Beam parameter processing method, device, electronic equipment, storage medium and product

Abstract

The present invention discloses a beam parameter processing method, device, electronic device, storage medium and product. The beam parameter processing method includes: inputting the expected beam state value into a parameter prediction model to obtain a predicted parameter value of at least one first component in the ion implanter, the predicted parameter value including the parameter value required to achieve the numerical value of the first component corresponding to the beam state; for each first component, inputting the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state into the time correlation model corresponding to the first component to obtain a set of ramping data for the first component; for each first component, setting the corresponding set of ramping data to the first component in sequence until the first component reaches the corresponding predicted parameter value. Avoiding the reduction of the life of the first component or even damage caused by setting the current parameter value to the predicted parameter value. Improving the safety of setting the parameter value of the first component.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a beam parameter processing method, device, electronic equipment, storage medium and product. Background Art

[0002] Ion implanters are a versatile piece of equipment used extensively in semiconductor manufacturing, optimizing material properties, and manufacturing optical devices. During the ion implantation process, specific ions are accelerated to energies of hundreds of kiloelectronvolts to form a beam.

[0003] Beam generation and regulation is a complex, parameter-intensive process, requiring adjustments to the parameters of various components within the ion implanter to achieve a stable, uniform beam. However, since each part of the ion implanter is composed of precision components, directly assigning predicted parameter values to each component can significantly shorten its lifespan or even damage it. Summary of the Invention

[0004] The present invention provides a beam parameter processing method, device, electronic equipment, storage medium and product to avoid damaging components in an ion implanter, thereby reducing their service life or even damaging them.

[0005] According to one aspect of the present invention, a beam parameter processing method is provided, comprising:

[0006] Inputting a desired beam state value into a parameter prediction model to obtain a predicted parameter value of at least one first component in the ion implanter, wherein the predicted parameter value includes a parameter value required to achieve a value of the first component corresponding to the beam state value;

[0007] For each first component, inputting the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state into the time correlation model corresponding to the first component to obtain a set of climbing data of the first component;

[0008] For each of the first components, a corresponding set of climbing data is sequentially set to the first component until the first component reaches the corresponding predicted parameter value.

[0009] According to another aspect of the present invention, there is provided a beam parameter processing device, comprising:

[0010] A first input module is configured to input a desired beam state value into a parameter prediction model to obtain a predicted parameter value of at least one first component in the ion implanter, wherein the predicted parameter value includes a parameter value required to achieve the value of the first component corresponding to the beam state value;

[0011] a second input module, configured to input, for each first component, a predicted parameter value corresponding to the first component and a current parameter value of the first component in a current state into a time correlation model corresponding to the first component, to obtain a set of ramp data of the first component;

[0012] The setting module is used to sequentially set a corresponding set of climbing data to each first component until the first component reaches the corresponding predicted parameter value.

[0013] According to another aspect of the present invention, an electronic device is provided, comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the method according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method described in any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, a computer program product is provided. The computer program product comprises a computer program. When the computer program is executed by a processor, the computer program implements the method according to any embodiment of the present invention.

[0019] The technical solution of the embodiment of the present invention determines a predicted parameter value for each first component in the ion implanter through a parameter prediction model to achieve the desired beam state, thereby improving the accuracy of determining the parameter value required for the first component. After determining the predicted parameter value of each first component, the time correlation model corresponding to the first component is used for each first component to determine the ramp data that the first component needs to go through to reach the predicted parameter value from the current parameter value, so that the first component sequentially goes through the ramp data and changes the current parameter value to the predicted parameter value, avoiding the direct setting of the current parameter value to the predicted parameter value, which may reduce the life of the first component or even cause damage. This improves the safety of setting the parameter value of the first component.

[0020] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 This is a flowchart of a beam parameter processing method provided in accordance with the first embodiment of the present invention;

[0023] Figure 2 This is a flowchart of another beam parameter processing method provided in accordance with the second embodiment of the present invention;

[0024] Figure 3 is a structural schematic diagram of an ion implanter provided by a beam parameter processing method according to an embodiment of the present invention;

[0025] Figure 4 This is a structural diagram of a beam parameter processing device provided according to a third embodiment of the present invention;

[0026] Figure 5 It is a structural diagram of an electronic device for implementing the beam parameter processing method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] The ion implanter uses an ion source to ionize dopant atoms to form ions. These ions are drawn out and accelerated by an electric field to form an ion beam with a certain energy and direction.

[0030] An ion beam is composed of charged particles—atoms or molecules that have lost or gained at least one electron and therefore react to electric and magnetic fields. The stability and uniformity (also known as homogeneity) of the ion beam are crucial factors in determining the quality of the finished product.

[0031] Ion implanters use electromagnetic fields generated by a series of components to filter and guide the ion beam. Therefore, beam generation and adjustment is a complex process involving numerous parameters. Traditional tuning methods require operators to have extensive experience and knowledge of how changes in the parameters of each component in the ion implanter affect the ion beam. However, manually adjusting the parameters of each component is tedious and difficult to achieve a stable and uniform beam. Even an experienced operator cannot quickly and accurately adjust all variable parameters to achieve a stable and uniform beam.

