale distribution is calculated in the vertical slit direction, and the target slit mark image is converted into a one-dimensional slit signal which exhibits a gaussian-like distribution. And obtaining the gray level distribution of the target slit region by calculating the gray level value of each row of the target slit region in the target slit mark image. As shown in fig. 5, fig. 5 is a schematic view of gray scale distribution of the target slit region. After the gray value of each line in the target slit region is obtained, the left and right boundaries of the distribution are judged according to a peak half-width method, and then the intermediate value is taken as the central position coordinate of the target slit mark image. The relationship of the second center position variation amount (i.e., slit center position variation amount) and the second defocus amount can be expressed as: Wherein (1)>Is the second center position variation (i.e., slit center position variation), +.>And A is the magnification of the optical magnifier group, and theta is the included angle between the incident light and the vertical line on the surface of the silicon wafer. I.e. the second amount of change in center position and the second amount of defocus are in a linear relationship.

In one implementation, the target slit marker image is a partial image corresponding to a slit array located in the extended region.

In one implementation, the target slit mark image is a partial image corresponding to the third slit array. As shown in fig. 6, fig. 6 is a schematic diagram of gray scale distribution of a partial image corresponding to the third slit array. There are two peaks in the gray scale distribution, and the peak widths are ordered from small to large. And detecting the gray level distribution of the target slit mark image to obtain the central position coordinate of the target slit mark image, and obtaining the defocus amount of the silicon wafer to be detected according to the central position coordinate of the target slit mark image and the central position coordinate information of the target reference slit region. According to the invention, the slit mark of the expansion area is arranged on the mask plate, so that the measurement range can be greatly expanded, and the defocusing direction of the silicon wafer can be rapidly and accurately judged by detecting the distribution of peak width.

In one embodiment of the present invention, the calculating and controlling module obtains a final defocus amount of the silicon wafer to be measured according to the second central position variation, including:

the calculation control module obtains a second defocus amount of the silicon wafer to be measured according to the second central position variation, and determines a target position according to the second defocus amount and the calibration curve;

the calculation control module controls the movable displacement platform to move to a target position;

the calculation control module acquires a second target image, wherein the second target image is a target image corresponding to the silicon wafer to be detected, generated by the optical path imaging device, when the movable displacement platform moves to a target position, and the second target image is provided with a grating mark;

the calculation control module divides the second target image to obtain a second target grating image, and processes the second target grating image to obtain a second digital grating corresponding to the second target grating image, wherein the second digital grating is stored in a matrix form;

the calculation control module converts the second target grating image into a matrix form to obtain a second grating image matrix, multiplies the second grating image matrix and the second digital grating to obtain a second grating superposition matrix, and processes the second grating superposition matrix to obtain a second grating superposition image;

The calculation control module carries out spline interpolation processing on the second grating superimposed image to obtain a second smooth image, fits the second smooth image to obtain a second fitting curve, and obtains the center coordinates of the second grating superimposed image according to the second fitting curve;

the calculation control module obtains a corresponding second reference grating image in the reference image according to the second target grating image, processes the second reference grating image to obtain a second reference grating superimposed image, processes the second reference grating superimposed image to obtain a center coordinate of the second reference grating superimposed image, compares the center coordinate of the second grating superimposed image with the center coordinate of the second reference grating superimposed image to obtain a second center position variation of the silicon wafer to be measured, and obtains a final defocus amount of the silicon wafer to be measured according to the second center position variation.

Specifically, a second defocus amount of the silicon wafer to be measured is obtained according to the second central position variation, a target position is determined according to the second defocus amount and the calibration curve, and the movable displacement table is controlled to move to the target position. At the moment, the optical path generating device generates a second target image corresponding to the silicon wafer to be detected, and the second target image comprises a grating mark. The invention realizes the defocus compensation of larger amplitude by adjusting the movable displacement table according to the second defocus amount, and then processes the second target image to further accurately perform defocus compensation. When the final defocus amount is obtained, the final defocus amount is in the order of hundred nanometers to nanometers, and the current value of the objective lens coil in the electron beam exposure machine needs to be modified to perform defocus compensation.

