Multi-pipeline wafer cleaning system and control method thereof

Abstract

The invention relates to the technical field of wafer cleaning, in particular to a multi-pipeline wafer cleaning system and a control method thereof. The system comprises a cleaning chamber, a ring net which is detachably mounted on the periphery of the chamber and provided with uniform fine holes, a plurality of independent air draft pipelines connected to different areas of the ring net, and a control unit connected with all the pipelines. The method comprises the following steps: monitoring environmental parameters in a chamber through a sensor, and intelligently and independently adjusting air draft parameters of each pipeline based on the environmental parameters. By constructing the distributed air draft field, the inherent defects of non-uniform air flow distribution, low water vapor extraction efficiency, easiness in causing secondary splashing pollution, high maintenance cost and the like in the traditional single-point air draft technology are effectively overcome. Uniform and controllable distribution of air flow in the cavity can be achieved, the drying effect and consistency of wafer cleaning are remarkably improved, the defect generation risk is reduced, meanwhile, the device has the advantages of being self-adaptive in adjustment and convenient to maintain, and the device has great significance in improving the semiconductor wafer manufacturing yield.

Description

Technical Field

[0001] This invention relates to the field of wafer cleaning technology, and more specifically to a multi-channel wafer cleaning system and its control method. Background Technology

[0002] In the semiconductor wafer manufacturing industry, wafer cleaning is a crucial process step, as its cleanliness directly determines the yield of the final product. During wafer cleaning, deionized water or various chemical solutions are used to spray or immerse the wafer surface, generating a large amount of water vapor, droplets, and particulate contaminants that are stripped off. To maintain a clean environment within the chamber and ensure effective drying, these contaminants must be removed promptly and effectively.

[0003] Currently, most traditional wafer cleaning machines widely used in the industry employ a "single exhaust port" design. This design involves placing an exhaust port at a specific location within the chamber and connecting it to a single exhaust duct, relying on a centralized negative pressure point to extract contaminants from the entire chamber. However, this simple exhaust method has gradually revealed several inherent technical flaws in practical applications. First, because the exhaust power originates from a single point, it is difficult to cover the entire chamber space, resulting in severely uneven airflow distribution within the chamber. This easily creates airflow dead zones in areas far from the exhaust port, preventing the effective removal of moisture and particles, leading to water stains and affecting drying efficiency. Second, when the concentrated high-speed airflow impacts the chamber's inner wall or liquid surface, it may cause "secondary splashing" of already attached water droplets or particles, resulting in secondary contamination of the wafer surface. These two drawbacks combined directly increase the risk of defects such as oxidation and uneven etching on the wafer surface, posing a continuous threat to product yield. Furthermore, under high load operation, all pollutants in a single-point exhaust system pass through a single pipe, which can easily lead to the rapid accumulation of residues on the inner wall of the pipe. This not only increases flow resistance and reduces efficiency, but also causes the equipment to require more frequent shutdowns for cleaning and maintenance, thus affecting the overall efficiency of the production line.

[0004] While existing technologies have attempted to improve extraction efficiency by increasing ventilation power or optimizing duct routing, these improvements have not overcome the fundamental limitations of single-point ventilation and cannot simultaneously resolve the contradiction between uneven airflow distribution and secondary contamination. Therefore, there is an urgent need in the field for a novel ventilation system design that can achieve uniform and controllable airflow distribution within the chamber from a structural perspective. This would allow for efficient removal of contaminants while completely avoiding the negative impacts of airflow disturbance, thus meeting the increasingly stringent requirements for cleaning quality and stability in semiconductor manufacturing processes.

[0005] Therefore, existing technologies still need further development. Summary of the Invention

[0006] The purpose of this invention is to overcome the above-mentioned technical deficiencies and provide a multi-channel wafer cleaning system and its control method to solve the problems existing in the prior art.

[0007] To achieve the above-mentioned technical objectives, according to a first aspect of the present invention, the present invention provides a multi-pipeline wafer cleaning system, including a cleaning chamber for accommodating wafers for cleaning, and a fine-perforated ring mesh detachably installed around the periphery of the cleaning chamber, the fine-perforated ring mesh having uniformly distributed fine holes for allowing the exhaust airflow to pass through evenly. Multiple exhaust pipes, each connected to a different area of ​​the fine-pore ring network, are used to independently extract water vapor and particles generated during the cleaning process; The control unit, connected to each exhaust duct, is used to independently control the exhaust parameters of each exhaust duct in order to adjust the distribution of exhaust airflow.

