Position closed loop control and levitation uniform pressure deposition system of wire bar evaporation source

Abstract

The application relates to the technical field of vacuum thin film preparation, and particularly discloses a position closed-loop control and suspension uniform pressure deposition system of a wire rod evaporation source; the position closed-loop control device is composed of a high-frequency induction heating coil, a laser displacement sensor, a precision lifting driving mechanism and a controller, the relative position between the top end of the wire rod and the coil is kept constant by driving the wire rod to axially move through an incremental PID algorithm, the impedance matching network is automatically tuned by using a gradient descent method, the deposition side is provided with an edge chamfered negative bias electrode, a uniform pressure plate which is suspended below a substrate and is connected with an adjustable bias power supply, and a double-layer insulation protection ring which is composed of a PTFE inner layer ring and an alumina ceramic outer layer ring. The application solves the problems of thermal decomposition carbonization, rate fluctuation and large-area electric field edge distortion during evaporation of easily thermally decomposed organic materials, and realizes high raw material utilization, high stability and large-area uniform deposition.

Description

Technical Field

[0001] This invention relates to the field of vacuum thin film preparation technology, and in particular to a closed-loop control system for the position of a wire rod evaporation source and a suspension pressure equalization deposition system. Background Technology

[0002] Organic semiconductor thin films are widely used in photovoltaic devices, OLED displays, thin-film transistors, and other fields. Vacuum thermal evaporation is the mainstream method for preparing high-quality organic thin films. Among them, C60 (fullerene), as a typical electron transport material, has the characteristics of high sublimation temperature (500-800℃) and poor thermal stability during evaporation.

[0003] C60 manufacturers have adopted bar evaporation for large-area coating, but the core closed-loop position control technology is monopolized by foreign equipment manufacturers. According to publicly available literature (such as ULVAC's "Technical Report on Vacuum Evaporation Equipment" 2019 edition, Section 3.2.4), existing commercial equipment generally adopts a process scheme of deeply inserting C60 bars into the coil heating zone to prevent the heating point from drifting as the bar is consumed. However, this scheme has inherent defects such as thermal decomposition and carbonization of raw materials and large fluctuations in evaporation rate.

[0004] Chinese patent CN106835029A (Shenzhen Huaxing Optoelectronics Technology Co., Ltd.) discloses a high-frequency induction evaporation source device that uses OLED material as the evaporation source. Radio frequency induced heating components are installed inside the crucible to disperse the heating area and avoid localized overheating. This solution does not involve real-time position detection and feed control of the evaporating material, and its dispersed arrangement of heating components aims to prevent localized overheating of the powder material. This is fundamentally different from the continuous feed evaporation of the present invention in terms of material morphology, feeding method, and control strategy.

[0005] The US patent US2018 / 0347030A1 (family patent) further describes the arrangement of the radio frequency heating components, but its evaporation source still uses batch loading of powder / granular materials into the crucible as the application object. It does not address the problem of evaporation rate fluctuation caused by the drift of heating point during the consumption of rod-shaped materials, nor does it propose technical means for real-time position detection and closed-loop compensation.

[0006] Chinese patent application CN121575356A (Wuxi Boda New Energy Technology Co., Ltd., application date December 2025) discloses a method for stabilizing the evaporation rate of fullerenes. This method suppresses localized overheating and abrupt rate changes by mixing high-melting-point medium powders (such as tungsten or molybdenum) into the fullerene powder. While this method addresses the problem at the material formulation level, it alters the purity of the evaporation film and cannot provide real-time closed-loop control of the evaporation rate. Furthermore, this method is applicable to powdered materials and cannot be directly applied to the continuous feed evaporation process of this invention.

[0007] In existing high-frequency induction heating C60 bar processes, to prevent the heating point from drifting out of the coil heating zone as the bar is consumed, the C60 bar must be deeply inserted into the coil preheating zone (the insertion depth is typically 1.5 to 2 times the coil height). This process is documented in ULVAC's publicly available equipment technical data (see ULVAC Technical Journal No. 75, 2016, "Development of High-Rate Evaporation Source for C60", which clearly states that the heating zone design of the C60 bar must ensure an insertion depth of at least 1.5 times the coil height to guarantee continuous evaporation).

[0008] In the prior art, those skilled in the art have long faced a technical contradiction between "preventing drift" and "avoiding carbonization": increasing the insertion depth can prevent drift, but it inevitably leads to an increase in temperature in the non-evaporation zone and intensified carbonization; decreasing the insertion depth can reduce the risk of carbonization, but the heating point is prone to drifting out of the effective heating zone of the coil, resulting in drastic fluctuations in the evaporation rate. This technical bias has led those skilled in the art to habitually adopt the "deep insertion preheating" approach, without considering breaking this contradiction through active control methods. Summary of the Invention

[0009] In view of this, embodiments of the present invention provide a closed-loop control system for the position of a bar evaporation source and a suspension pressure equalization deposition system to solve the problems of material carbonization, unstable evaporation rate, and low raw material utilization caused by traditional deep insertion preheating methods in the evaporation deposition process of organic materials such as C60, as well as uneven film thickness caused by edge distortion of the large-area bias electric field.

[0010] In a first aspect, embodiments of the present invention provide a closed-loop control system for the position of a bar evaporator, comprising: The vacuum chamber is maintained under vacuum during operation. A high-frequency induction heating coil is used to locally heat the top of the evaporation source; A laser displacement sensor is installed outside the vacuum chamber and detects the position of the top of the evaporation source in real time through an observation window, and outputs a position signal; A lifting mechanism is provided, wherein the evaporation source is fixedly mounted on the moving end of the lifting mechanism, and the lifting mechanism is used to drive the evaporation source to move along the axial direction of the vacuum chamber; The controller is electrically connected to the laser displacement sensor and the precision lifting drive mechanism respectively. The controller uses a closed-loop control algorithm to control the lifting mechanism based on the deviation between the position signal and the preset target position, so that the relative position between the top of the evaporation source and the high-frequency induction heating coil is kept constant.

[0011] Preferably, the high-frequency induction heating coil is made of hollow copper tube and has cooling water circulating inside; The high-frequency induction heating coil is arranged as a tapered variable diameter coil, and the bottom inner diameter of the high-frequency induction heating coil is larger than the top inner diameter of the high-frequency induction heating coil, and the top inner diameter of the high-frequency induction heating coil is larger than the diameter of the evaporation source. The operating frequency of the high-frequency induction heating coil is adaptively adjusted by the load through an impedance matching network. The number of turns of the spiral coil is set to 3-8 turns.

[0012] Preferably, the impedance matching network is an L-type matching network, including an adjustable capacitor and a fixed inductor. The adjustable capacitor has an adjustment range of 10-100pF, and the fixed inductor has an inductance of 2μH. The impedance matching network adopts an automatic tuning algorithm based on the principle of minimizing reflected power, and optimizes the matching network parameters through gradient descent method, with a tuning time of less than 100ms. Preferably, the gradient descent method for optimizing the matching network parameters includes: Step 1: Set the initial adjustable capacitor value and measure the current reflected power; Step 2: Adjust the value of the adjustable capacitor in steps of 2pF, measure the adjusted reflected power, and calculate the rate of change of reflected power with respect to the capacitor value, dPr / dC. Step 3: Iteratively adjust the value of the adjustable capacitor along the direction of decreasing reflected power, and remeasure the reflected power after each iteration; Step 4: Repeat steps 2 and 3 until the reflected power is less than 0.1W, thus completing the automatic tuning of the impedance matching network.

[0013] Preferably, the evaporation source is a wire rod made of organic material; The organic material wire rod is made of at least one of C60 fullerene, Alq3 (8-hydroxyquinoline aluminum), NPB (N,N'-diphenyl-N,N'-(1-naphthyl)-1,1'-biphenyl-4,4'-diamine), and metal phthalocyanine.

[0014] Preferably, the closed-loop control algorithm includes: Step 1: Set the target position. Set the top target position of the evaporation source below the center line of the high-frequency induction heating coil. Step 2: Detect the current position of the top of the evaporation source in real time using a laser displacement sensor at a preset sampling frequency; Step 3: Subtract the target position from the current position to obtain the position deviation value. ; Step 4: Determine the deviation range and select the control mode: If the position deviation value If the absolute value is greater than 1.5mm, the system enters a rapid approach mode, and the evaporation source is driven to move toward the target position at the maximum speed V_max via the lifting mechanism. If the position deviation value If the absolute value is less than or equal to 1.5mm, then the incremental PID control mode is entered, and steps 5 to 7 are executed. Step 5: Calculate the incremental PID control output and calculate the control increment Δu according to the following formula: in, This is a proportionality coefficient, with a value ranging from 0.6 to 1.0; is the integral coefficient, with a value range of [0.03, 0.05, 0.07]s⁻¹; is the differential coefficient, with a value range of 0.08-0.16s; T is the control period, with a value ranging from 10 to 100 ms; This represents the current cycle position deviation. This represents the positional deviation from the previous cycle. This represents the positional deviation between the previous two cycles; Step 6: If the position deviation value The absolute value is less than the integral separation threshold. Then the control increment Integral terms Set to zero; Step 7: Transfer the speed command from the previous cycle With the control increment Add them together to get the current cycle speed command. and the speed command Amplitude limiting is applied; Step 8: Send the speed command The output is sent to the lifting mechanism, which drives the evaporation source to move axially; Step 9: Adjust the current cycle deviation Store as ,Will Store as It records the current cycle speed command and then returns to step 2 to form a closed-loop control.