According to the invention, the target image is processed, defocus compensation is respectively carried out according to the content of the target image, when the target image only contains the grating mark, the defocus amount is smaller, the first defocus amount is obtained through the target image, and the current value of an objective lens coil in the electron beam exposure machine is modified according to the first defocus amount to complete defocus compensation; when the target image only contains the slit mark, the defocus amount is larger, a second defocus amount is obtained through the target image, the movable displacement table is adjusted according to the second defocus amount, the second target image is further generated, the second target image is processed, a third defocus amount is obtained, and the current value of an objective lens coil in the electron beam exposure machine is modified according to the third defocus amount to perform defocus compensation.

In one implementation, the processing module for defocus compensation in the e-beam exposure machine may be a CPU or GPU.

In one implementation, the processing module for defocus compensation in the electron beam exposure machine is an FPGA module, and the FPGA module performs image processing and calculates defocus.

In one implementation, a reference silicon wafer and a plurality of silicon wafers to be measured are fixed on a stage of a movable displacement stage. And processing the reference silicon wafers to obtain a calibration curve and reference images corresponding to the reference silicon wafers, and then sequentially processing and calibrating the plurality of silicon wafers to be tested until the defocus compensation of all the silicon wafers to be tested is completed.

The invention carries out multiple defocusing detection on the same silicon wafer, and as shown in figure 7, the measurement error precision is not more than ten nanometers.

The present invention also provides a computer-readable storage medium storing a computer program executable for implementing the steps of the defocus compensation method as described above; as described in detail above.

In summary, the present invention discloses an exposure apparatus and a defocus compensation method, the exposure apparatus comprising: an optical path imaging device and an electron beam exposure machine; the optical path imaging device comprises a light source, a beam expanding collimation device, an iris diaphragm, a first projection device, a first reflection device, a second projection device, an optical amplifying assembly and an image sensor which are sequentially arranged; a mask plate is arranged between the iris diaphragm and the first projection device, and a grating array and a plurality of slit arrays with different widths are arranged on the mask plate. According to the invention, the mask plate with the grating array and the slit arrays with different widths is used, a plurality of slit marks with different widths are introduced into the light beam, when the defocus amount is large, the image sensor can generate partial images with slits with different widths, and the partial images with the slits with different widths contain more defocus information, so that the high-precision defocus detection of an expanded range is realized on the electron beam exposure machine.

It is to be understood that the invention is not limited in its application to the examples described above, but is capable of modification and variation in light of the above teachings by those skilled in the art, and that all such modifications and variations are intended to be included within the scope of the appended claims.

Claims (10)

1. An exposure apparatus, characterized by comprising: an optical path imaging device and an electron beam exposure machine; the optical path imaging device comprises a light source, a beam expanding collimation device, an iris diaphragm, a first projection device, a first reflection device, a second projection device, an optical amplifying assembly and an image sensor which are sequentially arranged; when the exposure equipment is in a detection state and the target silicon wafer is positioned between the first projection device and the second projection device, the light source, the beam expansion collimation device, the iris diaphragm, the first projection device, the first reflection device, the target silicon wafer, the second reflection device, the second projection device, the optical amplification assembly and the image sensor form a light path; a mask plate is arranged between the iris diaphragm and the first projection device, and a grating array and a plurality of slit arrays with different widths are arranged on the mask plate and used for introducing grating marks and a plurality of slit marks with different widths into the optical path imaging optical path; the optical path imaging device is used for generating a target image with a grating mark and a slit mark;

The electron beam exposure machine comprises a movable displacement table and an exposure machine body; the movable displacement platform is positioned between the first reflecting device and the second reflecting device and is used for placing a target silicon wafer, and the exposure machine body is positioned above the movable displacement platform and is used for performing defocus compensation on the target silicon wafer and exposing the target silicon wafer according to a target image.

