Ion implantation parameter optimization method and device, electronic equipment and storage medium
By training a network model and optimizing ion implantation parameters using deep learning, the problem of poor doping quality during ion implantation was solved, achieving precise doping control and improving the ion implantation effect.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUNAN UNIV
- Filing Date
- 2023-08-07
- Publication Date
- 2026-05-19
AI Technical Summary
Existing technologies make it difficult to precisely control the doping quality during ion implantation, resulting in poor collision processes and travel paths of ions in silicon carbide, which affects the doping effect.
Ion implantation was simulated by training a network model, and the ion implantation parameters were optimized by combining a deep learning network. The parameters were fine-tuned by comparing the simulation results with the actual operation results. A multi-physics coupling model of cluster dynamics and charge model was established to optimize the ion implantation process.
It enables precise control of the ion implantation process, significantly improving the quality of ion doping and enhancing the doping effect.
Smart Images
Figure CN116936008B_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of semiconductor materials technology, specifically relating to an ion implantation parameter optimization method, apparatus, electronic device, and storage medium. Background Technology
[0002] Silicon carbide (SiC), as the most mature wide-bandgap semiconductor material currently available, is one of the core materials of the third generation of semiconductors after Si and GaAs. Compared with traditional semiconductor materials represented by Si and GaAs, silicon carbide has advantages such as wide bandgap, high breakdown field strength, high thermal conductivity, high saturated electron drift velocity, and high bonding energy. Its excellent performance can meet the requirements of modern electronic technology for high temperature, high frequency, high power, high voltage, and radiation resistance, and is therefore regarded as one of the most promising semiconductor materials in the field of microelectronics.
[0003] Ion implantation is a crucial step in the fabrication of integrated circuits. Dopant ions, implanted into a target under a high electric field, lose energy after colliding with atoms and electrons, eventually ceasing their motion—a random process. Since solids are composed of atomic nuclei and electrons, ions implanted into the solid must collide with these nuclei and electrons. During these collisions, the ions' direction of motion is deflected, and they lose energy. Therefore, ions with initial energy, entering the target, undergo a series of collisions with atomic nuclei and electrons in the substrate, losing energy and traversing a highly tortuous path before finally settling within the substrate, ultimately stopping at a specific point on the target. The collision process and travel path of ions in silicon carbide significantly impact the doping quality. Summary of the Invention
[0004] The purpose of this application is to provide an ion implantation parameter optimization method, apparatus, electronic device, and storage medium to improve the doping quality of ion implantation.
[0005] According to a first aspect of the embodiments of this application, an ion implantation parameter optimization method is provided, the method comprising:
[0006] Obtain the initial injection parameters of the target material;
[0007] Based on the initial injection parameters, the trained network model is used to repeatedly simulate ion implantation to obtain simulation results.
[0008] The simulation results are compared with the actual running results, and the initial injection parameters are adjusted based on the comparison results to obtain the optimized injection parameters.
[0009] In some optional embodiments of this application, the trained network model is obtained by training the following method:
[0010] Obtain prior information, which is a set including reaction events and parameters;
[0011] Based on prior information and using a simulation system to simulate ion implantation, the initial implantation vacancy and ion concentration distribution are obtained;
[0012] A multiphysics coupling model, including a cluster dynamics model and a charge model, was established using the initial injected vacancy and ion concentration distribution. The partial differential equations of the multiphysics coupling model were then solved and trained using a deep learning network to obtain the trained network model.
[0013] In some optional embodiments of this application, obtaining prior information includes:
[0014] Obtain the geometric parameters of the system to be simulated, including the incident particle parameters and the target layer material parameters;
[0015] Based on the target material parameters, a target simulation region is established;
[0016] By using the incident particle parameters and the target layer simulation region, a set of reaction events and parameters is established to obtain prior information.
[0017] In some optional embodiments of this application, obtaining the geometric parameters of the system to be simulated includes:
[0018] Obtain simulation parameters.
[0019] The incident particle parameters are created based on the simulation parameters and using the Ion class;
[0020] Target material parameters are created based on simulation parameters and using the Target class.
