Preparation method of organic silicon foam with conductive function

Abstract

The invention provides a preparation method of organic silicon foam with a conductive function. The preparation method comprises the following steps: step 1, raw material preparation; step 2, mixing and dispersing; step 3, foaming and cross-linking reaction, foaming environment control and cross-linking reaction control; the conductive filler is uniformly dispersed and the coating is optimized, so that the conductivity of the foam is remarkably improved, multi-parameter control and optimization strategies are realized, the problem of non-uniformity of cell size and wall thickness is remarkably reduced, the heat resistance and mechanical property of the foam are ensured through accurate cross-linking and curing processes, the parameters are optimized by utilizing machine learning, and the mechanical property of the foam is improved. Compared with the prior art, the preparation method has the advantages that the test time and material waste are reduced, the production efficiency is improved, the problems of nonuniform foam hole distribution, poor conductive filler dispersity, unstable foaming process and the like in the prior art are solved, and the conductivity, the mechanical property and the environmental adaptability of the foam are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of conductive foam, and particularly to a preparation method of a silicone foam with conductive function. Background Art

[0002] Conductive foam is a functional material with both conductivity and elasticity characteristics. Due to its excellent electrical conductivity, elastic recovery ability and processing adaptability, it is widely used in the electromagnetic shielding, grounding protection, sealing and buffering, and shock absorption and vibration reduction of electronic components. With the development of miniaturization and high performance of electronic devices, the performance requirements of conductive foam are continuously improved. It is necessary to meet the requirements of lightweight, high conductivity and excellent shielding effect, and also consider good mechanical properties, durability, environmental adaptability and plasticity. Most of the conductive foams on the market use rubber matrix or polyurethane as the main materials, and the electrical conductivity can be achieved by adding metal powders, carbon nanomaterials or conductive coatings. Silicone materials, with their excellent flexibility, high temperature resistance, aging resistance and chemical stability, have become a conductive foam matrix material that has received much attention. Compared with traditional materials, silicone materials perform better in harsh environments and are especially suitable for use in working conditions with high temperature, high humidity and strong corrosion.

[0003] However, at present, the cell size distribution of silicone foam is uneven, the cell size in some areas is too large or the closed cell rate is too high, which affects the uniformity of electrical conductivity and the stability of material properties. There are also often inconsistent cell wall thicknesses, and local defects may lead to the failure of material properties. The main reason is that the decomposition rate of the chemical blowing agent is not well controlled, resulting in too fast or too slow foaming rate. The temperature or pressure control is unstable, causing over-expansion or collapse during the foaming process. After adding conductive fillers, the filler particles may interfere with the uniformity of the gas released by the blowing agent. There is an urgent need for an improved preparation method to achieve uniform dispersion of conductive fillers, stable cell structure, improved electrical conductivity and comprehensive properties of the material by optimizing the raw material ratio, filler pretreatment process and foaming process control, so as to meet the requirements of the electronics industry for high-performance conductive foam. Summary of the Invention

[0004] The purpose of the present invention is to provide a preparation method of a silicone foam with conductive function, which realizes uniform dispersion of conductive fillers, stable cell structure, and improved electrical conductivity and comprehensive properties of the material by optimizing the raw material ratio, filler pretreatment process and foaming process control.

[0005] To achieve the above object, the present invention proposes the following technical solution: A preparation method of a silicone foam with conductive function, comprising the following steps:

[0006] Step 1, raw material preparation: Prepare silicone matrix, conductive filler, crosslinking agent, foaming agent and dispersant. The amounts of the silicone matrix, conductive filler, crosslinking agent, foaming agent and dispersant are 70 - 80%, 5 - 15%, 1 - 2%, 3 - 5% and 2 - 5% of the total mass respectively;

[0007] Step 2, mixing and dispersion: First, pre-treat the conductive filler. Place the conductive filler in a vacuum drying oven and dry it at 80 - 100 °C for 2 hours to remove moisture, and then surface-clean the conductive filler with ethanol or acetone;

[0008] Prepare the basic mixture. In a variable-frequency stirrer, add the silicone matrix, dispersant and the dried conductive filler to the stirring container in proportion, and stir at a speed of 1500 - 2000 rpm for 10 - 20 minutes to preliminarily disperse the conductive filler;

[0009] Ultrasonic dispersion: Transfer the mixture to an ultrasonic dispersion device, set the power to 500 - 1000 W, and the dispersion time to 30 minutes to ensure that the filler is evenly distributed in the matrix and form a stable conductive network;

[0010] Add the crosslinking agent and foaming agent. Under the condition of low-speed stirring at 500 - 800 rpm, add the crosslinking agent and foaming agent to the basic mixture in sequence and stir evenly for 5 - 10 minutes;

[0011] Step 3, foaming and crosslinking reaction: Control the foaming environment and crosslinking reaction;

[0012] Step 4, curing and forming: First, quickly inject the foamed mixture into a mold preheated to 80 - 100 °C, maintain the temperature at 120 °C in the mold and cure for 1 hour to ensure the preliminary forming of the foam. Then transfer the preliminarily cured foam to a hot air circulation oven, set the temperature to 150 - 180 °C and keep it for 2 - 4 hours;

[0013] Step 5, surface treatment and post-treatment: Uniformly spray a silver paste coating on the surface of the foam, control the coating thickness within 5 - 10 μm, and place the treated foam in a heat treatment at 180 °C for 30 minutes.

[0014] Furthermore, in the present invention, the silicone matrix is selected as room temperature vulcanized silicone rubber, the conductive filler is selected as carbon nanotubes or graphene, the crosslinking agent is benzoyl peroxide, the foaming agent is selected as azodicarbonamide or AC foaming agent, and the dispersant is a silane coupling agent.

[0015] Furthermore, in the present invention, in the ultrasonic dispersion step, the ultrasonic dispersion device is a device with a circulating cooling function, and the temperature during the dispersion process is controlled within the range of 30 - 50 °C to avoid the adverse effects of local overheating caused by ultrasonic dispersion on the performance of the matrix and conductive filler.

