Method for determining arrangement of multiple shot areas on a substrate, exposure apparatus, article manufacturing method, program, and information processing apparatus

The method uses Bayesian estimation to update regression models for high-precision alignment of substrate regions, addressing the trade-off between accuracy and productivity in exposure devices by correcting high-order deformations with fewer measurement points.

JP7580297B2Active Publication Date: 2024-11-11CANON KK
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2021026641
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-02-22
Publication Date
2024-11-11
Estimated Expiration
2041-02-22

AI Technical Summary

Technical Problem

Existing alignment methods in exposure devices face a trade-off between alignment accuracy and productivity due to overfitting with high-order polynomial models and increased measurement time with more measurement points, necessitating a high-precision arrangement technique with a small number of measurement points.

Method used

A method involving Bayesian estimation to update a regression model using posterior distributions from a prior distribution based on limited initial measurement data, allowing for high-precision arrangement of substrate regions by correcting high-order deformations with a reduced number of measurement points.

Benefits of technology

Enhances alignment accuracy while maintaining productivity by reducing overfitting and measurement time, improving chip yield and reducing defects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007580297000008
    Figure 0007580297000008
  • Figure 0007580297000009
    Figure 0007580297000009
  • Figure 0007580297000010
    Figure 0007580297000010
Patent Text Reader

Abstract

To provide a technology advantageous for accurately obtaining an arrangement of a shot region on a substrate.SOLUTION: A method for determining an arrangement of a plurality of regions on a substrate characterized by comprising: a first step of obtaining a first posterior distribution showing a probability distribution of parameters from first positional measurement data on a mark assigned to a sample region of a first number of a plurality of regions on a first substrate, by using an anterior distribution showing a probability distribution of parameters of a regression model for estimating the arrangement; a second step of obtaining a second posterior distribution showing a probability distribution of the parameters from second positional measurement data on a mark assigned to a sample region of a second number smaller than the first number of a plurality of regions on a second substrate processed after the first substrate, by using the first posterior distribution as an anterior distribution showing the probability distribution of parameters; and a third step of determining the parameters based on the second posterior distribution and renovating the regression model, and obtaining an arrangement of a plurality of regions on the second substrate from the second positional measurement data by using the renovated regression model.SELECTED DRAWING: Figure 5
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a method for determining an arrangement of a plurality of shot areas on a substrate, an exposure apparatus, a method for manufacturing an article, a program, and an information processing apparatus. [Background technology]

[0002] The exposure device transfers 10 or more layers of patterns (circuit patterns) onto the substrate in an overlapping manner, but if the overlapping accuracy of the patterns between the layers is not good, problems may occur with the circuit characteristics. In such cases, the chip will not meet the desired characteristics, resulting in a defective chip and a reduced yield. Therefore, it is necessary to precisely align each of the multiple areas to be exposed on the substrate with the original pattern.

[0003] In an exposure apparatus, alignment marks arranged in each region on a substrate are detected, and each region on the substrate is aligned with the pattern on the original based on position information of the alignment marks and position information of the pattern on the original. Ideally, the most accurate alignment is possible by detecting alignment marks in all regions on the substrate, but this is not realistic from the viewpoint of productivity. Therefore, the global alignment method is currently the mainstream method for aligning a substrate and an original (see Patent Documents 1 and 2).

[0004] In the global alignment method, it is assumed that the relative position of each area on the substrate can be expressed by a function model of the position coordinates of the area, and the positions of alignment marks arranged only in a plurality of sample areas (4 to 16) on the substrate are measured. Next, parameters of the function model are estimated using regression analysis-like statistical calculation processing from the assumed function model and the measurement results of the alignment mark positions. Then, alignment is performed by calculating the position coordinates of each area in the stage coordinate system (arrangement of areas on the substrate) using the parameters and the function model. In the global alignment method, a polynomial model with the stage coordinates as a variable is generally used, and scaling, rotation, uniform offset, etc., which are first-order polynomials of the stage coordinates, are mainly used (see Patent Document 3). In addition, a technology using a regression model that also considers high-order components of the array of areas on the substrate as parameters has also been proposed (see Patent Document 4). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Unexamined Patent Publication No. 61-44429 [Patent Document 2] Japanese Patent Application Publication No. 62-84516 [Patent Document 3] Japanese Patent Application Publication No. 6-349705 [Patent Document 4] Patent No. 3230271 Summary of the Invention [Problem to be solved by the invention]

