Discussion and Application of High Temperature Resistant PI Polyimide Materials
In the era of highly thin, multifunctional, and integrated electronic devices, it is inevitable to accumulate heat inside composite materials, which seriously affects the stable operation and service life of equipment. How to achieve fast and efficient thermal conductivity and heat dissipation of dielectric materials has become a key issue affecting the development of electronic devices. The traditional polyimide has a low intrinsic thermal conductivity, which limits its application in fields such as electrical equipment and smart grids. The development of new high thermal conductivity polyimide dielectric film materials has become a research focus both domestically and internationally. This article introduces the thermal conduction mechanism of composite materials, summarizes the research progress and development status of thermal conductive polyimide films in recent years, and focuses on the influence of thermal conductive fillers, interface compatibility, and molding processes on the thermal conductivity of materials. Finally, combined with the future development needs of thermal conductive polyimide composite dielectric materials, some key scientific and technological issues in research are summarized and discussed.
01 Introduction
Polymer materials are widely used in fields such as electronics, communications, military equipment manufacturing, aerospace, etc. due to their excellent electrical insulation, chemical corrosion resistance, lightweight, and low density. Polyimide (PI) is an aromatic heterocyclic polymer compound constructed from chain links containing imide groups [- C (O) - N (R) - C (O) -]. It has excellent electrical insulation, radiation resistance, mechanical properties, and is known as a "problem-solving expert". PI, as a structural or functional material, has great development prospects, especially for PI film materials, which are known as the "golden film". It is one of the earliest polyimide products developed and applied, widely used in printed circuit boards, electronic packaging, interlayer media, display panels and other fields (see Figure 1).
The high integration and high power of modern electronic devices, industrial devices represented by chips, hybrid electric vehicles, and light-emitting diodes have led to a gradual reduction in product size, resulting in an increasingly prominent problem of doubled heat generation, which seriously affects the operational performance and service life of products. The efficient thermal conduction and heat dissipation of thermal management systems have attracted widespread attention.
Related studies have shown that for every 2 ℃ increase in temperature of electronic devices, reliability decreases by 10%; The temperature increase of 8-12 ℃ reduces the service life by half, and the thermal conductivity of the material has become an important parameter affecting the normal operation of the equipment. Polymer materials have shown great potential in solving thermal conductivity and heat dissipation problems, but the intrinsic thermal conductivity of polyimide materials is relatively low, usually below 0.2 W/(m · K), far lower than materials such as metals, carbon, ceramics, etc., greatly limiting the application of PI films in high-tech fields. It is of great significance to seek appropriate methods to improve the thermal conductivity of polyimide materials in order to ensure the normal operation and safety of equipment. In order to solve the thermal conductivity and heat dissipation problems of polyimide materials, researchers mainly carry out work from two aspects. One is to modify the PI matrix body, starting from the perspective of molecular structure design, based on PI's 1-3 level structure design and construction of ordered structures; Inducing the formation of ordered structures through mechanical stretching, shearing, centrifugation, spinning, and other methods; Based on intermolecular interactions, especially the advantage of hydrogen bonding, interpenetrating and entanglement structures are formed between molecular chains, as well as hydrogen bonding interactions between side groups. The strategy to improve the intrinsic thermal conductivity of polyimide is to change the morphology of the matrix chain structure, so that the curled molecular chains present a stretched state, improve the orderliness of chain segment aggregation, and create a pathway for phonon transmission, thereby improving the intrinsic thermal conductivity of the matrix.
The second is to use PI as the matrix and add high thermal conductivity fillers in the matrix, which is also an effective strategy to improve thermal conductivity. Currently, theoretical research and industrial production of high thermal conductivity polyimide composites at home and abroad mainly focus on filled PI composites. Thermal conductive fillers are interconnected in the PI matrix to form ordered thermal pathways, reducing scattering during phonon transfer and achieving rapid heat transfer.