[0016] Further, in the present invention, in step four, after the foam removed from the mold during the curing process is cooled to room temperature, it is subjected to surface spraying treatment again. The spraying uniformity is controlled by a spin coating device to ensure that the conductivity of the silver paste coating on the surface of the foam meets the requirement of resistivity ≤ 0.1 Ω·cm.

[0017] Further, in the present invention, in the heat treatment process of step five, a programmable temperature control device is used for precise temperature control, and a temperature gradient is set, gradually increasing from 150 °C to 180 °C, and the heating time is 30 minutes, so as to improve the adhesion and conductivity between the silver paste coating and the surface of the foam.

[0018] Further, in the present invention, step three specifically is:

[0019] Foaming environment control: Transfer the mixture to a closed foaming kettle, set the reaction temperature to 120 - 160 °C, the pressure to 0.1 - 0.3 Mpa, and maintain the constant temperature and pressure conditions for 10 - 20 minutes to ensure uniform decomposition of the foaming agent and form a stable cell structure;

[0020] Crosslinking reaction control: Gradually increase the temperature to 180 - 200 °C and maintain for 30 minutes to promote the complete reaction between the silicone matrix and the crosslinking agent and form a uniform three-dimensional crosslinked network.

[0021] Further, in the present invention, the foaming environment control in step three is controlled by a monitoring and control system. The monitoring and control system includes a data acquisition unit, a control unit, and an execution unit. The control unit includes a single-chip microcomputer;

[0022] Among them, the control unit controls by the following steps:

[0023] Step S1, data acquisition and feature selection: Collect key process parameters and data related to the cell structure during the foaming process to provide a training data set for the machine learning model, and collect the input feature X. The input feature X includes: foaming temperature T f , foaming pressure P f , foaming agent concentration C f , mixing shear rate R m , conductive filler particle size distribution D p , foaming agent release rate S a , and the output feature Y is the cell size uniformity index Us w , with the unit of μm, and is expressed by the formula: Us w : Weighted index of cell size uniformity, with the unit of μm, V i : Volume of the i-th cell, with the unit of μm 3 , d i : Diameter of the i-th cell, μm The average diameter of all the cells, N: The total number of cells.

[0024] Collect process data X i and Y i , to form a data set for model training;

[0025] Step S2, establish a machine learning model, select a regression model to predict the uniformity index U of the cell structure during the foaming process s , regression model formula: f(X; θ) is the model function, representing the cell size uniformity U s The non-linear relationship between U and the input feature X, θ is the model parameter, including weights and biases, which need to be optimized through the training data, is the predicted cell size uniformity index;

[0026] Use the mean squared error (MSE) as the loss function to optimize the model parameter θ: L(θ): The value of the loss function, used to evaluate the error between the model prediction and the actual value, M is the number of training samples, w i : The weight of the i-th sample, used to adjust the influence of different samples on the model, U s，i is the uniformity index measured experimentally, is the predicted uniformity index by the model, λ: Regularization coefficient, used to control the model complexity and avoid overfitting, The L 2 norm of the model parameter, representing the sum of the squares of the weights. By minimizing L(θ), the model can obtain the optimal parameters.

[0027] Step S3, model optimization and hyperparameter tuning, adjust the process parameters: foaming temperature T f , foaming pressure P f , blowing agent concentration C f , mixing shear rate R m , conductive filler particle size distribution D p , blowing agent release rate S a， to minimize the uniformity index U s , while ensuring the stability of the foaming process, objective function:

[0028] w 1 w 2 w 3 : Weight coefficients, used to balance the priorities of uniformity, stability, and cost objectives;

[0029] The predicted cell size uniformity index, the smaller the better;

[0030] S f:Cell stability index, defined as the standard deviation of the cell wall thickness, unit: μm;

[0031] S min :Set stability threshold, representing the minimum requirement for the uniformity of the cell wall thickness;

[0032] Process cost: Process parameter adjustment cost. Using genetic algorithm or Bayesian optimization method, combined with the model prediction value, to find the optimal parameter combination that meets the process stability condition.

[0033] Furthermore, in the present invention, in the step S3, the cell stability index S f is defined as the distribution deviation of the cell wall thickness, then:

[0034] S f : Weighted index of cell wall thickness stability, unit: μm, V i : Volume of the i-th cell, unit: μm 3 , t i is the wall thickness of the i-th cell, is the average wall thickness of all cells, N: Number of cells;

[0035] Predict its release rate according to the decomposition kinetic equation of the blowing agent:

[0036]

[0037] Sa opt : Blowing agent release rate, unit: mol / s, k: Reaction rate constant, C f : Blowing agent concentration, unit: wt%, E a : Activation energy of blowing agent decomposition, unit: kJ / mol, R: Gas constant, 8.314 J / (mol·K), T: Reaction temperature, unit: K, D: Blowing agent diffusion rate, unit: μm / s, L: Blowing agent diffusion path length, unit: μm, P target : Target pressure, unit: Pa, P: Real-time pressure, unit: Pa, by adjusting C f and T, the blowing agent release rate can be optimized to improve the foaming stability.

[0038] Furthermore, in the present invention, the data acquisition unit includes:

[0039] High-precision temperature sensor, which monitors the temperature in the foaming kettle in real time through the high-precision temperature sensor and transmits the temperature data to the single-chip microcomputer;

[0040] Strain type pressure sensor, which monitors the pressure sensor in the foaming kettle in real time through the pressure sensor and transmits the pressure data to the single-chip microcomputer;

[0041] A mass flowmeter detects changes in the mass flow rate of the fluid, monitors the release rate of the foaming agent in real time, and transmits the monitoring data to the single-chip microcomputer;

[0042] A laser particle size analyzer irradiates particles with a laser, analyzes the intensity distribution of the scattered light to calculate the particle size, and the results can be used to adjust the filler uniformity and transmit the detection data to the single-chip microcomputer;

[0043] A microimaging instrument. After foaming is completed, the microimaging instrument takes cross-sectional images of the cell holes from the material section. The image analysis software calculates the cell hole diameter and distribution through an edge detection algorithm and outputs the cell size uniformity index U s for evaluating the overall uniformity of foaming.

