Lithography model establishment method and device, storage medium and program product

Abstract

The invention discloses a photoetching model establishment method and device, a storage medium and a program product, and relates to the technical field of semiconductors. The method comprises the following steps: acquiring actual morphology measurement data of photoresist in a photoetching simulation process and N physical and chemical effect functions of the photoresist corresponding to the photoetching simulation process; based on the N physical and chemical effect functions, establishing an initial photoetching model; based on the initial photoetching model and the actual morphology measurement data, a target parameter combination with fitness meeting a first preset condition is obtained through iterative solution by running a genetic algorithm, and the fitness is determined and obtained based on a loss function and / or a root-mean-square error of the initial photoetching model; and based on the target parameter combination, determining to obtain a target photoetching model of the photoresist under the photoetching simulation process. According to the embodiment of the invention, the method can achieve the building of the photoetching model more efficiently and accurately, thereby effectively improving the building efficiency of the whole photoetching model, and improving the prediction precision of the photoresist morphology.

Description

Technical Field

[0001] The present application belongs to the field of semiconductor technology, and in particular, relates to a method, device, storage medium and program product for establishing a lithography model. Background Art

[0002] In the field of semiconductor technology, photolithography is a key process in integrated circuit manufacturing. It mainly transfers the target pattern to the silicon wafer substrate through steps including substrate preparation, glue coating, pre-baking, alignment, exposure, post-baking, development, measurement, etching, etc. With the continuous deepening of technology nodes, the size of the target pattern of photolithography is getting smaller and smaller, and the optical diffraction effect and the physical and chemical effects occurring in the photoresist are increasingly affecting the shape of the pattern on the substrate after photolithography. Therefore, optical proximity correction (OPC) of the design pattern plays an increasingly important role in the manufacturing process of very large-scale integrated circuits.

[0003] At present, in actual production, the main correction method is to perform OPC processing on the design graphics based on empirical rules. With the continuous advancement of process nodes, it is becoming more and more difficult to determine the correction rules. The traditional rule-based OPC method is increasingly difficult to meet the needs of actual processes. The OPC method based on the lithography model solves this problem. The lithography model is an empirical model obtained by fitting the actual process morphology measurement data based on actual physical and chemical effects. The lithography model can simulate the process from exposure to development in the process, simulate the specific morphology of the photoresist after development, and further correct the design graphics. Therefore, the accuracy of the lithography model prediction has become the key to affecting the OPC correction capability based on the lithography model. However, the existing lithography model establishment methods often have technical defects that are not efficient and accurate enough, and the lithography models obtained are still lacking in accurate prediction.

[0004] Based on this, the industry is still in urgent need of a more efficient lithography model building solution to achieve the goal of more accurate simulation and prediction of photoresist morphology data in the lithography simulation process. Summary of the invention

[0005] The embodiments of the present application provide a lithography model establishment method, device, storage medium and program product, which can realize lithography model establishment more efficiently and accurately, thereby effectively improving the overall lithography model establishment efficiency and improving the accuracy of photoresist morphology prediction.

[0006] In a first aspect, an embodiment of the present application provides a method for establishing a lithography model, the method for establishing a lithography model comprising:

[0007] Acquire actual morphology measurement data of the photoresist in the photolithography simulation process and N physical and chemical effect functions of the photoresist corresponding to the photolithography simulation process, where the physical and chemical effect functions are used to characterize the physical and chemical effects of the photoresist in the photolithography simulation process, and N is a positive integer;

[0008] Based on the N physical and chemical effect functions, an initial lithography model is established, wherein the initial lithography model includes a plurality of parameters to be solved in the N physical and chemical effect functions;

[0009] Based on the initial lithography model and actual morphology measurement data, a target parameter combination whose fitness meets the first preset condition is obtained by iteratively running a genetic algorithm, the fitness being determined based on a loss function and / or a root mean square error of the initial lithography model, and the target parameter combination including target values ​​corresponding to a plurality of parameters to be solved;

[0010] Based on the target parameter combination, the target lithography model of the photoresist in the lithography simulation process is determined.

[0011] In some possible implementations, based on the initial lithography model and actual morphology measurement data, a target parameter combination whose fitness meets the preset conditions is obtained by running a genetic algorithm for iterative solution, including:

[0012] A plurality of chromosomes are randomly generated as the initial population, and any of the plurality of chromosomes is composed of a plurality of binary-coded genes connected in series, and the plurality of genes respectively correspond to random values ​​of a plurality of parameters to be solved;

[0013] Repeatedly executing the calculation based on the actual morphology measurement data and the initial lithography model to obtain the fitness of multiple chromosomes in the current round population, and selecting some chromosomes from the multiple chromosomes based on the fitness; and,

[0014] Determine the next round of population based on the selected part of chromosomes, and iterate and update the current round of population to the next round of population, until the iteration obtains the target population that meets the iteration termination condition;

[0015] A target chromosome that minimizes the fitness of the initial lithography model is selected from the target population, and the target chromosome is decoded into a corresponding target parameter combination.

[0016] In some possible implementations, the iteration termination conditions include: the number of population iterations reaches the maximum number of iterations, the fitness of the chromosomes appearing in the population reaches a predetermined fitness target, or the population iteration running time reaches a preset running time; the lithography model establishment method also includes:

[0017] Determine the target convergence speed parameter required for population iteration based on the maximum number of iterations, the predetermined fitness target or the preset running time;

[0018] Based on the correspondence between the preset convergence speed parameter and the population screening ratio, determining the target population screening ratio corresponding to the target convergence speed parameter;

[0019] Based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in the current population is calculated one by one, and some chromosomes are selected from the multiple chromosomes based on the fitness, including:

[0020] Based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in this round of population is calculated one by one, and some chromosomes are selected from multiple chromosomes based on the fitness and the target population screening ratio.

[0021] In some possible implementations, based on actual morphology measurement data and an initial lithography model, the fitness of multiple chromosomes in the current round population is calculated one by one, and some chromosomes are selected from the multiple chromosomes based on the fitness and the target population screening ratio, including:

[0022] Based on the actual morphology measurement data and the initial photolithography model, the fitness of multiple chromosomes in this round of population is calculated one by one;

[0023] Compare the fitness of multiple chromosomes one by one, and determine the fitness ranking results of multiple chromosomes;

[0024] Based on the fitness ranking results, some chromosomes are selected from multiple chromosomes according to the target population screening ratio.

