Metal ion source beam intensity and purity two-parameter adaptive control method

Abstract

The invention discloses a metal ion source beam intensity and purity two-parameter adaptive control method, and solves the problems of two-parameter coupling conflict, poor dynamic adaptability and low control precision in the prior art. The method comprises the following steps: initializing beam intensity and purity target values and control parameters; two-parameter data are collected and preprocessed in real time, and normalized deviation is calculated; dynamically distributing weights based on the purity deviation to generate a comprehensive deviation; pID parameters are adaptively optimized according to the comprehensive deviation and the change rate; and performing closed-loop iteration control until the two parameters meet the precision threshold. Through dynamic weight distribution and PID parameter adaptive optimization, cooperative control of beam intensity and purity is realized, beam intensity fluctuation can be controlled within + / -0.8%, purity fluctuation can be controlled within + / -0.1%, and the method is suitable for ion implanters and physical vapor deposition equipment in the fields of semiconductor manufacturing, metal film deposition and the like. And the process stability and the product yield are obviously improved.

Description

Technical Field

[0001] This invention belongs to the field of metal ion source control technology, specifically relating to an adaptive control method based on dual parameter feedback of beam intensity and purity, which is particularly suitable for precise control of ion sources in metal ion implanters and metal thin film deposition equipment in fields such as semiconductor manufacturing and integrated circuit chip processing. Background Technology

[0002] Metal ion sources are core components in processes such as metal ion implantation and metal thin film deposition. The intensity and purity of the output ion beam directly determine the process quality. For example, in semiconductor chip manufacturing, the stability of the metal ion implantation beam intensity affects the uniformity of doping concentration; insufficient purity introduces impurities, leading to device leakage or performance degradation. In metal thin film deposition, beam intensity fluctuations cause uneven film thickness, and low purity increases film resistivity and reduces adhesion. Therefore, achieving high-precision synergistic control of metal ion source beam intensity and purity is crucial for improving the stability of related processes and product yield.

[0003] Current state of technology and existing technical problems Currently, the main technical shortcomings of metal ion source control methods are as follows, making it difficult to meet the requirements of high-precision processes: 1. Limitations of single-parameter control: Ignores the coupling relationship between two parameters. Existing technologies mostly adopt a "single-parameter priority control" strategy, that is, focusing solely on beam intensity or purity as the core control objective: If beam intensity is controlled in priority, the proportion of impurity ions may increase and the purity may decrease due to excessive adjustment of the ion source operating parameters. If purity is prioritized, excessive restrictions on ion extraction conditions may lead to drastic fluctuations in beam intensity.

[0004] The essence of the problem is that there is a strong coupling relationship between beam intensity and purity, and single-parameter control cannot balance the requirements of both, resulting in a situation where one is prioritized at the expense of the other.

[0005] 2. Fixed-weight dual-parameter control: unable to dynamically adapt to changes in operating conditions. Some improved solutions attempt to use "fixed-weight dual-parameter control," that is, preset beam intensity deviation weights and purity deviation weights, and calculate the comprehensive deviation through weighted summation. However, this method has inherent drawbacks: Static weights cannot cope with dynamic operating conditions: for example, during the startup phase of an ion source, it is necessary to prioritize improving purity, while during the stable operation phase, it is necessary to focus on suppressing intensity fluctuations, but fixed weights cannot switch priorities. Control failure under extreme deviations: When the purity deviation suddenly increases, the fixed weight still allocates control resources according to the preset ratio, which cannot quickly bring the purity back to the target range, resulting in process interruption.

[0006] 3. Fixed control parameters: Insufficient adaptive capability In existing technologies, the core parameters of PID control are mostly fixed values ​​tuned offline, which cannot adapt to the nonlinear and time-varying characteristics of ion sources. Nonlinear characteristics: The relationship between the beam intensity and purity of the ion source and the operating parameters is not nonlinear, and the control accuracy of fixed PID parameters varies significantly at different operating points; Time-varying characteristics: After long-term operation, the dynamic response characteristics of the ion source change, and fixed parameters will lead to an increase in overshoot or a longer settling time.

[0007] 4. Lack of data processing and anomaly response mechanisms Existing technologies do not pay enough attention to the preprocessing and anomaly handling of collected data: Unsuppressed noise interference: The data collected by the beam intensity detector and mass spectrometer contain random noise, which will introduce control oscillations if used directly for deviation calculation; No response to abnormal data: When the sensor fails or the ion source suddenly malfunctions, the system lacks a rapid identification and backup control strategy, which can easily lead to equipment damage or process scrapping.