[0044] Furthermore, in the present invention, the execution unit includes:

[0045] A PID temperature controller and a heater. The heater is arranged on the foaming kettle. The temperature controller calculates the temperature difference adjustment signal and drives the heater to output the corresponding power. If the temperature is too high, the heating power is reduced; if it is too low, the power is increased to ensure that the temperature is stable at the set value;

[0046] An electromagnetic valve and a pressure controller. The electromagnetic valve is arranged on the foaming kettle. The pressure controller controls the opening degree of the electromagnetic valve according to the collected pressure data, adjusts the gas injection or discharge rate to maintain the pressure in the foaming kettle within a stable range, and avoids the deformation of the cell holes due to pressure fluctuations;

[0047] A variable-frequency stirrer. The variable-frequency stirrer is arranged on the foaming kettle. The variable-frequency motor controls the rotation speed of the stirrer and adjusts the mixing intensity according to the shear rate Rm recommended by the model;

[0048] A flow controller. The flow controller adjusts the release rate of the foaming agent according to the data collected by the flowmeter, and ensures the stability of the cell hole structure by controlling the dosage and release time.

[0049] Advantageous effects. The technical solution of this application has the following technical effects:

[0050] The present invention provides a preparation method of an organosilicon foam with conductive function. By uniformly dispersing conductive fillers and optimizing the coating treatment, the conductive performance of the foam is significantly improved. Multiple parameter control and optimization strategies significantly reduce the non-uniformity problems of cell hole size and wall thickness. Precise cross-linking and curing processes ensure the heat resistance and mechanical properties of the foam. Using machine learning to optimize parameters reduces the test time and material waste, and improves the production efficiency. It not only overcomes the problems of uneven cell hole distribution, poor dispersion of conductive fillers and unstable foaming process in the prior art, but also significantly improves the conductive performance, mechanical properties and environmental adaptability of the foam.

[0051] It should be understood that all combinations of the foregoing concepts and additional concepts described in greater detail below are considered to be part of the inventive subject matter of the present disclosure as long as such concepts are not mutually inconsistent.

[0052] The foregoing and other aspects, embodiments, and features of the teachings of the present invention can be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the present invention, such as the features and / or beneficial effects of exemplary embodiments, will be apparent from the following description or will be learned from the practice of specific embodiments in accordance with the teachings of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in various figures may be represented by the same reference numeral. For clarity, not every component is labeled in each figure. Now, embodiments of various aspects of the present invention will be described by way of example and with reference to the accompanying drawings, in which:

[0054] Figure 1 is a schematic diagram of the method flow of the present invention.

[0055] Figure 2 is an SEM image of the cell structure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] To better understand the technical content of the present invention, specific embodiments are hereby given and described in conjunction with the accompanying drawings as follows. In the present disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. The embodiments of the present disclosure do not necessarily define all aspects of the present invention. It should be understood that the various concepts and embodiments introduced above, as well as those concepts and embodiments described in greater detail below, can be implemented in any of a number of ways, since the concepts and embodiments disclosed in the present invention are not limited to any particular embodiment. Additionally, some aspects of the present invention may be used alone or in any suitable combination with any other aspects of the present invention.

[0057] The embodiment provides a method for preparing a silicone foam having a conductive function, comprising the following steps:

[0058] Step 1, raw material preparation, preparing a silicone matrix, a conductive filler, a crosslinking agent, a foaming agent, and a dispersant, wherein the amounts of the silicone matrix, the conductive filler, the crosslinking agent, the foaming agent, and the dispersant are 70-80%, 5-15%, 1-2%, 3-5%, and 2-5% of the total mass, respectively;

[0059] Step 2, Mixing and Dispersion: First, pre-treat the conductive filler. Place the conductive filler in a vacuum drying oven and dry it at 80 - 100 °C for 2 hours to remove moisture. Then, perform surface cleaning on the conductive filler with ethanol or acetone. The surface cleaning process removes moisture and surface contaminants, enhancing the compatibility between the conductive filler and the matrix.

[0060] Prepare the basic mixture. In a variable-frequency stirrer, add the silicone matrix, dispersant, and dried conductive filler to the stirring container in proportion and stir at a speed of 1500 - 2000 rpm for 10 - 20 minutes to preliminarily disperse the conductive filler. By drying and surface cleaning the conductive filler, impurities and moisture are removed, improving the binding ability between the filler and the silicone matrix. High-speed stirring preliminarily disperses the filler, reducing the possibility of filler aggregation.

[0061] Ultrasonic dispersion: Transfer the mixture to an ultrasonic dispersion device, set the power to 500 - 1000 W, and the dispersion time to 30 minutes to ensure that the filler is evenly distributed in the matrix, forming a stable conductive network. Through the optimized combination of the ultrasonic dispersion step and the dispersant, the present invention significantly improves the dispersion of conductive fillers (such as carbon nanotubes, graphene) in the silicone matrix, ensuring uniform distribution of the filler and forming a stable conductive network. Using the cavitation effect of ultrasonic waves, the filler particles are further depolymerized and evenly dispersed, effectively avoiding the problem of filler agglomeration. In particular, by controlling the temperature (30 - 50 °C) during the dispersion process, damage to the material properties caused by high temperature is prevented. The silane coupling agent forms chemical bonds on the surface of the filler, improving the interfacial compatibility between the filler and the silicone matrix, thereby further enhancing the conductive performance and the mechanical properties of the foam.

[0062] Add the cross-linking agent and the foaming agent. Under the condition of low-speed stirring at 500 - 800 rpm, add the cross-linking agent and the foaming agent to the basic mixture in sequence and stir evenly for 5 - 10 minutes.

[0063] Step 3, Foaming and Cross-linking Reaction: Control the foaming environment and the cross-linking reaction. In this step, control the temperature and pressure in a closed foaming kettle to ensure uniform decomposition of the foaming agent and form a stable cell structure. The temperature gradient controls the cross-linking reaction to form a uniform three-dimensional network structure, improving the strength and stability of the foam. Precise control of temperature and pressure prevents cell collapse or excessive expansion, ensuring the uniformity of cell size. The cross-linking reaction further enhances the mechanical properties and durability of the material. Real-time monitor the data during the foaming process and optimize the parameters by combining machine learning to significantly improve production consistency.

