CN-117784517-B - Mask precision measurement method and system

CN117784517BCN 117784517 BCN117784517 BCN 117784517BCN-117784517-B

Abstract

The invention discloses a mask precision measurement method and a mask precision measurement system, which relate to the technical field of semiconductors and solve the problems that the existing mask precision measurement is easily influenced by abnormal point positions and the measurement area is small; the method comprises the steps of constructing a measuring matrix according to measuring data, constructing a design matrix according to the design data, wherein the measuring matrix and the design matrix are two-dimensional matrices, the row-column sequence of the two-dimensional matrices corresponds to the row-column sequence of the point positions, elements of the two-dimensional matrices are combinations of an abscissa X and an ordinate Y of the corresponding point positions, calculating an X total length variance component, a Y total length variance component, an X direction vertical deviation variance component, a Y direction vertical deviation variance component and an XY combined vertical deviation according to the measuring matrix and the design matrix, calculating various precision values by constructing the measuring matrix and combining a calculation formula, and meanwhile, reducing the influence of abnormal point positions on a measuring result and expanding a measuring range.

Inventors

LI GANG
ZHENG YUCHEN
SUN DUOWEI

Assignees

成都路维光电有限公司

Dates

Publication Date: 20260512
Application Date: 20231227

Claims (8)

1. The mask precision measurement method is characterized by comprising the following steps of: selecting a plurality of point positions from the region to be measured, and acquiring measurement data and design data of each point position; Constructing a measurement matrix according to the measurement data, and constructing a design matrix according to the design data, wherein the measurement matrix and the design matrix are two-dimensional matrices, the row-column sequence of the two-dimensional matrices corresponds to the row-column sequence of the point positions, and the elements of the two-dimensional matrices are the combination of the abscissa X and the ordinate Y of the corresponding point positions; Calculating an X total length variance component, a Y total length variance component, an X direction vertical deviation variance component, a Y direction vertical deviation variance component and an XY merging vertical deviation according to the measurement matrix and the design matrix; Calculating an X total length variance component, a Y total length variance component, an X direction vertical deviation variance component, a Y direction vertical deviation variance component and an XY merging vertical deviation according to the measurement matrix and the design matrix, wherein the method comprises the following steps: Calculating a deviation matrix of the abscissa based on the measurement matrix and the design matrix And an ordinate deviation matrix ; Deviation matrix according to abscissa Calculating row deviation mean matrix of abscissa Column deviation mean matrix of sum abscissa Deviation matrix according to ordinate Calculating a row bias mean matrix of ordinate Column deviation mean matrix of ordinate ; Line deviation mean matrix according to abscissa Calculating row deviation mean of abscissa Column deviation mean matrix according to abscissa Calculating column deviation mean of abscissa Line deviation mean matrix according to ordinate Calculating row bias mean of ordinate Column deviation mean matrix according to ordinate Calculating column deviation mean of ordinate ; Calculating a row design mean matrix of an abscissa according to the design matrix And column design mean matrix of ordinate ; According to the abscissa of the design matrix Line design mean matrix of abscissa Column deviation mean matrix of abscissa And column deviation mean of abscissa Calculating to obtain an X total length variance component Scale X; According to the ordinate of the design matrix Column design mean matrix of ordinate Line deviation mean matrix of ordinate And the row deviation mean of the ordinate Calculating to obtain a Y total length variance component Scale Y; According to the ordinate of the design matrix Column design mean matrix of ordinate Line deviation mean matrix of abscissa And row deviation mean of abscissa Calculating to obtain an X-direction vertical deviation variance component Ortho X; according to the abscissa of the design matrix Line design mean matrix of abscissa Column deviation mean matrix of ordinate And column deviation mean of ordinate Calculating a Y-direction vertical deviation variance component Ortho Y; and calculating to obtain an XY combined vertical deviation Ortho according to the X-direction vertical deviation variance component Ortho X and the Y-direction vertical deviation variance component Ortho Y.
2. The mask precision measurement method according to claim 1, wherein the X total length variance component Scale X and the Y total length variance component Scale Y are: Scale X= ; Scale Y= ; Wherein, the For the purpose of designing the abscissa of the matrix, A mean matrix is designed for the rows of the abscissa, Is a column deviation average matrix of the abscissa, Is the column deviation average value of the abscissa, In order to measure the number of columns of the matrix, In order to design the ordinate of the matrix, A mean matrix is designed for the columns on the ordinate, Is a row deviation average matrix with an ordinate, Is the mean value of the row deviation on the ordinate, For measuring the number of rows of the matrix.
3. The reticle precision measurement method according to claim 1, wherein the X-direction vertical deviation variance component orthox and the Y-direction vertical deviation variance component orthoy are: Ortho X= ; Ortho Y= ; Wherein, the In order to design the ordinate of the matrix, A mean matrix is designed for the columns on the ordinate, Is a row deviation average matrix of the abscissa, Is the row deviation average value of the abscissa, In order to measure the number of rows of the matrix, For the purpose of designing the abscissa of the matrix, A mean matrix is designed for the rows of the abscissa, Is a column deviation average matrix with an ordinate, Is the column deviation average value of the ordinate, For measuring the number of columns of the matrix.
