Composite pressing process of high multilayer rigid-flex circuit board

Abstract

The invention belongs to the technical field of circuit boards, and discloses a composite pressing process of a high multilayer rigid-flex circuit board, which comprises the following steps: performing microstructure processing on the surface of a rigid base material to form a three-dimensional reinforced interface, and performing chemical grafting modification treatment on a flexible base material to generate a multilayer pretreated substrate; identifying the positioning mark of each layer of substrate through an optical positioning system, establishing a lamination stress prediction model in combination with the thermodynamic parameters of the material, and outputting a lamination configuration scheme containing an interlayer matching relationship; placing the configured laminated structure in a multi-temperature-zone press, and executing a three-stage gradient heating and pressurizing program to realize interface fusion between adjacent layers of substrates; monitoring an interlayer deformation difference value of the substrate in real time through a sensing network, and eliminating the deformation difference to obtain a press-fit molded circuit substrate; through heterogeneous interface strengthening design, multi-physical field cooperative pressing and real-time deformation compensation, accurate interface fusion and low-warpage forming of the high-multilayer rigid-flex circuit board are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit boards, and particularly to a composite pressing process for high multi-layer rigid-flexible printed circuit boards. Background Art

[0002] With the rapid development of fields such as millimeter-wave communication, satellite payloads, and high-end medical equipment, high multi-layer rigid-flexible printed circuit boards have become the core carriers of millimeter-wave antenna arrays and high-density interconnect modules due to their three-dimensional wiring ability and high-frequency signal integrity advantages. In the power amplifier module of a communication base station, it is necessary to integrate more than 20 layers of alternating rigid and flexible structures within a limited space, and at the same time, the transmission loss at the 28 GHz frequency band is required to be lower than 0.15 dB / cm, which poses strict requirements on the interfacial bonding strength of heterogeneous materials, the uniformity of the interlayer dielectric, and the structural stability.

[0003] Existing composite pressing processes for high multi-layer circuit boards mostly adopt single-temperature zone hot pressing and forming technology, and the laminated substrates are integrally pressed through unified temperature-pressure parameters. For example, in the disclosed lamination process, FR4 and polyimide substrates are synchronously pressed under the conditions of a constant 135 °C and 2.0 MPa. Although such methods can achieve basic interlayer bonding, they do not consider the thermal deformation differences between rigid materials (FR4 has a thermal expansion coefficient of 13 ppm / °C) and flexible materials (PI has a thermal expansion coefficient of 35 ppm / °C). During the pressing and cooling process, due to the interlayer stress mismatch caused by the shrinkage rate differences of heterogeneous materials, it is easy to cause interfacial delamination and overall warping, resulting in fluctuations in the thickness of the high-frequency signal transmission layer medium, an increase in the insertion loss at the 28 GHz frequency band, and impedance jump failure occurs when the thermal cycle life is less than 800 times. This technical bottleneck seriously restricts the reliable application of high-density rigid-flexible printed circuit boards in millimeter-wave scenarios.

[0004] In view of this, it is necessary to address the technical problem of interlayer deformation mismatch caused by differences in the thermal expansion coefficients of the plates during the composite pressing process of high multi-layer circuit boards in the prior art. Summary of the Invention

[0005] The purpose of the present invention is to provide a composite pressing process for high multi-layer rigid-flexible printed circuit boards to solve the above technical problems.

[0006] To achieve this purpose, the present invention adopts the following technical solutions: A composite pressing process for high multi-layer rigid-flexible printed circuit boards includes the following steps: S1, in the substrate pretreatment stage, microstructural processing is performed on the surface of the rigid substrate to form a three-dimensional strengthening interface, and at the same time, chemical grafting modification treatment is performed on the flexible substrate to generate a multi-layer pretreated substrate; S2, Laminating Configuration Stage: Identify the positioning marks of each layer of substrate through an optical positioning system, establish a lamination stress prediction model by combining material thermodynamics parameters, output a lamination configuration plan including the interlayer matching relationship, and configure and form a laminated structure according to the lamination configuration plan; S3, Gradient Pressing Stage: Place the configured laminated structure in a multi-temperature zone press and execute a three-stage gradient heating and pressing process, including low-temperature pre-flow, medium-temperature main curing, and high-temperature shaping processes, to achieve interfacial fusion between adjacent substrate layers; S4, Dynamic Deformation Regulation Stage: Real-time monitor the deformation difference between substrate layers through a sensing network. When the preset deformation threshold is detected, trigger the local temperature control module and pressure compensation device to eliminate the deformation difference and obtain a pressed and formed circuit substrate.

[0007] Optionally, the specific steps of step S1 include: S11, Surface Cleaning Treatment of Rigid Substrate: Clean the surface of the rigid substrate using an ultrasonic cleaning device, and then remove the residual particles on the surface through an air knife to obtain a clean rigid substrate; S12, Etch a wavy microgroove array on the surface of the rigid substrate using a laser, and simultaneously monitor the etching morphology online using a confocal microscope to generate a three-dimensional anchoring interface with mechanical interlocking function; S13, Place the flexible substrate into a vacuum plasma reaction chamber, introduce an Ar / O2 mixed gas with a volume ratio of 4:1, and process it at a set power for 120 seconds to obtain an activated flexible substrate with a target surface oxygen element content rate and surface energy; S14, Immerse the activated flexible substrate in an ethanol solution containing 3-aminopropyltriethoxysilane and react it in a constant temperature bath at 80°C for 30 minutes to complete the grafting of amino functional groups and form a molecular bonding layer with a target grafting density; S15, Composite Surface Treatment: Spray a composite treatment solution containing nano-silica on the three-dimensional anchoring interface of the rigid substrate, and simultaneously coat a polyimide-epoxy blend transition layer on the grafting layer of the flexible substrate. After pre-curing at 80°C, a gradient chemical bonding interface is formed.