2. The exposure apparatus according to claim 1, comprising: the mask plate is provided with a grating array, a first slit array, a second slit array, a third slit array, a fourth slit array, a fifth slit array and a sixth slit array; the grating array is positioned at the center of the mask plate, the first slit array is positioned at a first position of the grating array, the second slit array is positioned at a second position of the grating array, the third slit array and the fourth slit array are arranged in parallel and are positioned at a third position of the grating array, and the third slit array is positioned between the grating array and the fourth slit array; the fifth slit array is arranged in parallel with the sixth slit array and is positioned in a fourth direction of the grating array, and the fifth slit array is positioned between the grating array and the sixth slit array.

3. The exposure apparatus according to claim 2, comprising: the first slit array comprises first slits; the second slit array comprises a second slit; the third slit array comprises a third slit and a fourth slit, the third slit is positioned between the grating array and the fourth slit, and the width of the third slit is smaller than the width of the fourth slit; the fourth slit array comprises three fifth slits, a sixth slit and a seventh slit with sequentially increasing widths, wherein the fifth slit is positioned between the fourth slit and the sixth slit, and the sixth slit is positioned between the fifth slit and the seventh slit; the fifth slit array comprises an eighth slit and a ninth slit, the eighth slit is positioned between the grating array and the ninth slit, and the width of the eighth slit is smaller than that of the ninth slit; the sixth slit array comprises a tenth slit, an eleventh slit and a twelfth slit with sequentially increasing widths, wherein the tenth slit is positioned between the ninth slit and the eleventh slit, and the eleventh slit is positioned between the tenth slit and the eleventh slit.

4. A defocus compensation method realized based on the exposure apparatus according to any one of claims 1 to 3, comprising:

detecting a silicon wafer to be detected by using an optical path imaging device to obtain a target image;

and processing the target image to obtain the defocus amount of the silicon wafer to be detected, and performing defocus compensation according to the defocus amount.

5. The defocus compensation method of claim 4, wherein said electron beam exposure machine further comprises a calculation control module; the target image is processed to obtain the defocus amount of the silicon wafer to be detected, and defocus compensation is carried out according to the defocus amount, and the method comprises the following steps:

the calculation control module performs image recognition on the target image to obtain a recognition result;

if the identification result is that the target image has the grating mark, the calculation control module processes the target image to obtain a first defocus amount of the silicon wafer to be detected, and the exposure machine body performs defocus compensation on the silicon wafer to be detected according to the first defocus amount;

if the identification result is that the target image does not have the grating mark, the calculation control module divides the target image to obtain a target slit mark image, processes the target slit mark image to obtain the final defocus amount of the silicon wafer to be detected, and the exposure machine body performs defocus compensation on the silicon wafer to be detected according to the final defocus amount.

6. The defocus compensation method of claim 5, wherein before the target image is obtained by detecting the silicon wafer to be detected by the optical path imaging device, further comprising:

detecting a reference silicon wafer by using an optical path imaging device to obtain a reference image and a plurality of non-reference images;

the calculation control module processes the plurality of non-reference images to obtain a calibration curve of the reference silicon wafer;

the reference silicon wafer is placed on the upper surface of the movable displacement table, the reference image is an image corresponding to the reference silicon wafer when the reference silicon wafer is located at the reference position of the movable displacement table, the reference position is a position where the defocus amount of the reference silicon wafer is zero, the non-reference image is an image corresponding to the reference silicon wafer when the reference silicon wafer is located at the non-reference position of the movable displacement table, the defocus amount of the reference silicon wafer is non-zero when the reference silicon wafer is located at the non-reference position of the movable displacement table, and the calibration curve comprises a corresponding relation between the defocus amount and the position of the movable displacement table.