4. The method for measuring mask precision according to claim 1, wherein the XY merging vertical deviation Ortho is: Ortho =Ortho X + Ortho Y; where orthox is the X-direction vertical deviation variance component and orthoy is the Y-direction vertical deviation variance component.
5. A mask precision measurement system is characterized by comprising: the point position selection module is used for selecting a plurality of point positions from the region to be measured and acquiring measurement data and design data of each point position; The measuring matrix/design matrix construction module is used for constructing a measuring matrix according to the measuring data and constructing a design matrix according to the design data, wherein the measuring matrix and the design matrix are two-dimensional matrices, the row-column sequence of the two-dimensional matrices corresponds to the row-column sequence of the point positions, and the elements of the two-dimensional matrices are the combination of the abscissa X and the ordinate Y of the corresponding point positions; The precision measurement module is used for calculating an X total length variance component, a Y total length variance component, an X-direction vertical deviation variance component, a Y-direction vertical deviation variance component and an XY merging vertical deviation according to the measurement matrix and the design matrix; The precision measurement module comprises: A deviation matrix calculation module for calculating a deviation matrix of the abscissa based on the measurement matrix and the design matrix And an ordinate deviation matrix ; The deviation average value matrix calculation module is used for calculating a deviation matrix according to the abscissa Calculating row deviation mean matrix of abscissa Column deviation mean matrix of sum abscissa Deviation matrix according to ordinate Calculating a row bias mean matrix of ordinate Column deviation mean matrix of ordinate ; The deviation average value calculation module is used for calculating a row deviation average value matrix according to the abscissa Calculating row deviation mean of abscissa Column deviation mean matrix according to abscissa Calculating column deviation mean of abscissa Line deviation mean matrix according to ordinate Calculating row bias mean of ordinate Column deviation mean matrix according to ordinate Calculating column deviation mean of ordinate ; A design mean matrix calculation module for calculating a row design mean matrix of abscissa according to the design matrix And column design mean matrix of ordinate ; An X total length variance component calculation module for calculating the X total length variance component according to the abscissa of the design matrix Line design mean matrix of abscissa Column deviation mean matrix of abscissa And column deviation mean of abscissa Calculating to obtain an X total length variance component Scale X; a Y total length variance component calculation module for calculating the vertical coordinate of the design matrix Column design mean matrix of ordinate Line deviation mean matrix of ordinate And the row deviation mean of the ordinate Calculating to obtain a Y total length variance component Scale Y; An X-direction vertical deviation variance component calculation module for calculating the vertical coordinate of the design matrix Column design mean matrix of ordinate Line deviation mean matrix of abscissa And row deviation mean of abscissa Calculating to obtain an X-direction vertical deviation variance component Ortho X; A Y-direction vertical deviation variance component calculation module for calculating the horizontal coordinate of the design matrix Line design mean matrix of abscissa Column deviation mean matrix of ordinate And column deviation mean of ordinate Calculating a Y-direction vertical deviation variance component Ortho Y; And the XY merging vertical deviation calculating module is used for calculating the XY merging vertical deviation Ortho according to the X-direction vertical deviation variance component Ortho X and the Y-direction vertical deviation variance component Ortho Y.
6. The reticle precision measurement system of claim 5, wherein the X total length variance component Scale X and the Y total length variance component Scale Y in the X total length variance component calculation module are: Scale X= ; Scale Y= ; Wherein, the For the purpose of designing the abscissa of the matrix, A mean matrix is designed for the rows of the abscissa, Is a column deviation average matrix of the abscissa, Is the column deviation average value of the abscissa, In order to measure the number of columns of the matrix, In order to design the ordinate of the matrix, A mean matrix is designed for the columns on the ordinate, Is a row deviation average matrix with an ordinate, Is the mean value of the row deviation on the ordinate, For measuring the number of rows of the matrix.
7. The reticle precision measurement system of claim 5, wherein the X-direction vertical deviation variance component orthox in the X-direction vertical deviation variance component calculation module and the Y-direction vertical deviation variance component orthoy in the Y-direction vertical deviation variance component calculation module are: Ortho X= ; Ortho Y= ; Wherein, the In order to design the ordinate of the matrix, A mean matrix is designed for the columns on the ordinate, Is a row deviation average matrix of the abscissa, Is the row deviation average value of the abscissa, In order to measure the number of rows of the matrix, For the purpose of designing the abscissa of the matrix, A mean matrix is designed for the rows of the abscissa, Is a column deviation average matrix with an ordinate, Is the column deviation average value of the ordinate, For measuring the number of columns of the matrix.
8. The reticle precision measurement system of claim 5, wherein the XY merge vertical deviation Ortho in the XY merge vertical deviation calculation module is: Ortho =Ortho X + Ortho Y; where orthox is the X-direction vertical deviation variance component and orthoy is the Y-direction vertical deviation variance component.