[0032] To address the above problems, the present invention proposes an automatic tuning method based on deep learning to improve the quality and efficiency of ion beam generation.

[0033] Example 1

[0034] Figure 1 This is a flow chart of a beam parameter processing method according to Embodiment 1 of the present invention. This embodiment is applicable to controlling the beam formation in ion implanters. Ion implanters are key equipment in semiconductor manufacturing. They accelerate specific ions to energies of several hundred kiloelectronvolts to form a beam for doping or surface modification of materials. A beam is a flow formed by a group of particles (such as electrons, ions, or atoms) moving in a specific direction and speed in space.

[0035] The beam parameter processing method can be executed by a beam parameter processing device, which can be implemented in hardware and / or software and can be configured in an electronic device, such as a mobile phone, a computer, or a server.

[0036] like Figure 1 As shown, the method includes:

[0037] S110 , inputting a desired beam state value into a parameter prediction model to obtain a predicted parameter value of at least one first component in the ion implanter.

[0038] The beam state value includes a numerical value that characterizes the state of the beam formed by the ion implanter. The beam state value can be used to describe the characteristics of the beam. For example, in an ion implanter, the beam state value may include indicators such as the beam intensity and energy.

[0039] In one embodiment, the beam state value includes one or more of the following:

[0040] Beam impurity percentage, beam current value, beam uniformity measurement information and beam stability measurement information.

[0041] The beam impurity percentage includes the proportion of impurities in the beam. The beam impurity percentage can be calculated by measuring the number or energy of impurity and non-impurity particles in the beam. For example, the ratio of impurity particles to the total number of particles is used to determine the beam impurity percentage.

[0042] The beam current value includes the amount of charge in the beam that passes through a specific cross-section per unit time. The beam current value reflects the relationship between the charge carried by particles in the beam and time. The beam charge value includes the ratio of the total charge to time.

[0043] Beam uniformity measurement information includes information measuring beam uniformity. Beam uniformity measurement information includes indicators measuring whether the beam is spatially uniform. Beam uniformity includes the uniformity of particle distribution across the beam's cross-section, which can be assessed by measuring the beam density or intensity distribution. Beam uniformity measurement includes measuring the spatial current distribution or intensity distribution of the beam to assess its uniformity. For example, the proportion of locations with uniform beam current can be determined as beam uniformity. In another example, beam uniformity can be measured using parameters such as the variance of the current at each measurement point in the beam.

[0044] Beam stability measurement information includes information measuring beam stability. Beam stability includes indicators that measure whether a particle beam remains stable in time or space. Beam stability can refer to the degree of variation in parameters such as intensity, energy, or position over time or space. Beam stability reflects the ability of key beam parameters (such as intensity, energy, and position) to maintain a stable state over time. Beam stability measurement information includes multiple measurements of the beam's set parameters and the percentage of stability determined.

[0045] The desired beam state includes the beam state value that the ion implanter needs to achieve, that is, the state value achieved by the beam formed by the pre-taken ion implanter. The parameter prediction model includes a model for predicting the parameter values of components in the ion implanter. Based on the input beam state value, the parameter prediction model can determine the predicted parameter values that the components in the ion implanter need to achieve in order to achieve the beam state value. The predicted parameter value includes the parameter value that the numerical value of the first component corresponding to the predicted parameter value needs to achieve in order to achieve the beam state value. That is, after the numerical value of the first component in the ion implanter is set to the predicted parameter value, the ion implanter can form a beam that meets the beam state value.

[0046] In this operation, the input of the desired beam state value can be obtained. The input method is not limited. It can be input through the human-computer interaction interface, through voice input, or through an input device. There is no limitation here. After obtaining the desired beam state value, the beam state value can be input into the parameter prediction model. After processing by the parameter prediction model, the predicted parameter value of at least one first component in the ion implanter is output. Different beam state values can correspond to the same or different first components. The first components include the components required for the ion implanter to achieve the desired beam state value.

[0047] The parameter prediction model can output a predicted parameter value for at least one first component in order to achieve a beam state value. Each first component corresponds to a predicted parameter value. The predicted parameter value is used to configure the corresponding first component so that the first component operates at the predicted parameter value, thereby enabling the ion implanter to form a beam that meets the beam state value.

[0048] S120 . For each first component, input the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state into a time correlation model corresponding to the first component to obtain a set of climbing data of the first component.

[0049] Each first component has a current state. The current state includes the current state of the first component. The current parameter value includes the parameter value of the first component in the current state. The current parameter value includes the initial parameter value of the first component after it is changed to the predicted parameter value. The first component will change from the current parameter value to the predicted parameter value.

[0050] Time-dependent models include time-related models. Components have different stable values and acceptable fluctuation ranges at different stages of their lifecycle. Time-dependent models can capture periodic data changes and set ramping ranges for components at different stages to achieve optimal performance.

[0051] In the present invention, ramping includes gradually increasing (or decreasing) the value of the first component to the predicted parameter value by gradually adjusting the parameter value. For example, ramping of data includes gradually adjusting the first component from the current parameter value to the predicted parameter value.

[0052] The ramp-up data includes data required to be experienced in the process of gradually adjusting the parameter value of the first component.

