A control method, control system and control device of a reactive sputtering process

Abstract

The application discloses a control method, a control system and a control device of a reactive sputtering process. The control method of the reactive sputtering process comprises the following steps: collecting real-time data of a target voltage in a preset initial time period after the reactive sputtering process is started, calculating a standard deviation of the target voltage, and setting a dynamic gradient threshold value according to the standard deviation; after the preset initial time period, the current target voltage is acquired in real time at a preset sampling period, and a real-time change rate of the target voltage is calculated; the absolute value of the real-time change rate is compared with the dynamic gradient threshold value; when the absolute value of the real-time change rate is greater than the dynamic gradient threshold value, a target poisoning precursor is determined, and a negative gas flow pulse signal is output to a gas mass flow controller to reduce the flow of the reaction gas. The application can respond in advance, avoid target poisoning, reduce the false trigger rate and improve the adaptability.

Description

Technical Field

[0001] This invention relates to the field of magnetron sputtering technology, and in particular to a control method, control system and control device for reactive sputtering processes. Background Technology

[0002] Reactive magnetron sputtering is an important technique for preparing compound thin films (such as nitride and oxide films). In this process, a reactive gas (such as oxygen, nitrogen, or air) is introduced into a vacuum chamber, causing the sputtered metal atoms to react chemically with the reactive gas and deposit onto the substrate to form a compound film. However, target poisoning is a common problem in reactive sputtering: when the reactive gas flow rate is too high, an insulating compound layer forms on the target surface, leading to a sharp decrease in sputtering rate and frequent arcing, which seriously affects film quality and process stability.

[0003] To suppress target poisoning, various control strategies have been developed in existing technologies. For example, Jimei University's invention patent (publication number: CN120924901A) discloses a method and system for combating target poisoning based on multi-stage pulse reactive sputtering, proposing a control scheme using air as the sole reactive gas. This technology divides the reactive sputtering process into three stages: the first stage uses low-power pre-sputtering to stabilize ignition and clean the target surface; the second stage switches to medium power and duty cycle to increase nitrogen ionization and suppress oxygen ionization; the third stage is a sputtering voltage feedback control stage, where when the sputtering voltage drop reaches 15% of the initial voltage, the power is automatically increased and the duty cycle is reduced, using high-power pulses to strip the insulating compound layer from the target surface. Once the voltage recovers to more than 85% of the initial value, it returns to the parameters of the second stage. This technology eliminates the need for complex equipment, reducing gas costs and improving target utilization to some extent.

[0004] However, the aforementioned existing technologies still have the following shortcomings in practical applications: hysteresis problem: using a 15% drop in absolute voltage as the trigger threshold, when the voltage drop reaches this level, the target surface is often already covered by a large area of ​​insulating compound layer, entering an avalanche-like poisoning state; crude judgment criteria and risk of misjudgment: using a fixed proportional coefficient cannot effectively distinguish between factors such as target poisoning, power supply ripple, vacuum fluctuations, and thermal drift, easily leading to false triggering; limited applicability of gases and materials: limiting the use of air as the single reactant gas makes it impossible to prepare compound films with precise requirements for nitrogen-oxygen ratios, and it is also difficult to apply to pure nitride or pure oxide coating processes.

[0005] Therefore, how to achieve rapid, precise, and adaptive control of reactive sputtering processes, intervene at the initial signs of target poisoning to avoid deep poisoning, while being compatible with various reactive gases and metal targets and reducing dependence on complex power supply hardware, is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] This invention provides a control method, control system, and control device for reactive sputtering processes to solve the problems of delayed target poisoning response, high misjudgment rate, and poor adaptability caused by the use of a fixed absolute voltage threshold in the prior art.