[0008] Specifically, the perforated ring mesh is detachably installed by clips or bolts, and the diameter of the perforations ranges from 5 mm to 10 mm.

[0009] Specifically, the fine-pore ring mesh is made of a corrosion-resistant material, including stainless steel or polytetrafluoroethylene.

[0010] Specifically, each exhaust duct includes a flow control valve, which is electrically connected to the control unit and is used to adjust the exhaust flow rate.

[0011] Specifically, it also includes an environmental sensor, which is installed in the cleaning chamber to monitor at least one parameter, namely humidity and particle concentration, and sends the monitoring data to the control unit.

[0012] According to a second aspect of the present invention, a control method for a multi-channel wafer cleaning system is provided, comprising: S100: Real-time monitoring of environmental parameters within the cleaning chamber via environmental sensors; S200. Based on the environmental parameters, the exhaust parameters of each exhaust duct are independently adjusted by the control unit to ensure uniform airflow distribution and prevent water vapor backflow or secondary splashing.

[0013] Specifically, the environmental parameters include humidity and particulate concentration values, and S100 includes: An alarm signal is triggered when the humidity value exceeds the preset threshold or the particle concentration value is abnormal.

[0014] Specifically, S200 includes: The exhaust speed of each exhaust duct is dynamically adjusted based on changes in humidity levels, with the exhaust speed being positively correlated with the humidity level.

[0015] Specifically, S200 further includes: Adjust the opening sequence or ventilation duration of the exhaust ducts according to the particle concentration value to prioritize the extraction of high concentration areas.

[0016] Specifically, after S200, it also includes: S300: Based on the cumulative operating time of the exhaust duct or historical data of environmental parameters, the periodic disassembly and cleaning of the fine-pore ring network is automatically triggered.

[0017] Beneficial effects: The multi-pipeline wafer cleaning system and its control method provided by this invention have a series of significant advantages compared with traditional single-point ventilation technology. These advantages stem from the close integration of its innovative system structure design and intelligent control strategy.

[0018] The primary benefit of this invention lies in fundamentally improving the uniformity of airflow and moisture extraction within the chamber. By constructing a ring network with uniformly spaced fine holes around the perimeter of the chamber and connecting multiple independently controlled exhaust pipes, a distributed exhaust field is created. This design ensures that the negative pressure from the exhaust acts uniformly throughout the entire circumferential space of the chamber, effectively eliminating the airflow dead zones present in traditional single-point exhaust. This ensures that all areas of the wafer surface are under uniform downward airflow, thereby achieving efficient and consistent extraction of moisture and particles, and greatly reducing the risk of water stains remaining due to uneven local drying.

[0019] The second key benefit of this invention is its superior suppression of secondary contamination. Because the exhaust power is distributed across multiple pipes, the airflow velocity in each pipe can be precisely controlled, avoiding secondary splashing caused by excessively high airflow velocity in a single pipe impacting the chamber walls or liquid surface. The uniform and gentle airflow smoothly carries contaminants away from the wafer surface and out of the chamber, significantly reducing the likelihood of contaminants that have detached from the wafer falling back onto its surface, thus providing a solid guarantee for improving the final yield of wafer cleaning.

[0020] Third, the intelligent control method of this invention brings about adaptive optimization and improved energy efficiency. By integrating environmental sensors to monitor the chamber's operating conditions in real time and employing an advanced algorithm combining fuzzy rules and PID control, the system can dynamically and intelligently adjust the ventilation parameters of each pipeline independently. This means that the system can automatically adapt to different cleaning formulas, process stages, and sudden contamination events, avoiding unnecessary energy waste while ensuring cleaning effectiveness, thus achieving a balance between high-efficiency cleaning and low operating costs.