[0008] Summary of existing technical issues In summary, existing metal ion source control methods suffer from shortcomings such as neglecting the two-parameter coupling relationship, lacking dynamic priority adjustment, having fixed control parameters, and having imperfect data processing mechanisms. These limitations prevent them from meeting the demands of semiconductor manufacturing, high-precision thin film deposition, and other fields for ion sources that require "high purity, high intensity stability, and high adaptability." Therefore, developing a control method that can dynamically balance beam intensity and purity control priorities, achieve adaptive optimization of control parameters, and possess robust data processing capabilities has become a pressing technical problem to be solved in this field. Summary of the Invention

[0009] To address the shortcomings of existing metal ion source control technologies, such as dual-parameter coupling conflicts, poor adaptability to dynamic operating conditions, insufficient adaptive capability of control parameters, and lack of data robustness, the present invention aims to provide a dual-parameter adaptive control method for beam intensity and purity of metal ion sources. Through the collaborative design of dynamic weight allocation, real-time optimization of control parameters, data preprocessing, and anomaly response mechanisms, high-precision coordinated control of beam intensity and purity is achieved, meeting the requirements of high-precision processes such as semiconductor manufacturing and metal thin film deposition for the stability and reliability of ion sources.

[0010] To achieve the above objectives, this invention provides a dual-parameter adaptive control method for the beam intensity and purity of a metal ion source. Its core lies in achieving coordinated control of beam intensity and purity through a three-level control logic: "dual-parameter dynamic weight allocation - adaptive optimization of control parameters - closed-loop iteration and robust processing." The specific technical solution is as follows: A dual-parameter adaptive control method for the beam intensity and purity of a metal ion source includes the following steps: S1. Parameter Initialization: Set the target beam current value according to the type of metal ion source and the target process requirements. and purity target value Initialize the initial PID control coefficients , , Adaptive adjustment factor and initial weighting coefficients , ( ); S2. Real-time acquisition of dual parameters: The actual value of the beam intensity output by the metal ion source is acquired in real time through a beam intensity detector. The actual value of ion beam purity is acquired in real time by mass spectrometry. The sampling period is After data collection, a moving average filter is applied to denoise the data. The filter window size is [not specified]. Eliminate the influence of electromagnetic interference and other noise to ensure the accuracy of deviation calculation; S3. Deviation Calculation: Based on the data collected in step S2 , and the steps set in S1 , Calculate beam intensity deviation and purity deviation The formula is:

[0011] ; S4. Dual-parameter integrated deviation and adaptive weight allocation: The beam intensity deviation weight coefficient is dynamically adjusted according to the current operating status of the ion source. And purity deviation weighting coefficient And calculate the overall deviation. The formula is:

[0012]

[0013]

[0014] in, Purity deviation influence factor ( ); S5. Adaptive control parameter calculation: based on comprehensive deviation and its rate of change, dynamically adjusting the proportional coefficient. Integral coefficient and differential coefficients The formula is:

[0015]

[0016]

[0017] in, The overall deviation change rate; S6. Control Output and Ion Source Parameter Adjustment: Based on the adaptive control parameters calculated in step S5, the adjustment amount of the ion source operating parameters is calculated using a PID control algorithm. The formula is:

[0018] And according to Adjust the operating parameters of the metal ion source, wherein the operating parameters include at least one of arc voltage, arc current, magnetic field strength and reaction gas flow rate; S7. Closed-loop iteration: Repeat steps S2-S6 until the actual beam intensity value is reached. With target value deviation And the actual purity value With target value deviation ( , (Preset precision threshold).

[0019] Furthermore, in step S1, the target value of the beam intensity and purity target value Based on the type of metal ions and process requirements: for aluminum ions and titanium ions... , For copper ions and tungsten ions, , .

[0020] Furthermore, in step S2, before the real-time acquisition of the dual parameters, a data preprocessing step is also included: processing the acquired actual beam intensity values... and actual purity value Noise denoising is achieved using a moving average filter, with the following formula:

[0021]

[0022] in, For the size of the filter window ( ), , The filtered values ​​represent the actual beam intensity and purity. The deviation calculation in step S3 uses the filtered values.

[0023] Furthermore, in step S3, the deviation calculation also includes normalization processing: normalizing the beam intensity deviation. and purity deviation Normalize to interval The normalization formula is:

[0024] in, , The upper and lower limits of the process allowable beam current intensity. , The process allows for upper and lower limits on purity.