Description

Mask precision measurement method and system Technical Field The invention relates to the technical field of semiconductors, in particular to a mask precision measurement method and a mask precision measurement system. Background With the development of display technology and semiconductor technology, the product precision requirement of mask industry manufacture is higher and higher. The existing measuring mask precision is to use a four-point measuring method, namely four points on the mask are selected to form a rectangle, and the deviation values of the four points on an abscissa X and an ordinate Y are calculated by comparing the deviation of the total length of the four points and the total length of the design, so as to obtain precision values such as total length deviation, verticality and the like. However, because the selected points are fewer, the influence of abnormal points is easy to generate measurement fluctuation, and the effective measurement area is smaller and limited to four points. Based on the above, the application provides a mask precision measurement method and a mask precision measurement system, which solve the problems. Disclosure of Invention The application aims to provide a mask precision measurement method and a mask precision measurement system, which solve the problems that the existing mask precision measurement is easily influenced by abnormal point positions and has small measurement area, and can reduce the influence of the abnormal point positions on measurement results and expand the measurement range by constructing a measurement matrix and combining a calculation formula to calculate various precision values. The application firstly provides a mask precision measurement method, which comprises the following steps: selecting a plurality of point positions from the region to be measured, and acquiring measurement data and design data of each point position; Constructing a measurement matrix according to the measurement data, and constructing a design matrix according to the design data, wherein the measurement matrix and the design matrix are two-dimensional matrices, the row-column sequence of the two-dimensional matrices corresponds to the row-column sequence of the point positions, and the elements of the two-dimensional matrices are the combination of the abscissa X and the ordinate Y of the corresponding point positions; Calculating an X total length variance component, a Y total length variance component, an X direction vertical deviation variance component, a Y direction vertical deviation variance component and an XY merging vertical deviation according to the measurement matrix and the design matrix; Wherein the X total length variance component is calculated by the sum of the deviations of the abscissas and the sum of the squares of the deviations of the ordinates and the abscissas, the Y total length variance component is calculated by the sum of the sums of the deviations of the ordinates and the abscissas and the sum of the squares of the deviations of the ordinates and the abscissas, the X vertical deviation variance component is calculated by the sum of the sums of the deviations of the ordinates and the abscissas, and calculating a Y-direction vertical deviation variance component through the sum of the deviation of the horizontal coordinate and the vertical coordinate and the square sum of the deviation of the horizontal coordinate and the vertical coordinate, and obtaining the XY merging vertical deviation through the X-direction vertical deviation variance component and the Y-direction vertical deviation variance component. By adopting the technical scheme, a measuring matrix and a design matrix are constructed by selecting a plurality of points from a region to be measured, the calculation of an X total length variance component, a Y total length variance component, an X direction vertical deviation variance component, a Y direction vertical deviation variance component and an XY merging vertical deviation is realized through the calculation of the measuring matrix and the design matrix, the data of the plurality of points are combined to construct the measuring matrix and the design matrix, the total length variance component of each two points and the variance component of each point on an abscissa X and an ordinate Y are calculated through the matrix, the vertical deviation is not easily influenced by an abnormal value, the measuring precision is greatly improved, and the result error is only influenced by the precision of a measuring instrument. In one possible implementation, calculating an X total length variance component, a Y total length variance component, an X-direction vertical deviation variance component, a Y-direction vertical deviation variance component, and an XY-merged vertical deviation from the measurement matrix and the design matrix includes: Calculating a deviation matrix delta x ij of an abscissa and a deviation matrix del