CN105467781B - A kind of mark and alignment methods with focusing and slant correction design - Google Patents

Abstract

The invention discloses a mark with focus adjustment and tilt correction design and an alignment method. The mark includes an alignment mark and at least one pair of focus marks, and the center of the alignment mark is located at any pair of focus adjustment marks. At the midpoint of the marking line, the center of any pair of focusing marks is symmetrical to the two sides of the alignment mark, the alignment mark is a cross-shaped mark or a rice-shaped mark, and the focusing mark is a square type focusing mark or grating type focusing mark, the line width of the alignment mark is greater than the discretization granularity, the line width of the alignment mark is greater than twice the PSF width, and the present invention also provides an alignment mark applied to the mark The alignment method achieves high-precision focusing and leveling of the alignment marks. Compared with the prior art, the mark provided by the present invention eliminates the influence of tilt on the mark while adjusting the focus, and improves the measurement reproducibility; in addition, the mechanism of the influence of distortion on the measurement reproducibility is analyzed, and the The limited condition of the mark width further improves the measurement reproducibility.

Description

A Marking and Alignment Method with Focus Adjustment and Tilt Correction Design

technical field

The present invention relates to a mark with focus adjustment and tilt correction design and an alignment method, in particular to a mark comprising an alignment mark and at least one pair of focus adjustment marks and an alignment method thereof.

Background technique

In the integrated circuit manufacturing process, a complete chip usually requires multiple photolithography exposures to complete. Except for the first layer of lithography, the other layers of lithography must be accurately positioned before exposure to the graphics of this level and the graphics (ie marks) left by the exposure of the previous level, so as to ensure that there is a correct relationship between the graphics of each layer. The relative position, that is, the overlay accuracy.

In the prior art, an alignment system based on the principle of optical imaging is one of the commonly used systems in lithography machines, such as Nikon FIA system, Ultratec MVS and so on. The marking style of FIA is single, as shown in Figure 1. Ultratec MVS has a label learning function, and the label has no fixed form. In addition, US6344698B2 and CN102103336 considered the influence of technology on marking, and respectively designed marks less affected by the technology.

In fact, in addition to the process affecting the measurement accuracy, there are many factors that will affect the measurement accuracy.

First, distortion is a common aberration, and its influence on measurement reproducibility cannot be ignored in optical imaging systems. Its influence mechanism is that the nonlinear coupling between the uncertainty of the position of the measurement mark in the field of view and the distortion affects the measurement reproducibility, as shown in Figure 2. Although the FIA system reduces the influence of distortion through multiple iterations so that the marker is always measured at a fixed position, the iterations consume a lot of time.

Secondly, as shown in Figure 3, the defocus tilt effect of the mark will also affect the measurement reproducibility, and the tilt angle a of the measured object will produce an error D. Although the focus and leveling system in the photolithography machine can realize the correction of tilt and defocus, because the focus and leveling measurement surface is the upper surface of the photoresist, and the alignment mark is sometimes located on the lower surface of the photoresist, the photoresist The thickness of the film has certain fluctuations, so the focus and leveling system is not enough to achieve high-precision focus and leveling of the alignment mark.

In addition, in the optical imaging system, although the auto-focus technology has been widely used in the alignment sensor, the sensor that realizes the tilt correction while auto-focusing has not been invented yet.

Contents of the invention

The object of the present invention is to provide a mark and alignment method with focus adjustment and tilt correction design, the mark includes an alignment mark and at least one pair of focus adjustment marks, which are used to eliminate the influence of tilt on the mark while adjusting the focus , while reducing the influence of distortion on the label.

In order to achieve the above purpose, the present invention provides a marker with focus adjustment and tilt correction design, including:

an alignment mark and at least one pair of focus marks, the center of the alignment mark is located at the midpoint of any pair of the focus marks, and the center of any pair of the focus marks is symmetrical to the alignment mark on both sides.