[0015] Secondly, a suspension pressure equalization deposition system is provided, comprising: The aforementioned closed-loop control system for the position of the bar evaporator source; A substrate support stage is set inside a vacuum chamber and is used to place the substrate to be processed. A negative bias electrode is disposed in the vacuum cavity and located between the high-frequency induction heating coil and the processing substrate, and is used to apply a DC negative high voltage to generate a deposition electric field; A floating equalizing plate is disposed below the substrate support stage, parallel to and insulated from the substrate support stage, and is connected to an adjustable DC bias power supply to keep it in a floating potential state. A double-layer insulating protective ring is arranged around the periphery of the suspended equalizing plate, including an inner ring and an outer ring, wherein the dielectric constant of the inner ring is smaller than that of the outer ring.

[0016] Preferably, the bias electrode includes an electrode body disposed between the evaporation source and the processing substrate, the electrode body being opposite to and parallel to the processing substrate; The electrode body is flat or slightly arc-shaped, and the material of the electrode body is 316L stainless steel with a nickel plating layer on the surface.

[0017] Preferably, the suspended pressure equalizing plate comprises: The pressure equalization plate body is disposed below the substrate support stage, and the pressure equalization plate body is parallel to and insulated from the substrate support stage; The equalizing plate body is made of aluminum alloy 6061 and the surface is treated with hard anodizing.

[0018] Preferably, the double-layer insulating protective ring comprises: The inner ring and the outer ring are concentric ring structures and are arranged around the periphery of the suspended pressure equalizing plate; The inner ring is made of polytetrafluoroethylene; the outer ring is made of alumina ceramic; the dielectric constant ratio of the inner ring to the outer ring is 1:3 to 1:5.

[0019] The closed-loop position control and suspension pressure equalization deposition system for a bar evaporation source provided by this invention has the following beneficial effects: In this embodiment, through closed-loop position control, the relative positional deviation between the top of the evaporation source and the high-frequency induction heating coil can be controlled within ±0.2mm. This eliminates the need for deep insertion preheating, preventing heating point drift and fundamentally resolving the technical contradiction between "preventing drift" and "avoiding carbonization." This significantly reduces the risk of thermal decomposition and carbonization of organic materials such as C60, and evaporation rate fluctuations can be controlled within ±1%. Raw material utilization is increased from 30%~40% in traditional processes to over 70%. Furthermore, by adding a suspended equalizing plate with a double-layer insulating protective ring below the substrate support stage, the electric field distortion problem in the edge region during large-area deposition can be effectively improved, increasing the uniformity of the electric field across the substrate by over 15%. This significantly improves the uniformity of the organic thin film thickness, meeting the requirements for the fabrication of large-area photovoltaic devices and OLED display panels. This invention breaks through long-standing technical biases in the field. By using active position closed-loop control, it overcomes the inherent contradictions of traditional processes. At the same time, it optimizes the deposition electric field distribution through a suspended pressure equalization structure. It has the advantages of high raw material utilization, stable evaporation rate, and good film thickness uniformity, making it suitable for large-scale industrial production applications. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, and these are all within the protection scope of the present invention.

[0021] Figure 1 This is a schematic diagram of a closed-loop control system for the position of a bar evaporator. Figure 2 This is a schematic diagram of the overall architecture of a linear bar evaporation source location closed-loop control and suspension pressure equalization deposition system; Figure 3 This is a schematic block diagram of a position closed-loop control system for a bar evaporator. Figure 4 It is a block diagram of closed-loop control and process flow; Figure 5 This is a comparison curve of evaporation rate stability over 120 minutes; Figure 6 It is a 9-point film thickness distribution curve; Figure 7 This is a closed-loop control logic block diagram; Figure 8 The diagram shows a comparison of the temperature distribution in the heating zone of the C60 rod (a) the existing technology of "deep insertion preheating"; b) the present invention of "top local heating"). Figure 9 These are comparison diagrams of simulated electric field distribution (a) traditional parallel plate; b) the equalizing plate + protective ring of this invention; c) surface electric field distribution curve. Figure 10 These are experimental data curves (a) evaporation rate stability curve; b) 9-point film thickness distribution curve; c) large-area contour plot. Figure 11 It is a complete time-series process flow diagram; Figure 12 This is a cross-sectional view of the double-layer protective ring; Figure 13 This is a schematic diagram of a multi-source array layout; Figure 14 This is a flowchart of the incremental PID control algorithm; Figure 15 A comparison of the electric field intensity distribution curves on the substrate surface with and without a pressure equalizing plate; Figure 16 This is a diagram showing the correspondence between the 3×3 evaporator source array and the equalizing plate partitions; Figure 17 This is a comparison diagram of the temperature gradient between a tapered variable-diameter coil and a constant-diameter coil; Figure 18 It is a graph showing the relationship between the dielectric constant ratio and the edge / center thickness ratio; Figure 19 This is a comparison of the Raman spectra of the carbonization products; Figure 20 This is a comparison chart of step response curves from PID parameter tuning experiments; Figure 21 This is a graph showing the standard deviation of the electric field strength in the voltage equalization plate potential optimization experiment. Figure 22 It is a 72-hour long-term operational stability curve; Figure 23 This is a graph showing the effect of vacuum level on film performance; Parts and component numbers in the diagram: 100-Vacuum cavity, 200-High frequency induction heating coil, 300-Laser displacement sensor, 400-Lifting mechanism, 500-Evaporation source. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. In the description of the present invention, it should be understood that the terms "center," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, the element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Where there is no conflict, embodiments of the present invention and the various features thereof can be combined with each other, all of which are within the scope of protection of the present invention.

[0023] Example 1 Please see Figure 1 , Figure 2 and Figure 3 This invention provides a closed-loop control system for the position of a bar evaporator, comprising: The vacuum chamber 100 is kept in a vacuum state during operation; A high-frequency induction heating coil 200 is used to locally heat the top of the evaporation source 500; A laser displacement sensor 300 is installed outside the vacuum chamber 100 and detects the position of the top of the evaporation source 500 in real time through an observation window, and outputs a position signal. A lifting mechanism 400 is provided, wherein the evaporation source 500 is fixedly disposed at the moving end of the lifting mechanism 400, and the lifting mechanism 400 is used to drive the evaporation source 500 to move along the axial direction of the vacuum chamber 100. The controller is electrically connected to the laser displacement sensor 300 and the precision lifting drive mechanism respectively. The controller uses a closed-loop control algorithm to control the lifting mechanism 400 to operate based on the deviation between the position signal and the preset target position, so that the relative position between the top of the evaporation source 500 and the high-frequency induction heating coil 200 is kept constant.

[0024] In use, first set the heating area of ​​the high-frequency induction heating coil 200, then fix the evaporation source 500 on the moving end of the lifting mechanism 400, calibrate the laser displacement sensor 300 by aligning it with the detection position at the top of the evaporation source 500, then evacuate the vacuum chamber 100 to the required vacuum level, and start the high-frequency induction heating coil 200 to heat the top of the evaporation source 500, so that the material in the evaporation source 500 is gradually heated and evaporated. During this process, the laser displacement sensor 300 will continuously collect the real-time position information of the top of the evaporation source 500 through the observation window and transmit the corresponding position signal to the controller. The controller will compare and calculate the real-time position with the preset target position, and after obtaining the position deviation, it will output an adjustment command to the lifting mechanism 400 through the closed-loop control algorithm, driving the lifting mechanism 400 to move the evaporation source 500 upward along the axial direction, compensating for the downward movement of the top position caused by material consumption during the evaporation process, and always keeping the relative position between the top of the evaporation source 500 and the high-frequency induction heating coil 200 constant, ensuring stable heating temperature, and thus obtaining a stable and uniform evaporation rate.

[0025] Furthermore, the high-frequency induction heating coil 200 is made of hollow copper tube and has cooling water circulating inside; the high-frequency induction heating coil 200 is a tapered variable diameter coil, and the bottom inner diameter of the high-frequency induction heating coil 200 is larger than the top inner diameter of the high-frequency induction heating coil 200, and the top inner diameter of the high-frequency induction heating coil 200 is larger than the diameter of the evaporation source 500; the operating frequency of the high-frequency induction heating coil 200 is adaptively adjusted by the load through an impedance matching network; wherein, the number of turns of the spiral coil is set to 3-8 turns.

[0026] The high-frequency induction heating coil 200 is a spiral coil wound with a hollow copper tube, through which cooling water flows. The flow rate of the cooling water is 2-5L / min, and the temperature difference between the inlet and outlet water is <10℃. The spiral coil has 3-8 turns, preferably 5 turns, and its inner diameter is 3-8mm larger than the diameter of the evaporation source 500.

[0027] The high-frequency induction heating coil 200 operates at a frequency of 100-500kHz, preferably 200-400kHz, with an adjustable power of 1-5kW. It can achieve load self-adaptation through an impedance matching network.

