Oxidation layer uniformity control method and device, electronic equipment, medium and product
By optimizing the process parameters and utilizing multi-objective optimization algorithms and response surface models, the growth of oxide layers in the deposition equipment was controlled, solving the problem of insufficient uniformity in silicon oxide films and improving cost-effectiveness and equipment compatibility.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-29
- Publication Date
- 2026-04-07
Smart Images
Figure CN121815990A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor technology, and more specifically to a method, apparatus, electronic device, dielectric, and product for controlling oxide layer uniformity. Background Technology
[0002] As semiconductor manufacturing process nodes continue to shrink (from 45nm to more advanced processes), the aspect ratio of shallow trench isolation (STI) structures has increased significantly. When forming oxide layers (e.g., silicon oxide films) in the trenches using low-pressure chemical vapor deposition (LPCVD) or plasma-enhanced chemical vapor deposition (PECVD) processes, voids are easily generated on the oxide layers.
[0003] To address the aforementioned issues, a high aspect ratio process (HARP) combined with sub-atmospheric chemical vapor deposition (SACVD) is currently used to form the oxide layer. However, the uniformity of the silicon oxide film generated by SACVD needs further improvement. Related technologies improve the uniformity of the silicon oxide film by modifying the deposition equipment structure, but this is costly and has low compatibility with different equipment models. Summary of the Invention
[0004] This invention provides a method, apparatus, electronic device, medium, and product for controlling the uniformity of oxide layers, in order to solve the problems of high cost and low compatibility when improving the uniformity of silicon oxide films by modifying the structure of deposition equipment.
[0005] In a first aspect, the present invention provides a method for controlling oxide layer uniformity, comprising: obtaining a target range of performance parameters and a constraint range of each process parameter among multiple process parameters, wherein the performance parameters are parameters characterizing oxide layer thickness uniformity, and the process parameters are parameters affecting oxide layer thickness uniformity; determining a target parameter combination based on multiple constraint ranges and a target optimization model, wherein the target parameter combination is one of at least one set of parameter combinations that makes the values of the performance parameters fall within the target range, the parameter combination is a combination of values of multiple process parameters, and the target optimization model is used to characterize the mapping relationship between the values of multiple process parameters and the values of the performance parameters; and controlling a deposition equipment to grow oxide on the wafer surface according to the target parameter combination to obtain a target oxide layer.
[0006] The oxide layer uniformity control method provided in this embodiment, after obtaining the target range of performance parameters and the constraint range of each process parameter among multiple process parameters, determines a combination of multiple process parameter values (target parameter combination) that makes the value of the first parameter fall within the target range based on multiple constraint ranges and a target optimization model. Then, it controls the deposition equipment to grow oxide on the wafer surface according to the target parameter combination to obtain the target oxide layer. This invention, under the constraints of multiple constraint ranges, adjusts the values of multiple process parameters in conjunction with a target optimization model. By optimizing the values of multiple process parameters to make the performance parameters characterizing thickness uniformity fall within the target range, it can improve the thickness uniformity of the final oxide layer without changing the SACVD deposition equipment structure, reducing production costs and improving equipment compatibility.
[0007] In one optional implementation, the target parameter combination is determined based on multiple constraint ranges and target optimization models, including: optimizing the values of multiple process parameters using a multi-objective optimization algorithm based on multiple constraint ranges and target optimization models to obtain multiple sets of parameter combinations; and determining the parameter combination that minimizes the value of the performance parameter among the multiple sets of parameter combinations as the target parameter combination.
[0008] In this embodiment, based on multiple constraint ranges and target optimization models, the values of multiple process parameters are optimized through a multi-objective optimization algorithm, which can more efficiently and accurately determine the combination of target parameters.
[0009] In one alternative implementation, the multi-objective optimization algorithm includes a multi-objective genetic algorithm or a multi-objective particle swarm optimization algorithm.
[0010] In this embodiment, the multi-objective genetic algorithm has good global search capabilities, which can quickly find the optimal solution set that makes the performance parameters fall within the target range, improve the efficiency of parameter determination, and avoid getting trapped in local optima.
[0011] In an optional implementation, before determining the target parameter combination, the method for controlling oxide layer uniformity further includes: acquiring a sample database and a response surface model, wherein the sample database includes multiple samples, each sample consisting of a set of actual parameter combinations and the values of performance parameters corresponding to the actual parameter combinations; training the response surface model based on the multiple samples until the trained response surface model reaches the convergence condition; and determining the response surface model that has reached the convergence condition as the target optimization model.
[0012] In this embodiment, the response surface model is trained by Gaussian process regression. The kernel function of Gaussian process regression can capture more subtle parameter interaction effects, which can improve the prediction accuracy of the optimization model with a defined objective.
