Method, device and system for correcting measurement errors caused by asymmetry of overlay marks

Abstract

The invention provides a method, device and system for correcting measurement errors caused by asymmetry of overlay marks, and the method comprises the steps: obtaining to-be-measured structure information through employing optical imaging equipment based on an I BO technology, and carrying out the calculation through employing a preset algorithm, and obtaining product analysis items and an asymmetry value of each product analysis item; obtaining an initial asymmetric evaluation value according to the asymmetric value of each product analysis item, and obtaining a target asymmetric evaluation value meeting a preset error condition and iterated hardware parameters and software parameters by adjusting different combinations of the hardware parameters and the software parameters; the hardware parameters acquire the relationship between the product analysis item and the software parameters through a large amount of data in the process of configuring the product analysis item, and the software parameters are used for carrying out enhanced image processing on to-be-measured structure information so as to be matched with the hardware parameters to obtain the product analysis item and a target asymmetric evaluation value, and then correction is carried out; more accurate measurement is provided, the repair requirement is remarkably reduced, and the efficiency of a production line and the product quality are improved.

Description

Technical Field

[0001] The present application relates to the field of semiconductor manufacturing technology, and in particular to a method, device and system for correcting measurement errors caused by asymmetric overlay marks. Background Art

[0002] In the semiconductor manufacturing and precision machining process, optical measurement systems are widely used to detect and evaluate various process parameters. However, in practical applications, the asymmetry of the target often brings great challenges to measurement. This asymmetry is particularly common in thick glue processes, especially in thick glue processes in the fields of automotive chips, new energy power chips, etc. The targets in the thick glue process usually have irregular geometric shapes or surface characteristics, which makes it difficult for traditional measurement methods to obtain accurate measurement results. The asymmetric shape makes it difficult for the optical system to capture the complete information of the target during imaging, thus affecting the accuracy of the measurement.

[0003] The current traditional IBO technology is limited by resolution and noise when facing thick glue processes and target asymmetry, and it is difficult to provide accurate measurement results. Although DBO has high precision, its equipment is expensive, the technology is complex, and its application is limited in thick glue and large line width processes.

[0004] Therefore, a new solution for correcting the measurement error caused by the asymmetry of the overlay marks is needed. Summary of the invention

[0005] In view of this, the embodiments of this specification provide a method, device and system for correcting measurement errors caused by asymmetric overlay marks.

[0006] The embodiments of this specification provide the following technical solutions:

[0007] The embodiment of this specification provides a method for correcting measurement errors caused by asymmetric overlay marks, including:

[0008] An optical imaging device is used to obtain information of the structure to be measured, and a preset algorithm is used to calculate the product analysis items and the asymmetry value of each product analysis item; wherein the asymmetry value of the product analysis item is used to reflect the global and local asymmetry degree in the target overlay mark corresponding to the structure to be measured;

[0009] An initial asymmetry evaluation value is obtained according to the asymmetry value of each product analysis item, and a target asymmetry evaluation value is obtained by adjusting different combinations of hardware parameters and software parameters when the initial asymmetry evaluation value meets a preset error condition, and the iterative hardware parameters and software parameters are obtained; wherein, the hardware parameters obtain the relationship between the product analysis items and the software parameters through a large amount of data during the configuration of the product analysis items, and the software parameters are used to perform enhanced image processing on the structural information to be measured, so as to cooperate with the hardware parameters to obtain the product analysis items and the target asymmetry evaluation value;

[0010] The hardware parameters and software parameters after the iteration are adjusted respectively to correct the asymmetric evaluation value in the process of measuring the overlay mark.

[0011] The embodiment of this specification also provides a measurement error correction device caused by asymmetric overlay marks, including: an imaging and calculation module, an imaging stage optimization module and a calculation stage optimization module;

[0012] An imaging phase optimization module and a computing phase optimization module are respectively connected to the imaging and computing modules in communication;

[0013] An imaging and calculation module, used to obtain information of the structure to be measured by using an optical imaging device, and to calculate the product analysis items and the asymmetry value of each product analysis item by using a preset algorithm; wherein the asymmetry value of the product analysis item is used to reflect the global and local asymmetry degree in the target overlay mark corresponding to the structure to be measured;

[0014] An initial asymmetry evaluation value is obtained according to the asymmetry value of each product analysis item, and a target asymmetry evaluation value is obtained when the initial asymmetry evaluation value meets a preset error condition by adjusting different combinations of hardware parameters and software parameters, and iterative hardware parameters and software parameters are obtained, so as to adjust them respectively according to the iterative hardware parameters and software parameters;

[0015] The hardware parameters acquire the relationship between the product analysis items and the software parameters through a large amount of data during the configuration of the product analysis items. The software parameters are used to enhance the image processing of the structural information to be measured, so as to cooperate with the hardware parameters to obtain the product analysis items and the target asymmetry evaluation value;

[0016] An imaging stage optimization module is used to adjust the hardware parameters after iteration when correcting the asymmetry of the overlay mark in the structure to be measured, so as to correct the initial asymmetry evaluation value by adjusting the software parameters to meet the preset error conditions;

[0017] The calculation phase optimization module is used to adjust the software parameters after iteration when correcting the asymmetry of the overlay mark in the structure to be measured, so as to correct the initial asymmetry evaluation value by adjusting the hardware parameters to meet the preset error conditions.