[0021] Fourth, the detachable fine-hole ring network design and the predictive maintenance mechanism based on operational data of this invention bring a dual improvement in maintenance convenience and equipment reliability. The modular design of the ring network makes cleaning and maintenance work simple and quick, significantly reducing equipment downtime. The system's automatically recorded operating parameters and triggered maintenance reminders shift the maintenance mode from passive response to proactive prevention, effectively avoiding unexpected downtime and unplanned wafer scrapping caused by component blockage or performance degradation, thereby improving the overall equipment efficiency and service life of the production line. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the structure of the multi-channel wafer cleaning system provided in a specific embodiment of the present invention; Figure 2 This is a flowchart illustrating the control method of the multi-channel wafer cleaning system provided in a specific embodiment of the present invention; The above figures contain the following reference numerals: 1. Fine-mesh ring network; 2. Exhaust duct. Detailed Implementation

[0023] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Based on the embodiments in this application, other similar embodiments obtained by those skilled in the art without creative effort should all fall within the scope of protection of this application. Furthermore, directional terms mentioned in the following embodiments, such as "up," "down," "left," and "right," are only for reference to the directions in the accompanying drawings; therefore, the directional terms used are for illustrative purposes and not for limiting the invention.

[0024] The present invention will be further described below with reference to the accompanying drawings and preferred embodiments.

[0025] Please see Figure 1 This invention provides a multi-channel wafer cleaning system, including a cleaning chamber for accommodating wafers for cleaning, and further comprising: A fine-pore ring mesh 1 is detachably installed around the periphery of the cleaning chamber. The fine-pore ring mesh 1 has uniformly distributed fine pores to allow the exhaust airflow to pass through evenly. Multiple exhaust pipes 2, each exhaust pipe 2 is connected to a different area of ​​the fine-pore ring network 1, for independently extracting water vapor and particles generated during the cleaning process; The control unit is connected to each exhaust duct 2 and is used to independently control the exhaust parameters of each exhaust duct 2 in order to adjust the distribution of exhaust airflow.

[0026] It should be further explained that the specific design of the cleaning chamber is as follows: The chamber is cylindrical to match the circular structure of a wafer (e.g., a 12-inch wafer). The chamber's inner diameter is preferably 400mm ± 10mm, and its height is 300mm ± 10mm. It is made of 316L stainless steel with a thickness of 5mm to ensure corrosion resistance and structural strength. A wafer inlet is located at the top of the chamber, and a drain outlet with a diameter of 50mm is located at the bottom. A perforated ring mesh 1 is installed around the perimeter of the chamber's inner wall. The ring mesh is 80mm wide and annular in shape, with an inner diameter matching the chamber's inner diameter. The ring mesh is fixed by supports spaced 150mm apart. The perforations in the perforated ring mesh 1 are evenly distributed, with a density of 30 holes per square centimeter. The perforation diameter is preferably 8mm ± 1mm. The number of multiple exhaust pipes 2 is preferably six, evenly distributed around the ring mesh, with adjacent pipes forming an angle of 60 degrees. Each exhaust duct 2 has a diameter of 25mm and is made of PVC (polyvinyl chloride). The duct length is adjusted according to the installation location, typically ranging from 500mm to 1000mm. The control unit uses a PLC (Programmable Logic Controller), such as a Siemens S7-1200, with digital and analog input / output modules. The control unit is connected to the actuator (such as a flow control valve) of each exhaust duct 2 via a 4-20mA current signal to independently control the exhaust parameters, including exhaust speed, flow rate, and start / stop time.

[0027] Specifically, the ventilation parameter adjustment range is designed as follows: The ventilation speed is 0-10m / s, and the flow rate is 0-20m³ / h.

[0028] Understandably, independent control of multiple pipelines enables precise adjustment of airflow distribution. Experiments show that water vapor extraction efficiency is improved by more than 35% (compared to traditional single-pipeline systems), avoiding localized water vapor residue. The modular design of the system facilitates maintenance and reduces equipment downtime. Uniform airflow distribution reduces the risk of secondary splashing, improving wafer cleaning yield by approximately 5%. Those skilled in the art can directly process and assemble the chambers, ring networks, and pipelines based on the above dimensions and parameters without requiring creative effort.

[0029] Specifically, the perforated ring mesh 1 is detachably installed by clips or bolts, and the diameter of the perforations ranges from 5 mm to 10 mm.