[0064] Step 4, Curing and Molding: First, quickly inject the foamed mixture into a mold preheated to 80 - 100°C, maintain the temperature at 120°C in the mold for 1 hour to ensure the preliminary molding of the foam. Then, transfer the preliminarily cured foam to a hot air circulation oven, set the temperature to 150 - 180°C, and maintain for 2 - 4 hours. Control the basic shape of the foam through preliminary curing in the mold. Complete the deep curing in the hot air circulation oven to further enhance the mechanical strength and thermal stability of the foam. The staged curing effectively reduces thermal stress and avoids the generation of foam deformation or internal defects. High-temperature curing promotes the complete reaction of the cross-linking agent with the matrix, improving the service life and electrical conductivity of the foam.

[0065] Step 5, Surface Treatment and Post-treatment: Uniformly spray a silver paste coating on the surface of the foam, control the coating thickness within 5 - 10μm, and heat-treat the treated foam at 180°C for 30 minutes. Spraying the silver paste coating on the surface of the foam improves the surface electrical conductivity and corrosion resistance. Improve the adhesion and electrical conductivity of the coating through heat treatment to ensure that the overall performance of the foam reaches the target. The silver paste coating increases the conductive path on the surface of the foam, further reducing the resistivity. Heat treatment reduces the internal stress of the coating by gradually increasing the temperature, improving the durability and electrical conductivity of the coating.

[0066] Furthermore, in the present invention, the silicone matrix is selected as room temperature vulcanized silicone rubber, the conductive filler is selected as carbon nanotubes or graphene, the cross-linking agent is benzoyl peroxide, the foaming agent is selected as azodicarbonamide or AC foaming agent, and the dispersant is a silane coupling agent.

[0067] In the embodiment, in the ultrasonic dispersion step, the ultrasonic dispersion equipment uses a device with a circulating cooling function, and the temperature during the dispersion process is controlled within the range of 30 - 50°C to avoid adverse effects of local overheating generated by ultrasonic dispersion on the performance of the matrix and conductive filler. Adopt high-speed stirring and ultrasonic dispersion technology to evenly distribute the conductive filler and construct a stable conductive network. Ultrasonic dispersion further improves the uniform distribution effect of the filler, ensuring consistent electrical conductivity.

[0068] In the embodiment, in Step 4, after the foam demolded from the mold during the curing process is cooled to room temperature, it is subjected to surface spraying treatment again. The spraying uniformity is controlled by a spin coating device to ensure that the electrical conductivity of the silver paste coating on the surface of the foam meets the requirement of resistivity ≤ 0.1Ω·cm.

[0069] In the embodiment, during the heat treatment process in Step 5, a programmable temperature control device is used for precise temperature control, and a temperature gradient is set to gradually increase the temperature from 150°C to 180°C over a heating time of 30 minutes, thereby improving the adhesion and electrical conductivity between the silver paste coating and the surface of the foam.

[0070] In the embodiment, Step 3 is specifically as follows:

[0071] Foaming environment control: Transfer the mixture to a closed foaming kettle, set the reaction temperature to 120 - 160 °C, the pressure to 0.1 - 0.3 Mpa, and maintain the constant temperature and pressure conditions for 10 - 20 minutes to ensure uniform decomposition of the foaming agent and form a stable cell structure;

[0072] Crosslinking reaction control: Gradually increase the temperature to 180 - 200 °C and maintain for 30 minutes to promote the complete reaction of the silicone matrix with the crosslinking agent and form a uniform three-dimensional crosslinked network.

[0073] Furthermore, in the present invention, by introducing a foaming environment monitoring and control system and a parameter optimization method based on machine learning, the uniformity and stability of the cells during the foaming process are significantly improved. The foaming environment control in step three is controlled by the monitoring and control system, and the monitoring and control system includes a data acquisition unit, a control unit, and an execution unit, and the control unit includes a single-chip microcomputer;

[0074] The control unit controls according to the following steps:

[0075] Step S1, data acquisition and feature selection: Collect key process parameters and data related to the cell structure during the foaming process to provide a training data set for the machine learning model, collect the input feature X, and the input feature X includes: foaming temperature T f , foaming pressure P f , foaming agent concentration C f , mixing shear rate R m , conductive filler particle size distribution D p , foaming agent release rate S a , and the output feature Y is the cell size uniformity index Us w , with the unit of μm, and is expressed by the formula: Us w : weighted index of cell size uniformity, with the unit of μm, V i : volume of the i-th cell, with the unit of μm 3 , d i : diameter of the i-th cell in μm, average diameter of all cells, N: total number of cells;

[0076] Quantifying the cell uniformity using the formula can accurately measure the distribution of the volume and diameter of the cells. By optimizing parameters such as foaming temperature, pressure, foaming agent concentration, etc., to make Us wReaches the minimum value, thus achieving a highly uniform cell structure and enhancing the stability of material properties. This formula takes into account the influence of cell volume on uniformity, making it more in line with the actual evaluation of material properties. The weighted calculation reduces the excessive influence of small cells on the uniformity index and improves the credibility of the index. It better reflects the influence of cell size on material properties and is suitable for the special requirements of conductive foam.

[0077] Collect process data X i and Y i , forming a data set for model training;

[0078] Step S2, establish a machine learning model, select a regression model to predict the uniformity index U of the cell structure during the foaming process s , regression model formula: f(X; θ) is the model function, representing the non-linear relationship between the cell size uniformity U s and the input feature X. θ is the model parameter, including weights and biases, which need to be optimized through training data. is the predicted cell size uniformity index; By introducing a monitoring and control system, including a thermostat, a pressure controller, a flow meter, etc., precise control of the foaming process is achieved, and the process parameters are optimized in combination with the machine learning model, enhancing the stability and intelligent level of the preparation process. Using the machine learning regression model, analyze the non-linear relationship between the input features and the cell size uniformity, and achieve the optimal model parameters by minimizing the mean squared error MSE. This method can quickly and accurately predict the influence of process parameters on the cell structure, significantly reducing the number of experiments and shortening the development cycle.