[0013] In one alternative implementation, multiple process parameters include the wafer-to-nozzle spacing, deposition time, deposition rate, deposition temperature, and chamber pressure, and the response surface model is a quinary nonlinear response surface model.
[0014] In one alternative embodiment, the spacing is constrained to a range of 180 mils to 300 mils, the deposition time is constrained to a range of 200 s to 250 s, the deposition rate is constrained to a range of 19 Å / s to 23 Å / s, the deposition temperature is constrained to a range of 530 ℃ to 570 ℃, and the chamber pressure is constrained to a range of 200 Tor to 400 Tor.
[0015] Secondly, the present invention provides a device for controlling oxide layer uniformity, comprising: an acquisition module for acquiring a target range of performance parameters and a constraint range of each process parameter among multiple process parameters, wherein the performance parameters are parameters characterizing oxide layer thickness uniformity, and the process parameters are parameters affecting oxide layer thickness uniformity; a parameter determination module for determining a target parameter combination based on multiple constraint ranges and a target optimization model, wherein the target parameter combination is one of multiple parameter combinations that makes the values of the performance parameters fall within the target range, the parameter combination is a combination of values of multiple process parameters, and the target optimization model is used to characterize the mapping relationship between the values of multiple process parameters and the values of the performance parameters; and a control module for controlling a deposition equipment to grow oxide on the wafer surface according to the target parameter combination to obtain a target oxide layer.
[0016] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the oxide layer uniformity control method of the first aspect or any corresponding embodiment described above.
[0017] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the oxide layer uniformity control method of the first aspect or any corresponding embodiment described above.
[0018] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the oxide layer uniformity control method of the first aspect or any corresponding embodiment described above. Attached Figure Description
[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0020] Figure 1 This is a schematic diagram of the panel structure of a deposition equipment before optimization. Figure 2 This is a schematic diagram of the panel structure of an optimized deposition equipment; Figure 3 This is a structural diagram showing the location of a horizontal adjustment device. Figure 4 This is a flowchart illustrating a method for controlling oxide layer uniformity according to an embodiment of the present invention. Figure 5 This is a schematic diagram of SACVD growth of silicon oxide thin films according to an embodiment of the present invention; Figure 6 This is a flowchart illustrating another method for controlling oxide layer uniformity according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the user interface when inputting parameters according to an embodiment of the present invention; Figure 8 This is a schematic diagram of the user interface when outputting parameters according to an embodiment of the present invention; Figure 9 This is a structural block diagram of an oxide layer uniformity control device according to an embodiment of the present invention; Figure 10 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention.
[0021] Reference numerals: 11, panel; 111, center hole; 12, heater; 13, leveling screw; 14, wafer; 15, reaction chamber. Detailed Implementation
[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0023] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.
[0024] Shallow trench isolation (STI) structures achieve electrical isolation by forming silicon oxide-filled trenches between the active regions of a semiconductor device, effectively preventing current interference between adjacent transistors. Compared to other isolation processes, shallow trench isolation structures can suppress the formation of parasitic field-effect transistors and improve latch-up effects, making them an indispensable process step in chip manufacturing.
[0025] As mentioned in the background section, in order to improve the problem of voids easily generated in the silicon oxide film in the STI structure formed by LPCVD and PECVD processes, a high aspect ratio process (HARP) combined with sub-atmospheric pressure chemical vapor deposition (SACVD) is used to fill the trench with a silicon oxide film.
[0026] Specifically, tetraoxoethylsilane (TEOS) and ozone (O3) can be used as reactant gases to deposit silica (SiO2) under conditions of a chamber temperature of 540°C, a substrate temperature of 430°C, and a chamber pressure of 600 Torr. Using a sub-atmospheric pressure environment (below atmospheric pressure) extends the free path of gas molecules, allowing TEOS / O3 to mix thoroughly and uniformly cover the trench sidewalls and bottom. The chemical formula for the TEOS / O3 reaction is: TEOS + O3 → SiO2 + byproduct. The byproduct is generated and discharged (or volatilizes) during the reaction process.
[0027] However, the uniformity of silicon oxide films generated by SACVD needs further improvement. The uniformity of silicon oxide film thickness directly affects the process stability of subsequent chemical mechanical polishing (CMP) and wet etching (WET). Excessive edge thickness may lead to residual defects after CMP, such as silicon nitride (SIN) residue defects, and may even cause short circuits or leakage in semiconductor devices.
