Highly efficient heat dissipating circuit board and method of manufacturing the same

Abstract

The present application relates to a kind of high-efficiency heat dissipation circuit board and its manufacturing method, which constructs raw material performance database by the structuralization collection and evaluation of high thermal conductivity nano filler and resin matrix physical parameters, guides the self-adapting migration and positioning of nano filler along gradient by adjusting partition solvent evaporation rate and interfacial tension gradient, forms the partition ordered high-efficiency heat conduction filler network.Further combined with in-situ multimodal detection and automatic feedback optimization, the uniformity of filler dispersion and heat conduction channel continuity are guaranteed, and the preparation of circuit board with uniform heat dissipation performance is realized through step-by-step curing and locking structure of low temperature.

Description

Technical Field

[0001] This invention relates to the field of "high thermal conductivity nanocomposite material coating and circuit board high-efficiency thermal management manufacturing technology", and particularly to a high-efficiency heat dissipation circuit board and its manufacturing method. Background Technology

[0002] High thermal conductivity circuit boards (PCBs) are core components for heat dissipation management in high-performance electronic devices. Their thermal management capabilities largely depend on the thermal conductivity and uniformity of the high thermal conductivity nanocomposite coating applied to the substrate surface. Currently, the mainstream methods for preparing high thermal conductivity nanocomposite coatings in the industry mainly include direct mixing and dispersion, chemical grafting dispersion, template-assisted thermal channel construction, and layered or gradient composite coating. A common technical approach involves first mechanically mixing high thermal conductivity nanofillers (such as boron nitride, alumina, and graphene) with a resin matrix or organic polymer solution, supplemented by ultrasonic dispersion, surface coating, or interface modification to improve filler dispersibility. Then, traditional methods such as scraping, casting, and spraying are used to deposit the nanofillers onto the PCB substrate surface, followed by thermal / photocuring or cross-linking reactions to construct thermal channels. With the increasing demands for uniform thermal conductivity and heat dissipation consistency in terminal applications such as 5G and high-power modules, the industry has recently explored technologies such as interface biomimetic modification, in-situ polymerization, molecular self-assembly, and multi-component gradient composites, significantly expanding the theoretical and application boundaries of high thermal conductivity coatings.

[0003] Although existing technologies can improve the dispersion of nanofillers in polymer coatings on circuit boards to some extent, reducing single-point agglomeration and localized thermal resistance, they generally suffer from the following shortcomings and technical bottlenecks in mass production and large-scale manufacturing environments:

[0004] (1) Nanofillers are prone to agglomeration, and the heat conduction channels are easily blocked by nano agglomerates, resulting in discontinuous heat channel structure, local failure of heat dissipation effect, and failure to meet the requirement of uniform heat conduction.

[0005] (2) Relying solely on mechanical or ultrasonic dispersion makes it difficult to consistently ensure the uniformity of the initial distribution of fillers, resulting in large process fluctuations and low batch-to-batch consistency.

[0006] (3) Traditional layered gradient method or template method preparation has a complex process flow and insufficient flexibility, making it difficult to meet the needs of high-speed automation and multi-specification production lines. The process parameters are difficult to adapt to changes in the material system.

[0007] (4) The technology for continuous construction of intelligent, template-free heat conduction channel networks is not yet mature, and the ability to dynamically control the dispersion of fillers, detect in real time and optimize the process in a closed loop is insufficient, resulting in large fluctuations in the final thermal performance distribution of the product.

[0008] (5) The lack of automatic evaluation and feedback of the whole process of decentralized / heat conduction network makes it difficult to monitor, respond to and optimize potential problems such as blockage of heat dissipation channels in a timely manner, increasing failure rate and maintenance costs. Summary of the Invention

[0009] This application provides a high-efficiency heat dissipation circuit board and its manufacturing method, aiming to solve one of the problems or issues of the prior art mentioned in the background.

[0010] This application provides a method for manufacturing a high-efficiency heat dissipation circuit board, specifically including:

[0011] S1: Obtain high thermal conductivity nanofillers and resin matrix raw materials of different batches and specifications, and collect their physical performance parameters and batch labels in a structured manner to ensure that the material properties meet the requirements of heat dissipation circuit board manufacturing.

[0012] S2: Based on the collected physical performance parameters of the high thermal conductivity nanofiller and resin matrix, surfactants of different concentrations and types are prepared and surfactant-resin precursor liquid systems are prepared in groups according to the preset longitudinal or thickness direction concentration gradient rules to obtain surfactant gradient liquids with high material compatibility.

[0013] S3: High thermal conductivity nanofillers are added to various surfactant-resin precursor liquid systems. By controlling the parameters of the dispersion equipment (mechanical stirring or ultrasonic power and time), enhanced uniform initial dispersion treatment is performed to obtain a uniform coating of nanofillers in multi-gradient precursor liquids.

[0014] S4: The obtained multi-gradient surfactant-coated nanofiller system was applied to the substrate surface of different locations on the circuit board in different time periods or layers using multi-channel synchronous gradient coating technology. The corresponding substrate area, coating rate and thickness label were recorded to achieve ordered gradient coating deposition in multiple areas.

[0015] S5: Under constant humidity and temperature control, the solvent evaporation rate of each substrate area is adjusted. By stepwise heating or partitioned evacuation, the interfacial tension gradient is activated, causing the nanofillers already coated in the precursor liquid to undergo adaptive migration and orderly positioning in the high interfacial surfactant region, thereby realizing the self-assembly of the partitioned gradient dispersed thermally conductive filler network.

[0016] S6: Using non-contact methods such as infrared thermography, optical interference and fluorescent labeling, in-situ multi-mode detection is carried out on each substrate area to collect data on filler distribution uniformity, thermal conductivity channel continuity and interface structure consistency, and generate evaluation parameters for zoned filler dispersion and channel performance.

[0017] S7: Determine whether the uniformity of filler dispersion and the continuity evaluation parameters of heat conduction channels in each region meet the preset consistency and high-efficiency heat conduction threshold. If the threshold is abnormal, it will be fed back to the process control parameters such as the surfactant gradient parameter in S2 and the coating rate in S4, and automated optimization iteration will be performed until the technical indicators meet the requirements.

[0018] S8: After the coating is evenly dispersed in each substrate area and the continuity of the heat conduction channel meets the standard, a medium-low temperature stepwise curing process is used to heat-set and cross-link the coating in each area, lock the optimized gradient distribution structure of the nanofiller, and finally obtain a finished circuit board with uniform heat dissipation performance.

[0019] This application also provides a high-efficiency heat dissipation circuit board, manufactured using the above-described manufacturing method.

[0020] The high-efficiency heat dissipation circuit board and its manufacturing method provided in this application have the following beneficial effects:

[0021] (1) This invention breaks through the limitations of traditional uniform mixing and coating methods, which often result in the agglomeration and localized accumulation of nanofillers leading to the breakage of thermal conductive chains and the blockage of channels, by combining precise control of surfactant gradients, synchronous multi-channel gradient coating, environmental humidity and temperature control with dynamic partitioned solvent evaporation. Utilizing the interfacial tension gradient-induced mechanism, the invention achieves adaptive, continuous dispersion, and orderly directional migration of nanofillers, forming a highly continuous thermal conductive pathway network free from agglomeration. Quantitative testing shows that the spatial dispersion uniformity index and thermal conductive channel connectivity score of the nanofillers are both improved by more than 30%, significantly superior to traditional methods.

[0022] (2) Innovative multi-mode in-situ online detection, fluorescence three-dimensional tracking, and infrared thermography evaluation methods construct a closed-loop data system for the entire process, enabling real-time monitoring of filler dispersion and thermal conductivity channel status within the coating, and supporting zoned adaptive optimization. Compared with existing single physical mixing and static batch detection technologies, the overall heat dissipation performance differences between different regions are significantly reduced, and the thermal conductivity variation range of the final product is ≤±8% (compared to ±25% for conventional process errors). This achieves highly controllable and consistent performance transformation of high-efficiency heat dissipation circuit boards across large areas and multiple batches, effectively supporting the scalable and highly reliable thermal management requirements of high-power electronic devices.

[0023] (3) This invention employs a step-by-step, synchronous multi-channel automated coating and intelligent curing system, which greatly reduces process dependence and batch-to-batch human intervention, facilitating integration with existing automated circuit board production lines. The interface gradient construction and self-assembly processes involved are based on solvent evaporation / adaptive distribution of interfacial active materials, without relying on external physical fields (such as magnetic fields or electric fields) or complex template guidance, significantly simplifying the process and reducing operational risks. The production line has wide applicability, can flexibly adapt to different material systems and substrate geometries, and has excellent potential for mass production expansion.

[0024] (4) Traditional technologies cannot suppress the agglomeration of nanofillers and the breakpoints in the thermal conductivity chain from the perspective of the process itself. They rely on high-energy physical dispersion and the addition of excessive fillers at one time, which leads to the deterioration of the rheological properties of the system and an increased risk of coating cracking. This invention uses the surfactant gradient barrier and spontaneous interfacial tension field to drive the migration of fillers, which fundamentally solves the bottleneck of difficult elimination of nanofiller agglomeration and discontinuous channels. Under the same process, the maximum thermal conductivity filler volume fraction is increased by 10-20%, while the apparent rheology and film quality of the system remain excellent. Attached Figure Description

[0025] Appendix Figure 1 This is the main flowchart of a high-efficiency heat dissipation circuit board manufacturing method.

[0026] Appendix Figure 2 This is a sub-flowchart of a method for manufacturing a high-efficiency heat dissipation circuit board.

[0027] Appendix Figure 3 This is another sub-flowchart of a method for manufacturing a high-efficiency heat dissipation circuit board. Detailed Implementation

[0028] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0029] The following disclosure provides many different embodiments or examples for implementing different structures of the invention. To simplify the disclosure, specific examples of components and arrangements are described below. Of course, these are merely examples and are not intended to limit the invention. Furthermore, reference numerals and / or letters may be repeated in different examples; such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. In addition, examples of various specific processes and materials are provided in this invention, but those skilled in the art will recognize the application of other processes and the use of other materials.

[0030] As attached Figure 1 As shown, this application provides a method for manufacturing a high-efficiency heat dissipation circuit board, specifically including:

[0031] S1: Obtain high thermal conductivity nanofillers and resin matrix raw materials of different batches and specifications, and collect their physical performance parameters and batch labels in a structured manner to ensure that the material properties meet the requirements of heat dissipation circuit board manufacturing.

[0032] S2: Based on the collected physical performance parameters of the high thermal conductivity nanofiller and resin matrix, surfactants of different concentrations and types are prepared and surfactant-resin precursor liquid systems are prepared in groups according to the preset longitudinal or thickness direction concentration gradient rules to obtain surfactant gradient liquids with high material compatibility.

[0033] S3: High thermal conductivity nanofillers are added to various surfactant-resin precursor liquid systems. By controlling the parameters of the dispersion equipment (mechanical stirring or ultrasonic power and time), enhanced uniform initial dispersion treatment is performed to obtain a uniform coating of nanofillers in multi-gradient precursor liquids.