[0008] Optionally, the composite treatment solution contains 15% nano-silica and 2% silane coupling agent by mass fraction, and forms a stable suspension after ultrasonic dispersion.

[0009] Optionally, the specific steps of step S2 include: S21, Scan the positioning holes of each layer of substrate through an optical imaging system to generate a positioning data matrix including XY-axis coordinate deviation and rotation angle; S22, Collect the thermal expansion coefficient, elastic modulus, and dielectric constant parameters of each layer of substrate, and construct a digital twin model of the three-dimensional laminated structure by combining the positioning data; S23. Apply a temperature-pressure coupling field to the digital twin model based on the finite element analysis algorithm, predict the stress distribution contour map of the laminated interface, and mark the potential deformation risk areas. S24. Based on the marked deformation risk areas, use the Monte Carlo algorithm to iteratively optimize the interlayer matching relationship, and output the optimal laminated configuration scheme including the hierarchical alternating order, interlayer offset compensation amount, and pre-deformation amount. S25. According to the optimal laminated configuration scheme, use a vacuum adsorption mechanism to grab the corresponding pre-treated substrate, and cooperate with a six-axis alignment platform to complete the physical stacking of the preset number of layers, forming a laminated structure to be pressed.

[0010] Optionally, step S3 specifically includes: S31. Transfer the configured laminated structure to the multi-temperature zone press cavity, calibrate the position deviation between the laminate and the platen through the alignment system to ensure the XY-axis offset. S32. Set the partition temperature field, divide independent temperature control units according to the distribution of hard and soft substrates, set 120°C ± 2°C for the rigid area and 105°C ± 2°C for the flexible area to form a gradient thermal field distribution. S33. Low-temperature pre-flow process, load the laminated structure with an initial pressure of 0.8 MPa, heat it up to 110°C at a rate of 1.5°C / min, and simultaneously increase the pressure to 1.2 MPa at a gradient of 0.05 MPa / min, and maintain for 20 min to make the epoxy resin fill the three-dimensional anchoring interface. S34. Medium-temperature main curing process, switch to the dual-temperature zone mode: heat the rigid area to 135°C and the flexible area to 125°C, increase the pressure to 2.5 MPa and keep it for 30 min, and monitor the resin curing degree in real time through an infrared thermal imager during this period. S35. High-temperature shaping process, raise the overall temperature to 150°C, apply a high pressure of 3.0 MPa for shaping for 10 min, and at the same time turn on the ultrasonic interface monitoring system to detect the interlayer bonding state to ensure that the acoustic reflection coefficient is less than the preset coefficient value. S36. Gradient cooling control, cool down to 80°C at a rate of 0.8°C / min, simultaneously reduce the pressure step by step to 0.5 MPa at a gradient of 0.1 MPa / min, and then introduce nitrogen to cool it down to room temperature to form a pressed substrate with interfacial fusion.

[0011] Optionally, the multi-temperature zone press includes 16 independent heating modules, and the 16 heating modules are arranged in an array, and each heating module covers an area of N*N mm 2 of the area.

[0012] Optionally, step S4 specifically includes: S41. Sample the laminated substrate, embed distributed fiber optic strain sensors between the layers of the laminated structure of the sampled laminated substrate, synchronously install a piezoelectric film array, and construct a sensing network for real-time monitoring of interlayer deformation; S42. Deformation threshold judgment. Obtain the strain data of each layer in real time through a multi-channel data acquisition system. When the deformation difference between adjacent layers is greater than the preset deformation threshold, trigger a compensation instruction, and at the same time predict the deformation trend in the preset time period in the future based on the LSTM neural network; S43. For the laminated substrate that has not been sampled, locate the target area with excessive deformation according to the compensation instruction, apply local temperature control and piezoelectric ceramic pressure compensation to the target area, and the synchronization adjustment time does not exceed 30 seconds to perform dynamic compensation; S44. After completing the dynamic compensation, perform secondary gradient annealing, set the annealing parameters as 100°C → 60°C, and the rate as 0.5°C / min. Detect the overall warpage through a laser interferometer, and output the laminated and formed circuit substrate.

[0013] Optionally, after step S4, the following is further included: S5. Quality verification stage. Perform interlayer bonding scanning detection, electrothermal performance simulation modeling, and reliability prediction on the laminated and formed substrate in sequence, and generate a quality assessment report including process parameter optimization suggestions.

[0014] Optionally, step S5 specifically includes: S51. Nondestructive detection of interlayer bonding. Perform three-dimensional tomographic scanning on the laminated substrate using a tomographic scanning system, analyze the reflected wave characteristics through a deep learning algorithm, and generate interlayer bonding data including porosity and delamination area; S52. Based on the results of the three-dimensional tomographic scanning, construct a three-dimensional electrothermal coupling model, load the working conditions of the power amplifier module, calculate the current density distribution and thermal gradient field of the high-frequency signal transmission path, and output the predicted transmission loss value; S53. Intelligent reliability prediction. Input the interlayer bonding data and the predicted transmission loss value into a pre-trained recurrent neural network, and combine the accelerated aging test data to predict the failure probability curve of the circuit substrate during the preset number of years of working period; S54. Process parameter optimization feedback. According to the failure probability curve, use the decision tree algorithm to compare the life data in the historical process database, and generate a process parameter optimization plan.