[0006] As devices become finer and more highly integrated, there is a demand for improved alignment accuracy, so it is necessary to increase the degree of freedom of the polynomial order of the function model by using higher-order components. However, if there are fewer measurement points for measuring the positions of alignment marks in a substrate compared to the degrees of freedom of the function model, overfitting occurs, and the correction error in the unmeasured area increases. On the other hand, if the number of measurement points for measuring the positions of alignment marks is increased to suppress overfitting, the measurement time increases and productivity decreases. Since there is a trade-off between these two, there is a demand for a technology that can predict the arrangement of areas on a substrate that contain higher-order components with high accuracy using a function model with fewer measurement points and a high degree of freedom.

[0007] The present invention has been made in consideration of the problems with the conventional techniques, and has an exemplary object to provide a technique that is advantageous for determining the arrangement of regions on a substrate with high accuracy. [Means for solving the problem]

[0008] In order to achieve the above object, one aspect of the present invention is a method for determining an arrangement of a plurality of regions on a substrate, the method including: indicating a probability distribution of parameters of a regression model for estimating the arrangement from first position measurement data of a mark assigned to a first number of sample regions among the plurality of regions on a first substrate; No. 1 A first step of obtaining a first posterior distribution indicating a probability distribution of the parameter by using a prior distribution, and a second step of obtaining the first posterior distribution indicating the probability distribution of the parameter from second position measurement data of marks assigned to a second number of sample areas smaller than the first number among a plurality of areas on a second substrate processed after the first substrate. Second The method includes a second step of obtaining a second posterior distribution indicating a probability distribution of the parameter by using the second posterior distribution as a prior distribution, and a third step of determining the parameter based on the second posterior distribution to update the regression model, and determining an arrangement of multiple regions on the second substrate from the second position measurement data using the updated regression model.

[0009] Further objects or other aspects of the present invention will become apparent from the following description of the embodiments with reference to the accompanying drawings. Effect of the Invention

[0010] According to the present invention, for example, it is possible to provide a technique that is advantageous for determining the arrangement of regions on a substrate with high accuracy. [Brief description of the drawings]

[0011] [Figure 1] 1 is a schematic diagram showing a configuration of an exposure apparatus according to one aspect of the present invention. [Diagram 2] 2 is a schematic diagram showing the configuration of an alignment optical system of the exposure apparatus shown in FIG. [Diagram 3] 2 is a flowchart for explaining an exposure process in the exposure apparatus shown in FIG. [Figure 4] FIG. 2 is a diagram showing an arrangement of shot areas on a substrate. [Diagram 5] 11A and 11B are diagrams for explaining optimization of parameters of a model for estimating a shot arrangement; [Figure 6] 11A and 11B are diagrams for explaining optimization of parameters of a model for estimating a shot arrangement; [Figure 7] FIG. 6 is a diagram showing a result of applying the sequence shown in FIG. 5 to an actual process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, the embodiments will be described in detail with reference to the attached drawings. Note that the following embodiments do not limit the invention according to the claims. Although the embodiments describe a number of features, not all of these features are essential to the invention, and the features may be combined in any manner. Furthermore, in the attached drawings, the same reference numbers are used for the same or similar configurations, and duplicated descriptions are omitted.