[0007] According to one aspect of the present invention, a method for controlling a reactive sputtering process is provided, the method comprising: Within a preset initial time period after the reactive sputtering process is started, real-time data of the target voltage is collected, the standard deviation of the target voltage is calculated, and a dynamic gradient threshold is set based on the standard deviation. After the preset initial time period, the current target voltage is acquired in real time with a preset sampling period, and the real-time rate of change of the target voltage is calculated. The absolute value of the real-time rate of change is compared with the dynamic gradient threshold; When the absolute value of the real-time rate of change is greater than the dynamic gradient threshold, it is determined to be a precursor to target poisoning, and a negative gas flow pulse signal is output to the gas mass flow controller to reduce the flow rate of the reactant gas.

[0008] Optionally, the control method for the reactive sputtering process also includes: When the absolute value of the real-time rate of change is less than or equal to the dynamic gradient threshold, the negative gas flow pulse signal is not output or is stopped from being output to the gas mass flow controller, and the flow rate of the reactant gas is not adjusted.

[0009] Optionally, when the absolute value of the real-time rate of change is less than the dynamic gradient threshold, after stopping the output of the negative gas flow pulse signal to the gas mass flow controller and not adjusting the flow rate of the reactant gas, the method further includes: Monitor the deviation between the current target voltage and the target reference voltage; If the deviation value exceeds the preset steady-state range, the gas mass flow controller is controlled to perform slow flow fine-tuning to compensate for voltage drift; wherein the flow adjustment rate is 0.1 sccm.

[0010] Optionally, the process of acquiring real-time target voltage data and calculating the standard deviation of the target voltage includes: The real-time data of the target voltage collected is low-pass filtered using a moving average filtering algorithm, and the standard deviation of the target voltage is calculated based on the processed real-time data of the target voltage. The step of acquiring the current target voltage in real time with a preset sampling period and calculating the real-time rate of change of the target voltage includes: The current target voltage is low-pass filtered using a moving average filtering algorithm, and the real-time rate of change of the target voltage is calculated based on the processed current target voltage.

[0011] Optionally, the setting of the dynamic gradient threshold specifically includes: The dynamic gradient threshold is determined based on the standard deviation of the target voltage and a preset gradient coefficient.

[0012] Optionally, the amplitude of the negative gas flow pulse signal is proportional to the magnitude of the real-time rate of change.

[0013] Optionally, the pulse width of the negative gas flow pulse signal is 1ms-100ms, and the pulse amplitude of the negative gas flow pulse signal is 10%-50% of the current set flow rate of the reaction gas.

[0014] Optionally, the preset initial time period is the first 1-3 seconds after the process begins; The real-time rate of change is obtained by performing discrete difference operations on the filtered target voltage data.

[0015] According to another aspect of the present invention, a control system for a reactive sputtering process is provided, the control system comprising: Voltage acquisition module, used to acquire target voltage in real time during sputtering process; A gas mass flow controller is used to control the input flow rate of the reaction gas. The central processing unit is connected to the voltage acquisition module and the gas mass flow controller; The central processing unit is used to execute the control method for the reactive sputtering process as described in any one of the preceding aspects.

[0016] According to another aspect of the present invention, a control device for a reactive sputtering process is provided, the control device comprising: The dynamic gradient threshold generation module is used to collect real-time data of the target voltage during a preset initial time period after the reactive sputtering process is started, calculate the standard deviation of the target voltage, and set the dynamic gradient threshold based on the standard deviation. The real-time rate of change calculation module is used to acquire the current target voltage in real time after the preset initial time period with a preset sampling period, and to calculate the real-time rate of change of the target voltage. A comparison module is used to compare the absolute value of the real-time rate of change of the target voltage with the dynamic gradient threshold. The reactive gas flow control module is used to determine that when the absolute value of the real-time change rate is greater than the dynamic gradient threshold, it is a precursor to target poisoning. The module then outputs a negative gas flow pulse signal to the gas mass flow controller to reduce the flow rate of the reactive gas.