[0025] In some possible implementations, determining the next round population based on the selected partial chromosomes, and iteratively updating the current round population to the next round population includes:

[0026] Performing chromosome mutation and / or chromosome crossover operations on the selected part of the chromosomes to generate a plurality of chromosomes after the operations;

[0027] Select chromosomes whose fitness meets the second preset condition from the selected part of chromosomes to obtain M chromosomes, where M is a non-negative integer;

[0028] Based on the chromosomes processed by multiple operations and the M chromosomes, the next round of population is constructed, and the current round of population is iteratively updated to the next round of population.

[0029] In some possible implementations, before obtaining actual morphology measurement data of the photoresist in the photolithography simulation process, the photolithography model establishment method further includes:

[0030] Acquire initial morphology measurement data of the photoresist after the development step of the photolithography simulation process;

[0031] Perform data cleaning on the initial topography measurement data according to preset data cleaning rules to obtain actual topography measurement data;

[0032] The preset data cleaning rules include: eliminating data whose deviation of multiple measurement results is greater than a preset deviation threshold, and / or eliminating data that deviates from a preset critical size range.

[0033] In some possible implementations, the fitness includes: a root mean square error between the predicted morphology measurement data of the photoresist in the lithography simulation process and the actual morphology measurement data of the photoresist by the initial lithography model;

[0034] The fitness calculation formula is:

[0035]

[0036] Among them, rms is the fitness, model_CD1 is the first key dimension in the predicted shape measurement data, gauge_CD1 is the first key dimension in the actual shape measurement data, model_CD n To predict the nth critical dimension in the topography measurement data, gauge_CD n is the nth critical dimension in the actual topography measurement data.

[0037] Based on the same inventive concept, in a second aspect, an embodiment of the present application provides a photolithography model building device, the photolithography model building device comprising:

[0038] A first acquisition module is used to acquire actual morphology measurement data of the photoresist in the photolithography simulation process and N physical and chemical effect functions of the photoresist corresponding to the photolithography simulation process, where the physical and chemical effect functions are used to characterize the physical and chemical effects of the photoresist in the photolithography simulation process, and N is a positive integer;

[0039] A first establishing module is used to establish an initial lithography model based on N physical and chemical effect functions, wherein the initial lithography model includes a plurality of parameters to be solved in the N physical and chemical effect functions;

[0040] A first solving module is used to obtain a target parameter combination whose fitness meets a first preset condition by iteratively solving by running a genetic algorithm based on an initial lithography model and actual topography measurement data, wherein the fitness is determined based on a loss function and / or a root mean square error of the initial lithography model, and the target parameter combination includes target values ​​corresponding to a plurality of parameters to be solved;

[0041] The first determination module is used to determine a target lithography model of the photoresist in a lithography simulation process based on a target parameter combination.

[0042] In a third aspect, an embodiment of the present application provides a lithography model building device, the lithography model building device comprising:

[0043] a processor and a memory storing computer program instructions;

[0044] When the processor executes the computer program instructions, the lithography model establishment method provided in any one of the above-mentioned embodiments of the present application is implemented.

[0045] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, a lithography model establishment method as provided in any one of the above-mentioned embodiments of the present application is implemented.

[0046] In a fifth aspect, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes a lithography model building method as provided in any one of the above-mentioned embodiments of the present application.

[0047] It can be known from the above description that a method, device, storage medium and program product for establishing a photolithography model provided by an embodiment of the present application establish an initial photolithography model based on N physicochemical effect functions of the photoresist corresponding to the photolithography simulation process, and then based on the actual morphology measurement data of the photoresist obtained by measurement, the initial photolithography model is iteratively solved by using a genetic algorithm, so as to quickly obtain a target parameter combination whose fitness meets the first preset condition under the premise of ensuring a reasonable amount of calculation. In this way, the target photolithography model of the photoresist under the photolithography simulation process is determined based on the target parameter combination. In addition, the design of the present application adopts the fitness obtained by determining the loss function and / or the root mean square error based on the initial photolithography model to iteratively obtain the solution value of the photolithography model parameter more accurately, so as to finally determine to obtain a more efficient and accurate target photolithography model, which can make a more accurate prediction of the photoresist morphology size. Overall, the method, device, storage medium and program product for establishing a photolithography model in an embodiment of the present application can realize the establishment of a photolithography model more efficiently and accurately, so as to effectively improve the efficiency of establishing the overall photolithography model and improve the prediction accuracy of the photoresist morphology. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0049] Figure 1 It is a schematic diagram of the process of establishing a lithography model provided by an embodiment of the present application;

[0050] Figure 2 is a comparative schematic diagram of the fitness of a lithography model provided in an embodiment of the present application;

[0051] Figure 3 It is a structural schematic diagram of a photolithography model building device provided in one embodiment of the present application;

[0052] Figure 4 It is a structural schematic diagram of a lithography model building device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0053] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0054] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0055] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0056] As mentioned in the background technology section, the accuracy of lithography model prediction is the key to the OPC correction capability based on the lithography model, and the existing lithography model building methods often have technical defects of being inefficient and inaccurate. Specifically, when determining the lithography model, the most accurate parameter combination of the model is often calculated through traversal search or local optimization algorithm based on gradient descent.

[0057] However, although the above traversal search method can search for the global optimal parameter combination, the amount of calculation is too large when the number of parameters involved is large, so it is not suitable for the establishment of a lithography model containing a large number of parameters. The local optimization algorithm based on gradient descent can quickly find a better parameter combination based on the gradient of the loss function, but it is easy to fall into the local optimum, resulting in poor prediction effect of the final lithography model.

[0058] In view of the above, in order to solve the problems of the prior art, the embodiments of the present application provide a method, device, storage medium and program product for establishing a lithography model, so as to achieve the purpose of more accurately simulating and predicting the photoresist morphology data in the lithography simulation process. It should be noted that the embodiments provided in this application are not intended to limit the scope of the disclosure of this application.

[0059] The following first introduces the lithography model establishment method provided in the embodiment of the present application.

[0060] Figure 1 The schematic diagram of the process of establishing a lithography model provided by an embodiment of the present application is shown. The lithography model establishing method is applied to an electronic device, which may include a server or a user terminal. Figure 1 As shown, the lithography model establishment method includes the following steps:

[0061] S110, obtaining actual morphology measurement data of the photoresist in the photolithography simulation process and N physicochemical effect functions of the photoresist corresponding to the photolithography simulation process, where the physicochemical effect functions are used to characterize the physicochemical effects of the photoresist in the photolithography simulation process, and N is a positive integer;

[0062] S120, establishing an initial lithography model based on the N physical and chemical effect functions, wherein the initial lithography model includes a plurality of parameters to be solved in the N physical and chemical effect functions;

[0063] S130, based on the initial lithography model and the actual morphology measurement data, a target parameter combination whose fitness meets the first preset condition is obtained by iteratively running a genetic algorithm, the fitness being determined based on a loss function and / or a root mean square error of the initial lithography model, and the target parameter combination including target values ​​corresponding to a plurality of parameters to be solved;

[0064] S140, determining a target lithography model of the photoresist in the lithography simulation process based on the target parameter combination.