[0018] Compared with the prior art, the at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects:

[0019] The embodiments of this specification are based on IBO technology. After quantifying asymmetry, the asymmetry is used as a target to optimize the measurement conditions. By coupling the measurement conditions with hardware parameters and software parameters, more accurate measurements are provided, the demand for rework is significantly reduced, and the efficiency of the production line and product quality are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 It is a flow chart of a method for correcting measurement errors caused by asymmetric overlay marks provided by the present application;

[0022] Figure 2 It is a schematic diagram of a measurement error correction caused by asymmetric overlay marks provided by the present application;

[0023] Figure 3 It is a schematic diagram of optimizing measurement conditions based on an asymmetric function provided in this application. DETAILED DESCRIPTION

[0024] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0025] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0026] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present application, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number and aspect described herein can be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this device and / or practice this method.

[0027] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. The drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0028] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the examples can be practiced without these specific details.

[0029] In the process of semiconductor manufacturing and precision machining, optical measurement systems are widely used to detect and evaluate various process parameters. However, in practical applications, the asymmetry of the target often poses a great challenge to measurement. This asymmetry is particularly common in the fields of thick glue process, automotive chips, new energy power chips, etc., especially in the lithography and etching links of semiconductor manufacturing. Asymmetric targets lead to a significant decrease in measurement accuracy, affecting product quality control and process stability. Specifically, the main problems of asymmetry include the following aspects:

[0030] Irregularity of target shape: Targets in thick glue processes usually have irregular geometric shapes or surface characteristics, which makes it difficult for traditional measurement methods to obtain accurate measurement results. The asymmetric shape makes it difficult for the optical system to capture the complete information of the target during imaging, thus affecting the accuracy of the measurement.

[0031] Measurement errors caused by asymmetry: In traditional optical measurement systems, uniform illumination conditions cannot fully adapt to the structural differences of asymmetric targets. Common measurement errors include edge blur, focus offset, and uneven light intensity distribution. These problems lead to unstable measurement results, which can seriously affect the product yield, especially in the demanding Fab production.

[0032] Repeated process rework is time-consuming and labor-intensive: Due to the measurement error of asymmetric targets, products that do not meet the standards often need to be reworked during the production process. This rework not only increases production costs, but also affects the stability and efficiency of the process, especially in the fields of automotive regulations and new energy power chips. Frequent target asymmetry caused by thick glue processes makes rework a normal occurrence, consuming a lot of time and resources, and reducing the overall efficiency of the production line.

[0033] In the Fab production line, optical measurement is mainly used to accurately detect the key parameters of the product to ensure the stability of the process and the quality of the product. In the manufacturing process of automotive chips and new energy power chips, thick glue process has become one of the commonly used technologies, but due to the thickness variation of the thick glue layer and the asymmetric characteristics of the target, the existing measurement system cannot effectively adapt to these changes, often resulting in large measurement errors.

[0034] Automotive chip manufacturing: Automotive chips require extremely high reliability and consistency. Any slight process deviation will affect the stability and safety of the chip during vehicle operation. During the production process, due to the use of thick photolithography, asymmetry problems frequently occur, resulting in inaccurate measurements and failure to detect process defects in a timely manner, which in turn affects the quality control of the entire production process.

[0035] New energy power chip manufacturing: In power chips, thick glue processes are widely used to manufacture power components with high voltage resistance and high reliability. These chips usually require higher precision to ensure their stable operation under high voltage and high current environments. However, asymmetric targets often appear in such processes, which brings great challenges to measurement, and repeated rework further exacerbates the waste of cost and time.

[0036] The current traditional IBO technology is limited by resolution and noise when facing thick glue processes and target asymmetry, and it is difficult to provide accurate measurement results. Although DBO has high precision, its equipment is expensive, the technology is complex, and its application is limited in thick glue and large line width processes.

