A method of constructing a magnetron sputtering system

Abstract

This invention discloses a method for constructing a magnetron sputtering system, relating to the field of vacuum coating equipment design technology. The method includes: S1, parametric modeling to obtain the spatial potential distribution and working gas ionization characteristic distribution within the cavity, calculating the argon ion trajectory to determine the sputtering area on the target surface, and calculating the secondary electron trajectory for damage assessment; S2, calculating target particle transport and static film thickness distribution based on the sputtering area; S3, calculating the dynamic film thickness distribution and coating uniformity of the carrier plate based on the movement mode of the transport mechanism, and obtaining the coating efficiency based on deposition time and average film thickness; S4, comparing the coating efficiency, coating uniformity, and secondary electron damage assessment results with preset design indicators, iteratively optimizing until the indicators are met, and outputting the optimized design parameters of the magnetron sputtering system. This method solves the modeling and calculation convergence problem under strong coupling conditions of multi-physics fields, significantly reducing the debugging cost and structural redundancy after equipment manufacturing.

Description

Technical Field

[0001] This invention relates to the field of vacuum coating equipment design technology, and in particular to a method for constructing a magnetron sputtering system. Background Technology

[0002] Magnetron sputtering is a thin film fabrication technology widely used in the manufacturing of semiconductors, displays, and optical devices. Its equipment design involves complex coupling of multiple physical fields such as electric field, magnetic field, plasma physics, particle transport, and thin film growth, and there is a highly nonlinear relationship between the design parameters and the final performance.

[0003] Existing magnetron sputtering system design methods, such as the patent with publication number CN115323335B, have the following drawbacks: First, the design process heavily relies on experience and trial-and-error with local parameters, lacking a systematic and quantitative simulation method for the complete physical chain of "electric field-magnetic field-plasma-particle transport-film growth," resulting in the inability to accurately predict key performance indicators such as coating uniformity and efficiency during the design phase. Second, for secondary electronic damage, a critical factor affecting equipment reliability, existing technologies lack effective non-invasive prediction methods, and its assessment often lags behind equipment manufacturing, leading to potential risks. Finally, due to the large size and high manufacturing cost of the equipment, performance verification relies entirely on "post-production testing" of physical prototypes. If the design fails to meet standards, the modification costs are extremely high and the cycle is extremely long, forcing designers to be conservative, resulting in structural redundancy and resource waste. Therefore, there is an urgent need for a method that can collaboratively predict and optimize the design of multiple key performance aspects of a magnetron sputtering system before manufacturing. Summary of the Invention

[0004] This invention proposes a method for constructing a magnetron sputtering system. Addressing specific technical challenges in magnetron sputtering equipment design, such as the difficulty in measuring secondary electronic damage, simulating dynamic operating conditions, and balancing multiple objectives, this invention constructs a system-level, quantitative, and iterative pre-design method. It solves the modeling and computational convergence problem under strong coupling conditions of multiple physics fields, and significantly reduces the debugging cost and structural redundancy after equipment manufacturing.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for constructing a magnetron sputtering system, comprising: S1, Parametric modeling, obtains the spatial potential distribution and working gas ionization characteristic distribution in the cavity, calculates the argon ion trajectory to determine the sputtering area on the target surface, and calculates the secondary electron trajectory for damage assessment. S2, calculate the target particle transport and static film thickness distribution based on the sputtering area; S3. Calculate the dynamic film thickness distribution and coating uniformity of the carrier plate according to the movement mode of the transmission mechanism, and obtain the coating efficiency based on the deposition time and average film thickness. S4 compares the coating efficiency, coating uniformity, and secondary electronic damage assessment results with the preset design indicators, iterates and optimizes until the indicators are met, and outputs the optimized design parameters of the magnetron sputtering system.

[0006] Preferably, in step S1, a parameterized model of the magnetron sputtering system is constructed. The model includes at least a three-dimensional model of the cavity, a three-dimensional model of the cathode, a cathode power supply type, a cathode structure type, anode position, grounding position, target parameters, substrate parameters, transport mechanism parameters, and vacuum system parameters, thereby achieving geometric parameterization.

[0007] Preferably, in step S1, the cathode power supply type includes one of radio frequency power supply, pulse power supply and DC power supply; the cathode structure type includes one of rotating target and planar target; the pumping system parameters include the pumping rate of the molecular pump; the transmission mechanism parameters include the carrier plate transport path and movement speed; the substrate parameters include the substrate material type and the adsorption energy of the sputtered material on the substrate material; and the vacuum system parameters include the pumping rate of the molecular pump.

[0008] Preferably, in step S2, calculating the trajectory of argon ions includes: setting up an argon ion particle swarm in the argon ionization region, with the initial velocity of the particles randomly distributed according to Maxwell's law, and the number of particles per velocity direction being no less than 40; and determining the collision position with the target surface by tracking the trajectory of argon ions under the action of electric and magnetic fields.