[0017] The technical solution of this invention utilizes the target voltage change rate as the core judgment indicator. It can rapidly identify early signs of poisoning within a timeframe when the poisoned layer on the target surface has just begun to form and before the absolute voltage value has significantly deviated. It immediately cuts off the continued supply of reactive gas via a negative gas flow pulse, thereby maintaining the process at its optimal operating point in the metal sputtering mode or transition zone. This avoids the lag inherent in traditional absolute voltage feedback strategies, providing an advanced response and preventing target poisoning. A gradient threshold is dynamically set based on the natural fluctuation standard deviation of the target voltage during the initial stable phase. This threshold can adaptively adjust with changes in process conditions, exhibiting strong anti-interference capabilities and significantly reducing the false trigger rate. This control method has strong compatibility and requires no power supply modification. In summary, this invention solves the problems of delayed target poisoning response, high false trigger rate, and poor adaptability caused by using a fixed absolute voltage threshold in existing technologies. It offers the beneficial effects of advanced response to avoid target poisoning, reduced false trigger rate, and strong adaptability.

[0018] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a control method for a reactive sputtering process provided according to an embodiment of the present invention; Figure 2 This is a flowchart of another control method for reactive sputtering process provided by an embodiment of the present invention; Figure 3 This is a flowchart of another control method for reactive sputtering process provided by an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a control system for a reactive sputtering process according to an embodiment of the present invention; Figure 5 This is a schematic diagram of a control device for a reactive sputtering process provided according to an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Figure 1 This is a flowchart of a control method for a reactive sputtering process according to an embodiment of the present invention. This method can be executed by a control device for the reactive sputtering process. Figure 1 As shown, the method includes: S110. During the preset initial time period after the reactive sputtering process is started, collect real-time data of the target voltage, calculate the standard deviation of the target voltage, and set a dynamic gradient threshold based on the standard deviation.

[0024] Specifically, this embodiment provides an intelligent stabilization method for industrial magnetron sputtering coating equipment, taking the reaction sputtering of titanium dioxide thin films in an oxygen / argon mixed atmosphere as an example. The equipment includes: a DC magnetron sputtering power supply (with a millisecond-level voltage sampling output interface), a vacuum chamber, a mass flow controller (MFC), which controls the oxygen flow, a data acquisition card, and a central processing unit (PLC or industrial computer).

[0025] The control method for reactive sputtering is not dependent on a specific reactive gas (applicable to oxygen, nitrogen, air, etc.) and is applicable to various metal targets with hysteresis characteristics, such as titanium, aluminum, and zirconium. More importantly, this invention only requires a conventional DC power supply to provide a millisecond-level voltage sampling interface, without the need for a power supply with high-frequency pulse output or complex multi-stage power switching capabilities, and can be directly deployed on existing industrial DC sputtering equipment.

[0026] The initial time period is preset to the first 1-3 seconds after the process starts. During the first 1-3 seconds after the reactive sputtering process starts, the system is in pure metal sputtering mode (no oxygen or very small amount of oxygen is introduced). After the system power climbs to the set value, the central processing unit continuously acquires the target voltage signal at a sampling period of 1ms for 1-3 seconds, obtaining a total of 3000 voltage data points. The arithmetic mean of these voltage data points is calculated. and the standard deviation of natural fluctuations Set a dynamic gradient threshold. (gradient coefficients) (It can be adjusted between 3 and 10 based on the actual process noise level). The dynamic gradient threshold represents the upper limit of the allowable fluctuation of the target voltage change rate under normal process conditions.

[0027] S120. After a preset initial time period, the current target voltage is acquired in real time with a preset sampling period, and the real-time rate of change of the target voltage is calculated.

[0028] Specifically, after a preset initial time period, the current target voltage is continuously acquired in real time with a sampling period of 10-100ms, and the voltage change rate of the current target voltage is calculated. .

[0029] S130. Compare the absolute value of the real-time rate of change with the dynamic gradient threshold.