[0015] Compared with the prior art, the present invention has the following beneficial effects: First, a three-dimensional strengthening interface is formed by laser micro-structuring of a rigid substrate, and at the same time, plasma activation graft modification is performed on a flexible substrate to prepare a pre-treated substrate with an active bonding surface; then, an optical positioning system is used to identify the positioning marks of each layer, and a laminated stress prediction model is established in combination with the coefficient of thermal expansion of the material to generate a laminated configuration scheme with interlayer matching; the configured laminated structure is placed into a multi-temperature zone press, and a three-stage gradient pressing process including low-temperature pre-flow, medium-temperature main curing, and high-temperature shaping is performed; then, the interlayer deformation difference is monitored in real time through a distributed optical fiber sensing network. When a deformation threshold exceeding the preset value is detected, a local temperature control module is triggered to perform temperature compensation and adjust the pressure field distribution to obtain a press-formed circuit board with tight interlayer bonding; through the design of heterogeneous interface strengthening, multi-physical field collaborative pressing, and real-time deformation compensation, this process realizes the precise interface fusion and low-warpage forming of high-multilayer rigid-flex printed circuit boards. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0017] The structures, ratios, sizes, etc. shown in the drawings of this specification are only used to cooperate with the content disclosed in the specification for those who are familiar with this technology to understand and read, and are not used to limit the limiting conditions under which the present invention can be implemented. Therefore, they do not have technical substance. Any modification of the structure, change of the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope that can be covered by the technical content disclosed in the present invention.

[0018] Figure 1 It is one of the flow diagrams of the composite pressing process of the high-multilayer rigid-flex printed circuit board of this embodiment; Figure 2 It is another flow diagram of the composite pressing process of the high-multilayer rigid-flex printed circuit board of this embodiment; Figure 3 It is the third flow diagram of the composite pressing process of the high-multilayer rigid-flex printed circuit board of this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the object, features, and advantages of the present invention more obvious and understandable, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the following described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "upper", "lower", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention. It should be noted that when a component is considered to be "connected" to another component, it may be directly connected to the other component or there may be intermediate components present simultaneously.

[0021] The following further illustrates the technical solutions of the present invention with reference to the accompanying drawings and through specific embodiments.

[0022] Combined Figures 1 to 3 As shown, the embodiments of the present invention provide a composite pressing process for high multi-layer rigid-flex printed circuit boards, including the following steps: S1. Substrate pretreatment stage: Microstructure processing is performed on the surface of the rigid substrate to form a three-dimensional strengthening interface, and at the same time, chemical grafting modification treatment is performed on the flexible substrate to generate a multi-layer pretreated substrate. Laser microstructure processing is performed on the rigid substrate to form a three-dimensional anchoring interface, increasing the mechanical interlocking area; plasma activation and chemical grafting are performed on the flexible substrate to introduce active functional groups to enhance the molecular bond force. This pretreatment lays a dual combination foundation of physical anchoring and chemical bonding for subsequent pressing, and can effectively improve the thermal stress distribution at the interface between rigid and flexible materials, reducing the risk of delamination.

[0023] S2. Laminating configuration stage: The positioning marks of each layer of substrate are identified through an optical positioning system, and a lamination stress prediction model is established in combination with the material thermodynamics parameters, and a lamination configuration scheme including the interlayer matching relationship is output, and a laminated structure is configured according to the lamination configuration scheme. High-precision lamination optimization is realized based on optical positioning and digital modeling technologies. The positioning marks of each layer are captured through a multi-spectral imaging system, and a lamination stress prediction model is constructed in combination with parameters such as the coefficient of thermal expansion and elastic modulus of the material to intelligently optimize the interlayer matching relationship. This step predicts the stress distribution during the hot pressing process and actively designs the lamination offset compensation amount, thereby reducing the interlayer misalignment and warping problems caused by the difference in thermal deformation of materials.

[0024] S3, Gradient pressing stage: Place the configured laminated structure in a multi-temperature zone press and execute a three-stage gradient heating and pressing process, including low-temperature pre-flow, medium-temperature main curing, and high-temperature shaping processes, to achieve interfacial fusion between adjacent hierarchical substrates; Adopt a multi-temperature zone segmented pressing strategy to achieve co-curing of heterogeneous materials. Promote resin flow to fill the interfacial microstructure in the low-temperature stage, complete the main cross-linking reaction in the medium-temperature stage, and release residual stress in the high-temperature stage. The gradient heating and pressing process matches the difference in glass transition temperature of hard and soft materials, avoiding local over-curing or under-curing phenomena caused by single-parameter pressing, and ensuring the uniformity of interfacial fusion.

[0025] S4, Dynamic deformation regulation stage: Real-time monitor the deformation difference between substrate layers through a sensing network. When the preset deformation threshold is detected, trigger the local temperature control module and pressure compensation device to eliminate the deformation difference, and obtain a pressed and formed circuit substrate.

[0026] Control the interlayer deformation through an embedded sensing network and a real-time compensation mechanism. Use distributed fiber optic sensors to monitor the strain distribution of each layer, combine with a prediction model to predict the deformation trend. When the detected deformation exceeds the tolerance, trigger the local temperature control and pressure compensation modules for dynamic adjustment.

[0027] S5, Quality verification stage: Sequentially perform interlayer bonding scanning detection, electrothermal performance simulation modeling, and reliability prediction on the pressed and formed substrate, and generate a quality assessment report including process parameter optimization suggestions.