[0013] 1 is a schematic diagram showing the configuration of an exposure apparatus 1 according to one aspect of the present invention. The exposure apparatus 1 is a lithography apparatus used in the manufacturing process of devices such as semiconductor elements. In this embodiment, the exposure apparatus 1 projects a pattern of an original 2 (reticle or mask) onto a substrate 4 via a projection optical system 3 to expose the substrate 4 to light.

[0014] 1, the exposure apparatus 1 has a projection optical system 3 that projects (reduced projection) a pattern formed on an original 2, and a chuck 5 that holds a substrate 4 on which a base pattern and alignment marks have been formed in a previous process. The exposure apparatus 1 also has a substrate stage 6 that holds the chuck 5 and positions the substrate 4 at a predetermined position, an alignment optical system 7 that measures the position of an alignment mark provided on the substrate 4, a control unit CN, and a storage unit SU.

[0015] The control unit CN is composed of, for example, a computer (information processing device) including a CPU, memory, etc., and performs overall control of each unit of the exposure apparatus 1 in accordance with a program stored in the storage unit SU, etc. In this embodiment, the control unit CN functions as a processing unit that determines the arrangement (shot arrangement, area arrangement) of multiple shot areas (multiple areas on the substrate) on the substrate, in addition to controlling the exposure process that exposes the substrate 4 via the original 2.

[0016] The storage unit SU stores programs and various information (data) required to control each part of the exposure apparatus 1 and perform an exposure process for exposing the substrate 4. The storage unit SU also stores programs and various information (data) required for the control unit CN to determine a shot arrangement.

[0017] 2 is a schematic diagram showing the configuration of the alignment optical system 7. The alignment optical system 7 has a function of optically detecting marks assigned to each shot area of ​​the substrate 4 to obtain position measurement data, and in this embodiment, includes a light source 8, a beam splitter 9, lenses 10 and 13, and a sensor 14.

[0018] Light from the light source 8 is reflected by the beam splitter 9 and passes through a lens 10 to illuminate an alignment mark 11 or 12 provided on the substrate 4. The light diffracted by the alignment mark 11 or 12 passes through the lens 10, the beam splitter 9, and a lens 13 and is received by a sensor 14.

[0019] With reference to FIG. 3, the exposure process in the exposure apparatus 1 will be described. Here, an overview of the process from aligning the substrate 4 to exposing it will be described. In S101, the substrate 4 is loaded into the exposure apparatus 1. In S102, pre-alignment is performed. Specifically, alignment marks 11 for pre-alignment provided on the substrate 4 are detected by the alignment optical system 7, and the position of the substrate 4 is roughly obtained. At this time, detection of the alignment marks 11 is performed for multiple shot areas of the substrate 4, and the overall shift and first-order linear components (magnification and rotation) of the substrate 4 are obtained.

[0020] In S103, fine alignment is performed. Specifically, first, based on the result of the pre-alignment, the substrate stage 6 is driven to a position where the alignment mark 12 for fine alignment can be detected by the alignment optical system 7. Then, the alignment mark 12 provided in each of the multiple shot areas of the substrate 4 is detected by the alignment optical system 7, and the overall shift and first-order linear components (magnification and rotation) of the substrate 4 are precisely obtained. At this time, by obtaining the positions of many shot areas, it is also possible to precisely obtain high-order deformation components of the substrate 4. This makes it possible to obtain the precise position of each shot area of ​​the substrate 4, i.e., the shot arrangement.

[0021] In S104, the substrate 4 is exposed to light. Specifically, after fine alignment is performed, the pattern of the original 2 is transferred to each shot area of ​​the substrate 4 via the projection optical system 3. In S105, the substrate 4 is unloaded from the exposure apparatus 1.

[0022] In this embodiment, if distortion occurs in the substrate 4, high-order deformation components are corrected in the fine alignment of S103. Here, a third-order polynomial model is taken as an example of a regression model for estimating the shot arrangement, but the present invention is not limited to this. For example, any order model may be used as the regression model, or a model other than a polynomial (a trigonometric function model or a logarithmic model) may be used.