[0065] It can be seen from the above description that the method for establishing a lithography model provided in the embodiment of the present application establishes an initial lithography model based on N physical and chemical effect functions of the photoresist corresponding to the lithography simulation process, and then based on the actual morphology measurement data of the measured photoresist, the initial lithography model is iteratively solved by using a genetic algorithm, so as to quickly obtain a target parameter combination whose fitness meets the first preset condition under the premise of ensuring a reasonable amount of calculation. In this way, the target lithography model of the photoresist under the lithography simulation process is determined based on the target parameter combination.

[0066] Furthermore, the embodiment of the present application is designed to use the fitness determined based on the loss function and / or the root mean square error of the initial lithography model to iteratively obtain the solution value of the lithography model parameter more accurately, so as to finally determine a more efficient and accurate target lithography model, which can make a more accurate prediction of the photoresist shape size. In general, the lithography model establishment method of the embodiment of the present application can realize the lithography model establishment more efficiently and accurately, thereby effectively improving the overall lithography model establishment efficiency and improving the photoresist shape prediction accuracy.

[0067] The specific implementation of the above steps 110 to 140 is described in detail below.

[0068] In S110, during specific implementation, actual morphology measurement data of the photoresist in the lithography simulation process and N physicochemical effect functions of the photoresist corresponding to the lithography simulation process are obtained, and the physicochemical effect functions are used to characterize the physicochemical effects of the photoresist in the lithography simulation process.

[0069] In the semiconductor field, the actual photoresist lithography simulation process generally includes the steps of photoresist coating, pre-baking, alignment and exposure, post-baking, and development. In this embodiment, when obtaining the actual morphology measurement data of the above-mentioned photoresist, a test mask is manufactured according to the process parameters used in the actual photolithography simulation and the chip type design. After the photoresist is exposed and developed using the test mask, the morphology size of the developed photoresist is measured by a scanning electron microscope (SEM) and other equipment to obtain the above-mentioned actual morphology measurement data.

[0070] The above-mentioned actual morphology measurement data may be characteristic dimension information such as line width, line end shortening, corner rounding, etc. of the actual photoresist after the development step, and no strict restrictions are imposed here. In addition, in order to ensure that the prediction effect of the subsequent lithography model determined based on the measurement data is better, the amount of the above-mentioned actual morphology measurement data can generally be selected to be around 1000-5000.

[0071] The above N physical and chemical effect functions are related to the photoresist reaction steps involved in the photolithography simulation process, and can be selected based on the design graphics and process conditions. The above N physical and chemical effect functions represent a series of physical and chemical effects that occur in the photoresist during processes such as exposure and development, and can reflect the diffusion and mechanical properties of the photoresist. For example, the Dill model can be used as one of the above N physical and chemical effect functions, which describes the changes in the chemical properties of the photoresist during optical exposure.

[0072] These physical and chemical effect functions are the basis for the role of photoresist in the lithography process. There are often a series of parameters to be solved in the physical and chemical effect functions. The different values ​​of these parameters can reflect the strength of the physical effect of the photoresist. It is necessary to solve them in the subsequent lithography model determination process to obtain the parameter combination that makes the lithography model prediction effect best.

[0073] It should be added that, considering that the specific steps involved in different lithography simulation processes may be different, the above-mentioned N physical and chemical effect functions will also change accordingly. Therefore, the specific contents of the above-mentioned N physical and chemical effect functions are not limited here.

[0074] Optionally, in some feasible embodiments of the present application, the accuracy of the model depends on the quality and quantity of the experimental data. Therefore, in order to fully guarantee the data quality and reliability of the above-mentioned actual morphology measurement data so as to improve the subsequent model prediction accuracy, before obtaining the actual morphology measurement data of the photoresist in the above-mentioned lithography simulation process, the lithography model establishment method further includes:

[0075] Acquire initial morphology measurement data of the photoresist after the development step of the photolithography simulation process;

[0076] Perform data cleaning on the initial topography measurement data according to preset data cleaning rules to obtain actual topography measurement data;

[0077] The preset data cleaning rules include: eliminating data whose deviation of multiple measurement results is greater than a preset deviation threshold, and / or eliminating data that deviates from a preset critical size range.

[0078] In this embodiment, the initial morphology measurement data of the photoresist after the development step of the lithography simulation process is obtained by preliminary measurement using a scanning electron microscope. Considering that there may be operating errors or unsatisfactory lithography simulation results in the above measurement process, there may be deviation data in the initial morphology measurement data, and these deviation data need to be cleaned to avoid affecting the prediction accuracy and reliability of the subsequent lithography model.

[0079] During the specific data cleaning, the above-mentioned initial morphology measurement data is pre-processed by using the preset data cleaning rules, so as to delete the data with low reliability. In particular, considering the actual lithography simulation scenario, the reliability of data with large deviations in multiple measurement results is low, so the data with measurement result deviations greater than the preset deviation threshold in the initial morphology measurement data can be deleted.

[0080] Also, considering that the actual lithography model has a limited coverage, using measurement data that is too large or too small (such as line width) may involve different physical and chemical effects, which may affect the accuracy of the lithography model solution. Moreover, in actual applications, measurement data with extremely large deviations rarely appear on the same layer of the chip. Therefore, in order to ensure that a higher-precision lithography model can be obtained, the above-mentioned preset critical dimension range is set in advance in this embodiment, so that the data outside the preset critical dimension range in the initial morphology measurement data can be deleted in the data cleaning step, thereby fully ensuring the data quality and reliability of the above-mentioned actual morphology measurement data, which in turn helps to improve the prediction accuracy of the subsequent lithography model.

[0081] It should be added that the above-mentioned preset deviation threshold and preset critical dimension range are flexibly set according to actual lithography simulation requirements, manufacturing process limitations or lithography modeling experience, and are not strictly limited here.

[0082] In S120 , during specific implementation, an initial lithography model is established based on the N physical and chemical effect functions, and the initial lithography model includes a plurality of parameters to be solved in the N physical and chemical effect functions.

[0083] After obtaining the N physical and chemical effect functions, the initial lithography model is established by integrating or splicing the functions. Each of the N physical and chemical effect functions has a corresponding parameter to be solved, so the initial lithography model also includes multiple parameters to be solved in the N physical and chemical effect functions, and the values ​​of the multiple parameters to be solved need to be solved in subsequent steps.