[0030] Specifically, when When, is defined as a slow, normal drift of the system, when This is defined as a sudden precursor to poisoning. For example: Scenario A (normal drift): The target material etching deepens, and the target voltage... The voltage drops slowly by 2V over one minute. At this point, ≈-0.03V / s, judgment: Ignore this and do not adjust the flow rate. Scenario B (early signs of poisoning): Tiny nitrided areas appear on the target surface, and the voltage suddenly drops by 0.5V within 0.1 seconds. At this time... =-5.0V / s, judgment: This triggers an alert. It should be noted that... It is not a constant value and can be set according to the actual process type.

[0031] S140. When the absolute value of the real-time rate of change is greater than the dynamic gradient threshold, it is determined to be a precursor to target poisoning. A negative gas flow pulse signal is output to the gas mass flow controller to reduce the flow rate of the reactant gas.

[0032] Specifically, the target voltage will drop rapidly in the event of poisoning. Therefore, when the absolute value of the real-time rate of change exceeds the dynamic gradient threshold, a steep drop in voltage can be identified as a precursor to poisoning. When the absolute value of the real-time rate of change exceeds the dynamic gradient threshold, it indicates that the target voltage is experiencing an abnormally steep drop (negative gradient). At this point, the system determines that an insulating compound layer has just begun to form on the target surface, i.e., a precursor to poisoning. Upon determining a precursor to target poisoning, the central processing unit can output a negative gas flow pulse to the oxygen MFC within a microsecond or millisecond reaction time: the pulse width is set to 5ms, and the pulse amplitude is 20% of the current set oxygen flow rate. This negative pulse causes the MFC to instantaneously reduce the oxygen flow rate by 20% within 5ms, and then automatically restore the original set value. Due to the extremely short pulse duration, the overall pressure within the chamber remains almost unchanged, but this is sufficient to interrupt the chain reaction of poisoning and inhibit further growth of the compound layer. Under this intervention, the target voltage typically recovers to normal levels within tens of milliseconds, completely avoiding downstream avalanche poisoning. By introducing the voltage change gradient as the core control variable, the system can instantly identify the precursors of poisoning layer formation and intervene at the millisecond level when the absolute value of the voltage has not deviated significantly but the slope of the change is abnormally steep, thereby avoiding target poisoning and ensuring the quality of the deposited film.

[0033] At the microsecond mark when the poison layer is just beginning to form, the voltage exhibits a rapid change. Even if the absolute voltage value is still within the normal range, the system can identify and intervene in advance to promptly suppress the avalanche phenomenon (without adjusting power parameters, only transient suppression of the MFC flow rate is performed, and a negative gas flow pulse is output through millisecond-level calculation to instantly cut off the reactive gas). By introducing a dynamic gradient threshold and microsecond to millisecond-level gas flow pulse suppression, the problem of target poisoning response lag in reactive sputtering is fundamentally solved, and it possesses excellent anti-interference and versatility.

[0034] The technical solution of this invention utilizes the target voltage change rate as the core judgment indicator. It can rapidly identify early signs of poisoning within a timeframe when the poisoned layer on the target surface has just begun to form and before the absolute voltage value has significantly deviated. It immediately cuts off the continued supply of reactive gas via a negative gas flow pulse, thereby maintaining the process at its optimal operating point in the metal sputtering mode or transition zone. This avoids the lag inherent in traditional absolute voltage feedback strategies, providing an advanced response and preventing target poisoning. A gradient threshold is dynamically set based on the natural fluctuation standard deviation of the target voltage during the initial stable phase. This threshold can adaptively adjust with changes in process conditions, exhibiting strong anti-interference capabilities and significantly reducing the false trigger rate. This control method has strong compatibility and requires no power supply modification. In summary, this invention solves the problems of delayed target poisoning response, high false trigger rate, and poor adaptability caused by using a fixed absolute voltage threshold in existing technologies. It offers the beneficial effects of advanced response to avoid target poisoning, reduced false trigger rate, and strong adaptability.

[0035] Figure 2 This is a flowchart of another control method for a reactive sputtering process provided according to an embodiment of the present invention, see reference. Figure 2 The control methods for reactive sputtering processes include: S210. During the preset initial time period after the reactive sputtering process is started, collect real-time data of the target voltage, calculate the mean and standard deviation of the natural fluctuation of the target voltage, and set a dynamic gradient threshold based on the standard deviation.