Further, the alignment mark is a cross-shaped mark or a V-shaped mark.

Further, the line width of the alignment mark is larger than the discretization granularity.

Further, the line width of the alignment mark is greater than twice the PSF width.

Further, the focus adjustment mark is a square type focus adjustment mark or a grating type focus adjustment mark.

Further, the grating-type focusing marks are horizontal grating-type focusing marks or vertical grating-type focusing marks.

Further, the marks include alignment marks and focus marks, and the focus marks include a pair of square-type focus marks, a pair of horizontal grating-type focus marks, and a pair of vertical grating-type marks. Focus mark.

The present invention also proposes an alignment method, which is applied to the mark, and the alignment method includes the following steps:

(1) The alignment system initially aligns the mark on the workpiece with focus adjustment and tilt correction design;

(2) Move the workpiece table vertically with a predetermined step distance, and obtain the optimal focal plane position {P _i , Q _i } of each pair of focusing marks according to the focal plane criterion and its optimal focal plane determination method, that is, the two focusing marks The vertical best focus plane position of the focus marks, where P _i and Q _i are the positions of the ith pair of focus marks, where i≥3;

(3) Obtain the original value M _i of the best focus plane position of the alignment mark and the original value T _i of its inclination according to the best focus plane position {P _i , Q _i } of each pair of focusing marks, wherein,

(4) Determine the optimal position of the workpiece (silicon wafer) in the field of view according to the original value M _i of the best focal plane position of the alignment mark and the original value T _i of its inclination through mean filtering or median filtering method The focal plane position M and its inclination T;

(5) According to the M of the best focal plane position of the workpiece (silicon wafer) in the field of view and its inclination T, move the workpiece table vertically to compensate the vertical direction of the alignment mark to be compensated among the plurality of alignment marks Location.

Further, the focal plane criterion includes a gradient magnitude method and a PSF width method.

Further, the focus adjustment mark is a square type focus adjustment mark or a grating type focus adjustment mark.

Further, the focal plane criterion of the grating-type focusing mark is the gradient magnitude of the mark image, that is, the image is convolved with a Sobel operator and accumulated.

Further, the focal plane criterion of the square-shaped focusing mark is the PSF width of the mark. The specific method is to extract the gray value of a row or rows of the square-shaped focusing mark to obtain gray degree distribution, and solve the h value, according to the size of the h value to determine the focal plane position.

Further, the method for determining the best focus plane obtains {P _i , Q _i } of any pair of the focusing marks in the alignment marks, and the first method may be adopted, and the first method includes:

(1) Move the workpiece table with a predetermined step;

(2) Take an image at each location;

(3) extracting the focal plane criterion value from the image;

(4) Find the best focal plane position by fitting the curve.

Further, the method for determining the best focus plane obtains {P _i , Q _i } of any pair of the focusing marks in the alignment marks, and a second method may also be used, and the second method includes:

(1) Calibrate the relationship between the focal plane criterion value and the vertical position;

(2) Take an image at the current position of the workpiece table;

(3) extracting the focal plane criterion value from the image;

(4) Obtain the distance d of the current position from the focal plane according to the calibration data;

(5) On the basis of the current position, move the workpiece table by d and -d respectively, and take images respectively to obtain focal plane criterion values V ₁ and V ₂ ;

(6) Compare V ₁ and V ₂ to determine the focal plane position.

Compared with the prior art, the invention discloses a mark and an alignment method with focus adjustment and tilt correction design, and realizes high-precision focus adjustment and leveling of the alignment mark. On the one hand, it eliminates the influence of tilt on the mark while adjusting the focus, and improves the measurement reproducibility; on the other hand, the mechanism of the influence of distortion on the measurement reproducibility is analyzed, and accordingly the width of the alignment mark is given. Restricted conditions further improve measurement reproducibility.