[0021] In some optional embodiments of this application, a multiphysics coupling model including a cluster dynamics model and a charge model is established using the initial implanted vacancy and ion concentration distribution, including:
[0022] The reaction events between the incident particle and the target layer ions are generated by utilizing the initial injected vacancies and ion concentration distribution.
[0023] In some optional embodiments of this application, the simulation results are compared with the actual running results, and the initial injection parameters are tuned based on the comparison results to obtain optimized injection parameters, including:
[0024] Using actual operating data of the ion implanter as prior information, a parameter optimization dataset was established.
[0025] Based on the parameter tuning dataset, a solution network model is established by combining fully connected and convolutional network structures.
[0026] The training parameters are set and trained according to the network model to achieve convergence, thus obtaining a converged model.
[0027] The initial injection parameters are tuned using a convergence model to obtain optimized injection parameters.
[0028] According to a second aspect of the embodiments of this application, an ion implantation method is provided, which uses the ion implantation parameter optimization method shown in any embodiment of the first aspect to obtain optimized implantation parameters, and performs ion implantation according to the optimized implantation parameters.
[0029] According to a third aspect of the embodiments of this application, an ion implantation parameter optimization apparatus is provided, the apparatus comprising:
[0030] The acquisition module is used to acquire the initial injection parameters of the target material;
[0031] The simulation module is used to repeatedly simulate ion implantation using a trained network model based on initial injection parameters, and obtain simulation results.
[0032] The optimization module compares the simulation results with the actual running results, and optimizes the initial injection parameters based on the comparison results to obtain the optimized injection parameters.
[0033] According to a fourth aspect of the embodiments of this application, an electronic device is provided, which may include:
[0034] processor;
[0035] Memory used to store processor-executable instructions;
[0036] The processor is configured to execute instructions to implement the ion implantation parameter optimization method as shown in any embodiment of the first aspect or the ion implantation method as shown in the embodiment of the second aspect.
[0037] According to a fifth aspect of the embodiments of this application, a storage medium is provided, wherein when instructions in the storage medium are executed by a processor of an information processing device or a server, the information processing device or server implements the ion implantation parameter optimization method as shown in any embodiment of the first aspect or the ion implantation method as shown in the embodiments of the second aspect.
[0038] The above-mentioned technical solution of this application has the following beneficial technical effects:
[0039] The method described in this application first uses a trained network model to perform simulated injection to obtain simulation results. Then, it uses actual production process detection data and simulation results to train a deep learning network model, thereby obtaining optimized ion implantation parameters. This method can precisely control the ion implantation process, thereby greatly improving the ion doping quality. Attached Figure Description
[0040] Figure 1 This is a flowchart illustrating an exemplary embodiment of the ion implantation parameter optimization method in this application;
[0041] Figure 2 This is a schematic diagram of the ion implantation parameter optimization device in an exemplary embodiment of this application;
[0042] Figure 3 This is a schematic diagram of the electronic device structure in an exemplary embodiment of this application;
[0043] Figure 4 This is a schematic diagram of the hardware structure of an electronic device in an exemplary embodiment of this application. Detailed Implementation
[0044] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to specific embodiments and accompanying drawings. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of this application. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.
[0045] The accompanying drawings illustrate layer structure diagrams according to embodiments of this application. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.
[0046] Obviously, the described embodiments are only a part of the embodiments of this application, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0047] In the description of this application, it should be noted that the terms "first", "second", and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0048] Furthermore, the technical features involved in the different embodiments of this application described below can be combined with each other as long as they do not conflict with each other.
[0049] The following description, in conjunction with the accompanying drawings, details the ion implantation parameter optimization method, apparatus, electronic device, and storage medium provided in this application through specific embodiments and application scenarios.
[0050] like Figure 1As shown, in a first aspect of the embodiments of this application, an ion implantation parameter optimization method is provided, which may include:
[0051] S110: Obtain the initial injection parameters of the target material;
[0052] S120: Based on the initial injection parameters, the trained network model is used to repeatedly simulate ion implantation to obtain simulation results;
[0053] S130: Compare the simulation results with the actual running results, and adjust the initial injection parameters based on the comparison results to obtain the optimized injection parameters.