[0028] The height of the heating zone of the high-frequency induction heating coil 200 is 10-30mm above the axial height of the coil, forming a clear hot spot area with an axial temperature gradient >50℃ / mm.

[0029] The spiral coil adopts a tapered variable diameter design, with a larger inner diameter at the bottom and a slightly smaller inner diameter at the top, which makes the axial magnetic field distribution more uniform and improves the accuracy of heating point position control.

[0030] To verify the effect of tapered variable diameter coils on improving axial magnetic field distribution, a comparative experiment was conducted.

[0031] Experimental conditions: The C60 wire rod has a diameter of 10mm, 5 coil turns, a frequency of 300kHz, and a power of 2.5kW. Temperature distribution was measured by inserting thermocouples at 5mm intervals along the 500-degree axial direction of the evaporation source using two different coil types: a uniform inner diameter coil (15mm) and a tapered variable diameter coil (16mm bottom inner diameter, 13mm top inner diameter, and approximately 1:10 taper).

[0032] Experimental results: Equal diameter coil: axial temperature gradient 48℃ / mm, axial length of heating zone (>500℃) approximately 12mm; Tapered variable diameter coil: axial temperature gradient 56℃ / mm, axial length of heating zone approximately 9mm, heating zone location more concentrated.

[0033] Specifically, please see Figure 17 When using a tapered variable-diameter coil, the magnetic field distribution is more concentrated, increasing the signal-to-noise ratio of the position sensor from 32dB to 38dB, and reducing the steady-state error of position control from ±0.08mm to ±0.05mm. The experimental data above demonstrates that the tapered variable diameter design effectively improves the spatial resolution and position control accuracy of the heating point by compressing the axial length of the heating zone and increasing the temperature gradient.

[0034] Furthermore, the impedance matching network is an L-type matching network, including an adjustable capacitor and a fixed inductor. The adjustable capacitor has an adjustment range of 10-100pF, and the fixed inductor has an inductance of 2μH. The impedance matching network adopts an automatic tuning algorithm based on the principle of minimizing reflected power, and the matching network parameters are optimized by the gradient descent method, with a tuning time of less than 100ms.

[0035] The high-frequency power supply output impedance of the impedance matching network is 50Ω, and the coil load impedance varies with temperature in the range of 0.5-2Ω. The matching network topology is an L-type matching network, which includes an adjustable capacitor of 10-100pF and a fixed inductor of 2μH.

[0036] The gradient descent method optimizes the matching network parameters, including: Step 1: Set the initial adjustable capacitor value to C=50pF and measure the current reflected power Pr; Step 2: Adjust the value of the adjustable capacitor in steps of ΔC=2pF, measure the adjusted reflected power, and calculate the rate of change of reflected power with respect to the capacitor value, dPr / dC. Step 3: Iteratively adjust the value of the adjustable capacitor along the direction of decreasing reflected power, and remeasure the reflected power after each iteration; Step 4: Repeat steps 2 and 3 until the reflected power is less than 0.1W, thus completing the automatic tuning of the impedance matching network.

[0037] A typical tuning process requires 15-25 iterations and takes less than 100ms at a 100kHz control frequency.

[0038] Experimental verification: Tested under 100 random initial conditions, the average tuning time was 82ms and the maximum tuning time was 96ms, which meets the technical specification of <100ms.

[0039] Please see Figure 1 and Figure 3 Furthermore, the evaporation source 500 is configured as an organic material wire rod; The organic material wire rod is made of at least one of the following: C60 fullerene, Alq3 (8-hydroxyquinoline aluminum), NPB (N,N'-diphenyl-N,N'-(1-naphthyl)-1,1'-biphenyl-4,4'-diamine), and metal phthalocyanines (such as CuPc, ZnPc), which are easily thermally decomposed organic evaporation materials.

[0040] The specifications of the organic material wire rods are: diameter 5-15mm, length 100-300mm, formed by cold isostatic pressing of high-purity organic material powder (≥99.5%).

[0041] Molding process parameters: pressure 200-300MPa, holding time 5-15min, density after molding reaches 85-90% of theoretical density.

[0042] The initial position of the organic material rod is set so that the top of the rod is 2-5 mm below the upper edge of the coil, with only the top 5-10 mm area entering the effective heating zone, while the lower part is in the low-temperature zone (<200℃). As evaporation proceeds, the rod is fed upwards in real time to maintain a constant relative position between the top and the coil. The feed rate is automatically matched according to the evaporation rate, with a typical value of 0.1-0.8 mm / min.

[0043] Specifically, the precision lifting drive mechanism is driven by a servo motor and a precision ball screw with a lead of 2mm, with an encoder resolution of 131072 pulses / revolution and a linear displacement resolution of ≤0.01mm.

[0044] The transmission structure is a magnetohydrodynamic sealed transmission shaft with a leakage rate of <1×10-9Pa·m³ / s, ensuring the reliability of the vacuum dynamic seal.

[0045] Technical parameters of the magnetohydrodynamic sealed drive shaft: Magnetofluid model: Ferrotec APG-J10, saturation magnetization 40mT; Sealing structure: multi-stage magnetic circuit, with a magnetic gap of 0.2 mm per stage and a magnetic fluid filling volume of 0.5 mL per stage; Pressure resistance: >0.2MPa; Leakage rate: <1×10⁻⁹ Pa·m³ / s; To verify the reliability of the magnetohydrodynamic seal under continuous operating conditions, an 8-hour continuous operation test was conducted.

[0046] Test conditions: The drive shaft rotates at 15 rpm, corresponding to a feed rate of 0.5 mm / min, and the vacuum chamber pressure is 5 × 10⁻⁴ Pa. The leakage rate at the seal was measured using a helium mass spectrometer at three time points: at the start of the test, after 4 hours of operation, and after 8 hours of operation.

[0047] Test results: 0 hours: Leakage rate 8.5 × 10⁻¹⁰ Pa·m³ / s 4 hours: Leakage rate 9.2 × 10⁻¹⁰ Pa·m³ / s 8 hours: Leakage rate 9.8 × 10⁻¹⁰ Pa·m³ / s The leakage rate remained within the specified range of <1×10-9 Pa·m³ / s for 8 hours, and the vacuum level of the cavity remained stable at 5×10-4 Pa, with no vacuum deterioration caused by seal leakage. This test proves that the magnetohydrodynamic sealing structure meets the reliability requirements for long-term continuous operation of mass production equipment.

[0048] Travel range: 100-300mm, matching the full length of the line bar.

[0049] Speed ​​range: 0.05-5mm / min stepless speed regulation, maximum acceleration 50mm / s².

[0050] Furthermore, the sensor is configured as a laser triangulation range sensor with a range of ±10mm, linearity of ±0.02%FS, and repeatability of 0.01μm. The sensor is installed outside the vacuum chamber 100 and non-contact measurement is performed through a quartz observation window. The optical path is at a 30° angle to the evaporating vapor flow to avoid vapor deposition and contamination of the lens.

[0051] The quartz observation window has a diameter of 50mm and a thickness of 5mm, and is made of JGS1 far-ultraviolet quartz. Optical path design: incident angle 30°, reflection angle 30°, to avoid direct vapor flow; Pollution prevention measures: Quartz windows are equipped with a nitrogen purging device (flow rate 0.5L / min), which automatically cleans every 2 hours; Pollution monitoring: An alarm is triggered to indicate the need for cleaning when the sensor reading drifts by more than 0.1 mm; Sampling frequency: 100Hz (matching the controller scan cycle).

[0052] Further, please see Figure 7 and Figure 14 The closed-loop control algorithm includes: Step 1: Set the target position. Set the top target position of the evaporation source 500 below the center line of the high-frequency induction heating coil 200. Step 2: The current position of the top of the evaporation source 500 is detected in real time using the laser displacement sensor 300 at a preset sampling frequency; Step 3: Subtract the target position from the current position to obtain the position deviation value. ; Step 4: Determine the deviation range and select the control mode: If the position deviation value If the absolute value is greater than 1.5mm, the system enters a rapid approach mode, and the evaporation source 500 is driven to move toward the target position at the maximum speed V_max via the lifting mechanism 400. If the position deviation value If the absolute value is less than or equal to 1.5mm, then the incremental PID control mode is entered, and steps 5 to 7 are executed. Step 5: Calculate the incremental PID control output and calculate the control increment Δu according to the following formula: in, This is a proportionality coefficient, with a value ranging from 0.6 to 1.0; is the integral coefficient, with a value range of [0.03, 0.05, 0.07]s⁻¹; is the differential coefficient, with a value range of 0.08-0.16s; T is the control period, with a value ranging from 10 to 100 ms; This represents the current cycle position deviation. This represents the positional deviation from the previous cycle. This represents the positional deviation between the previous two cycles; Step 6: If the position deviation value The absolute value is less than the integral separation threshold. Then the control increment Integral terms Set to zero; Step 7: Transfer the speed command from the previous cycle With the control increment Add them together to get the current cycle speed command. and the speed command Amplitude limiting is applied; Step 8: Send the speed command The output is sent to the lifting mechanism 400, which in turn drives the evaporation source 500 to move axially. Step 9: Adjust the current cycle deviation Store as ,Will Store as It records the current cycle speed command and then returns to step 2 to form a closed-loop control.