[0053] A set of ramping data may include a current parameter value and a predicted parameter value, and may also include intermediate parameter values that need to be set during the process of ramping from the current parameter value to the predicted parameter value. Intermediate parameter values include parameter values between the current parameter value and the predicted parameter value. The number of intermediate parameter values may be at least one. When setting the parameter value of a first component, the parameter value may be set sequentially from the current parameter value to the intermediate parameter value, and finally to the predicted parameter value, thereby achieving a gradient parameter value change and avoiding damage to the first component caused by directly setting the parameter value to the predicted parameter value.

[0054] In the present invention, a set of ramp data may include or exclude the current parameter value. In the case of retaining the current parameter value, since the first component is currently at the current parameter value, the first component can be set starting from the parameter value subsequent to the current parameter value.

[0055] In this operation, for each predicted parameter value output by the parameter prediction model, the first component corresponding to the predicted parameter value is determined, and then the time correlation model corresponding to the first component is determined. For each first component, the following operation is performed: the predicted parameter value and the current parameter value of the first component are input into the time correlation model, so that a set of ramp data is output for the first component through the time correlation model to facilitate parameter value setting for the first component.

[0056] When outputting the ramp-up data, the time-correlation model may combine the historical ramp-up data of the first component to determine the ramp-up data in the current state.

[0057] In this example, there is one parameter prediction model and multiple time-related models. The number of time-related models is associated with the number of components in the ion implanter, such as the number of time-related models is equal to the number of components in the ion implanter.

[0058] The predicted parameter values output by the parameter prediction model are sequentially input into the corresponding time correlation model to determine the ramp data. The time correlation model corresponding to the predicted parameter value includes the time correlation model corresponding to the first component corresponding to the predicted parameter value.

[0059] S130 : For each first component, sequentially set a corresponding set of climbing data to the first component until the first component reaches the corresponding predicted parameter value.

[0060] For each first component for which ramp data has been determined, after determining the ramp data for the first component, this operation can traverse each data in a set of ramp data. The traversed data is sequentially set to the first component until the first component is set to the predicted parameter value, thereby avoiding damage to the first component.

[0061] For example, a set of ramp data corresponding to the first component is 0, 100, and 200. 0 corresponds to the current parameter value, 200 corresponds to the predicted parameter value, and 100 corresponds to the intermediate parameter value. Furthermore, a set of ramp data may not include the current parameter value. For example, a set of ramp data may be 100 and 200.

[0062] This operation can set 100 to the first component, so that the parameter value of the first component is changed from 0 to 100. After the parameter value of the first component is changed to 100, 200 is continued to be set to the first component.

[0063] In this operation, when the climbing data is sequentially set to the first component, for each set climbing data, after the first component reaches the climbing data, the next climbing data is continued to be set until the predicted parameter value is reached.

[0064] The beam parameter processing method provided by the present invention determines a predicted parameter value for each first component in an ion implanter through a parameter prediction model to achieve a desired beam state, thereby improving the accuracy of determining the parameter value required for the first component. After determining the predicted parameter value for each first component, the time correlation model corresponding to the first component is used to determine the ramp data required for the first component to reach the predicted parameter value from the current parameter value. This allows the first component to sequentially experience the ramp data and change the current parameter value to the predicted parameter value, avoiding the direct setting of the current parameter value to the predicted parameter value, which could reduce the lifespan of the first component or even cause damage. This improves the safety of setting the parameter values of the first components.

[0065] Based on the above embodiment, a modified embodiment of the above embodiment is proposed. It should be noted that, in order to simplify the description, only the differences from the above embodiment are described in the modified embodiment.

[0066] In one embodiment, the set of ramp-up data includes the predicted parameter value; and for each first component, sequentially setting the corresponding set of ramp-up data to the first component until the first component reaches the corresponding predicted parameter value includes:

[0067] For each of the first components, selecting data having the largest difference with the predicted parameter value corresponding to the first component from a set of ramp data corresponding to the first component as target data;

[0068] Setting the target data to the first component;

[0069] After the data of the first component reaches the target data, it returns to continue selecting the next target data from the remaining data of a set of climbing data corresponding to the first component, and performs the setting operation until the set of climbing data corresponding to the first component is traversed.

[0070] In this embodiment, a set of climbing data includes at least the predicted parameter value. In the process of setting the first component, the setting method of each first component is the same, and the description here takes one first component as an example.

[0071] The target data includes the data in the slope data with the largest difference from the predicted parameter value. If the slope data is sorted in ascending or descending order, the target data includes the data selected sequentially from the slope data, including the first data in the slope data. The predicted parameter value may be the last data in the slope data.

[0072] In one example, target data is selected sequentially starting from a high position in a set of climbing data until a predicted parameter value is selected, or target data is selected sequentially starting from a high position in a set of climbing data until a predicted parameter value is selected.

[0073] After selecting the target data, the target data can be set to the first component. After the first component reaches the target data, the next target data is selected from the remaining data in the set of ramp-up data, and the operation of setting the target data to the first component is continued until the set of ramp-up data is completely traversed. The predicted data in the set of ramp-up data is the last data traversed. After traversing the set of ramp-up data, the parameter value of the first component is set to the predicted parameter value. The remaining data can be considered as the data in the set of ramp-up data that was not selected.

