Substrate preparation method and application based on low-temperature plasma modification and one-step dry film

Abstract

The invention belongs to the technical field of FPC (flexible printed circuit) substrate preparation, and discloses a substrate preparation method and application based on low-temperature plasma modification and one-step dry film. The preparation method of the base material comprises the steps of FPC base material pretreatment, PI / PET base material cutting, low-temperature plasma modification and drying. Preparing a TSF dry film, mixing the elastic resin, the photosensitizer and the monomer, and adjusting the viscosity by the solvent; carrying out dry film lamination, hot-pressing lamination and cooling shaping; photoetching and forming; and reliability testing and FPC finished product cutting are carried out. The invention provides a series of TSF dry film products with extreme flexibility, ultrahigh reliability, excellent processability and high suitability of FPC (Flexible Printed Circuit) process, so as to meet the strict requirements of folding display, automotive electronics, advanced packaging and other high-end FPC application scenes.

Description

Technical Field

[0001] This invention belongs to the field of FPC substrate preparation technology, and particularly relates to a substrate preparation method and application based on low-temperature plasma modification and one-step dry film. Background Technology

[0002] The flexible, thin, and three-dimensional assembly characteristics of FPCs place special requirements on the solder mask layer: Limitations of traditional screen-printed solder mask: After repeated dynamic bending, the coating cured from the liquid ink may develop microcracks, blistering, or peel off from the copper surface due to poor CTE (coefficient of thermal expansion) matching with the substrate or internal stress issues, especially in bending areas. TSF dry film solution: Dedicated TSF dry film, through formulation design, possesses higher elastic modulus, better ductility, and better mechanical compatibility with flexible substrates (such as PI), thus providing more reliable protection during repeated bending and significantly extending product lifespan. This is its core value in dynamically bending FPCs.

[0003] Limitations of traditional screen-printed solder mask: In chip-scale packaging FPCs and ultra-fine pitch (e.g., <40μm) connector areas, it is difficult to guarantee the thickness uniformity and pattern accuracy of extremely fine solder mask bridges, which can easily lead to short circuits or insufficient strength. TSF dry film solution: The uniform thickness of dry film is an inherent advantage (e.g., 20±2μm), enabling the formation of robust and precise fine solder mask bridges, providing reliable insulation for ultra-high density wiring. At the same time, ultra-thin TSF dry films (e.g., below 15μm) can meet the requirements of ultra-thin FPCs with extremely stringent thickness requirements.

[0004] Limitations of traditional screen-printed solder mask: In applications requiring multiple lead-free reflow soldering cycles, high-temperature aging, or exposure to high-temperature and high-humidity environments (such as near automotive engine compartments or outdoor equipment), ordinary solder mask may discolor, degrade, and experience decreased adhesion. TSF dry film solution: TSF dry film can withstand more reflow soldering cycles (>5) while maintaining color and performance stability, providing long-term chemical resistance, moisture resistance, and insulation reliability, meeting automotive-grade (AEC-Q) or industrial-grade certification requirements.

[0005] The demand for solder resist photosensitive dry film directly depends on the production volume and application expansion of FPC. Current main application areas and targets include: The trend towards thinner and foldable consumer electronics: Smartphones (especially foldable phones), tablets, and wearable devices (TWS earphones, smartwatches) have extremely high requirements for internal space utilization, leading to a continuous increase in the use of FPC (Flexible Printed Circuit). The hinge of foldable phones requires multiple high-precision FPCs, representing a significant growth point for high-end dry film.

[0006] Automotive Electrification and Intelligentization: The rise of new energy vehicles (electrification, intelligent driving, and smart cockpits) has led to a surge in the use of FPCs (Flexible Printed Circuits) in automobiles, applied to battery management systems (BMS), in-vehicle displays, sensors, LED headlights, and more. The extremely high reliability and durability requirements of automotive electronics are driving the demand for high-performance dry films.

[0007] Product demand trends: High resolution to meet the needs of fine-line, high-density interconnect FPC manufacturing. High flexibility to adapt to repeated bending and dynamic use scenarios of FPC, requiring no cracking and strong adhesion after dry film curing.

[0008] High reliability, heat resistance, chemical resistance (electrolyte and sweat resistance), and high insulation, especially suitable for automotive and military applications.

[0009] Low dielectric constant and low loss factor: suitable for high-frequency and high-speed FPCs.

[0010] In summary, the technical defects of the existing technology are as follows: when using a vacuum laminator, the mesh and circuit areas are not fully pressed and have air bubbles after lamination; the pad openings are misaligned and have poor uniformity after exposure; and there is residue due to overdevelopment. Summary of the Invention

[0011] To overcome the problems existing in related technologies, the present invention discloses an embodiment of a substrate preparation method and its application based on low-temperature plasma modification and one-step dry film.

[0012] The technical solution is as follows: a method for preparing a substrate based on low-temperature plasma modification and one-step dry film, the method comprising: S1, FPC substrate pretreatment, PI / PET substrate cutting, low-temperature plasma modification, drying; S2, TSF dry film preparation: elastic resin, photosensitizer, monomer mixture, and solvent viscosity adjustment; S3, dry film lamination, hot pressing, cooling and shaping; S4, photolithography, exposure, development, and thermal curing; S5, reliability testing, bending cycle test, high temperature and high humidity test, circuit accuracy test; S6, FPC finished product cutting, appearance inspection, packaging and shipping.

[0013] In step S1, the low-temperature plasma modification includes: S101, surface activation is performed using low-voltage pulsed radio frequency plasma with a power of 50-200W; S102, micro-hole etching is performed using a DBD plasma of an oxygen / argon mixed gas with a gas ratio of O2:Ar=1:9; S103 utilizes a nitrogen-hydrogen plasma gradient treatment method with a pressure gradient of 0.1-1 Pa for interface strengthening.