The thermal conductivity of composite materials is determined by factors such as the structure of the PI matrix and the properties of the fillers, the arrangement of fillers in the matrix, and the interaction between the matrix and fillers. At the same time, the influence of the construction of thermal conductivity pathways and preparation processes on the thermal conductivity of materials should also be considered.
02 Heat conduction mechanism
Heat is the energy generated by the movement, rotation, and vibration of microscopic particles such as molecules, atoms, and electrons inside a material. The thermal conductivity mechanism of a material is closely related to the collision and transfer of microscopic particles inside it. The carriers of heat conduction include molecules, electrons, phonons (energy quanta of lattice vibrations), and photons. Heat is transferred from the high-temperature part of the material to the low-temperature part, and in essence, it can be considered as molecules and atoms with larger amplitudes driving molecules and atoms with smaller amplitudes to vibrate. The conduction process is shown in Figure 2.
图 2 粒子碰撞在材料中的热传导
The thermal conduction mechanism of different materials is different, mainly depending on the role of the thermal carrier in the material. There are a large number of freely moving electrons inside the metal, which transfer heat through interactions or collisions. Metals are also crystals, and the process of heat conduction is completed through the vibration of the lattice, that is, phonon conduction still exists. However, the heat transfer efficiency of free electrons is much higher than that of phonon heat transfer. Therefore, the heat conduction carrier of metals is mainly electrons.
In non-conductive crystals, molecules or atoms are orderly distributed on the lattice, and the thermal conduction mode is mainly phonon conduction. Its thermal conductivity mainly depends on the degree of crystallization and orientation of the material, and is believed to depend on the scattering degree of phonons from a mechanistic perspective. The main reasons for phonon scattering are: high entanglement of molecular chains, voids in molecular structures, interface and structural defects, and weak interactions between molecular chains.
The static scattering of phonons is caused by various defects, while the dynamic scattering is caused by the non harmonic vibration of molecular chains. The rotation of molecular chains and the entanglement between them can intensify the non harmonic vibration. At the same time, the various conformations generated by the rotation within the chain segment can also cause phonon scattering.
Most polymers are saturated systems where there are no free moving electrons or intense collisions between electrons, and heat is mainly transferred through phonons. Molecular chains vibrate when heated, and thermal conduction mainly relies on the thermal vibrations around molecules or atoms at fixed positions, transferring heat to adjacent molecules or atoms in sequence. The thermal conduction of polymers is shown in Figure 3.
图 3 聚合物的导热机理
Polymers have the characteristics of complex and easily entangled molecular chains, high molecular weight dispersion, and large molecular weight. The crystallinity is not very high, and it is very difficult for phonons to propagate and move internally. In addition, defects, interfaces, voids in the crystal structure, as well as disordered parts in amorphous materials, will cause phonon scattering, which has a significant impact on the thermal conductivity of composite materials. Therefore, the thermal conductivity of polymers is generally low, and the thermal conductivity of common polymers is shown in Table 1.
表 1 常见聚合物的热导率
Polyimide is mainly prepared by the reaction of dianhydride and diamine. Common synthesis methods include one-step method, two-step method, three-step method, and gas-phase deposition method. The two-step method is widely used in laboratory and industrial production. The most commonly used dianhydride and diamine are phthalic anhydride (PMDA) and diamine (ODA), respectively. Taking this as an example, the route for synthesizing phthalic anhydride type PI is shown in Figure 4, where DMAC is N, N-dimethylacetamide, used as a solvent; Thermal imidization refers to thermal imidization.
Like polymers such as epoxy resin, polyvinylidene chloride, and polydimethylsiloxane, polyimide has a long-chain molecular composition and randomly arranged molecular structure. The main chain of the molecule contains a large number of aromatic and nitrogen-containing pentagonal rings, as well as a certain number of ether bonds. From the perspective of rigidity and flexibility, polyimide molecules exhibit high rigidity. The conjugation effect of the rigid structure and aromatic heterocycles endows PI with excellent heat resistance and stability, while also suppressing the internal rotation of the chain links and reducing phonon scattering to a certain extent. There are also studies indicating that the thermal conductivity of polymer chains is closely related to the type of monomer. Polymers with aromatic rings have a thermal conductivity that is even five times that of polyethylene. However, due to the low bonding strength and uneven mass distribution of polymer chains, the thermal conductivity of the material is greatly reduced.