[0079] Use the mean squared error (MSE) as the loss function to optimize the model parameter θ: L(θ): The value of the loss function, used to evaluate the error between the model prediction and the actual value. M is the number of training samples, w i : The weight of the i-th sample, used to adjust the influence of different samples on the model. U s，i is the uniformity index measured experimentally. is the uniformity index predicted by the model. λ: Regularization coefficient, used to control the model complexity and avoid overfitting. The L of the model parameter 2 norm, representing the sum of the squares of the weights. By minimizing L(θ), the model can obtain the best parameters;

[0080] This formula improves the fitting ability for key samples through the dynamic setting of w i . The regularization term λ controls the model complexity and improves the generalization ability. It can handle unbalanced data sets and enhance the processing effect for extreme samples.

[0081] Step S3, model optimization and hyperparameter adjustment, adjusting process parameters including foaming temperature T f , foaming pressure P f , foaming agent concentration C f , mixing shear rate R m , particle size distribution D of conductive filler p , foaming agent release rate S a， to minimize the uniformity index U s , while ensuring the stability of the foaming process. The objective function:

[0082] w 1 , w 2 , w 3 : Weight coefficients, used to balance the priorities of uniformity, stability, and cost objectives;

[0083] Predicted cell size uniformity index, the smaller the better;

[0084] S f : Cell stability index, defined as the standard deviation of cell wall thickness, unit μm;

[0085] S min : Set stability threshold, representing the minimum requirement for cell wall thickness uniformity;

[0086] Process cost: The cost of process parameter adjustment. Using genetic algorithm or Bayesian optimization method, combined with model prediction values, to find the optimal parameter combination that meets the process stability conditions. This formula comprehensively considers cell uniformity, stability, and process cost, weighs different optimization objectives, ensures cell performance while reducing costs, and enhances the economy of the preparation process. Introducing genetic algorithm or Bayesian optimization method can quickly find the global optimal solution in the multi-dimensional parameter space and greatly improve the optimization efficiency.

[0087] Furthermore, in the present invention, in step S3, the cell stability index S f is defined as the distribution deviation of cell wall thickness, then:

[0088] S f : Weighted index of cell wall thickness stability, unit μm, V i : Volume of the i-th cell, unit μm 3 , t i is the wall thickness of the i-th cell, is the average wall thickness of all cells, N: number of cells; accurately evaluate the uniformity of cell wall thickness through the formula to ensure the consistency of the foaming structure in mechanical and electrical properties. S fThe decrease indicates that the distribution of the cell wall thickness is more uniform, reducing local stress concentration and enhancing the overall mechanical properties and long-term stability of the foam.

[0089] The above formula takes into account uniformity, stability, and cost simultaneously, ensuring that the optimization results are practical and feasible by adjusting the weight coefficients w 1 ,w 2 ,w 3 to adapt to different optimization requirements. Adding upper and lower limit conditions ensures that the results are within a reasonable range.

[0090] Predict its release rate according to the decomposition kinetic equation of the blowing agent:

[0091]

[0092] Sa opt : Blowing agent release rate, unit: mol / s, k: Reaction rate constant, C f : Blowing agent concentration, unit: wt%, E a : Activation energy of blowing agent decomposition, unit: kJ / mol, R: Gas constant, 8.314 J / (mol·K), T: Reaction temperature, unit: K, D: Blowing agent diffusion rate, unit: μm / s, L: Blowing agent diffusion path length, unit: μm, P target : Target pressure, unit: Pa, P: Real-time pressure, unit: Pa, by adjusting C f and T, the blowing agent release rate can be optimized to improve foaming stability. This formula combines the kinetic model of blowing agent decomposition and can accurately predict the gas release rate Sa opt ,by controlling the combination of C f and TT, making the gas release more uniform. Ensure the stability of the foaming process, avoiding foam collapse caused by too fast gas release or too high closed cell rate caused by too slow release. Adding diffusion rate and pressure factors makes the release rate closer to the actual foaming conditions. It can adjust the blowing agent release according to the real-time process conditions, improving cell uniformity and stability. Suitable for the process requirements of different types of blowing agents and matrix materials

[0093] The above formula is constructed through historical experimental data and can capture complex non-linear relationships. Combining machine learning and optimization algorithms can efficiently find the global optimal process parameters. The formula can be quickly adjusted according to different blowing agents or matrix materials, with strong adaptability. By predicting and optimizing the cell size distribution, the uniformity problem in the foaming process is significantly reduced. By controlling the blowing agent release rate and the cell wall thickness distribution, the stability of the foaming process is ensured. Reducing the number of experiments saves process development time and costs. Through the above steps, using machine learning can not only solve the instability problem in the foaming process but also achieve precise optimization, providing scientific support for the efficient and stable preparation of conductive silicone foam.

[0094] In the embodiment, the data acquisition unit includes:

[0095] A high-precision temperature sensor that monitors the temperature inside the foaming kettle in real time through the high-precision temperature sensor and transmits the temperature data to the single-chip microcomputer; the purpose of collecting the temperature is to monitor the temperature distribution inside the foaming kettle and ensure that the reaction temperature is maintained within a suitable range.

[0096] A strain gauge pressure sensor that monitors the pressure sensor in the foaming kettle in real time through the pressure sensor and transmits the pressure data to the single-chip microcomputer; collecting the pressure in real time is to control the stability of the pressure inside the foaming kettle and prevent uneven cell morphology caused by pressure fluctuations.

[0097] A mass flowmeter that monitors the release rate of the foaming agent in real time by detecting changes in the mass flow rate of the fluid and transmits the monitoring data to the single-chip microcomputer; the data is transmitted to the server in real time for judging whether the decomposition of the foaming agent is uniform and stable.

[0098] A laser particle size analyzer that irradiates particles with a laser, analyzes the intensity distribution of the scattered light to calculate the particle size, and the results can be used to adjust the filler uniformity and transmits the detection data to the single-chip microcomputer; this data is used to adjust the dispersion of the filler to ensure the uniformity of the conductive function.