[0037] Limitations of IBO Technology

[0038] As an overlay measurement method based on optical imaging, IBO is widely used to measure overlay marks in the chip manufacturing process. The IBO system uses optical imaging of lithography marks to determine the offset of the pattern and then evaluate the overlay accuracy. However, IBO technology has the following main disadvantages when facing thick glue process and target asymmetry problems:

[0039] Mark noise has a great impact: The imaging process of IBO is easily affected by the structural noise of the mark itself, especially in the thick glue process, the glue wall around the mark often appears asymmetric (such as asymmetric side walls of the left and right glue walls, grooves at the bottom of the glue wall, etc.). These structural noises lead to blurred optical imaging, and it is impossible to accurately determine the exact position of the overlay mark.

[0040] The resolution is not high enough: Since IBO relies on optical imaging, its resolution is limited by the optical system, especially at the microscopic scale, and it is impossible to accurately quantify the angle difference of the glue wall and the depth of the groove and other fine structures. Target asymmetry occurs frequently in thick glue processes, and these subtle asymmetric structures are difficult to accurately capture in IBO imaging, resulting in large errors in the measurement results, which ultimately affects the overlay accuracy.

[0041] Unable to accurately quantify asymmetry: IBO technology focuses on qualitative analysis based on images rather than quantitative analysis of target asymmetry. Therefore, in the face of complex glue wall morphology and microscopic asymmetry, IBO is difficult to provide accurate quantitative indicators to optimize measurement conditions. This limitation makes IBO perform poorly in thick glue processes, especially when measuring asymmetric targets.

[0042] Limitations of DBO Technology

[0043] DBO is an overlay measurement technology based on grating diffraction, which is commonly used in 28nm and more advanced process nodes. DBO uses the diffraction image generated by the grating to determine the offset of the overlay mark, with high accuracy and stability. However, DBO has great limitations in thick-resistance processes and large-linewidth processes, which are mainly manifested in the following aspects:

[0044] Complex technology and expensive equipment: DBO technology requires grating marks and relies on high-precision lithography machines to make these marks. In practical applications, the complexity and equipment cost of the DBO system are high, and it is often difficult to be widely used in Fabs that manufacture thick glue or large line width processes. Especially in thick glue applications, the high-precision advantage of DBO is difficult to play a role, because these processes do not require extremely high grating accuracy, and the optical properties of thick glue materials will further interfere with the measurement accuracy of DBO.

[0045] Difficult to apply to thick glue process: DBO technology is mainly aimed at advanced process nodes of 28nm and below, and its precision advantage can be well demonstrated in micro-line width and micro-structure. However, in thick glue and large line width processes, DBO faces practical application challenges. The thick glue layer is thick and opaque, which will interfere with the diffraction effect of the grating, resulting in DBO's inability to accurately capture the diffraction image of the mark. In addition, the asymmetry of the glue wall under the thick glue process and the groove structure at the bottom of the glue wall will also make it difficult for DBO to effectively process complex diffraction signals. The measurement error caused by the asymmetric structure of thick glue products exceeds the design application range of DBO. DBO relies on the symmetry of the diffraction grating, but in the thick glue process, due to the frequent occurrence of glue layer thickness and asymmetric structure, the diffraction image is severely distorted, resulting in a large deviation in the measurement results. This makes the measurement error of DBO in processing thick glue processes much greater than IBO, and it cannot even be applied normally.

[0046] Based on this, the embodiments of this specification propose a new solution for correcting measurement errors caused by asymmetric overlay marks. After quantifying the asymmetry based on the IBO technology, the measurement conditions are optimized with it as the target. The measurement conditions are adjusted by coupling with hardware parameters and software parameters to provide more accurate measurements, significantly reduce the need for rework, and improve the efficiency of the production line and product quality.

[0047] This solution is of great significance in Fab production, especially in fields requiring high reliability such as automotive chips and new energy power chips. It can solve the bottleneck of current measurement technology (including large optical measurement noise and inaccurate measurement), improve the stability and consistency of the process, and thus bring higher economic value.

[0048] The following are explanations of professional terms:

[0049] Overlay (OVL) describes the deviation between the current layer of lithography and the previous layer of lithography in the lithography process, thereby achieving the function of monitoring and providing data to correct the process.

[0050] Image-Based-Overlay Measurement: Image-Based-Overlay measurement based on optical imaging signals.

[0051] Diffraction-Based-Overlay Measurement: Overlay measurement based on optical diffraction signals.

[0052] Overlay marks (OVL-Marks) are graphic marks used to measure overlay errors. These marks are distributed in the lithography unit in a certain number. In terms of measurement type, they are generally divided into imaging measurement (IBO, Image-based Overlay) and non-imaging diffraction measurement (DBO, diffraction based overlay). The overlay mark graphics will also be different.

[0053] Mark-Signal is a marker signal used to measure the OVL error. It is a spatial domain signal obtained by CIs and is divided into signals in the x and y directions.

[0054] CIS (CMOS-Image-Sensor) is a sensor used to receive light signals marked by test samples.

[0055] NA (Numerical Aperture) is the numerical aperture of an optical system, which refers to half of the angle range in which light can enter the optical path lens module.