[0009] Preferably, in step S2, calculating the trajectory of secondary electrons includes: based on the characteristics of electrons and argon ions being generated in pairs during argon ionization, setting up an electron particle swarm in the electron generation region, calculating the trajectory and energy change of secondary electrons under the combined action of electric and magnetic fields, and assessing the risk of damage to sensitive components on the substrate, device surface, or cavity by secondary electrons.

[0010] Preferably, in step S3, the calculation of the film thickness distribution of the carrier plate in motion adopts a static equivalent processing method of discrete sampling, that is, the state of the carrier plate at different positions is calculated separately, and then a weighted average is performed in combination with the actual residence time of the carrier plate at each position to obtain the deposition distribution in motion.

[0011] Preferably, in step S4, the coating uniformity, coating efficiency, and secondary electronic damage are optimized together, and at least one of the following is adjusted iteratively: target-substrate distance, cathode power supply parameters, magnetic field strength, transmission mechanism motion parameters, and anode position.

[0012] Preferably, in step S4, when the coating uniformity is not up to standard and the thickest film is at the center of the substrate, it is determined that the target-substrate distance is too large, and an optimization adjustment to reduce the target-substrate distance is performed; when the coating uniformity is not up to standard and the thickest film is at both sides of the substrate, it is determined that the target-substrate distance is too low, and an optimization adjustment to increase the target-substrate distance is performed.

[0013] Preferably, the optimization adjustment of the cathode power supply parameters includes: adjusting the duty cycle of the pulse power supply, changing the power supply bias value, or switching the power supply type; the optimization adjustment of the magnetic field strength includes: adjusting the magnetic flux density or changing the magnetic field distribution pattern; the optimization adjustment of the motion parameters of the transmission mechanism includes: adjusting the carrier plate movement speed, acceleration, or changing the movement trajectory path.

[0014] Preferably, step S4 includes: using a gradient scanning strategy, traversing multiple target-substrate distance values ​​within a reasonable range with a set step size, calculating the coating uniformity corresponding to each target-substrate distance, and selecting the target-substrate distance that meets the preset index and has the best overall performance as the final design parameter; the reasonable range is predetermined based on the cavity size, cathode structure, and process window.

[0015] The beneficial effects of this invention are as follows: By constructing a parameterized model and sequentially executing multiphysics coupling simulations, this invention can quantitatively predict coating uniformity and coating efficiency before the manufacture of magnetron sputtering equipment, and simultaneously assess the risk of secondary electronic damage, thereby bringing the performance evaluation and optimization process forward. This method establishes a clear mapping relationship from design parameters to performance indicators, making the optimization process data-driven, significantly reducing the number of physical trial-and-error attempts, shortening the design cycle, reducing costs, and contributing to achieving better overall equipment performance. Attached Figure Description

[0016] Figure 1 This is a general flowchart of the magnetron sputtering system design method of Embodiment 1 of the present invention.

[0017] Figure 2 This is a schematic diagram of the three-dimensional magnetron sputtering process cavity model in Example 1.

[0018] Figure 3 This is a diagram showing the spatial potential distribution and electron energy distribution in Example 1, where... Figure 3 (A) represents the spatial potential distribution along the cathode-substrate-auxiliary anode direction. Figure 3 (B) represents the characteristics of electron energy distribution.

[0019] Figure 4 Ar in Example 1 + Ion movement trajectory and sputtering area distribution on target surface.

[0020] Figure 5 This is a diagram of the trajectory of the secondary electron in Example 1.

[0021] Figure 6 Ar in the sputtering area of ​​Example 1 + Particle collision kinetic energy distribution and incident angle distribution diagram.

[0022] Figure 7 This is a distribution diagram of target particle deposition in Example 1, where... Figure 7 (A) shows the three-dimensional deposition distribution within the chamber. Figure 7 (B) shows the two-dimensional film thickness distribution at the substrate plane.

[0023] Figure 8 This is a comparison chart of film thickness distribution and optimization results under different target-substrate distances. Figure 8 (A) shows the film thickness distribution when the target-substrate distance is 120 mm. Figure 8 (B) shows the film thickness distribution when the target-substrate distance is 100 mm. Figure 8 (C) shows the film thickness distribution when the target-substrate distance is 80 mm. Figure 8 (D) is the gradient scan curve of coating uniformity in the target-substrate distance range of 80-120 mm. Detailed Implementation

[0024] Example 1

[0025] This embodiment provides a method for constructing a magnetron sputtering system, applicable to various magnetron sputtering equipment, and refers to... Figure 1 This method mainly includes the following steps.