[0028] In related technologies, the uniformity of silicon oxide films is mainly improved by modifying deposition equipment. For example, the uniformity of silicon oxide films is improved by adjusting the shape of the faceplate. Compared to... Figure 1 The traditional panel shown, such as Figure 2As shown, a center hole 111 is added to panel 11 to optimize gas distribution, allowing gas to reach the wafer surface uniformly and form a uniform silicon oxide film; for example, as... Figure 3 As shown, a leveling kit is installed in the heater 12. The leveling kit includes multiple leveling screws 13, which are used to adjust the levelness of the heater 12, keeping the wafer 14 on the heater 12 level, thereby improving the thickness uniformity of the silicon oxide film. Improving the thickness uniformity of the silicon oxide film by improving the deposition equipment is costly and has low compatibility with different equipment models.
[0029] In view of this, the present invention provides a method, apparatus, electronic device, medium and product for controlling oxide layer uniformity. By optimizing the values of multiple process parameters through a target optimization model, the performance parameters characterizing thickness uniformity are kept within the target range. This can improve the thickness uniformity of the final oxide layer without changing the structure of the SACVD deposition equipment.
[0030] According to an embodiment of the present invention, a method for controlling the uniformity of an oxide layer is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.
[0031] This embodiment provides a method for controlling the uniformity of the oxide layer, which can be used in a computer system. Figure 4 This is a flowchart of a method for controlling the uniformity of an oxide layer according to an embodiment of the present invention, such as... Figure 4 As shown, the process includes the following steps: Step S401: Obtain the target range of the performance parameters and the constraint range of each process parameter among multiple process parameters.
[0032] Among them, performance parameters are parameters that characterize the uniformity of oxide layer thickness. For example, performance parameters can be thickness range, thickness average, or thickness uniformity.
[0033] The thickness range (THK Range) can be the difference between the maximum and minimum thickness values of the oxide layer at multiple different measurement points (locations). ,in, This represents the maximum thickness among all measurement points on the oxide layer. This represents the minimum thickness among all measured points on the oxide layer; the average thickness. It can be the average thickness of all measurement points on the oxide layer; the thickness uniformity can be characterized by formula (1) or formula (2):
[0034] In the formula, Indicates the uniformity of oxide layer thickness. This represents the standard deviation of the thickness at all measurement points of the oxide layer. The thickness range is positively correlated with thickness uniformity; the smaller the thickness range, the smaller the thickness uniformity, indicating better thickness uniformity of the oxide layer.
[0035] The target range refers to the range of values that the performance parameters of the final generated target oxide layer should reach. The target range can be configured by the designer according to the processing requirements. For example, if the performance parameter is the thickness range, the target range can be 100 mils to 300 mils; if the performance parameter is the thickness average, the target range can be 5500 Å ± 200 Å; if the performance parameter is thickness uniformity, the target range can be Uniformity < 2.4%.
[0036] Process parameters are parameters that affect the uniformity of oxide layer thickness. The constraint range of process parameters is the adjustable range of their values, which can be configured by the designer. Multiple process parameters may include at least two of the following: wafer-to-nozzle spacing, deposition time, deposition rate, deposition temperature, and chamber pressure.
[0037] For example, the constraint range of process parameters can be an adjustable range of process parameter values without affecting the growth quality of silicon oxide thin films. The constraint range of the wafer-to-nozzle spacing can be 180 mils to 300 mils, the constraint range of deposition time can be 200 s to 250 s, the constraint range of deposition rate can be 19 Å / s to 23 Å / s, the constraint range of deposition temperature can be 530 ℃ to 570 ℃, and the constraint range of chamber pressure can be 200 Tor to 400 Tor.
[0038] Step S402: Determine the combination of target parameters based on multiple constraint ranges and the target optimization model.
[0039] The target parameter combination is one of at least one set of parameter combinations that makes the value of the performance parameter fall within the target range. The parameter combination is a combination of the values of multiple process parameters. The target optimization model is used to characterize the mapping relationship between the values of multiple process parameters and the values of the performance parameter.
[0040] Specifically, changing the value of any one of the multiple process parameters can alter the thickness uniformity of the oxide layer. Variations in the wafer-nozzle spacing affect the gas flow distribution and diffusion efficiency of the reactive gases. Uneven gas flow distribution can lead to thicker edges and thinner centers in the silicon oxide film, while reduced diffusion efficiency can cause a decrease in thickness uniformity due to a larger gas concentration gradient. Adjusting the spacing can balance gas flow uniformity and deposition efficiency. Deposition time and deposition rate affect the thickness of the silicon oxide film. Adjusting at least one of these factors can achieve a preset thickness. Changes in the deposition rate affect the concentration distribution of the reactive gases; uneven concentration distribution can worsen the thickness uniformity of the silicon oxide film. Deposition temperature affects the reactivity of the reactive gases, thus affecting the deposition rate. Changes in chamber pressure affect gas flow uniformity, which in turn affects thickness uniformity.