[0056] The technical solutions provided by various embodiments of the present application are described below in conjunction with the accompanying drawings.

[0057] like Figure 1-Figure 3 As shown, the embodiment of this specification provides a method for correcting measurement errors caused by asymmetric overlay marks, including steps S101-S103. In step S101, an optical imaging device is used to obtain information of the structure to be measured, and a preset algorithm is used to calculate the product analysis item and the asymmetry value of each product analysis item; wherein, the asymmetry value of the product analysis item is used to reflect the global and local asymmetry degree of the target overlay mark corresponding to the structure to be measured. Step S102, an initial asymmetry evaluation value is obtained according to the asymmetry value of each product analysis item, and by adjusting different combinations of hardware parameters and software parameters, when the initial asymmetry evaluation value meets the preset error adjustment, the target asymmetry evaluation value is obtained, and the iterated hardware parameters and software parameters are obtained; wherein, the hardware parameters obtain the relationship between the product analysis items and the software parameters through a large amount of data in the process of configuring the product analysis items, and the software parameters are used to enhance the image processing of the structure information to be measured, so as to cooperate with the hardware parameters to obtain the product analysis items and the target asymmetry evaluation value. Step S103, according to the iterated hardware parameters and software parameters, they are adjusted respectively to correct the measurement error caused by the asymmetry of the overlay marks.

[0058] Specifically, in step S101, the optical imaging device is used to obtain signals of the measurement object (i.e., the structural information to be measured), and these signals include images of the surface or internal structure of the object, optical reflection data, etc. The hardware of the optical imaging device includes but is not limited to an illumination module, an imaging device, and an optical adjustment module, and the structural information to be measured is obtained by adjusting the parameters of the hardware device. In some embodiments, high-quality optical signals are obtained by adjusting conditions such as illumination uniformity, illumination numerical aperture (NA), illumination center offset, wavelength, and focal depth offset.

[0059] The obtained structure information to be measured is calculated using a preset algorithm to obtain product analysis items and asymmetry values ​​of each product analysis item. The asymmetry values ​​of the product analysis items are used to reflect the global and local asymmetry degree in the target overlay mark corresponding to the structure to be measured. The preset algorithm includes Taylor expansion.

[0060] In some embodiments, a preliminary center position is obtained based on information of the structure to be measured, a function of the sampling signal and the reference information is obtained through a known reference signal, and then Taylor expansion is performed near a certain point of the target structure to obtain a global asymmetry degree such as the first-order derivative and the second-order derivative, and a local asymmetry degree such as the third-order or fourth-order derivative and higher-order derivatives, and a specific asymmetry value is obtained by calculating the specific values ​​of the derivatives of each order at the certain point.

[0061] In step S102, the morphology of the target overlay mark is analyzed in combination with the above-mentioned embodiment, and the asymmetry value of each product analysis item is obtained through parameters such as geometric morphology and brightness distribution. An initial asymmetry evaluation value is obtained based on the asymmetry value of each product analysis item, such as obtaining the initial asymmetry evaluation value based on the coupling of each product analysis item.

[0062] Among them, the asymmetry evaluation value or asymmetry evaluation function performs a detailed analysis on the product analysis items to obtain a measurement target asymmetry evaluation value to correct the measurement error caused by the asymmetry of the overlay mark. For example, the asymmetry angle of the side wall of the glue wall, the depth of the groove at the bottom of the glue wall, etc. are analyzed and refined to generate an asymmetry evaluation value or an asymmetry evaluation function, and then the hardware and / or software parameters are adjusted in a targeted manner according to the quantified asymmetry evaluation value to reduce the asymmetry evaluation value to an acceptable range, so as to correct the measurement error caused by the asymmetry of the overlay mark.

[0063] Based on the coupling of the optical imaging device and the preset algorithm used in the acquisition process of the above product analysis items, all factors such as hardware parameters and software parameters are directly bound, and the optimal asymmetric evaluation value (i.e., the target asymmetric evaluation value) is obtained by adjusting different combinations of hardware parameters and software parameters. The optimal asymmetric evaluation value corresponds to the hardware parameters and software parameters after iteration. The optical imaging device corresponds to the hardware parameters, and the preset algorithm corresponds to the software parameters.

[0064] In some embodiments, the hardware parameters obtain the relationship between the product analysis items and the software parameters through a large amount of data during the process of configuring the product analysis items. The software parameters are used to enhance the image processing of the measured structural information, that is, in the process of optimizing the asymmetry evaluation value, the image quality is improved to cooperate with the hardware parameters to obtain the product analysis items and the target asymmetry evaluation value.