[0028] Construct a multi-dimensional quality assessment system to achieve process closed-loop optimization. Use terahertz wave tomography to detect interlayer bonding defects, evaluate the high-frequency signal transmission performance through electrothermal coupling simulation, and combine with a reliability prediction model to predict the long-term service performance. Finally, reverse-optimize the pressing parameters based on the detection data to form an iterative upgrade link of "manufacturing - detection - feedback", continuously improving the process stability.

[0029] The working principle of the present invention is as follows: First, laser micro-structuring is performed on the rigid substrate to form a three-dimensional strengthening interface, and at the same time, plasma activation grafting modification is carried out on the flexible substrate to prepare a pre-treated substrate with an active bonding surface; then, each layer of positioning marks is identified through an optical positioning system, and a laminated stress prediction model is established in combination with the coefficient of thermal expansion of the material to generate a laminated configuration plan with interlayer matching; the configured laminated structure is placed into a multi-temperature zone press, and a three-stage gradient pressing process including low-temperature pre-flow, medium-temperature main curing, and high-temperature shaping is executed; then, the interlayer deformation difference is monitored in real time through a distributed optical fiber sensing network. When the detected deformation exceeds the preset deformation threshold, the local temperature control module is triggered to perform temperature compensation and adjust the pressure field distribution to obtain a press-formed circuit board with a tightly bonded interlayer; through the strengthening design of the heterogeneous interface, the collaborative pressing of multiple physical fields, and real-time deformation compensation, this process realizes the precise interface fusion and low-warpage forming of high-multi-layer rigid-flex printed circuit boards.

[0030] In this embodiment, specifically, step S1 specifically includes: S11, cleaning the surface of the rigid substrate. The surface of the rigid substrate is cleaned using an ultrasonic cleaning device, and then the residual particles on the surface are removed by an air knife to obtain a clean rigid substrate; Through the synergistic effect of ultrasonic cleaning and air knife treatment, the cleanliness of the substrate surface is ensured. The ultrasonic cavitation effect can efficiently peel off the surface grease and particles, and the non-contact blowing of the air knife avoids secondary contamination. Compared with traditional solvent wiping, it can better protect the microscopic structure of the substrate surface. The set ultrasonic frequency and air knife pressure (about 0.2 - 0.5 MPa) are optimized to reach the industrial-grade cleaning standard without damaging the substrate, providing an interference-free substrate for subsequent laser etching.

[0031] S12, using a laser to etch a wavy micro-groove array on the surface of the rigid substrate, and simultaneously using a confocal microscope to monitor the etching morphology online to generate a three-dimensional anchoring interface with a mechanical interlocking function; Laser etching realizes cold processing through ultra-short pulses, avoiding carbonization damage to the rigid substrate (FR4 material) in the heat-affected zone. The design of the wavy micro-grooves (preferably with an amplitude of 50 μm / wavelength of 200 μm) matches the subsequent resin flow characteristics, and its radius of curvature can enhance the mechanical anchoring effect. The online monitoring of the confocal microscope ensures the consistency of the etching depth through three-dimensional morphology reconstruction and real-time feedback adjustment of the laser parameters.

[0032] S13, placing the flexible substrate into a vacuum plasma reaction chamber, introducing an Ar / O2 mixed gas with a volume ratio of 4:1, and treating it at a set power for 120 seconds to obtain an activated flexible substrate with a target surface oxygen element content rate and surface energy; The plasma treatment of the Ar / O2 mixed gas (4:1) balances the requirements of physical bombardment and chemical modification. Argon ion sputtering removes the weak boundary layer on the surface, and oxygen radicals introduce polar groups such as carboxyl / hydroxyl groups. The combination of 50W power and 120 seconds of treatment time increases the surface oxygen content to the saturation threshold of 28 at%, while avoiding the embrittlement of the PI film caused by over-etching.

[0033] S14, Immerse the activated flexible substrate in an ethanol solution containing 3-aminopropyltriethoxysilane (concentration 1.5 wt%), and react in a constant temperature bath at 80 °C for 30 minutes to complete the grafting of amino functional groups and form a molecular bonding layer with the target grafting density; The concentration design of the 1.5 wt% APTES ethanol solution takes into account both the reaction efficiency and the quality of the self-assembled monolayer: too low a concentration results in insufficient grafting density, while too high a concentration causes multi-layer aggregation. The hydrolysis and condensation reaction of silane is accelerated at 80 °C, and the 30-minute reaction time enables the amino grafting density to reach the saturation value of 8×10 14 groups / cm 2 (verified by XPS), constructing an active interface that can react with epoxy resin at the molecular level. Compared with the room-temperature grafting process, the reaction efficiency is improved and the grafting uniformity is enhanced.

[0034] S15, Composite surface treatment. Spray a composite treatment solution containing nano-silica on the three-dimensional anchoring interface of the rigid substrate, and at the same time coat a polyimide-epoxy blend transition layer on the grafting layer of the flexible substrate. After pre-curing at 80 °C, a gradient chemical bonding interface is formed.

[0035] The coefficient of thermal expansion of the polyimide-epoxy blend transition layer is designed to be 8 ppm / °C (between 35 ppm / °C of PI and 13 ppm / °C of FR4), forming a gradient buffer layer. The 80 °C pre-curing preliminarily crosslinks the treatment solution, retaining sufficient fluidity to adapt to the subsequent lamination process.

[0036] In this embodiment, it is further illustrated that the composite treatment solution contains 15% by mass of nano-silica and 2% of silane coupling agent, and forms a stable suspension after ultrasonic dispersion.

[0037] The addition amount (15 wt%) of nano-silica (particle size 20 nm) is optimized by rheology, which improves the mechanical strength of the epoxy resin while maintaining the viscosity below 3000 cps to ensure sufficient filling of the micro-grooves.