[0026] Step S1: Parametric modeling is performed to obtain the spatial potential distribution and working gas ionization characteristic distribution within the cavity. The argon ion trajectory is calculated to determine the sputtering area on the target surface, and the secondary electron trajectory is calculated for damage assessment.

[0027] Constructing a parameterized model of the magnetron sputtering system includes the following sub-steps.

[0028] Step S1.1, Construction of the three-dimensional model of the cavity.

[0029] Establish a three-dimensional geometric model of the process cavity, such as Figure 2 As shown. The internal dimensions of the process chamber are: length 1720 mm, width 1070 mm, height 300 mm, and wall thickness 30 mm. The chamber material is 304 stainless steel, and the corresponding electrical conductivity and magnetic permeability parameters are set in the model. Two vacuum pump ports are provided on the bottom surface of the chamber, each with a diameter of 200 mm, for connecting to the vacuum pumping system.

[0030] Step S1.2, construction of the three-dimensional model of the cathode.

[0031] Two rotating cathodes are arranged symmetrically along the length of the cavity. Each rotating cathode has a diameter of 100 mm and a length of 600 mm, with a center-to-center distance of 400 mm between the two cathodes. An array of magnets is installed inside the rotating cathodes to form a closed magnetic field on the target surface, with a magnetic flux density of 0.05 T at the rotating cathodes. The rotational speed of the rotating cathodes is set to 5 rpm to ensure uniform utilization of the target material.

[0032] Step S1.3, Electrode and power supply parameter settings.

[0033] The rotating cathode is connected to the negative electrode, and the target power supply uses a pulsed DC power supply with a negative bias of -350 V, a duty cycle of 80%, and a frequency of 40 kHz. The substrate is grounded, forming an electric field distribution between the cathode potential and the ground potential. An auxiliary anode is placed between the two rotating cathodes. This auxiliary anode is grounded and is used to adsorb negative ions and secondary electrons during the magnetron sputtering process, reducing damage to the substrate.

[0034] Step S1.4, setting the parameters of the target and substrate.

[0035] The target material used is ITO (indium tin oxide), with a density of 7.14 g / cm³. 3 The target material, with a molecular weight of 462.34 g / mol, is attached to the surface of the rotating cathode. A silicon substrate, measuring 156 mm × 156 mm and with a thickness of 200 μm, is placed on a transport belt above the cathode. The adsorption energy of ITO on the silicon substrate surface is 5 eV, which is used to subsequently calculate the deposition stability of the target particles on the substrate surface.

[0036] Step S1.5, setting transmission mechanism parameters.

[0037] The distance between the substrate and the rotating cathode is 120 mm, i.e., the initial target-substrate distance is 120 mm. The substrate begins its initial movement from a stationary state, starting at a speed of 0.2 m / s² from a distance of -0.8 m from the cavity center. 2 The substrate is accelerated to 0.1 m / s², with an acceleration stroke of 25 mm; subsequently, it moves at a constant speed of 0.1 m / s² to a position 0.775 m from the center of the cavity; then it moves at -0.2 m / s². 2 The acceleration decelerates and stops 0.8 m from the center of the cavity; after stopping, the above motion process is repeated in the opposite direction, thus forming a reciprocating motion trajectory. The single complete stroke of the carrier plate is 1600 mm, and the single motion cycle is 32.5 s.

[0038] Step S1.6, Vacuum system parameter settings.

[0039] Two vacuum pump ports located on the bottom surface of the cavity are each connected to a molecular pump. The pumping speed of a single molecular pump is 1800 L / s, and the rated vacuum degree is 5 × 10⁻⁶. -5 Pa. The process gas is argon (Ar), with a flow rate set at 100 sccm and a working pressure maintained at 0.3 Pa.

[0040] Step S1.7: Model mesh generation and boundary condition setting.

[0041] The parametric model described above was discretized, and the cavity space was divided using an unstructured tetrahedral mesh. The mesh size was refined to 2 mm on the cathode surface and near the substrate, and widened to 10 mm on the cavity wall away from the process area, with a total mesh size of approximately 2.8 million. Boundary conditions were set as follows: the outer wall of the cavity was grounded, the vacuum pump port was set as the evacuation boundary, and the process gas inlet was set as the mass flow inlet.

[0042] Through the above parametric modeling steps, a complete digital twin model including geometric structure, electromagnetic properties, kinematic parameters, and material properties was established, providing a foundation for subsequent calculation of spatial potential distribution, particle trajectory tracking, and film thickness distribution prediction.

[0043] Step S1.8: Calculate the spatial potential distribution and ionization characteristic distribution.

[0044] Based on the model, the spatial electric potential distribution is calculated in three-dimensional space, such as... Figure 3 As shown in (A), the potential decreases along the gradient from cathode to substrate and from cathode to auxiliary anode. The electron energy distribution is as follows: Figure 3 As shown in (B), electrons are controlled between the auxiliary anode and the cathode, which is consistent with the purpose of the auxiliary anode design: to control negative ions from colliding with the substrate.