[0025] Furthermore, in step S2, the beam intensity detector is a Faraday cup or a semiconductor detector, and the sampling frequency is... The mass spectrometer is a quadrupole mass spectrometer with a mass resolution of [missing information]. ( ).

[0026] Furthermore, in step S4, the purity deviation influence factor The range of values ​​is And when the beam intensity deviation With purity deviation When they are in the same direction (both positive or both negative), Increase To enhance purity control priority; when reversed, Decrease To balance the two-parameter control.

[0027] Furthermore, in step S1, the adaptive adjustment factors α, β, and γ are determined through optimization using a genetic algorithm: based on the ion source stabilization time. and control error For the objective function By initializing the population (population size N=50), selection (roulette wheel selection), and crossover (single-point crossover, crossover probability)... ), mutation (probability of mutation) Iterative optimization yields... , , .

[0028] Furthermore, the method also includes step S8: stability determination, when continuous N Each sampling period ( All of them satisfy and When the ion source is determined to have entered a stable operating state, control parameter adjustments are paused, and only data acquisition and status monitoring are performed; among which, , .

[0029] Furthermore, it also includes a data anomaly handling step: when the actual value of the beam intensity collected in step S2... Or actual purity value satisfy or ( , If the data is deemed abnormal, immediately switch to the backup control parameters. , , It also issues a fault warning signal.

[0030] Furthermore, the method can be applied to the control of sputtering ion sources or electron bombardment ion sources, wherein the metal ion source is a core component of a metal ion implanter or metal thin film deposition equipment, and the metal ions include aluminum ions, titanium ions, copper ions, tungsten ions, or tantalum ions.

[0031] Due to the adoption of the above technical solution, the beneficial effects of the present invention are as follows: 1. Resolve the coupling conflict between two parameters to achieve coordinated control. Through the dynamic weight allocation mechanism of step S4 ( , Based on real-time adjustment of purity deviation, this invention can automatically switch control priorities according to operating conditions: When the purity deviation is small ( ), Maintain the initial weight (e.g., 0.6) and focus on stabilizing the beam intensity; When the purity deviation increases ( ), Automatically increase to above 0.8 to prioritize suppressing purity deterioration.

[0032] Practical results: In the verification of copper ion implantation process, when the purity suddenly drops (from 99.9% to 99.0%), this method can detect the drop within 3 sampling cycles. Within 15 cycles, the purity was brought back to 99.8%, while the traditional fixed-weight method required 15 cycles, thus increasing the response speed by 5 times.

[0033] 2. Adaptive optimization of control parameters improves adaptability to nonlinear operating conditions. By adjusting the adaptive control parameters in step S5, this invention can dynamically adapt to the nonlinear and time-varying characteristics of the ion source: When the overall deviation is large ( ), Increase (enhance the proportion adjustment effect) Reduce (avoid integral saturation) and quickly suppress deviation; When the rate of change of deviation is large ( ), Increase (enhance differential damping effect) and reduce overshoot.

[0034] Actual results: In scenarios where the tungsten ion source operates for a long time (with 50% target consumption), the control error of this method is reduced by more than 75% compared with the traditional fixed parameter method.

[0035] 3. Data preprocessing and anomaly handling enhance system robustness. The moving average filtering (S2 step) can suppress beam intensity noise from ±3% to ±0.5% and purity noise from ±0.8% to ±0.1%, avoiding control oscillations caused by noise. The fault handling mechanism can switch to backup parameters within 0.2 seconds in the event of a sensor failure, ensuring that the ion source operating parameters do not exceed the limits and avoiding equipment damage.

[0036] Practical results: In the semiconductor aluminum ion implantation process, this method extends the continuous and stable operation time of the ion source from 8 hours in the traditional method to 24 hours, and reduces the process interruption rate by 70%.

[0037] 4. Strong process compatibility and wide range of applications This invention uses target values , The flexible configuration (S1 step) and the expansion of metal ion types can adapt to different process requirements: For processes requiring low to medium purity aluminum ions, titanium ions, etc. ( (Can be set to a larger size) Prioritize ensuring strength stability; For processes requiring high purity of copper ions, tungsten ions, etc. ( (Can be set to a larger size) To enhance the sensitivity to purity deviation.

[0038] Practical Results: In metal thin film deposition equipment, this method is compatible with the control of three ion sources: aluminum, titanium, and copper. The process yield (film thickness uniformity ±1%, resistivity fluctuation ±2%) is improved compared to traditional methods. .