[0023] When the deformation of the substrate 4 is expressed by a third-order polynomial model, the positional shift (ShiftX, ShiftY) of each shot area is expressed by the following formula (1). Note that the positional shift of each shot area can also be considered as a correction value for correcting such positional shift.

[0024]

number

[0025] In formula (1), x and y indicate the position (explanatory variables) of the shot area of ​​the substrate 4. From the actual position measurement data of each shot area of ​​the substrate 4, k1 to k 20 Then, based on the equation (1) in which the coefficients have been determined, the positional deviation of each shot area is calculated.

[0026] In order to obtain position measurement data, the alignment optical system 7 detects alignment marks assigned to some shot areas, so-called sample shot areas (sample areas), of the multiple shot areas, as shown in FIG. 4, for example. In FIG. 4, the number of sample shot areas is set to 14. In order to correct high-order deformation components of the substrate 4, many shot areas need to be set as sample shot areas. However, since an increase in the number of sample shot areas is in a trade-off relationship with the measurement time (alignment time), in practice the number of sample shot areas is determined taking into consideration the productivity of the device.

[0027] Hereinafter, with reference to FIG. 5, the optimization of parameters of a model for estimating a shot arrangement (updating of the model) in this embodiment will be described.

[0028] First, for the first substrate 4A (first substrate) in the lot, the number of sample shot areas is set to a number (first number) that is sufficient for the model's degree of freedom and does not overlearn the model's degree of freedom. Then, the alignment marks assigned to each sample shot area of ​​the substrate 4A are detected by the alignment optical system 7 to obtain first position measurement data.

[0029] Next, the data likelihood (first likelihood data) P(D|θ) is calculated from the first position measurement data (D), the model, and the parameter θ of the model. The parameter θ is the coefficients k1 to k in Equation (1). 20 This means:

[0030] Next, the prior distribution P(θ) of the parameter θ is specified. When there is no prior information on the data, a non-informative prior distribution is specified as the prior distribution P(θ). Representative non-informative prior distributions include a uniform distribution, a normal distribution with a large variance, and a Jeffreys prior distribution.

[0031] Next, the posterior distribution (first posterior distribution) p(θ|D) of the parameter θ is calculated from Bayesian estimation using the prior distribution P(θ) and the data likelihood P(D|θ). Note that Bayesian theorem is expressed by the following formula (2).

[0032]

number

[0033] In formula (2), θ is a model parameter, D is position measurement data, and p(θ|D) is the posterior distribution of parameter θ when new position measurement data is obtained. In addition, p(D|θ) is a data likelihood (likelihood function) that indicates the probability that data is generated from parameter θ, p(θ) is a prior distribution of parameter θ, and p(D) is the marginal likelihood of position measurement data D.

[0034] In addition, when the marginal likelihood p(D) cannot be analytically calculated in the calculation of the posterior distribution p(θ|D), approximate inference may be used. Approximate inference includes, for example, the Markov chain Monte Carlo method (MCMC: Markov chain Monte Carlo), which is a sampling method, and variational inference, which is analytically calculated using an approximate probability distribution. The posterior distribution p(θ|D) is a probability distribution (probability distribution of the parameters of the model) determined by the model set by the designer, and the characteristic of Bayesian estimation is that the variance of the distribution is interpreted as the "confidence" of the value of each parameter. In this embodiment, the model is a polynomial model with the board coordinates (board position) as variables, and there are 10 parameters for each of ShiftX and ShiftY.

[0035] Next, for the second or subsequent substrate 4B in the lot (second substrate processed after the first substrate), the number of sample shot areas is set to a number (second number) smaller than the number of sample shots set for substrate 4A. Then, the alignment marks assigned to each sample shot area of ​​substrate 4B are detected by alignment optical system 7 to obtain first position measurement data.