[0045] Step S1.9: Calculate the trajectory of Ar+ ions and the sputtering area.

[0046] Based on the potential distribution and electron energy, the Ar ionization region is located above and in the middle of the cathode. This is because Ar ionization produces one Ar+ ion and one electron, and the region where the electron is produced is the ionization region. Therefore, an Ar+ particle swarm is placed at this location, with the initial velocities following a Maxwell random distribution. To ensure accuracy, the number of particles per velocity direction in space is no less than 40. The calculated motion of Ar+ is as follows... Figure 4 As shown, Ar+ ions accelerate towards the target under the attraction of the cathode, while the central ions move outward under the repulsion of the auxiliary anode. After 100,000 sets of particle motion simulations, the collision region on the target was obtained; this is the etching region. To ensure uniform utilization of the target material, the rotating target rotates at 5 revolutions per minute.

[0047] Furthermore, the trajectory of electrons in space was calculated to assess the damage caused by secondary electrons from magnetron sputtering, and the results are as follows: Figure 5 As shown, as time increases, the secondary electrons ionized in space move towards the auxiliary anode and are attracted by the auxiliary anode, reducing sputtering damage.

[0048] Step S2: Calculate the target particle transport and static film thickness distribution based on the sputtering area.

[0049] Based on the sputtering area obtained in step S1 and the kinetic energy distribution and incident angle distribution of the incident particles at each location, the sputtering and transport process of the target particles is calculated, specifically including the following sub-steps.

[0050] Step S2.1, Calculation of target particle sputtering parameters.

[0051] Based on the Ar values ​​at each location within the sputtering area as statistically analyzed in step S1.3 + The calculated results of the kinetic energy distribution and incident angle distribution of the ions are as follows: Figure 6 As shown, the sputtering parameters of the target particles are calculated by combining the momentum conservation relationship.

[0052] Specifically, for each discrete location on the circumference of the sputtering region, according to Ar + The collision kinetic energy and incident direction of the particles determine the sputtering rate of the target particles. + The statistical range of particle collision kinetic energy is 300 electron volts to 450 electron volts. According to the principle of momentum transfer, Ar... + After a particle collides with the target surface, some of its momentum is transferred to the target atoms or molecules, giving them enough energy to escape from the target surface. The sputtering rate of the target particles is related to the Ar... + The incident velocity, incident direction, and Ar of the particles + The mass ratio of the incident ions to the target particles is relevant; the lighter incident ions transfer more momentum to the heavier target particles, but the transfer efficiency is affected by the mass difference between the two.

[0053] The sputtered target particles move in a direction symmetrical to the normal of the incident direction; that is, relative to the normal to the target surface, the reflection angle is equal to the incident angle. The calculation of the incident angle is based on Ar... + The ratio of the particle velocity component along the target axis to the component along the target radial direction is determined, where the axial component is the velocity component along the length of the cathode and the radial component is the velocity component pointing towards the substrate.

[0054] Step S2.2: Determine the sputtering position and initial motion direction.

[0055] Based on the sputtering region distribution obtained in step S1.3, the spatial location of sputtering is determined on the target surface of the dual rotating cathode. Since the rotating cathode rotates at 5 rpm, the circumferential direction is divided into 360 discrete angles, and the axial direction is divided at 10 mm intervals, establishing a two-dimensional coordinate system for the sputtering location. For each sputtering location, the Ar value at that location is... + The incident angle distribution is a statistical distribution that assigns the initial motion direction of the target particles, with the normal direction accounting for 60%, the deviation from the normal within ±30° accounting for 35%, and the remaining angles accounting for 5%.

[0056] Step S2.3, Calculation of target particle transport path.

[0057] The transport path of the target particles in the cavity space was calculated by combining the particle's own parameters (ITO atomic mass, initial kinetic energy). Considering the gas scattering effect at a working pressure of 0.3 Pa, the Monte Carlo method was used to track the motion of the target particles from the sputtering position to the substrate plane. The initial kinetic energy of the target particles was set to be in the range of 10-50 eV (based on Ar). + (Statistics on collision energy and momentum transfer efficiency) During the transport process, collisions occur with background argon atoms. After the collision, the orientation is randomized, and the energy is lost according to elastic collision loss.

[0058] For each ITO particle sputtered from the target, its trajectory is tracked until: (1) it is deposited on the substrate surface; (2) it is deposited on the cavity wall or other structural surface; (3) it returns to the target surface (self-sputtering); (4) it is pumped away by the vacuum pump. The probability distribution of each outcome is statistically analyzed, and particles deposited on the substrate surface are counted as effective deposits.

[0059] Step S2.4, Calculation of static film thickness distribution on the substrate plane.