[0041] It should be understood that variations in the values of multiple process parameters also affect the growth quality of the deposited silicon oxide film. However, by limiting multiple constraint ranges (constraint conditions), the thickness uniformity of the silicon oxide film can be optimized without affecting the growth quality.
[0042] In some embodiments, based on multiple constraint ranges and target optimization models, a multi-objective optimization algorithm can be used to obtain at least one set of parameter combinations and the values of performance parameters corresponding to each set of parameter combinations. Then, one of the parameter combinations whose performance parameter values fall within the target range is determined as the target parameter combination.
[0043] In other embodiments, multiple sets of parameter combinations can be randomly generated under the constraints of multiple ranges. Then, each set of parameter combinations is input into the target optimization model. The value of the performance parameter corresponding to each set of parameter combinations is determined according to the output of the target optimization model. The parameter combination corresponding to the value of the multiple performance parameters that is within the target range is determined as the target parameter combination.
[0044] For example, the performance parameter is the thickness range, and multiple process parameters include the wafer-to-nozzle spacing, deposition time, deposition rate, deposition temperature, and chamber pressure. If the target range is less than or equal to 200 mils, then the target parameter combination can include: S 289 mils, t 281s, R 20 Å / s, T 539℃, and P 240 Tor. Here, S represents the wafer-to-nozzle spacing, t represents the deposition time, R represents the deposition rate, T represents the deposition temperature, and P represents the chamber pressure.
[0045] Step S403: Control the deposition equipment to grow oxide on the wafer surface according to the target parameter combination to obtain the target oxide layer.
[0046] The target oxide layer is an oxide layer whose performance parameters are within the target range.
[0047] Specifically, after obtaining the target parameter combination, the process parameters of the SACVD deposition equipment are adjusted according to the target parameter combination, and then a reaction gas is introduced to grow oxide on the wafer surface to obtain the target oxide layer.
[0048] like Figure 5 As shown, the wafer 14 is first placed on the heater 12 in the reaction chamber 15 of the SACVD deposition equipment, and then the reaction gas is introduced into the reaction chamber 15. The process of growing a silicon oxide thin film by SACVD can be as follows: ① The gas mixture participating in the reaction is transported to the vicinity of the wafer (substrate), ② The reactant diffuses to the substrate surface through the boundary layer, ③ The reactant molecules are adsorbed on the substrate surface, ④ Surface reaction, including chemical decomposition and reaction, migration of adsorbed molecules on the substrate surface, and growth of silicon oxide thin film on the substrate surface, ⑤ Residual reactant and by-products are desorbed from the substrate surface, ⑥ Residual reactant and by-products are removed from the substrate surface and return to the main gas flow through the boundary layer by diffusion, ⑦ Residual reactant and by-products are discharged from the deposition area.
[0049] The oxide layer uniformity control method provided in this embodiment, after obtaining the target range of performance parameters and the constraint range of each process parameter among multiple process parameters, determines a combination of multiple process parameter values (target parameter combination) that makes the value of the first parameter fall within the target range based on multiple constraint ranges and a target optimization model. Then, it controls the deposition equipment to grow oxide on the wafer surface according to the target parameter combination to obtain the target oxide layer. This invention, under the constraints of multiple constraint ranges, adjusts the values of multiple process parameters in conjunction with a target optimization model. By optimizing the values of multiple process parameters to make the performance parameters characterizing thickness uniformity fall within the target range, it can improve the thickness uniformity of the final oxide layer without changing the SACVD deposition equipment structure, reducing production costs and improving equipment compatibility.
[0050] This embodiment also provides another method for controlling oxide layer uniformity, which can be used in a computer system. Figure 6 This is a flowchart of another method for controlling oxide layer uniformity according to an embodiment of the present invention, such as... Figure 6 As shown, the process includes the following steps: Step S601: Obtain the target range of the performance parameters and the constraint range of each process parameter among multiple process parameters.
[0051] Please see details Figure 4 Step S401 of the illustrated embodiment will not be described again here.
[0052] Step S602: Obtain the sample database and response surface model.
[0053] The sample database includes multiple samples, each consisting of a set of actual parameter combinations and the corresponding performance parameter values. Samples can be determined experimentally or obtained from a database of factors affecting silicon oxide film growth. This database includes several sets of factors influencing silicon oxide film growth and corresponding uniformity values for each set. Each set of influencing factors includes spacing (S), deposition time (t), deposition rate (R), deposition temperature (T), and chamber pressure (P).