[0074] After a set of climbing data has selected target data once, the target data can be removed from the climbing data so that the remaining target data can be selected from the remaining climbing data. Alternatively, after a set of climbing data has selected target data once, the next target data can be selected from the remaining unselected data.

[0075] This embodiment traverses each target data in a set of climbing data, and gradually traverses to the predicted parameter value. Each time it traverses, the traversed data is set to the first component, so that the first component gradually reaches the predicted parameter value, thereby improving the safety of the first component.

[0076] In one embodiment, for each first component, the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state are input into the time correlation model corresponding to the first component to obtain a set of ramp data of the first component, including:

[0077] For each first component, inputting the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state into the time correlation model corresponding to the first component;

[0078] The ramp data from the current parameter value to the predicted parameter value is determined and output by combining the time correlation model with the data of each ramp during the life cycle of the first component.

[0079] This embodiment refines the operation of outputting a set of ramp-up data for each first component through a time correlation model.

[0080] In this embodiment, when determining the ramp-up data, the time-correlation model combines each historical ramp-up data in the life cycle of the first component to determine a set of ramp-up data that the first component needs to go through to change from the current parameter value to the predicted parameter value in the current state.

[0081] Because each part of the ion implantation system is composed of precision components, directly assigning predicted parameter values to each component could potentially damage the device. Furthermore, simply setting the numerical ramp to a linear increase within a constant time interval is inappropriate. For example, while a value from 0 to 100 might only take 1 second, a safe and stable increase from 100 to 200 might require 5 seconds. Therefore, setting component parameters should be a nonlinear process.

[0082] In order to extend the service life of the ion implanter, the present invention uses a time-dependent model, such as a long short-term memory model, to set the values of components safely and stably until the values of the components reach the predicted parameter values of the model.

[0083] In typical neural network models, each layer's nodes are interconnected. As data related to time and periods passes through such models, their past and future relevance is lost. Temporal correlation models incorporate memory units, which maintain information flow across long sequences. They can generate values for the next time point based on data from multiple past time points, while also ensuring relatively smooth fluctuations between these values.

[0084] Example 2

[0085] Figure 2This is a flow chart of another beam parameter processing method provided according to the second embodiment of the present invention. This embodiment provides training operations for parameter prediction models and time correlation models. Figure 2 As shown, the method includes:

[0086] S210: Obtain a first training set.

[0087] The first training set includes parameter values of the second component and corresponding beam state values when generating and adjusting the ion beam.

[0088] The second components include components included in the ion implanter. The first components and the second components can be components of the same type or model. The ion implanter in the parameter prediction model training phase and the ion implanter in the application phase of the present invention can be ion implanters of the same type or model.

[0089] The parameter values of the second component include the parameter values that need to be set in order for the beam to reach the state value. The state value of the beam can correspond to the parameter values of multiple second components. The parameter values of the second component correspond to the predicted parameter values. The parameter values of the second component include the parameter values of the parameter prediction model stage. The predicted parameter values include the parameter values outputted during the parameter prediction model application stage. The expected beam state value corresponds to the state value of the beam in the first training set. The expected beam state value includes the value inputted during the parameter prediction model application stage. The state value of the beam in the first training set includes the value inputted during the parameter prediction model training stage.

[0090] The data in the first training set includes data from the process of generating and adjusting the ion beam. The parameter values of the second component and the state value of the corresponding beam can be used as sample pairs to train the model to obtain a trained parameter prediction model.

[0091] In one example, parameters of a series of components (as parameter values of the second component) and various state values of the final beam are collected each time an ion beam is generated and adjusted as training data for the model.

[0092] S220: Train a first initial model based on the first training set to obtain a parameter prediction model.

[0093] The first initial model includes an untrained basic model for determining the prediction parameter value. A trained parameter prediction model can be obtained by training the first initial model with the first training set.

[0094] The first initial model uses randomly generated data when creating the model. If these random values are not adjusted through training, the predicted values will be far from the expected values.

[0095] In this operation, during the process of training the first initial model, an optimization algorithm is used to train the model. The initially generated random data can be fine-tuned through a large amount of training data in the first training set, so that the final output value of the model is closer to the optimal value actually required.

[0096] In the process of optimizing the model training, gradient descent and hyperparameter adjustment can be performed to change the amplitude of the random value while making the predicted value close to the expected value. Stochastic gradient descent and hyperparameter adjustment can change the amplitude of this change. For example, when the difference between the predicted value and the expected value is large, a larger change amplitude is used; and when the difference is small, a smaller change amplitude is selected to make the model more fit the training data.

[0097] S230: Obtain a second training set.

[0098] The second training set includes a target parameter value, climbing data of a third component in a current state, and a current parameter value of the third component. The target parameter value is the parameter value that the third component needs to eventually reach.

[0099] The third component includes components included in the ion implanter. The second component and the third component include components of the same type or model. The second component and the third component also include the same components.

[0100] In this embodiment, the time correlation model and the parameter prediction model can be trained jointly or separately.

[0101] The target parameter values include empirical values and parameter values output during the parameter prediction model training process.

[0102] The ramp-up data of the third component in the current state includes the ramp-up data that the third component needs to go through in order to reach the target parameter value.