[0084] In a specific example, the initial lithography model established may be set to be in the form of:

[0085] M=f1(x)+f2(x)+…f N (x)

[0086] Among them, f1(x), f2(x), f N(x) corresponds to the above-mentioned N physicochemical effect functions, which represent a series of physicochemical effects that occur in the photoresist during exposure and development. The strength of these physicochemical effects is determined by the parameters in the function, and this application will specifically use genetic algorithms to solve and determine these parameters in the future. These parameters may include the sensitivity, development rate, diffusion coefficient, etc. of the photoresist.

[0087] In S130, based on the initial lithography model and actual morphology measurement data, a target parameter combination whose fitness meets the first preset condition is obtained by running a genetic algorithm for iteration, and the fitness is determined based on the loss function and / or root mean square error of the initial lithography model. The target parameter combination includes target values ​​corresponding to multiple parameters to be solved.

[0088] In this embodiment, a global optimization genetic algorithm (GA) is used to iteratively solve the parameters to be solved in the initial lithography model. Compared with the local optimization algorithm in the existing solution, it is less likely to fall into the local optimum, and does not need to involve a large amount of calculation when the amount of data is large, as in the traversal method. Therefore, the genetic algorithm is used to globally search and obtain the optimal parameter combination of the lithography model, thereby improving the efficiency of parameter solution and the accuracy of lithography model prediction.

[0089] In specific implementation, a genetic algorithm is run iteratively based on the initial lithography model to obtain a target parameter combination whose fitness meets the first preset condition. The above-mentioned actual morphology measurement data can be used in the fitness solution corresponding to different parameter combinations during the genetic algorithm iteration process.

[0090] The fitness is determined based on the loss function and / or the root mean square error of the initial lithography model. In this embodiment, the fitness corresponding to the parameter combination can reflect the stability and prediction accuracy of the lithography model under the parameter combination.

[0091] The above-mentioned first preset condition can be, for example: the fitness is the smallest among multiple fitnesses corresponding one-to-one to multiple parameter combinations when the iteration termination round is reached, or the fitness is the smallest among multiple fitnesses corresponding one-to-one to multiple parameter combinations of the current round when the iteration termination time is reached, etc. This embodiment does not make strict restrictions here.

[0092] Optionally, in some feasible embodiments of the present application, in order to more effectively implement the iterative solution of the above target parameter combination based on the genetic algorithm, the above target parameter combination whose fitness meets the preset conditions is obtained by iteratively solving the initial lithography model and the actual morphology measurement data by running the genetic algorithm, including:

[0093] A plurality of chromosomes are randomly generated as the initial population, and any of the plurality of chromosomes is composed of a plurality of binary-coded genes connected in series, and the plurality of genes respectively correspond to random values ​​of a plurality of parameters to be solved;

[0094] Repeatedly executing the calculation based on the actual morphology measurement data and the initial lithography model to obtain the fitness of multiple chromosomes in the current round population, and selecting some chromosomes from the multiple chromosomes based on the fitness; and,

[0095] Determine the next round of population based on the selected part of chromosomes, and iterate and update the current round of population to the next round of population, until the iteration obtains the target population that meets the iteration termination condition;

[0096] A target chromosome that minimizes the fitness of the initial lithography model is selected from the target population, and the target chromosome is decoded into a corresponding target parameter combination.

[0097] The basic concepts in genetic algorithms include genes, chromosomes, populations, and fitness. Genes are the most basic units in the algorithm, chromosomes are composed of multiple genes, and populations contain multiple chromosomes. Fitness guides the direction of algorithm optimization.

[0098] In the genetic algorithm used for determining the lithography model in the present application, the gene may correspond to the value of any parameter to be solved among the multiple parameters to be solved in the above-mentioned initial lithography model, and the gene is specifically expressed in the form of binary coding. The above-mentioned chromosomes may correspond to the parameter value combination of multiple parameters to be solved in the initial lithography model, and the chromosomes are formed by connecting all binary-coded genes in series. The population is the sum of all chromosomes in a round of iterations, and each round of iteration of the population is obtained by processing the part of the chromosomes with the best fitness in the previous round of population. The fitness can be calculated based on the loss function or root mean square error of the initial lithography model under the parameter combination.

[0099] In specific implementation, when performing iterative parameter solving based on genetic algorithm, multiple chromosomes are first randomly generated as the initial population, and any chromosome among the multiple chromosomes is composed of multiple binary-coded genes connected in sequence, and the multiple genes correspond to random values ​​of multiple parameters to be solved. The binary codes of each gene in different chromosomes are not completely consistent, and different chromosomes represent different value combinations of the above multiple parameters to be solved in the parameter space.

[0100] After the initial population is generated, the fitness of multiple chromosomes in the current population is calculated one by one, and then some chromosomes are selected from the multiple chromosomes for processing according to the fitness to construct the next round of population. Thus, the current population is iterated to the next round of population constructed and generated.

[0101] By analogy, the fitness of the chromosomes in the current population is calculated and the chromosomes are screened to achieve iterative update of the population until the target population that meets the iteration termination condition is obtained. After the target population is obtained, the target chromosome that minimizes the fitness of the initial lithography model is selected from the target population, and the target chromosome is decoded into a corresponding target parameter combination, which can effectively guarantee the performance of the final lithography model.

[0102] It is worth noting that in this embodiment, considering that the relevant parameters of the physicochemical effect function to be optimized by the genetic algorithm include both continuous values ​​and discrete values, the binary encoding method can more easily convert between these two data forms. In addition, when the gene is encoded in binary, the operations such as chromosome crossover and mutation involved in the iterative process are relatively simple and the amount of calculation is small, and information can be effectively transmitted between individuals. In addition, for some more complex situations, the number of parameters involved in the model calculation is large, resulting in a huge solution space, and the binary encoding form can effectively narrow the search space.

[0103] Optionally, in some feasible embodiments of the present application, in combination with the actual lithography model iteration requirements, in order to ensure the effective realization of the optimal parameter solution under different iteration requirements, the above iteration termination conditions include: the number of population iterations reaches the maximum number of iterations, the fitness of the chromosomes appearing in the population reaches a predetermined fitness target, or the population iteration running time reaches a preset running time; the lithography model establishment method also includes:

[0104] Determine the target convergence speed parameter required for population iteration based on the maximum number of iterations, the predetermined fitness target or the preset running time;

[0105] Based on the correspondence between the preset convergence speed parameter and the population screening ratio, determining the target population screening ratio corresponding to the target convergence speed parameter;

[0106] Based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in the current population is calculated one by one, and some chromosomes are selected from the multiple chromosomes based on the fitness, including:

[0107] Based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in this round of population is calculated one by one, and some chromosomes are selected from multiple chromosomes based on the fitness and the target population screening ratio.