[0054] The response surface model is a mathematical model that describes the functional relationship between a response variable (dependent variable) and multiple independent variables, enabling process optimization. In this embodiment, the dependent variable is a performance parameter, and the independent variables are process parameters. When multiple process parameters include the wafer-to-nozzle spacing, deposition time, deposition rate, deposition temperature, and chamber pressure, the response surface model is a five-element nonlinear response surface model. Taking the performance parameter as the thickness range value as an example, the expression of the five-element nonlinear response surface model can be shown in formula (3):
[0055] In the formula, Indicates the total thickness value; Indicates spacing The linear term coefficients are used to characterize the spacing. The influence of variables on thickness range; Indicates deposition time The linear coefficients are used to characterize the deposition time. The influence of variables on thickness range; Indicates deposition rate The linear coefficients are used to characterize the deposition rate. The influence of variables on thickness range; Indicates deposition temperature The linear coefficients are used to characterize the deposition temperature. The influence of variables on thickness range; Indicates chamber pressure The linear coefficients are used to characterize the chamber pressure. The influence of variables on thickness range; Indicates spacing and sedimentation time The second-order interaction coefficients are used to characterize the spacing. and sedimentation time The influence of two variables on the thickness range; Indicates spacing and deposition temperature The second-order interaction coefficients are used to characterize the spacing. and deposition temperature The influence of two variables on the thickness range; This represents a higher-order interaction term, used to characterize the combined effect of three or more process parameters on the thickness range.
[0056] In this embodiment, the five process parameters are optimized in a coordinated manner, which can more accurately adjust the thickness uniformity of the oxide layer.
[0057] Step S603: Train the response surface model based on multiple samples until the trained response surface model reaches the convergence condition.
[0058] Step S604: The response surface model that has reached the convergence condition is determined as the target optimization model.
[0059] For example, based on multiple samples, a Box-Behnken experimental design using response surface methodology can be employed to generate a training dataset. This training dataset includes multiple training samples selected from the existing dataset. Box-Behnken is a statistical method for optimizing experimental conditions in multivariate systems, specifically for selecting experimental points. Through Box-Behnken, representative experimental points (samples) can be selected from multiple samples, avoiding redundancy in full-factor experiments, covering key regions of the parameter space with the fewest possible experiments, and ensuring sufficient data information.
[0060] After obtaining multiple training datasets, the nonlinear response surface can be fitted using Gaussian Process Regression (GPR). This involves training the response surface model using Gaussian Process Regression on multiple training datasets until a convergence condition is met. The convergence condition can be a preset number of iterations or the confidence interval (reliability) of the prediction results meeting design requirements.
[0061] Specifically, firstly, a kernel function (such as a squared exponential kernel) is defined to describe the similarity between different combinations of input parameters (multiple process parameters); then, based on the training dataset, the parameters of the kernel function are learned through Bayesian inference to establish a probabilistic mapping relationship between multiple process parameters and performance parameters, and finally, a nonlinear response surface (objective optimization model) is generated.
[0062] In this embodiment, Gaussian process regression (GPR) can capture more subtle parameter interaction effects through kernel functions, improve the prediction accuracy of the defined target optimization model, and thus more reliably guide the subsequent process parameter optimization process.
[0063] Step S605: Determine the combination of target parameters based on multiple constraint ranges and the target optimization model.
[0064] Specifically, step S605 includes: Step S6051: Based on multiple constraint ranges and target optimization models, the values of multiple process parameters are optimized using a multi-objective optimization algorithm to obtain multiple sets of parameter combinations.
[0065] Step S6052: Determine the parameter combination that minimizes the value of the performance parameter among multiple parameter combinations as the target parameter combination.
[0066] Among them, multi-objective optimization algorithms include multi-objective genetic algorithm (NSGA-II) or multi-objective particle swarm optimization algorithm. Multi-objective genetic algorithm (NSGA-II) is a genetic algorithm based on the concept of Pareto optimality, used to solve multi-objective optimization problems. It maintains the diversity of the population through non-dominated sorting and crowding calculation, and adopts an elitist strategy to retain superior individuals.
[0067] Taking a multi-objective optimization algorithm as an example, step S6051 above may include the following steps: Step a1: Generate an initial population based on multiple constraint ranges.
[0068] The initial population consists of multiple individuals, each corresponding to a configuration scheme of multiple process parameters that satisfy the constraints.
[0069] Step a2: Determine the fitness of each individual among the multiple individuals based on the fitness function.
[0070] The fitness function is the objective optimization model. After obtaining multiple individuals, the individuals are input into the fitness function, and the output of the fitness function is the fitness of the individual.