[0014] In step S2, the optimized formula for the elastic resin ratio is obtained through resin crosslinking degree simulation. Specific steps include: S201, Molecular structure modeling and cross-linking network design, including: by introducing a reversible molecular spring structure with flexible segments and dynamic response stress at cross-linking points, based on target properties including elasticity and tear strength, a molecular dynamics model of elastic resin is constructed by using dynamic cross-linking gradient, interfacial energy control parameters, and hyperelastic model coefficients to simulate the distribution of cross-linking points and three-dimensional network structure. S202, Multi-scale simulation and parameter optimization; employing layered simulation: Macroscopic scale, predicting the overall mechanical behavior of materials, including tensile strength and elongation at break; At the microscale, the effect of local crosslinking density on performance was analyzed using Representational Volume Units (RVEs). Combined with Direct FE 2 Multi-scale coupling technology enables collaborative simulation of macroscopic and microscopic deformation; S203, Crosslinking Degree-Performance Mapping Database Construction; Predicting the Crosslinking Degree of Resin / Functional Nanoparticles under Different Formulations through Quantum Chemical Calculations and Correlating it with Key Performance Indicators such as Tear Strength and Bending Resistance; The optimized formulation scheme for elastic resin includes: (1) A single-component elastic resin formulation using FlexOne series resins, which replaces the traditional two-component system by optimizing the molecular chain structure; (2) Synergistic ratio of resin and nanofiller; add functional nanoparticles to enhance interfacial bonding through in-situ polymerization. The addition ratio is: the mass fraction of carbon nanotubes (CNTs) in the resin matrix is ​​usually 0.1%-2.0%.

[0015] In step S201, the reversible molecular spring structure includes: A reversible three-dimensional network structure is formed by flexible segments between cross-linking points. At the microscale, the flexible segments connect the cross-linking points through dynamic bonds. When subjected to force, the segments extend and disperse the stress, and return to their original state after unloading. Dynamic bond selection: Reversible covalent bonds or supramolecular interactions are used to ensure that crosslinking points can dynamically respond to stress.

[0016] The steps of multi-parameter collaborative modeling include: (1) Dynamic crosslinking gradient design; crosslinking point distribution optimization, through molecular dynamics MD simulation, to regulate the spatial distribution of crosslinking agents; (2) Integration of interface energy regulation parameters; modeling of interface interaction, introducing interface energy parameters, and optimizing the interface between carbon nanotube filler and matrix; (3) Calibration of hyperelastic model coefficients; embedding constitutive equations to map hyperelastic model coefficients to MD simulation: elastic coefficients, obtained by uniaxial tensile simulation to obtain stress-strain curves and fit model parameters; viscoelastic parameters, combined with the generalized Maxwell / Kelvin model to describe frequency-dependent response; time-temperature equivalence: using the WLF equation to correlate modulus changes under different temperature / frequency domains to predict long-term service performance. The process of building a molecular dynamics model includes: (i) Initial structure generation: a mixture of resin monomer and curing agent was constructed using Materials Studio / PackMol, and flexible segments were introduced as spring units. (ii) Dynamic cross-linking reaction simulation: set cross-linking degree target, optimize network topology through energy minimization and annealing cycle; monitor path length between cross-linking points and Voronoi volume to evaluate network looseness; (iii) Performance verification and optimization, including: mechanical response simulation, dynamic performance verification, and reversibility verification.

[0017] The hot-pressing bonding in the dry film bonding process in step S3 includes: S301, multi-stage pressure curve control; adopts segmented pressurization instead of traditional constant pressure, divided into three stages: pre-pressurization, main pressure, and pressure holding. S302, dynamic temperature compensation bonding; an infrared temperature measurement module is integrated into the hot press head to monitor the substrate temperature in real time and adjust the heating power accordingly. S303, vacuum lamination bonding for local reinforcement.

[0018] In step S301, the pre-compression stage is characterized by a pressure of 0.1-0.2 MPa, a time of 3-5 seconds, and a temperature of 80°C. Main pressure stage: Pressure rises to 0.8-1.2 MPa, temperature rises to 100-110℃, duration 8-12 seconds; Pressure holding stage: The pressure drops to 0.3-0.5 MPa, and the temperature is maintained for 5-8 seconds.

[0019] In step S302, the temperature fluctuation of the substrate is monitored in real time and controlled within ±2℃; Step S303, local reinforcement in vacuum lamination, includes: pre-bonding the cover film and the substrate in the vacuum chamber; adding FR4 / PI reinforcing plates to high-stress areas such as connectors; maintaining the vacuum level at 0.05-0.1 MPa during hot pressing and curing; and using a vacuum lamination temperature of 170-190℃ and a pressure of 0.8-1.2 MPa.

[0020] In step S4, a CCD vision positioning system is introduced, and LED surface light source power is controlled by partitioning. In step S4, during development, the concentration of the developer, temperature, spray pressure, and development time parameters are optimized by evaluating the ultrathin dry film.

[0021] In step S5, during the line accuracy test, no residue was found in the 30μm line.

[0022] Another objective of this invention is to provide an application of the substrate preparation method based on low-temperature plasma modification and one-step dry film in the preparation of TSF dry film products.

[0023] Combining all the above technical solutions, the beneficial effects of this invention are as follows: This invention provides a series of TSF dry film products with extreme flexibility, ultra-high reliability, excellent processability, and high adaptability to FPC processes, to meet the stringent requirements of high-end FPC application scenarios such as foldable displays, automotive electronics, and advanced packaging. This invention's dry film photosensitive cover film replaces existing cover film and photosensitive ink processes, eliminating the entire cover film process; this invention proposes TSF application process technology; this invention analyzes the reliability capabilities of TSF materials applied to FPC products. Attached Figure Description

[0024] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure; Figure 1 This is a flowchart of a substrate preparation method based on low-temperature plasma modification and one-step dry film provided in an embodiment of the present invention. Detailed Implementation

[0025] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0026] Example 1, such as Figure 1 As shown, the substrate preparation method based on low-temperature plasma modification and one-step dry film provided in this embodiment of the invention includes: S1, FPC substrate pretreatment, PI / PET substrate cutting, low-temperature plasma modification, drying; S2, TSF dry film preparation: elastic resin, photosensitizer, monomer mixture, and solvent viscosity adjustment; S3, dry film lamination, hot pressing, cooling and shaping, shaping temperature is 25℃; S4, photolithography, exposure (ultraviolet light), development, thermal curing; S5, reliability testing, bending cycle test, high temperature and high humidity test, line accuracy test (no residue in 30μm line); S6, FPC finished product cutting, appearance inspection (no cracks, warping), packaging and shipping.

[0027] As can be seen from the above embodiments, the present invention adopts a one-step dry film covering process to replace the traditional covering film (material cutting-drilling-punching-lamination-pressing-curing) + photosensitive ink process, saving 6 processes, improving production efficiency and reducing energy consumption; This invention develops an integrated process of low-temperature plasma modification and precise hot-pressing bonding, which improves the bonding strength of FPC substrate surface, reduces board warping rate, and ensures mass production stability.