[0053] To determine the optimal PID parameters, a system parameter tuning experiment was conducted: Phase 1: Preliminary determination of critical proportionality using the method First set Ki and Kd to 0, then gradually increase Kp. When Kp=1.2, the system exhibits constant amplitude oscillation. The critical gain Kc=1.2 and the oscillation period Pc=0.8s are recorded. The initial parameters were calculated using the Ziegler-Nichols empirical formula: Kp = 0.6 × Kc = 0.72; Ki = 2Kp / Pc = 1.8 s⁻¹ → Ki × T = 0.018 (T = 10 ms); Kd = Kp × Pc / 8 = 0.12 s; Phase Two: Orthogonal Experiment Optimization Using steady-state error, overshoot, and settling time as evaluation indicators, a three-factor, three-level orthogonal experiment was conducted within the following ranges: Kp: [0.6, 0.8, 1.0]; Ki: [0.03, 0.05, 0.07] (unit: s⁻¹); Kd: [0.08, 0.12, 0.16] (unit: s).

[0054] Orthogonal experimental results (L9(3) 4 )): The overall scoring weights are: steady-state error 40%, overshoot 30%, and settling time 30%. Experiment 5 (Kp=0.8, Ki=0.05, Kd=0.1) had the highest overall score and was determined to be the optimal parameters.

[0055] To verify the impact of the control cycle on position control accuracy, a comparative experiment was conducted under different control cycles. Test conditions: C60 wire rod diameter 10mm, feed speed 0.5mm / min, control cycles were set to 5ms, 10ms, 20ms, 50ms, 100ms, and 150ms respectively, each cycle was run for 60min, and steady-state position error and evaporation rate CV value were recorded. Control cycle steady-state position error Overshoot Evaporation rate CV evaluate 5ms ±0.04mm 3.50% 2.20% High processor load 10ms ±0.05mm 4.20% 2.30% Preferred 20ms ±0.06mm 4.50% 2.40% good 50ms ±0.07mm 5.10% 2.60% Acceptable 100ms ±0.09mm 6.30% 3.00% Boundary values 150ms ±0.15mm 8.80% 4.10% Exceeding the permitted range

[0056] Experiments show that when the control cycle is shorter than 10ms, the improvement in control accuracy is limited, but the processor load increases significantly; when the control cycle exceeds 100ms, the position error and evaporation rate fluctuations increase significantly. Therefore, a control cycle range of 10-100ms is the optimal operating range for this invention, preferably 10-50ms.

[0057] Comparative analysis of control algorithms: Please see Figure 20 In other embodiments of the present invention, the controller may employ a fuzzy PID control algorithm or a model predictive control (MPC) algorithm. However, experimental verification has shown that for the evaporation control of easily thermally decomposable organic materials such as C60, the incremental PID control algorithm exhibits the best overall performance in terms of response speed, steady-state accuracy, and anti-interference capability. The fuzzy PID algorithm may generate overshoot during parameter self-tuning, leading to fluctuations in the rod position; the MPC algorithm has high computational complexity and is difficult to complete the optimization solution within a 10ms control cycle. Therefore, the present invention preferably employs an incremental PID control algorithm. Control Algorithm Overshoot Adjusting time steady-state error Anti-interference capability Applicability evaluation Incremental PID <5% <30s ±0.05mm powerful Preferred Fuzzy PID 8-12% 25-40s ±0.08mm middle Available but not optimal MPC <3% 20-35s ±0.03mm powerful High computational complexity

[0058] Control performance indicators: steady-state position error < ±0.05mm, overshoot < 5%, settling time < 30s (from initial position to set position).

[0059] Example 2 Please see Figure 2 , Figure 3 and Figure 4 This invention provides a suspension pressure equalization deposition system, comprising: The closed-loop control system for the position of the bar evaporator source described in Example 1; A substrate support stage is set inside the vacuum chamber 100 and is used to place the substrate to be processed. A negative bias electrode, disposed within the vacuum cavity 100 and located between the high-frequency induction heating coil 200 and the substrate, is used to apply a negative DC high voltage to generate a deposition electric field. A DC high voltage of -10kV to -30kV is applied between the evaporation source 500 and the substrate to generate a directional electric field, causing partial ionization of the organic material vapor and accelerating it towards the substrate, thus improving the film density. Its chamfered edges (R≥5mm) can initially mitigate tip discharge.

[0060] A floating equalizing plate is disposed below the substrate support stage, parallel to and insulated from the substrate support stage. The floating equalizing plate is connected to an adjustable DC bias power supply to keep it in a floating potential state. It is disposed 1-5mm below the substrate and a floating potential of -2kV to -5kV is applied to reshape the equipotential surface of the substrate, flatten the distorted electric field lines at the edge, and reduce the electric field strength at the edge of the substrate from 180kV / m in the traditional scheme to about 118kV / m. The edge / center film thickness ratio is increased from 0.68 to more than 0.92.

[0061] A double-layer insulating protective ring is arranged around the periphery of the suspended equalizing plate, including an inner ring and an outer ring. The dielectric constant of the inner ring is smaller than that of the outer ring. It is composed of an inner layer of PTFE (low dielectric constant) and an outer layer of alumina ceramic (high dielectric constant). By utilizing the refraction effect of the electric field lines caused by the abrupt change in dielectric constant, it further suppresses the concentration of residual field strength around the equalizing plate, eliminates corona discharge, and lengthens the creepage distance (equivalent to an increase of 32mm).

[0062] Furthermore, the bias electrode includes an electrode body disposed between the evaporation source 500 and the processing substrate, the electrode body being opposite to and parallel to the processing substrate; the electrode body is flat or micro-arc-shaped, the material of the electrode body is stainless steel 316L, and the surface is plated with a nickel layer.

[0063] Specifically, the electrode body is located between the coil and the substrate, 100-300mm away from the substrate, and a DC negative high voltage of -5kV to -60kV is applied, preferably -10kV to -30kV. The power supply ripple is <0.1%, and it has arc detection and fast shutdown functions.

[0064] The electrode is in the shape of a flat plate or a slightly curved surface, with a polished surface Ra≤0.4μm and an edge chamfer R≥5mm. The electrode material is 316L stainless steel, with a nickel plating to prevent oxidation.

[0065] Positive bias substrate: The conductive glass is grounded or a positive bias voltage of 0~+100V is applied to form a directional deposition electric field.

[0066] Ionization mechanism: Organic material vapor is partially ionized by electron collisions in a strong electric field, with an ionization rate of about 1-5%. Positive ions are repelled by the negative electrode and accelerated towards the substrate, gaining additional kinetic energy and improving the film density.

[0067] Furthermore, the suspended equalizing plate includes: an equalizing plate body disposed below the substrate support stage, the equalizing plate body being parallel to and insulated from the substrate support stage; the equalizing plate body is made of aluminum alloy 6061 and its surface is hard anodized.

[0068] The high-pressure equalizing plate is installed 3mm below the processing substrate, with an adjustable range of 1-5mm, and is strictly parallel to the processing substrate.

[0069] Material and treatment: Aluminum alloy 6061, hard anodized with a surface thickness of 20μm, insulation resistance >1GΩ.

[0070] Floating potential control: Active potential regulation is achieved by connecting a series of high-resistance resistors to an adjustable DC bias power supply.

[0071] To determine the optimal suspension potential, a systematic single-factor optimization experiment was conducted: Experimental conditions: negative electrode -20kV, substrate grounded, electrode spacing 200mm, equalizing plate size 320mm×320mm, to test the uniformity of electric field distribution under different suspension potentials.

[0072] Test method: The electric field strength is measured radially along the substrate surface using an electrostatic probe, and the standard deviation σ and uniformity index U=(Emax-Emin) / Eavg are calculated.

[0073] Experimental results: When the levitation potential is -3kV, the standard deviation of the electric field strength on the substrate surface is the smallest (8.2kV / m), and the edge / center thickness ratio reaches 0.92, which is the optimal value. Considering the process margin and adaptability to different substrate sizes, the adjustable range was finally selected as -2kV to -5kV, preferably -3kV to -3.5kV.

[0074] Precise control circuit for the floating potential of the equalizing plate: High-resistance resistor chain: 10 x 10MΩ resistors connected in series, total resistance 100MΩ Adjustable bias power supply: Output range 0 to -5kV, resolution 10V, ripple <0.01% Potential monitoring: The voltage of the equalizing plate is monitored in real time via a high-voltage divider (1000:1). Closed-loop regulation: Automatically optimizes the voltage equalization plate potential based on substrate film thickness monitoring feedback (optional function). Please see Figure 9Electric field simulation results: COMSOL Multiphysics was used to simulate the electrostatic field and compare the electric field distribution with and without the equalizing plate.

[0075] Simulation parameters: negative electrode potential -20kV, substrate grounded, electrode spacing 200mm, equalizing plate size 320mm×320mm, floating potential -3.5kV.