[0071] Step a3: Based on the fitness of each individual, perform crossover, mutation, and elite operations on the initial population to generate the next generation population.
[0072] Specifically, after determining the fitness of each individual, the NSGA-II algorithm can be used to perform non-dominated sorting based on fitness, dividing the individuals in the initial population into different levels, such as selecting individuals from the Pareto front, level 2, level 3, ..., level n. The Pareto front, representing level 1, refers to a set of optimal solutions that cannot be further improved; individuals from the Pareto front are preferentially selected for the next generation of the population.
[0073] After dividing the individuals in the initial population into different levels, the candidate solutions (individuals) with higher levels can be retained as parents. Then, through crossover (randomly combining the parameters of the parents to generate new candidate solutions), mutation (randomly fine-tuning the parameters of some offspring) and elitist strategy (retaining the candidate solutions with the Pareto optimal frontier in the previous generation), 100 new individuals (the next generation population) are generated from multiple individuals in the initial population (taking 100 as an example).
[0074] Step a4: Determine whether the next generation population has reached the convergence condition or the maximum number of iterations.
[0075] Step a5: When the next generation population reaches the convergence condition or the maximum number of iterations, multiple individuals in the next generation population are identified as multiple sets of parameter combinations.
[0076] If the convergence condition or the maximum number of iterations is not met, return to step a2 to continue the optimization iteration.
[0077] In this embodiment, the multi-objective genetic algorithm has good global search capabilities, which can quickly find the Pareto optimal solution set that makes the performance parameters within the target range, improve the efficiency of parameter determination, and avoid getting trapped in local optima.
[0078] Multi-objective particle swarm optimization is a multi-objective optimization method based on particle swarm optimization. It finds the optimal solution by simulating the foraging behavior of bird flocks. The core of it is to maintain an external archive to store non-dominated solutions, that is, solutions that are not dominated by other solutions on any objective.
[0079] The multi-objective particle swarm optimization (PSO) algorithm mainly includes the following steps: Initialization: Randomly generate an initial particle swarm and calculate the fitness value of each particle, saving non-dominated solutions to an external archive; Update particle velocity and position: Update the velocity and position of particles based on individual optimal solutions and the global optimal solution; Update external archive: Add a new generation of non-dominated solutions to the external archive and remove solutions dominated by the new solutions; Archive truncation operation: When the size of the external archive exceeds a preset value, delete redundant solutions using crowding sorting and grid partitioning methods to maintain the size of the archive and the diversity of solutions; Iteration termination judgment: If the iteration number or other termination conditions are met, output the set of non-dominated solutions in the external archive, i.e., obtain multiple sets of parameter combinations.
[0080] In some embodiments, in order to improve the efficiency of determining the target parameter combination, the value of at least one process parameter can be fixed, and then the values of the non-fixed process parameters can be optimized by a multi-objective optimization algorithm to obtain multiple sets of parameter combinations.
[0081] Specifically, in the process of optimizing the values of multiple process parameters to improve the uniformity of oxide layer thickness, the deposition rate will also decrease to some extent as the thickness range decreases. Therefore, in addition to minimizing the thickness range, the optimization objective is to maximize the deposition rate.
[0082] It should be noted that in this embodiment, the adjustable range (constraint range) of the spacing is usually large (180 mils-300 mils), and the uniformity of the silicon oxide film thickness changes significantly within the adjustable range of the spacing. Therefore, in this embodiment, the spacing is the main control variable, and other process parameters are secondary control variables that coordinate with each other. Furthermore, in normal film uniformity control, the adjustable range of other variables besides the spacing is small, and fine-tuning is usually performed to accommodate uniformity changes.
[0083] Step S606: Control the deposition equipment to grow oxide on the wafer surface according to the target parameter combination to obtain the target oxide layer.
[0084] Please see details Figure 4 Step S403 of the illustrated embodiment will not be described again here.
[0085] The oxide layer uniformity control method provided in this embodiment trains a response surface model using Gaussian process regression. The kernel function of Gaussian process regression can capture more subtle parameter interaction effects, thereby improving the prediction accuracy of the defined target optimization model. Moreover, based on multiple constraint ranges and target optimization models, the values of multiple process parameters are optimized through a multi-objective optimization algorithm, which can more efficiently and accurately determine the combination of target parameters.
[0086] The following uses several second process parameters, including spacing S, deposition time t, deposition rate R, deposition temperature T, and chamber pressure P, as examples to illustrate the process of improving the thickness uniformity of silicon oxide films formed by SACVD deposition equipment. Thickness uniformity is characterized by the thickness range value.
[0087] The process of improving the uniformity of silicon oxide film thickness may include the following steps: S1. Initialize the current process state Specifically, initialize (reset) the current process status of the SACVD deposition equipment.