[0054] The method in this embodiment first uses a trained network model to simulate ion implantation and obtain simulation results. Then, it uses actual production process detection data and simulation results to train a deep learning network model, thereby obtaining optimized ion implantation parameters. This method can precisely control the ion implantation process, thereby greatly improving the ion doping quality.
[0055] To illustrate this more clearly, the steps described above will be explained in detail below:
[0056] The first step is S110: Obtain the initial injection parameters of the target material.
[0057] The initial injection parameters obtained in this step can be imported custom input parameter files. The geometric parameters of the system to be simulated are then read, including the atomic composition and lattice structure of the target material and the energy, orientation, and distribution of the incident particles. For example, the parser function in the initialization module reads the user-defined parameter input file, performs bytecode conversion, and saves the processed parameters to a list.
[0058] For example, each parameter in the input file is read into a list by the parser, including the parameter name and value before and after the "=" sign, and a parameter list is generated. The data in the list is then read using the `read_initial_parameters` function in the input module, and the parameters are initialized. Parameters not set in the input will use default values.
[0059] The parameter list can be read, and the values of user-defined parameters can be assigned to the corresponding variables using regular expressions for parameter names. Undefined values will be used.
[0060] The main parameters may include the number, element, energy, incident angle, azimuth angle, and concentration of injected ions, the thickness of the injected target layer, the number of layers, the elements in each layer, and the elemental composition, as well as other simulation-related parameters.
[0061] Then, the simulation system is initialized using the parameter table, the target layer simulation region is established, and the incident ions are initialized.
[0062] For example, based on the initialized simulation parameters, incident ion parameters are created using the Ion class, including the initialized ion coordinates, velocity, ordinal number, and element-related parameters. Based on the initialized incident ion element parameters, the corresponding element data in the element table file element.csv is read from the get_ion_element_by_symbol method of the Element class, and an Element class parameter is created, including the element's symbol, ordinal number, atomic mass, and density. The Element class, defined by the class function, uses the read element symbol for regularization and maps it to the data table. The process involves reading all relevant parameters from the element rows. Position and velocity parameters of the incident ions are created and initialized using the Position and Velocity classes. This step uses the Position and Velocity classes (defined by class functions) to create a three-dimensional solution domain. A Cartesian coordinate system is used to define the solution domain, dividing the Position class into three directions (x, y, z) and initializing them. Similarly, the Velocity class is divided into three directions (u, v, w) and initialized. Based on the element class parameters, position parameters, and velocity parameters of the incident ions, the overall incident ion parameters are created using the Ion class. For example, the previously read parameters can be assigned to the corresponding elements using the Ion class (defined by class functions), with the Position and Velocity classes included as subclasses, and ion IDs set for counting.
[0063] Based on the initialized simulation parameters, target layer parameters are created using the `Target` class, including the parameters of each layer and the overall layer thickness. For example, based on the initialized target layer element parameters, the `get_ion_element_by_symbol` method in the `Element` class is used to look up the corresponding element data in the `element.csv` file, and an `Element` class parameter is created, including the element's symbol, index, atomic mass, and density. Using the `Element` class defined by the class function, the read element symbols are regularized and mapped to the element rows in the data table, and all relevant parameters within each row are read. Based on the incident ion, the element mixture parameters for one layer of the target layer are created using the `Compound` class, including the proportion of each element, element parameters, and element threshold shifts. For example, using the `Compound` class defined by the class function, the previously read relevant parameters are assigned to the corresponding elements, including element proportions and threshold shifts. Based on the target layer single-layer mixture parameters, target layer parameters are created using the `Target` class, including the mixture, thickness, and layer density of each layer. For example, the Target class, defined through a class function, includes the Layer class as a subclass, representing the overall target layer, and adds information such as target layer thickness. Based on the individual parameters of the target layer, the Layer class creates parameters for each individual layer within the target layer, including the individual target layer parameters and target layer density. Similarly, the Layer class, defined through a class function, can include the Compound class as a subclass, representing a single target layer, and adding information such as target layer thickness and layer density.