[0065] In step S103, the iterated hardware parameters and software parameters are obtained by adjusting the hardware parameters and software parameters in combination with the above embodiment, and then the iterated hardware parameters and software parameters are adjusted respectively, so as to correct the measurement error caused by the asymmetry of the overlay mark.

[0066] In view of the above limitations of the existing IBO and DBO technologies, this application provides a measurement optimization system and method based on target asymmetry quantification, which solves the problem that the traditional technology cannot accurately measure in the face of thick glue process, target asymmetry and other scenarios, and has the following beneficial significance:

[0067] Accurately quantify asymmetry: The present invention can identify and calculate subtle structural features such as glue wall sidewall angle difference, groove depth, etc. by quantitatively analyzing the asymmetry of the target. This quantitative index can not only effectively compensate for the existing IBO measurement results, but also provide a basis for optimizing measurement conditions and reducing measurement errors caused by asymmetric structures.

[0068] Optimize measurement conditions and improve measurement accuracy: By combining hardware adjustment (such as illumination uniformity, NA, focal depth offset, etc.) and software processing (such as image enhancement, threshold method, etc.), the measurement parameters can be adjusted dynamically, so that the system can achieve higher measurement accuracy when facing asymmetric targets. Compared with IBO technology, which is limited by the resolution of the optical system, it provides a more flexible adjustment method and effectively improves measurement accuracy.

[0069] Applicable to thick glue process and reduce rework: In the thick glue process of automotive chips and new energy power chips, the frequent occurrence of asymmetric structures poses a huge challenge to measurement. The present invention can effectively identify and quantify these asymmetric features, and then optimize the measurement conditions, thereby reducing measurement errors and avoiding repeated process rework. Compared with the complex and expensive DBO system, the present invention has lower costs and a wider range of applications, and also has good application effects in thick glue and large line width processes.

[0070] Adaptable to various process requirements: The present invention is not only applicable to thick glue processes, but can also be applied in other process scenarios that require high-precision measurement. Through flexible solutions for asymmetry quantification and optimization of measurement conditions, the present invention can be widely used in the manufacturing processes of automotive chips, new energy power chips, and other complex structures.

[0071] In some embodiments, when applied to thick glue processes, product analysis items include but are not limited to glue wall angle, side wall bottom edge width, target signal brightness uniformity, groove depth, groove radius, and side wall height. Product analysis items are used to form an asymmetry evaluation function to obtain the corresponding relationship between hardware parameters, software parameters, and product analysis items through the asymmetry evaluation function.

[0072] Combined with the above embodiments, the embodiments of this specification are applied to thick glue processes and can also be applied in other process scenarios that require high-precision measurement. Through the flexible solution of asymmetry quantification and optimization of measurement conditions, it can be widely used in the manufacturing process of automotive chips, new energy power chips, and other complex structures.

[0073] The degree of asymmetry reflected by the product analysis items is used to form an asymmetry evaluation function. For example, the asymmetry evaluation function is expressed as Formula 1:

[0074] Asymmetric=f(θ,w,I,d,r,h)

[0075] Among them, θ, w, I, d, r, and h represent the morphological parameters of the target overlay mark, including the glue wall angle, the bottom edge width of the side wall, the uniformity of the target signal brightness, the groove depth, the groove radius, and the side wall height.

[0076] In some embodiments, the asymmetry evaluation function is expressed as the sum of derivatives of various orders of Taylor expansion.

[0077] The product analysis items constitute an asymmetric evaluation function, and the corresponding relationship between hardware parameters, software parameters and product analysis items is obtained through the asymmetric evaluation function.

[0078] In some embodiments, a machine learning algorithm is used to iterate the initial asymmetry evaluation value according to different combinations of hardware parameters and software parameters to obtain a target asymmetry evaluation value, wherein the preset error condition includes obtaining the minimum initial asymmetry evaluation value or setting a preset threshold value according to the allowable process error, and there is an initial asymmetry evaluation value greater than the preset threshold value.

[0079] In combination with the above embodiments, the asymmetric evaluation function constructed according to the product analysis items automatically adjusts the combination of different hardware parameters and software parameters to gradually reduce the asymmetric evaluation value. The optimization algorithms used include but are not limited to genetic algorithms, particle swarm optimization algorithms, and gradient descent methods. These optimization algorithms can reduce the complexity of traversal optimization to a certain extent, which is actually an improvement in efficiency. Relatively optimal results can be obtained without all data sets.

[0080] Genetic algorithm: Find the optimal parameter combination through population selection, crossover, mutation and other mechanisms.

[0081] Particle Swarm Optimization Algorithm: Based on the group behavior of particles with multiple parameters, it quickly converges to the optimal solution.

[0082] Gradient descent method: By calculating the gradient of the evaluation function, the parameters are quickly adjusted to make the evaluation value decrease continuously.