[0088] S2. Call the optimization parameters (optimize_parameters) S21: Call the database of factors affecting silicon oxide film growth. The database includes several groups of factors affecting silicon oxide film growth and the corresponding silicon oxide film uniformity values. Each group of factors includes spacing S, deposition time t, deposition rate R, deposition temperature T, and chamber pressure P.
[0089] S3. Training to form a target optimization model S31: Invoke the five-element nonlinear response surface model:
[0090] S32: Use a database of factors affecting silicon oxide film growth to train a five-element nonlinear response surface model to form a target optimization model. The target optimization model can generate the optimal values of the influencing factors based on the uniformity values of the silicon oxide film.
[0091] S4. Run the objective optimization model under constraints (constraint range) to determine the optimal parameter combination.
[0092] S41: Determine the constraint range of influencing factors. The target range of influencing factors includes the following: spacing range: 180 mils-300 mils, deposition time range: 200 s-250 s, deposition rate range: 19 Å / s-23 Å / s, temperature range: 530℃-570℃, and pressure range: 200 Tor-400 Tor.
[0093] S42: Determine the target uniformity value of the silicon oxide film. Based on the target uniformity value of the silicon oxide film and the target optimization model, determine at least one set of silicon oxide film growth influencing factors that meet the target range of influencing factors.
[0094] The target uniformity values and influencing factors of the silicon oxide thin film can be input by the user into the computer system via a human-computer interaction interface, allowing the computer system to obtain the relevant parameters. When inputting parameters, the user interface (UI) can be as follows: Figure 7 As shown; after inputting the target range, the values of various process parameters can be obtained through the target optimization model, and these values can be displayed on the UI interface. At this point, the user interface can be as follows: Figure 8 As shown.
[0095] In some examples, one of the multiple process parameters can be fixed (e.g., a fixed temperature of 540℃), while the other unfixed process parameters are output, making the application more flexible. The parameters input to the user interface and the parameters output to the user interface can be shown in Table 1: Table 1. Values of each parameter
[0096] Among them, uniformity is positively correlated with the thickness range (THK_Range), and uniformity is determined based on the thickness range corresponding to the output parameter.
[0097] This embodiment also provides a device for controlling the uniformity of the oxide layer, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.
[0098] This embodiment provides a device for controlling the uniformity of the oxide layer, such as... Figure 9 As shown, it includes: The acquisition module 901 is used to acquire the target range of the performance parameters and the constraint range of each process parameter among multiple process parameters, wherein the performance parameters are parameters characterizing the uniformity of oxide layer thickness, and the process parameters are parameters affecting the uniformity of oxide layer thickness. The parameter determination module 902 is used to determine the target parameter combination based on multiple constraint ranges and the target optimization model. The target parameter combination is one of multiple parameter combinations that make the performance parameter values fall within the target range. The parameter combination is a combination of the values of multiple process parameters. The target optimization model is used to characterize the mapping relationship between the values of multiple process parameters and the values of performance parameters. The control module 903 is used to control the deposition equipment to grow oxides on the wafer surface according to the target parameter combination to obtain the target oxide layer.
[0099] In some alternative implementations, the parameter determination module 902 includes: The optimization unit is used to optimize the values of multiple process parameters based on multiple constraint ranges and objective optimization models, and to obtain multiple sets of parameter combinations through multi-objective optimization algorithms. The parameter determination unit is used to determine the parameter combination that minimizes the value of the performance parameter among multiple parameter combinations as the target parameter combination.
[0100] In some alternative implementations, the multi-objective optimization algorithm includes a multi-objective genetic algorithm or a multi-objective particle swarm optimization algorithm.
[0101] In some alternative embodiments, the apparatus further includes: The model acquisition module is used to acquire a sample database and a response surface model. The sample database includes multiple samples, each consisting of a set of actual parameter combinations and the values of the performance parameters corresponding to those combinations. The training module is used to train the response surface model based on multiple samples until the trained response surface model reaches the convergence condition. The model determination module is used to determine the response surface model that has reached the convergence condition as the target optimization model.
[0102] In some alternative implementations, multiple process parameters include wafer-to-nozzle spacing, deposition time, deposition rate, deposition temperature, and chamber pressure, and the response surface model is a quinary nonlinear response surface model.
[0103] In some alternative implementations, the spacing is constrained to 180 mils to 300 mils, the deposition time is constrained to 200 s to 250 s, the deposition rate is constrained to 19 Å / s to 23 Å / s, the deposition temperature is constrained to 530 ℃ to 570 ℃, and the chamber pressure is constrained to 200 Tor to 400 Tor.