[0036] Here, for example, in the third-order polynomial model, if the number of sample shot areas of substrate 4A is set to 16, the number of sample shot areas of substrate 4B is set to 4 to 8. If the number of sample shot areas for the third-order polynomial model is 4 to 8, the number of measurement points relative to the degree of freedom of the model is small, so that the possibility of overfitting occurring becomes extremely high. In order to suppress such overfitting, in this embodiment, the following process is performed.

[0037] A data likelihood (second likelihood data) P(D|θ) is calculated from the second position measurement data (D), the model, and the model parameter θ. Then, the posterior distribution P(θ|D) obtained for the substrate 4A is used as the prior distribution P(θ), and the data likelihood P(D|θ) calculated from the second position measurement data is used to calculate a posterior distribution (second posterior distribution) p(θ|D) from Bayesian estimation using Bayes' theorem. As shown in FIG. 6, the posterior distribution p(θ|D) is a distribution (posterior distribution) in which the distribution (prior distribution) obtained from the first position measurement data (multiple point data) and the distribution (data likelihood) obtained from the second position measurement data (few point data) are mixed. When the shapes of the distortions of the substrates in a lot are similar, the distribution obtained in this way is expected to be closer to the actual distortion amount of the substrate than the distribution obtained only from the second position measurement data. The actual positional shift (ShiftX, ShiftY) of each shot area is calculated by statistical estimation from a specified model based on a statistical value such as the mode, average, or median of the posterior distribution of the parameter θ.

[0038] The parameters of the model are determined based on the posterior distribution calculated in this manner, the model is updated, and the updated model is used to determine the positional deviation (shot arrangement) of each shot area of ​​the substrate 4B from the second position measurement data, thereby making it possible to correct higher-order deformation components.

[0039] According to this embodiment, for the first substrate, it is necessary to set a sufficient number of sample shot areas with respect to the degree of freedom of the model, but for the second and subsequent substrates, it is possible to reduce the number of sample shot areas while suppressing overlearning. Also, when the variation in distortion of substrates within a lot is large, the posterior distribution obtained by setting a sufficient number of sample shot areas with respect to the degree of freedom of the model for multiple substrates may be used as the prior distribution for the subsequent substrates.

[0040] If the model is a polynomial linear regression model and the prior distribution is a multivariate normal distribution, the posterior distribution can be analytically solved. For example, assume that ShiftX, which is a deformation component of the substrate, is modeled by a polynomial as shown in formula (1). A design matrix G (the number of measurement data x, a matrix of the number of parameters) used in multivariate regression analysis is created from the polynomial model and the measured substrate positions (x, y), and the position measurement data is represented as y (a vector of the number of measurement data), and the error variance of the position measurement data is represented as σ. In this case, the data likelihood P(D|θ) for the set of position measurement data D and the model parameter θ is expressed by the following formula (3) based on the theory of statistical regression analysis.

[0041]

number

[0042] In addition, the prior distribution P(θ) is assumed to follow a multivariate normal distribution, as shown in the following equation (4).

[0043]

number

[0044] In equation (4), μ0 represents the mean vector, and Σ0 represents the variance-covariance matrix. Therefore, the posterior distribution p(θ|D) is expressed by the following equation (5).

[0045]

number

[0046] Rearranging equation (5) leads to the conclusion that it follows a multivariate normal distribution, as shown in equation (6) below.

[0047]

number

[0048] Equation (6) is the parameter of the multivariate normal distribution derived from equation (5), μ is the mean vector of the posterior distribution (mean vector μ), and Σ is the variance-covariance matrix of the posterior distribution (variance-covariance matrix Σ).

[0049] It is possible to calculate the optimal parameters of the model by sequentially calculating formulas (3), (4), (5), and (6) according to the sequence shown in Fig. 5. However, for the first substrate, it is necessary to set, as the non-informative prior distribution, μ0 as a zero vector, and a symmetric matrix with large values ​​for the diagonal components of the variance-covariance matrix and zero for the other components.