[0034] S4: The obtained multi-gradient surfactant-coated nanofiller system was applied to the substrate surface of different locations on the circuit board in different time periods or layers using multi-channel synchronous gradient coating technology. The corresponding substrate area, coating rate and thickness label were recorded to achieve ordered gradient coating deposition in multiple areas.

[0035] S5: Under constant humidity and temperature control, the solvent evaporation rate of each substrate area is adjusted. By stepwise heating or partitioned evacuation, the interfacial tension gradient is activated, causing the nanofillers already coated in the precursor liquid to undergo adaptive migration and orderly positioning in the high interfacial surfactant region, thereby realizing the self-assembly of the partitioned gradient dispersed thermally conductive filler network.

[0036] S6: Using non-contact methods such as infrared thermography, optical interference and fluorescent labeling, in-situ multi-mode detection is carried out on each substrate area to collect data on filler distribution uniformity, thermal conductivity channel continuity and interface structure consistency, and generate evaluation parameters for zoned filler dispersion and channel performance.

[0037] S7: Determine whether the uniformity of filler dispersion and the continuity evaluation parameters of heat conduction channels in each region meet the preset consistency and high-efficiency heat conduction threshold. If the threshold is abnormal, it will be fed back to the process control parameters such as the surfactant gradient parameter in S2 and the coating rate in S4, and automated optimization iteration will be performed until the technical indicators meet the requirements.

[0038] S8: After the coating is evenly dispersed in each substrate area and the continuity of the heat conduction channel meets the standard, a medium-low temperature stepwise curing process is used to heat-set and cross-link the coating in each area, lock the optimized gradient distribution structure of the nanofiller, and finally obtain a finished circuit board with uniform heat dissipation performance.

[0039] Step S1 involves obtaining high thermal conductivity nanofillers and resin matrix raw materials of different batches and specifications, and structurally collecting their respective physical performance parameters and batch labels to ensure that the material properties meet the requirements for heat dissipation circuit board fabrication. Specifically, this includes:

[0040] S1.1: Perform specification screening on the high thermal conductivity nanofiller raw materials to be used in the preparation of high-efficiency heat dissipation circuit boards. Based on physical performance parameters such as particle size distribution, electrical conductivity, and particle surface energy, batch identification and structured data acquisition of the physical performance parameters of high thermal conductivity nanofillers are achieved by using a nanoparticle size analyzer and a surface characteristic analyzer, so as to obtain a dataset of original parameters of high thermal conductivity nanofillers that can be used for subsequent processes.

[0041] The batch of high thermal conductivity nanofiller raw material is used as the input object, including the source manufacturer, preliminary particle size range, appearance and primary physicochemical parameter labels.

[0042] A high-precision nanoparticle size analyzer (based on laser scattering principle, with a measurement range of 10nm-10μm and a resolution better than ±2nm) was used to determine the particle size distribution of high thermal conductivity nanofillers, enabling physical detection of particle size uniformity and fine particle distribution between batches of raw materials.

[0043] Furthermore, the intrinsic conductivity of different batches of high thermal conductivity nanofillers was tested in batches using a high-sensitivity conductivity meter (four-electrode method, measurement range 10^-6 to 10^2 S / cm, resolution better than 0.01 S / cm) to obtain key thermal conductivity performance indicators.

[0044] The surface energy, hydrophilicity / hydrophobicity, and surface polarity components of each batch of high thermal conductivity nanofiller raw materials were characterized by a high surface energy automatic analyzer (microdroplet contact angle method, measurement range 10-150 mN / m, repeatability better than ±0.2 mN / m).

[0045] Based on the above physical performance parameter collection results, a batch physical performance parameter matrix of high thermal conductivity nanofillers is constructed:

[0046] ParameterMatrix = [(d i ,σ i ,γ i )] n

[0047] Where, d i Let σ be the average particle size of the i-th batch of nanofillers. i For its intrinsic conductivity, γ i Let n be the surface energy of its particles, and n be the number of collection batches.

[0048] Data is structured and a process labeling system is adopted that corresponds one-to-one with batch numbers and original test data to achieve parameter traceability.

[0049] The structured physical performance parameters and batch labels are input into the hierarchical database management system to generate a dataset of original parameters for high thermal conductivity nanofillers that can be used for process parameter scheduling and adaptability analysis.

[0050] Through the above steps and data flow, we can achieve efficient identification of the physical properties of batches of high thermal conductivity nanofiller raw materials, structuring of process parameters, data traceability, and provide a basis for subsequent raw material selection.

[0051] For example, during the collection of a batch of high thermal conductivity boron nitride (BN) nanofillers, batch number BN-A20240401 was collected. The particle size analyzer measured d = 158 nm, the conductivity test result σ = 8.5 × 10⁻² S / cm, and the surface energy analyzer measured γ = 36.7 mN / m. The following batch structured record was established for the above physical parameters: BN-A20240401–[158 nm, 8.5 × 10⁻² S / cm, 36.7 mN / m]. For the three parallel BN batches, the structured data was summarized into the following parameter dataset: [[158 nm, 8.5 × 10⁻², 36.7], [145 nm, 8.1 × 10⁻², 33.8], [170 nm, 8.8 × 10⁻², 40.2]]. In subsequent process matching and gradient formulation optimization, batches of nanofillers with specified particle size distribution and surface energy range can be screened to ensure the quantitative consistency of raw material performance and improve the inherent basis for the dispersion uniformity and thermal conductivity continuity of high thermal conductivity coatings.

[0052] S1.2: Chemical composition analysis and thermal analysis methods (such as FT-IR and DSC) are used to collect and process key physical properties such as molecular structure, thermal decomposition temperature and polarity parameters of the resin matrix raw materials suitable for circuit board substrates. This enables the structured correlation record between the molecular parameters of the resin matrix raw materials and batch labels, providing data output for the compatibility assessment of high thermal conductivity nanofillers and the design of coating precursor liquid system composition.

[0053] The input includes samples of suitable resin matrix raw materials for circuit board substrates obtained through supplier delivery or inventory purchase, resin raw material batch labels, preset chemical structure labels, and initial appearance parameters.

[0054] Fourier transform infrared spectroscopy (FT-IR, wavenumber range 4000-400 cm⁻¹) was used for analysis. -1 Resolution better than 2cm -1 The molecular structure characteristic spectrum of the resin matrix raw material is collected, and the main characteristic absorption peaks (such as functional groups such as hydroxyl, carboxyl, epoxy, and amino groups) are marked to achieve qualitative and quantitative differentiation of the molecular backbone type and functional group composition, and generate a set of characteristic parameters of the resin matrix molecular structure.

[0055] Furthermore, the thermal decomposition temperature (T0) of the resin matrix raw material was obtained by differential scanning calorimetry (DSC, heating rate 10 K / min, test temperature range -50℃ to 300℃). d ), glass transition temperature (T)g Thermophysical parameters such as T are used. Based on the DSC output curve, the linear fitting method is used to accurately extract T. g With T d The data was collected and recorded as an indicator for analyzing the thermal stability of high thermal conductivity coatings.

[0056] The polarity parameters (molecular dipole moment, surface polarization energy, etc.) of the resin raw material were calculated using an electrochemical analyzer (automatic polarization curve scanning) to obtain the material polarity parameter vector P. r And establish an intrinsic connection with molecular structure data.

[0057] Based on the above FT-IR, DSC, and polarity analysis results, a multidimensional parameter structured algorithm was used to perform feature encoding, resulting in the following resin matrix performance parameter vector:

[0058] ResinParam j =[S j ,T g,j ,T d,j ,P r,j ]

[0059] Where S j T is the molecular structure code for the j-th batch of resin raw materials. g,j T is the glass transition temperature. d,j P is the thermal decomposition temperature. r,j This is a polarity parameter.

[0060] Based on the above parameter vector, a structured association table of resin matrix raw material parameters and batch labels is established by using a multi-dimensional one-to-one correspondence index of batch number, product brand and supply source.

[0061] Through data integration and database writing, standardized output of resin matrix raw material parameters is generated that can be directly called upon in subsequent process steps such as high thermal conductivity nanofiller compatibility modeling and gradient precursor liquid design.

[0062] Through multi-level parameter analysis and structured data processing, the molecular structure and thermophysical properties are transformed into a database of molecular parameters for resin matrix raw materials, enabling closed-loop output of basic data for high thermal conductivity nanofiller matching evaluation and high-performance coating system.

[0063] For example, FT-IR analysis was performed on a sample of epoxy resin with batch number EP-20240401, with a detection wavenumber of 3450 cm⁻¹. -1 A hydroxyl absorption peak was detected at 1600 cm⁻¹. -1 The peaks at this location indicate a characteristic aromatic ring skeleton, and the molecular structure is qualitatively determined to be a substituted aromatic cyclic epoxy group. DSC analysis yielded T... g =113℃, T d =298℃. The polarity parameter P was measured using a polarization energy analyzer.r =6.3±0.2Debye. After multidimensional parameter structuring, the molecular parameter record for this batch was generated: [S='1,3-dihydroxybenzoepoxide',T g =113℃,T d =298℃, P r =6.3Debye], associated with batch label EP-20240401. For multiple batches of EP series resin raw materials of the same type, the batch number, parameter vector and molecular structure are collected in a structured manner to form a standard database format: [['EP-20240401','1,3-dihydroxybenzo[epoxy]',113,298,6.3],['EP-20230416','linear aliphatic epoxy',96,265,4.1],…]. The above parameter set is directly used as input for the high thermal conductivity nanofiller-resin matrix compatibility model, realizing dynamic screening of material adaptability and process compatibility of multiple batches and multiple sources of resin raw materials, significantly improving the targeting of subsequent precursor liquid gradient system component design and the consistency of thermally conductive coating performance.

[0064] S1.3: Based on the collected high thermal conductivity nanofiller parameter dataset and resin matrix molecular parameter data, material compatibility modeling algorithms (such as interface energy matching evaluation model and dispersion stability prediction algorithm) are applied to score and label the interface compatibility between the two types of raw materials, forming a high thermal conductivity nanofiller-resin matrix raw material compatibility label, which provides modeling input for the parameter configuration of surfactant type and concentration.

[0065] The structured database of high thermal conductivity nanofiller parameter dataset and resin matrix molecular parameter data is used as input objects. It includes the particle size, electrical conductivity, and surface energy parameter matrix of each batch of nanofillers, as well as the molecular structure, thermal decomposition temperature, glass transition temperature, and polarity parameters of the resin matrix.

[0066] An interface-matching evaluation model is adopted (parameter input: particle size d). i Surface energy γ i Resin polarity parameter P r,j Molecular structure encoding S j This allows for scoring of the interfacial thermodynamic compatibility between highly thermally conductive nanofillers and resin matrices.

[0067] Furthermore, by employing a dispersion stability prediction algorithm (inputs: surface energy of the high thermal conductivity nanofiller, polarity and viscosity parameters of the resin matrix, and affinity constants of both), the aggregation risk of the nanofiller within the resin system is quantified, and the dispersion stability index (DSI) is obtained. ij .

[0068] Using the algorithmic steps, the following scoring process is performed on the batch pairs [i,j] that the interface can match for evaluation:

[0069]

[0070] Where γ i For the surface energy of the filler, It represents the surface energy or polarity index of the resin, and is used to form a score for the interfacial energy differences between batches.