[0064] Next, using the initialized incident particles and target layer, ions are emitted into the target layer. In each step after entering the target layer, the ions interact with the atoms in the target layer through processes such as scattering, stopping power, and energy loss. The probability of each interaction is calculated from the defined physical model using the Monte Carlo method.
[0065] For example, based on the initialized incident particles and target layer, reaction events involving clusters are generated by using known reaction principles and cluster-related reaction mechanisms to generate ions in the incident particles and target layer. These events include both non-cluster events and cluster-related events. This process is mainly based on point defect recombination, kick-out mechanisms, Frank-Turnbull mechanisms, and cluster-related reaction mechanisms as theoretical references to establish corresponding reaction events. Based on the initialized incident particles and target layer, all relevant parameters are categorized into material-related and defect-related parameters and organized into a parameter list. The two main categories of parameters are: material parameters, used to characterize material properties and distinguish different materials, such as lattice constants and dielectric constants; and defect parameters, used to characterize different defect types and physical properties, such as defect formation energy and migration energy.
[0066] Next, using the reaction events and parameter set as prior information, ions are emitted into the target layer. In each step after entering the target layer, the ions interact with the atoms in the target layer through processes such as scattering, stopping power, and energy loss. The probability of each interaction is calculated from the defined physical model using the Monte Carlo method to obtain the initial injected vacancy and ion concentration distribution.
[0067] For example, based on the initialized incident particles and target layer, the ion implantation simulation begins through the `run_simulation` module. The simulation can be started with a loop based on the previously set target layer and incident ions, and the number of incident particles. The `simulate_ions_stopping` function of the simulation module checks the ion energy. If the ion energy is less than the energy required for movement, the velocity becomes 0; if the ion energy is sufficient for movement, a displacement is performed. This process checks the ion energy; if the energy is insufficient, the displacement is stopped; if sufficient, the displacement continues. The `simulate_collision` function of the simulation module calculates ion collisions, using the Monte Carlo binary collision approximation theory to calculate the direction and velocity of the ions after the collision, and updates their state to obtain the initial implanted vacancies and ion concentration distribution.
[0068] Finally, using the initial distribution generated during the injection process, a multiphysics coupling model including a cluster dynamics model and a charge model is established. The partial differential equation model of the multiphysics coupling is then trained and solved using a deep learning network.
[0069] For example, based on the initial distribution, a multiphysics coupling model including a cluster dynamics model and a charge model is established, and the boundary conditions of its coupling field are determined. Based on the physical fields that occur during ion implantation, such as cluster dynamics and charge model, the PDE equations are coupled to establish a multiphysics coupling model. Then, the boundary condition set of the entire solution domain is established through the boundary conditions of each physical field.
[0070] Based on the definition of PDE equations by coupled physical fields, a solution geometric domain is established, boundary conditions are established, and initial conditions are set. The PDE equations are initialized according to the physical fields and conditions to be solved. Multiple PDE equation sets and boundary condition sets can be characterized by a program to establish PDE equations and a geometric solution domain including boundary judgment.
[0071] Based on the PDE equation, a solution network model is established by combining fully connected neural networks and Fourier neural operators. The geometric solution domain is then used to build the corresponding network model through fully connected network layers and Fourier neural operator network layers.
[0072] The training parameters are set according to the solution network model, including the loss calculation method, the number of training iterations and the selection of the optimizer, and the network model is trained until it converges. The hyperparameters in this process are mainly modified based on the training results, and the optimal hyperparameter selection scheme is determined after repeated parameter tuning.
[0073] The next step is step S120: Based on the initial injection parameters, the trained network model is used to repeatedly simulate ion implantation to obtain simulation results.
[0074] This step involves repeatedly simulating a sufficient number of ions using a trained network model to obtain and analyze the ion implantation simulation results, such as the depth distribution of implanted ions, the distribution of ion energy, and the damage distribution caused by ion implantation to the target layer.