[0050] With reference to FIG. 7(a), FIG. 7(b), and FIG. 7(c), the result of applying the sequence shown in FIG. 5 to an actual process will be described. Here, 12 process substrates were used for evaluation. First, for the first substrate, the number of sample shot areas was set to 36, and the posterior distribution p(θ|D) of the parameters was calculated using a 5th-order polynomial model. For the second and subsequent substrates, the number of sample shot areas was reduced to 8, the posterior distribution calculated for the first substrate was used as the prior distribution, and the posterior distribution was calculated for each substrate from the data likelihood p(D|θ) calculated for each substrate, and alignment was performed based on the posterior distribution. Each of FIG. 7(a) and FIG. 7(b) shows the X component (ShiftX) and Y component (ShiftY) of the alignment error for each substrate. Each of FIG. 7(a) and FIG. 7(b) also shows the result of performing alignment using a 1st-order polynomial model with the number of sample shot areas set to 8 as a comparative example (conventional technology) to this embodiment. Fig. 7(c) shows the average alignment error in this embodiment and the average alignment error in the comparative example. Referring to Fig. 7(c), it can be seen that the alignment accuracy is improved in this embodiment compared to the conventional technology, which is the comparative example.

[0051] The manufacturing method of an article according to an embodiment of the present invention is suitable for manufacturing articles such as liquid crystal display elements, semiconductor elements, flat panel displays, and MEMS. The manufacturing method includes a step of exposing a substrate coated with a photosensitive agent using the above-mentioned exposure apparatus 1, and a step of developing the exposed photosensitive agent. In addition, an etching step, an ion implantation step, and the like are performed on the substrate using the developed photosensitive agent pattern as a mask, and a circuit pattern is formed on the substrate. These steps of exposure, development, etching, and the like are repeated to form a circuit pattern consisting of multiple layers on the substrate. In a post-process, dicing (processing) is performed on the substrate on which the circuit pattern is formed, and chip mounting, bonding, and inspection steps are performed. In addition, the manufacturing method may include other well-known steps (oxidation, film formation, deposition, doping, planarization, resist peeling, and the like). The manufacturing method of an article according to the present embodiment is advantageous in at least one of the performance, quality, productivity, and production cost of the article compared to the conventional method.

[0052] The present invention can also be realized by a process in which a program for implementing one or more of the functions of the above-described embodiments is supplied to a system or device via a network or a storage medium, and one or more processors in a computer of the system or device read and execute the program. The present invention can also be realized by a circuit (e.g., ASIC) for implementing one or more of the functions.

[0053] The invention is not limited to the above-described embodiments, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, the following claims are appended to apprise the public of the scope of the invention. [Explanation of symbols]

[0054] 1: exposure device 2: original 3: projection optical system 4: substrate 5: substrate chuck 6: substrate stage 7: detection optical system CN: control unit SU: memory unit

Claims

1. 1. A method for determining an alignment of a plurality of regions on a substrate, comprising the steps of: a first step of obtaining a first posterior distribution indicating a probability distribution of parameters of a regression model for estimating the arrangement from first position measurement data of marks assigned to a first number of sample areas among a plurality of areas on a first substrate, using a first prior distribution indicating a probability distribution of the parameters; a second step of obtaining a second posterior distribution indicating a probability distribution of the parameter from second position measurement data of marks assigned to a second number of sample regions, the second number being smaller than the first number, among a plurality of regions on a second substrate processed after the first substrate, by using the first posterior distribution as a second prior distribution indicating a probability distribution of the parameter; a third step of determining the parameters based on the second posterior distribution to update the regression model, and determining an arrangement of a plurality of regions on the second substrate from the second position measurement data using the updated regression model; The method according to claim 1, further comprising:

2. The method of claim 1 , wherein the regression model comprises a polynomial model that includes the position of the substrate as an explanatory variable.

3. 3. The method according to claim 1, wherein in the first step, the first posterior distribution is obtained from Bayes' theorem using a non-informative prior distribution that is the first prior distribution, the first position measurement data, and first likelihood data obtained from the regression model.