[0071] Further, the following derivation of dispersion stability is adopted:

[0072] DSI ij =f(γ) i ,P r,j ,η j )

[0073] Where η j Given the resin viscosity, the dispersion index of the filler-resin system is calculated based on the superimposed weights of diffusion theory and DLVO theory.

[0074] Combining the above two steps, a performance-weighted aggregation tag algorithm is used to generate material compatibility tags for input to the gradient surfactant configuration model. ij The logical relationship is as follows:

[0075] Tag ij =F(C ij DSI ij ,S j ,d i ,σ i )

[0076] Wherein, F is a logical mapping function that integrates interfacial energy, dispersion index, structural parameters and filler performance to achieve batch-to-batch material compatibility labeling.

[0077] Using the aforementioned material compatibility modeling algorithm, a multi-to-one multidimensional compatibility matrix is ​​formed between batches of high thermal conductivity nanofillers and resin matrices, and a structured compatibility tag library is output as the modeling input for surfactant type and concentration allocation, thereby achieving automated recommendation of downstream formulation parameters and optimization of compatibility.

[0078] For example, a dataset containing three sets of highly thermally conductive boron nitride nanofillers (BN-A, BN-B, BN-C) and two sets of epoxy resin matrices (EP-X, EP-Y) is given. The surface energies are known to be 36.7 mN / m for BN-A, 33.8 mN / m for BN-B, and 40.2 mN / m for BN-C. The polarity parameter of EP-X is 6.3 Debye, and that of EP-Y is 4.1 Debye. The interfacial energy matching evaluation model uses the following formula:

[0079]

[0080] Substituting BN-A and EP-X, we get C. A,X =1 / (1+|36.7-6.3|)≈0.027, similarly calculate pairs such as BN-B-EP-X, and get C. B,X and C C,X .

[0081] Dispersion stability prediction uses DSI ij =γ i / P r,j , BN-A-EP-X gets DSI A,X =36.7 / 6.3≈5.83. Generate a suitable tag such as: Tag_{A,X}=[0.027,5.83,structure1,158,8.5E-2]. Following the above process, complete all batch combinations to form a tag matrix, which serves as the automatic optimization input for subsequent surfactant grouping and concentration prediction, achieving optimal parameter configuration of the coating system for specified material properties and consistent output of high thermal conductivity dispersion.

[0082] S1.4: In accordance with the batch management system, the structured performance parameters and compatibility labels of high thermal conductivity nanofillers and resin matrix raw materials are generated through the raw material batch label generation process. The entire process is tracked and identified based on the traceability management platform to output a raw material batch performance database that is compatible with the high-efficiency manufacturing requirements of heat dissipation circuit boards.

[0083] S1.5: Perform field consistency verification and data deduplication on the structured record database of high thermal conductivity nanofiller-resin matrix raw material properties and batch label library. Utilize the database management system verification algorithm to eliminate abnormal or redundant raw material data, and finally output big data on raw material properties through multi-level structured integration, providing input support for subsequent surfactant gradient configuration and initial dispersion process design.

[0084] Step S2: Based on the collected physical performance parameters of the high thermal conductivity nanofiller and resin matrix, surfactants of different concentrations and types are prepared, and surfactant-resin precursor liquid systems are prepared in groups according to a preset longitudinal or thickness direction concentration gradient to obtain surfactant gradient liquids with high material compatibility. Specifically, this includes:

[0085] S2.1: Perform data structured analysis on the physical property parameters of the high thermal conductivity nanofiller and the physical property parameters of the resin matrix, and use the material compatibility analysis algorithm to generate the distribution model parameters required for the target gradient formulation, so as to obtain the surfactant distribution group suitable for different material systems.

[0086] S2.2: Based on the surfactant distribution groups obtained from the material compatibility analysis algorithm, surfactant types with compatibility and synergistic effects (such as hydrophilic and hydrophobic coupling agents) are selected. High-precision ratio control technology is used to generate a primary ratio dataset of surfactants with different types of synergistic effects, laying a synergistic foundation for subsequent concentration gradient construction.

[0087] Using the distribution group data of surfactants obtained from the material compatibility analysis algorithm as input, the surfactant candidate library is collaboratively screened using type screening rules. Hydrophilic and hydrophobic coupling agents with high compatibility with high thermal conductivity nanofillers and resin matrices are selected to form a basic surfactant type set.

[0088] The synergistic effect evaluation method (parameters: surface energy matching factor, dispersion stability index, synergistic interface energy) is adopted to analyze the synergistic effect of the above-mentioned basic surfactant types in pairs or multiple groups, so as to achieve a quantitative evaluation of the interfacial binding ability, dispersion medium stability and interfacial heat transfer channel compensation effect of different surfactant combinations in nanofiller-resin composite systems.

[0089] Furthermore, through high-precision proportioning control technology (parameters: surfactant solubility, miscibility, and expected distribution model proportioning coefficient), the primary proportioning parameters of the surfactant type schemes screened through collaborative analysis are set according to material compatibility labels and target distribution models, forming a primary proportioning dataset of multiple types of surfactants.

[0090] The numerical simulation method of surfactant synergy was adopted. The primary ratio parameters of surfactant were input into the synergistic distribution prediction model to calculate the effective binding ratio and dispersion stability of each type of surfactant with nanofiller, resin matrix and composite interface, and obtain multiple sets of surfactant-filler-resin composite interface performance synergistic score matrices.

[0091] The synergy effect index is calculated using the following formula:

[0092] SEI k =α k C f,k +β k C r,k +γ k C fr,k

[0093] Among them, SEI k C represents the synergistic effect index of the k-th type of synergistic surfactant combination. f,k C represents the score for the effect of packing dispersion. r,k C represents the score for the effect of resin polarity. fr,k α represents the collaborative score for interface integration. k β k γk These are the weighting coefficients.

[0094] Furthermore, a multi-objective optimization algorithm (parameters: dispersion uniformity index, interfacial energy reduction rate, and thermal resistance minimization objective parameter) is used to perform multiple rounds of trade-off calculations on the SEI index of all combinations within the primary formulation dataset, outputting the optimal or suboptimal primary formulation dataset of multiple types of synergistic surfactants. This dataset contains the type number and primary formulation value of each surfactant under each fitted action layer (such as a hydrophilic-hydrophobic bipolar interface).

[0095] Through the above-mentioned chain-like collaborative screening, proportioning, collaborative numerical simulation and multi-objective optimization, a primary proportioning dataset of synergistic activators that meets the requirements of material compatibility, interface control and subsequent concentration gradient distribution is obtained, which precisely empowers the concentration gradient construction process and provides basic technical support for the efficient dispersion and interface coupling of nanofillers.

[0096] For example, for the master adaptor material system of boron nitride nanofiller with batch number BN-A20240401 and epoxy resin EP-20240401, the material compatibility analysis algorithm was used to identify two types of coupling agents that need to cooperate synchronously: hydrophilic silane (APTES) and hydrophobic silane (ODTES). A synergistic effect evaluation was used to obtain the synergistic index C between APTES and ODTES. f,1 =0.82, C r,1 =0.69, C fr,1 =0.74, weighting coefficients α1=0.4, β1=0.3, γ1=0.3, substituting into the synergy effect index formula:

[0097] SEI1=0.4×0.82+0.3×0.69+0.3×0.74=0.328+0.207+0.222=0.757

[0098] The synergistic effect index was calculated for other candidate synergistic combinations using the same method, and the primary formulation combination with an SEI > 0.75 was selected as the output. A high-precision formulation control system was used, setting the mass ratio of APTES to ODTES at 5:3 based on the material system and target gradient. After multi-objective optimization, this ratio was adjusted to 4.5:3.5 to maximize distribution uniformity. Finally, a primary formulation dataset of synergistic surfactants, including the synergistic index, specific formulation parameters, and type labels, was generated. In practical application testing, this component configuration, used for a synergistic anti-agglomeration coating system for uniform dispersion of nanofillers, achieved a 12% improvement in the uniformity of the subsequent concentration gradient distribution and an 18% reduction in interfacial thermal resistance, providing support for meeting the consistent design requirements of efficient heat dissipation processes for circuit boards.

[0099] S2.3: Using the primary surfactant ratio dataset as input, the concentration gradient design and parameter optimization model is applied to implement the partition ratio according to the preset gradient distribution law in the longitudinal or thickness direction, thereby obtaining multiple sets of surfactant concentration gradient parameters, and generating a unique surfactant concentration coordinate for each specified gradient layer.

[0100] S2.4: Input multiple sets of surfactant concentration gradient parameters into the resin precursor liquid precision mixing control system to perform highly uniform micro-mixing treatment, forming a structured and layered surfactant-resin precursor liquid system, and achieving highly controllable multi-level gradient spatial concentration.

[0101] S2.5: High-resolution component monitoring and gradient consistency detection are performed on the surfactant-resin precursor liquid system. The gradient distribution uniformity evaluation algorithm is applied to obtain the distribution consistency parameters in real time, ensuring that each group of surfactant-resin precursor liquids meets the preset concentration gradient and material compatibility output requirements, thus providing a guarantee for the downstream nanofiller dispersion treatment.

[0102] Step S3: High thermal conductivity nanofillers are added to various surfactant-resin precursor liquid systems, and enhanced uniform initial dispersion treatment is performed by controlling the parameters of the dispersion equipment (mechanical stirring or ultrasonic power and time) to obtain a uniformly coated nanofiller in the multi-gradient precursor liquid. Figure 2 As shown, it specifically includes:

[0103] S3.1: The configured multi-gradient surfactant-resin precursor liquid system is numbered and labeled. According to the preset surfactant concentration gradient labels, the batch raw materials of high thermal conductivity nanofiller, which are the input objects, are measured separately. The batch is prepared by using an automatic weighing system to obtain the initial mixing mixture of various surfactant-resin precursor liquid systems and high thermal conductivity nanofiller, so as to ensure the compatibility of the precursor liquid and the initial consistency of the filler distribution.

[0104] The initial inputs are a multi-gradient surfactant-resin precursor liquid system with pre-structured proportions and groupings, high thermal conductivity nanofiller raw materials in the batch management system, and a matching automatic weighing platform.

[0105] An algorithm for identifying surfactant concentration gradient labels (parameters: batch number, group gradient, target concentration range) is used to perform full numbering and coding on multiple surfactant-resin precursor liquid systems, thereby realizing a structured association between gradient formulations and physical groupings.

[0106] Furthermore, through the batch label-driven automatic feeding control system for high thermal conductivity nanofillers (parameters: raw material batch, target added mass, metering accuracy threshold), the system automatically selects and adds matching batches of high thermal conductivity nanofiller raw materials based on the number of each type of precursor liquid system and the corresponding target mass ratio of the filler, thereby obtaining the initial addition amount of high thermal conductivity nanofiller under each gradient group.

[0107] High-precision automatic weighing technology (parameters: single weighing error ≤0.1%, minimum resolution 0.1mg) is adopted to perform real-time batch detection and addition result correction of high thermal conductivity nanofillers in the sub-step process. The difference between the target filler addition amount and the actual weighing value is verified through digital closed-loop feedback loop, and multi-gradient preliminary batching qualified labels are generated for each batch.