[0030] It should be further explained that the buckle type is a quick-release stainless steel buckle, with 8 buckles evenly distributed around the perimeter of the ring mesh, with a buckle spacing of 157mm (based on a 400mm perimeter). The buckle dimensions are 20mm long and 10mm wide, and the buckles are fixed to the ring mesh by welding. An alternative to bolt installation is to use 12 M5 stainless steel bolts with a bolt spacing of 104mm and a bolt hole diameter of 5.5mm, with an allowable tolerance of ±0.1mm. During installation, the ring mesh is first aligned with the positioning groove on the inner wall of the cavity, and then the buckles or bolts are tightened. The orifice diameter range of 5mm to 10mm is the optimal value after experimental optimization. When the diameter is less than 5mm, the orifice is easily blocked by submicron particles, shortening the cleaning cycle to less than 50 hours; when the diameter is greater than 10mm, the airflow uniformity coefficient (defined as the variance of airflow per unit area) exceeds 0.1, resulting in uneven air extraction. More preferably, the diameter of the fine holes is 8 mm, a value determined based on computational fluid dynamics (CFD) simulation, which balances airflow uniformity and anti-clogging properties. The fine holes are machined using laser drilling, with a hole depth equal to the thickness of the ring network (1 mm) and a hole spacing of 16 mm to ensure continuous airflow.

[0031] Understandably, the detachable design extends the ring network cleaning cycle to over 300 hours (reducing maintenance time by 60% compared to fixed installation); optimized orifice size reduces the airflow uniformity coefficient to below 0.05, improving cleaning consistency. Those skilled in the art can select the installation method based on the chamber size, such as clips for small-batch production and bolts for large-batch production, and utilize standard laser equipment for drilling.

[0032] Specifically, the fine-pore ring mesh 1 is made of a corrosion-resistant material, including stainless steel or polytetrafluoroethylene.

[0033] It should be further noted that the preferred stainless steel is 316L stainless steel, with the following composition: carbon ≤0.03%, chromium 16-18%, nickel 10-14%, molybdenum 2-3%, tensile strength ≥520MPa, suitable for cleaning solutions with pH values ​​of 2-12 (such as SC-1 solution: ammonia + hydrogen peroxide). The ring mesh thickness is 1.0mm±0.1mm, and the surface roughness Ra≤0.8μm to reduce particle adhesion. An alternative to polytetrafluoroethylene (PTFE) is available: PTFE density is 2.2g / cm³, tensile strength ≥20MPa, applicable temperature -200℃ to 260℃, and ring mesh thickness is 1.5mm±0.2mm. Material selection is based on the cleaning environment: for high-temperature (>80℃) or strong acid cleaning (such as DHF: hydrofluoric acid), PTFE is preferred; for neutral cleaning at room temperature, stainless steel is selected. The ring mesh manufacturing process includes: stainless steel ring mesh is formed by stamping and then laser-drilled, while PTFE ring mesh is formed by injection molding and then drilled.

[0034] Understandably, corrosion-resistant materials extend the lifespan of the ring network to over 3 years (reducing the corrosion rate by 90% compared to ordinary steel); the material's inertness prevents the introduction of metal contamination, and the wafer surface defect density is reduced to below 0.1 defects / cm². Those skilled in the art can select materials based on the cleaning solution formulation and determine processing parameters by referring to the material data sheet.

[0035] Specifically, each exhaust duct 2 includes a flow control valve, which is electrically connected to the control unit and is used to adjust the exhaust flow rate.

[0036] It should be further noted that the flow control valve is preferably an electric proportional valve, such as the SMCPVQ31-5G, with a valve diameter of 25mm (matching the pipe diameter), a flow regulation range of 0.5-20m³ / h, and an accuracy of ±0.1m³ / h. The preferred connection method between the valve and the control unit is to control the valve opening (0-100%) via a 4-20mA analog signal, with a response time <0.5 seconds. The control algorithm uses PID (Proportional-Integral-Derivative) control, with the following preferred parameters: proportional coefficient Kp=0.8, integral time Ti=15 seconds, derivative time Td=2 seconds. Kp=0.8 provides fast response without overshoot (based on the Ziegler-Nichols tuning method), Ti=15 seconds eliminates steady-state error (for the slow dynamics of the cleaning process), and Td=2 seconds suppresses oscillation. The PID algorithm steps include: 1. Read the flow rate setpoint (from the control unit) and the actual flow rate (from the flow sensor); 2. Calculate the error e(t) = set value - actual value; 3. Calculate the PID output: u(t) = Kp e(t)+Ki ∫e(t)dt+Kd de(t) / dt, where Ki=Kp / Ti, Kd=Kp Td; 4. Convert u(t) into a 4-20mA signal to drive the valve.