[0108] In this embodiment, the user can set the desired genetic algorithm iteration termination condition. When different users have different iteration termination conditions, even if the aforementioned initial lithography model and the actual morphology measurement data are consistent, the fitness of the target parameter combinations obtained under different iteration termination conditions may be significantly different, resulting in uneven performance of the lithography model obtained in the end. Therefore, it is possible to consider regulating the convergence speed of the genetic algorithm.

[0109] Specifically, the speed of convergence of the genetic algorithm will affect the accuracy of the lithography model obtained after a certain number of iterations. When the convergence is faster, fewer iterations (i.e., less computing resources and less computing time) can be used to obtain better results. When the convergence is slower, more iterations can be used to obtain better results. Therefore, how to adjust the convergence speed to balance the consumption of computing resources and model accuracy is an issue that needs to be studied.

[0110] However, the present application found that the convergence speed of the genetic algorithm is closely related to the number of individuals selected for the next generation of population generation. Therefore, in view of the above situation, in order to ensure the good performance of the final lithography model under different iteration termination conditions, the corresponding relationship between the above preset convergence speed parameters and the population screening ratio can be set in advance by using actual experience and simulation tests.

[0111] After the user sets the iteration termination condition for the genetic algorithm used for the lithography model, the convergence speed required by the user can be determined based on the iteration termination condition. The iteration termination condition can specifically include that the number of population iterations reaches the maximum number of iterations, the fitness of the chromosomes appearing in the population reaches a predetermined fitness target, or the population iteration running time reaches a preset running time. When determining the convergence speed parameter, illustratively, if the maximum number of iterations indicated in the iteration termination condition is greater than 50, it can be determined that the user requires a slower convergence speed; if the maximum number of iterations indicated in the iteration termination condition is less than 20, it can be determined that the user requires a faster convergence speed.

[0112] After determining the target convergence speed parameter required by the user, the target population screening ratio corresponding to the target convergence speed parameter can be determined according to the correspondence between the preset convergence speed parameter and the population screening ratio. The target population screening ratio is the proportion of chromosome individuals in the previous round of population (current population) selected for the generation of the new population (next round of population). The target population screening ratio is applied to the iterative update of the population to play a good role in regulating the convergence speed.

[0113] Combine the following Figure 2 Make an introduction, Figure 2FIG. 1 is a schematic diagram showing the fitness comparison of the lithography model provided in an embodiment of the present application, taking the fitness as the root mean square error (RMS) as an example. Figure 2 The waveform shown can illustrate the influence of the convergence speed of the genetic algorithm caused by different proportions of individuals of the previous generation selected for the generation of new populations on the fitness of the lithography model.

[0114] like Figure 2 As shown, in the scheme provided in this embodiment, it is recommended to set the proportion of individuals selected for the next generation of population generation to 30-50% of the proportion of individuals in the previous generation, that is, the above-mentioned target population screening ratio is set in the range of 30-50%, wherein the setting of 30% is suitable for fast convergence, and the setting of 50% is suitable for slow convergence, and no strict limitation is made here.

[0115] Optionally, in some feasible embodiments of the present application, in order to more reasonably realize the screening of chromosomes in the population for constructing the next round of population, so as to fully guarantee the reliability of population iteration and global optimization of the genetic algorithm, the fitness of multiple chromosomes in the current round of population is calculated based on the actual morphology measurement data and the initial lithography model, and some chromosomes are selected from the multiple chromosomes based on the fitness and the target population screening ratio, including:

[0116] Based on the actual morphology measurement data and the initial photolithography model, the fitness of multiple chromosomes in this round of population is calculated one by one;

[0117] Compare the fitness of multiple chromosomes one by one, and determine the fitness ranking results of multiple chromosomes;

[0118] Based on the fitness ranking results, some chromosomes are selected from multiple chromosomes according to the target population screening ratio.

[0119] In this embodiment, the fitness of multiple chromosomes in the current round population can be calculated by calculating the root mean square error or loss function of the photolithography model under each chromosome. Then, the fitness ranking results of multiple chromosomes can be determined by comparing and ranking the fitness of multiple chromosomes in the current round population.

[0120] Next, considering that the smaller the fitness of the lithography model is, the more stable the performance of the model is and the more accurate the prediction effect is, based on the fitness ranking results of multiple chromosomes in this round of population, some chromosomes with the smallest fitness are selected according to the target population screening ratio.

[0121] For example, taking the target population screening ratio as 30%, 30% of the chromosomes with the smallest fitness are selected from multiple chromosomes in the current round population. In this way, the chromosomes used to generate a new round of population can be reasonably screened in the population.

[0122] Optionally, in some feasible embodiments of the present application, in order to fully iteratively update a new round of population based on the partial chromosomes obtained by the above screening, thereby realizing rapid global optimization of individuals in the new round of population, so as to fully guarantee the reliability of population iteration and global optimization of genetic algorithm, the above determination of the next round of population based on the selected partial chromosomes, and iteratively updating the current round of population to the next round of population, includes:

[0123] Performing chromosome mutation and / or chromosome crossover operations on the selected part of the chromosomes to generate a plurality of chromosomes after the operations;

[0124] Select chromosomes whose fitness meets the second preset condition from the selected part of chromosomes to obtain M chromosomes, where M is a non-negative integer;

[0125] Based on the chromosomes processed by multiple operations and the M chromosomes, the next round of population is constructed, and the current round of population is iteratively updated to the next round of population.

[0126] For ease of understanding, the following is a specific example to describe the generation method of the next round of population in this embodiment. Exemplarily, the first 30-50% of the chromosomes with the smallest fitness in the current round of population are selected for the next round of population generation. A chromosome crossover operation is performed on the selected first 30-50% of the chromosomes, and a chromosome mutation operation is performed on the selected first 30-50% of the chromosomes, thereby generating chromosomes after multiple operation processing.

[0127] The second preset condition is, for example, a small number of chromosomes with the smallest fitness among the selected partial chromosomes. In this way, M chromosomes are further screened from the first 30-50% of the selected partial chromosomes, and the M chromosomes are, for example, a small number of chromosomes with the smallest fitness in the first 10% of the population in this round.

[0128] Thus, the next round of population is constructed based on the chromosomes processed by the above multiple operations and M chromosomes. The chromosomes in the next round of population are composed of three parts: 1) The chromosomes directly retained from the previous round to this round, and the best few chromosomes (fitness top 10%) in the previous round are retained intact to this round; 2) The chromosomes of the best part of the chromosomes in the previous round (fitness top 30-50%) are generated through crossover operation; 3) The chromosomes of the best part of the chromosomes in the previous round (fitness top 30-50%) are generated through mutation operation. These three parts each account for a certain proportion to form a new round of population.