[0108] By using a mixing ratio consistency algorithm (parameters: surfactant concentration distribution, target filler mass ratio), the compatibility of the surfactant-resin precursor liquid system and the mixture of high thermal conductivity nanofillers is verified, enabling the detection and evaluation of the initial distribution uniformity and original ingredient ratio under different gradient systems.

[0109] Furthermore, by adopting a traceable link coding system for raw material distribution, each group of qualified batches of ingredients is numbered and stored in the database, forming a preliminary batching mixture list of surfactant-resin-high thermal conductivity nanofiller that can be used for subsequent enhanced dispersion and refined coating treatments.

[0110] By integrating structured batching and automatic weighing control, the surfactant-resin precursor liquid system with different concentration gradient groups is efficiently matched with the corresponding high thermal conductivity nanofiller, achieving batch compatibility, precise proportioning and initial uniform distribution of the mixture, significantly improving the subsequent dispersion uniformity, and providing a reliable raw material basis for optimizing the high-efficiency thermally conductive coating process.

[0111] For example, for three groups of graded surfactant-resin precursor liquid systems (numbered G1, G2, and G3), according to the previous zoning allocation plan, the target filler addition mass is set to 1.50g for group G1, 1.80g for group G2, and 2.00g for group G3. All groups use hexagonal boron nitride nanopowder with batch number BN-A20240401. The automatic weighing system is set to a maximum permissible error of ±0.003g per weighing. The control platform automatically retrieves, measures, and adds BN-A20240401 raw material based on the surfactant graded labels of each group. Online weighing results show that G1 weighs 1.501g (qualified), G2 weighs 1.799g (qualified), and G3 weighs 2.004g (qualified), all within the acceptable error range. All weighed materials were labeled into batch ingredient lists G1-MIX-20240401, G2-MIX-20240401, and G3-MIX-20240401, respectively. Through consistency testing of the ingredient ratios, it was confirmed that the initial dispersion index of each mixture group was >0.98 (out of 1), and the overall proportioning error was <0.2%. Finally, three preliminary batch mixtures with clearly defined quantitative batches, concentration gradients, and consistent filler ratios were output. This provides highly consistent and traceable raw material base batches for downstream mechanical stirring and ultrasonic-enhanced dispersion processes, ensuring quality control and physical compatibility at the starting point of the gradient grouping.

[0112] S3.2: Based on the interfacial physical properties of high thermal conductivity nanofiller particles and the type parameters of surfactants, the multi-gradient precursor liquid mixture obtained in the first step is premixed using mechanical stirring. The viscosity and surface tension parameters of the slurry are collected online in real time to form a pre-dispersion liquid of surfactant-resin-high thermal conductivity nanofiller with preliminary dispersion, providing rheological data input for subsequent ultrasonic dispersion optimization.

[0113] The experiment uses a preliminary batching mixture of multi-gradient surfactants, resins, and high thermal conductivity nanofillers that has been numbered and confirmed in batches. The input parameters include batch labels of each component, surfactant concentration gradient numbers, target system component ratios, and physical properties of the high thermal conductivity nanofiller particle interface and surfactant type parameters.

[0114] A high-precision mechanical stirring method (parameters: stirring speed range 300-1500 rpm, time 3-30 min, temperature control range 20-35℃, and adaptive matching of stirring blade shape and container size) was adopted to perform uniform premixing treatment on each group of preliminary batching mixtures, making full use of shear and convection to achieve preliminary dispersion of nanofillers in different gradient precursor liquid systems.

[0115] Furthermore, a real-time rheological parameter acquisition module integrated with online sensors (parameters: slurry viscosity measurement range 1–10000 mPa·s, sampling frequency ≥1 Hz, error ≤2%) is used to monitor the viscosity changes of the stirred slurry and obtain real-time characterization data of the uniformity of nanofiller distribution under mechanical stirring. The following viscosity uniformity quantification algorithm is used:

[0116]

[0117] Among them, U γ σ represents the viscosity uniformity index. η The standard deviation of online viscosity at multiple points within a batch. This represents the average viscosity of the batch.

[0118] Furthermore, an online surface tension measuring device (parameters: interfacial tension resolution 0.01 mN / m, automatic calibration function) is applied to synchronously detect the surface tension change process of the slurry, and to obtain the distribution response of the surfactant and the filler coating level in real time during the stirring process.

[0119] Surface tension uniformity evaluation algorithm is adopted:

[0120]

[0121] Among them, U γ σ is an index of surface tension uniformity. γ To measure the standard deviation of tension values ​​within the measurement cycle, The average surface tension is denoted as .

[0122] Based on the above-mentioned online acquisition of viscosity and surface tension parameters and the quantification of uniformity, dynamic discrimination of the dispersion state during the mechanical stirring stage is achieved.

[0123] The dispersion threshold determination rule (parameter: viscosity uniformity U) is used. γ >0.90, surface tension uniformity U γ >0.90 is considered acceptable), outputting the batches that meet the dispersion standards of the mechanical stirring process and the structured parameters of the pre-dispersion liquid, as the rheological basis data input for the subsequent ultrasonic dispersion and refinement stages.

[0124] By integrating mechanical stirring with online rheology / interface detection algorithms, the initial uniform dispersion of each group of multi-gradient precursor liquid mixtures and the process confirmation of the surface coating state of nanofillers are achieved. This effectively improves the controllability of subsequent ultrasonic dispersion efficiency and dispersion uniformity, and provides initial assurance for preventing nanofiller agglomeration and ensuring the continuity of heat conduction channels.

[0125] For example, for the medium-gradient surfactant-resin-boron nitride nanopowder system with serial number G2-MIX-20240401, the mechanical stirring speed was set to 1200 rpm, the mixing time to 20 minutes, and the ambient temperature to 25°C. A real-time online rheological monitoring device detected an initial viscosity of 3500 mPa·s, which decreased to 3400 mPa·s by the 15th minute. The viscosity fluctuation at 20 minutes was ≤12 mPa·s. The standard deviation σ was calculated. η =9.8, average viscosity Based on this viscosity uniformity U η =1 - 9.8 / 3405 ≈ 0.9971. The surface tension measurement and control module recorded an average value of 27.5 mN / m, with a fluctuation standard deviation σ. γ =0.12, then U γ =1-0.12 / 27.5≈0.9956, both indicators are significantly higher than the 0.90 threshold. The pre-dispersion liquid of batch G2-MIX-20240401 meets the standard, and the output structured parameters include batch number, process viscosity uniformity, surface tension uniformity, and actual physical parameter curves, which are used as inputs for subsequent ultrasonic dispersion. The application results show that after subsequent ultrasonic reprocessing, the particle size distribution of the nanofiller in this batch converged from the initial D90=215nm to D90=92nm, and the dispersion uniformity improved by 57%, which significantly suppressed the agglomeration phenomenon and laid the foundation for the continuous structure of the heat conduction path of the circuit board.

[0126] S3.3: Using the surfactant-resin-high thermal conductivity nanofiller pre-dispersion obtained in S3.2 as input, ultrasonic dispersion technology (setting specific ultrasonic power / time) is used to perform secondary dispersion treatment on the high thermal conductivity nanofiller particles in the slurry, which promotes the refinement and uniformity of the nanofiller particle size in the multi-gradient system, improves the uniformity of the surfactant coating reaction on the filler surface, and obtains an ultrasonically enhanced high thermal conductivity nanofiller dispersion with uniform particle size distribution.

[0127] S3.4: Samples of the ultrasonically enhanced high thermal conductivity nanofiller dispersion obtained in S3.3 are taken and the particle size distribution and surfactant coating status are detected by dynamic light scattering (DLS) or scanning electron microscopy (SEM). Based on the real-time monitoring data, the dispersion uniformity parameters and interfacial binding effect are calculated, and abnormal or unqualified batches are screened to provide data support for closed-loop feedback of the dispersion process.

[0128] S3.5: The ultrasonically enhanced high thermal conductivity nanofiller dispersions that have passed the S3.4 inspection are grouped and archived according to their respective surfactant concentration gradients. A standard surfactant-resin-high thermal conductivity nanofiller uniform coating configuration list is prepared for use in subsequent multi-channel synchronous gradient coating processes, so as to realize the structured transfer of dispersion uniformity and gradient parameters throughout the entire process.

[0129] Step S4: The obtained multi-gradient surfactant-coated nanofiller system is applied to the substrate surface of different locations on the circuit board at different times or in layers using multi-channel synchronous gradient coating technology. The corresponding substrate area, coating rate, and thickness label are recorded to achieve ordered gradient coating deposition in multiple areas. Figure 3 As shown, it specifically includes:

[0130] S4.1: Using a multi-gradient surfactant-coated nanofiller system as the initial input, and based on the spatial partitioning labels on the surface of the circuit board substrate, a multi-channel synchronous coating equipment is invoked to perform a preset partitioned thermally conductive filler system separation and loading operation to obtain a uniformly distributed raw material pool at the coating station, providing a partitioned gradient composition guarantee for subsequent orderly deposition processes.

[0131] The initial input object is a uniformly coated body of multi-gradient surfactant-resin-high thermal conductivity nanofiller that has been grouped and archived. The input parameters include batch labels of raw materials, gradient numbers, surfactant types and concentration distribution information for each group.

[0132] A spatial partitioning recognition algorithm (parameters: circuit board substrate area division scheme, target thermal conductivity threshold, partition coordinate label) is adopted to realize standardized modeling of spatial partitions on the surface of the circuit board substrate, and output clear physical boundaries of each partition and mapping labels within the partition.

[0133] The automatic scheduling module of the multi-channel synchronous coating equipment (parameters: number of channels, loading capacity, gradient group number) performs preset partition material guiding path planning for each partition of the thermally conductive filler system, separates and loads the corresponding gradient raw material group with a unique gradient number, and constructs a multi-channel raw material pool with consistent partition distribution.

[0134] Furthermore, a raw material pool consistency detection and partition ratio verification algorithm (parameters: batch number, gradient number, consistency criterion) is adopted to detect the gradient parameters, viscosity, and component consistency of the raw material pool in each coating channel / station in real time, generating a set of partition raw material pool consistency evaluation indicators to realize structured process control of each partition gradient assembly process.

[0135] An automatic material traceability and zone label binding mechanism is adopted to digitally identify the raw material pool number, raw material batch number and gradient level of each zone, generate a raw material pool scheduling list corresponding to each spatial zone and archive it, so as to provide data support for subsequent multi-channel orderly coating and parameter traceability.

[0136] Through the above chain process, the multi-gradient coatings that are grouped and archived are accurately allocated to the independent raw material pools of each spatial partition of the circuit board, laying the foundation for spatial gradient control and orderly construction of heat conduction channels in the downstream synchronous multi-channel coating process, and realizing the consistency and traceability of the material partition gradient structure.