[0076] result: Without the equalizing plate: the electric field strength at the center of the substrate surface is 105 kV / m, and the electric field strength at 20 mm from the edge is 182 kV / m (73% enhancement).

[0077] With an equalizing plate: the electric field strength at the center of the substrate surface is 108kV / m, and the electric field strength at 20mm from the edge is 118kV / m (only 9% increase), and the edge electric field distortion is effectively suppressed.

[0078] Further, please see Figure 12 and Figure 21 The double-layer insulating protective ring includes an inner ring and an outer ring, wherein the inner ring and the outer ring are concentric ring structures and are arranged around the periphery of the suspended equalizing plate; The inner ring is made of polytetrafluoroethylene; the outer ring is made of alumina ceramic; the dielectric constant ratio of the inner ring to the outer ring is 1:3 to 1:5.

[0079] The material selection for the double-layer insulating protective ring is based on the following principles: The inner ring uses a low dielectric constant material with εr < 3, causing the electric field lines to refract in the normal direction at the PTFE-vacuum interface, reducing the tangential electric field component on the surface. The outer ring uses a high dielectric constant material with εr > 5, further extending the creepage distance. The dielectric constant ratio of the two layers needs to be greater than 3:1. If the ratio is too small, the electric field refraction effect will be insufficient, and the edge distortion suppression effect will be significantly reduced.

[0080] To verify the effect of dielectric constant ratio on edge electric field suppression, a comparative experiment was conducted with different ratio combinations. The inner ring was fixed to PTFE (εr≈2.1), and the outer ring was formed with the following materials to form different ratios: quartz (εr≈3.8, ratio 1:1.8), alumina ceramic (εr≈9.8, ratio 1:4.7), and barium titanate ceramic (εr≈1500, ratio 1:714).

[0081] Experimental conditions: negative electrode -20kV, substrate grounded, electrode spacing 200mm, electric field strength and edge / center ratio of film thickness measured 10mm from the edge of the substrate. Outer ring material Dielectric constant ratio Edge electric field strength Edge / center thickness ratio evaluate No protective ring --- 182kV / m 0.68 Unacceptable quartz 01:01.8 145kV / m 0.78 Improve shortcomings 99% aluminum oxide 01:04.7 118kV / m 0.92 Preferred Barium titanate 0.5375 112kV / m 0.93 Through design

[0082] Experiments show: Please refer to Figure 18 When the dielectric constant ratio is less than 2:1 (quartz outer ring), the edge / center thickness ratio only increases to 0.78, which is insufficient. When the ratio reaches 1:4.7, the edge / center ratio reaches 0.92, meeting mass production requirements. Further increasing the ratio to 1:714 provides limited improvement but significantly increases material costs. Therefore, a dielectric constant ratio range of 1:3 to 1:5 is the optimal range for this invention.

[0083] Structure: A double-layer concentric ring structure surrounding the pressure equalization plate, with a rectangular cross-section.

[0084] Inner ring: Polytetrafluoroethylene (PTFE, dielectric constant εr≈2.1), 8 mm thick, 18 mm high (extending 10 mm beyond the top surface of the equalizing plate). PTFE has an extremely low secondary electron emission coefficient (<1.5) and a high surface flashover voltage.

[0085] Outer ring: 99% alumina ceramic (dielectric constant εr≈9.8), 15mm thick, 22mm high (14mm above the top surface of the equalizing plate). The high dielectric constant of the ceramic can further extend the creepage distance.

[0086] Dielectric constant ratio: The dielectric constant ratio of the inner ring to the outer ring ranges from 1:3 to 1:5 (1:4.7 in this embodiment).

[0087] Electric field simulation comparison: Single-layer PTFE ring (thickness 8mm): The peak electric field strength at the outer edge of the protective ring is 85kV / m, and the creepage distance is equivalent to an increase of 15mm.

[0088] Double-ring structure (PTFE + ceramic): The peak electric field strength at the outer edge of the protective ring is reduced to 48kV / m, and the creepage distance is effectively increased by 32mm.

[0089] Stepped electric field distribution: The double-layer ring forms two abrupt changes in dielectric constant, and the electric field lines are refracted at the interface, which effectively avoids the concentration of field strength in the single-layer structure.

[0090] Please see Figure 4 and Figure 11 Specifically, the workflow is as follows: Step 1: Initial loading and positioning Insert the C60 wire bar into the vacuum chamber 100, and set the initial position of the top end 2-5mm below the upper edge of the coil; calibrate the position sensor to zero and set the target maintenance position (e.g., move the coil center line down 2mm).

[0091] Step 2: Vacuuming Start the vacuum system and pump the cavity to a vacuum level of ≤10-3 Pa (preferably ≤5×10-4 Pa).

[0092] Step 3: Start heating and closed-loop control The high-frequency power supply is started, and the coil heats the top of the C60 bar to the sublimation temperature (500-800℃). The position sensor monitors the top position in real time; As the vapor evaporates and is consumed, the rod length shortens, and the sensor detects a decrease in position. When the drop exceeds a threshold (e.g., 1mm), the PLC controller starts the lifting motor to drive the bar to move upward to compensate. Maintain a constant position of the heating point relative to the coil to ensure a stable evaporation rate.

[0093] Step 4: Bias Deposition and Isostatic Protection After stable evaporation is achieved, start the negative bias power supply and apply -10kV to -30kV; C60 vapor is partially ionized and accelerated towards the substrate under the action of an electric field; The equalizing plate is suspended below the processing substrate to homogenize the surface potential; The double-layer protective ring suppresses edge electric field distortion, enabling large-area uniform deposition.

[0094] Step 5: End and Alarm When the C60 rod is consumed to the lower limit (e.g., 20mm remaining), the system will automatically stop heating. The lifting mechanism resets at 400° and issues a material change alarm.

[0095] Example 3 Verification of basic closed-loop control system: Experimental setup: Organic material bar: C60, diameter 10mm, length 200mm, purity 99.9% Coil: 5 turns, inner diameter 15mm, height 20mm, frequency 300kHz, power 2.5kW Position sensor: KeyenceLK-G152, mounted outside the vacuum chamber, measures through a φ50mm quartz window. Lifting mechanism 400: Panasonic servo motor + THK ball screw, speed setting 0.5mm / min Controller: Siemens S7-1200 PLC, executing incremental PID algorithm; Vacuum degree: 5×10⁻⁴ Pa Test method: C60 wire rods were continuously evaporated for 120 min, and the evaporation rate was recorded in real time using a quartz crystal film thickness monitor (Inficon IC6) with a sampling interval of 1 s. Simultaneously, a silicon wafer was placed on the substrate to collect the film layer, and the film thickness distribution was measured using a profilometer. The particle defect density was statistically analyzed using an optical microscope.

[0096] Experimental results: Evaporation rate stability: Please refer to Figure 5 The average evaporation rate over 120 min was 0.52 nm / s, with a standard deviation σ = 0.012 nm / s and a coefficient of variation CV = 2.3%. In contrast, the control group using the traditional "deep insertion preheating" process with an insertion depth of 35 mm and no position feedback had a CV of 12.7%.

[0097] Raw material utilization rate: After the experiment, the remaining rods were weighed. The initial mass was 28.5g, the remaining mass was 1.2g, and the actual consumption was 27.3g. The mass of C60 deposited on the substrate was 26.1g, and the raw material utilization rate was 26.1 / 27.3 = 95.6%. The utilization rate of the control group was 68%.

[0098] Temperature distribution in the non-evaporation zone: Thermocouples were inserted into the side of the rod at distances of 20mm, 40mm, and 60mm from the top, and the temperatures during stable evaporation were recorded. The temperature at 20mm was 180℃, at 40mm it was 95℃, and at 60mm it was 52℃. All of these temperatures are below the thermal decomposition initiation temperature of C60, thus preventing carbonization in the non-evaporation zone.

[0099] Film defect density: Observed under an optical microscope, the particle defect density was <3 particles / mm², while the control group was >50 particles / mm².

[0100] To confirm the properties of the carbonization products generated in the non-evaporation zone during the deep insertion preheating process, samples were taken from C60 rods 30 mm from the top after 120 min of the conventional process, and Raman spectroscopy analysis was performed. Pristine C60 powder and commercial graphite powder were used as references.

[0101] Please see Figure 19 Comparison of spectral characteristic peaks: sample C60 characteristic peak (Ag(2)) D peak G Peak ID / IG Original C60 powder 1469cm-1 (Strong) Not detected Not detected --- 30mm after deep insertion and preheating 1469cm - 1 (weak) 1350cm-1 1580cm-1 1.12 Commercial graphite powder Not detected 1350cm-1 1580cm-1 0.95 The results showed that the preheated rod after deep insertion had obvious amorphous carbon (D peak) and graphitic carbon (G peak) at 30 mm, confirming the occurrence of thermal decomposition carbonization reaction. The carbonization products were released in particulate form during the subsequent evaporation process, causing defects in the film layer.