Description of drawings

Figure 1 is a schematic diagram of the FIA mark;

Figure 2 is a schematic diagram of the influence of distortion on measurement reproducibility;

Figure 3 is a schematic diagram of the defocus tilt effect;

Fig. 4 is a schematic diagram of marking in Embodiment 1 of the present invention;

Fig. 5 is a schematic diagram of extracting the gray value of a mark in Embodiment 1 of the present invention;

Fig. 6 is a schematic diagram of the gray distribution of the gray value of the extracted mark in Embodiment 1 of the present invention;

7 is a schematic diagram of the positional relationship between the alignment mark and the focus mark in Embodiment 1 of the present invention;

Fig. 8 is a schematic diagram of calculating the vertical pose Mi and Ti of each alignment mark according to the obtained vertical positions Pi and Qi of each focusing mark;

FIG. 9 is a schematic diagram of calculating the position of the workpiece (silicon wafer) according to the positions X and Y of several alignment marks.

Among them, a: the inclination angle of the measured object, D: error, 10: square type focus mark, 30: square type focus mark, 11: horizontal grating type focus mark, 31: horizontal grating type focus mark, 12: Vertical grating type focus mark, 32: vertical grating type focus mark, 20: alignment mark, h: PSF width, 1: best focus plane of one side focus mark, 2: best alignment mark Focal plane, 3: The plane of best focus on the other side of the focusing mark.

detailed description

The specific implementation manner of the present invention will be described in more detail below with reference to schematic diagrams. Advantages and features of the present invention will be apparent from the following description and claims. It should be noted that all the drawings are in a very simplified form and use imprecise scales, and are only used to facilitate and clearly assist the purpose of illustrating the embodiments of the present invention.

Embodiment one

As shown in Figure 4, Embodiment 1 of the present invention provides a mark with a focus adjustment and tilt correction design, the mark includes an alignment mark 20 and at least one pair of focus marks, and the focus mark is a square focus mark or a grating type focus mark, the grating type focus mark is a horizontal grating type focus mark or a vertical grating type focus mark, and the focus mark includes a pair of the square type focus marks, a pair of the Horizontal grating type focus mark and a pair of said vertical grating type focus mark, said focus mark includes square type focus mark 10, square type focus mark 30, horizontal grating type focus mark 11, horizontal grating type A focus mark 31 , a vertical raster-type focus mark 12 and a vertical raster-type focus mark 32 . Among them, the square-shaped focusing mark 10 and the square-shaped focusing mark 30 are a pair of square-shaped focusing marks, and the horizontal grating-shaped focusing mark 11 and the horizontal grating-shaped focusing mark 31 are a pair of grating-shaped focusing marks. The raster type focus adjustment mark 12 and the vertical raster type focus adjustment mark 32 are a pair of raster type focus adjustment marks. The alignment mark 20 is a cross-shaped mark or a rice-shaped mark for alignment. In Embodiment 1, the alignment mark 20 is a cross-shaped mark, and the cross-shaped mark includes horizontal lines and vertical lines. The horizontal lines and the vertical lines are perpendicular to each other, and the center of the alignment mark 20 is located on the midpoint of any pair of the focusing marks, and the center of any pair of the focusing marks is symmetrical to both sides of the alignment mark, The alignment marks 20 and the focus marks are used in combination to achieve high-precision focus and leveling of the alignment marks. When determining the best focal plane of the focus mark, it is necessary to select the focal plane criterion. There are two common focal plane criteria, namely the gradient amplitude method and the PSF (point spread function) width method.

In the first embodiment, the focal plane criterion of the grating-type focusing mark adopts the gradient magnitude method, that is, the image is convolved with the Sobel operator and accumulated. The formula of the focal plane criterion is as follows:

Dx=Image*SobelX

Dy=Image*SobelY

Among them, Image is the original image, SobelX and SobelY are the horizontal and vertical Sobel operators, respectively, Dx and Dy are the horizontal and vertical edge detection images, and V is the gradient approximation value, that is, the focal plane criterion value.