[0137] For example, for a circuit board with a coverage area of ​​200×150mm, the surface is divided into three thermal management regions: A, B, and C, based on heat dissipation simulation and functional requirements. Region A (high load) is assigned gradient I (high surfactant concentration, numbered G1), region B (medium load) is assigned gradient II (medium surfactant concentration, numbered G2), and region C (low load) is assigned gradient III (low surfactant concentration, numbered G3).

[0138] After the spatial partitioning identification system divides the circuit board surface, the multi-channel synchronous coating equipment schedules three independent feed tanks, automatically identifies the gradient number and loads three sets of raw material pools, G1, G2 and G3, which correspond to areas A, B and C respectively.

[0139] The raw material pool consistency testing module performed gradient parameter testing on the three groups of raw material pools. The measured values ​​were: G1 group viscosity 4200 mPa·s, surfactant concentration 8.0%, and nanofiller volume fraction 32%; G2 group viscosity 3500 mPa·s, concentration 6.0%, and volume fraction 28%; and G3 group viscosity 2950 mPa·s, concentration 4.0%, and volume fraction 22%. Viscosity fluctuation and component characteristics were all below the 1% threshold, and the consistency evaluation index was >0.98, meeting the requirements for zoned assembly.

[0140] The automatic tag binding system establishes a one-to-one correspondence between area A and pool G1, area B and pool G2, and area C and pool G3, and archives the scheduling tags A-G1-R20240415, B-G2-R20240415, and C-G3-R20240415 in the production database.

[0141] The final output consists of three raw material pools with clearly defined numbers, consistent gradient parameters, and unique spatial partitions, providing technical support for subsequent multi-channel synchronous gradient coating to achieve precise spatial partition loading, data traceability, and efficient pre-construction of thermally conductive gradient structures.

[0142] S4.2: For the raw material pool of the coating station after multi-channel separation and loading, the gradient structure parameters such as the concentration of surfactant, gradient type, and physical properties of precursor liquid of each station are set through the gradient parameter control module. The partition parameter matrix is ​​generated to ensure that the subsequent deposition process strictly follows the preset gradient distribution law in the spatiotemporal distribution dimension, forming an orderly gradient structure configuration in each region.

[0143] S4.3: Using the generated partition parameter matrix as input, the corresponding partition gradient raw material precursor liquid is applied to the substrate surface using a synchronous multi-channel coating process. The coating head is intermittently switched to achieve multi-layer gradient deposition in different time periods, or a layered coating method is used to achieve gradient structure stacking in the thickness direction, so as to obtain the initial structure of the gradient coating with layered or partitioned distribution in physical space.

[0144] Using the partition parameter matrix set by the gradient parameter control module as input, the numbering data of the multi-gradient surfactant-resin-high thermal conductivity nanofiller dispersion precursor liquid corresponding to each region is imported into the synchronous multi-channel coating process automatic control platform. The parameters include partition number, gradient group number, surfactant concentration, raw material batch and target coating thickness.

[0145] The synchronous multi-channel coating equipment (parameters: number of independent channels ≥3, channel switching response <200ms, equipped with a high-precision servo coating head) is used to plan and distribute the multi-gradient raw material precursor liquid for each zone according to the zone parameter matrix. It automatically adjusts each coating head to the specified substrate zone and physical coordinates, and applies the predetermined gradient group of precursor liquid to the zone respectively.

[0146] Furthermore, through the intermittent switching dynamic switching control algorithm of the coating head (parameters: timing switching rate, coating head switching threshold, and region boundary coordinates), the automatic application of different gradient precursor liquids in the same region at multiple time periods and in multiple batches layer by layer is realized, ensuring the continuous gradient transition between physical space segments and forming a multi-layered gradient deposition structure in different time periods.

[0147] By using a layered coating method and an automatic stacking algorithm in the thickness direction (parameters: total thickness target, single layer thickness setting, interlayer interval time), precursor liquids of different gradient concentrations are sequentially stacked and applied to the substrate surface. The thickness of each gradient layer and the stacking order are controlled to achieve an initial structure with a gradual change in the thickness direction.

[0148] Furthermore, by utilizing online coating width and thickness detection sensors (parameters: thickness detection accuracy ±0.5μm, width detection accuracy ±0.1mm), the deviation between the actual and target values ​​of the width and thickness of each gradient coating layer is monitored in real time. When abnormal locations are detected, the coating head output is automatically adjusted immediately, and the closed loop maintains the consistency of deposition in each zone / layer.

[0149] The automatic archiving module of the spatiotemporal distribution deposition database binds the actual process parameters of each zone / layer deposition (such as coating start / end time, switching sequence, concentration gradient number, raw material batch, thickness, etc.) to the structured output, and fully maps the preliminary structure of the multi-zone / multi-layer gradient coating in physical space.

[0150] Through the above multi-level chain processing, an initial gradient coating structure with controllable thickness direction, orderly partition distribution, and traceable gradient parameters and batches covering the entire space of the substrate is obtained, laying a key physical foundation for subsequent adaptive migration and the continuity of the final heat conduction channel.

[0151] For example, on a printed circuit board measuring 200mm × 150mm, region A is designated with a high surfactant concentration coating (G1, concentration 8%, target total thickness 80μm, 4 layers, 20μm each), region B with a medium concentration (G2, concentration 6%, total thickness 60μm, 3 layers), and region C with a low concentration (G3, concentration 4%, total thickness 40μm, 2 layers). A synchronous multi-channel servo coating machine automatically performs multi-layer intermittent switching coating on regions A, B, and C sequentially, with a 3-minute temperature and humidity controlled resting period between each layer. Throughout the process, coating thickness sensors monitor the thickness of each layer: region A G1 is 20.2±0.3μm, region B G2 is 20.0±0.2μm, and region C G3 is 19.8±0.4μm. The uniformity of the number of layers and thickness across all regions is better than 2%. The data archiving module generates labels with the numbers A-G1-4L-20240415, B-G2-3L-20240415, and C-G3-2L-20240415. This step creates an initial gradient coating structure on the board surface with ordered, physically continuous, multi-layered distribution along the thickness direction or in specific regions. This provides a precise and controllable physical starting point for subsequent gradient-induced migration and self-assembly of heat conduction channels. After migration induction and detection in steps S5-S6, the continuity of the heat conduction network is improved by 34%, and the heat dissipation consistency index is more than 20% better than that of comparable circuit board samples in the industry.

[0152] S4.4: For the already deposited partitioned gradient coating structure, the online process parameter recording module is used to collect manufacturing process parameters such as coating speed, application thickness, substrate temperature, and precursor liquid viscosity in each substrate area in real time, generate structured data tags, and map each process data to the physical substrate partitions to form a corresponding multi-zone distributed manufacturing process database.

[0153] The input object is a multi-zone gradient coating structure completed by multi-channel synchronous coating. The input data includes the deposition area number of each zone, the corresponding precursor liquid batch label, the real-time operation data acquisition signal of the coating equipment, and the real-time process status parameters.

[0154] A real-time online process parameter acquisition method (parameters: high-precision position encoder, thickness detection sensor, zone identification label, online temperature / viscosity sensor array) is adopted to achieve dynamic acquisition of key process data for each substrate area.

[0155] Furthermore, by using a regional coating speed monitoring algorithm (parameters: servo motor speed feedback signal, actual operating speed of coating head, and region length coordinates), continuous monitoring of the coating speed of each zone is achieved, and the speed is compared with a preset process standard library to determine the match between the actual speed and the target process.

[0156] Furthermore, through thickness detection and data archiving algorithms (parameters: thickness laser measurement system, partitioned spatial coordinates, consistency error compensation threshold), the physical thickness information of the gradient coating corresponding to each deposition area is obtained in real time, and the thickness acquisition data stream is output.

[0157] Furthermore, a multi-channel data acquisition device for substrate temperature and precursor liquid viscosity (parameters: thermocouple array, accuracy ±0.1℃, and rotational viscometer module, measurement range 1~10000mPa·s) is applied to achieve multi-point monitoring of substrate temperature gradient and actual viscosity of gradient precursor liquid during coating operation.

[0158] The data structure tag generation algorithm (parameters: spatial partition number, precursor liquid batch, timestamp, and collection parameter set) is used to bind the process parameters of each partition (including coating speed, coating thickness, substrate temperature, precursor liquid viscosity, etc.) collected above one by one, generating data structure tags with physical spatial partition attributes.

[0159] Furthermore, by leveraging the automatic archiving module of the manufacturing process database (parameters: real-time transmission protocol, data integrity check code, batch traceability label), all collected and labeled data are incorporated into a multi-region distributed manufacturing process database, enabling digital traceability of process parameters and quality process monitoring throughout the entire process.

[0160] The online process parameter recording module converts real-time process parameters during the gradient coating deposition process in each zone into structured data tags, enabling accurate collection and digital archiving of multi-zone gradient manufacturing process information, and providing a scientific data foundation for subsequent process data quality comparison, process optimization, and traceability.

[0161] For example, on a 200×150mm three-zone gradient deposition circuit board sample, three thickness laser detection heads and three sets of servo coating speed tracking encoders were configured. The measured coating speeds in zones A, B, and C during the test cycle were 1.80±0.04m / min, 1.40±0.03m / min, and 1.00±0.02m / min, respectively. Simultaneously, the thickness fluctuations of each layer, detected by the thickness sensor, were all below ±0.7μm. Temperature monitoring arrays showed the station temperature in zone A to be 28.6±0.1℃, in zone B 27.8±0.2℃, and in zone C 27.2±0.2℃. Real-time monitoring of the precursor liquid viscosity showed 4150±8mPa·s in zone A, 3540±7mPa·s in zone B, and 2990±6mPa·s in zone C. All collected data were associated with partition number, batch, and timestamp to generate structured tags (e.g., A-G1-20240415-01), with a data integrity rate of 100%. The entire process data of the archived batch was added to the manufacturing process database, providing full-process, digital data support for subsequent evaluation of coating deposition consistency quality in various regions, tracing of regional parameter trends, and anomaly source tracing and optimization.

[0162] S4.5: By combining the partitioned structured tags in the process database, the formed multi-zone gradient coating structure is compared with the preset surfactant gradient and high thermal conductivity nanofiller distribution model to evaluate the step consistency and gradient distribution accuracy of the deposition structure, obtain the deposition process quality assessment output, and provide data reference and process optimization support for subsequent adaptive migration and thermal channel construction.

[0163] Step S5: Under constant humidity and temperature control, the solvent evaporation rate of each substrate region is adjusted. The interfacial tension gradient is activated by stepwise heating or zoned vacuuming, causing the nanofillers already coated in the precursor solution to adaptively migrate and orderly position themselves in the high-interfacial-surfactant region, achieving the self-assembly of a zoned, gradually dispersed, thermally conductive filler network. Specifically, this includes:

[0164] S5.1: For the gradient coating area on the substrate surface where the gradient surfactant-coated high thermal conductivity nanofiller-resin precursor liquid is deposited, environmental control settings are executed based on the ambient temperature and humidity parameters measured by on-site sensors to obtain a constant humidity and temperature control process window, providing precise input conditions for the controllability of the subsequent solvent evaporation rate. The goal is to form a constant process environment parameter input.