[0036] S220. After a preset initial time period, the current target voltage is acquired in real time with a preset sampling period, and the real-time rate of change of the target voltage is calculated.

[0037] S230. Compare the absolute value of the real-time rate of change with the dynamic gradient threshold.

[0038] S240. When the absolute value of the real-time rate of change is less than or equal to the dynamic gradient threshold, the negative gas flow pulse signal is not output or is stopped from being output to the gas mass flow controller, and the flow rate of the reactant gas is not adjusted.

[0039] Specifically, when the absolute value of the real-time rate of change is less than or equal to the dynamic gradient threshold, normal drift occurs and the flow rate is not adjusted.

[0040] Figure 3 This is a flowchart of another control method for a reactive sputtering process provided according to an embodiment of the present invention, see reference. Figure 3 After step S240, the method further includes: S250: Monitor the deviation between the current target voltage and the target reference voltage.

[0041] S260. If the deviation value exceeds the preset steady-state range, the gas mass flow controller is controlled to adjust the flow rate to compensate for voltage drift; wherein the flow rate adjustment rate is 0.1 sccm.

[0042] Specifically, the target reference voltage refers to the ideal voltage value corresponding to the optimal process point in the transition region on the hysteresis curve obtained from previous process experiments. The preset steady-state range is a dead zone threshold (for example, set to ±1% to ±3% of the target reference voltage) that is comprehensively set based on the allowable tolerance of the quality of the prepared film (such as refractive index, transmittance, etc.) and the level of low-frequency fluctuations in the system background, in order to avoid over-adjustment caused by normal small fluctuations.

[0043] If the voltage drifts slowly and normally over a long period (corresponding to scenario A above), although a gradient alarm will not be triggered, the absolute value will deviate from the target. In this case, the gas mass flow controller will initiate slow flow integral regulation, for example, fine-tuning by 0.1 sccm (standard milliliters / minute) to counteract thermal drift.

[0044] After the flow rate of the reactant gas is not adjusted, the system enters steady-state monitoring mode. At this time, the deviation ΔU between the current target voltage and the target reference voltage is calculated. If |ΔU| exceeds the preset steady-state range, it indicates that voltage drift is caused by changes in target surface roughness due to target etching, increased cavity temperature, changes in pumping speed, gas pressure fluctuations, or temperature drift of internal power supply components. In this case, the central processing unit sends a slow adjustment command to the MFC, gradually correcting the oxygen flow rate at a rate of 0.1% per 100ms until the deviation returns to the steady-state range, thus compensating for voltage drift caused by systematic factors such as target consumption and changes in cavity state. This slow adjustment is independent of the aforementioned fast pulse suppression, and the two together constitute a hierarchical control architecture.

[0045] This embodiment, based on the rapid pulse suppression of poisoning precursors, also introduces a slow flow fine-tuning mechanism to compensate for voltage drift caused by target etching, cavity temperature changes, etc., thus achieving a balance between transient anti-poisoning and long-term process stability.

[0046] Optionally, real-time data of the target voltage is collected, and the standard deviation of the target voltage is calculated, including: The real-time target voltage data is low-pass filtered using a moving average filtering algorithm, and the standard deviation of the target voltage is calculated based on the processed real-time target voltage data. The process of acquiring the current target voltage in real time with a preset sampling period and calculating the real-time rate of change of the target voltage includes: using a moving average filtering algorithm to perform low-pass filtering on the acquired current target voltage, and calculating the real-time rate of change of the target voltage based on the processed current target voltage.

[0047] Specifically, during the actual sputtering process, the central processing unit continues to acquire real-time target voltage at a preset initial time period. To eliminate power supply ripple (typically 20~50kHz) and high-frequency plasma oscillation noise, a moving average filtering algorithm is used: the arithmetic mean of the most recent N=10 sampling points is taken as the filtered voltage value at the current moment. This filtering process does not introduce significant delay but effectively filters out high-frequency glitches.