[0083] In some embodiments, the preset error condition includes obtaining a minimum initial asymmetry evaluation value; correspondingly, by adjusting different combinations of hardware parameters and software parameters, when the initial asymmetry evaluation value meets the preset error condition, a target asymmetry evaluation value is obtained, including: traversing each item of the hardware parameter and each item of the software parameter respectively; obtaining multiple initial asymmetry evaluation values ​​when different combinations of hardware parameters and software parameters are adjusted, and by comparing each initial asymmetry evaluation value, taking the minimum initial asymmetry evaluation value as the target asymmetry evaluation value.

[0084] Among them, the hardware parameters include lighting uniformity, lighting NA, wavelength, focal depth offset, lighting center-west offset, etc. The optimization direction of each hardware parameter adjustment is as follows:

[0085] Lighting uniformity adjustment: Reduce errors caused by uneven lighting by adjusting the uniformity of the light source.

[0086] Illumination NA adjustment: adjust the numerical aperture to optimize the resolution and light intensity distribution of the optical system.

[0087] Focus Depth Shift: Adjust the focus position to optimize the clarity of the image.

[0088] Illumination center offset: change the position of the irradiation center so that the target area can be accurately illuminated; and other conditions.

[0089] The software parameters include threshold method, enhancement method, cross-correlation method, etc. The optimization direction of each software parameter adjustment is as follows:

[0090] Threshold method: adjust the binarization threshold of the image to highlight the key structure of the target.

[0091] Enhancement: Apply image enhancement techniques, such as contrast enhancement and sharpening, to improve the visibility of structures.

[0092] Cross-correlation method: Utilize the correlation analysis of multiple image regions to further improve the recognition accuracy of asymmetric structures.

[0093] In combination with the above embodiment, according to the quantized initial asymmetry evaluation value, by traversing the hardware parameters and software parameters respectively, different initial asymmetry evaluation values ​​corresponding to different combinations of the hardware parameters and software parameters are obtained, and the minimum initial asymmetry evaluation value is used as the target asymmetry evaluation value by comparison, that is, by optimizing the initial asymmetry evaluation value, the iterative hardware parameters and software parameters are obtained to correct the measurement error caused by the asymmetry of the overlay mark. For example, the minimum initial asymmetry evaluation value is used as the target asymmetry evaluation value, and the hardware parameters corresponding to the target asymmetry evaluation values ​​are the illumination center offset to adjust the target area to accurately illuminate, and the software parameters are to highlight the key structure of the target area.

[0094] In some embodiments, the preset error condition includes setting a preset threshold value according to the allowable process error; correspondingly, by adjusting different combinations of hardware parameters and software parameters, when the initial asymmetry evaluation value meets the preset error condition, the target asymmetry evaluation value is obtained, including: traversing each item of the hardware parameters and each item of the software parameters respectively; when the number of traversals does not exceed the maximum number of optimizations, obtaining the initial asymmetry evaluation value when each different combination of hardware parameters and software parameters is adjusted; when an initial asymmetry evaluation value is greater than the preset threshold, the initial asymmetry evaluation value is used as the target asymmetry evaluation value.

[0095] Among them, the preset threshold is set according to the allowable process error. For example, if the thick glue asymmetry error in the 0.18 process is expected to be controlled within the range of 20nm, then the preset threshold is set to 20nm according to the allowable process error. According to the strictness of the process requirements, the preset threshold can be set to a smaller value to meet more precise accuracy. The specific different adjustment combinations of traversing hardware parameters and software parameters are described above and will not be repeated here.

[0096] Combination Figure 3 In the embodiment of this specification, the preset error condition is met when the number of traversals does not exceed the maximum number of optimizations. When the hardware parameters and software parameters are adjusted, there is a set of initial asymmetric evaluation values ​​corresponding to the hardware parameters and software parameters. The initial asymmetric evaluation value is greater than the preset threshold. Then the initial asymmetric evaluation value is used as the target asymmetric evaluation value. 0 ) exceeds the preset threshold A threshold , and it is judged that the maximum number of optimization times has not been exceeded, then the target asymmetry evaluation value is obtained.

[0097] In some embodiments, if the current asymmetry evaluation value is still greater than a preset threshold after the number of traversals exceeds the maximum number of optimizations, an alarm mechanism is entered. After multiple adjustments and optimizations, the asymmetry evaluation value is reduced to an acceptable threshold. If the evaluation value cannot be reduced to the threshold within a certain number of times, it is determined that the asymmetry of the target is too large and may have been damaged, and an alarm message needs to be issued to inform the process engineer for further processing.