[0039] In summary, this invention effectively addresses the core shortcomings of existing technologies through a collaborative design involving dual-parameter dynamic weight allocation, adaptive optimization of control parameters, and robust data processing. It provides a practical and feasible technical solution for high-precision control of metal ion sources and has significant industrial application value. Attached Figure Description

[0040] Figure 1 This is a flowchart of a dual-parameter adaptive control method for the intensity and purity of a metal ion source beam according to the present invention. Detailed Implementation

[0041] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. The embodiments are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0042] This embodiment uses ion source control in a copper ion implanter during semiconductor chip manufacturing as an application scenario. The target material is high-purity copper (99.999% purity), and the process targets are beam current intensity = 500μA (allowable fluctuation ±10μA) and ion purity = 99.9% (allowable fluctuation ±0.1%). The sampling period is... The aim is to solve the problems of purity and strength coupling conflict, slow dynamic response and low control accuracy in traditional control methods.

[0043] I. Overall Scheme of the Implementation Examples This embodiment is based on Figure 1 The control flow shown (parameter initialization module 10 → dual parameter acquisition and preprocessing module 20 → deviation calculation module 30 → dynamic weight allocation module 40 → adaptive control parameter optimization module 50 → control output module 60 → closed-loop iteration and stability judgment module 70) achieves dual-parameter coordinated control of the copper ion source through the following steps. Specific parameters and equipment selection are shown in Table 1: Table 1 Core parameters and equipment configuration of the embodiment

[0044] II. Detailed Implementation Steps of the Adaptive Control Method for Dual Parameters of Metal Ion Source Beam Intensity and Purity in This Embodiment S1. Parameter Initialization The following parameters are preset by controlling the host computer of the system (an industrial PC, configured with an Intel i5 processor and 8GB of memory): Target value: Beam intensity ,purity ; Initial weights: (Initial weight of beam intensity) (Initial purity weight), satisfying ; Purity deviation influencing factors: (Determined through preliminary experiments:) (At the same time, the sensitivity of weight adjustment is matched with the system response speed to avoid over-adjustment). Initial PID parameters: (Proportion coefficient) (Integral coefficient) (Differential coefficients) (Based on offline tuning of the static characteristic curve of the copper ion source to ensure initial control stability); Adaptive adjustment factor: (Proportional coefficient adjustment factor). (Integral coefficient adjustment factor). (Differential coefficient adjustment factor) (obtained through genetic algorithm optimization, with the objective function being "minimize control error + minimize settling time"). Accuracy threshold: Beam intensity deviation threshold (i.e., ±0.4%) purity deviation threshold (i.e., ±0.05%); Stability determination period N =10 (stable is determined if the threshold is met for 10 consecutive sampling periods).

[0045] S2. Real-time acquisition and preprocessing of dual parameters Data Acquisition: The copper ion source (model IS-500, sputtering ion source) was activated, and the real-time beam intensity was acquired using a Faraday cup (FC-200). Sampling frequency 10 Hz (i.e. Copper ions in the ion beam were collected using a quadrupole mass spectrometer (QMS-300). , ) and impurity ions (such as , ) strength ratio, calculate purity .

[0046] Moving average filtering: Apply a moving average filter to the raw data, with a filter window of n=5 (i.e., averaging the data from the current time step and the previous four time steps). The formula is as follows:

[0047] Effect: The original beam intensity data fluctuated due to electromagnetic interference. (±0.6%), after filtering volatility decreased (±0.1%); Original purity data fluctuated by ±0.3%, after filtering The fluctuation was reduced to ±0.05%, meeting the accuracy requirements for deviation calculation.

[0048] S3. Deviation Calculation and Normalization Deviation calculation: based on filtered data , Calculate the real-time deviation from the target value:

[0049] For example: at a certain moment ,but ; ,but (The negative sign indicates that the purity is lower than the target).

[0050] Normalization: To eliminate the impact of dimensional differences on weight allocation, the deviation is normalized to... Interval: Beam intensity normalization: based on process-allowed upper and lower limits , Normalization bias (because The denominator takes the upper limit of the deviation; if Take the denominator ); Purity normalization: based on process allowable range Normalization bias (The denominator is the difference between the lower limit of purity and the target value, which is 4.9%).

[0051] S4. Two-parameter dynamic weight allocation and comprehensive deviation calculation Based on the normalized absolute value of purity deviation Calculated using dynamic weight formula and : Beam intensity weighting :

[0052] Purity weight :

[0053] Note: Purity deviation at this time (Process warning value) The system automatically increases the purity weight to 0.706 and reduces the beam intensity weight to 0.294, prioritizing purity control.