[0075] Finally, step S130: compare the simulation results with the actual running results, and adjust the initial injection parameters according to the comparison results to obtain the optimized injection parameters.
[0076] This step is based on simulation results. It uses the actual operating data monitored and recorded by the sensors on the existing ion implanter as prior information, and uses the comparison results with the simulation data to optimize the parameter set and obtain the optimized parameter list.
[0077] For example, based on the simulation results, the actual operating data monitored and recorded by the sensors on the existing ion implanter is used as prior information to establish a parameter tuning dataset. The dataset mainly tunes the relevant parameters of ion implantation, including parameters that can change the simulation results, such as incident ion concentration, direction, energy, and target layer structure. The actual operating data obtained from existing production is used as prior information for reference to generate a training dataset.
[0078] Based on the training dataset, a solution network model is established by combining fully connected, convolutional, and other network structures.
[0079] The training parameters are set according to the solution network model, including the loss calculation method, the number of training iterations and the selection of the optimizer, and the network model is trained until it converges. The hyperparameters are mainly modified based on the training results, and the optimal hyperparameter selection scheme is determined after repeated parameter tuning.
[0080] Based on the trained neural network model, the parameter set is fine-tuned using the comparison results with the simulation data to obtain an optimized parameter list. This process provides appropriate injection-related parameters such as energy, angle, and concentration based on the type of particle and the type of target layer, resulting in an AI-tuned parameter list.
[0081] In some embodiments, the trained network model is obtained by training the following method:
[0082] Obtain prior information, which is a set including reaction events and parameters;
[0083] Based on prior information and using a simulation system to simulate ion implantation, the initial implantation vacancy and ion concentration distribution are obtained;
[0084] A multiphysics coupling model, including a cluster dynamics model and a charge model, was established using the initial injected vacancy and ion concentration distribution. The partial differential equations of the multiphysics coupling model were then solved and trained using a deep learning network to obtain the trained network model.
[0085] In some embodiments, obtaining prior information includes:
[0086] Obtain the geometric parameters of the system to be simulated, including the incident particle parameters and the target layer material parameters;
[0087] Based on the target material parameters, a target simulation region is established;
[0088] By using the incident particle parameters and the target layer simulation region, a set of reaction events and parameters is established to obtain prior information.
[0089] In some embodiments, obtaining the geometric parameters of the system to be simulated includes:
[0090] Obtain simulation parameters.
[0091] The incident particle parameters are created based on the simulation parameters and using the Ion class;
[0092] Target material parameters are created based on simulation parameters and using the Target class.
[0093] In some embodiments, a multiphysics coupling model including a cluster dynamics model and a charge model is established using the initial implanted vacancy and ion concentration distribution, including:
[0094] The reaction events between the incident particle and the target layer ions are generated by utilizing the initial injected vacancies and ion concentration distribution.
[0095] In some embodiments, the simulation results are compared with the actual running results, and the initial injection parameters are tuned based on the comparison results to obtain optimized injection parameters, including:
[0096] Using actual operating data of the ion implanter as prior information, a parameter optimization dataset was established.
[0097] Based on the parameter tuning dataset, a solution network model is established by combining fully connected and convolutional network structures.
[0098] The training parameters are set and trained according to the network model to achieve convergence, thus obtaining a converged model.
[0099] The initial injection parameters are tuned using a convergence model to obtain optimized injection parameters.
[0100] According to a second aspect of the embodiments of this application, an ion implantation method is provided, which uses the ion implantation parameter optimization method shown in any embodiment of the first aspect to obtain optimized implantation parameters, and performs ion implantation according to the optimized implantation parameters.
[0101] It should be noted that the ion implantation parameter optimization method provided in this application can be executed by an ion implantation parameter optimization device, or a control module within that ion implantation parameter optimization device for executing the ion implantation parameter optimization method. This application uses an ion implantation parameter optimization device executing the ion implantation parameter optimization method as an example to illustrate the ion implantation parameter optimization device provided in this application.