[0129] Optionally, in some feasible embodiments of the present application, more specifically, the actual photoresist profile measurement involves the measurement of multiple types of key dimensions. Therefore, in order to enable the fitness of the present application to better reflect the performance of the lithography model, the fitness may specifically include: the root mean square error between the predicted profile measurement data of the photoresist in the lithography simulation process and the actual profile measurement data of the initial lithography model;

[0130] The fitness calculation formula is:

[0131]

[0132] Among them, rms is the fitness, model_CD1 is the first key dimension in the predicted shape measurement data, gauge_CD1 is the first key dimension in the actual shape measurement data, model_CD n To predict the nth critical dimension in the topography measurement data, gauge_CD n is the nth critical dimension in the actual topography measurement data. The above critical dimension may be, for example, feature dimension information such as line width.

[0133] It should be added that the fitness can also be set as a more complex function to more effectively verify the accuracy and universality of the model. For example, the fitness can specifically include evaluation indicators such as the root mean square error and the stability of the lithography model, which are not limited here.

[0134] In S140, when it is specifically implemented, after determining to obtain the above target parameter combination, based on the target parameter combination, a target lithography model of the photoresist in the lithography simulation process is determined. Specifically, the values ​​of the multiple parameters to be solved in the initial lithography model are set one by one as the target values ​​of the multiple parameters to be solved in the target parameter combination, so as to obtain the target lithography model after the parameter solution is completed.

[0135] The target lithography model can be effectively applied to OPC (optical proximity correction) software to predict and correct possible distortions in the lithography process, such as line width variation, line end shortening, and corner rounding. In this way, the accuracy and yield of the lithography process can be significantly improved through the OPC technology based on the target lithography model.

[0136] In order to facilitate understanding of the lithography model establishment method provided by the above embodiment, the above method is described below with a specific scenario embodiment. In this scenario embodiment, the local optimization algorithm in the prior art and the genetic algorithm provided by the embodiment of the present application are respectively used to solve the parameters that make the lithography model prediction effect optimal.

[0137] In the comparative experiment, the scenario embodiment may specifically include the following steps:

[0138] Step 1: Collect a set of actual morphology measurement data of the pattern on the photoresist after actual development. Please refer to Table 1. Table 1 shows some actual morphology measurement data of the photoresist after development, as follows:

[0139] Table 1

[0140] index gauge startx starty endx endy mcd wafercd 1 gauge_1 3248182 163419.2 3248240 163419.2 58 58 2 gauge_2 3248211 163390.2 3248211 163448.2 58 58 3 gauge_3 3230453 163329.2 3230510 163329.2 57 51.66 4 gauge_4 3266362 163239.2 3266421 163239.2 59 63.13 5 gauge_5 3284361 163149.2 3284421 163149.2 60 68.1 6 gauge_6 3230481 163300.7 3230481 163357.7 57 52.22 7 gauge_7 3266391 163209.7 3266391 163268.7 59 63.74 8 gauge_8 3284391 163119.2 3284391 163179.2 60 68.3 9 gauge_9 3176089 144994 3176143 144994 54 50.25 10 gauge_10 3194089 145249 3194144 145249 55 55.93

[0141] In Table 1, index is index, gauge is gauge / measurement tool, startx is starting X coordinate, starty is starting Y coordinate, endx is ending X coordinate, endy is ending Y coordinate, mcd is mean critical dimension (Mean Critical Dimension) and wafercd is wafer critical dimension (Wafer Critical Dimension).

[0142] Step 2: Input the optical parameters, photoresist process parameters and the actual morphology measurement data in Table 1, and use the OPC tool based on the local optimization algorithm to perform simulation to obtain the lithography model and CD that can be used for OPC. model Please refer to Table 2, which shows the CD obtained based on the local optimization algorithm. model , as follows:

[0143] Table 2

[0144]

[0145]

[0146] Table 2 shows the wafer critical dimension wafercd, the predicted morphology value model_cd of the lithography model based on the local optimization algorithm, and the corresponding lithography model prediction error model_error.

[0147] Step 3: Binary encode the parameters used in all physical effect functions as "genes", and then splice the "genes" together to form "chromosomes". These binary vectors are used as the input of the OPC tool based on the genetic algorithm. At the same time, input the optical parameters, photoresist process parameters and the actual morphology measurement data in Table 1, and use the OPC tool based on the genetic algorithm for simulation to obtain the lithography model and CD that can be used for OPC. model The values ​​are shown in Table 3. Please refer to Table 3. Table 2 shows the CD obtained based on the genetic algorithm. model , as follows:

[0148] Table 3

[0149] index gauge wafercd model_cd model_error 1 gauge_1 58 57.966 -0.034 2 gauge_2 58 58.098 0.098 3 gauge_3 51.66 52.784 1.124 4 gauge_4 63.13 62.958 -0.172 5 gauge_5 68.1 68.196 0.096 6 gauge_6 52.22 52.894 0.674 7 gauge_7 63.74 63.12 -0.62 8 gauge_8 68.3 68.397 0.097 9 gauge_9 50.25 51.327 1.077 10 gauge_10 55.93 55.973 0.043

[0150] Table 2 shows the wafer critical dimension wafercd, the predicted morphology value model_cd of the lithography model based on the genetic algorithm, and the corresponding lithography model prediction error model_error.

[0151] Finally, based on the results obtained by the local optimal algorithm and the genetic algorithm, the root mean square error (RMS) of the lithography model using the two different optimization algorithms is calculated and compared. The RMS (1.148) of the lithography model using the local optimization algorithm is significantly higher than the RMS (0.958) of the lithography model using the genetic algorithm. Therefore, applying the genetic algorithm to the parameter solution of the lithography model can obviously obtain a lithography model with higher accuracy and better performance.

[0152] In this scenario embodiment, a genetic algorithm is used to optimize the parameters of the lithography model based on the physical effect function, which can greatly save the amount of calculation compared to the traversal search method. At the same time, compared with the local optimization algorithm based on the gradient descent of the loss function, a more accurate lithography model can be obtained.

[0153] Based on the lithography model establishment method provided in the above embodiment, out of the same inventive concept, the present application also provides a lithography model establishment device corresponding to the above lithography model establishment method. Figure 3 The photolithography model building device is introduced in detail.

[0154] Figure 3 A schematic structural diagram of a lithography model building device provided in one embodiment of the present application is shown.