[0099] A microscopic imager. After foaming is completed, the microscopic imager takes a cross-sectional image of the cell from the material section. The image analysis software calculates the cell diameter and distribution through an edge detection algorithm and outputs the cell size uniformity index U s , which is used to evaluate the overall uniformity of foaming, such as Figure 2 shown.

[0100] In the embodiment, the execution unit includes:

[0101] A PID temperature controller and a heater. The heater is arranged on the foaming kettle. The temperature controller calculates the temperature difference adjustment signal and drives the heater to output the corresponding power. If the temperature is too high, the heating power is reduced, and if it is too low, the power is increased to ensure that the temperature is stable at the set value;

[0102] A solenoid valve and a pressure controller. The solenoid valve is arranged on the foaming kettle. The pressure controller controls the opening degree of the solenoid valve according to the collected pressure data, adjusts the gas injection or discharge rate to maintain the pressure inside the foaming kettle within a stable range, and avoids cell deformation due to pressure fluctuations;

[0103] A variable-frequency stirrer. The variable-frequency stirrer is arranged on the foaming kettle. The variable-frequency motor controls the rotation speed of the stirrer and adjusts the mixing intensity according to the shear rate Rm recommended by the model;

[0104] A flow controller that adjusts the release rate of the foaming agent according to the data collected by the flowmeter, and ensures the stability of the cell structure by controlling the dosage and release time.

[0105] The specific process is as follows:

[0106] Data acquisition and analysis process: First, install and debug the equipment. Install temperature sensors and pressure sensors in the foaming kettle to ensure stable signals. Install a flowmeter in the foaming agent channel to monitor the release of the foaming agent. Configure the data acquisition system to integrate the equipment signals into a single-chip microcomputer, and the single-chip microcomputer transmits the data to the server. Then, conduct experimental operations, run the foaming process, and record the process parameters (temperature, pressure, foaming agent release rate, etc.) in real time. Collect data every fixed time (such as 1 second) to form time-series data. For data analysis, use image analysis software (such as ImageJ) to process microscopic images and extract the cell size distribution, average diameter, and standard deviation. After the data acquisition is completed, the input feature data needs to go through data standardization, noise reduction processing, and feature encoding. Data standardization normalizes the input feature X to the interval [0, 1]. Noise reduction processing uses smoothing filters (such as the moving average method or Kalman filter) to remove the noise during the equipment acquisition process. For categorical features (such as foaming agent types), use one-hot encoding (One-Hot Encoding). For continuous features, keep the numerical form as input. Finally, input the experimental results into the training set of the machine learning model as feature data.

[0107] After the data acquisition is completed, the input feature data needs to go through the following steps for processing: Conduct data analysis. The server runs the machine learning model and substitutes the collected data (input features) into the regression formula: Output the prediction results (cell uniformity and stability indicators). Then, perform feedback adjustment. Based on the model prediction values, adjust the process equipment parameters (such as the temperature of the temperature controller and the stirring speed) in real time to keep the foaming process stable and the cell structure uniform.

[0108] In other words, each sensor transmits the data to the server through the data acquisition system to form a real-time multi-dimensional process data stream. The server runs the machine learning model to predict the cell uniformity index and wall thickness stability, and generates an optimal parameter adjustment plan. Then, perform feedback adjustment. The temperature controller, pressure controller, and flow controller dynamically adjust the parameters according to the analysis results. The equipment adjustment process continues to cycle until the predicted value is consistent with the target value. Confirm whether the cell uniformity meets the expectations through microscopic images and image analysis. If the deviation is large, adjust the model training set and re-optimize.

[0109] Generally speaking, through the uniform dispersion of conductive fillers and the optimization of coating treatment, the above embodiments significantly improve the electrical conductivity of the foam. It has a uniform and stable cell structure, multiple parameter control and optimization strategies, significantly reducing the non-uniformity of cell size and wall thickness. The precise cross-linking and curing process ensures the heat resistance and mechanical properties of the foam. By using machine learning to optimize parameters, the test time and material waste are reduced, and the production efficiency is improved.

[0110] The embodiments of the present invention also designed the following experiments to verify that through multiple parameter control and machine learning optimization strategies, the non-uniformity of cell size and wall thickness can be significantly reduced, and the production efficiency can be improved, while reducing the test time and material waste.

[0111] Experimental Design

[0112] Materials and Equipment

[0113] Silicone matrix: room temperature vulcanized silicone rubber; conductive filler: graphene; cross-linking agent: benzoyl peroxide; foaming agent: azodicarbonamide; dispersant: silane coupling agent; equipment: variable frequency stirrer, ultrasonic disperser (with cooling function), foaming kettle (constant temperature and pressure control), microscopic imager, laser particle size analyzer, etc.

[0114] Experimental Procedures

[0115] 1. Experimental Group Settings

[0116] A single variable comparison experimental design was adopted, and the following groups were set up:

[0117] Examples 1-3: Foaming was carried out according to the method of the present invention with parameters optimized by machine learning.

[0118] Comparative Examples 1-2: Machine learning optimization was not adopted, and foaming was carried out only based on empirical parameters.

[0119] 2. Machine Learning Model Training A regression model was trained through historical experimental data and real-time monitoring data (such as foaming temperature, pressure, shear rate, etc.) to optimize the cell size uniformity index and cell wall thickness stability index.

[0120] Experimental Results

[0121] Table 1 shows the comparison of cell size uniformity and wall thickness stability

[0122]

[0123] Analysis and Conclusion: The cell size uniformity index of the example group optimized by machine learning is significantly lower than that of the comparative example group, and the uniformity is improved by about 40%-50%. The cell wall thickness stability index of the example group is significantly optimized, reducing the wall thickness deviation by about 50%-60% and enhancing the stability of the cell structure.

[0124] The resistivity of the example group is lower (≤0.1 Ω·cm), indicating that the distribution of the conductive network is more uniform.

[0125] The average material waste rate of the example group is 2.2%, which is much lower than 7% of the comparative example. The number of tests is reduced by about 50%-60%, proving that machine learning optimization significantly shortens the R & D cycle.

[0126] Through multi-parameter control and machine learning optimization, the uniformity of cell size and wall thickness can be significantly improved, the number of tests and material waste can be reduced, and the production efficiency can be significantly increased. This method has strong adaptability and can be extended to the preparation process of various foaming materials.