In Embodiment 1, the focal plane criterion of the square-shaped focusing mark adopts the PSF width method, that is, the gray value of a certain row or several rows of the square mark is extracted, as shown in the dotted line box in Figure 5, and the result shown in Figure 6 is obtained. For the gray distribution shown, solve the PSF width h in Figure 6, that is, the focal plane criterion value, and then judge the focal plane position according to the PSF width h value.

After the focal plane criterion is selected, the method for determining the best focal plane is the first method or the second method.

The first method includes the following steps:

(1) Move the workpiece table with a predetermined step;

(2) Take an image at each location;

(3) extracting the focal plane criterion value from the image;

(4) Find the best focal plane position by fitting the curve.

The second method includes the following steps:

(1) Calibrate the relationship between the focal plane criterion value and the vertical position;

(2) Take an image at the current position of the workpiece table;

(3) extracting the focal plane criterion value from the image;

(4) Obtain the distance d of the current position from the focal plane according to the calibration data;

(5) On the basis of the current position, move the workpiece table by d and -d respectively, and take images respectively to obtain focal plane criterion values V ₁ and V ₂ ;

(6) Compare V ₁ and V ₂ to determine the focal plane position.

As shown in Figure 7, the alignment mark 20 is in the middle of a pair of focus marks, and the best focus plane 2 of the alignment mark 20 is the best focus plane 1 of one side focus mark and the best focus plane 1 of the other side focus mark. Mean of focal plane 3. In addition, the inclination of the alignment mark 20 can also be obtained from the best focus plane 1 of the focus mark on one side and the best focus plane 3 of the focus mark on the other side.

The alignment method of the mark comprises the following steps:

(1) After placing the workpiece on the workpiece table, the first step of focusing and leveling is realized through the focusing and leveling system (FLS);

(2) moving the workpiece table to bring the alignment mark 20 with focusing and tilt correction design into the field of view of the alignment system, and aligning to complete the preliminary measurement, thereby determining the horizontal position parameter of the focusing mark;

(3) Move the workpiece table vertically with a predetermined step distance, and obtain the optimal focal plane position {P _i , Q _i } of each pair of focusing marks according to the focal plane criterion and its optimal focal plane determination method, where,

P _i and Q _i are the positions of the ith pair of focusing marks, where i≥3;

(4) According to the best focus plane positions {P _i , Q _i } of each pair of focus marks, that is, the vertical best focus plane positions of the two focus marks, the best focus of the alignment mark 20 is obtained. The original value M _i of the surface position and the original value T _i of its inclination, where,

Please refer to Figure 8 for this step;

(5) According to the original value M _i of the best focus plane position of the alignment mark 20 and the original value T _i of its inclination, determine the position in the field of view surrounded by a plurality of alignment marks 20 by mean filtering or median filtering method The best focal plane position M and its inclination T of the workpiece (silicon wafer);

(6) According to the M of the best focal plane position of the workpiece (silicon wafer) in the field of view and its inclination T, move the workpiece table vertically to compensate the vertical position of the alignment mark to be compensated among the plurality of alignment marks ;

(7) Recalculate the horizontal position of the alignment mark 20, its horizontal position is x _WZCS , y _WZCS , and record the workpiece table position during measurement at the same time; please refer to Figure 9 for this step

(8) Obtain the position of the workpiece (silicon wafer) in the workpiece table.

Among them, step (1) is the first step of focusing and leveling, which is used for preliminary adjustment, and steps (3)-(8) are the second step of focusing and leveling, which is used to achieve high-precision focusing and leveling of alignment marks .

The above content completes the design of the mark with tilt correction, and gives the alignment method of the mark, but does not consider the influence of distortion on the alignment accuracy of the alignment mark.