[0155] Figure 3 The lithography model building apparatus 300 shown includes:

[0156] A first acquisition module 310 is used to acquire actual morphology measurement data of the photoresist in the photolithography simulation process and N physical and chemical effect functions of the photoresist corresponding to the photolithography simulation process, where the physical and chemical effect functions are used to characterize the physical and chemical effects of the photoresist in the photolithography simulation process, and N is a positive integer;

[0157] A first establishing module 320 is used to establish an initial lithography model based on N physical and chemical effect functions, wherein the initial lithography model includes a plurality of parameters to be solved in the N physical and chemical effect functions;

[0158] A first solving module 330 is used to obtain a target parameter combination whose fitness meets a first preset condition by iteratively solving by running a genetic algorithm based on an initial lithography model and actual topography measurement data, wherein the fitness is determined based on a loss function and / or a root mean square error of the initial lithography model, and the target parameter combination includes target values ​​corresponding to a plurality of parameters to be solved;

[0159] The first determination module 340 is used to determine a target lithography model of the photoresist in a lithography simulation process based on a target parameter combination.

[0160] The embodiment of the present application provides a lithography model establishment device, which establishes an initial lithography model based on N physical and chemical effect functions of the photoresist corresponding to the lithography simulation process by setting corresponding functional modules, and then iterates and solves the initial lithography model based on the actual morphology measurement data of the photoresist obtained by measurement by using a genetic algorithm, so as to quickly obtain a target parameter combination whose fitness meets the first preset condition under the premise of ensuring a reasonable amount of calculation. In this way, the target lithography model of the photoresist under the lithography simulation process is determined based on the target parameter combination.

[0161] Furthermore, the embodiment of the present application is designed to use the fitness determined based on the loss function and / or the root mean square error of the initial lithography model to iteratively obtain the solution value of the lithography model parameter more accurately, so as to finally determine a more efficient and accurate target lithography model, which can make a more accurate prediction of the photoresist shape size. In general, the lithography model establishment device of the embodiment of the present application can more efficiently and accurately establish the lithography model, thereby effectively improving the overall lithography model establishment efficiency and improving the accuracy of photoresist shape prediction.

[0162] Optionally, in some feasible embodiments of the present application, the first solution module 330, based on the initial lithography model and the actual morphology measurement data, obtains a target parameter combination whose fitness meets the preset conditions by running a genetic algorithm for iterative solution, including:

[0163] The first generation submodule can be used to randomly generate multiple chromosomes as an initial population, wherein any chromosome among the multiple chromosomes is composed of multiple binary-coded genes connected in series, and the multiple genes correspond to random values ​​of multiple parameters to be solved;

[0164] The first iterative submodule can be used to repeatedly execute the calculation of the fitness of multiple chromosomes in the current round population based on the actual morphology measurement data and the initial lithography model, and select some chromosomes from the multiple chromosomes based on the fitness; and

[0165] Determine the next round of population based on the selected part of chromosomes, and iterate and update the current round of population to the next round of population, until the iteration obtains the target population that meets the iteration termination condition;

[0166] The first selection submodule can be used to select a target chromosome from the target population that minimizes the fitness of the initial lithography model, and decode the target chromosome into a corresponding target parameter combination.

[0167] Optionally, in some feasible embodiments of the present application, the iteration termination condition includes: the number of population iterations reaches the maximum number of iterations, the fitness of the chromosomes appearing in the population reaches a predetermined fitness target, or the population iteration running time reaches a preset running time; the lithography model building device also includes:

[0168] The second determination module may be used to determine a target convergence speed parameter required for population iteration based on a maximum number of iterations, a predetermined fitness target, or a preset running time;

[0169] The third determination module may be used to determine a target population screening ratio corresponding to a target convergence speed parameter based on a correspondence between a preset convergence speed parameter and a population screening ratio;

[0170] Based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in the current round population is calculated one by one, and some chromosomes are selected from the multiple chromosomes based on the fitness, including:

[0171] Based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in this round of population is calculated one by one, and some chromosomes are selected from multiple chromosomes based on the fitness and the target population screening ratio.

[0172] Optionally, in some feasible embodiments of the present application, based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in the current round population is calculated one by one, and some chromosomes are selected from the multiple chromosomes based on the fitness and the target population screening ratio, including:

[0173] Based on the actual morphology measurement data and the initial photolithography model, the fitness of multiple chromosomes in this round of population is calculated one by one;

[0174] Compare the fitness of multiple chromosomes one by one, and determine the fitness ranking results of multiple chromosomes;

[0175] Based on the fitness ranking results, some chromosomes are selected from multiple chromosomes according to the target population screening ratio.

[0176] Optionally, in some feasible embodiments of the present application, determining the next round population based on the selected part of the chromosomes, and iteratively updating the current round population to the next round population includes:

[0177] Performing chromosome mutation and / or chromosome crossover operations on the selected part of the chromosomes to generate a plurality of chromosomes after the operations;

[0178] Select chromosomes whose fitness meets the second preset condition from the selected part of chromosomes to obtain M chromosomes, where M is a non-negative integer;

[0179] Based on the chromosomes processed by multiple operations and the M chromosomes, the next round of population is constructed, and the current round of population is iteratively updated to the next round of population.

[0180] Optionally, in some feasible embodiments of the present application, before obtaining the actual morphology measurement data of the photoresist in the photolithography simulation process, the photolithography model building device further includes:

[0181] The second acquisition module can be used to obtain the initial morphology measurement data of the photoresist after the development step of the photolithography simulation process;

[0182] The first cleaning module can be used to clean the initial topography measurement data according to a preset data cleaning rule to obtain actual topography measurement data;

[0183] The preset data cleaning rules include: eliminating data whose deviation of multiple measurement results is greater than a preset deviation threshold, and / or eliminating data that deviates from a preset critical size range.

[0184] Optionally, in some feasible embodiments of the present application, the fitness includes: a root mean square error between the predicted morphology measurement data and the actual morphology measurement data of the photoresist in the lithography simulation process by the initial lithography model;

[0185] The fitness calculation formula is:

[0186]

[0187] Among them, rms is the fitness, model_CD1 is the first key dimension in the predicted shape measurement data, gauge_CD1 is the first key dimension in the actual shape measurement data, model_CD n To predict the nth critical dimension in the topography measurement data, gauge_CD n is the nth critical dimension in the actual topography measurement data.

[0188] Based on the lithography model establishment method provided in the above embodiment, out of the same inventive concept, the present application also provides a lithography model establishment device corresponding to the above lithography model establishment method. Figure 4 A detailed introduction to the photolithography model building equipment is given.