[0048] To accurately reflect the natural fluctuation level of the target voltage and avoid false amplification of the standard deviation due to noise interference, the system performs a moving average filter on the raw acquired data. This filtering is applied to all raw data points to obtain the filtered voltage sequence. Then, the mean and standard deviation are calculated based on the filtered voltage sequence. Finally, a dynamic gradient threshold is set.

[0049] During the actual sputtering process, the system continues to acquire the current target voltage at a sampling period of a preset initial time interval. Similarly, before calculating the real-time rate of change, the current target voltage is first filtered using a moving average. Then, based on two adjacent filtered voltage values, the real-time rate of change of the target voltage is calculated using discrete difference operations.

[0050] Existing technologies use a fixed proportional coefficient as the criterion, which cannot distinguish between power supply ripple, vacuum fluctuations, thermal drift, and true target poisoning, easily leading to misjudgment or failure after process parameter drift. This embodiment establishes a dynamic gradient model, which automatically calculates the current reference voltage drift curve, reacting only to sudden high gradient changes, automatically filtering out slow thermal drift and high-frequency power supply noise, and achieving precise control (the threshold adopts a dynamic following mechanism based on the standard deviation of the background noise, eliminating false triggering caused by fixed empirical values).

[0051] Optionally, the setting of the dynamic gradient threshold specifically includes: The dynamic gradient threshold is determined based on the standard deviation of the target voltage and the preset gradient coefficient.

[0052] Specifically, based on the natural fluctuation standard deviation of the target voltage during the preset initial time period. and preset gradient coefficients Determine the dynamic setting of the gradient threshold Furthermore, this threshold can be adaptively adjusted to follow changes in process conditions. Dynamic gradient threshold. The adaptive noise immunity threshold setting is always aligned with the noise level of the current operating condition (i.e., the standard deviation of the natural voltage fluctuation). It is directly linked. When the system noise increases, the threshold is automatically widened; when the system is stable, the threshold is automatically tightened, thus effectively avoiding false triggering caused by fluctuations in the system's background noise.

[0053] Compared with the fixed percentage threshold of existing technologies, this embodiment can effectively distinguish between normal fluctuations such as power supply ripple and thermal drift and true precursors of target poisoning, significantly reducing the false trigger rate and having strong anti-interference capabilities.

[0054] Optionally, the amplitude of the negative gas flow pulse signal is proportional to the magnitude of the real-time rate of change.

[0055] Optionally, the pulse width of the negative gas flow pulse signal is 1ms-100ms, and the pulse amplitude of the negative gas flow pulse signal is 10%-50% of the current set flow rate of the reactant gas.

[0056] Specifically, once a high gradient voltage change rate is detected... The algorithm's action is to immediately output a negative gas flow pulse signal proportional to its magnitude (e.g., the nitrogen gas flow rate is instantly reduced by a set value). This allows the tiny nitrided spots on the target surface to be quickly sputtered away, and the voltage slope to return to positive quickly.

[0057] Optionally, the initial time period is preset to the first 1-3 seconds after the process begins; The real-time rate of change is obtained by performing discrete difference operations on the filtered target voltage data.

[0058] Specifically, in the first 1-3 seconds after the process begins (in metal sputtering mode), the system records the average voltage. and the standard deviation of natural fluctuations Discrete differential calculations are performed on the filtered target voltage data to obtain the real-time rate of change of the target voltage. By utilizing the mathematical characteristic that gradient signals are insensitive to voltage drift, slow voltage changes caused by target wear or thermal effects can be automatically filtered out, and only instantaneous changes deviating from steady state can be responded to, thereby achieving adaptive tracking of the new process window after drift.