[0038] In this embodiment, specifically, step S2 specifically includes: S21, Scan the positioning holes of each layer of the substrate through an optical imaging system to generate a positioning data matrix containing the XY-axis coordinate deviation and the rotation angle; The multispectral optical imaging system overcomes the positioning mark recognition error caused by the semi-transparency of the substrate by fusing visible light and near-infrared band imaging. Compared with the monochromatic light positioning technology, multispectral imaging can eliminate the reflection interference on the material surface and improve the positioning accuracy of complex laminated structures.

[0039] S22, collect the thermal expansion coefficient, elastic modulus and dielectric constant parameters of each layer of the substrate, and combine the positioning data to construct a digital twin model of the three-dimensional laminated structure; The multi-parameter fusion modeling of the thermal expansion coefficient, elastic modulus and dielectric constant accurately characterizes the coupling behavior of the material under the temperature-pressure field. By combining the physical positioning data (XY deviation, rotation angle) with the material constitutive equation, the constructed three-dimensional digital twin model can simulate the thermo-mechanical response of the real laminate. This model breaks through the limitations of traditional single physical field simulation and provides high-fidelity input for subsequent stress prediction, especially suitable for the anisotropic analysis of hard-soft alternating structures.

[0040] S23, apply a temperature-pressure coupling field to the digital twin model based on the finite element analysis algorithm, predict the stress distribution nephogram of the laminated interface and mark the potential deformation risk areas; The 80 - 150 °C temperature gradient field simulates the thermal load changes during the actual pressing process, and the 0.5 - 3.0 MPa pressure gradient field covers the mechanical conditions during the entire stage from resin flow to curing. Through multi-field coupling calculation, the finite element analysis identifies the interface shear stress concentration areas (such as the junction of hard and soft materials) and bending risk areas (such as the edges of large-sized rigid layers). Compared with the estimation by empirical formulas, this method can quantitatively predict the maximum interlayer stress value and its spatial distribution characteristics, providing a targeted goal for optimization.

[0041] S24, based on the marked deformation risk areas, use the Monte Carlo algorithm to iteratively optimize the interlayer matching relationship, and output the optimal laminated configuration scheme including the hierarchical alternating order, interlayer offset compensation amount and pre-deformation amount; The Monte Carlo algorithm searches for the Pareto optimal solution in the multi-variable space of hard-soft alternating order, interlayer offset compensation, pre-deformation amount, etc. through iterative global random search. Set the double constraint conditions of interlayer stress < 15 MPa and warpage < 0.3% to ensure that the scheme meets the requirements of both mechanical strength and deformation control.

[0042] S25, according to the optimal laminated configuration scheme, use a vacuum adsorption mechanism to grab the corresponding pre-treated substrate, and cooperate with a six-axis alignment platform to complete the physical stacking of the preset number of layers to form a laminated structure to be pressed.

[0043] The vacuum adsorption mechanism senses the curvature of the substrate through a flexible tactile sensor and adaptively adjusts the adsorption force to avoid wrinkling of thin-layer materials (0.1 mm PI). The six-axis alignment platform has a closed-loop linkage with the optical positioning data for repeated positioning accuracy.

[0044] In this embodiment, specifically, step S3 specifically includes: S31. Transfer the configured laminated structure to the multi-temperature zone press cavity, and calibrate the position deviation between the laminate and the platen through the alignment system to ensure the XY-axis offset; Calibrate the relative position between the laminate and the platen through the laser alignment system, control the XY-axis offset within the ideal range, and ensure uniform pressure transmission during the hot pressing process. This accuracy requirement matches the die tolerance of the multi-temperature zone press to avoid stress concentration in the edge area caused by initial misalignment. Compared with mechanical stop positioning, laser dynamic calibration can adapt to the micro-displacement caused by the thermal expansion of the laminate.

[0045] S32. Set the zonal temperature field. Divide independent temperature control units according to the distribution of hard and soft substrates. Set the rigid zone at 120°C ± 2°C and the flexible zone at 105°C ± 2°C to form a gradient thermal field distribution; The independent temperature control units are divided according to the distribution of hard and soft substrates (for example, an independent heating module is set for each 50×50 mm 2 area). The 120°C ± 2°C in the rigid zone matches the pre-curing requirement of the FR4 glass transition temperature (Tg≈140°C), and the 105°C ± 2°C in the flexible zone avoids premature shrinkage of the PI film (Tg≈250°C). The gradient design with a double-temperature zone ΔT = 15°C not only meets the rheological property differences of heterogeneous materials but also prevents thermal stress cracks from occurring in the interface area due to excessive temperature difference.

[0046] S33. Low-temperature pre-flow process. Load the laminated structure with an initial pressure of 0.8 MPa, heat it to 110°C at a rate of 1.5°C / min, and simultaneously increase the pressure to 1.2 MPa at a gradient of 0.05 MPa / min and maintain it for 20 min to allow the epoxy resin to fill the three-dimensional anchoring interface; The combination of an initial pressure of 0.8 MPa and a heating rate of 1.5°C / min causes the viscosity of the epoxy resin to gradually decrease from 3000 cps to 800 cps (in the 80 - 110°C range). The gradient pressure increase of 0.0 MPa / min synchronously matches the resin flow front expansion rate. Maintaining for 20 min ensures that the resin fully fills the micro-grooves of the three-dimensional anchoring interface (depth 50 μm) to form a continuous bonding interface. Compared with constant pressure loading, the gradient pressure increase strategy reduces the interface poor glue defect caused by resin extrusion.