[0104] The oxide layer uniformity control device provided in this embodiment of the invention can execute the oxide layer uniformity control method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.
[0105] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.
[0106] The following is a detailed reference. Figure 10 This diagram illustrates a suitable structural schematic for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from memory 1008 into random access memory (RAM) 1003. The RAM 1003 also stores various programs and data required for the operation of the electronic device. The processor 1001, ROM 1002, and RAM 1003 are interconnected via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.
[0107] Typically, the following devices can be connected to the I / O interface 1005: input devices 1006 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 1007 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; memory devices 1008 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 10 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.
[0108] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 1009, or installed from a memory 1008, or installed from a ROM 1002. When the computer program is executed by the processor 1001, it performs the functions defined in the oxide layer uniformity control method of the embodiments of the present invention.
[0109] Figure 10 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.
[0110] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium after being downloaded via a network. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the oxide layer uniformity control method shown in the above embodiments is implemented.
[0111] A portion of this invention can be applied to computer program products, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installation program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.
[0112] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.
Claims
1. A method for controlling the uniformity of an oxide layer, characterized in that, include: Obtain the target range of the performance parameters and the constraint range of each process parameter among multiple process parameters, wherein the performance parameters are parameters characterizing the uniformity of oxide layer thickness, and the process parameters are parameters affecting the uniformity of oxide layer thickness; Based on multiple constraint ranges and target optimization models, a target parameter combination is determined, wherein the target parameter combination is one of at least a set of parameter combinations that makes the value of the performance parameter fall within the target range, the parameter combination is a combination of the values of the multiple process parameters, and the target optimization model is used to characterize the mapping relationship between the values of the multiple process parameters and the values of the performance parameter; The deposition equipment is controlled to grow oxides on the wafer surface according to the target parameter combination to obtain the target oxide layer.
2. The method for controlling the uniformity of the oxide layer according to claim 1, characterized in that, The step of determining the combination of target parameters based on multiple constraint ranges and target optimization models includes: Based on the multiple constraint ranges and the target optimization model, the values of the multiple process parameters are optimized using a multi-objective optimization algorithm to obtain multiple sets of parameter combinations; The parameter combination that minimizes the value of the performance parameter among the multiple parameter combinations is determined as the target parameter combination.
3. The method for controlling the uniformity of the oxide layer according to claim 2, characterized in that, The multi-objective optimization algorithm includes a multi-objective genetic algorithm or a multi-objective particle swarm optimization algorithm.
4. The method for controlling the uniformity of the oxide layer according to any one of claims 1 to 3, characterized in that, Before determining the target parameter combination, the method for controlling oxide layer uniformity further includes: Obtain a sample database and a response surface model, wherein the sample database includes multiple samples, and each sample consists of a set of actual parameter combinations and the values of the performance parameters corresponding to the actual parameter combinations; The response surface model is trained based on the multiple samples until the trained response surface model reaches the convergence condition. The response surface model that meets the convergence condition is determined as the target optimization model.
5. The method for controlling the uniformity of the oxide layer according to claim 4, characterized in that, The multiple process parameters include the wafer-to-nozzle spacing, deposition time, deposition rate, deposition temperature, and chamber pressure, and the response surface model is a quinary nonlinear response surface model.
6. The method for controlling the uniformity of the oxide layer according to claim 5, characterized in that, The spacing is constrained to 180 mils to 300 mils, the deposition time is constrained to 200 s to 250 s, the deposition rate is constrained to 19 Å / s to 23 Å / s, the deposition temperature is constrained to 530 ℃ to 570 ℃, and the chamber pressure is constrained to 200 Tor to 400 Tor.
7. A device for controlling the uniformity of an oxide layer, characterized in that, include: The acquisition module is used to acquire the target range of the performance parameters and the constraint range of each process parameter among multiple process parameters, wherein the performance parameters are parameters characterizing the uniformity of oxide layer thickness, and the process parameters are parameters affecting the uniformity of oxide layer thickness. The parameter determination module is used to determine a target parameter combination based on multiple constraint ranges and a target optimization model. The target parameter combination is one of multiple parameter combinations that make the value of the performance parameter fall within the target range. The parameter combination is a combination of the values of the multiple process parameters. The target optimization model is used to characterize the mapping relationship between the values of the multiple process parameters and the values of the performance parameter. The control module is used to control the deposition equipment to grow oxides on the wafer surface according to the target parameter combination to obtain the target oxide layer.
8. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the method for controlling the uniformity of the oxide layer according to any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the method for controlling the uniformity of the oxide layer according to any one of claims 1 to 6.
10. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the method for controlling the uniformity of the oxide layer according to any one of claims 1 to 6.