[0028] For example, in step S1, compared with traditional wet chemical treatment or thermal plasma technology, the core innovative steps of low-temperature plasma modification in the pretreatment of FPC substrates are as follows: S101. Surface activation is performed using low-pressure pulsed radio frequency plasma with a power of 50-200W; traditional atmospheric pressure plasma treatment has a power of >500W. This invention avoids thermal deformation of the polyimide substrate, precisely controls the surface energy (up to 70mN / m), and reduces carbon fiber damage; the electron temperature (5-10eV) and ion temperature (≈300K) are separated, and the selective reaction of free radicals reduces the thermal effect. S102 utilizes a DBD plasma of an oxygen / argon mixed gas with a gas ratio of O2:Ar=1:9 for micropore etching; the traditional technique is chemical etching solution (strong acid / alkali); this invention achieves submicron-level aperture control (0.2-0.5μm), eliminates waste liquid pollution, and enhances the adhesion of copper foil; high-energy electrons (>8eV) excite oxygen free radicals for selective oxidation, with argon ion physical sputtering acting synergistically; S103 utilizes a nitrogen-hydrogen plasma gradient treatment with a pressure gradient of 0.1-1 Pa to strengthen the interface, whereas traditional techniques involve chemical copper plating pretreatment. This invention generates a copper nitride transition layer in situ, improving dry film adhesion by 20% and reducing electroplating voids. Plasma induces surface chemical bond reconstruction, and the low-pressure gradient enhances the penetration of active particles.

[0029] Specifically, the principle basis of the process parameters of this invention is as follows: pulse radio frequency power control; adjusting the electron energy distribution by pulse duty cycle (10%-30%) to maintain electron temperature (T e ≥5eV while suppressing ion temperature (T) i≤50℃ to avoid thermal damage to the FPC substrate. The non-thermal equilibrium characteristics of cold plasma (T...) e / T i >100).

[0030] Gas composition optimization; O2 / Ar mixing ratio: Argon ion sputtering removes surface impurities (physical action), oxygen free radicals achieve controllable oxidation etching (chemical action), and a ratio of 1:9 achieves a balance between etching rate (50nm / min) and precision.

[0031] Gradient pressure design; 0.1-1Pa gradient change: The low pressure zone (0.1Pa) enhances the ability of high-energy particles to penetrate micropores, while the high pressure zone (1Pa) promotes the uniform diffusion of active species, achieving deep / breadth synergistic modification.

[0032] The advantages of the improved process parameters compared to existing technologies are shown in Table 1.

[0033] Table 1 Comparison of Technological Advantages For example, in step S2, the optimized ratio of the elastic resin is obtained through resin crosslinking degree simulation. The specific steps for obtaining the optimized ratio of the elastic resin through resin crosslinking degree simulation include: S201, Molecular structure modeling and crosslinking network design, includes: introducing a "reversible molecular spring structure" with flexible segments and dynamic response stress at crosslinking points; based on target properties (such as elasticity and tear strength), utilizing dynamic crosslinking gradients, interfacial energy control parameters, and hyperelasticity model coefficients to collaboratively construct a molecular dynamics model of the elastic resin, simulating the distribution of crosslinking points and the three-dimensional network structure. This invention introduces a "reversible molecular spring structure" (flexible segments and dynamic response stress at crosslinking points) to replace the traditional static crosslinking model.

[0034] Among these features, the dynamic crosslinking gradient and segmented curing experiments optimize the uniformity of the crosslinking network by testing the degree of curing under different temperature / time combinations using DSC. This invention avoids stress concentration and improves bending resistance (up to 1 million cycles). Interfacial energy control parameters, contact angle measurement, and molecular dynamics simulation optimize the compatibility of the resin with fillers / reinforcing phases and reduce interfacial defects. This invention improves tear resistance (>50kN / m, twice that of traditional resins). Hyperelasticity model coefficients and cyclic compression / tension tests fit the Yeoh hyperelasticity model parameters (C1, C2, C3) to verify resilience. This invention ensures that the material suffers no permanent damage under repeated deformation (250% elongation at break).

[0035] For example, the modeling basis for reversible molecular spring structures includes: Molecular design principle: Referencing the "molecular spring" structure design of iSUN3D elastic resin, its core is the formation of a reversible three-dimensional network structure through flexible segments between cross-linking points. At the microscale, flexible segments (such as polyurethane chains) connect cross-linking points through dynamic bonds (hydrogen bonds, coordination bonds, etc.). When subjected to force, the segments extend and disperse stress, and return to their original shape after unloading.

[0036] Dynamic bond selection: Reversible covalent bonds (such as Diels-Alder bonds) or supramolecular interactions (such as metal coordination and hydrogen bonding) are employed to ensure that crosslinking points can dynamically respond to stress. For example, existing technologies achieve the performance switching between plastics and elastomers by controlling the crosslinking density: low crosslinking density gives chain segments freedom of movement, forming a spring-like rebound.

[0037] Exemplary key steps in multi-parameter collaborative modeling; (1) Dynamic crosslinking gradient design; crosslinking point distribution optimization: The spatial distribution of crosslinking agents is controlled through molecular dynamics (MD) simulation (e.g., gradient increasing structure). For example: high crosslinking region: improves tear strength (e.g., iSUN3D resin tear strength > 50kN / m); low crosslinking region: increases the proportion of flexible segments to achieve high elasticity (bending resistance > 1 million times). Algorithm implementation: Perl / Python scripts are used to dynamically control the crosslinking reaction, combined with distance and angle criteria (e.g., CO bond distance < 2.5Å and included angle > 120°) to avoid non-physical configurations.

[0038] (2) Integration of interface energy control parameters; modeling of interface interactions: introducing interface energy parameters (such as van der Waals forces and electrostatic potential) to optimize the interface between filler and matrix (such as carbon nanotubes / epoxy resin): strong interface: improves stress transfer efficiency (refer to polyurethane-epoxy resin braided network); weak interface: promotes molecular chain slippage and dissipates energy (such as hydrogen bond recombination of helical peptide hydrogels). Force field selection: using PCFF or ReaxFF force fields to accurately describe the energy changes in the dynamic bond breaking / recombination process.