[0127] The embodiments of the present invention also designed the following experiments to verify that by optimizing the dispersion method of conductive fillers, coating treatment, and controlling the cross-linking and curing processes, the conductive performance of silicone foam can be significantly improved while ensuring its heat resistance and mechanical properties.

[0128] Experimental materials

[0129] Silicone matrix: room temperature vulcanized silicone rubber; conductive filler: carbon nanotubes or graphene; cross-linking agent: benzoyl peroxide; foaming agent: azodicarbonamide (AC foaming agent); dispersant: silane coupling agent; silver paste coating: for conductive surface treatment.

[0130] Experimental steps

[0131] Raw material preparation: Prepare materials according to the formula ratio.

[0132] Pretreatment of conductive filler: The conductive filler is dried at 80-100 °C for 2 hours and washed with ethanol.

[0133] Mixing and dispersion: Use a variable frequency stirrer to pre-disperse the filler, and then further uniformly distribute it through ultrasonic dispersion.

[0134] Add cross-linking agent and foaming agent: Stir evenly at low speed.

[0135] Foaming and cross-linking reaction: Carry out in a closed environment, controlling temperature and pressure.

[0136] Curing and molding: Cure and mold at different temperatures, and finally heat-treat at 180 °C for 2-4 hours.

[0137] Surface coating treatment: Spray silver paste coating and carry out heat treatment.

[0138] Experimental examples and comparative experiments

[0139] Example 4: According to the method of the present invention, use conductive foam with ultrasonic dispersion and optimized coating treatment.

[0140] Comparative example 3: Without ultrasonic dispersion, only use mechanical stirring.

[0141] Comparative Example 4: Without silver paste coating treatment, only the matrix material is used.

[0142] Table 2 Performance test results:

[0143]

[0144] Conclusion analysis: In Example 4, due to the uniformly dispersed conductive fillers and the optimized silver paste coating, the conductivity is greatly improved, and the resistivity reaches 0.08 Ω·cm; while the conductivity of the untreated Comparative Example 3 and Comparative Example 4 is poor, being 0.15 Ω·cm and 0.20 Ω·cm respectively.

[0145] Example 1 has better heat resistance and tensile strength, indicating that the cross-linking and curing processes ensure the structural stability and thermal stability of the silicone foam. The ultrasonic dispersion process significantly improves the uniformity of the foam cells in Example 1. Through these data, it can be demonstrated that uniformly dispersing conductive fillers and optimizing the coating treatment significantly improve the conductivity of the foam, and the precise cross-linking and curing processes ensure its heat resistance and mechanical properties.

[0146] Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Those with ordinary knowledge in the technical field to which the present invention pertains can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be determined by the scope defined in the claims.

Claims

1. A method for preparing a silicone foam having a conductive function, characterized in that: The steps include: Step 1, raw material preparation, preparing an organosilicon matrix, a conductive filler, a crosslinking agent, a foaming agent and a dispersing agent, wherein the amounts of the organosilicon matrix, the conductive filler, the crosslinking agent, the foaming agent and the dispersing agent are 70-80%, 5-15%, 1-2%, 3-5% and 2-5% of the total mass respectively; Step 2: Mixing and dispersing: pretreat the conductive filler first, place the conductive filler in a vacuum drying oven, dry it at 80-100°C for 2 hours to remove moisture, and clean the surface of the conductive filler with ethanol or acetone; Prepare a basic mixture, add the silicone matrix, dispersant and dried conductive filler into a mixing container in proportion in a variable frequency stirrer, stir at a speed of 1500-2000 rpm for 10-20 minutes to preliminarily disperse the conductive filler; Ultrasonic dispersion: transfer the mixture to an ultrasonic dispersion device, set the power to 500-1000W, and the dispersion time to 30 minutes to ensure that the filler is evenly distributed in the matrix to form a stable conductive network; Add the crosslinking agent and the foaming agent, add the crosslinking agent and the foaming agent to the base mixture in sequence under low-speed stirring conditions of 500-800 rpm, and stir evenly for 5-10 minutes; Step 3: foaming and cross-linking reaction, foaming environment control and cross-linking reaction control; Step 4: Curing and molding: First, quickly inject the foamed mixture into a mold preheated to 80-100°C, maintain the temperature at 120°C in the mold, and cure for 1 hour to ensure that the foam is initially formed. Move the initially cured foam to a hot air circulation oven, set the temperature to 150-180°C, and keep it for 2-4 hours. Step 5: Perform surface treatment and post-treatment, spray a silver paste coating evenly on the foam surface, control the coating thickness to 5-10 μm, and heat treat the treated foam at 180° C. for 30 minutes.

2. The method for preparing a conductive silicone foam according to claim 1, characterized in that: The organic silicon matrix is ​​selected from room temperature vulcanized silicone rubber, the conductive filler is selected from carbon nanotubes or graphene, the crosslinking agent is selected from benzoyl peroxide, the foaming agent is selected from azodicarbonamide or AC foaming agent, and the dispersant is a silane coupling agent.

3. The method for preparing a conductive silicone foam according to claim 1, characterized in that: In the ultrasonic dispersion step, the ultrasonic dispersion equipment is equipped with a circulating cooling function, and the dispersion process temperature is controlled within the range of 30-50° C. to avoid local overheating caused by ultrasonic dispersion from adversely affecting the performance of the matrix and the conductive filler.

4. The method for preparing a conductive silicone foam according to claim 1, characterized in that: In step 4, the foam demoulded from the mold during the curing process is cooled to room temperature and then sprayed again. The spraying uniformity is controlled by a rotary coating device to ensure that the conductivity of the silver paste coating on the foam surface meets the resistivity requirement of ≤0.1Ω·cm.

5. The method for preparing a conductive silicone foam according to claim 1, characterized in that: During the heat treatment process in step five, a programmable temperature control device is used for precise temperature control, and a temperature gradient is set, gradually increasing the temperature from 150°C to 180°C within 30 minutes, thereby improving the adhesion and conductivity between the silver paste coating and the foam surface.