As shown in FIG. 2 , the alignment mark has a certain line width, and the line width in FIG. 2 is the length of the LR line segment. The positions on both sides of the alignment mark are L and R, the center position is C, the actual imaging position of the alignment mark is L’ and R’, and the corresponding center position is CC. The calibration algorithm can establish the positional relationship between the ideal imaging point and the actual imaging point, as shown by the arrow in Figure 2. Due to the non-linearity of the distortion, point CC deviates from the imaging point C' corresponding to point C. Therefore, the actual mark test position CC deviates from the actual position of the mark. However, if the deviation is kept constant, the reproducibility can still be guaranteed. But in fact, with the position of the field of view, the amount of deviation will also change, thus affecting the measurement reproducibility.

The calculation method for the influence of distortion on alignment accuracy includes the following steps:

(1) Obtain the offset curve O(t) of the lens at each position of the field of view relative to the ideal imaging position through optical software such as Zemax and CODE V;

(2) Generate a window function according to the marked line width, and the width is T;

(3) Divide the window from the center into two adjacent sub-windows, W(t) and W(t-T/2), both of width T/2;

(4) Convolve the sub-windows W(t) and W(t-T/2) with O(t) respectively to obtain C(t) and D(t);

(5) Take E(t)=C(t)-D(t);

(6) Find the maximum value Max and the minimum value Min in E(t).

In the image space, the impact of distortion on the accuracy can be calculated as Err=Max-Min, and the corresponding object space value is Err/magnification.

From the above analysis, it can be seen that the smaller the line width of the alignment mark 20 is, the smaller the influence of distortion on precision is. However, the alignment mark 20 cannot be infinitely small, because the image is discretized by a CCD (Charge Coupled Device) at the back end, and the line width of the alignment mark 20 needs to be larger than the discretization granularity (grain size) of the image.

In addition, since the internal brightness value of the alignment mark 20 remains unchanged, there is no information about the position of the alignment mark 20 . Only at the edge of the alignment mark 20 are there features indicating the position of the alignment mark 20 . In order to accurately locate the edge of the alignment mark 20, within the range covered by the edge (i.e. the PSF width), 4 sampling points are required, so the line width of the alignment mark 20 needs to be greater than 2 times the PSF width, so as to determine the alignment mark 20. line width.

Embodiment two

Different from Embodiment 1, in Embodiment 2, the center of the alignment mark is not necessarily located at the midpoint of any pair of focusing marks, but only the distance to the two focusing marks is known. The best focus plane position of the alignment mark is obtained by using the known distance from the alignment mark to the two focusing marks. Compared with the first embodiment, the second embodiment provides a mark with less restrictive conditions, which meets the diverse needs of users.

In Embodiment 2, except for the above content, the marking and alignment methods are consistent with Embodiment 1, so details will not be repeated here.

Embodiment three

Different from the first embodiment, in the third embodiment, the alignment mark is a horizontal line mark or a vertical line mark, which is used to measure the position in one direction. Compared with Embodiment 1, Embodiment 3 provides a marker for measuring a position in one direction, which meets the diverse needs of users.

In the third embodiment, except for the above content, the marking and alignment methods are the same as those in the first embodiment, so no more details are given here.

Embodiment four

Different from the first embodiment, in the fourth embodiment, the alignment mark is a Pozier-shaped mark. Compared with Embodiment 1, Embodiment 4 provides a Pozier-shaped alignment mark, which satisfies the diverse needs of users.

In the fourth embodiment, except for the above content, the marking and alignment methods are the same as those in the first embodiment, so no more details are given here.

To sum up, the present invention discloses a mark and an alignment method with focus adjustment and tilt correction design, which realizes high-precision focus and leveling of the alignment mark. On the one hand, it eliminates the tilt while adjusting the focus, and improves the measurement reproducibility; on the other hand, it analyzes the mechanism of the influence of distortion on the measurement reproducibility, and accordingly gives the limited conditions for the mark width. Improved measurement reproducibility.