[0039] (3) Hyperelastic model coefficient calibration; Constitutive equation embedding: Mapping the coefficients of hyperelastic models (such as Mooney-Rivlin, Ogden) to MD simulation: Elastic coefficients: Obtaining stress-strain curves through uniaxial tensile simulation and fitting model parameters; Viscoelastic parameters: Combining the generalized Maxwell / Kelvin model to describe the frequency-dependent response. Time-temperature equivalence principle: Using the WLF equation to correlate modulus changes under different temperature / frequency domains to predict long-term service performance.

[0040] An example is the molecular dynamics model construction process; (i) Initial structure generation: A mixed system of resin monomers (such as DGEBA) and curing agents (such as DETDA) is constructed using Materials Studio / PackMol; flexible segments (such as polydimethylsiloxane) are introduced as “spring units”.

[0041] (ii) Dynamic cross-linking reaction simulation: Set a cross-linking degree target (e.g., 80%), optimize the network topology through energy minimization and annealing cycle; monitor the path length between cross-linking points and Voronoi volume, and evaluate the network looseness (existing techniques can be used).

[0042] (iii) Performance Verification and Optimization: Mechanical Response: Simulate tear testing (ASTM D624 standard) and cyclic loading to analyze stress distribution and energy dissipation (existing technologies can be used); Dynamic Performance: Calculate mean square displacement (MSD) and free volume fraction (existing technologies can be used) to verify segment mobility; Reversibility Verification: Track dynamic bond recombination rate (e.g., hydrogen bond recovery rate >90%, existing technologies can be used). For example, the multi-component network integration of this invention: Referring to the asymmetric entanglement strategy of Huang Feihe's team, rigid segments (e.g., epoxy resin) and flexible segments (e.g., polyurethane) are woven together to achieve rigid-elastic synergy. Machine Learning-Assisted Optimization: Using the GA-ANN model, with target performance (tear strength, resilience) as the optimization objective, the crosslinking gradient and interfacial energy parameters are screened in reverse. For example, experimental verification and application; performance benchmarking, see Table 2; Table 2 Performance Benchmarking Industrial applications: Combined with HSL photopolymerization 3D printing, it realizes the entire process of "design-printing-molding" and is used for flexible products such as integrated products.

[0043] This invention constructs a "reversible molecular spring" resin model that combines high elasticity and high strength by dynamically controlling the network topology through crosslinking gradient, optimizing stress transfer through interface energy parameters, and calibrating the macroscopic response using a hyperelastic model.

[0044] S202, multi-scale simulation and parameter optimization; employing hierarchical simulation: macroscopic scale: predicting the overall mechanical behavior of materials (e.g., tensile strength, elongation at break); microscopic scale: analyzing the impact of local crosslinking density on performance through representative volume elements (RVE). Combined with Direct FE. 2 Multi-scale coupling technology enables collaborative simulation of macroscopic and microscopic deformation.

[0045] For example, Direct FE 2 The core steps of multi-scale coupling technology include: (a) Scale model construction and coupling mechanism; Macroscopic model establishment: Define the geometry and boundary conditions of the overall composite material structure (such as laminated plates) and generate macroscopic finite element meshes.

[0046] Microscopic RVE embedding: Microscopic representative volumetric units (RVEs) are embedded at each integration point of the macroscopic model. The RVEs must contain the microstructure of the material (such as functional nanoparticles or fibers and elastoplastic matrix).

[0047] Degree of freedom coupling: The degrees of freedom of macroscopic nodes and microscopic RVE boundary nodes are directly connected through linear multi-point constraints (MPCs), realizing the real-time transmission of displacement / strain information and avoiding traditional nested iteration.

[0048] (b) Material constitutive model and numerical algorithm implementation; Microscopic component modeling: Different components in microscopic RVE (such as functional nanoparticles, fibers, or matrix) need to have their own material models defined (e.g., the matrix adopts an elastoplastic constitutive model).

[0049] Nonlinear solution algorithm: A numerical algorithm based on the radial reflection method to derive the plastic constitutive model of the missile is used to calculate the uniform tangent stiffness matrix and ensure the convergence efficiency of the Newton-Raphson iterative method.

[0050] For example, the Radial Return Mapping Algorithm is the most mainstream implicit stress update strategy for solving elastoplastic constitutive equations. Its physical essence is to "pull back" the predicted elastic stress along the normal (i.e., radial) direction of the yield surface in the deviatoric stress space to the yield surface, ensuring that plastic deformation satisfies the consistency condition and volume invariance. This method is widely used for modeling metals, composite materials, and geotechnical materials under J2 flow theory (such as the Mises yield criterion) and has become the standard paradigm for UMAT subroutines in commercial software such as ABAQUS.

[0051] For example, the core content of the algorithm; This numerical algorithm typically includes the following four progressive steps, forming a closed-loop iterative process: Elastic prediction step: Assuming there is no plastic deformation in the current increment step, the trial stress and trial strain are calculated directly according to Hooke's law; Yield criterion step: Substitute the yield function (e.g., f(σ)) trial If f(q)=0, determine whether the plastic phase has been entered; if f>0, then a callback is required. The plasticity parameter solution step involves solving for the plasticity multiplier Δλ using the Newton-Raphson iteration. Its incremental form is derived from the consistency condition f=0; for example, for the Mises model, Δλ=f(σ...). trial ,q) / ( f / σ:C ep : f / σ; Stress / state variable update step: Correct stress, plastic strain, and hardening variables (such as equivalent plastic strain ε) using Δλ. p Back stress α), and calculate the uniform tangent stiffness matrix C. ep This ensures that the Newton iteration converges twice.

[0052] To improve computational stability, fully implicit schemes (such as the backward Euler method) are often used in practical implementations to avoid the phenomenon of predicted stress drift yield surface. Meanwhile, for singular points (such as the Mohr-Coulomb corner), the literature proposes principal stress spatial backtracking or smoothing processing. The core of the numerical algorithm based on the radial backtracking method is a four-step closed loop of "elastic prediction - yield discrimination - plastic parameter iteration - state update," and its success depends on the selection of the yield function, the strict implementation of the consistency condition, and the accurate derivation of the consistent tangent stiffness matrix. Currently, this algorithm has been maturely applied in multiple fields such as composite materials, metal forming, geotechnical engineering, and frozen soil dynamics, and has been deployed in engineering applications on platforms such as ABAQUS through UMAT subroutines.

[0053] User subroutine development: Write user-defined material subroutines such as UMAT / VUMAT and integrate them into the finite element platform (such as ABAQUS) to achieve cross-scale calculations.