In addition, due to the difficulty in effectively controlling the progress of the polymerization reaction, regulating the composition of the crystalline and amorphous regions, and preventing the production of by-products, the entanglement of polyimide molecular chains, uneven molecular weight distribution, defects, voids, impurities, etc. in the amorphous structure all cause changes in the energy, momentum, and direction of motion of phonons during conduction, thereby reducing the average free path of phonons.
In addition, the symmetry of PI molecules prepared with different dianhydride and diamine is different, and the structure and arrangement of side groups, as well as the introduction of asymmetric chain links, can weaken the directional arrangement of the molecular structure, reduce the degree of ordered stacking in space, and have a serious impact on the improvement of thermal conductivity of polyimide. At present, the main theories explaining the thermal conduction mechanism of polymer composite materials include: thermal conductivity pathway theory, thermal conductivity percolation theory, and thermal elasticity coefficient theory. The theory of thermal conductivity pathway is most commonly used to explain the thermal conductivity mechanism of filled polymers. This theory suggests that the formation of thermal conductivity pathway is due to the contact between the thermal conductivity filler and the interior of the polymer matrix, and the heat flow is transmitted through phonons along paths or networks with lower thermal resistance.
When the filler content of the system is low, the fillers are separated from each other, the spacing between particles is large, the interaction is weak, and a continuous thermal conductivity path cannot be formed in contact with each other. The fillers are wrapped by the matrix, forming a structure similar to a "sea island". Heat is transmitted along the path of matrix filler matrix in the polymer matrix, and phonons scatter at the unconnected parts of the fillers, which is not conducive to improving the thermal conductivity of the material.
As the filler content increases, the fillers come into contact with each other, forming a thermal chain locally, and heat propagates along the thermal conduction path composed of thermal conductive particles. As the filler further increases, local thermal conductive chains overlap with each other to form a complete thermal conductive network, and heat is transferred along the network through phonons, as shown in Figure 5.
03 Simulation of Thermal Conductivity of Composite Materials
The selection of polymer matrix and thermal conductive filler is diverse, and appropriate models or methods can limit the number of experiments and predict the performance of candidate materials. Therefore, a mature thermal conductivity theory for composite materials has guiding significance for the conduct of scientific experiments. It helps guide us to conduct in-depth research on the intrinsic mechanisms of thermal conductivity, factors affecting the thermal conductivity of composite materials, the relationship between structural design and performance balance, and the screening of potential materials with more development prospects, providing strong theoretical support for the development of new high conductivity thermoelectric dielectric composite materials.
The thermal conductivity model has certain reference value for a deeper understanding of the factors that affect the thermal conductivity performance of materials and the design of specific thermal conductivity materials. The construction of the thermal conductivity model is closely related to factors such as the composition, structure, and properties of the matrix itself and the thermal conductivity filler. Therefore, theoretical research and model building play an important role in analyzing and screening materials. Thermal conductivity models can typically predict the "ideal" thermal conductivity of many simple systems, but these models are mostly based on assumptions such as phonons passing through interfaces without thermal resistance and ideal filler morphology, and the calculations of the models are often qualitative.
Many scholars have constructed composite material thermal conductivity models with added filler particles based on different assumptions. Xu Xingxing, Jiang Tao, and others have summarized in detail the widely used particle filler thermal conductivity models. This article will not provide too much description in this section. In addition to the research on particle packing models, scholars have also studied fiber packing models and multi shape packing models. These models can determine the dependence of thermal conductivity on the loading amount of fillers and the thermal conductivity of fillers and matrices.