[0103] The target parameter value corresponds to the predicted parameter value, and the target parameter value includes the data input during the time correlation model training process. The predicted parameter value includes the data input during the time correlation model application process. The current parameter value of the third component can correspond to the current parameter value, and the current parameter value includes the data input during the time correlation model application process. The current parameter value of the third component includes the data input during the time correlation model training process. The ramp data of the third component in the current state corresponds to a set of ramp data of the first component, the former being the output of the time correlation model training phase, and the latter being the output of the time correlation model application phase.

[0104] The second training set includes empirical values and also includes data formed by adjusting component parameters based on manual history to obtain the required beam current.

[0105] S240: Train a second initial model related to time based on the second training set to obtain a time correlation model.

[0106] The second initial model includes an untrained basic model for determining the hill climbing data. The trained time correlation model can be obtained by training the second initial model with the second training set.

[0107] During the training process, the second initial model can also perform gradient descent and hyperparameter adjustment to improve the training effect, which will not be described in detail here.

[0108] In one embodiment, the target parameter value includes the parameter value output during the parameter prediction model training process.

[0109] In this embodiment, the parameter prediction model and the time correlation model can be jointly trained. During the training process, the parameter value output by the first initial model can be input into the second initial model as the target parameter value, and then processed in combination with the ramp data of the third component in the current state input into the second initial model. The ramp data for this round is output, and this ramp data is processed together with the ramp data of the third component in the current state to adjust the parameters of the time correlation model. That is, during the training of the parameter prediction model and the time correlation model, the output of the parameter prediction model is input into the corresponding time correlation model to achieve joint training, so that the output of each round of the first initial model is output to the second initial model trained in that round. The corresponding relationship can be determined based on the second component corresponding to the ramp data output by the parameter prediction model and the third component corresponding to the time correlation model.

[0110] S250 , inputting the expected beam state value into a parameter prediction model to obtain a predicted parameter value of at least one first component in the ion implanter.

[0111] S260 . For each first component, input the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state into a time correlation model corresponding to the first component to obtain a set of climbing data of the first component.

[0112] S270 : For each of the first components, sequentially set a corresponding set of climbing data to the first component until the first component reaches the corresponding predicted parameter value.

[0113] In one example, ramp data is sequentially set to each first component, and after determining that the set ramp data is reached, the next ramp data is set until the predicted parameter value is reached. After the predicted parameter value is reached, stability and uniformity can be measured.

[0114] The beam parameter processing method provided in this embodiment obtains a parameter prediction model through training with a first training set and a time correlation model through training with a second training set. The predicted parameter values of a first component in an ion implanter are determined based on the parameter prediction model and the time correlation model, thereby achieving a desired beam state value, improving beam stability, and enhancing the safety of the first component.

[0115] Based on the above embodiment, a modified embodiment of the above embodiment is proposed. It should be noted that, in order to simplify the description, only the differences from the above embodiment are described in the modified embodiment.

[0116] In one embodiment, the predicted parameter values include one or more of the following:

[0117] Gas type, gas flow, solid type, evaporator heating temperature, arc chamber voltage, arc chamber current, cathode voltage, cathode current, filament current, source magnet current, extraction electrode vertical tilt angle, lateral left and right position, gap position, extraction suppression voltage, extraction suppression current, analysis magnet current, high-voltage discharge rod voltage, high-voltage discharge rod current, extraction voltage, extraction current, accelerator voltage, accelerator current, quadruple lens voltage, quadruple lens current, steering electrode voltage, steering electrode current, scanning amplifier voltage, parallel lens voltage, parallel lens current, scanning suppression voltage, scanning suppression current, angular energy focusing voltage, angular energy focusing current, energy slit width.

[0118] Gases are used in ion sources to generate the desired ions. These gases are ionized in the ion source of an ion implanter to form a charged ion beam. The desired ions are then isolated and implanted into the material. Predictable parameters for the ion source include gas type and gas flow rate.

[0119] The gas type includes the type of gas used in the ion implanter. The gas flow rate includes the flow rate of the gas used in the ion implanter.

[0120] Solids in an ion implanter can be converted to a gaseous state by heating in an evaporator for use. The solid type includes the type of solid used in the ion implanter. Prediction parameters for the evaporator include the solid type. Prediction parameters for the evaporator may also include the evaporator heating temperature, i.e., the temperature to which the evaporator is heated.

[0121] Figure 3 is a structural diagram of an ion implanter provided by a beam parameter processing method according to an embodiment of the present invention, see Figure 3 The ion source 1 includes an arc chamber, a cathode, a filament, and a source magnet. The arc chamber comprises the cavity of the ion source, which is used to contain the working gas. The cathode can be used to emit electrons. The filament comprises a component that heats the cathode. The source magnet comprises a component that generates a magnetic field within the arc chamber.

[0122] The filament is located within the arc chamber, the filament is located below the arc chamber, and the source magnet is located around the arc chamber. In the description of the present invention, it should be understood that the terms "upper" and "lower" and the like, indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended only to facilitate and simplify the description of the present invention. They do not indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention.