[0047] S34. Medium-temperature main curing process. Switch to the double-temperature zone mode: heat the rigid zone to 135°C and the flexible zone to 125°C, increase the pressure to 2.5 MPa and maintain it for 30 min. During this period, monitor the resin curing degree in real time through an infrared thermal imager; The rigid region approaches the Tg point of FR4 at 135°C to activate the cross-linking reaction, and the flexible region is below the Tg point of PI at 125°C to avoid thermal shrinkage mismatch. A constant pressure of 2.5 MPa is applied to make the resin curing degree reach 85% ± 3%, forming a preliminary network structure while retaining an appropriate amount of unreacted functional groups. The synergistic effect of the dual-temperature zones reduces the curing rate difference in the soft-hard interface region and reduces the accumulation of interlayer internal stress.

[0048] S35, high-temperature shaping process, raise the overall temperature to 150°C, apply a high pressure of 3.0 MPa for shaping for 10 min, and at the same time turn on the ultrasonic interface monitoring system to detect the interlayer bonding state to ensure that the acoustic reflection coefficient is less than the preset coefficient value; The high temperature of 150°C breaks through the reaction energy barrier of the residual functional groups, and the high pressure of 3.0 MPa promotes the densification of molecular chains, synchronously eliminating the microvoids caused by curing shrinkage. The ultrasonic monitoring determines the interface bonding quality through the reflection coefficient threshold (<0.15), and its wavelength matches the thickness of the interface transition layer, which can effectively identify delamination defects.

[0049] S36, gradient cooling control, cool down to 80°C at a rate of 0.8°C / min, simultaneously step down the pressure to 0.5 MPa at a rate of 0.1 MPa / min in steps, and then introduce nitrogen to cool down to room temperature forcibly to form a laminated substrate with interface fusion.

[0050] The slow cooling rate of 0.8°C / min matches the glass transition range (100 - 120°C) of the epoxy resin, and more than 80% of the residual stress is released through the slow relaxation of molecular chains. The forced nitrogen cooling replaces the traditional air convection, which not only avoids high-temperature oxidation but also realizes directional heat dissipation. Stepping down the pressure to 0.5 MPa prevents interlayer debonding caused by cooling shrinkage and controls the warpage of the substrate.

[0051] In this embodiment, it is further illustrated that the multi-temperature zone press includes 16 independent heating modules, and the 16 heating modules are arranged in an array, and each heating module covers an area of N*Nmm 2 of the region.

[0052] The multi-temperature zone press adopts an array arrangement design of 16 independent heating modules, and each module covers an area of N×N mm 2 region (such as 50×50mm 2 ) and realizes precise regulation of the thermal field through independent temperature control of each zone. This structure can dynamically divide different temperature control units according to the distribution characteristics of the soft and hard substrates in the laminate (such as setting a high temperature in the rigid region and a low temperature in the flexible region), so that the temperature of each region matches the glass transition characteristics of the corresponding material.

[0053] In this embodiment, it is specifically illustrated that step S4 specifically includes: S41. Take samples of the laminated substrate, embed distributed fiber optic strain sensors between the layers of the laminated structure of the sampled laminated substrate, synchronously install a piezoelectric film array, and construct a sensing network for real-time monitoring of interlayer deformation; By embedding distributed fiber optic strain sensors (spacing 10mm×10mm) between the layers of the laminated substrate and combining with a piezoelectric film array, a high-density deformation monitoring network is constructed. The fiber optic sensor realizes microstrain detection through wavelength coding technology, and the piezoelectric film array captures changes in local pressure distribution. The two cooperate to provide mechanical state data for the entire field.

[0054] S42. Deformation threshold judgment. Real-time obtain the strain data of each layer through a multi-channel data acquisition system. When the deformation difference between adjacent layers is greater than the preset deformation threshold, trigger a compensation instruction, and at the same time predict the deformation trend in the future preset time period based on the LSTM neural network; Set the deformation difference threshold between adjacent layers to 0.03%. When the detected value exceeds it, trigger the compensation mechanism. The LSTM neural network predicts the deformation trend in the next 5 minutes based on historical process data (such as temperature and pressure time series curves) to achieve advanced regulation. This threshold setting balances the process tolerance and reliability requirements, and avoids energy loss caused by overcompensation. The LSTM model captures the thermal hysteresis effect through time series analysis to improve the timeliness of deformation prediction.

[0055] S43. For the laminated substrate without sampling, locate the target area with excessive deformation according to the compensation instruction, apply local temperature control and piezoelectric ceramic pressure compensation to the target area, and the synchronous adjustment time does not exceed 30 seconds to perform dynamic compensation; Complete the compensation response within 30 seconds, match the kinetic characteristics of the resin curing process, and the compensation range of the target area (50×50mm 2 ) is aligned with the heating module partition of the multi-temperature zone press to ensure the regulation accuracy.

[0056] S44. After completing the dynamic compensation, perform secondary gradient annealing, set the annealing parameters as 100℃→60℃, the rate is 0.5℃ / min, detect the overall warpage degree through a laser interferometer, and output the laminated and formed circuit substrate.

[0057] The secondary gradient annealing (100℃→60℃, 0.5℃ / min) eliminates the local stress generated by the compensation through molecular chain relaxation. The laser interferometer (accuracy 0.1μm / m) detects the overall warpage degree, and its wavelength matches the surface roughness to avoid misjudgment. The alternating magnetic field (50Hz / 0.3T) induces dipole rearrangement to further reduce the residual stress, and finally the output substrate meets the flatness requirements of the high-frequency transmission layer.