[0102] Example 4 Verification of the electric field homogenization system: Experimental setup: Negative bias electrode: 300mm in diameter, 5mm in thickness, 316L stainless steel, 200mm from the substrate, voltage -20kV; Equalizing plate: 320mm×320mm, aluminum alloy + hard anodized, suspended 3mm below the substrate, with a suspension potential of -3.5kV; Protective ring: Inner layer PTFE 8mm×18mm, outer layer ceramic 15mm×22mm; Substrate used for fabrication: 400mm×500mm FTO glass, grounded; Evaporation source 500: The closed-loop control evaporation source 500 of Example 1 is adopted.

[0103] Test method: A C60 thin film with a target thickness of 100 nm was deposited. The film thickness was measured at 9 points along the long side of the substrate using a spectroscopic ellipsometry, and the edge / center thickness ratio was calculated. The surface roughness of the film was measured using an atomic force microscope.

[0104] Experimental results: Film thickness uniformity: Please refer to Figure 6 The thickness of the central region measured at 9 points was 102.3 nm, while the thicknesses of the edge regions were 95.1 nm and 94.7 nm, respectively, with edge / center thickness ratios of 0.93 and 0.92. The edge / center ratio of the control group without the equalizing plate / protective ring was 0.68.

[0105] Surface roughness: AFM scanning showed that the root mean square roughness of the film was Rq=1.2nm, while that of the control group was Rq=2.8nm.

[0106] Edge blurring phenomenon: According to visual observation, the color of the film layer at the edge of the substrate in the experimental group is uniform and there is no blurring or mottled phenomenon; the control group has a color difference transition zone of about 20mm at the edge.

[0107] Consistency between electric field simulation and actual measurement: The surface potential distribution of the substrate was measured using an electrostatic probe, and the deviation from the COMSOL simulation results was less than 8%, verifying the accuracy of the simulation.

[0108] To further verify the effect of electric field homogenization on improving the functional performance of C60 thin films, electrical performance comparison tests were conducted on the thin films prepared in Example 2 (experimental group) and the thin films prepared in Comparative Example 2 (control group without equalizing plate). Electron mobility was measured using the space charge confined current (SCLC) method, and the device structure was ITO / C60 (100nm) / Al.

[0109] Please see Figure 15 Test results: sample Electron mobility cm² / V·s Electrical conductivity (S / cm) Breakthrough field strength (MV / cm) Experimental group (with equalizing plate) 1.8×10-3 2.3×10-7 2.1 Control group (without pressure equalization plate) 1.2×10-3 1.5×10-7 1.6 Compared with the control group, the experimental group showed a 50% increase in electron mobility, a 53% increase in conductivity, and a 31% increase in breakdown field strength. This is because the electric field homogenization resulted in a uniform kinetic energy distribution of C60 ions on the substrate surface, leading to consistent film density and crystallinity, thereby improving charge transport performance. These results demonstrate that the electric field homogenization scheme of this invention not only improves film thickness uniformity but also significantly enhances the electrical functional properties of the film.

[0110] Example 5: Large-scale mass production system validation: Experimental setup: Please see Figure 13 and Figure 16 Evaporation source array 500: 3×3 layout, source spacing 180mm, each source has independent closed-loop control. Processing substrate size: 1100mm × 1300mm ITO glass; Equalizing plate: zoned design, divided into 9 independent zones, each zone's suspension potential is independently adjustable, and the absolute value of the suspension potential in the edge zone is 0.5-1.5kV greater than that in the center zone to compensate for the difference in vapor flow density at the edge. Source spacing optimization method: Based on the cosine distribution model of steam flow, the source spacing is determined by superposition calculation. For point sources, the film thickness distribution is d∝cosⁿθ, and the value of n is determined experimentally. In this experiment, n=1.8 is adopted for C60.

[0111] Collaborative control scheme: Each source uses the same setpoint, but the PID parameters are fine-tuned according to the differences in heat load of each source. The evaporation rate of each source is fed back through an independent quartz crystal monitor, and the host computer coordinates the rate deviation of each source to be less than 5%.

[0112] Test method: C60 thin films were deposited with a target thickness of 100 nm. The entire substrate was scanned using a fully automated ellipsometry with a step size of 100 mm to calculate the film thickness uniformity.

[0113] Experimental results: Film thickness statistics at 121 points across the entire surface: average thickness 101.5nm, standard deviation 4.8nm, relative standard deviation RSD=4.7%, i.e. film thickness uniformity ±4.7%, which meets mass production requirements.

[0114] Long-term stability: After 8 hours of continuous operation, with one sampled substrate per hour, the film thickness RSD remained within the range of 4.5%-5.1%.

[0115] Raw material utilization rate: The initial load of raw materials was 250g, and after 8 hours, the remaining material was 18g, with a consumption of 232g. The mass of the film layer collected from the processed substrate was 221g, resulting in a utilization rate of 95.3%.

[0116] Example 6: Material scalability verification (Alq3 green fluorescent material): Using the same apparatus as in Example 1, the C60 bar was replaced with an Alq3 bar, 10 mm in diameter, 150 mm in length, and 99.5% pure. The sublimation temperature of Alq3 is approximately 300°C, and its thermal decomposition temperature is approximately 380°C.

[0117] Key parameter adjustments: Heating power: 1.2kW; Target location: 3mm below the upper edge of the coil; Feed rate: 0.8 mm / min; Experimental results: Evaporation rate stability: CV = 2.8% Temperature in the non-evaporation zone: The temperature 15mm from the top is only 160℃, which is far below the decomposition temperature; Raw material utilization rate: 94.8%; PL spectroscopy test: The fluorescence peak of the deposited Alq3 film is 530 nm, and the full width at half maximum (FWHM) is 65 nm, which is consistent with the standard Alq3 film, proving that there is no contamination by thermal decomposition products.

[0118] This embodiment demonstrates that the present invention is also applicable to other organic evaporation materials that are prone to thermal decomposition.

[0119] Example 5 Endpoint parameter validation: To support the broad range of parameters in the claims, the following endpoint tests are performed: Detailed data from frequency endpoint tests: frequency power Temperature gradient Position accuracy CV value Raw material utilization rate 100kHz 2kW 35℃ / mm ±0.08mm 3.20% 93.50% 200kHz 2kW 48℃ / mm ±0.05mm 2.50% 95.10% 300kHz 2.5kW 55℃ / mm ±0.05mm 2.30% 95.60% 400kHz 2kW 62℃ / mm ±0.04mm 2.20% 95.40% 500kHz 2kW 68℃ / mm ±0.03mm 2.10% 95.20% Note: 300kHz is the preferred frequency, balancing heating efficiency and position control accuracy.

[0120] Bias endpoint test: bias Film compactness (Rq) Edge / Center Ratio Anti-splashing situation Overall evaluation -5kV 2.1nm 0.88 none The effect is average. -10kV 1.8nm 0.9 none good -20kV 1.2nm 0.92 none Preferred -30kV 0.9nm 0.91 trace amounts good -60kV 0.9nm 0.87 obvious Available but not preferred Note: Backsputtering is more pronounced at -60kV, with a slight decrease in edge / center ratio compared to -30kV, but film fabrication is still possible. For applications requiring the highest film thickness uniformity, a range of -10kV to -30kV is recommended; for special applications that allow for some uniformity loss but require the highest density, -60kV remains an option.

[0121] Example 6 NPB material evaporation verification: Experimental setup: Same as in Example 1, except that the C60 wire rod was replaced with an NPB wire rod, 10 mm in diameter, 180 mm in length, and 99.5% pure. NPB (N,N'-diphenyl-N,N'-(1-naphthyl)-1,1'-biphenyl-4,4'-diamine) is a hole transport material with a sublimation temperature of approximately 350°C and a thermal decomposition temperature of approximately 420°C.

[0122] Key parameter adjustments: Heating power: 1.5kW; Target location: 2.5mm below the upper edge of the coil; Feed rate: 0.6 mm / min; PID parameter adjustment: Kp=0.75, Ki=0.04, Kd=0.12; Experimental results: Evaporation rate stability: After 120 min of continuous evaporation, the average rate was 0.48 nm / s, and the CV was 2.6%. Temperature in the non-evaporation zone: 165℃ at 20mm from the top and 88℃ at 40mm, both below the decomposition temperature; Raw material utilization rate: 94.2%; Film properties: PL spectroscopy tests showed that the NPB film had a fluorescence peak at 435 nm and a full width at half maximum (FWHM) of 52 nm, consistent with the standard NPB film. Hole mobility: Measured by the space charge confinement current (SCLC) method, the mobility was 5.2 × 10⁻⁴ cm² / V·s, consistent with the literature value; This embodiment demonstrates that the present invention is applicable to hole transport materials such as NPB, and its technical effect is comparable to that of C60.

[0123] Example 7: CuPc material evaporation verification: Experimental setup: Same as in Example 1, except that the C60 wire rod was replaced with a CuPc (copper phthalocyanine) wire rod, 12 mm in diameter, 200 mm in length, and 99% pure. CuPc is a p-type organic semiconductor with a sublimation temperature of approximately 450°C and a thermal decomposition temperature of approximately 550°C.