[0165] S5.2: Taking the surfactant-resin precursor liquid coated with highly thermally conductive nanofillers in each substrate region as the treatment object, the solvent evaporation rate of each region is adjusted stepwise according to the parameters of the distributed heating unit or the zoned vacuum system, based on the preset heating curve or vacuum program. The goal is to generate a physical environment with differentiated solvent removal rates in each region, thereby establishing an energy basis for the activation of the interfacial tension gradient.

[0166] The interface surfactant-resin precursor liquid coated with highly thermally conductive nanofillers in each substrate area is used as the input object. The environmental parameters of the current area are set, including the board partition number, precursor liquid characteristic label, real-time temperature and humidity data of the substrate area, and the preset solvent evaporation gradient target of the partition.

[0167] A distributed heating unit parameter closed-loop control method (parameters: heating unit number, target temperature rise curve, spatial coordinates, heating output power) is adopted to perform dynamic temperature rise curve setting for each substrate area, thereby achieving regional temperature gradual rise and providing a basic driving force for differentiated solvent evaporation rates.

[0168] Furthermore, by using a differentiated control algorithm for the zoned gas extraction system (parameters: zoned airtight gate control, vacuum target value, gas extraction rate, pressure gradient), independent gas extraction control is implemented for different zones, dynamically adjusting the gas pressure gradient above the zones, enhancing the driving force for solvent removal between zones, and precisely maintaining the controllability of the atmosphere in the zoned physical environment.

[0169] Furthermore, an online solvent evaporation rate monitoring and feedback parameter adjustment method is adopted (parameters: real-time solvent mass evaporation rate sensor, feedback compensation factor, target evaporation rate range) to collect the solvent release amount on the surface of each zone in real time. Based on the deviation between the measured evaporation rate and the theoretical gradient target, the output of the heating unit and the parameters of the pumping system are automatically adjusted to form a closed-loop coordination of the zone gradient solvent evaporation rate.

[0170] Furthermore, based on the regional physical environment data and solvent evaporation rate distribution, the solvent removal curves of each region are fitted in real time using the regional evaporation dynamic modeling formula to obtain the following gradient evaporation model:

[0171] R i (t)=k i ·(T i (t)-T0)·(P s,i (t)-P a,i (t))

[0172] Among them, R i (t) represents the solvent evaporation rate of the i-th partition at time t, k i T is the partition mass transfer coefficient. i (t) represents the current temperature of the region, T0 represents the ambient reference temperature, and P s,i (t) represents the saturated vapor pressure of the partitioned solvent, P a,i (t) represents the pressure of the gas phase zone.

[0173] Through the above model iterations, the zonal heating and extraction parameters are continuously optimized to achieve a differentiated and controlled solvent removal gradient.

[0174] By using a gradient control algorithm, the solvent removal rate distribution of each partition is transformed into dynamic physical environment parameters, providing an energy basis for subsequent activation and formation of an interfacial tension gradient field, thereby realizing the adaptive migration of high thermal conductivity nanofillers and the ability to construct microstructures driven by interfaces.

[0175] For example, in a 200mm × 150mm three-zone circuit board manufacturing scenario, zone A (high load heat dissipation zone) uses a distributed heating unit with a set heating curve [25℃→38℃, 15min uniform heating rate], and the target vacuum degree of the vacuum system is 0.08MPa. Zone B (medium load) has a set heating curve [25℃→33℃, 15min] with a vacuum degree of 0.06MPa, and zone C (low load) is heated [25℃→28℃, 15min] with a vacuum degree of 0.03MPa. Three solvent evaporation sensors collect the average evaporation rates of zones A, B, and C in real time, which are 0.92 g·m³. -2 ·min -1 0.66 g·m -2 ·min -1 0.35 g·m -2 ·min -1 Fitting the gradient model yields k A =0.85, k b =0.62, k C =0.33, the deviation between the R value and the target gradient in each region is lower than 0.33.

[0176] ±5%. The feedback parameter adjustment mechanism slightly reduces the heating output based on the short-term high evaporation rate in zone A, and automatically fine-tunes the vacuum pumping rate in zone B by ±0.005 MPa, achieving dynamic stability of the spatially partitioned gradient solvent removal environment. After this step, all regions achieve enhanced physical environment gradients, which can effectively activate the subsequent interfacial tension gradient field, guide the precise migration of nanofillers and the orderly construction of the channel network, ultimately improving the continuity and heat dissipation consistency of the heat conduction channels on the plate surface. The measured surfactant-induced migration rate in zone A is approximately 38% higher than that of the traditional gradient-free process.

[0177] S5.3: Based on the already achieved gradual change in solvent evaporation rate across different substrate regions, the interfacial tension distribution curves of the high thermal conductivity nanofiller-surfactant-resin system in each substrate region are calculated in real time. The gradient intensity distribution is evaluated using liquid phase interface science algorithms, and the interfacial tension gradient activation parameters are output. The goal is to generate an interfacial tension gradient control signal that dynamically adjusts with process evolution.

[0178] S5.4: Based on the acquired interfacial tension gradient control signal, spontaneous migration of the high thermal conductivity nanofiller group is induced. The interfacial tension gradient field drives the coated nanofillers to migrate along the gradient direction to regions with high surfactant concentrations, achieving adaptive and ordered positioning of the nanofillers. The goal is to form a network of highly efficient thermally conductive channels in the thickness direction or regionally.

[0179] S5.5: The migration kinetics of high thermal conductivity nanofillers under surfactant gradient guidance are analyzed. By combining data on residual solvent in different zones and the evolution trajectory of the interfacial tension field, migration convergence is controlled to ensure stable distribution of the nanofillers and prevent secondary agglomeration. The goal is to obtain a self-assembled structure of a zoned, gradually dispersed, thermally conductive filler network, achieving continuity of the thermal conduction path.

[0180] S5.6: Utilize a multi-channel data acquisition device to monitor in real time the migration and positioning behavior of high thermal conductivity nanofillers in each zone, the concentration gradient distribution of surfactants, and the final thermally conductive filler network structure. Generate multi-dimensional migration distribution evaluation parameters to provide data support for subsequent structure finalization and automatic quality feedback optimization. The goal is to achieve accurate process output of layered / zoned nanofiller distribution information.

[0181] Step S6: Using non-contact methods such as infrared thermography, optical interference, and fluorescent labeling, in-situ multi-mode detection is performed on each substrate region to collect data on filler distribution uniformity, thermal conductivity channel continuity, and interface structure consistency, generating evaluation parameters for zoned filler dispersion and channel performance. Specifically, this includes:

[0182] S6.1: Perform infrared thermal imaging detection on the pre-cured coating samples in each substrate region. By acquiring the infrared thermal distribution signal in the region, and processing it based on the heat flux density distribution algorithm, extract the initial heat conduction channel continuity map to obtain the spatial continuity distribution characteristics of the heat conduction network in the region.

[0183] Using the pre-curing gradient coating areas of the substrate, which are deposited in each zone after self-assembly migration of the high thermal conductivity nanofiller-surfactant-resin system, as the detection input object, the infrared thermal imaging detection method (parameter settings: infrared thermal imager model, measurement resolution, radiation calibration coefficient, detection area zone number) is used to collect in-situ non-contact thermal distribution signals in each area, forming a high-resolution infrared thermal imaging raw dataset of the substrate in multiple zones.

[0184] A heat flux density distribution algorithm (parameters: regional pixel temperature field matrix, material thermal conductivity, coating thickness, and boundary temperature conditions) is used to perform pixel-level processing on the acquired infrared thermal image data. The surface temperature distribution value of each detection area is mapped to the internal heat flux conduction intensity. The initial heat flux density of each pixel is calculated using the following analytical heat flux density model:

[0185]

[0186] Where, q i Let ΔT be the heat flux density corresponding to the i-th pixel, k be the effective thermal conductivity of the material in this region, and ΔT be the heat flux density corresponding to the i-th pixel. i Δx represents the pixel temperature gradient, and Δx represents the physical thickness between pixels.

[0187] Furthermore, through a three-dimensional heat flux density field reconstruction algorithm (parameters: two-dimensional heat flux density matrix, thickness layer domain partition label, thermally conductive filler distribution model), the connectivity of the coating layer to the heat flux channel in each region is analyzed by spatial vector analysis, and the system outputs the initial thermal conductivity channel continuity map of the region.

[0188] Furthermore, a channel connectivity evaluation algorithm (parameters: heat flow field vector diagram, shortest heat conduction path, continuous domain scale index, conduction angle parameter) is used to perform a zoned and layered quantitative evaluation of the initial heat flow path, generating a regional heat conduction channel spatial continuity distribution feature set.

[0189] By using the structured output of thermal channel continuity maps and distribution feature sets, the microstructure of thermally conductive channels formed by the adaptive dispersion of high thermal conductivity nanofillers and driven by interfacial tension gradients is transformed into quantifiable spatial continuity parameters, enabling spatial visualization and data-driven characterization of thermally conductive networks in different partitions and thicknesses.

[0190] For example, on a 210mm × 140mm three-zone circuit board coating sample, regions A, B, and C were scanned using a high-sensitivity (NETD < 30mK) infrared thermal imager with a resolution of 640 × 480. The emissivity was set to 0.95, and each scanned region was 70mm × 140mm. After acquiring the surface temperature distribution, the thermal conductivity kJ of the material was used to determine the temperature distribution of each region. A =3.3W·m -1 ·K-1,k B =2.6W·m -1 ·K-1,k C =1.8W·m -1 K-1 is used as input, with a thickness range of 2.3 mm. Based on the average temperature difference between pixels obtained from the detection, ΔT in area A is... i The K value is 1.1K in region A, 1.5K in region B, and 2.0K in region C. Applying the analytical heat flux density model, the main channel temperature in region A is q... main =-3.3×(1.1 / 2.3)≈-1.58W·m -2 Area B q main ≈-1.70W·m -2 C area q main ≈-1.57W·m -2Furthermore, three-dimensional heat flux density field analysis yielded a channel connectivity index of 0.93 (excellent continuity) for region A, 0.77 (good) for region B, and 0.65 (moderate) for region C. A thermal channel continuity map for each region was output, corresponding to the distribution characteristics of the heat dissipation bus network, providing fundamental data for subsequent optical and multimodal fusion criteria. This detection process allows for high-resolution, non-contact, and quantitative assessment of the dispersion uniformity and thermal conductivity continuity of each region, providing proactive detection methods and data support for quality control and intelligent optimization of gradient coating structures.

[0191] S6.2: For the same coating area after infrared thermal imaging, an interferogram of the coating micro-uniformity in the substrate area is obtained based on an optical interferometric detection device, and a multi-wavelength interferometric analysis algorithm is used for data fitting to perform quantitative analysis of the uniformity of the filler micro-region distribution.

[0192] For the same coating area after infrared thermal imaging detection, multidimensional infrared thermal imaging data of that area is acquired as the input signal, and the partition number, pixel spatial coordinates, heat flux density parameters, and previously obtained thermal channel continuity results are used as the data basis.

[0193] A high-precision optical interferometry detection device (parameters: incident light wavelength range, illumination angle, spatial resolution, interference intensity calibration coefficient) is used to perform in-situ multi-wavelength interferometry acquisition on the partitioned area of ​​the substrate under test, generating a complete set of interference fringe images of the detection area, thereby achieving non-destructive detection of the uniformity of the coating surface and the nanostructure of the thickness micro-area.