[0102] like Figure 2 As shown, in a third aspect of the embodiments of this application, an ion implantation parameter optimization apparatus is provided, which may include:
[0103] The acquisition module 210 is used to acquire the initial injection parameters of the target material;
[0104] Simulation module 220 is used to repeatedly simulate ion implantation based on the initial injection parameters using a trained network model to obtain simulation results;
[0105] The optimization module 230 compares the simulation results with the actual running results, and optimizes the initial injection parameters based on the comparison results to obtain the optimized injection parameters.
[0106] The ion implantation parameter optimization device in this application embodiment can be a device, or it can be a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, mobile electronic devices can be mobile phones, tablets, laptops, PDAs, in-vehicle electronic devices, wearable devices, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), etc., while non-mobile electronic devices can be servers, network-attached storage (NAS), personal computers (PCs), televisions (TVs), ATMs, or self-service machines, etc. This application embodiment does not impose specific limitations.
[0107] The ion implantation parameter optimization device in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit it.
[0108] The ion implantation parameter optimization device provided in this application embodiment can achieve… Figure 1 The various processes implemented in the method implementation examples will not be described again here to avoid repetition.
[0109] Optionally, such as Figure 3 As shown, this application embodiment also provides an electronic device 300, including a processor 301, a memory 302, and a program or instructions stored in the memory 302 and executable on the processor 301. When the program or instructions are executed by the processor 301, they implement the various processes of the above-described ion implantation parameter optimization method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.
[0110] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.
[0111] Figure 4 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of this application.
[0112] The electronic device 400 includes, but is not limited to, components such as: radio frequency unit 401, network module 402, audio output unit 403, input unit 404, sensor 405, display unit 406, user input unit 407, interface unit 408, memory 409, and processor 410.
[0113] Those skilled in the art will understand that the electronic device 400 may also include a power supply (such as a battery) for supplying power to various components. The power supply may be logically connected to the processor 410 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.
[0114] It should be understood that, in this embodiment, the input unit 404 may include a graphics processing unit (GPU) 4041 and a microphone 4042. The GPU 4041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 406 may include a display panel 4061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit 407 includes a touch panel 4071 and other input devices 4072. The touch panel 4071 is also called a touch screen. The touch panel 4071 may include a touch detection device and a touch controller. Other input devices 4072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, joysticks, etc., which will not be described in detail here. The memory 409 can be used to store software programs and various data, including but not limited to applications and operating systems. The processor 410 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understandable that the aforementioned modem processor may not be integrated into the processor 410.
[0115] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described ion implantation parameter optimization method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.
[0116] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.
[0117] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
[0118] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0119] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.
Claims
1. A method for optimizing ion implantation parameters, characterized in that, include: Obtain the initial injection parameters of the target material; Based on the initial injection parameters, the trained network model is used to repeatedly simulate ion implantation to obtain simulation results. The simulation results are compared with the actual running results, and the initial injection parameters are tuned based on the comparison results to obtain optimized injection parameters; the trained network model is trained using the following method: Obtain prior information, which is a set including reaction events and parameters; Based on the prior information and using a simulation system to perform simulated ion implantation, the initial implantation vacancy and ion concentration distribution are obtained; A multiphysics coupling model including a cluster dynamics model and a charge model is established using the initial injected vacancy and ion concentration distribution. The partial differential equations of the multiphysics coupling model are solved and trained using a deep learning network to obtain a trained network model. The trained network model is obtained through the following steps: Based on the initial distribution, a multiphysics coupling model including a cluster dynamics model and a charge model is established, and the boundary conditions of the coupling field are determined. Based on the physical fields that occur during ion implantation in the cluster dynamics and charge model, the PDE equations are coupled to establish a multiphysics coupling model. Then, the boundary condition set of the entire solution domain is established through the boundary conditions of the physical fields. Based on the definition of PDE equations by coupled physical fields, a solution geometric domain is established, boundary conditions are established, and initial conditions are set. The PDE equations are initialized according to the solved physical fields and boundary conditions. The multiple PDE equation sets and boundary condition sets are characterized by the program, and the PDE equations and the geometric solution domain including boundary judgment are established. Based on the PDE equation, a solution network model is established by combining a fully connected neural network and a Fourier neural operator network structure. The geometric solution domain is then used to build the corresponding network model through fully connected network layers and Fourier neural operator network layers.