[0124] Key parameter adjustments: Heating power: 2.8kW; Target position: 2mm below the upper edge of the coil; Feed rate: 0.3 mm / min; PID parameter adjustment: Kp=0.85, Ki=0.06, Kd=0.08; Heating frequency: 350kHz; Experimental results: Evaporation rate stability: After 120 min of continuous evaporation, the average rate was 0.35 nm / s, and the CV was 2.9%. Temperature in the non-evaporation zone: 195℃ at 20mm from the top and 102℃ at 40mm from the top, both far below the decomposition temperature; Raw material utilization rate: 93.8%; Film properties: UV-Vis absorption spectroscopy shows that the Q band absorption peaks are located at 678 nm and 610 nm, consistent with standard CuPc films; XRD shows characteristic peaks of α crystal form. Electrical performance: Field-effect transistor (FET) test, mobility 1.8 × 10⁻³ cm² / V·s, on / off ratio 10 5 ; This embodiment demonstrates that the present invention is applicable to metal phthalocyanine materials, and the technical effect meets the requirements for device fabrication.

[0125] Example 8 Control algorithm comparison and verification: Experimental objective: To verify the performance differences of incremental PID, fuzzy PID, and MPC control algorithms in this application scenario.

[0126] Experimental setup: Same as Example 1, only the control algorithm is changed. The implementation details of the three algorithms are as follows: Incremental PID: As described in Section 5.2.4, Kp=0.8, Ki=0.05, Kd=0.1, control period 10ms; Fuzzy PID: A two-dimensional fuzzy controller is used, with inputs being the error e and the rate of change of error ec, and outputs being ΔKp, ΔKi, and ΔKd. The fuzzy rule base contains 49 rules, and Mamdani inference and the centroid method are used for defuzzification. MPC: Employs a state-space model, with prediction time domain Np=10 and control time domain Nc=3. The objective function includes penalties for position error and velocity rate of change. The optimal control sequence is solved using quadratic programming, with a computation period of 50ms. Test method: Each algorithm was run for 120 minutes, and the position control accuracy, overshoot, and settling time were recorded. Simulated disturbance: A +0.2mm positional disturbance was artificially introduced at 60 minutes, and the recovery time was recorded; Repeat 3 times and take the average. Experimental results: Control Algorithm steady-state error Overshoot Adjusting time Interference recovery time CV value at 120 min Processor load Incremental PID ±0.05mm 4.20% 28s 12s 2.30% 35% Fuzzy PID ±0.08mm 9.50% 35s 18s 3.10% 58% MPC ±0.04mm 2.80% 32s 10s 2.50% 89% Note: The calculation cycle of the MPC algorithm is 50ms, which exceeds the preferred control cycle range (10-100ms) of this invention.

[0127] Computational complexity analysis: Incremental PID: Each iteration requires 6 multiplications and 4 additions, with a computational load of <100 FLOPs and a processor load of <35% within a 10ms cycle; Fuzzy PID: Each iteration requires fuzzification, rule matching, inference, and defuzzification, with a computational load of approximately 5000 FLOPs and a processor load of approximately 58%. MPC: Each iteration requires solving a quadratic programming problem, with a computational load of approximately 500,000 FLOPs and a processor load of up to 89%, making it difficult to complete within 10ms; Conclusion: Incremental PID has the best overall performance in this application scenario, with overshoot <5% and settling time <30s, meeting the requirements of real-time control, and with moderate processor load; Please see Figure 10 Fuzzy PID generates a large overshoot during parameter self-tuning, which leads to fluctuations in the position of the bar and a decrease in the stability of the evaporation rate. The MPC algorithm has high control precision, but its computational complexity is too high. It is difficult to complete the optimization solution within a 10ms control cycle, and it has too high requirements for processor performance, making it unsuitable for this application scenario. Therefore, the present invention preferably employs an incremental PID control algorithm.

[0128] Example 9 Please see Figure 22 Long-term operational stability verification (72 hours): Experimental setup: Same as in Example 5, using a 3×3 evaporation source 500 array and a G5 fabricated substrate.

[0129] Test method: Run continuously for three production shifts for 72 hours, and record the following parameters every 4 hours: evaporation rate and CV value of each source, position control error, vacuum degree, pressure equalization plate potential stability, and film thickness uniformity. Experimental results: runtime average rate Rate CV Position error vacuum degree Film thicknessRSD (h) (nm / s) (%) (mm) (Pa) (%) 0-8 0.52 2.3 ±0.05 4.8×10-4 4.7 8-16 0.51 2.4 ±0.05 5.1×10-4 4.8 16-24 0.51 2.5 ±0.06 5.2×10-4 4.9 24-32 0.5 2.6 ±0.06 5.3×10-4 5 32-40 0.5 2.7 ±0.07 5.4×10-4 5.1 40-48 0.49 2.8 ±0.07 5.5×10-4 5.2 48-56 0.49 2.9 ±0.08 5.6×10-4 5.3 56-64 0.48 3 ±0.08 5.7×10-4 5.4 64-72 0.48 3.1 ±0.09 5.8×10-4 5.5 Results analysis: The evaporation rate slowly decreased from 0.52 nm / s to 0.48 nm / s within 72 hours, a decrease of 7.7%, mainly due to the gradual deposition of carbides on the coil surface, which led to a slight decrease in coupling efficiency. The rate CV value slowly increased from 2.3% to 3.1%, still meeting the technical target of <3%; The position control error remained within ±0.1mm, demonstrating the long-term stability of the closed-loop control system. The vacuum level was maintained at <6×10-4 Pa, and no obvious leakage was observed; The film thickness uniformity slowly deteriorated from 4.7% to 5.5%, and maintenance cleaning is recommended every 48 hours; This embodiment demonstrates that the present invention has the ability to operate stably for 72 consecutive hours, meeting the continuous production requirements of mass production lines.

[0130] Example 10 Performance verification under different vacuum conditions: Experimental setup: Same as in Example 3, the vacuum level of the chamber is controlled at three levels by adjusting the pumping speed of the vacuum pump group.

[0131] Test conditions: Low vacuum: 1×10⁻³ Pa; Medium vacuum: 5 × 10⁻⁴ Pa; High vacuum: 1×10⁻⁴ Pa; Experimental results: Results analysis: Increased vacuum helps reduce the defect density of the film layer. The evaporation rate decreases slightly with increasing vacuum, the rate stability improves slightly with increasing vacuum, and the electron mobility increases with increasing vacuum.

[0132] Please see Figure 23 This embodiment demonstrates that the present invention can operate normally in the range of 1×10-3 Pa to 1×10-4 Pa, with a preferred vacuum degree of ≤5×10-4 Pa.

[0133] Comparative Example 1: Traditional Deep Insertion Preheating Process Experimental setup: Same as in Example 3, but using a conventional control method; C60 wire insertion depth: 35mm; No position sensor feedback; Without a lifting compensation mechanism, the bar remains stationary; Other parameters are the same: 300kHz, 2.5kW, vacuum degree 5×10-4Pa; Experimental results: Evaporation rate fluctuation: CV = 12.7% within 120 min, with the rate gradually decreasing from 0.65 nm / s to 0.42 nm / s; Raw material carbonization: The temperature reaches 380℃ at 30mm from the top, forming a distinct carbonized layer with a thickness of about 5mm; Raw material utilization rate: 68%; Film defects: Particle defect density > 50 particles / mm², black particles visible to the naked eye; Film quality: Surface roughness Rq = 3.5 nm, which is much higher than 1.2 nm in Example 3; Please see Figure 8 The comparative conclusion is that the closed-loop position control technology of the present invention is significantly better than the traditional deep insertion preheating process, with a 40.6% increase in raw material utilization and a 5.5-fold increase in rate stability.

[0134] Comparative Example 2: Electric Field Deposition without Equalizing Plate Experimental setup: Same as in Example 4, but without the equalizing plate and double protective ring. Traditional parallel plate electrode structure; Negative bias electrode -20kV, substrate grounded during processing; The electrode spacing is 200mm, and the electrode diameter is 300mm. All other parameters are the same; Experimental results: Film thickness uniformity: edge / center thickness ratio 0.68, with a thickness of only 68nm in the 20mm edge region; Electric field distribution: The field strength at the edge is 182 kV / m, which is 1.82 times that of the central region (100 kV / m); Blurred edges: A 20mm wide color difference transition zone is visible to the naked eye, and the film layer appears mottled; Corona discharge: Local discharge occurs at the electrode edge, and the vacuum level fluctuates to 8×10-3 Pa; Film density: Due to excessively high ion kinetic energy, the edge region exhibits abnormal density and microcracks. Comparative conclusion: The equalizing plate + double-layer protective ring structure of the present invention significantly improves the uniformity of the electric field, increasing the edge / center ratio from 0.68 to over 0.92 and eliminating the edge blurring phenomenon.

[0135] Comparative Example 3: Verification of differences from existing patented technical solutions Comparative Example 3-1: Verification of the differences between this and the CN106835029A scheme To verify the substantial differences between this invention and the scheme disclosed in CN106835029A, a comparative device simulating that patent was constructed. According to the description in the patent specification, radio frequency heating components dispersed inside the crucible are used to load C60 powder into the crucible, with a heating power of 2.5kW, and no position sensor or lifting mechanism 400 is provided.