[0123] The arc chamber voltage includes the voltage of the arc chamber in the ion source 1. The arc chamber current includes the current in the arc chamber in the ion source 1. The cathode voltage includes the voltage of the cathode in the ion source 1. The cathode current includes the current in the cathode in the ion source 1. The filament current includes the current in the filament in the ion source 1. The source magnet current includes the current in the source magnet in the ion source 1.

[0124] The extraction electrode 2 in an ion implanter, also known as the extraction electrode, comprises the electrode assembly used for extraction. Corresponding prediction parameters include the extraction electrode's vertical tilt angle, lateral position, and gap position. Extraction voltage and current can also be included.

[0125] The extraction electrode vertical tilt angle includes the angle between the extraction electrode 2 and the vertical direction. The lateral position includes the horizontal position adjustment of the extraction electrode 2. The gap position includes the distance between the extraction electrode 2 and the adjacent component. The extraction voltage includes the voltage of the extraction electrode 2. The extraction current includes the current of the extraction electrode 2.

[0126] The extracted suppression voltage includes the voltage of the electrode used for suppression. The extracted suppression current includes the current of the electrode used for suppression.

[0127] The predicted parameter value of the analysis magnet 3 in the ion implanter may include the analysis magnet current, that is, the current of the analysis magnet 3. In addition, the predicted parameter value of the analysis magnet 3 may also include the analysis magnet voltage. The analysis magnet 3 can be used to separate and screen ions.

[0128] The components of the ion implanter also include a high-voltage discharge rod, a quadruple lens, and an accelerator. Accordingly, the corresponding expected parameter values include: high-voltage discharge rod voltage, high-voltage discharge rod current, accelerator voltage, accelerator current, quadruple lens voltage, and quadruple lens current.

[0129] The high-voltage discharge rod can be used to release residual high-voltage charge, where high voltage refers to voltages above a certain voltage level, such as 1 kV or above. The quadruple lens can be used to adjust the shape and focusing effect of the ion beam. The accelerator can accelerate the ions.

[0130] The high-voltage discharge rod voltage includes the voltage of the high-voltage discharge rod in the ion implanter. The high-voltage discharge rod current includes the current of the high-voltage discharge rod in the ion implanter. The accelerator voltage includes the voltage of the accelerator in the ion implanter. The accelerator current includes the current of the accelerator in the ion implanter. The quadruple lens voltage includes the voltage of the quadruple lens in the ion implanter, and the quadruple lens current includes the current of the quadruple lens in the ion implanter.

[0131] The components of the ion implanter also include a scanning amplifier and a parallel lens. The scanning amplifier includes a device that scans and amplifies the ion beam. The parallel lens includes a device that maintains each ion beam in parallel. Predicted parameters corresponding to the scanning amplifier include: scanning amplifier voltage, scanning suppression voltage, and / or scanning suppression current. A scanning amplifier current may also be included. Predicted parameters corresponding to the parallel lens include: parallel lens voltage and / or parallel lens current.

[0132] The scan amplifier voltage includes a voltage signal for controlling ion beam deflection. The scan amplifier current includes a current signal for controlling ion beam deflection. The scan suppressor voltage includes a voltage for limiting the ion beam scan range, preventing the ion beam from exceeding a predetermined range during scanning. The scan suppressor current is a parameter used in conjunction with the scan suppressor voltage to control the current intensity during scanning, preventing uneven ion beam deflection caused by current fluctuations during scanning.

[0133] The parallel lens voltage includes the voltage applied to the parallel lens electrodes, and the parallel lens current includes the current flowing through the parallel lens electromagnetic coil.

[0134] The steering electrode 4 in an ion implanter controls the direction of the ion beam. Corresponding predicted parameters include the steering electrode voltage and the steering electrode current. The steering electrode voltage generates an electric field that deflects the ion beam in a specific direction. The steering electrode current and voltage work together to influence the strength and stability of the electric field.

[0135] The angle energy filter 5 in the ion implanter adjusts the ion beam angle to ensure that the ion beam is implanted onto the wafer. Predicted parameters corresponding to the angle energy filter 5 include: angle energy focusing voltage, angle energy focusing current, and / or energy slit width. Adjusting the angle energy filter 5 using the predicted parameters focuses the energy, forming an energy slit.

[0136] The angle energy focusing voltage includes the voltage applied to the angle energy filter 5 , the angle energy focusing current includes the current applied to the angle energy filter 5 , and the energy slit width includes the width of the energy slit generated by the angle energy filter 5 .

[0137] The beam parameter processing method provided by the present invention includes an ion beam automatic tuning method based on a deep learning model. In order to reproduce this stable and uniform beam under different conditions, the present invention proposes an automatic tuning method based on a deep learning model.

[0138] The parameter prediction model receives the expected beam state value and generates a series of predicted component parameter values based on it. After setting these predicted parameter values for each component, it can determine whether the beam current meets the expected settings and whether the beam stability and uniformity meet the standards. If so, the process ends. Simultaneously, the set of input and output parameters serves as subsequent training data for the parameter prediction model, allowing for continuous model adjustments to facilitate subsequent predictive analysis.

[0139] If the standard is not met, the model can be trained further, or the expected beam state value can be input to obtain the predicted parameter value, and then the ramp data can be determined through the time correlation model, and the ramp data can be set to the first component to determine whether the standard is met.