[0060] A two-dimensional grid with a resolution of 5 mm × 5 mm was established on the plane containing the substrate (120 mm above the center of the cathode), covering an area consistent with the cavity width (1070 mm × 1070 mm). For each target particle sputtered from the dual rotating cathodes, if it ultimately deposited on this plane, the deposition location was recorded and the deposition mass was accumulated. Considering the superposition effect of the two cathodes, the film thickness distribution acting independently of each cathode was calculated, and then linearly superimposed to obtain the overall static film thickness distribution.

[0061] The static film thickness distribution refers to the deposition distribution when the carrier plate is stationary at a fixed position. Based on the calculated velocity and direction of the target material at different positions on the target surface after collision, the deposition distribution of target atoms within the cavity can be obtained, such as... Figure 7 As shown in (A); the particle deposition at the substrate plane, as follows Figure 7 As shown in (B).

[0062] The film thickness is higher near the two targets, and gradually decreases along the direction away from the center of the targets.

[0063] Through the above steps, the spatial distribution characteristics of film thickness under static conditions of the carrier plate were obtained, providing basic data for subsequent dynamic working condition calculations and coating uniformity evaluation.

[0064] Step S3: Calculate the dynamic film thickness distribution and coating uniformity of the carrier plate according to the movement mode of the transmission mechanism, and obtain the coating efficiency based on the deposition time and average film thickness.

[0065] To achieve uniform film thickness, the carrier plate will undergo reciprocating motion. The motion in this case is as follows: The substrate begins to move from a stationary state in the initial stage, starting at a speed of 0.2 m / s² from a distance of -0.8 m from the center of the cavity. 2 The substrate is accelerated to 0.1 m / s², with an acceleration stroke of 25 mm; subsequently, it moves at a constant speed of 0.1 m / s² to a position 0.775 m from the center of the cavity; then it moves at -0.2 m / s². 2 The acceleration decelerates and stops 0.8 m from the center of the cavity; after stopping, the above motion process is repeated in the opposite direction, thus forming a reciprocating motion trajectory.

[0066] This calculation method employs a static equivalent processing approach with discrete sampling for the motion process. Specifically, it calculates the state of the carrier plate at different positions and then performs a weighted average based on the actual residence time of the carrier plate at each position to obtain the deposition distribution under motion conditions. The calculated coating distribution at a target-substrate distance of 120 mm is shown below. Figure 8 As shown in Figure A, the coating uniformity is calculated to be 2.39%, which is greater than the coating uniformity target of 2%.

[0067] Based on the actual working conditions and the movement mode of the transmission mechanism, the film thickness distribution, coating uniformity and coating efficiency of the carrier plate in motion are calculated, which includes the following sub-steps.

[0068] Step S3.1: Discretize the motion trajectory of the carrier plate.

[0069] Based on the carrier plate motion parameters set in step S1.5, the reciprocating variable speed motion process is discretized and sampled. The 1600 mm single complete stroke of the carrier plate is divided into 160 discrete position points, with an interval of 10 mm between adjacent position points. The instantaneous velocity and dwell time of the carrier plate at each position point are calculated according to the kinematic equations: In the acceleration phase, the velocity increases linearly from 0 to 0.1 m / s in the range of 0-25 mm, and the dwell time decreases as the velocity increases; in the constant velocity phase, the velocity remains at 0.1 m / s in the range of 25 mm-1575 mm, and the dwell time is 0.1 s for each 10 mm stroke; in the deceleration phase, the velocity decreases linearly from 0.1 m / s to 0 in the range of 1575 mm-1600 mm, and the dwell time increases as the velocity decreases. The carrier plate stops for 0.5 s at each of the two turning points and then moves in the opposite direction.

[0070] Step S3.2, calculate the static film thickness distribution at each discrete location.

[0071] For each discrete position defined in step S3.1, the substrate is translated to that position, and the calculation process of step S2 is repeated to obtain the film thickness distribution when the carrier plate is stationary at that position. Since the carrier plate moves along the length of the cavity and the dual rotating cathodes are symmetrically arranged in the width direction of the cavity, the main difference between each discrete position is reflected in the change of the relative distance between the substrate and the two cathodes. When the substrate is located at the center of the cavity, the distance to the two cathodes is equal, and the film thickness distribution shows a symmetrical bimodal characteristic; when the substrate is shifted to one side, the film thickness near the cathode on that side increases, and the film thickness away from the cathode on the other side decreases.

[0072] Step S3.3, weighted average calculation of dynamic film thickness distribution.