[0098] In combination with the above embodiments, the embodiments of this specification provide a measurement error correction device caused by asymmetry of overlay marks, including an imaging and computing module, an imaging stage optimization module and a computing stage optimization module. The imaging stage optimization module and the computing stage optimization module are respectively communicated with the imaging and computing modules to form a coupling system, wherein the imaging stage optimization module is for hardware parameters, and the computing stage optimization module is for software parameters. The coupling system does not perform a decoupling analysis, but directly binds all hardware parameters and software parameters to adjust the hardware parameters and software parameters in a targeted manner, optimizes the quantized initial asymmetry evaluation value, and obtains an acceptable target asymmetry evaluation value that meets the preset error conditions.

[0099] Specifically, the imaging and calculation module is used to obtain the structure information to be measured by using an optical imaging device, and to calculate the product analysis items and the asymmetry value of each product analysis item by using a preset algorithm; wherein the asymmetry value of the product analysis item is used to reflect the global and local asymmetry degree in the target overlay mark in the structure to be measured; wherein the structure information to be measured includes the surface or internal structure image of the structure object to be measured, optical reflection data, etc., and the preset algorithm includes Taylor expansion.

[0100] An initial asymmetry evaluation value is obtained according to the asymmetry value of each product analysis item, and a target asymmetry evaluation value is obtained when the initial asymmetry evaluation value meets a preset error condition by adjusting different combinations of hardware parameters and software parameters, and iterative hardware parameters and software parameters are obtained, so as to adjust them respectively according to the iterative hardware parameters and software parameters;

[0101] The hardware parameters acquire the relationship between the product analysis items and the software parameters through a large amount of data during the configuration of the product analysis items. The software parameters are used to enhance the image processing of the structural information to be measured, so as to cooperate with the hardware parameters to obtain the product analysis items and the target asymmetry evaluation value;

[0102] An imaging stage optimization module is used to adjust the hardware parameters after iteration when correcting the asymmetry of the overlay mark in the structure to be measured, so as to correct the initial asymmetry evaluation value by adjusting the software parameters to meet the preset error conditions;

[0103] The calculation phase optimization module is used to adjust the software parameters after iteration when correcting the asymmetry of the overlay mark in the measured structure, so as to adjust the initial asymmetry evaluation value to meet the preset error conditions together with the hardware parameters. Figure 1 As shown, no further description is given here.

[0104] The embodiments of this specification are intended to solve the problem of inaccurate measurement caused by target asymmetry in Fab production. By quantifying the target asymmetry, a system and method for optimizing measurement conditions are proposed to improve the accuracy of optical measurement and reduce the process rework problem caused by it. The significance of this technology is reflected in the following aspects:

[0105] Quantification and compensation of asymmetry: This invention proposes an optimization compensation scheme based on quantified asymmetry, which can adjust the measurement conditions according to the asymmetric characteristics of the target object, including hardware parameters (such as illumination uniformity, illumination NA, illumination center offset, wavelength, focal depth offset, etc.) and software processing methods (such as different solution algorithms, image enhancement, etc.). By accurately quantifying the degree of asymmetry and providing compensation measures, the accuracy of the measurement results can be ensured.

[0106] Improve measurement accuracy and reduce rework: This invention reduces measurement errors caused by asymmetry and avoids repeated rework by optimizing hardware and software conditions. In the manufacturing of automotive chips and new energy power chips, this optimization solution can significantly improve process efficiency, reduce production costs, and increase the final yield of products.

[0107] Enhance the stability and consistency of the Fab process: By providing a dynamically optimized measurement solution, the present invention helps to enhance the process stability of the Fab production line, especially when facing the frequent asymmetry problems of thick glue processes, it can significantly improve the measurement accuracy and reliability, and ensure that every link of the production is under precise control.

[0108] The same or similar parts between the various embodiments in this specification can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the product embodiment described later, since it corresponds to the method, the description is relatively simple, and the relevant parts can be referred to the partial description of the system embodiment.

[0109] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for correcting measurement errors caused by asymmetric overlay marks, characterized in that: include: An optical imaging device is used to obtain information of the structure to be measured, and a preset algorithm is used to calculate the product analysis items and the asymmetry value of each product analysis item; wherein the asymmetry value of the product analysis item is used to reflect the global and local asymmetry degree in the target overlay mark corresponding to the structure to be measured; An initial asymmetry evaluation value is obtained according to the asymmetry value of each product analysis item, and a target asymmetry evaluation value is obtained by adjusting different combinations of hardware parameters and software parameters when the initial asymmetry evaluation value meets a preset error condition, and the iterative hardware parameters and software parameters are obtained; wherein, the hardware parameters obtain the relationship between the product analysis items and the software parameters through a large amount of data during the configuration of the product analysis items, and the software parameters are used to perform enhanced image processing on the structural information to be measured, so as to cooperate with the hardware parameters to obtain the product analysis items and the target asymmetry evaluation value; The hardware parameters and software parameters after iteration are adjusted respectively to correct the measurement error caused by the asymmetry of the overlay mark.