[0054] (c) Parallel solution and scale synchronization; single analytical framework solution: macroscopic and microscopic models are calculated in parallel within the same finite element analysis, and stress and strain data are exchanged in real time, unlike traditional FE. 2 Separation and iteration. Adaptive scale transfer: Dynamically adjust inter-scale information through upscaling (microscopic response → macroscopic stiffness) and downscaling (macroscopic strain → microscopic boundary conditions) operators.

[0055] (d) Validation and Model Optimization; Multi-scale Model Validation: The accuracy of the multi-scale method is validated by comparing the prediction results (such as stress-strain curves) of a single macroscopic element, a multi-directional laminate, and a full-scale model. Computational Efficiency Optimization: GPU acceleration and adaptive mesh refinement techniques are employed to balance computational accuracy and resource consumption.

[0056] For example, the technical functions and advantages of the above-mentioned technical features include: High-precision cross-scale prediction simultaneously captures macroscopic mechanical responses (such as the stress-strain curve of laminates) and microscopic evolution mechanisms (such as matrix plastic deformation and fiber stress concentration), providing a basis for material design.

[0057] Example: The simulation results of T300 / 7901 composite materials are in excellent agreement with experimental data, and the evolution of the micro-stress field can guide fatigue-resistant design. Computational efficiency is significantly improved, and direct coupling avoids the problems associated with traditional FE (Features, Materials, and Materials). 2 The iterative convergence process reduces computation time; parallel solution frameworks (such as ABAQUS's built-in parallelism) accelerate the analysis of large-scale problems. Traditional nested FE 2 It requires repeated calls to microscopic calculations, while DirectFE 2 Achieve single-step solution through MPC. It possesses capabilities for handling complex boundaries and nonlinearities, capable of processing strongly nonlinear behaviors (such as plastic deformation and damage evolution) and complex geometric boundaries (such as fluid-structure interactions in microfluidic chips). Furthermore, it can incorporate machine learning to optimize scale transfer parameters, improving the simulation accuracy of multi-physics couplings (such as thermo-mechanical-electrical).

[0058] S203, Crosslinking Degree-Performance Mapping Database Construction: The crosslinking degree under different formulations (resin / functional nanoparticles) is predicted by quantum chemical calculations and correlated with key performance indicators (tear strength, flexural strength).

[0059] For example, the optimized formulation of elastic resin and its technical effects include: (1) Single-component elastic resin formulation; using FlexOne series resins, replacing the traditional two-component system by optimizing the molecular chain structure (such as introducing flexible polyurethane segments). It is ready to use immediately, avoiding mixing errors of two-component systems; residual resin can be recycled, reducing waste.

[0060] (2) Synergistic formulation of resin and nanofiller; addition of functional nanoparticles (such as carbon nanotubes) to enhance interfacial bonding through in-situ polymerization. At the same time, it enhances mechanical strength and toughness, avoiding the decrease in modulus caused by traditional toughening.

[0061] For example, typical addition ratios; the mass fraction of carbon nanotubes (CNTs) in the resin matrix is ​​typically 0.1%-2.0%, exceeding this range can easily lead to agglomeration and performance degradation. In epoxy resin / CNTs systems, common ratios are 0.3%-1.5% (e.g., 0.3wt%, 0.5wt%, 1.0wt%, 1.5wt%). Addition ratios by resin type are shown in Table 3. Table 3 shows the addition ratio by resin type. Key influencing factors; 1) Nanoparticle dispersibility: When the concentration is above 2.0%, CNTs tend to agglomerate and form stress concentration points, reducing interfacial bonding. Surface modification (such as carboxylation or silane coupling agent treatment) is needed to improve dispersibility, allowing for higher addition levels (such as 1.5%-2.0%).

[0062] 2) Target performance requirements; Conductivity: 0.5%-1.0% can reach the percolation threshold (e.g., 0.7wt% in epoxy resin achieves 10). - 3 S / m electrical conductivity). Mechanical reinforcement: 1.0%-1.5% significantly improves tensile strength / modulus (e.g., epoxy resin + 1.0wt% CNTs increases strength by 40%).

[0063] In-situ polymerization process optimization; pre-dispersion: CNTs are first ultrasonically treated in monomers (e.g., with acetone solvent assistance) before polymerization. Gradient addition: Nanoparticles are added in batches to avoid excessively high local concentrations. Initial experimental ratio: Tests are started at 0.5 wt%, and the concentration is increased in increments of 0.3% to 1.5%, monitoring the dispersion state and performance changes. Critical value control: In most systems, the viscosity increases sharply after exceeding 1.5 wt%, requiring adjustment of process parameters. Necessity of surface modification: If the addition amount is ≥1.0%, CNTs must be functionalized (e.g., -COOH, -NH2). (3) Optimize the photocuring process parameters; match the 385-405nm wavelength light source, control the photoinitiator anchoring concentration and exposure dose, and improve curing efficiency.

[0064] An exemplary experimental verification method for the optimization scheme.

[0065] (1) Mechanical performance verification; ASTM standard test: D624: tear strength >50kN / m; D412: tensile strength 20MPa, elongation at break 250%; bending fatigue test: bending resistance >1 million times (traditional resin only 10,000-50,000 times).

[0066] (2) Microstructure verification; FTIR / XRD analysis: confirming the formation of cross-linked networks and the absence of phase separation; AFM characterization: observing the uniformity of cross-linking at the nanoscale. Key breakthroughs compared to existing technologies; Innovative process: Multi-stage curing process: segmented curing achieves molecular-level cross-linking control, replacing single-stage curing; Cross-scale design: full-process integration from molecular spring structure (microscopic) to integrated supply chain closed loop (macroscopic). This solution significantly shortens the R&D cycle through simulation-driven experiments, promoting the application of elastic resins in high-end manufacturing.

[0067] For example, step S3, the hot pressing bonding in dry film bonding includes: S301, multi-stage pressure curve control; implementation steps: segmented pressurization replaces the traditional constant pressure, divided into three stages: pre-pressurization (low-pressure contact), main pressure (high-pressure penetration), and holding pressure (stabilized curing). Improved parameters: Pre-pressurization stage: pressure 0.1-0.2MPa, time 3-5 seconds, temperature 80℃; Main pressure stage: pressure increases to 0.8-1.2MPa, temperature increases to 100-110℃, duration 8-12 seconds; Holding pressure stage: pressure decreases to 0.3-0.5MPa, temperature is maintained, time 5-8 seconds. It can be seen that the pre-pressurization stage removes air from the substrate surface, reducing bubbles; the main pressure stage ensures sufficient filling of the micropores between the dry film and the copper foil; the holding pressure stage cures interfacial stress, reducing the risk of delamination.