6. The method for preparing a conductive silicone foam according to claim 1, characterized in that: The step three is specifically as follows: Foaming environment control: transfer the mixture to a closed foaming kettle, set the reaction temperature to 120-160°C, the pressure to 0.1-0.3Mpa, and maintain constant temperature and pressure conditions for 10-20 minutes to ensure uniform decomposition of the foaming agent and form a stable pore structure; Cross-linking reaction control: the temperature is gradually increased to 180-200°C and maintained for 30 minutes to promote complete reaction between the silicone matrix and the cross-linking agent to form a uniform three-dimensional cross-linking network.

7. The method for preparing a conductive silicone foam according to claim 6, characterized in that: The foaming environment control in step 3 is controlled by a monitoring and control system, which includes a data server, a collection unit, a control unit and an execution unit, and the control unit includes a single chip microcomputer; The control unit uses the following steps to perform control: Step S1, data collection and feature selection, collect key process parameters and cell structure related data in the foaming process, provide a training data set for the machine learning model, collect input features X, and input features X include: foaming temperature T f , Foaming pressure P f , Foaming agent concentration C f , mixing shear rate R m , Conductive filler particle size distribution D p , foaming agent release rate S a , the output feature Y is the cell size uniformity index Us w , in μm, expressed by the formula: Us w : Weighted index of cell size uniformity, in μm, V i : The volume of the ith cell, in μm 3 , d i : The diameter of the ith cell μm, The average diameter of all cells, N: the total number of cells; Collect process dataX i and Y i , forming a data set for model training; Step S2, establish a machine learning model and select a regression model to predict the uniformity index U of the cell structure during the foaming process. s , regression model formula: f(X;θ) is the model function, which represents the uniformity of cell size U s The nonlinear relationship with the input feature X, θ is the model parameter, including weights and biases, which needs to be optimized through training data. is the cell size uniformity index predicted by the model; The mean square error (MSE) is used as the loss function to optimize the model parameters θ: L(θ): loss function value, used to evaluate the error between model prediction and actual value, M is the number of training samples, w i : The weight of the i-th sample, used to adjust the impact of different samples on the model, U s，i is the uniformity index measured experimentally, is the uniformity index of model prediction, λ is the regularization coefficient, which is used to control the complexity of the model and avoid overfitting. The L2 norm of the model parameters represents the sum of squared weights. By minimizing L(θ), the model can obtain the optimal parameters. Step S3: Model optimization and hyperparameter adjustment, adjusting the process parameter foaming temperature T f , Foaming pressure P f , Foaming agent concentration C f , mixing shear rate R m , Conductive filler particle size distribution D p , foaming agent release rate S a， To minimize the uniformity index U s , while ensuring the stability of the foaming process, the objective function is: w1, w2, w3: weight coefficients used to balance the priorities of uniformity, stability, and cost objectives; The predicted cell size uniformity index, the smaller the better; S f : Cell stability index, defined as the standard deviation of cell wall thickness, in μm; S min : The set stability threshold represents the minimum requirement for uniformity of cell wall thickness; Process cost: The cost of adjusting process parameters, using genetic algorithms or Bayesian optimization methods, combined with model prediction values, to find the optimal parameter combination that meets the process stability conditions.

8. The method for preparing a conductive silicone foam according to claim 7, characterized in that: In the step S3, the cell stability index S f Defined as the distribution deviation of cell wall thickness, then: S f : Weighted index of cell wall thickness stability, in μm, V i : The volume of the ith cell, in μm 3 , t i is the wall thickness of the ith cell, is the average wall thickness of all cells, N: number of cells; The release rate of the foaming agent is predicted based on the decomposition kinetic equation: Sa opt : Foaming agent release rate, unit is mol / s, k: reaction rate constant, C f : Foaming agent concentration, in wt%, E a : activation energy of decomposition of blowing agent, in kJ / mol, R: gas constant, 8.314 J / (mol·K), T: reaction temperature, in K, D: blowing agent diffusion rate, in μm / s, L: blowing agent diffusion path length, in μm, P target : target pressure, in Pa, P: real-time pressure, in Pa, by adjusting C f and T, the blowing agent release rate can be optimized to improve the foaming stability.

9. The method for preparing a conductive silicone foam according to claim 8, characterized in that: The data acquisition unit comprises: High-precision temperature sensor, which monitors the temperature in the foaming kettle in real time and transmits the temperature data to the single-chip microcomputer; The strain gauge pressure sensor monitors the pressure sensor in the foaming kettle in real time through the pressure sensor and transmits the pressure data to the single chip microcomputer; The mass flow meter monitors the release rate of the foaming agent in real time by detecting the change in the fluid mass flow rate and transmits the monitoring data to the single chip microcomputer; Laser particle size analyzer, which uses laser to illuminate particles and analyzes the intensity distribution of scattered light to calculate particle size. The results can be used to adjust the uniformity of the filler and transmit the detection data to the microcontroller; Microscopic imager: After foaming is completed, the microscopic imager takes a cross-sectional image of the pores from the material slice. The image analysis software calculates the pore diameter and distribution through the edge detection algorithm and outputs the pore size uniformity index U s , used to evaluate the overall uniformity of foaming.

10. The method for preparing a conductive silicone foam according to claim 9, characterized in that: The execution unit comprises: PID thermostat and heater, the heater is set on the foaming kettle, the thermostat calculates the temperature difference adjustment signal, drives the heater to output corresponding power, if the temperature is too high, the heating power is reduced, if it is too low, the power is increased to ensure that the temperature is stable at the set value; A solenoid valve and a pressure controller, wherein the solenoid valve is arranged on the foaming kettle, and the pressure controller controls the opening degree of the solenoid valve according to the collected pressure data, and adjusts the gas injection or discharge rate to maintain the pressure in the foaming kettle within a stable range and avoid deformation of the foam cells due to pressure fluctuations; A variable frequency stirrer, which is arranged on the foaming kettle, and the variable frequency motor controls the stirrer speed, and adjusts the mixing intensity according to the shear rate Rm recommended by the model; A flow controller adjusts the release rate of the foaming agent according to the data collected by the flow meter, and ensures the stability of the foam structure by controlling the dosage and release time.

Images