7. The defocus compensation method of claim 6, wherein if the recognition result is that the target image has a grating mark, the calculation control module processes the target image to obtain a first defocus amount of the silicon wafer to be measured, and the exposure machine body performs defocus compensation on the silicon wafer to be measured according to the first defocus amount, and the method comprises:

If the identification result is that the target image has a grating mark, the calculation control module divides the target image to obtain a first target grating image, and processes the first target grating image to obtain a first digital grating corresponding to the first target grating image, wherein the first digital grating is stored in a matrix form;

the calculation control module converts the first target grating image into a matrix form to obtain a first grating image matrix, multiplies the first grating image matrix and the first digital grating to obtain a first grating superposition matrix, and processes the first grating superposition matrix to obtain a first grating superposition image;

the calculation control module carries out spline interpolation processing on the first grating superimposed image to obtain a first smooth image, fits the first smooth image to obtain a first fitting curve, and obtains the center coordinate of the first grating superimposed image according to the first fitting curve;

the calculation control module obtains a corresponding first reference grating image in a reference image according to a first target grating image, processes the first reference grating image to obtain a first reference grating superimposed image, processes the first reference grating superimposed image to obtain a central coordinate of the first reference grating superimposed image, compares the central coordinate of the first grating superimposed image with the central coordinate of the first reference grating superimposed image to obtain a first central position variation of the silicon wafer to be measured, and obtains a first defocus amount of the silicon wafer to be measured according to the first central position variation;

And the exposure machine body adjusts the current value of the objective lens coil in the exposure machine body according to the first defocus amount so as to perform defocus compensation on the silicon wafer to be detected.

8. The defocus compensation method of claim 6, wherein if the recognition result is that the target image does not have a grating mark, the calculation control module segments the target image to obtain a target slit mark image, processes the target slit mark image to obtain a final defocus amount of the silicon wafer to be measured, and the exposure machine body performs defocus compensation on the silicon wafer to be measured according to the final defocus amount, and comprises:

if the identification result is that the target image does not have the grating mark, the calculation control module divides the target image to obtain a target slit mark image, and the size of the target slit mark image is smaller than that of the target image;

the calculation control module determines a target slit region in the target slit mark image, acquires a gray value of each row in the target slit region, and determines the center coordinate of the target slit region according to the gray value of each row in the target slit region;

The calculation control module determines a corresponding target reference slit region in the reference image according to the target slit region, processes the target reference slit region to obtain a center coordinate of the target reference slit region, and compares the center coordinate of the target slit region with the center coordinate of the target reference slit region to obtain a second center position variation of the silicon wafer to be detected;

the calculation control module obtains the final defocus amount of the silicon wafer to be measured according to the second central position variation;

and the exposure machine body adjusts the current value of the objective lens coil according to the final defocus amount so as to perform defocus compensation on the silicon wafer to be detected.

9. The defocus compensation method of claim 8, wherein the calculating control module obtains a final defocus amount of the silicon wafer to be measured according to the second central position variation, comprising:

the calculation control module obtains a second defocus amount of the silicon wafer to be measured according to the second central position variation, and determines a target position according to the second defocus amount and the calibration curve;

the calculation control module controls the movable displacement platform to move to a target position;

The calculation control module acquires a second target image, wherein the second target image is a target image corresponding to the silicon wafer to be detected, generated by the optical path imaging device, when the movable displacement platform moves to a target position, and the second target image is provided with a grating mark;

the calculation control module divides the second target image to obtain a second target grating image, and processes the second target grating image to obtain a second digital grating corresponding to the second target grating image, wherein the second digital grating is stored in a matrix form;

the calculation control module converts the second target grating image into a matrix form to obtain a second grating image matrix, multiplies the second grating image matrix and the second digital grating to obtain a second grating superposition matrix, and processes the second grating superposition matrix to obtain a second grating superposition image;

the calculation control module carries out spline interpolation processing on the second grating superimposed image to obtain a second smooth image, fits the second smooth image to obtain a second fitting curve, and obtains the center coordinates of the second grating superimposed image according to the second fitting curve;

The calculation control module obtains a corresponding second reference grating image in the reference image according to the second target grating image, processes the second reference grating image to obtain a second reference grating superimposed image, processes the second reference grating superimposed image to obtain a center coordinate of the second reference grating superimposed image, compares the center coordinate of the second grating superimposed image with the center coordinate of the second reference grating superimposed image to obtain a second center position variation of the silicon wafer to be measured, and obtains a final defocus amount of the silicon wafer to be measured according to the second center position variation.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program executable for implementing the steps of the defocus compensation method of any one of claims 4-9.