[0073] A static equivalent processing method using discrete sampling is employed, weighting the static film thickness distribution at each discrete location according to the actual residence time to obtain the comprehensive film thickness distribution under motion conditions. Specifically, for each spatial location on the substrate surface, the amount of deposition received during the entire motion cycle is the sum of the products of the film thickness at each discrete location and the corresponding residence time, divided by the total motion cycle time. This equivalent processing is based on the quasi-steady-state assumption, that is, the carrier plate's motion speed is much smaller than the target particle transport speed, and the discharge state and particle transport process within the cavity tend to stabilize within the time it takes for the carrier plate to move one discrete distance.

[0074] Calculations showed that the dynamic film thickness distribution under a target-substrate distance of 120 mm was approximately 78 nm in the central region and approximately 62 nm in the edge region, exhibiting a slightly higher thickness at the center and decreasing towards both sides. Compared to the static distribution, the motion process smoothed out the bimodal structure caused by the dual cathodes to some extent, but the central region still maintained a relatively high film thickness due to being covered by both cathodes.

[0075] Step S3.4, Calculation of coating uniformity.

[0076] Based on the dynamic film thickness distribution obtained in step S3.3, the coating uniformity is calculated. Eighty-one uniformly distributed sampling points are selected within the effective coating area of ​​the substrate, and the film thickness value at each point is extracted. The average film thickness and standard deviation of film thickness are then calculated. Coating uniformity is defined as the percentage of the standard deviation of film thickness to the average film thickness. The calculation shows that at a target-substrate distance of 120 mm, the average film thickness is 70 nm, the standard deviation of film thickness is 1.67 nm, and the coating uniformity is 2.39%, which is greater than the preset design target of less than 2%.

[0077] Step S3.5, Calculate the coating efficiency.

[0078] The deposition rate is calculated based on deposition time and average film thickness. A single reciprocating motion cycle is 32.5 s, with an effective deposition time of 32 s (minus the 1.5 s stop time at both ends). With an average film thickness of 70 nm, the deposition rate is 2.15 nm / s. Based on the target thickness requirements (e.g., 200 nm), the theoretical deposition time per substrate is calculated to be 93 s. Combining the effective loading area of ​​the carrier plate and the actual production cycle, the deposition efficiency under the current design parameters is derived to be 38 substrates per hour.

[0079] Step S3.6, design indicator compliance judgment.

[0080] The coating uniformity of 2.39% and coating efficiency obtained in steps S3.4 and S3.5 are compared with the preset design targets. The coating uniformity does not meet the requirement of less than 2%, while the coating efficiency is within the acceptable range. Based on the film thickness distribution characteristics, the thickest position is located at the center of the substrate, which is a typical characteristic of a large target-substrate distance. Therefore, the process proceeds to step S4 to perform parameter optimization.

[0081] Step S4: Compare the coating efficiency, coating uniformity and secondary electronic damage assessment results with the preset design indicators, iterate and optimize until the indicators are met, and output the optimized design parameters of the magnetron sputtering system.

[0082] The thickest film was determined to be located at the center of the substrate, indicating a large target-substrate distance. Therefore, the target-substrate distance was reduced. The calculated film distribution at a target-substrate distance of 100 mm is as follows: Figure 8 As shown in Figure B, the calculated coating uniformity is 1.41%, which meets the requirement of coating uniformity being less than 2%. Output the device parameters, film thickness distribution, and calculation results for coating uniformity.

[0083] When the coating uniformity is substandard, and the thickest film is located on both sides of the substrate, this indicates that the target-substrate distance is too low. Figure 8 As shown in C, the target-substrate distance is 80 mm, and the coating uniformity is 2.27%.

[0084] The target-substrate distance should be within a reasonable range; too large or too small is not good. The coating uniformity results when the target-substrate distance is 80-120 mm are as follows: Figure 8 As shown in D. Example 2

[0085] This embodiment provides a method for constructing a magnetron sputtering system. In step S4 of this embodiment, a gradient scanning strategy is adopted to traverse multiple target-substrate distance values ​​within a reasonable range with a set step size, calculate the coating uniformity corresponding to each target-substrate distance, and select the target-substrate distance that meets the preset index and has the best overall performance as the final design parameter; the reasonable range is predetermined based on the cavity size, cathode structure, and process window.

[0086] Based on the coating uniformity and coating efficiency obtained in step S3 and the secondary electronic damage assessment results obtained in step S1, the results are compared with the preset design indicators. The target parameters are adjusted through iterative optimization until all design indicators are met. The specific steps include the following sub-steps.

[0087] Step S4.1, preset design indicators.

[0088] The collaborative optimization objectives of this embodiment are set as follows: coating uniformity less than 2%, coating efficiency not less than 35 substrates per hour, and secondary electron damage index less than 0.3 (characterized by the normalized value of high-energy secondary electron flux received per unit area). Current calculation results: coating uniformity 2.39% (not meeting the target), coating efficiency 38 substrates / hour (meeting the target), secondary electron damage index 0.15 (meeting the target, the auxiliary anode effectively adsorbs secondary electrons). The main optimization direction is to reduce the coating uniformity value.