2. The ion implantation parameter optimization method according to claim 1, characterized in that, The acquisition of prior information includes: Obtain the geometric parameters of the system to be simulated, including incident particle parameters and target layer material parameters; Based on the target layer material parameters, a target layer simulation region is established; Using the incident particle parameters and the target layer simulation region, a set of reaction events and parameters is established to obtain the prior information.
3. The ion implantation parameter optimization method according to claim 2, characterized in that, The acquisition of the geometric parameters of the system to be simulated includes: Obtain simulation parameters. Based on the simulation parameters, incident particle parameters are created using the Ion class; Based on the simulation parameters, target layer material parameters are created using the Target class.
4. The ion implantation parameter optimization method according to claim 1, characterized in that, The establishment of a multiphysics coupling model, including a cluster dynamics model and a charge model, using the initial injected vacancy and ion concentration distribution includes: The initial injection of vacancies and ion concentration distribution generates a reaction event between the incident particle and the target layer ions.
5. The ion implantation parameter optimization method according to claim 1, characterized in that, The step of comparing the simulation results with the actual running results and optimizing the initial injection parameters based on the comparison results to obtain optimized injection parameters includes: Using actual operating data of the ion implanter as prior information, a parameter optimization dataset was established. Based on the parameters, the dataset is tuned, and a solution network model is established by combining fully connected and convolutional network structures. The training parameters are set and trained according to the solution network model to make it converge, thus obtaining a converged model; The initial injection parameters are tuned using the convergence model to obtain optimized injection parameters.
6. An ion implantation method, characterized in that, Optimized implantation parameters are obtained using the ion implantation parameter optimization method according to any one of claims 1-5, and ion implantation is performed according to the optimized implantation parameters.
7. An ion implantation parameter optimization device, characterized in that, include: The acquisition module is used to acquire the initial injection parameters of the target material; The simulation module is used to repeatedly simulate ion implantation using a trained network model based on the initial implantation parameters, and obtain simulation results. The optimization module compares the simulation results with the actual running results, and optimizes the initial injection parameters based on the comparison results to obtain optimized injection parameters. The trained network model was obtained through the following method: Obtain prior information, which is a set including reaction events and parameters; Based on the prior information and using a simulation system to perform simulated ion implantation, the initial implantation vacancy and ion concentration distribution are obtained; A multiphysics coupling model including a cluster dynamics model and a charge model is established using the initial injected vacancy and ion concentration distribution. The partial differential equations of the multiphysics coupling model are solved and trained using a deep learning network to obtain a trained network model. The trained network model is obtained through the following steps: Based on the initial distribution, a multiphysics coupling model including a cluster dynamics model and a charge model is established, and the boundary conditions of the coupling field are determined. Based on the physical fields that occur during ion implantation in the cluster dynamics and charge model, the PDE equations are coupled to establish a multiphysics coupling model. Then, the boundary condition set of the entire solution domain is established through the boundary conditions of the physical fields. Based on the definition of PDE equations by coupled physical fields, a solution geometric domain is established, boundary conditions are established, and initial conditions are set. The PDE equations are initialized according to the solved physical fields and boundary conditions. The multiple PDE equation sets and boundary condition sets are characterized by the program, and the PDE equations and the geometric solution domain including boundary judgment are established. Based on the PDE equation, a solution network model is established by combining a fully connected neural network and a Fourier neural operator network structure. The geometric solution domain is then used to build the corresponding network model through fully connected network layers and Fourier neural operator network layers.
8. An electronic device, characterized in that, include: A processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the ion implantation parameter optimization method as claimed in any one of claims 1-5 or the ion implantation method as claimed in claim 6.
9. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the ion implantation parameter optimization method as described in any one of claims 1-5 or the ion implantation method as described in claim 6.