[0037] Understandably, precise flow control allows the ventilation volume of each pipeline to adapt to changes, reducing energy consumption by 20% (compared to a fixed flow rate); the PID algorithm ensures stability, with flow fluctuations of <5%. Those skilled in the art can implement PID control using a programmable PLC without the need for complex models.

[0038] Specifically, it also includes an environmental sensor, which is installed in the cleaning chamber to monitor at least one parameter, namely humidity and particle concentration, and sends the monitoring data to the control unit.

[0039] It should be further noted that the humidity sensor is preferably a capacitive sensor, such as the Honeywell HIH8000, with a range of 0-100%RH and an accuracy of ±1.5%RH. It should be installed at the center of the top of the chamber to avoid direct water splashing. The particle concentration sensor uses a laser particle counter, such as the Lighthouse 3016, with a range of 0-1000 particles / m³ (for particles ≥0.3μm in diameter) and an accuracy of ±5%. It should be installed on the side wall of the chamber, 100mm from the bottom. Sensor data is transmitted to the control unit via an RS485 interface, with a sampling frequency of 2Hz (i.e., sampling once every 0.5 seconds). Regarding data preprocessing, the control unit performs a moving average filter on the raw data (window size = 5 samples) to reduce noise. The humidity alarm threshold is 85%RH, and the particle concentration alarm threshold is 100 particles / m³. 85%RH is the critical point for water vapor condensation (based on dew point calculations); exceeding this value easily leads to water stains. 100 particles / m³ is the cleanliness requirement for semiconductor cleaning processes (according to ITRS standards).

[0040] Understandably, real-time monitoring enables predictive control, triggering adjustments 10-30 seconds in advance, improving yield by 3%; sensor data recording is used for trend analysis to optimize maintenance plans. Those skilled in the art can install the sensors and set up the communication protocol without additional calibration.

[0041] Please see Figure 2 The present invention provides another embodiment, which provides a control method for a multi-pipe wafer cleaning system, the control method of the multi-pipe wafer cleaning system comprising: S100: Real-time monitoring of environmental parameters within the cleaning chamber via environmental sensors.

[0042] Further explanation is needed regarding the specific implementation of the S100: environmental parameters include humidity H (%RH), particle concentration P (particles / m³), and selectable temperature T (°C). The monitoring frequency is 2Hz, and data is stored in the control unit's SD card in CSV file format.

[0043] S200. Based on the environmental parameters, the exhaust parameters of each exhaust pipe 2 are independently adjusted by the control unit to ensure uniform airflow distribution and prevent water vapor backflow or secondary splashing.

[0044] Further explanation is needed regarding the specific implementation of the monitoring steps: environmental parameters include humidity H (%RH), particle concentration P (particles / m³), and selectable temperature T (°C). The monitoring frequency is 2Hz, and data is stored in the control unit's SD card in CSV file format. The detailed algorithm for the control steps employs a combination of fuzzy logic control and PID control, with the following steps: 1. Data Acquisition and Fuzzification: Sensor values ​​are read and filtered. Precise humidity values ​​H and particulate concentration values ​​P are converted into fuzzy linguistic variables. The fuzzy subset of humidity H is {low, medium, high}, and the fuzzy subset of particulate concentration P is {low, medium, high}. Their membership functions are defined using trigonometric functions. Humidity H (%RH): Low (L): Range 0-60, peak value 30; Medium (M): Range 40-85, peak value 62.5; High (H): Range 70-100, peak value 85; Particle concentration P (particles / m³): Low (L): Range 0-60, peak value 30; Medium (M): Range 40-100, peak value 70; High (H): Range 80-1000, peak value at 100; 2. Fuzzy reasoning (applying fuzzy rules): Reasoning is performed based on the following 9 fuzzy rule bases. The rule form is "IF (condition) THEN (conclusion)", using AND (minimum operation) to connect the premises. The fuzzy subset of the output variable "ventilation speed increment ΔV" is {negative small (NS), zero (ZO), positive small (PS), positive medium (PM), positive large (PB)}; Rule 1: IF (H is low) AND (P is low) THEN (ΔV is NS); Rule Explanation: When both humidity and particle concentration are low, it indicates that the cleaning process generates fewer contaminants, and the exhaust speed can be appropriately reduced to save energy.

[0045] Rule 2: IF (H is low) AND (P is medium) THEN (ΔV is ZO); Rule Explanation: Low humidity but moderate particulate concentration, maintain the current speed, focus on addressing particulate pollution.