[0089] Step S4.2, the target-base distance optimization strategy is determined.

[0090] Based on the film thickness distribution characteristics identified in step S3.4, the thickest film is located in the central region of the substrate, indicating a large target-substrate distance. When the target-substrate distance is large, the overlap of the sputtering regions of the two cathodes on the substrate plane increases, and the central region receives superimposed deposition from both cathodes, resulting in a higher central film thickness than the edge. Therefore, an optimization adjustment to reduce the target-substrate distance is performed to decrease the overlap of the sputtering regions of the two cathodes, making the film thickness distribution more even.

[0091] Step S4.3, gradient scan optimization execution.

[0092] The optimal target-substrate distance was determined using a gradient scanning strategy. Within a reasonable range of 80 mm to 120 mm, five scanning points (120 mm, 110 mm, 100 mm, 90 mm, and 80 mm) were set with a step size of 10 mm. The calculation process of steps S1 to S3 was repeated for each target-substrate distance to obtain the coating uniformity, coating efficiency, and secondary electronic damage assessment results under each target-substrate distance condition. The reasonable range was predetermined based on the cavity height limit, cathode dark area thickness, and electron mean free path. The lower limit of 80 mm ensured that the cathode sheath did not contact the substrate surface, and the upper limit of 120 mm was the initial design value.

[0093] The calculation results for each scanning point are as follows: Target-substrate distance 120 mm: Coating uniformity 2.39%, with the thickest film located at the center of the substrate; Target-substrate distance 110 mm: Coating uniformity 1.89%, with the film thickness distribution tending to be flat; Target-substrate distance 100 mm: Coating uniformity 1.41%, with the best film thickness distribution uniformity; Target-substrate distance 90 mm: Coating uniformity 1.78%, with the thickest film shifting to both sides; Target-substrate distance 80 mm: Coating uniformity 2.27%, with the thickest film located on both sides of the substrate, exhibiting a saddle-shaped distribution with high edges and low center.

[0094] Step S4.4, Optimal parameter selection and verification.

[0095] Based on the gradient scanning results, the coating uniformity is 1.41% when the target-substrate distance is 100 mm, meeting the design target of less than 2%. The coating efficiency is 39 substrates per hour (slightly improved compared to 120 mm, due to the increased deposition rate after the target-substrate distance decreases), and the secondary electron damage index is 0.18 (still below the limit of 0.3, indicating that the auxiliary anode is still effectively working). Considering these three indicators, a target-substrate distance of 100 mm is the optimal design parameter.

[0096] To further pinpoint the optimal value, a finer scan was performed in the 90 mm to 110 mm range with a step size of 5 mm, confirming that 100 mm remained the optimal solution within this subdivision range. The uniformity was 1.52% at a target-substrate distance of 105 mm and 1.55% at a target-substrate distance of 95 mm, both worse than the results at 100 mm.

[0097] Step S4.5: Optimize the output of design parameters.

[0098] The iterative calculation is terminated, and the optimized design parameters and corresponding performance evaluation indicators of the magnetron sputtering system are output: Optimized key design parameters: Target-base distance: 100 mm (optimized from the initial 120 mm); Cathode power supply parameters: pulsed DC power supply, negative bias -350 V, duty cycle 80%, frequency 40 kHz (remains constant); Rotating cathode speed: 5 rpm (remains constant); Motion parameters of the transmission mechanism: acceleration 0.2 m / s² 2 Maximum speed 0.1 m / s, stroke 1600 mm (unchanged); Auxiliary anode position: middle of the dual cathodes, ground potential (remains unchanged); Corresponding performance evaluation metrics: Coating uniformity: 1.41% (meets the requirement of less than 2%); Coating efficiency: 39 pieces / hour (meets the requirement of not less than 35 pieces / hour); Secondary electronic damage index: 0.18 (meets the requirement of being below 0.3); Average film thickness: 82 nm (within a single cycle).

[0099] Step S4.6, design results sensitivity analysis.

[0100] Sensitivity analysis was performed on the optimized design parameters to assess the impact of fluctuations in key parameters on performance indicators: when the target-substrate distance deviation was ±5 mm, the coating uniformity variation ranged from 1.41% to 1.78%, which was still within the target range; when the cathode power supply duty cycle deviation was ±5%, the coating efficiency variation ranged from 37 to 41 pieces / hour, and the uniformity variation was less than 0.1%; when the transmission speed deviation was ±0.02 m / s, the impact on uniformity was negligible, and the efficiency variation was within ±3 pieces / hour.

[0101] The analysis results show that the optimized design parameters have a good process window tolerance and are suitable for actual production environments.