[0194] Furthermore, through an optical interference intensity-phase mapping algorithm (parameter: center wavelength λ of the multi-wavelength light source), i Incident / reflected optical path length L j Pixel Intensity S i,j Interference phase This enables multi-wavelength phase demodulation and quantitative calculation of relative thickness changes for each spatial pixel in the interference image, and constructs a spatially resolved two-dimensional or three-dimensional coating micro-uniformity distribution data matrix.

[0195] A multi-wavelength interferometry algorithm (parameters: wavelength combination of each light source, spatial phase change function, statistical fitting window width) was used to fit the spatial fluctuation curves of the interferometric image sequence at multiple wavelengths. The uniformity of the nanofiller distribution in different micro-regions was quantitatively expressed by the following formula relating interference fringe density to thickness variation:

[0196]

[0197] Where, Δd x,y Let ΔN be the thickness variation at pixel position (x,y). x,y For the corresponding pixel stripe number change, For the weighted average of each wavelength, n eff The effective refractive index of the coating system.

[0198] Furthermore, by using a data fitting algorithm (parameters: least squares fitting, fitting residual threshold, spatial coefficient of variation threshold) to perform zonal statistical processing on the data set of stripe period, phase difference and spatial thickness variation obtained from fitting all pixels, quantitative parameters of micro-area filler distribution uniformity are obtained, including multivariate indicators such as uniformity coefficient of variation (CV), maximum / minimum thickness deviation, and spatial skeleton connectivity coefficient.

[0199] Through the above multi-step algorithm processing, the spatially resolved dataset generated by the optical interferometry detection process is transformed into a quantitative micro-uniformity distribution map of the coating, realizing accurate quantitative characterization of the distribution uniformity of high thermal conductivity nanofillers in each partition and micro-region.

[0200] For example, for the A, B, and C three-section coatings on a 210mm×140mm substrate, a multi-channel interferometric detection system with an incident wavelength range of 400-700nm and a spatial resolution of 1μm was used to collect multi-wavelength interferometric data of the 70mm×140mm region of area A. The average spatial variation amplitude of the fringe density was 0.095. n eff =1.45. Substituting the data into the above formula, the maximum three-dimensional thickness variation Δd in region A is... max The minimum value is Δd, which is 145nm. min The thickness was 88 nm, and the coefficient of variation (CV) for spatial uniformity was 0.061, indicating a highly uniform filler distribution. For region B, due to significant fluctuations in coating thickness, the CV was measured to be 0.119; in region C, due to local aggregation along the nanofiller migration path, the CV increased to 0.154. These multi-region data indicate that the filler uniformity in region A is superior to other regions. The final multi-region optical interferometry quantitative analysis results provide a high-precision uniformity foundation for subsequent fluorescence labeling three-dimensional imaging and multimodal data fusion, enabling quantitative assessment of filler uniformity at the microstructure level and decision-making for process optimization.

[0201] S6.3: Using the substrate area after optical interference detection as the detection object, fluorescent label tracking is implemented. The three-dimensional distribution signal of the nanofiller is acquired in real time through fluorescence imaging technology. The fluorescence signal intensity and three-dimensional imaging results are combined with the spatial correlation recognition algorithm to generate analysis parameters of the actual migration path and aggregation trend of the filler.

[0202] S6.4: The thermal channel continuity data, filler uniformity data and filler migration path analysis parameters obtained in steps S6.1 to S6.3 are fused together. A multimodal data fusion algorithm is applied to integrate the filler distribution and thermal conduction channel related multivariate parameter set of each substrate region to form a regional distribution and thermal conduction performance integrated performance parameter matrix.

[0203] S6.5: Based on the obtained integrated performance parameter matrix, a comprehensive evaluation is performed on the uniformity of filler distribution, continuity of heat conduction channels, and consistency of interface structure in each substrate area. The evaluation parameters of the performance of filler dispersion and heat conduction channels in each area are output through parameter weighted scoring and zonal statistical algorithms. These evaluation parameters are used as an important decision-making basis for subsequent automatic optimization of the process.

[0204] Step S7: Determine whether the filler dispersion uniformity parameters and heat conduction channel continuity evaluation parameters in each region meet the preset consistency and high-efficiency heat conduction thresholds. If threshold abnormalities occur, feedback is given to process control parameters such as the surfactant gradient parameter in S2 and the coating rate in S4, and automated optimization iterations are performed until the technical indicators meet the requirements. Specifically, this includes:

[0205] S7.1: The parameters of uniformity of packing dispersion and evaluation parameters of thermal conductivity channel continuity in the in-situ multi-mode test results are structured and aggregated to extract functional index data of each substrate region, providing a quantitative basis for subsequent consistency judgment.

[0206] S7.2: Based on industry standards or user-defined thresholds, a zoning judgment algorithm is used to perform qualification judgment on the filler dispersion uniformity parameter and the thermal conductivity channel continuity evaluation parameter, generate qualification mark information for each substrate area, and give an out-of-limit abnormality mark.

[0207] S7.3: Classify and summarize the qualified marking information and abnormal identification results, generate a regional consistency evaluation report, and calculate the global packing distribution uniformity evaluation index and the heat conduction channel continuity total evaluation index to realize the automatic quantification of the global process consistency level.

[0208] Using the conformity marking information and out-of-limit anomaly identification results of each substrate area generated in step S7.2 as input data, the partition attribute aggregation method (parameters: partition number, judgment label, functional evaluation parameters) is adopted to realize the structured grouping and classification summary of the detection information of each partition.

[0209] Furthermore, an algorithm for generating regional consistency evaluation reports (parameters: zoning label, compliance status, anomaly frequency, and quantitative evaluation parameter set) is applied to automatically sort and summarize the compliance and anomaly markers of all regions, and output a structured regional distribution consistency evaluation report. The report content includes evaluation parameters for the uniformity of packing dispersion, evaluation parameters for the continuity of heat conduction channels, interface structure consistency score, and corresponding compliance / anomaly status entries for each region.

[0210] Furthermore, using a global functional evaluation index calculation algorithm (parameters: functional quantification parameters of all zones and weight allocation coefficients), the evaluation parameters {U_i} (i=1,...,N) of packing dispersion uniformity and the evaluation parameters {C_i} (i=1,...,N) of heat conduction channel continuity are weighted and aggregated to calculate the global packing distribution uniformity evaluation index U. global and the overall evaluation index C of the continuity of the heat conduction channel global ,in:

[0211]

[0212] Where N is the total number of substrate partitions, U i C is the score for the uniformity of packing dispersion in region i. i The continuity score of the heat conduction channel in region i is given.

[0213] Furthermore, by evaluating the threshold comparison and automatic quantization processing module (parameters: industry standard or user-defined target threshold), U... global and C global It automatically compares with the pre-set high-efficiency heat dissipation technology threshold to form a global consistency parameter status mark, thereby realizing quantitative monitoring and output of the global consistency level of the coating gradient process.

[0214] Through the above chain derivation steps, the zonal-level qualification and anomaly judgment results are transformed into zonal consistency evaluation reports, and further aggregated to form global packing distribution uniformity evaluation index and heat conduction channel continuity overall evaluation index, realizing automated and quantitative evaluation of process consistency level, and providing decision support for subsequent process anomaly location and feedback optimization.

[0215] Exemplarily, on a highly efficient heat-conducting circuit board of 210mm×140mm, the detection area is divided into three major zones: A (70×140mm), B, and C. Through the previous step S7.2, zone A is qualified, the continuity score of the heat-conducting channels in zone B is 0.80, lower than the threshold of 0.85, and the uniformity score of the filler dispersion in zone C is 0.76, lower than the threshold of 0.80, and they are respectively marked as abnormal. Using the zone attribute aggregation method, zone A is automatically sorted as qualified, and zones B / C are abnormal. The summary evaluation report archives the quantitative dispersion scores (A: 0.89, B: 0.81, C: 0.76) and channel scores (A: 0.93, B: 0.80, C: 0.66) of each zone. According to the formula:

[0216]

[0217] Automatically compare with the high-efficiency heat dissipation standard thresholds of the user's preset filler dispersion uniformity of 0.85 and channel continuity of 0.85, identify that both the global uniformity and continuity indicators are lower than the target threshold, automatically output a determination of insufficient overall consistency, and attach a quantitative report of the abnormal zones, preparing a structured input for the abnormal cause analysis and subsequent closed-loop feedback optimization steps (S7.4~S7.6). This example verifies the technical effectiveness and process support capabilities of the method described in sub-step S7.3 in aspects such as automatic aggregation of multi-zone data, global target quantification, and triggering of the quality control closed-loop.

[0218] S7.4: For the zones with abnormalities, automatically associate the previous process parameters, including surfactant gradient parameters, coating rate, and precursor solution component parameters, and locate the key process factors causing the abnormalities through parameter backtracking and causal analysis algorithms.

[0219] S7.5: Based on the causal relationship between the process parameters of the abnormal zones and the abnormalities in their dispersion uniformity and heat-conducting channel continuity, call the automated process optimization model, dynamically adjust and feedback controllable process variables such as surfactant gradient parameters and coating rate, and generate an updated set of gradient coating process parameters.

[0220] S7.6: Input the optimized set of gradient coating process parameters into the intelligent production control system, iteratively update the surfactant concentration distribution scheme and coating rate, make them adaptively coordinated with the material physical property parameters and actual process status of the previous abnormal zones, and restart the closed-loop adaptive optimization iteration of the process flow.

[0221] Step S8: After the dispersion in each substrate area is uniform and the continuity of the heat-conducting channels meets the standard, use a medium-low temperature step-by-step curing procedure to thermally set and crosslink the coatings in each area, lock the optimized gradually changing distribution structure of the nano-fillers, and finally obtain a circuit board product with uniform high-efficiency heat dissipation performance. Specifically, it includes:

[0222] S8.1: For substrate areas where the filler is uniformly dispersed and the thermal conductivity channel continuity has met the standard, zoning labels are recorded. Based on the input zoning control parameters, an adaptive curing temperature profile is developed to ensure that the curing process accurately matches the gradual distribution structure of the nanofiller in different areas.

[0223] S8.2: Based on the partition control parameters and regional interface structure information, a medium-low temperature stepwise heating program is used to thermally induce crosslinking of the partition nanofiller gradient distribution coating, so that the polymer matrix gradually undergoes molecular chain movement and forms a dense crosslinked network, thereby solidifying the spatial distribution state of the high thermal conductivity nanofiller and realizing the physical locking of the thermal conduction channel structure.

[0224] S8.3: For each area of ​​the coating after step-by-step curing, in-situ dynamic infrared thermography and micro-area structure optical detection are performed to obtain the integrity of the thermal conductive channel network and the micro-interface structure parameters after curing, so as to realize real-time feedback on the effect of medium and low temperature curing process.

[0225] S8.4: Based on the integrity of the heat conduction channels and interface structure parameters obtained from in-situ detection, a quality threshold judgment is performed. If the parameters deviate from the preset range, the curing process parameters are automatically iterated and optimized in combination with the aforementioned partition labels and curing temperature curves to ensure that the final structure latch-up and thermal performance are consistent with the standards.