[0189] See below Figure 4 , Figure 4 It is a structural schematic diagram of a lithography model building device provided in one embodiment of the present application.

[0190] The lithography model building apparatus may include a processor 401 and a memory 402 storing computer program instructions.

[0191] Specifically, the processor 401 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0192] Memory 402 may include a large capacity memory for data or instructions. By way of example and not limitation, memory 402 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, memory 402 may include a removable or non-removable (or fixed) medium. In appropriate cases, memory 402 may be inside or outside of an integrated gateway disaster recovery device. In a specific embodiment, memory 402 is a non-volatile solid-state memory.

[0193] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0194] The processor 401 reads and executes computer program instructions stored in the memory 402 to implement any one of the photolithography model building methods in the above embodiments.

[0195] In one example, the data lithography model building device may further include a communication interface 403 and a bus 410. Figure 4 As shown, the processor 401, the memory 402, and the communication interface 403 are connected via a bus 410 and communicate with each other.

[0196] The communication interface 403 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0197] Bus 410 includes hardware, software or both, and the parts of lithography model building equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-side bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 410 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.

[0198] The lithography model building device executes the lithography model building method in the embodiment of the present application, thereby realizing the lithography model building method described in the embodiment of the present application.

[0199] In addition, in combination with the lithography model building method in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the lithography model building methods in the above embodiment is implemented.

[0200] Based on the lithography model establishment method in the above-mentioned embodiment, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the lithography model establishment method provided in any one of the above-mentioned embodiments of the present application.

[0201] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0202] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0203] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0204] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0205] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A method for establishing a photolithography model, characterized in that: The method comprises: Acquire actual morphology measurement data of the photoresist in the photolithography simulation process and N physicochemical effect functions of the photoresist corresponding to the photolithography simulation process, wherein the physicochemical effect functions are used to characterize the physicochemical effects of the photoresist in the photolithography simulation process, and N is a positive integer; Based on the N physical and chemical effect functions, an initial lithography model is established, wherein the initial lithography model includes a plurality of parameters to be solved in the N physical and chemical effect functions; Based on the initial lithography model and the actual topography measurement data, a target parameter combination whose fitness meets a first preset condition is obtained by iteratively running a genetic algorithm, wherein the fitness is determined based on a loss function and / or a root mean square error of the initial lithography model, and the target parameter combination includes target values ​​corresponding to the multiple parameters to be solved respectively; Based on the target parameter combination, a target lithography model of the photoresist in the lithography simulation process is determined.

2. The method according to claim 1, characterized in that The step of obtaining a target parameter combination whose fitness meets preset conditions by iteratively solving the initial lithography model and the actual morphology measurement data by running a genetic algorithm comprises: Randomly generate multiple chromosomes as an initial population, wherein any chromosome among the multiple chromosomes is composed of multiple binary-coded genes in series, and the multiple genes respectively correspond to random values ​​of the multiple parameters to be solved; Repeating the process of calculating the fitness of multiple chromosomes in the current round population based on the actual morphology measurement data and the initial lithography model, and selecting some chromosomes from the multiple chromosomes based on the fitness; and Determine the next round of population based on the selected part of chromosomes, and iteratively update the current round of population to the next round of population, until the iteration obtains the target population that meets the iteration termination condition; A target chromosome that minimizes the fitness of the initial lithography model is selected from the target population, and the target chromosome is decoded into the corresponding target parameter combination.

3. The method according to claim 2, characterized in that The iteration termination conditions include: the number of population iterations reaches the maximum number of iterations, the fitness of the chromosomes appearing in the population reaches a predetermined fitness target, or the population iteration running time reaches a preset running time; the method also includes: Determining a target convergence speed parameter required for population iteration based on the maximum number of iterations, the predetermined fitness target, or the preset running time; Based on the correspondence between the preset convergence speed parameter and the population screening ratio, determining a target population screening ratio corresponding to the target convergence speed parameter; The step of calculating the fitness of a plurality of chromosomes in the current round population in a one-to-one correspondence based on the actual morphology measurement data and the initial lithography model, and selecting a portion of chromosomes from the plurality of chromosomes based on the fitness, comprises: Based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in the current round population is calculated one by one, and some chromosomes are selected from the multiple chromosomes based on the fitness and the target population screening ratio.

4. The method according to claim 3, characterized in that The step of calculating the fitness of a plurality of chromosomes in the current round population based on the actual morphology measurement data and the initial lithography model, and selecting a portion of chromosomes from the plurality of chromosomes based on the fitness and the target population screening ratio includes: Based on the actual morphology measurement data and the initial lithography model, the fitness of multiple chromosomes in the current round population is calculated one by one; Comparing the fitness of the multiple chromosomes one by one, and determining the fitness ranking results of the multiple chromosomes; Based on the fitness ranking result, the partial chromosomes are selected from the plurality of chromosomes according to the target population screening ratio.

5. The method according to claim 2, characterized in that: The determining of the next round population based on the selected part of chromosomes, and iteratively updating the current round population to the next round population, comprises: Performing chromosome mutation and / or chromosome crossover operation processing on the selected part of chromosomes to generate a plurality of chromosomes after the operation processing; Selecting chromosomes whose fitness meets the second preset condition from the selected part of chromosomes to obtain M chromosomes, where M is a non-negative integer; Based on the chromosomes after the multiple operations and the M chromosomes, the next round population is constructed, and the current round population is iteratively updated to the next round population.

6. The method according to claim 1, characterized in that Before obtaining the actual morphology measurement data of the photoresist in the photolithography simulation process, the method further includes: Acquiring initial morphology measurement data of the photoresist after the development step of the photolithography simulation process; Performing data cleaning on the initial topography measurement data according to a preset data cleaning rule to obtain the actual topography measurement data; The preset data cleaning rules include: eliminating data with deviations of multiple measurement results greater than a preset deviation threshold, and / or eliminating data that deviates from a preset critical size range.

7. The method according to claim 1, characterized in that The fitness includes: the root mean square error between the predicted morphology measurement data of the photoresist in the lithography simulation process and the actual morphology measurement data of the initial lithography model; The calculation formula of the fitness is: Wherein, rms is the fitness, model_CD1 is the first key dimension in the predicted shape measurement data, gauge_CD1 is the first key dimension in the actual shape measurement data, model_CD n is the nth critical dimension in the predicted topography measurement data, gauge_CD n is the nth critical dimension in the actual topography measurement data.

8. A photolithography model building device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the lithography model establishment method according to any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the lithography model establishment method according to any one of claims 1 to 7 is implemented.

10. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the lithography model building method according to any one of claims 1 to 7.

Images