[0068] S302, Dynamic Temperature Compensated Lamination; Implementation Steps: An infrared temperature measurement module is integrated into the hot press head to monitor the substrate temperature in real time and adjust the heating power accordingly. Improved Parameters: Temperature fluctuation is controlled within ±2℃ (compared to ±5℃ in traditional processes), and the target temperature is dynamically adjusted according to the substrate thickness (e.g., 100℃ for thin boards, 120℃ for thick boards). This avoids localized overheating that could lead to dry film runoff or aging, and improves linewidth accuracy (up to ±5μm).

[0069] S303, vacuum lamination for localized reinforcement; pre-bonding of the cover film and substrate within a vacuum chamber; localized addition of FR4 / PI reinforcing plates to high-stress areas such as connectors; maintaining a vacuum level of 0.05-0.1 MPa during hot-pressing curing. Improved parameters: vacuum lamination temperature 170-190℃, pressure 0.8-1.2 MPa. It is evident that eliminating interlayer bubbles (bubble rate <0.1%) improves the peel strength of the reinforced area by 30%. The effects of this invention through the above technical features are shown in Table 4.

[0070] Table 4 Effects of the Invention As can be seen, this invention solves the bottleneck of micro-line processing by achieving uniform development of pads smaller than 50μm through precise temperature / pressure control, avoiding resist layer peeling caused by the "pond effect." It improves compatibility with dissimilar materials, with multi-stage pressure curves adapting to materials with large differences in thermal expansion coefficients, such as ceramic substrates and flexible PI, resulting in a peel strength ≥30N / cm after bonding. It reduces energy consumption and defect rate; dynamic temperature compensation reduces heat loss by 30%, and vacuum lamination reduces the bubble defect rate from 5% to 0.5%. For example, in step S4 exposure, a CCD vision positioning system (positioning accuracy ±5μm) is introduced, and LED surface light source power zoning control (error <5%) is adopted; this solves the problems of pad window misalignment and poor uniformity after exposure.

[0071] For example, the CCD vision positioning system integrates: High-precision dynamic compensation: Utilizing a 50-megapixel global shutter CCD camera (120fps frame rate) combined with a telecentric lens (distortion rate <0.01%), it achieves closed-loop control through "image acquisition - intelligent recognition - dynamic compensation". Sub-pixel-level positioning: Employing the Canny edge detection algorithm and multi-scale edge fusion technology, the detection accuracy reaches ±0.5μm, supporting automatic correction for ±1mm incoming material deviation. Real-time coordinate transformation: Employing a nine-point calibration method (repeatability ±0.3μm) and nonlinear distortion compensation (radial / tangential distortion correction) to achieve precise mapping between physical coordinates and pixel coordinates.

[0072] For example, the LED surface light source zoning control includes: Multispectral composite light source design: Configuring an adjustable red / blue / infrared light combination light source, supporting intelligent switching between coaxial light and ring light, and eliminating interference from solder pad mirror reflections. Independent power control for each zone: Dividing the light source into 9-12 independent temperature control zones, with a power error of <5% for each zone, and adjusting the current in real time through feedback from an illuminance sensor. Intelligent illuminance equalization algorithm: Based on a pre-scanned grayscale distribution map, automatically calculating the compensation coefficient for each zone to ensure that the light intensity uniformity of the exposure area is >98%.

[0073] Among them, the compensation coefficients for each zone are automatically calculated based on the pre-scanned grayscale distribution map, including: First, pre-scanning and grayscale distribution map generation; 1.1) Image Acquisition and Partitioning: The target area is scanned globally using a high-precision sensor (such as CMOS / CCD) to acquire the original grayscale image. The image is then divided into a dynamic grid (such as N×M partitions), with the partition size adjusted according to the required detection accuracy (e.g., 16×16 or 32×32 grids are commonly used in industrial circle detection).

[0074] 1.2) Gray-level feature extraction; calculate the average gray value, standard deviation, and gray-level co-occurrence matrix (GLCM) for each partition, and analyze texture features (such as contrast and energy). Use Local Binary Pattern (LBP) or Histogram of Oriented Gradients (HOG) to enhance the recognition of local illumination features.

[0075] Second, calculation of the zonal compensation coefficient; 2.1) Setting the reference illuminance, using the average gray level of the entire area or a preset ideal value (such as 128 gray levels) as the reference illuminance target value. Calculate the gray level deviation rate of each zone: Deviation rate = (average gray level of zone - reference value) / reference value; 2.2) Compensation coefficient generation; Linear mapping method: directly calculate the compensation gain based on the deviation rate (e.g., gain > 1 for low grayscale areas, gain < 1 for high grayscale areas). Adaptive weighting method: combine partition weights (e.g., higher weight for edge areas) and optimize the compensation coefficient using weighted least squares. Introduce dynamic threshold adjustment to avoid overcompensation (e.g., limit the gain range to 0.8-1.2).

[0076] Third, it achieves light intensity regulation and uniformity; performs real-time dimming; drives adjustable optical devices (such as LCD filters or LED arrays) to dynamically adjust the output intensity of the light source according to the zoning coefficient. It uses multi-channel PWM control to achieve millisecond-level zoning illuminance compensation.

[0077] Fourth, feedback verification and iteration; after compensation, perform a second scan and calculate the uniformity index: Uniformity = (1 - (maximum gray value - minimum gray value) / average gray value) × 100%; If the uniformity is less than 98%, the compensation coefficient is iteratively optimized using the gradient descent method. As can be seen from the above embodiments, the key points of process parameter optimization used to solve the problems of pad window misalignment and poor uniformity after exposure are shown in Table 5.

[0078] Table 5 Key Points for Optimizing Process Parameters For example, key processes for exposure include: (1) Pre-calibration and template learning; Automatic calibration process: Select the pad area by ROI, create a shape template (supports edge erasure optimization), and generate distortion correction file by combining with the nine-point calibration board. Multi-template management: Store calibration parameters for different pad models, support one-click switching, and reduce the changeover time to within 30 seconds.

[0079] For example, the distortion correction file (e.g., .descr) generated by the nine-point calibration board includes: the physical dimensions of the calibration board, the arrangement of feature points, the ratio of the center-to-diameter distance, the output path, and the visualization file (.ps). It does not directly contain distortion parameters, which need to be calculated in reverse after the acquired image is calibrated.