[0226] Using the integrity parameters of the heat conduction channel network and the micro-interface structure parameters obtained by in-situ dynamic infrared thermography and micro-area structure optical detection after curing as input, the partition quality threshold judgment process is performed.

[0227] A block-based data reduction and threshold comparison algorithm (parameters: partition number, heat conduction channel integrity parameter, interface structure parameter, preset threshold) is adopted to realize the automatic identification of the heat channel network and interface structure status of each partition.

[0228] Furthermore, through a continuous threshold difference judgment model (parameters: partition structural index, target range, deviation range), the integrity of the heat conduction channels and interface structure parameters of all partitions are gradually compared with the preset target values ​​to obtain the qualified / abnormal status label and deviation degree quantitative data of each partition.

[0229] Furthermore, a dynamic aggregation algorithm for partition labels (parameters: partition label, historical curing parameters, temperature control curve) is used to automatically retrieve the partition label before curing and the curing temperature curve for partitions with parameter deviations, thereby enabling anomaly tracing.

[0230] Furthermore, through intelligent process parameter correction and curing program adaptive optimization model (parameters: regional parameter deviation data, partition labels, curing temperature curves, time steps), the curing parameters of abnormal partitions are automatically iteratively adjusted to generate new curing temperature, heating rate and holding time schemes, which are then input into the partition process control system.

[0231] Furthermore, a localized heat setting and cross-linking process is performed under the modified parameter curing cycle, and the parameter convergence is monitored in real time until the integrity of the partitioned heat conduction channels and the interface structure parameters converge to the threshold range.

[0232] Through the above-mentioned chain-adaptive optimization algorithm, the partition process data with in-situ detection parameter deviations are transformed into a curing curve correction reference, so as to achieve the curing latching consistency and high-efficiency heat dissipation performance of the entire area heat conduction channel and the gradual distribution structure.

[0233] For example, after curing a 210mm×140mm high-efficiency heat dissipation circuit board, the thermal conductivity channel integrity parameters for zones A, B, and C were detected as 0.92, 0.77, and 0.88, respectively, and the interface structure parameters were 0.95, 0.80, and 0.91, respectively. Zone B's thermal conductivity channel integrity was below the threshold of 0.85, and its interface structure parameter was below the threshold of 0.83, automatically classifying Zone B as abnormal. The curing process database showed that Zone B used the first set of curing temperature curves (initial 60℃, stepped increase to 100℃, hold for 15 min), while Zones A and C performed normally. The system retrieved historical curing parameters for Zone B and, through the process parameter adaptive optimization model, suggested changing the curing temperature curve to a step-by-step process of 65℃ / 15 min followed by a re-curing at 105℃ / 15 min. After re-curing, Zone B's thermal conductivity channel integrity improved to 0.87, and its interface structure parameter improved to 0.86, both falling within the target threshold range. Ultimately, after the A, B, and C partitions are solidified, they all achieve structural latch consistency, outputting a quality compliance mark for all partitions, thus realizing the technical goal of uniformly achieving efficient heat dissipation performance.

[0234] S8.5: After all zones have met the curing quality threshold, the shaped circuit board is extracted and subjected to end mechanical and thermal stress release treatment to ensure that the high thermal conductivity nanofiller gradient distribution structure has long-term structural stability in actual application environment, and finally obtains a finished circuit board with high efficiency and uniform heat dissipation performance, and outputs the batch number and quality certification parameters for full process traceability.

[0235] This application also provides a high-efficiency heat dissipation circuit board, manufactured using the above-described manufacturing method.

[0236] For those skilled in the art, various other corresponding changes and modifications can be made based on the technical solutions and concepts described above, and all such changes and modifications should fall within the protection scope of the claims of this invention.

[0237] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this application pertains. The terms “first,” “second,” “third,” and similar terms used in this patent application specification and claims do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an” or “one,” etc., do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “comprising” or “including,” etc., mean that the element or object preceding “comprising” or “including” encompasses the element or object listed following “comprising” or “including” and its equivalents, and do not exclude other elements or objects. The “multiple” mentioned in the embodiments of this application refers to two or more. A and / or B indicate three possibilities: A; B; and A and B.

[0238] The above description is merely an exemplary embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for manufacturing high-efficiency heat dissipation circuit board, comprising the following steps: S1: obtaining high-thermal-conductivity nano-filler and resin matrix raw materials of different batches and specifications, and performing structured collection of physical performance parameters and batch labels; S2: configuring different concentrations and types of interfacial active agents according to the physical performance parameters of the high-thermal-conductivity nano-filler and the resin matrix, and preparing an interfacial active agent-resin precursor liquid system to obtain an active agent gradient liquid; S3: adding the high-thermal-conductivity nano-filler into each type of interfacial active agent-resin precursor liquid system, and performing uniform initial dispersion treatment by controlling dispersion equipment parameters to obtain a nano-filler uniform coating body in a multi-gradient precursor liquid; S4: applying the nano-filler uniform coating body to the surface of a circuit board substrate at different positions by using a multi-channel synchronous gradient coating technology in time or in layers to obtain a multi-zone distributed gradient coating deposition; S5: adjusting the solvent evaporation rate of each substrate region in a constant humidity and temperature environment, and activating the interfacial tension gradient by stepwise temperature rising or zonal air extraction; S6: performing in-situ multi-mode detection on each substrate region to collect data of filler distribution uniformity, thermal conduction channel continuity and interfacial structure consistency, and generate zonal filler dispersion uniformity parameters and channel performance evaluation parameters; S7: judging whether the filler dispersion uniformity parameters and the thermal conduction channel performance evaluation parameters of each region meet preset consistency and high-efficiency thermal conduction threshold values, and if threshold value abnormalities occur, performing automatic optimization iteration until the technical index requirements are met; and S8: after the dispersion uniformity of each substrate region and the thermal conduction channel continuity meet the standards, performing heat setting and crosslinking of the coating layer of each region by using a low-temperature step-by-step curing program to obtain a high-efficiency heat dissipation performance uniform circuit board product. The step S1 specifically comprises: performing specification screening processing on high-thermal-conductivity nano-filler raw materials for high-efficiency heat dissipation circuit board preparation, performing nano particle size analysis and surface property analysis according to physical performance parameters including particle size distribution, electrical conductivity and particle surface energy to obtain high-thermal-conductivity nano-filler original parameter data; performing chemical composition analysis and thermal analysis on resin matrix raw materials suitable for circuit board substrates to obtain structured correlation data of resin matrix raw material molecular parameters and batch labels; and performing interface matching score and labeling processing by material adaptability modeling algorithm to form high-thermal-conductivity nano-filler-resin matrix raw material adaptability labels. The step S2 specifically comprises: performing data structured analysis on high-thermal-conductivity nano-filler physical performance parameters and resin matrix physical performance parameters, generating distribution model parameters required by a target gradient formula by using a material adaptability analysis algorithm; generating primary proportioning data sets of different types of synergistic active agents according to the interface active agent distribution groups obtained by material adaptability analysis and by high-precision proportioning control; inputting the active agent primary proportioning data sets to perform concentration gradient design and parameter optimization to obtain a plurality of interfacial active agent concentration gradient parameters; and inputting the plurality of interfacial active agent concentration gradient parameters into a resin precursor liquid precision mixing control system to perform mixing processing to form a structured layered interfacial active agent-resin precursor liquid system. ​ ​ ​ ​ ​ ​ ​ ​ 2. The high-efficiency heat dissipating circuit board manufacturing method according to claim 1, wherein ​ ​ ​ ​ 3. The high-efficiency heat dissipating circuit board manufacturing method according to claim 1, wherein ​ ​ ​ ​ ​ 4. The high-efficiency heat dissipating circuit board manufacturing method according to claim 3, wherein: The interface active agent includes hydrophilic and hydrophobic silane coupling agents.

5. The high-efficiency heat dissipating circuit board manufacturing method according to claim 1, wherein In step S5, the environmental control setting is performed according to the environmental temperature and humidity parameters measured by the on-site sensor, so as to obtain a constant humidity and temperature control process window.

6. The high-efficiency heat dissipating circuit board manufacturing method according to claim 1, wherein The step S6 specifically includes: The infrared thermal image detection is performed on the cured coating sample in each substrate area, the spatial continuity distribution characteristics of the heat conduction network of the area are obtained through the heat flow density distribution algorithm processing; For the same coating area after infrared thermal image detection, the coating micro-uniformity interference spectrum is obtained, and data fitting is performed for quantitative analysis of the filler micro-area distribution uniformity; Real-time acquisition of three-dimensional distribution signals of nanofiller, using imaging and spatial correlation recognition algorithm to generate filler actual migration path and aggregation trend analysis parameters; The collected heat channel continuity data, filler uniformity data and filler migration path parameters are data fused, and the filler distribution and heat conduction channel multi-performance parameters of each substrate area are integrated to form an integrated performance parameter matrix; According to the integrated performance parameter matrix, the filler distribution uniformity, heat conduction channel continuity and interface structure consistency of each substrate area are comprehensively evaluated to obtain the partition filler dispersion and heat conduction channel performance evaluation parameters.

7. The high-efficiency heat dissipating circuit board manufacturing method according to claim 1, wherein: In step S7, the partition filler dispersion uniformity and channel continuity index are respectively calculated by using the global scoring method, and are automatically determined and closed-loop optimized according to the threshold value.

8. The high-efficiency heat dissipating circuit board manufacturing method according to claim 1, wherein The step S8 specifically includes: The substrate area with filler dispersion uniformity and heat conduction channel continuity up to standard is recorded by partition label, and the adaptive curing temperature curve is developed; According to the partition control parameters and area interface structure information, the heat-induced crosslinking of the partition nanofiller gradient distribution coating is performed by using the low-temperature step-by-step heating program, so as to obtain the spatial distribution state of the cured high-thermal-conductivity nanofiller; The in-situ dynamic infrared thermal image and microstructure optical detection are performed on each area coating after step-by-step curing treatment, and the heat conduction channel network integrity and micro-interface structure parameters after curing are obtained; According to the heat conduction channel integrity and interface structure parameters obtained by in-situ detection, the quality threshold is determined, if the parameters deviate from the preset category, the curing process parameters are automatically iteratively optimized combined with the partition label and the curing temperature curve; After the quality threshold of all partition curing is up to standard, the shaped circuit board is extracted for mechanical and thermal stress release treatment, and the circuit board product with uniform high-efficiency heat dissipation performance is obtained, and the full-process batch number and quality certification parameters are output.

9. The high-efficiency heat dissipating circuit board manufacturing method according to claim 8, wherein: The heat conduction channel structure integrity and partition consistency of each area coating after curing treatment are confirmed by in-situ infrared thermal image and structure optical detection.

10. The high-efficiency heat-dissipating circuit board manufacturing method according to claim 1, wherein: The activation of the interface tension gradient is controlled by the on-site temperature and humidity, partition temperature rise and customized pumping parameters, and the solvent removal rate and interface tension gradient of different partitions are dynamically adjusted.

Images