[0059] Figure 4 This is a schematic diagram of a control system for a reactive sputtering process according to an embodiment of the present invention. (Refer to...) Figure 4 The embodiments of the present invention also provide a control system for a reactive sputtering process, the control system for the reactive sputtering process comprising: Voltage acquisition module 301 is used to acquire the target voltage in real time during the sputtering process; Gas mass flow controller 302 is used to control the input flow rate of the reaction gas; The central processing unit 303 is connected to the voltage acquisition module 301 and the gas mass flow controller 302; The central processing unit 303 is used to execute the control method of the reactive sputtering process provided in any embodiment of the present invention.

[0060] Specifically, the voltage acquisition module 301 can employ a 16-bit high-precision analog-to-digital converter, connected to the analog voltage output of the magnetron sputtering power supply, with a sampling frequency of 1kHz. The gas mass flow controller 302 adopts a thermal MFC, with a control range of 0~200sccm and a response time of less than 10ms. The central processing unit 303 can employ an STM32F4 series microcontroller, integrating all algorithm modules of the above control methods, including: an initial self-learning module, a moving average filtering module, a differential calculation module, a gradient comparison module, a pulse generation module, and a PID slow-speed adjustment module. The central processing unit 303 can also be a PLC or an industrial computer. The industrial computer can integrate or run a human-machine interface, which can be a touch screen, used to display real-time target voltage, rate of change, current threshold, alarm records, and allow operators to adjust parameters (such as gradient coefficient k, pulse width, pulse amplitude, etc.). The central processing unit 303 executes the method steps as described in the above embodiments to achieve intelligent stabilization control of the reactive sputtering process.

[0061] Since the control system of the reactive sputtering process is used to execute the control method of the reactive sputtering process provided in any embodiment of the present invention, the beneficial effects of the control system of the reactive sputtering process and the control method of the reactive sputtering process are the same, and will not be repeated here.

[0062] Figure 5 This is a schematic diagram of a control device for a reactive sputtering process according to an embodiment of the present invention. (Refer to...) Figure 5 The embodiments of the present invention also provide a control device for a reactive sputtering process, the control device for the reactive sputtering process comprising: The dynamic gradient threshold generation module 401 is used to collect real-time data of the target voltage, calculate the standard deviation of the target voltage, and set the dynamic gradient threshold according to the standard deviation during a preset initial time period after the reactive sputtering process is started. The real-time change rate calculation module 402 is used to acquire the current target voltage in real time after a preset initial time period and at a preset sampling period, and to calculate the real-time change rate of the target voltage. Comparison module 403 is used to compare the absolute value of the real-time rate of change of the target voltage with the dynamic gradient threshold. The reaction gas flow control module 404 is used to determine that when the absolute value of the real-time change rate is greater than the dynamic gradient threshold, it is a sign of target poisoning. It then outputs a negative gas flow pulse signal to the gas mass flow controller to reduce the flow rate of the reaction gas.

[0063] The control device for reactive sputtering process provided in the embodiments of the present invention can execute the control method for reactive sputtering process provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0064] Continue to refer to Figure 5 The control device for the reactive sputtering process also includes: The normal drift module 405 is used to prevent or stop outputting negative gas flow pulse signals to the gas mass flow controller when the absolute value of the real-time rate of change is less than or equal to the dynamic gradient threshold, and to prevent the flow rate of the reactant gas from being adjusted.

[0065] The integral steady-state correction module 406 is used to monitor the deviation between the current target voltage and the target reference voltage. If the deviation exceeds the preset steady-state range, the gas mass flow controller will adjust the flow rate to compensate for voltage drift.