[0080] Nine-point calibration is a key technology in industrial vision for establishing the mapping between image coordinates and physical world coordinates, commonly used in image stitching, precision measurement, and other scenarios. It requires the use of a high-precision calibration board (such as a checkerboard or circular array), and "generating the calibration board file" is only the first step—it defines the mathematical description of the calibration board, providing a benchmark for subsequent image acquisition and parameter solving. The actual distortion correction file (such as CameraParameters in HALCON) needs to be generated after multi-angle image acquisition, feature point extraction, and nonlinear optimization are completed.

[0081] Taking HALCON gen_caltab as an example; according to the search results, when generating a .descr calibration board description file, the following five types of information must be specified: calibration board geometry: number of rows × number of columns (e.g., 7×7), actual spacing between circles / squares (unit: meters); feature point morphology parameters: the ratio of the radius of the circle to the distance between the center points (controls the relative position of the circle size, affecting detection robustness); file path and naming: specify the save path of the .descr description file and the accompanying PostScript visualization file (.ps); material and contrast settings: such as 'light_on_dark' indicates a bright circle on a dark background, ensuring that feature points in the image can be stably extracted; (implicit) coordinate system origin convention: HALCON defaults to using the first circle in the upper left corner as the world coordinate system origin, affecting subsequent pose calculations. It can be seen that the .descr file itself does not contain distortion coefficients (k1 / k2 / p1 / p2, etc.) or camera intrinsics (f, c). x c y These need to be calculated by the HALCON calibration assistant or the calibrate_cameras operator after acquiring ≥9 calibration images in different poses.

[0082] The generation process and key parameter comparison table; the following is the correspondence between each parameter in a typical gen_caltab function call and the content of the output file, see Table 6; Table 6 Correspondence Once this file is generated, it only completes the "modeling preparation"; the following steps are still required: ① Take at least 9 calibration images covering the entire field of view (at different tilts / rotations / distances); ② Load the .descr file and import the image in HALCON Calibration Assistant; ③ Extract marker points → Calculate errors → Generate CameraParam and CameraPose with distortion parameters.

[0083] As can be seen, the .descr file is essentially a "calibration board ID card," only describing its appearance and size; the actual distortion correction capability comes from the camera parameters calculated by subsequent image calibration. If image acquisition is skipped and the .descr file is used directly, no distortion correction effect can be obtained. The procedure is as follows: first generate the .descr file → then take 9+ high-quality images → finally run the calibration process. All corrections rely on the standard camera calibration output (such as the CameraParameters array from HALCON).

[0084] (2) Real-time closed-loop control process; including: graph TD A [CCD continuous image acquisition] --> B (intelligent recognition of pad feature points) B-->C {Position deviation > 5μm?} C--Yes-->D[Dynamic Compensation Mechanism Offset] C -- No --> E [Trigger Zoned Light Source Exposure] E-->F [Light intensity data fed back by the illuminance sensor] F --> G [Adjust LED zone power] G-->H [Exposure completed and quality inspection performed]; Automatic recalibration is triggered when positioning fails to avoid production line downtime.

[0085] (3) Post-verification and data traceability: The MES system records the exposure coordinate deviation and light uniformity data for each batch to achieve full-process quality traceability. Offline programming plugins (such as AutoCAD / CorelDRAW) are used to pre-generate exposure paths to reduce manual intervention.

[0086] For example, the effectiveness is verified as follows: Offset control: Pad opening position deviation decreased from ±20μm to within ±5μm, and rework rate decreased from 15% to below 1.5%. Uniformity improvement: Exposure energy fluctuation <3%, and edge clarity of micro QR codes (0.5mm×0.5mm) reaches 98%. A single device replaces three traditional models, reducing changeover time by 70%. The above solution integrates high-response visual positioning, adaptive light source control, and digital process management, and is suitable for precision electronic manufacturing scenarios such as PCB and MiniLED.

[0087] For example, in step S4, during development, by evaluating the ultrathin dry film, the concentration of the developer, temperature, spray pressure, development time, and other parameters are optimized, thus solving the problems of over-development and residue.

[0088] Example 2: This embodiment of the invention provides an application of a substrate preparation method based on low-temperature plasma modification and one-step dry film in the preparation of FPC finished products. The prepared FPC finished products can be used in smartphones, tablets, wearable devices, battery management systems (BMS), automotive displays, sensors, and LED vehicle lights.

[0089] The substrate preparation method based on low-temperature plasma modification and one-step dry film provided in this invention adopts a route of pre-research, development, trial production, verification, and standardization. The pre-research stage involves completing formula simulation (resin crosslinking degree simulation) and verifying the photosensitive principle, outputting preliminary formulas and process schemes. The process and testing development stage involves conducting formula experiments (monomer ratio optimization), coating and exposure process debugging, and developing an integrated testing method. The sample trial production stage involves small-batch trial production of TSF photosensitive solder resist dry film samples and completing core performance tests. The customer verification and standardization stage involves sending samples to the client for testing, collecting feedback for optimization, and forming a mass production standard process. The invention involved: completing the simulation of the elastic dry film formulation and verifying the modification of the FPC substrate; determining the raw material suppliers (elastic resin, PI substrate), and outputting preliminary formulations and process schemes; conducting formulation experiments (optimizing the elastic resin ratio), substrate modification, and adjusting the dry film lamination parameters; producing small-batch trial production of TSF dry film for FPC, and completing internal reliability testing after lamination onto the FPC substrate, with a pass rate of ≥95%; conducting reliability tests on flexural strength, solvent resistance, thermal shock, reflow soldering, insulation resistance, and adhesion; and conducting dry film filling, thickness, and ion migration tests; the TSF series photosensitive solder resist dry film of this invention is specifically designed to replace traditional cover film and ink processes, possessing both good heat resistance, flexural strength, and chemical resistance; and developing new process documents for vacuum lamination parameters, exposure energy, and development.

[0090] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention and within the spirit and principles of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for preparing a substrate based on low-temperature plasma modification and one-step dry film, characterized in that, The method includes the following steps: S1, FPC substrate pretreatment, PI / PET substrate cutting, low-temperature plasma modification, drying; S2, TSF dry film preparation: elastic resin, photosensitizer, monomer mixture, and solvent viscosity adjustment; S3, dry film lamination, hot pressing, cooling and shaping; S4, photolithography, exposure, development, and thermal curing; S5, reliability testing, bending cycle test, high temperature and high humidity test, circuit accuracy test; S6, FPC finished product cutting, appearance inspection, packaging and shipping.