The foregoing are only preferred embodiments of the present invention, and do not limit the present invention in any way. Any person skilled in the technical field, within the scope of the technical solution of the present invention, makes any form of equivalent replacement or modification to the technical solution and technical content disclosed in the present invention, which does not depart from the technical solution of the present invention. The content still belongs to the protection scope of the present invention.

Claims (7)

1. An alignment method, characterized in that the alignment method comprises the following steps: (1) A mark with focus adjustment and tilt correction design is set on the workpiece, and the mark with focus adjustment and tilt correction design includes an alignment mark and at least one pair of focus adjustment marks, and the center of the alignment mark is located at At the midpoint of any pair of focusing marks, the alignment system initially aligns the mark with focus and tilt correction design on the workpiece; (2) Move the workpiece table vertically with a predetermined step distance, and obtain the optimal focal plane position {P _i , Q _i } of each pair of focusing marks according to the focal plane criterion and its optimal focal plane determination method, where, P _i and Q _i are the positions of the ith pair of focusing marks, where i≥3; (3) Obtain the original value M _i of the best focus plane position of the alignment mark and the original value T _i of its inclination according to the best focus plane position {P _i , Q _i } of each pair of focusing marks, wherein, <mrow><msub><mi>M</mi><mi>i</mi></msub><mo>=</mo><mfrac><mrow><msub><mi>P</mi><mi>i</mi></msub><mo>+</mo><msub><mi>Q</mi><mi>i</mi></msub></mrow><mn>2</mn></mfrac><mo>,</mo><msub><mi>T</mi><mi>i</mi></msub><mo>=</mo><mi>arctan</mi><mrow><mo>(</mo><mfrac><mrow><msub><mi>P</mi><mi>i</mi></msub><mo>-</mo><msub><mi>Q</mi><mi>i</mi></msub></mrow><mrow><mi>D</mi><mi>i</mi><mi>s</mi><mrow><mo>(</mo><msub><mi>P</mi><mi>i</mi></msub><mo>,</mo><msub><mi>Q</mi><mi>i</mi></msub><mo>)</mo></mrow></mrow></mfrac><mo>)</mo></mrow><mo>;</mo></mrow> (4) Determine the best focus plane position M of the workpiece in the field of view according to the original value M _i of the best focus plane position of the alignment mark and the original value T _i of its inclination by mean filtering or median filtering method and its slope T; (5) Moving the workpiece stage vertically according to the best focus plane position M of the workpiece and its inclination T to compensate the vertical position of the alignment mark to be compensated among the plurality of alignment marks. 2. The alignment method according to claim 1, wherein the focal plane criterion comprises a gradient magnitude method and a PSF width method. 3. The alignment method according to claim 2, wherein the focusing mark is a square-type focusing mark or a raster-type focusing mark. 4. The alignment method according to claim 3, wherein the focal plane criterion of the grating-type focusing mark is a gradient magnitude of the mark image. 5. The alignment method according to claim 3, wherein the focal plane criterion of the square-shaped focusing mark is the PSF width of the mark. 6. The alignment method according to claim 1, wherein the best focus plane determination method obtains {P _i , Q _i } of any pair of the focusing marks in the alignment marks, which can be Using a first method, the first method includes: (1) Move the workpiece table with a predetermined step; (2) Take an image at each location; (3) Extract the focal plane criterion value from the image; (4) Find the best focal plane position by fitting the curve. 7. The alignment method according to claim 1, wherein the best focus plane determination method obtains {P _i , Q _i } of any pair of the focusing marks in the alignment marks, which can be Using a second method, the second method includes: (1) Calibrate the relationship between the focal plane criterion value and the vertical position; (2) Take an image at the current position of the workpiece table; (3) extracting the focal plane criterion value from the image; (4) Obtain the distance d of the current position from the focal plane according to the calibration data; (5) On the basis of the current position, move the workpiece table by d and -d respectively, and take images respectively to obtain focal plane criterion values V ₁ and V ₂ ; (6) Compare V ₁ and V ₂ to determine the focal plane position.