[0058] In this embodiment, specifically, step S5 specifically includes: S51, Non-destructive detection of interlayer bonding. Use a tomography scanning system to perform three-dimensional tomography on the laminated substrate, and analyze the characteristics of the reflected wave through a deep learning algorithm to generate interlayer bonding data including porosity and delamination area; Use a tomography scanning system (0.1 - 3THz frequency band) to perform three-dimensional tomography on the laminated substrate, and utilize its penetrability and wavelength characteristics to match the thickness of the dielectric layer to achieve non-destructive detection. The deep learning algorithm (such as ResNet-50) automatically identifies micron-scale pores and delamination defects by analyzing the time-frequency characteristics of the reflected wave. While maintaining the integrity of the substrate, it accurately quantifies the interlayer bonding quality (porosity, delamination area), providing physical defect distribution data for subsequent performance evaluation.

[0059] S52, Construct a three-dimensional electro-thermal coupling model based on the results of three-dimensional tomography, load the operating conditions of the power amplifier module, calculate the current density distribution and thermal gradient field of the high-frequency signal transmission path, and output the predicted value of transmission loss; Construct an electromagnetic-thermal-stress multi-physics field coupling model based on the results of three-dimensional tomography, load the operating conditions of the power amplifier module, and simulate the skin effect and Joule heat distribution of the high-frequency signal transmission path. By calculating the influence of conductor surface roughness on signal loss, output the predicted value of transmission loss. This model maps physical defects (such as pores) to the attenuation of electro-thermal performance, realizing the correlation analysis from structural defects to functional failures, and guiding the direction of process optimization.

[0060] S53, Intelligent prediction of reliability. Input the interlayer bonding data and the predicted value of transmission loss into a pre-trained recurrent neural network, and combine the accelerated aging test data to predict the failure probability curve of the circuit board during the preset number of years of operation; The recurrent neural network (RNN) captures the co-degradation law of interlayer defects and transmission loss through time series modeling, and combines the accelerated aging data (temperature cycle -40°C to 150°C, damp heat 85°C / 85%RH) to extrapolate the long-term service performance. The temperature cycle simulates thermal stress fatigue, and the damp heat test accelerates the hygroscopic expansion effect. The two jointly construct a multi-failure mode database. The hidden state mechanism of the RNN effectively correlates short-term test data with long-term failure trends, and outputs three key indicators: delamination risk, impedance shift, and thermal fatigue life, quantifying and predicting the reliability of the preset time period.

[0061] S54, Process parameter optimization feedback. Use the decision tree algorithm to compare the life data in the historical process database according to the failure probability curve, and generate a process parameter optimization plan.

[0062] The decision tree algorithm mines the causal relationship between process parameters and failure probability by comparing the historical process database (including parameters such as temperature, pressure, and lamination offset and the corresponding life data). Set the minimum number of samples in the leaf nodes to prevent overfitting, and generate an optimization plan including temperature gradient adjustment and pressure compensation correction.

[0063] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A composite pressing process for a high multi-layer rigid-flex printed circuit board, characterized in that, It includes the following steps: S1. Substrate pretreatment stage: Microstructure processing is performed on the surface of the rigid substrate to form a three-dimensional strengthening interface, and at the same time, chemical grafting modification treatment is carried out on the flexible substrate to generate a multi-layer pretreated substrate; S2. Laminating configuration stage: The positioning marks of each layer of substrate are identified through an optical positioning system, a lamination stress prediction model is established in combination with material thermodynamics parameters, a lamination configuration scheme including the interlayer matching relationship is output, and a laminated structure is configured according to the lamination configuration scheme; S3. Gradient pressing stage: The configured laminated structure is placed in a multi-temperature zone press, and a three-stage gradient heating and pressing procedure is executed, including low-temperature pre-flow, medium-temperature main curing, and high-temperature shaping processes, to achieve interface fusion between adjacent hierarchical substrates; S4. Dynamic deformation regulation stage: The deformation difference between substrate layers is monitored in real time through a sensing network. When a preset deformation threshold is detected, a local temperature control module and a pressure compensation device are triggered to eliminate the deformation difference, and a pressed and formed circuit substrate is obtained.

2. The composite pressing process of the high multi-layer rigid-flex printed circuit board according to claim 1, wherein The specific steps of step S1 include: S11. Surface cleaning treatment of the rigid substrate: The surface of the rigid substrate is cleaned by an ultrasonic cleaning device, and then the residual particles on the surface are removed by an air knife to obtain a clean rigid substrate; S12. Use a laser to etch a wavy microgroove array on the surface of the rigid substrate, and simultaneously use a confocal microscope to monitor the etching morphology online to generate a three-dimensional anchoring interface with a mechanical interlocking function; S13. Place the flexible substrate into a vacuum plasma reaction chamber, introduce an Ar / O2 mixed gas with a volume ratio of 4:1, and process it at a set power for 120 seconds to obtain an activated flexible substrate with a target surface oxygen element content rate and surface energy; S14. Immerse the activated flexible substrate in an ethanol solution containing 3-aminopropyltriethoxysilane, and react it in a constant temperature bath at 80°C for 30 minutes to complete the grafting of amino functional groups and form a molecular bonding layer with a target grafting density; S15. Composite surface treatment: Spray a composite treatment solution containing nano-silica on the three-dimensional anchoring interface of the rigid substrate, and at the same time coat a polyimide-epoxy blend transition layer on the grafting layer of the flexible substrate. After pre-curing at 80°C, a gradient chemical bonding interface is formed.

3. The composite lamination process of the high multi-layer rigid-flex printed circuit board according to claim 2, characterized in that, The composite treatment solution contains 15% nano-silica and 2% silane coupling agent by mass fraction, and a stable suspension is formed after ultrasonic dispersion.