2. The substrate preparation method based on low-temperature plasma modification and one-step dry film according to claim 1, characterized in that, In step S1, the low-temperature plasma modification includes: S101, surface activation is performed using low-voltage pulsed radio frequency plasma with a power of 50-200W; S102, micro-hole etching is performed using a DBD plasma of an oxygen / argon mixed gas with a gas ratio of O2:Ar=1:9; S103 utilizes a nitrogen-hydrogen plasma gradient treatment method with a pressure gradient of 0.1-1 Pa for interface strengthening.

3. The substrate preparation method based on low-temperature plasma modification and one-step dry film according to claim 1, characterized in that, In step S2, the optimized formula for the elastic resin ratio is obtained through resin crosslinking degree simulation. Specific steps include: S201, Molecular structure modeling and cross-linking network design, including: by introducing a reversible molecular spring structure with flexible segments and dynamic response stress at cross-linking points, based on target properties including elasticity and tear strength, a molecular dynamics model of elastic resin is constructed by using dynamic cross-linking gradient, interfacial energy control parameters, and hyperelastic model coefficients to simulate the distribution of cross-linking points and three-dimensional network structure. S202, Multi-scale simulation and parameter optimization; employing layered simulation: Macroscopic scale, predicting the overall mechanical behavior of materials, including tensile strength and elongation at break; At the microscale, the effect of local crosslinking density on performance was analyzed using Representational Volume Units (RVEs). Combined with Direct FE 2 Multi-scale coupling technology enables collaborative simulation of macroscopic and microscopic deformation; S203, Crosslinking Degree-Performance Mapping Database Construction; Predicting the Crosslinking Degree of Resin / Functional Nanoparticles under Different Formulations through Quantum Chemical Calculations and Correlating it with Key Performance Indicators such as Tear Strength and Bending Resistance; The optimization scheme for elastic resin formulation includes: (1) a single-component elastic resin formulation, using FlexOne series resin, and replacing the traditional two-component system by optimizing the molecular chain structure; (2) Synergistic ratio of resin and nanofiller; add functional nanoparticles to enhance interfacial bonding through in-situ polymerization. The addition ratio is: the mass fraction of carbon nanotubes (CNTs) in the resin matrix is ​​usually 0.1%-2.0%.

4. The substrate preparation method based on low-temperature plasma modification and one-step dry film according to claim 3, characterized in that, In step S201, the reversible molecular spring structure includes: A reversible three-dimensional network structure is formed by flexible segments between cross-linking points. At the microscale, the flexible segments connect the cross-linking points through dynamic bonds. When subjected to force, the segments extend and disperse the stress, and return to their original state after unloading. Dynamic bond selection: Reversible covalent bonds or supramolecular interactions are used to ensure that crosslinking points can dynamically respond to stress.

5. The substrate preparation method based on low-temperature plasma modification and one-step dry film according to claim 3, characterized in that, The steps of multi-parameter collaborative modeling include: (1) Dynamic crosslinking gradient design; crosslinking point distribution optimization, through molecular dynamics MD simulation, to regulate the spatial distribution of crosslinking agents; (2) Integration of interface energy regulation parameters; modeling of interface interaction, introducing interface energy parameters, and optimizing the interface between carbon nanotube filler and matrix; (3) Calibration of hyperelastic model coefficients; embedding constitutive equations to map hyperelastic model coefficients to MD simulation: The elastic modulus is obtained by uniaxial tensile simulation to obtain the stress-strain curve and fit the model parameters. Viscoelastic parameters, combined with the generalized Maxwell / Kelvin model, describe the frequency-dependent response; Time-temperature equivalence: Using the WLF equation to correlate modulus changes under different temperature / frequency domains to predict long-term service performance; The process of building a molecular dynamics model includes: (i) Initial structure generation: a mixture of resin monomer and curing agent was constructed using Materials Studio / PackMol, and flexible segments were introduced as spring units. (ii) Dynamic cross-linking reaction simulation: Set cross-linking degree target, and optimize network topology through energy minimization and annealing cycle; Monitor the path length between crosslinking points and the Voronoi volume to assess network looseness; (iii) Performance verification and optimization, including: mechanical response simulation, dynamic performance verification, and reversibility verification.

6. The substrate preparation method based on low-temperature plasma modification and one-step dry film according to claim 1, characterized in that, The hot-pressing bonding in the dry film bonding process in step S3 includes: S301, multi-stage pressure curve control; adopts segmented pressurization instead of traditional constant pressure, divided into three stages: pre-pressurization, main pressure, and pressure holding. S302, dynamic temperature compensation bonding; an infrared temperature measurement module is integrated into the hot press head to monitor the substrate temperature in real time and adjust the heating power accordingly. S303, vacuum lamination bonding for local reinforcement.

7. The substrate preparation method based on low-temperature plasma modification and one-step dry film according to claim 6, characterized in that, In step S301, the pre-compression stage is characterized by a pressure of 0.1-0.2 MPa, a time of 3-5 seconds, and a temperature of 80°C. Main pressure stage: Pressure rises to 0.8-1.2 MPa, temperature rises to 100-110℃, duration 8-12 seconds; Pressure holding stage: The pressure drops to 0.3-0.5 MPa, and the temperature is maintained for 5-8 seconds.

8. The substrate preparation method based on low-temperature plasma modification and one-step dry film according to claim 6, characterized in that, In step S302, the temperature fluctuation of the substrate is monitored in real time and controlled within ±2℃; Step S303, vacuum lamination bonding for local reinforcement, includes: The cover film and the substrate are pre-bonded inside the vacuum chamber; FR4 / PI reinforcing plates were added locally to high-stress areas such as connectors; The vacuum level is maintained at 0.05-0.1 MPa during hot pressing curing; Vacuum lamination temperature 170-190℃, pressure 0.8-1.2MPa.

9. The substrate preparation method based on low-temperature plasma modification and one-step dry film according to claim 1, characterized in that, In step S4, a CCD vision positioning system is introduced, and LED surface light source power is controlled by partitioning. In step S4, during development, the concentration of the developer, temperature, spray pressure, and development time parameters are optimized by evaluating the ultrathin dry film. In step S5, during the line accuracy test, no residue was found in the 30μm line.

10. The application of the substrate preparation method based on low-temperature plasma modification and one-step dry film as described in any one of claims 1-9 in the preparation of TSF dry film products.

Images