2. The method for correcting measurement errors caused by asymmetric overlay marks according to claim 1, characterized in that: Applied to thick glue process, the product analysis items include at least one of the following: glue wall angle, side wall bottom edge width, target signal brightness uniformity, groove depth, groove radius, side wall height; The product analysis items are used to form an asymmetry evaluation function, so as to obtain the corresponding relationship between the hardware parameters, the software parameters and the product analysis items through the asymmetry evaluation function; Among them, the preset algorithm includes Taylor expansion.

3. The method for correcting measurement errors caused by asymmetric overlay marks according to claim 1, characterized in that: The machine learning algorithm is used to iterate the initial asymmetric evaluation value according to different combinations of hardware parameters and software parameters to obtain the target asymmetric evaluation value. The machine learning algorithm includes at least one of the following: genetic algorithm, particle swarm optimization algorithm, and gradient descent method.

4. The method for correcting measurement errors caused by asymmetric overlay marks according to claim 1, characterized in that: The preset error condition includes obtaining a minimum initial asymmetry evaluation value; Correspondingly, by adjusting different combinations of hardware parameters and software parameters, when the initial asymmetry evaluation value meets the preset error condition, the target asymmetry evaluation value is obtained, including: Traverse each item of hardware parameters and each item of software parameters respectively; Obtaining multiple initial asymmetry evaluation values ​​when different combinations of hardware parameters and software parameters are adjusted; By comparing each initial asymmetry evaluation value, the minimum initial asymmetry evaluation value is taken as the target asymmetry evaluation value.

5. The method for correcting measurement errors caused by asymmetric overlay marks according to claim 1, characterized in that: The preset error condition includes setting a preset threshold value according to the process allowable error; Correspondingly, by adjusting different combinations of hardware parameters and software parameters, when the initial asymmetry evaluation value meets the preset error condition, the target asymmetry evaluation value is obtained, including: Traverse each item of hardware parameters and each item of software parameters respectively; When the number of iterations does not exceed the maximum number of optimizations, the initial asymmetric evaluation value is obtained when each different combination of hardware parameters and software parameters is adjusted; When an initial asymmetry evaluation value is greater than a preset threshold, the initial asymmetry evaluation value is used as a target asymmetry evaluation value.

6. The method for correcting measurement errors caused by asymmetric overlay marks according to any one of claims 1 to 5, characterized in that: The hardware parameters include at least one of the following: illumination uniformity, numerical aperture of the illumination optical system, wavelength, focal depth offset, and illumination center.

7. The method for correcting measurement errors caused by asymmetric overlay marks according to claim 6, characterized in that: The software parameters include at least one of the following: a threshold method, an enhancement method, and a cross-correlation method.

8. A device for correcting measurement errors caused by asymmetric overlay marks, characterized in that: include: Imaging and calculation module, imaging phase optimization module and calculation phase optimization module; An imaging phase optimization module and a computing phase optimization module are respectively connected to the imaging and computing modules in communication; An imaging and calculation module, used to obtain information of the structure to be measured by using an optical imaging device, and to calculate the product analysis items and the asymmetry value of each product analysis item by using a preset algorithm; wherein the asymmetry value of the product analysis item is used to reflect the global and local asymmetry degree in the target overlay mark in the structure to be measured; An initial asymmetry evaluation value is obtained according to the asymmetry value of each product analysis item, and a target asymmetry evaluation value is obtained when the initial asymmetry evaluation value meets a preset error condition by adjusting different combinations of hardware parameters and software parameters, and iterative hardware parameters and software parameters are obtained, so as to adjust them respectively according to the iterative hardware parameters and software parameters; The hardware parameters acquire the relationship between the product analysis items and the software parameters through a large amount of data during the configuration of the product analysis items. The software parameters are used to enhance the image processing of the structural information to be measured, so as to cooperate with the hardware parameters to obtain the product analysis items and the target asymmetry evaluation value; An imaging stage optimization module is used to adjust the hardware parameters after iteration when correcting the asymmetry of the overlay mark in the structure to be measured, so as to correct the initial asymmetry evaluation value by adjusting the software parameters to meet the preset error conditions; The calculation phase optimization module is used to adjust the software parameters after iteration when correcting the asymmetry of the overlay mark in the structure to be measured, so as to correct the initial asymmetry evaluation value by adjusting the hardware parameters to meet the preset error conditions.

9. A system for correcting measurement errors caused by asymmetric overlay marks, characterized in that: include: A memory, a processor and a computer program, wherein the computer program is stored in the memory, and the processor runs the computer program to execute the measurement error correction method caused by the asymmetry of the overlay mark according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, it is used to implement the measurement error correction method caused by the asymmetry of the overlay mark as described in any one of claims 1 to 7.

Images