[0102] Through the above iterative optimization process, the collaborative design of the magnetron sputtering system for three key indicators—coating uniformity, coating efficiency, and secondary electronic damage—was realized, and the optimal combination of design parameters that meets all preset indicators was output.

Claims

1. A method for constructing a magnetron sputtering system, characterized in that, Includes the following steps: S1, Parametric modeling, obtains the spatial potential distribution and working gas ionization characteristic distribution in the cavity, calculates the argon ion trajectory to determine the sputtering area on the target surface, and calculates the secondary electron trajectory for damage assessment. S2, calculate the target particle transport and static film thickness distribution based on the sputtering area; S3. Calculate the dynamic film thickness distribution and coating uniformity of the carrier plate according to the movement mode of the transmission mechanism, and obtain the coating efficiency based on the deposition time and average film thickness. S4 compares the coating efficiency, coating uniformity, and secondary electronic damage assessment results with the preset design indicators, iterates and optimizes until the indicators are met, and outputs the optimized design parameters of the magnetron sputtering system.

2. The method for constructing a magnetron sputtering system according to claim 1, characterized in that, In step S1, a parameterized model of the magnetron sputtering system is constructed. The model includes at least a three-dimensional model of the cavity, a three-dimensional model of the cathode, the cathode power supply type, the cathode structure type, the anode position, the grounding position, the target parameters, the substrate parameters, the transport mechanism parameters, and the vacuum system parameters.

3. The method for constructing a magnetron sputtering system according to claim 2, characterized in that, In step S1, the cathode power supply type includes one of radio frequency power supply, pulse power supply and DC power supply; the cathode structure type includes one of rotating target and planar target; the pumping system parameters include the pumping rate of molecular pump; the transmission mechanism parameters include the carrier plate transport path and movement speed; the substrate parameters include the substrate material type and the adsorption energy of sputtered material on the substrate material; and the vacuum system parameters include the pumping rate of molecular pump.

4. The method for constructing a magnetron sputtering system according to claim 1, characterized in that, In step S2, calculating the trajectory of argon ions includes: setting up an argon ion particle swarm in the argon ionization region, with the initial velocity of the particles randomly distributed according to Maxwell's law, and the number of particles per velocity direction being no less than 40. By tracking the trajectory of argon ions under the action of electric and magnetic fields, the collision position with the target surface is determined.

5. The method for constructing a magnetron sputtering system according to claim 1, characterized in that, In step S2, calculating the trajectory of secondary electrons includes: based on the characteristics of electrons and argon ions being generated in pairs during argon ionization, setting up an electron particle swarm in the electron generation region, calculating the trajectory and energy change of secondary electrons under the combined action of electric and magnetic fields, and assessing the risk of damage to sensitive components on the substrate, device surface, or cavity by secondary electrons.

6. The method for constructing a magnetron sputtering system according to claim 1, characterized in that, In step S3, the calculation of the film thickness distribution of the carrier plate in motion adopts a static equivalent processing method of discrete sampling, that is, the state of the carrier plate at different positions is calculated separately, and then a weighted average is performed by combining the actual residence time of the carrier plate at each position to obtain the deposition distribution in motion.

7. The method for constructing a magnetron sputtering system according to claim 1, characterized in that, In step S4, with coating uniformity, coating efficiency, and secondary electronic damage as the synergistic optimization objectives, the process is iterative by adjusting at least one of the following: target-substrate distance, cathode power supply parameters, magnetic field strength, motion parameters of the transmission mechanism, and anode position.

8. The method for constructing a magnetron sputtering system according to claim 7, characterized in that, In step S4, when the coating uniformity is not up to standard and the thickest film is at the center of the substrate, it is determined that the target-substrate distance is too large, and the optimization adjustment of reducing the target-substrate distance is performed; when the coating uniformity is not up to standard and the thickest film is at both sides of the substrate, it is determined that the target-substrate distance is too low, and the optimization adjustment of increasing the target-substrate distance is performed.

9. A method for constructing a magnetron sputtering system according to claim 7, characterized in that, The optimization and adjustment of the cathode power supply parameters include: adjusting the duty cycle of the pulse power supply, changing the power supply bias value, or switching the power supply type; the optimization and adjustment of the magnetic field strength includes: adjusting the magnetic flux density or changing the magnetic field distribution pattern; the optimization and adjustment of the motion parameters of the transmission mechanism includes: adjusting the carrier plate's motion speed, acceleration, or changing the motion trajectory path.

10. A method for constructing a magnetron sputtering system according to claim 1 or 8, characterized in that, Step S4 includes: using a gradient scanning strategy, traversing multiple target-substrate distance values ​​within a reasonable range with a set step size, calculating the coating uniformity corresponding to each target-substrate distance, and selecting the target-substrate distance that meets the preset index and has the best overall performance as the final design parameter; the reasonable range is predetermined based on the cavity size, cathode structure and process window.

Images