4. The composite pressing process of the high multi-layer rigid-flex printed circuit board according to claim 1, characterized in that, The specific steps of step S2 include: S21. Scan the positioning holes of each layer of substrate through an optical imaging system to generate a positioning data matrix including XY-axis coordinate deviation and rotation angle; S22. Collect the thermal expansion coefficient, elastic modulus, and dielectric constant parameters of each layer of substrate, and construct a digital twin model of the three-dimensional laminated structure in combination with the positioning data; S23. Apply a temperature-pressure coupling field to the digital twin model based on the finite element analysis algorithm, predict the stress distribution cloud map of the lamination interface, and mark potential deformation risk areas; S24. Based on the marked deformation risk areas, use the Monte Carlo algorithm to iteratively optimize the interlayer matching relationship, and output an optimal lamination configuration scheme including the hierarchical alternating order, interlayer offset compensation amount, and pre-deformation amount; S25. According to the optimal lamination configuration scheme, use a vacuum adsorption mechanism to grab the corresponding pre-treated substrate, and cooperate with a six-axis alignment platform to complete the physical stacking of the preset number of layers to form a laminated structure to be laminated.

5. The composite pressing process of the high multi-layer rigid-flex printed circuit board according to claim 1, wherein, The specific steps of step S3 include: S31. Transfer the configured laminated structure to the multi-temperature zone press cavity, and calibrate the position deviation between the laminate and the platen through the alignment system to ensure the XY-axis offset; S32. Set the partition temperature field, divide independent temperature control units according to the distribution of hard and soft substrates, set 120°C ± 2°C for the rigid area and 105°C ± 2°C for the flexible area to form a gradient thermal field distribution; S33. Low-temperature pre-flow process, load the laminated structure with an initial pressure of 0.8 MPa, heat it to 110°C at a rate of 1.5°C / min, and simultaneously increase the pressure to 1.2 MPa at a gradient of 0.05 MPa / min, and maintain it for 20 min to make the epoxy resin fill the three-dimensional anchoring interface; S34. Medium-temperature main curing process, switch to the dual-temperature zone mode: heat the rigid area to 135°C and the flexible area to 125°C, increase the pressure to 2.5 MPa and keep it for 30 min. During this period, monitor the resin curing degree in real time through an infrared thermal imager; S35. High-temperature shaping process, raise the overall temperature to 150°C, apply a high pressure of 3.0 MPa for shaping for 10 min, and at the same time turn on the ultrasonic interface monitoring system to detect the interlayer bonding state to ensure that the acoustic reflection coefficient is less than the preset coefficient value; S36. Gradient cooling control, cool down to 80°C at a rate of 0.8°C / min, simultaneously step down the pressure to 0.5 MPa at a rate of 0.1 MPa / min, and then introduce nitrogen to cool it down to room temperature to form a laminated substrate with interfacial fusion.

6. The composite lamination process of the high multi-layer rigid-flex printed circuit board according to claim 5, characterized in that, The multi-temperature zone press includes 16 independent heating modules, and the 16 heating modules are distributed in an array, with each heating module covering an area of N*Nmm 2 .

7. The composite pressing process of the high multi-layer rigid-flex printed circuit board according to claim 1, characterized in that, The specific steps of step S4 include: S41. Sample the laminated substrate, embed distributed fiber optic strain sensors between the layers of the laminated structure of the sampled laminated substrate, and install a piezoelectric film array at the same time to construct a sensing network for real-time monitoring of interlayer deformation; S42. Deformation threshold judgment, obtain the strain data of each layer in real time through a multi-channel data acquisition system. When the deformation difference between adjacent layers is greater than the preset deformation threshold, trigger a compensation instruction, and at the same time predict the deformation trend in the future preset time period based on the LSTM neural network; S43. For the non-sampled laminated substrate, locate the target area with excessive deformation according to the compensation instruction, apply local temperature control and piezoelectric ceramic pressure compensation to the target area, and the synchronous adjustment time does not exceed 30 seconds to perform dynamic compensation; S44. After completing the dynamic compensation, perform secondary gradient annealing, set the annealing parameters 100°C → 60°C, rate 0.5°C / min, detect the overall warpage through a laser interferometer, and output the laminated and formed circuit board.

8. The composite lamination process of the high multi-layer rigid-flex printed circuit board according to claim 1, characterized in that, After step S4, it further includes: S5. Quality verification stage, perform interlayer bonding scanning detection, electrothermal performance simulation modeling and reliability prediction on the laminated and formed substrate in sequence, and generate a quality assessment report including process parameter optimization suggestions.

9. The composite pressing process of the high multi-layer rigid-flex printed circuit board according to claim 8, characterized in that, The specific steps of step S5 include: S51. Interlayer bonding non-destructive testing: Use a tomography scanning system to perform three-dimensional tomography on the laminated substrate, analyze the characteristics of the reflected waves through a deep learning algorithm, and generate interlayer bonding data including porosity and delamination area. S52. Construct a three-dimensional electro-thermal coupling model based on the results of the three-dimensional tomography, load the operating conditions of the power amplifier module, calculate the current density distribution and thermal gradient field of the high-frequency signal transmission path, and output the predicted value of the transmission loss. S53. Reliability intelligent prediction: Input the interlayer bonding data and the predicted value of the transmission loss into a pre-trained recurrent neural network, and combine the accelerated aging test data to predict the failure probability curve of the circuit board during the preset number of years of operation. S54. Process parameter optimization feedback: According to the failure probability curve, use a decision tree algorithm to compare the life data in the historical process database and generate a process parameter optimization plan.

Images