[0054] Overall Deviation : Calculate the comprehensive control objective by combining normalized deviation and dynamic weights:

[0055] (A negative overall deviation indicates that the system needs to prioritize improving purity while slightly adjusting the beam intensity.)

[0056] S5. Adaptive Control Parameter Optimization Based on comprehensive deviation and its rate of change (through the previous moment) calculate, ), Optimize PID parameters: proportionality coefficient :

[0057] (because| Smaller, (Slightly larger than the initial value to enhance the sensitivity of proportional adjustment). Integral coefficient :

[0058] (The integral coefficient is slightly reduced to avoid integral saturation); Differential system :

[0059] (A positive rate of change in deviation indicates that the deviation is improving; a slight increase is made to enhance damping and reduce overshoot.)

[0060] S6. Control output and ion source parameter adjustment Based on the optimized PID parameters ( , , The adjustment amount of the ion source operating parameters is calculated using a PID algorithm. :

[0061] Calculations show that (The negative sign indicates that the arc voltage needs to be reduced to reduce the generation of impurity ions.) The control system drives the arc voltage regulator to reduce the arc voltage from the current 35.2 V to 35.15 V.

[0062] And according to Adjust the operating parameters of the metal ion source, wherein the operating parameters include at least one of arc voltage, arc current, magnetic field strength and reaction gas flow rate; S7. Closed-loop iteration The system repeats steps S2-S6: collecting new data. , Calculate the deviation → dynamically adjust the weights → optimize the PID parameters → output the control quantity. Within 10 consecutive sampling periods... and When the ion source is determined to be in a stable state, PID parameter adjustment is paused, and only data monitoring is performed.

[0063] III. Experimental Verification and Effect Comparison To verify the effectiveness of this invention, a comparative experiment was conducted on a copper ion implanter: the experimental group used the method of this invention, while the control group used the traditional "fixed-weight dual-parameter control" method. , (PID parameters fixed), the test scenarios included three typical operating conditions: "purity drop disturbance", "beam intensity fluctuation", and "long-term operational stability". The results are as follows: 1. Purity sudden drop disturbance scenario Disturbance setting: During stable operation, a brief introduction of a trace amount of oxygen (simulating a vacuum system leak) causes the purity to drop sharply from 99.9% to 99.0%. ); Control group: Under fixed weights, the system requires 15 sampling periods (1.5 s) to restore the purity to 99.8%; Experimental group: In the method of this invention, increased purity deviation leads to... The beam intensity was increased to 0.92, and the arc voltage and magnetic field strength were adjusted first. It recovered to 99.8% in just 3 sampling cycles (0.3 s), with a 5-fold increase in response speed and minimal beam intensity fluctuation. (Control group) ).

[0064] 2. Beam intensity fluctuation scenario Perturbation settings: Inhomogeneity of the target surface causes beam intensity to... Fluctuations within a range; Control group: Under fixed weights, the intensity fluctuation control accuracy is (±0.64%) Experimental group: when purity is stable ( ), The value has been restored to 0.58, with a focus on stabilizing the intensity and improving fluctuation control precision. (±0.16%), meeting the requirements for high-precision injection.

[0065] 3. Long-term operational stability scenarios Test conditions: Continuous operation for 24 hours, monitoring long-term fluctuations in beam intensity and purity; Control group: As the target material was consumed (50%), the intensity fluctuation decreased from... Increase to The purity fluctuation increased from ±0.08% to ±0.25%; Experimental group: Optimization through adaptive PID parameters The version was dynamically adjusted from 2.0 to 2.3. (Adjusted from 0.5 to 0.4), intensity fluctuations were consistently controlled within... Within a certain range, the purity fluctuation is within ±0.06%, and the long-term stability is improved by more than 70%.

[0066] In this embodiment, parameters (such as...) , , It can be adjusted according to specific process requirements: for example, for gold ion implantation with high purity requirements ( ), can be increased to 1.5 (to improve purity deviation sensitivity); for high beam scenarios (such as The moving average window n can be increased to 8 (to enhance the filtering effect). Those skilled in the art can adjust the parameters according to the ion type (aluminum, titanium, tungsten, etc.), equipment model, and process objectives through pre-experiments or simulation tools (such as MATLAB / Simulink), all of which fall within the protection scope of this invention.