[0046] Rule 3: IF (H is low) AND (P is high) THEN (ΔV is PS); Rule Explanation: Despite low humidity, high particulate concentration poses a risk of contamination, necessitating a slight increase in ventilation speed.

[0047] Rule 4: IF (H is medium) AND (P is low) THEN (ΔV is ZO); Rule Explanation: With moderate humidity and low particle concentration, maintaining the current speed is sufficient to maintain equilibrium.

[0048] Rule 5: IF (H is in) AND (P is in) THEN (ΔV is PS); Rule Explanation: With both humidity and particle concentration at moderate levels, slightly increase the exhaust speed to optimize cleaning performance.

[0049] Rule 6: IF (H is medium) AND (P is high) THEN (ΔV is PM); Rule Explanation: When humidity is moderate but particle concentration is high, the exhaust speed needs to be moderately increased to quickly remove the particles.

[0050] Rule 7: IF (H is high) AND (P is low) THEN (ΔV is PS); Rule Explanation: The humidity is high but the particle concentration is low. The main problem is water vapor. Slightly increase the speed to prevent water vapor residue.

[0051] Rule 8: IF (H is high) AND (P is medium) THEN (ΔV is PM); Rule Explanation: High humidity accompanied by moderate particulate concentration requires a moderate increase in speed to address both pollutants simultaneously.

[0052] Rule 9: IF (H is high) AND (P is high) THEN (ΔV is PB); Explanation of the rules: When both humidity and particle concentration are high, the conditions are the worst. The exhaust speed is increased significantly to ensure the cleaning effect and prevent secondary pollution to the greatest extent. 3. Defuzzification: The centroid method is used to convert the output fuzzy set "ventilation speed increment ΔV" obtained from fuzzy inference into a precise value ΔV_crisp. The formula is: ΔV_crisp = ∑(μ_i v_i) / ∑μ_i, where μ_i is the membership degree of the i-th rule conclusion, and v_i is the center value of the fuzzy set corresponding to the rule conclusion (for example, the center value of NS may be -0.5 m / s, ZO is 0, PS is 0.5 m / s, PM is 1.0 m / s, and PB is 1.5 m / s). in: μ_i: Represents the degree to which the i-th rule is activated, reflecting the degree of matching between the current state and the rule.

[0053] v_i: represents the typical value of the ventilation speed adjustment recommended by the i-th rule.

[0054] ΔV_crisp: Represents the final precise speed adjustment calculated after considering all activation rules; 4. PID Control: The reference speed V_base (e.g., calculated by the linear formula of claim 8) is added to the fuzzy increment ΔV_crisp to obtain the final exhaust speed setpoint V_set = V_base + ΔV_crisp. This V_set is used as the input to the PID controller to precisely adjust the valve opening of each pipeline.

[0055] Understandably, these nine complete fuzzy rules constitute an expert experience system, making the control strategy more aligned with human intuition and process requirements. Compared to simply using the linear formula of claim 8, fuzzy-PID hybrid control can better handle the complex situation of nonlinear coupling between humidity and particle concentration, resulting in a smoother and more intelligent dynamic response, further improving ventilation efficiency (approximately 5-10%) and system adaptability. Those skilled in the art can fine-tune the rules and membership function parameters according to the actual process.

[0056] Specifically, the environmental parameters include humidity and particle concentration. S100 includes triggering an alarm signal when the humidity exceeds a preset threshold or the particle concentration is abnormal.

[0057] Further explanation is needed regarding the preset thresholds: the humidity threshold is set to 85%RH, and abnormal particle concentration is defined as three consecutive sampling values ​​exceeding 100 particles / m³ or an instantaneous value exceeding 150 particles / m³. The preferred alarm signal types are audible and visual alarms (buzzer frequency 1kHz, flashing red LED) and software alarms (sending SNMP trap messages to the monitoring system via Ethernet). Alarm delay mechanism: to avoid false alarms, a delay time of 3 seconds is set (i.e., the condition must last for 3 seconds before triggering). The delay time is based on process stability analysis (the cleaning cycle is typically 60 seconds, with 3 seconds accounting for 5%, and false alarms are negligible). Alarm logs: the control unit records the alarm time, parameter values, and equipment status, storing them in a log file.