However, most models are only applicable to specific systems, and the model fitting results are not completely consistent with experimental data. Moreover, polymer based thermal conductive composite materials are developing towards the complexity of filler filling and matrix diversity. Simply studying theoretical models has certain limitations in predicting the thermal conductivity of real materials. For more complex composite material systems, simulation methods such as molecular dynamics models (MD), finite element integration simulations (FEA), and machine learning (ML) can better describe the thermal management of actual materials. Molecular dynamics simulation is mainly used to evaluate and predict the thermal conductivity of materials at the atomic and molecular level, and is widely applied to explore the thermal transport properties of substances. This method can effectively study the relationship between the thermal conductivity behavior and microstructure of graphene composite materials, and can better handle defective research systems.
Xu Jingcheng et al. studied the size effect of graphene/polyimide composite films using the opt Tesoff molecular force field and non-equilibrium molecular dynamics methods. They extended the micro scale graphene to the macro scale heat transfer system and explained the thermal conductivity differences caused by the composite film structure using phonon density of states. The thermal conductivity calculation model is shown in Figure 6, and the results are helpful for the design and large-scale preparation of high thermal conductivity polyimide films.
图 6 导热系数的计算模型
Molecular dynamics not only studies the relationship between thermal conductivity and microscopic properties of materials, but also calculates macroscopic performance parameters of materials. Wang et al. studied PI/boron nitride nanosheets (BNNS) and calculated the thermal conductivity, bulk modulus, dielectric constant, and breakdown strength of PI composites with different BNNS contents at different temperatures. In addition, the interfacial interaction strength was analyzed by calculating the surface binding energy of PI/BNNS composite materials, revealing the micro mechanism of the influence of filler doping on material properties from an energy perspective.
The thermal conductivity model analyzes the thermal conductivity of composite materials under various ideal conditions, limited to some special conditions. However, the actual shape of materials is more complex, and finite element simulation and machine learning can provide more accurate analysis and prediction. Finite element simulation accurately calculates the interface thermal resistance or thermal conductivity of composite materials based on the thermal conductivity model, taking into account the actual material interface, filler dispersion state, and the construction of the thermal conductivity network. Machine learning is training, screening, and systematically identifying materials with high thermal conductivity based on a large amount of data.
Guo et al. studied the columnar carbon nanotube (CPEG)/PI composite material grown on exfoliated graphite, and used finite element method to simulate the nano/micro scale heat transfer process of CPEG/PI material. The simulation results are shown in Figure 7, revealing the internal reasons for the enhanced thermal conductivity from a microscopic perspective. Based on the effective medium theory and the law of conservation of energy, a more widely applicable calculation model and empirical formula for the effective thermal conductivity of polymer based composite materials have been established, which better reflects the actual trend of thermal conductivity changes in CPEG/PI composite materials.
Ding et al. used multi dispersed hexagonal boron nitride (ae BN)/PI as the research system and predicted the in-plane thermal conductivity of ae BN/PI composite materials with different contents at different temperatures using machine learning techniques. By first principles analysis, the relationship between the inherent thermal conductivity of ae BN and temperature and thickness was derived. Based on this, considering the effects of operating temperature and filler content, a modified iterative EMT model was established to demonstrate the reduction of boundary thermal resistance in composite materials. The use of machine learning techniques in molecular design and material screening can predict the effective properties of new structures and accelerate the discovery of novel polymer functional materials.
Inspired by machine learning assisted development of polymers, researchers used trained molecular design algorithms to identify structure property relationships related to thermal conductivity and other target polymer properties from the perspective of the influence of chemical groups and spatial conformation on phonon scattering processes. They identified thousands of promising polymers and successfully discovered materials with high thermal conductivity.