[0066] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for controlling a reactive sputtering process, characterized in that, include: Within a preset initial time period after the reactive sputtering process is started, real-time data of the target voltage is collected, the standard deviation of the target voltage is calculated, and a dynamic gradient threshold is set based on the standard deviation. After the preset initial time period, the current target voltage is acquired in real time with a preset sampling period, and the real-time rate of change of the target voltage is calculated. The absolute value of the real-time rate of change is compared with the dynamic gradient threshold; When the absolute value of the real-time rate of change is greater than the dynamic gradient threshold, it is determined to be a precursor to target poisoning, and a negative gas flow pulse signal is output to the gas mass flow controller to reduce the flow rate of the reactant gas. The process of acquiring real-time target voltage data and calculating the standard deviation of the target voltage includes: The real-time data of the target voltage collected is low-pass filtered using a moving average filtering algorithm, and the standard deviation of the target voltage is calculated based on the processed real-time data of the target voltage. The step of acquiring the current target voltage in real time with a preset sampling period and calculating the real-time rate of change of the target voltage includes: The current target voltage is low-pass filtered using a moving average filtering algorithm, and the real-time rate of change of the target voltage is calculated based on the processed current target voltage. Specifically, the setting of the dynamic gradient threshold includes: The dynamic gradient threshold is determined based on the standard deviation of the target voltage and a preset gradient coefficient.

2. The method according to claim 1, characterized in that, Also includes: When the absolute value of the real-time rate of change is less than or equal to the dynamic gradient threshold, the negative gas flow pulse signal is not output or is stopped from being output to the gas mass flow controller, and the flow rate of the reactant gas is not adjusted.

3. The method according to claim 2, characterized in that, When the absolute value of the real-time rate of change is less than the dynamic gradient threshold, the output of the negative gas flow pulse signal to the gas mass flow controller is stopped, and the flow rate of the reactant gas is not adjusted. The process also includes: Monitor the deviation between the current target voltage and the target reference voltage; If the deviation value exceeds the preset steady-state range, the gas mass flow controller is controlled to adjust the flow rate to compensate for voltage drift; wherein the rate of flow adjustment is 0.1 sccm.

4. The method according to claim 1, characterized in that, The amplitude of the negative gas flow pulse signal is proportional to the magnitude of the real-time rate of change.

5. The method according to claim 1 or 4, characterized in that, The pulse width of the negative gas flow pulse signal is 1ms-100ms, and the pulse amplitude of the negative gas flow pulse signal is 10%-50% of the current set flow rate of the reactant gas.

6. The method according to claim 1, characterized in that, The preset initial time period is the first 1-3 seconds after the process begins; The real-time rate of change is obtained by performing discrete difference operations on the filtered target voltage data.

7. A control system for a reactive sputtering process, characterized in that, include: Voltage acquisition module, used to acquire target voltage in real time during the sputtering process; A gas mass flow controller is used to control the input flow rate of the reaction gas. The central processing unit is connected to the voltage acquisition module and the gas mass flow controller; The central processing unit is used to execute the control method for the reactive sputtering process as described in any one of claims 1-6.

8. A control device for a reactive sputtering process, characterized in that, include: The dynamic gradient threshold generation module is used to collect real-time data of the target voltage during a preset initial time period after the reactive sputtering process is started, calculate the standard deviation of the target voltage, and set the dynamic gradient threshold based on the standard deviation. The real-time rate of change calculation module is used to acquire the current target voltage in real time after the preset initial time period with a preset sampling period, and to calculate the real-time rate of change of the target voltage. The comparison module is used to compare the absolute value of the real-time rate of change of the target voltage with the dynamic gradient threshold. The reactive gas flow control module is used to determine that when the absolute value of the real-time change rate is greater than the dynamic gradient threshold, it is a precursor to target poisoning, and outputs a negative gas flow pulse signal to the gas mass flow controller to reduce the flow rate of the reactive gas. The process of acquiring real-time target voltage data and calculating the standard deviation of the target voltage includes: The real-time data of the target voltage collected is low-pass filtered using a moving average filtering algorithm, and the standard deviation of the target voltage is calculated based on the processed real-time data of the target voltage. The step of acquiring the current target voltage in real time with a preset sampling period and calculating the real-time rate of change of the target voltage includes: The current target voltage is low-pass filtered using a moving average filtering algorithm, and the real-time rate of change of the target voltage is calculated based on the processed current target voltage. Specifically, the setting of the dynamic gradient threshold includes: The dynamic gradient threshold is determined based on the standard deviation of the target voltage and a preset gradient coefficient.

Images