[0058] Understandably, the alarm function enables timely operator intervention, reducing wafer scrap rate by 10%; historical logs aid in fault diagnosis, reducing mean time to repair (MTTR) by 15%. Those skilled in the art can configure the alarm logic and use the timer function.

[0059] Specifically, S200 includes: dynamically adjusting the exhaust speed of each exhaust duct 2 according to the change in humidity value, wherein the exhaust speed is positively correlated with the humidity value.

[0060] It should be further explained that the dynamic adjustment uses a linear mapping formula: V=K×H+C, where V is the exhaust velocity (unit: m / s), H is the humidity value (unit: %RH), K is the proportionality coefficient, and C is the baseline velocity. The preferred value of K is 0.08 m / s / %RH, and the preferred value of C is 1.5 m / s. Through experimental regression analysis, K=0.08 ensures that for every 10% increase in humidity, the velocity increases by 0.8 m / s, covering a typical cleaning range (50-90%RH); C=1.5 provides basic exhaust capacity, avoiding zero velocity. In the formula, V represents the exhaust velocity, which directly affects airflow energy; H represents the ambient humidity, reflecting the water vapor load; K is the sensitivity coefficient, indicating the degree of influence of humidity on velocity; and C is the offset, ensuring minimum exhaust. The control unit updates the V value every 0.5 seconds. Example: When H=70%RH, V=0.08×70+1.5=7.1 m / s. If the humidity change rate dH / dt > 5%RH / s, then nonlinear compensation is enabled, and the V additional term is 0.1×dH / dt.

[0061] It is understandable that positive correlation adjustment optimizes energy consumption, saving 25% more energy than a fixed speed; dynamic response prevents water vapor accumulation, shortening drying time by 20%. Those skilled in the art can directly implement the formulas in PLC code.

[0062] Specifically, S200 further includes: adjusting the opening sequence or exhaust duration of the exhaust pipe 2 according to the particle concentration value, so as to prioritize the extraction of high concentration areas.

[0063] In a preferred embodiment of the present invention, the algorithm for adjusting the opening order includes: 1. Divide the chamber into 6 regions (corresponding to 6 pipelines), each region has a particle concentration value P_i (i=1-6).

[0064] 2. Calculate the ranking of P_i every 0.5 seconds, and select the region with the highest P_i to start first.

[0065] 3. Opening order rule: If max(P_i)>50 particles / m³, then the pipeline in that area is opened immediately; other pipelines are opened in descending order of P_i, with a delay time Δt=2 seconds×(rank-1).

[0066] Furthermore, the proposed adjustment plan for ventilation duration includes: The base duration is 10 seconds. For areas where P_i > 50 particles / m³, the ventilation duration is extended to 15 seconds; for areas where P_i > 100 particles / m³, it is extended to 20 seconds. The threshold of 50 particles / m³ is the critical point for particle deposition (based on SEMI standards), exceeding which increases the risk of contamination. The algorithm implementation preferably uses array sorting and timer control.

[0067] Understandably, prioritizing extraction from high-concentration areas increases particle removal efficiency by 40% and reduces cross-contamination; resource optimization reduces total ventilation energy consumption by 10%. Those skilled in the art can program using simple sorting algorithms (such as bubble sort).

[0068] Specifically, it also includes S300: automatically triggering the periodic disassembly and cleaning of the fine-pore ring network based on historical data of the cumulative operating time of the exhaust duct or environmental parameters.

[0069] The fine-pore ring network 1 is periodically disassembled and cleaned. The periodic cleaning is automatically triggered based on the cumulative operating time of the exhaust duct 2 or historical data of environmental parameters.

[0070] It should be further noted that the triggering conditions for regular cleaning include: The cumulative running time threshold is set to 250 hours (reason: experiments show that the ring network voltage drop increases by 15% after 250 hours, affecting performance).

[0071] Historical environmental parameter data: If the average humidity is >80%RH or the particle concentration variance is >10 for 24 consecutive hours, maintenance will be triggered.

[0072] Further maintenance procedures include: automatically prompting "Maintenance required" on the control unit display and sending an email. The cleaning method involves disassembling the ring network, cleaning it in deionized water for 10 minutes using an ultrasonic cleaner (40kHz frequency, 300W power), and then drying it (60°C, 30 minutes). Maintenance records preferably record the date and time of each maintenance session and generate a report.