[0140] During the application phase of the model, each set of input and output data can be recorded and incorporated into the first training set and the second data set to facilitate subsequent training.

[0141] Compared to manual tuning methods, this method efficiently adjusts the variable parameters of all components in the beam generation and regulation process to achieve a stable and uniform beam. Through a deep learning model, beam reproduction under different conditions becomes easier. This reduces beam generation costs, improves efficiency, enhances parameter adjustment capabilities, and provides greater adaptability.

[0142] Example 3

[0143] Figure 4 FIG. 1 is a schematic diagram of the structure of a beam parameter processing device according to the third embodiment of the present invention. Figure 4 As shown, the device includes:

[0144] A first input module 310 is configured to input a desired beam state value into a parameter prediction model to obtain a predicted parameter value of at least one first component in the ion implanter, wherein the predicted parameter value includes a parameter value required to achieve the value of the first component corresponding to the beam state value;

[0145] A second input module 320 is configured to input, for each first component, a predicted parameter value corresponding to the first component and a current parameter value of the first component in a current state into a time correlation model corresponding to the first component to obtain a set of ramp data for the first component;

[0146] The setting module 330 is configured to sequentially set a corresponding set of ramp data to each first component until the first component reaches the corresponding predicted parameter value.

[0147] The beam parameter processing method provided in this embodiment is to achieve the desired beam state. The first input module determines the predicted parameter value for each first component in the ion implanter through a parameter prediction model, thereby improving the accuracy of the parameter value determination required for the first component. After determining the predicted parameter value for each first component, the second input module uses the time correlation model corresponding to the first component for each first component to determine the climbing data that the first component needs to go through to reach the predicted parameter value from the current parameter value, so that the first component goes through the climbing data in sequence. The setting module changes the current parameter value to the predicted parameter value to avoid directly setting the current parameter value to the predicted parameter value, which may reduce the lifespan or even cause damage. This improves the safety of the parameter value setting of the first component.

[0148] In one embodiment, a set of climbing data includes the predicted parameter value; the setting module 330 is specifically configured to:

[0149] For each of the first components, selecting data having the largest difference with the predicted parameter value corresponding to the first component from a set of ramp data corresponding to the first component as target data;

[0150] Setting the target data to the first component;

[0151] After the data of the first component reaches the target data, it returns to continue selecting the next target data from the remaining data of a set of climbing data corresponding to the first component, and performs the setting operation until the set of climbing data corresponding to the first component is traversed.

[0152] In one embodiment, the beam parameter processing module includes a first training module, and the training operation of the parameter prediction model performed by the first training module includes:

[0153] Acquire a first training set, wherein the first training set includes parameter values of the second component and corresponding beam state values when generating and adjusting the ion beam;

[0154] A first initial model is trained based on the first training set to obtain a parameter prediction model.

[0155] In one embodiment, the beam parameter processing module includes a second training module, and the training operation of the time correlation model performed by the second training module includes:

[0156] Obtaining a second training set, where the second training set includes a target parameter value, climbing data of a third component in a current state, and a current parameter value of the third component, wherein the target parameter value is a parameter value that the third component ultimately needs to achieve;

[0157] A second initial model related to time is trained based on the second training set to obtain a time correlation model.

[0158] In one embodiment, the target parameter value includes a parameter value output during the parameter prediction model training process.

[0159] In one embodiment, the second input module 320 is specifically configured to:

[0160] For each first component, inputting the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state into the time correlation model corresponding to the first component;

[0161] The ramp data from the current parameter value to the predicted parameter value is determined and output by combining the time correlation model with the data of each ramp during the life cycle of the first component.

[0162] In one embodiment, the beam state value includes one or more of the following:

[0163] Beam impurity percentage, beam current value, beam uniformity measurement information and beam stability measurement information.

[0164] In one embodiment, the predicted parameter values include one or more of the following:

[0165] Gas type, gas flow, solid type, evaporator heating temperature, arc chamber voltage, arc chamber current, cathode voltage, cathode current, filament current, source magnet current, extraction electrode vertical tilt angle, lateral left and right position, gap position, extraction suppression voltage, extraction suppression current, analysis magnet current, high-voltage discharge rod voltage, high-voltage discharge rod current, extraction voltage, extraction current, accelerator voltage, accelerator current, quadruple lens voltage, quadruple lens current, steering electrode voltage, steering electrode current, scanning amplifier voltage, parallel lens voltage, parallel lens current, scanning suppression voltage, scanning suppression current, angular energy focusing voltage, angular energy focusing current, energy slit width.

[0166] The beam parameter processing device provided in the embodiment of the present invention can execute the beam parameter processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0167] Example 4

[0168] Figure 5The figure is a schematic diagram of an electronic device that implements the beam parameter processing method according to an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided for illustrative purposes only and are not intended to limit the implementation of the present inventions described and / or claimed herein.

[0169] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program executable by the at least one processor 11, and the computer program is executed by the at least one processor 11 so that the at least one processor 11 can perform the method provided by the present invention.

[0170] The processor 11 can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 12 or a computer program loaded from a storage unit 18 into a random access memory (RAM) 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0171] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0172] Processor 11 includes various general-purpose and / or specialized processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 executes the various methods and processes described above, such as the beam parameter processing method.