[0067] Through the above embodiments, those skilled in the art can clearly understand the technical solution and implementation details of the present invention. Its core technical features, such as dynamic weight allocation, adaptive parameter optimization, and robust data processing, effectively solve the key problem of dual-parameter coupling control of metal ion sources, and have significant industrial applicability and inventiveness.

Claims

1. A method for double-parameter adaptive control of beam intensity and purity of a metal ion source, characterized in that, Comprising the steps of: S1. Parameter initialization: according to the type of metal ion source and the target process requirement, set the target value of beam current intensity and the target value of purity , initialize the initial control coefficient 、 、 , the adaptive adjustment factor , and the initial weight coefficient 、 ( + =1) S2. Real-time acquisition of two parameters: real-time acquisition of the actual value of the beam intensity output by the metal ion source through the beam intensity detector Real-time acquisition of the actual value of the ion beam purity through the mass spectrometer The sampling period is ; S3. Deviation Calculation: Based on the data collected in step S2 , and the steps set in S1 , Calculate beam intensity deviation and purity deviation The formula is: ； S4. Double-parameter comprehensive deviation and weight adaptive allocation: dynamically adjust the beam intensity deviation weight coefficient according to the current running state of the ion source and the purity deviation weight coefficient , and calculate the comprehensive deviation , the formula is: wherein, is the purity bias influence factor (PBI) ); S5. Adaptive control parameter calculation: based on the integrated deviation and its rate of change, dynamically adjust the proportional coefficient , integral coefficient and differential coefficient , the formula is: wherein, is the integrated deviation rate of change; S6. Control amount output and ion source parameter adjustment: according to the adaptive control parameters calculated in step S5, the ion source operation parameter adjustment amount is calculated through the PID control algorithm , the formula is: and according to adjusting operating parameters of the metal ion source, the operating parameters including at least one of arc voltage, arc current, magnetic field strength, and reaction gas flow rate; S7. Closed loop iteration: repeat steps S2-S6 until the beam current actual value deviates from the target value by more than a preset accuracy threshold , and the purity actual value deviates from the target value by more than a preset accuracy threshold . , .

2. The method of claim 1, wherein, In step S1, the beam intensity target value and the purity target value are determined based on the metal ion species and process requirements: for aluminum ions, titanium ions, , ; for copper ions, tungsten ions, , .

3. The method of claim 1, wherein, In step S2, the double-parameter real-time acquisition further comprises a data preprocessing step before the acquisition of the beam intensity actual value and the purity actual value The sliding average filter is used for denoising, and the filter formula is: wherein, is the filter window size (N), , is the filtered beam intensity and purity actual value, which is used in the deviation calculation in step S3.​ 4. The method of claim 1, wherein, In step S3, the deviation calculation further comprises a normalization process: normalizing the beam intensity deviation and the purity deviation to the interval , with the normalization formula being: wherein , are process allowed upper and lower limits for beam intensity, , are process allowed upper and lower limits for purity.

5. The method of claim 1, wherein, In step S2, the beam intensity detector is a Faraday cup or a semiconductor detector, and the sampling frequency is 10 kHz ; the mass spectrometer is a quadrupole mass spectrometer, and the mass resolution is 1 ( ).

6. The method of claim 1, wherein, In step S4, the purity deviation influence factor is in the range of , and when the beam intensity deviation and the purity deviation are in the same direction, is increased to enhance the purity control priority; when in the opposite direction, is decreased to balance the double-parameter control.

7. The method of claim 1, wherein, In step S1, the adaptive adjustment factors a, b, g are determined by genetic algorithm optimization: with ion source stability time and control error as the objective function , by initializing the population (population size ), selection (roulette wheel selection), crossover (single-point crossover, crossover probability ), mutation (mutation probability ) iterative optimization, get , , .

8. The method of claim 1, wherein, It also includes step S8: stability determination, when continuous N Each sampling period ( All of them satisfy and When the ion source is determined to have entered a stable operating state, control parameter adjustments are paused, and only data acquisition and status monitoring are performed; among which, , .

9. The method of claim 1, wherein, Also included is a data abnormality processing step: when the actual value of the beam intensity or the actual value of the purity satisfies or ( , ), it is determined that the data is abnormal, immediately switches to the backup control parameter , , , and issues a fault warning signal.

10. The method according to any one of claims 1 to 9, characterized in that, The application is applied to the control of sputtering ion source or electron impact ion source, the metal ion source is the core component of metal ion implanter or metal thin film deposition equipment, and the metal ion includes aluminum ion, titanium ion, copper ion, tungsten ion or tantalum ion.

Images