[0073] Understandably, preventative maintenance extends the ring network lifespan to 4,000 hours and reduces the failure rate by 50%; automated triggering reduces human error and lowers maintenance costs by 20%. Those skilled in the art can set timers and data trend analysis algorithms, such as moving averages or exponential smoothing.

[0074] In a preferred embodiment, this application also provides an electronic device, the electronic device comprising: The computer device includes a memory and a processor, wherein the memory stores computer-readable instructions that, when executed by the processor, implement the control method of the multi-pipeline wafer cleaning system. The computer device can be broadly categorized as a server, terminal, or any other electronic device with the necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, memory, network interface, communication interface, etc., connected via a system bus. The processor of the computer device can be used to provide the necessary computing, processing, and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and internal memory. The non-volatile storage medium may store an operating system, computer programs, etc. The internal memory can provide an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface and communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, it performs the steps of the method of the present invention.

[0075] This invention can be implemented as a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the steps of the methods of embodiments of the invention to be performed. In one embodiment, the computer program is distributed across multiple network-coupled computer devices or processors, such that the computer program is stored, accessed, and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be executed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be executed by one or more computer devices or processors, and one or more other method steps / operations may be executed by one or more other computer devices or processors. One or more computer devices or processors may execute a single method step / operation, or execute two or more method steps / operations.

[0076] Those skilled in the art will understand that the method steps of this invention can be performed by a computer program instructing related hardware, such as a computer device or processor, to perform the steps of this invention when executed. Depending on the context, any references herein to memory, storage, databases, or other media may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid-state drive, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0077] The technical features described above can be combined arbitrarily. Although not all possible combinations of these technical features are described, any combination of these technical features should be considered to be covered by this specification, provided that such combination does not contain contradictions.

[0078] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A multi-channel wafer cleaning system, comprising a cleaning chamber for accommodating wafers for cleaning, characterized in that, Also includes: A fine-pore ring mesh is detachably installed around the periphery of the cleaning chamber. The fine-pore ring mesh has uniformly distributed fine pores to allow the exhaust airflow to pass through evenly. Multiple exhaust pipes, each connected to a different area of ​​the fine-pore ring network, are used to independently extract water vapor and particles generated during the cleaning process; The control unit, connected to each exhaust duct, is used to independently control the exhaust parameters of each exhaust duct in order to adjust the distribution of exhaust airflow.

2. The multi-channel wafer cleaning system according to claim 1, characterized in that, The perforated ring mesh is detachably installed by clips or bolts, and the diameter of the perforations ranges from 5 mm to 10 mm.

3. The multi-channel wafer cleaning system according to claim 2, characterized in that, The fine-pore ring network is made of a corrosion-resistant material, including stainless steel or polytetrafluoroethylene.

4. The multi-channel wafer cleaning system according to claim 3, characterized in that, Each exhaust duct includes a flow control valve, which is electrically connected to the control unit and is used to adjust the exhaust flow rate.

5. The multi-channel wafer cleaning system according to claim 4, characterized in that, It also includes an environmental sensor, which is installed in the cleaning chamber to monitor at least one parameter, namely humidity and particle concentration, and to send the monitoring data to the control unit.

6. A control method for a multi-channel wafer cleaning system, applied to the system as described in any one of claims 1-5, characterized in that, Includes the following steps: S100: Real-time monitoring of environmental parameters within the cleaning chamber via environmental sensors; S200. Based on the environmental parameters, the exhaust parameters of each exhaust duct are independently adjusted by the control unit to ensure uniform airflow distribution.

7. The control method according to claim 6, characterized in that, The environmental parameters include humidity and particulate concentration values, and S100 includes: An alarm signal is triggered when the humidity value exceeds the preset threshold or the particle concentration value is abnormal.

8. The control method according to claim 7, characterized in that, S200 includes: The exhaust speed of each exhaust duct is dynamically adjusted based on changes in humidity levels, with the exhaust speed being positively correlated with the humidity level.

9. The control method according to claim 8, characterized in that, The S200 further includes: Adjust the opening sequence or ventilation duration of the exhaust ducts according to the particle concentration value to prioritize the extraction of high concentration areas.

10. The control method according to claim 6, characterized in that, Following S200, the following is also included: S300: Based on the cumulative operating time of the exhaust duct or historical data of environmental parameters, the periodic disassembly and cleaning of the fine-pore ring network is automatically triggered.

Images