[0173] In some embodiments, the beam parameter processing method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the beam parameter processing method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the beam parameter processing method in any other suitable manner (e.g., via firmware).

[0174] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard parts (ASSPs), system on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system including at least one programmable processor, including a special-purpose or general-purpose programmable processor, that can receive data and instructions from a storage system, at least one input device, and at least one output device, and that can transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0175] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0176] In the context of the present invention, a computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the beam parameter processing method provided by the present invention when the instructions are executed.

[0177] The present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the computer program implements the method provided according to the embodiment of the present invention.

[0178] Computer-readable storage media include tangible media that can contain or store computer programs for use by or in conjunction with an instruction execution system, device, or apparatus. Computer-readable storage media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media include machine-readable signal media. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0179] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD monitor)) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can include any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0180] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), blockchain networks, and the Internet.

[0181] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship is established by computer programs running on the respective computers, creating a client-server relationship. Servers include cloud servers, also known as cloud computing servers or cloud hosts. These servers are a type of hosting product within the cloud computing service ecosystem, addressing the management difficulties and limited scalability of traditional physical hosts and virtual private server (VPS) services.

[0182] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0183] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A beam parameter processing method, characterized in that: include: Inputting a desired beam state value into a parameter prediction model to obtain a predicted parameter value of at least one first component in the ion implanter, wherein the predicted parameter value includes a parameter value required to achieve a value of the first component corresponding to the beam state value; For each first component, inputting the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state into the time correlation model corresponding to the first component to obtain a set of climbing data of the first component; For each of the first components, a corresponding set of climbing data is set to the first component in sequence until the first component reaches the corresponding predicted parameter value, and the first component is gradually adjusted from the corresponding current parameter value to the corresponding predicted parameter value.

2. The method according to claim 1, characterized in that The set of ramp-up data includes the predicted parameter value; and for each first component, sequentially setting the corresponding set of ramp-up data to the first component until the first component reaches the corresponding predicted parameter value, including: For each of the first components, selecting data having the largest difference with the predicted parameter value corresponding to the first component from a set of ramp data corresponding to the first component as target data; Setting the target data to the first component; After the data of the first component reaches the target data, it returns to continue selecting the next target data from the remaining data of a set of climbing data corresponding to the first component, and performs the setting operation until the set of climbing data corresponding to the first component is traversed.

3. The method according to claim 1, characterized in that The training operation of the parameter prediction model includes: Acquire a first training set, wherein the first training set includes parameter values of the second component and corresponding beam state values when generating and adjusting the ion beam; A first initial model is trained based on the first training set to obtain a parameter prediction model.

4. The method according to claim 1, wherein The training operation of the time association model includes: Obtaining a second training set, where the second training set includes a target parameter value, climbing data of a third component in a current state, and a current parameter value of the third component, wherein the target parameter value is a parameter value that the third component ultimately needs to achieve; A second initial model related to time is trained based on the second training set to obtain a time correlation model.

5. The method according to claim 4, characterized in that The target parameter value includes the parameter value output during the parameter prediction model training process.

6. The method according to claim 1, characterized in that For each first component, the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state are input into the time correlation model corresponding to the first component to obtain a set of climbing data of the first component, including: For each first component, inputting the predicted parameter value corresponding to the first component and the current parameter value of the first component in the current state into the time correlation model corresponding to the first component; The ramp data from the current parameter value to the predicted parameter value is determined and output by combining the time correlation model with the data of each ramp during the life cycle of the first component.

7. The method according to claim 1, characterized in that The beam state value includes one or more of the following: Beam impurity percentage, beam current value, beam uniformity measurement information and beam stability measurement information.

8. The method according to claim 1, characterized in that The predicted parameter values include one or more of the following: Gas type, gas flow, solid type, evaporator heating temperature, arc chamber voltage, arc chamber current, cathode voltage, cathode current, filament current, source magnet current, extraction electrode vertical tilt angle, lateral left and right position, gap position, extraction suppression voltage, extraction suppression current, analysis magnet current, high-voltage discharge rod voltage, high-voltage discharge rod current, extraction voltage, extraction current, accelerator voltage, accelerator current, quadruple lens voltage, quadruple lens current, steering electrode voltage, steering electrode current, scanning amplifier voltage, parallel lens voltage, parallel lens current, scanning suppression voltage, scanning suppression current, angular energy focusing voltage, angular energy focusing current, energy slit width.

9. A beam parameter processing device, characterized in that: include: A first input module is configured to input a desired beam state value into a parameter prediction model to obtain a predicted parameter value of at least one first component in the ion implanter, wherein the predicted parameter value includes a parameter value required to achieve the value of the first component corresponding to the beam state value; a second input module, configured to input, for each first component, a predicted parameter value corresponding to the first component and a current parameter value of the first component in a current state into a time correlation model corresponding to the first component, to obtain a set of ramp data of the first component; A setting module is used to set the corresponding set of climbing data to each first component in sequence until the first component reaches the corresponding predicted parameter value, and the first component is gradually adjusted from the corresponding current parameter value to the corresponding predicted parameter value.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method according to any one of claims 1 to 8 when the processor executes the computer instructions.

12. A computer program product, characterized in that The computer program product comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 8.

Images