4. 4. The method of claim 3, wherein the non-informative prior distribution comprises a uniform distribution, a normal distribution, or a Jefferies prior distribution.

5. The method according to any one of claims 1 to 4, characterized in that in the second step, the second posterior distribution is obtained from Bayes' theorem using the first posterior distribution, the second position measurement data, and second likelihood data obtained from the regression model.

6. 6. The method of claim 1, wherein the first prior distribution and the second prior distribution comprise a multivariate normal distribution.

7. The method according to any one of claims 1 to 5, wherein the first prior distribution and the second prior distribution comprise a plurality of multivariate normal distributions having different variance-covariance matrices.

8. 2. The method of claim 1, wherein the second step comprises obtaining the second posterior distribution from a Markov chain Monte Carlo method.

9. The prior distribution showing the probability distribution of the parameters is expressed as the mean vector μ 0 , the variance-covariance matrix is ​​Σ 0 where y is the second position measurement data, σ is the error variance of the second position measurement data, and G is the design matrix, the mean vector μ and variance-covariance matrix Σ of the second posterior distribution are given by 2. The method of claim 1, wherein the value is determined from

10. 10. The method according to claim 1, wherein in the third step, the parameters are determined and the regression model is updated by estimating the parameters based on a mode, a mean or a median of the second posterior distribution.

11. 1. An exposure apparatus that exposes a plurality of areas on a substrate through an original, comprising: A processing unit that determines an arrangement of the plurality of regions on the substrate; a stage that positions the substrate based on the arrangement determined by the processing unit, The processing unit includes: determining a first posterior distribution indicating a probability distribution of parameters of a regression model for estimating the arrangement from first position measurement data of marks assigned to a first number of sample regions among a plurality of regions on the first substrate, using a first prior distribution indicating a probability distribution of the parameters; obtaining a second posterior distribution indicating a probability distribution of the parameter from second position measurement data of marks assigned to a second number of sample regions, the second number being smaller than the first number, among a plurality of regions on a second substrate processed after the first substrate, by using the first posterior distribution as a second prior distribution indicating a probability distribution of the parameter; an exposure apparatus comprising: determining the parameters based on the second posterior distribution to update the regression model; and determining an arrangement of a plurality of regions on the second substrate from the second position measurement data using the updated regression model.

12. The method of claim 1, further comprising: determining an arrangement of a plurality of regions on a substrate; positioning the substrate based on the alignment; exposing the aligned substrate; developing the exposed substrate; producing an article from the developed substrate; A method for producing an article, comprising the steps of:

13. A program for causing a computer to execute a method for determining an arrangement of a plurality of regions on a substrate, comprising: The computer includes: a first step of obtaining a first posterior distribution indicating a probability distribution of parameters of a regression model for estimating the arrangement from first position measurement data of marks assigned to a first number of sample areas among a plurality of areas on a first substrate, using a first prior distribution indicating a probability distribution of the parameters; a second step of obtaining a second posterior distribution indicating a probability distribution of the parameter from second position measurement data of marks assigned to a second number of sample regions, the second number being smaller than the first number, among a plurality of regions on a second substrate processed after the first substrate, by using the first posterior distribution as a second prior distribution indicating a probability distribution of the parameter; a third step of determining the parameters based on the second posterior distribution to update the regression model, and determining an arrangement of a plurality of regions on the second substrate from the second position measurement data using the updated regression model; A program characterized by executing the above.

14. An information processing apparatus for executing the method according to any one of claims 1 to 10.

Citation Information

Patent Citations

  • Alignment method

    JP1986044429A

  • Alignment of position

    JP1987084516A

  • Method and device for alignment

    JP1994252027A

  • Alignment method

    JP1994349705A

  • Position detection method, exposure method, exposer, device manufacturing method, program, and information recording medium

    JP2004200632A