[0136] Experimental results: Evaporation rate stability: CV = 9.8% within 120 min, evaporation rate decreased from 0.55 nm / s to 0.41 nm / s; Raw material utilization rate: approximately 72%; Defect density in film layer: 15 defects / mm²; Difference Analysis: CN106835029A improves the local overheating problem of powder evaporation by using a dispersed heating component, but it cannot achieve continuous feeding and real-time closed-loop position control of the wire rod material. This invention differs fundamentally from it in three aspects: (1) Material form; (2) Material supply method; (3) Control strategy.

[0137] The significant difference in technical effects further proves that the two are not simply substitutes.

[0138] Comparative Example 3-2: Verification of the differences between this and the CN121575356A scheme This patent suppresses abrupt changes in evaporation rate by mixing high-melting-point dielectric powder into fullerene powder. According to its published information, C60 is evaporated using the mixed powder without applying a bias electric field.

[0139] Experimental results: Evaporation rate stability: CV = 5.2%; Membrane purity: EDS analysis showed that the membrane contained approximately 0.3 at% W impurities; Electron mobility: 1.0×10⁻³ cm² / V·s, lower than the 1.8×10⁻³ cm² / V·s of this invention, due to increased impurity scattering; Difference Analysis: CN121575356A addresses the problem at the material formulation level, sacrificing membrane purity for improved rate stability. This invention addresses the problem at the equipment and control level, maintaining high membrane purity without altering the evaporation material itself. The two technologies are fundamentally different, and this invention offers advantages in rate stability, membrane purity, and electrical performance.

[0140] In this embodiment, the degree of automation is improved, and no manual intervention is required to position the bar throughout the process, reducing the skill requirements of operators and human error.

[0141] Extended maintenance cycle: No carbonized products accumulate in the non-evaporation zone, the cleanliness of the inner wall of the vacuum chamber is improved, and the opening and cleaning cycle is extended from each batch to each month.

[0142] Economic Analysis: Taking a production line with an annual output of 100,000 G5 substrates (C60 electron transport layer) as an example, the raw material utilization rate increases from 68% to 95%, saving approximately 42 kg of C60 raw material annually (based on a consumption of 400 mg per substrate). With a C60 unit price of approximately 500 yuan / g, the annual raw material cost savings amount to approximately 21 million yuan. Simultaneously, the yield rate increases from 75% to 95%, increasing the annual output of high-quality products by 20,000 substrates, resulting in significant overall economic benefits.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A closed-loop control system for the position of a bar evaporator, characterized in that, include: The vacuum chamber (100) is kept in a vacuum state during operation; A high-frequency induction heating coil (200) is used to locally heat the top of the evaporation source (500); A laser displacement sensor (300) is installed outside the vacuum cavity (100) and detects the position of the top of the evaporation source (500) in real time through the observation window, and outputs a position signal; A lifting mechanism (400) is provided, wherein the evaporation source (500) is fixedly disposed at the moving end of the lifting mechanism (400), and the lifting mechanism (400) is used to drive the evaporation source (500) to move along the axial direction of the vacuum chamber (100); The controller is electrically connected to the laser displacement sensor (300) and the precision lifting drive mechanism respectively. The controller uses a closed-loop control algorithm to control the lifting mechanism (400) to operate according to the deviation between the position signal and the preset target position, so that the relative position between the top of the evaporation source (500) and the high-frequency induction heating coil (200) remains constant.

2. The closed-loop position control system for the bar evaporator source according to claim 1, characterized in that, The high-frequency induction heating coil (200) is made of hollow copper tube and has cooling water flowing inside; The high-frequency induction heating coil (200) is arranged as a tapered variable diameter coil, and the bottom inner diameter of the high-frequency induction heating coil (200) is larger than the top inner diameter of the high-frequency induction heating coil (200), and the top inner diameter of the high-frequency induction heating coil (200) is larger than the diameter of the evaporation source (500). The operating frequency of the high-frequency induction heating coil (200) is adaptively adjusted by the load through an impedance matching network; The number of turns of the spiral coil is set to 3-8 turns.

3. The closed-loop position control system for the bar evaporator source according to claim 2, characterized in that, The impedance matching network is an L-type matching network, including an adjustable capacitor and a fixed inductor. The adjustable capacitor has an adjustment range of 10-100pF, and the fixed inductor has an inductance of 2μH. The impedance matching network adopts an automatic tuning algorithm based on the principle of minimizing reflected power, and optimizes the matching network parameters through gradient descent, with a tuning time of less than 100ms.

4. The closed-loop position control system for the bar evaporator source according to claim 3, characterized in that, The gradient descent method optimizes the matching network parameters, including: Step 1: Set the initial adjustable capacitor value and measure the current reflected power; Step 2: Adjust the value of the adjustable capacitor in steps of 2pF, measure the adjusted reflected power, and calculate the rate of change of reflected power with respect to the capacitor value, dPr / dC. Step 3: Iteratively adjust the value of the adjustable capacitor along the direction of decreasing reflected power, and remeasure the reflected power after each iteration; Step 4: Repeat steps 2 and 3 until the reflected power is less than 0.1W, completing the automatic tuning of the impedance matching network.

5. The closed-loop position control system for the bar evaporator source according to claim 1, characterized in that, The evaporation source (500) is configured as an organic material wire rod; The organic material wire rod is made of at least one of C60 fullerene, Alq3 (8-hydroxyquinoline aluminum), NPB (N,N'-diphenyl-N,N'-(1-naphthyl)-1,1'-biphenyl-4,4'-diamine), and metal phthalocyanine.

6. The closed-loop control system for the position of the bar evaporator source according to claim 1, characterized in that, The closed-loop control algorithm includes: Step 1: Set the target position by setting the top target position of the evaporation source (500) below the center line of the high-frequency induction heating coil (200); Step 2: The current position of the top of the evaporation source (500) is detected in real time at a preset sampling frequency using a laser displacement sensor (300); Step 3: Subtract the target position from the current position to obtain the position deviation value. ; Step 4: Determine the deviation range and select the control mode: If the position deviation value If the absolute value is greater than 1.5mm, it enters the rapid approach mode and drives the evaporation source (500) to move towards the target position at the maximum speed V_max through the lifting mechanism (400); If the position deviation value If the absolute value is less than or equal to 1.5mm, then the incremental PID control mode is entered, and steps 5 to 7 are executed. Step 5: Calculate the incremental PID control output and calculate the control increment Δu according to the following formula: in, This is a proportionality coefficient, with a value range of [0.6, 0.8, 1.0]. s is the integral coefficient, with a value range of [0.03, 0.05, 0.07]. -1 ; s is the differential coefficient, with a value range of [0.08, 0.12, 0.16]. T is the control period, with a value ranging from 10 to 100 ms; This represents the current cycle position deviation. This represents the positional deviation from the previous cycle. This represents the positional deviation between the previous two cycles; Step 6: If the position deviation value The absolute value is less than the integral separation threshold. Then the control increment Integral terms Set to zero; Step 7: Transfer the speed command from the previous cycle With the control increment Add them together to get the current cycle speed command. and the speed command Amplitude limiting is applied; Step 8: Send the speed command The output is sent to the lifting mechanism (400) and drives the evaporation source (500) to move axially; Step 9: Adjust the current cycle deviation Store as ,Will Store as It records the current cycle speed command and then returns to step 2 to form a closed-loop control.

7. A suspension pressure equalization deposition system, characterized in that, include: The position closed-loop control system of the bar evaporator source according to any one of claims 1 to 6; A substrate support stage is set inside a vacuum chamber (100) and is used to place the substrate to be processed; A negative bias electrode is disposed in the vacuum cavity (100) and located between the high-frequency induction heating coil (200) and the processing substrate, for applying a DC negative high voltage to generate a deposition electric field; A floating equalizing plate is disposed below the substrate support stage, parallel to and insulated from the substrate support stage, and is connected to an adjustable DC bias power supply to keep it in a floating potential state. A double-layer insulating protective ring is arranged around the periphery of the suspended equalizing plate, including an inner ring and an outer ring, wherein the dielectric constant of the inner ring is smaller than that of the outer ring.

8. The suspension pressure equalization deposition system according to claim 7, characterized in that, The bias electrode includes an electrode body disposed between the evaporation source (500) and the processing substrate, the electrode body being opposite to and parallel to the processing substrate; The electrode body is flat or slightly arc-shaped, and the material of the electrode body is 316L stainless steel with a nickel plating layer on the surface.

9. The closed-loop position control and suspension pressure equalization deposition system for a bar evaporation source according to claim 7, characterized in that, The suspended pressure equalizing plate includes: The pressure equalization plate body is disposed below the substrate support stage, and the pressure equalization plate body is parallel to and insulated from the substrate support stage; The equalizing plate body is made of aluminum alloy 6061 and the surface is treated with hard anodizing.

10. The closed-loop position control and suspension pressure equalization deposition system for a bar evaporation source according to claim 7, characterized in that, The double-layer insulating protective ring includes: The inner ring and the outer ring are concentric ring structures and are arranged around the periphery of the suspended pressure equalizing plate; The inner ring is made of polytetrafluoroethylene; the outer ring is made of alumina ceramic; the dielectric constant ratio of the inner ring to the outer ring is 1:3 to 1:5.

Images