Regarding the simulation of thermal conductivity of composite materials, many scholars have conducted in-depth research based on theoretical prediction, model establishment, and device verification. The theory of thermal conductivity model research originated from Fourier's law, mainly studying the influence of the type, content, geometric shape, orientation, and aggregation state of fillers in filled thermal conductive polymers on the material's thermal conductivity. Among them, the thermal conductivity model of carbon nanotube/polymer system has been derived, mainly including mixed rule thermal conductivity model, homogeneous model, effective medium model, etc. These thermal conductivity models can preliminarily predict the thermal conductivity values of materials, but different models are based on certain theoretical assumptions and approximate or ignore the data. In addition, the thermal conductivity of materials is highly dependent on their microstructure and volume fraction, resulting in a certain gap between experimental and predicted values. But these simplified models are not without research significance. Finite element analysis is based on theoretical models and considers the influence of certain inherent properties of materials on thermal conductivity. At present, research on thermal conductivity models mainly focuses on further considering the influence of other factors on thermal conductivity based on previous models, constantly upgrading and improving the thermal conductivity models, and enriching the thermal conductivity theory of filled polymer composites.
Zhu et al. proposed a universal and easy-to-use effective thermal conductivity model by extending Meredith and Tobias' models, which can predict the thermal conductivity of non-contact spherical filler polymer materials. However, further research is needed to introduce shape factors and consider the influence of contact thermal resistance. Dong et al. proposed a new thermal conductivity model, the series parallel model, based on the classic series parallel model (Agari model). By introducing new parameters, the model was modified to have a higher fit with experimental data in the characteristic system. The introduction of new parameters improves the accuracy of prediction, but also reduces the generality of the model. The development of future thermal conductivity models should balance the accuracy of prediction and the universality of the model, and make more in-depth corrections to data calculations in order to provide more direct references for the preparation of high thermal conductivity composite materials.
Equilibrium Molecular Dynamics (EMD) and Non Equilibrium Molecular Dynamics (NEMD) are two methods for studying thermal conductivity in molecular dynamics. The EMD method is based on the Green Kubo formula and has strong robustness to the calculation details of external parameters, making it easy to use. However, it often leads to an independent relationship between thermal conductivity and simulation unit size. In addition, it is difficult to obtain thermal conductivity of materials with different shapes and sizes using EMD method. The NEMD method is based on Fourier's law and can explore the thermal conductivity of materials with different shapes, but its calculation process is more complex and difficult to determine the thermal conductivity of bulk materials. In practical applications, some scholars have used molecular dynamics simulation to quantitatively analyze the thermal conductivity of composite materials with structural damage and vacancy defects, evaluate the influence of different parameters on thermal conductivity behavior, and explore the effect of temperature gradient on thermal conductivity. MD simulation will develop towards higher computational efficiency, larger simulation systems, and more accurate simulation results. In future research, there should be a trend towards combining multiple simulation methods to improve the credibility of data and the accuracy of predictions. Ahmadi et al. used molecular dynamics simulations to evaluate the thermal conductivity of graphene sheets, and then evaluated the thermal conductivity of graphene reinforced polypropylene at different temperatures through finite element analysis. The use of finite element analysis to simulate heat transfer processes is widely used in the field of thermal conductivity to verify the actual thermal conductivity of materials. Generally, polyimide or other polymer films prepared are used as thermal interface materials between light emitting diode (LED) chips and heat sinks. The temperature changes of the LED over time are recorded by infrared thermography, and the heat dissipation and thermal management capabilities of the material are analyzed by finite element integration simulation. The influence of the structure of the prepared material on thermal conductivity and the mechanism of thermal conductivity enhancement are elucidated. For filled polymer composite materials, the addition of filler particles will inevitably have a certain impact on the mechanical properties of the material. Finite element analysis can simulate the stress distribution in the structural model of composite materials, thereby more profoundly explaining the improvement of material mechanical properties and elucidating the relationship between mechanical properties and thermal conductivity. Other applications of finite element simulation in the field of thermal conductivity include simulating the heat transfer of composite materials, characterizing and analyzing the in-plane and out of plane thermal conductivity behavior of 3D woven composite materials, and establishing new theoretical models for pore and particle shapes. Compared with the thermal conductivity theory model, finite element analysis can simulate various geometrically complex structures, consider the influence of interface thermal resistance and the construction of thermal conductivity pathways on thermal conductivity, and can also be used to prove the proposed theoretical model. Finite element analysis and optimization design are usually combined to reduce simulation calculation time and improve prediction accuracy.
But for calculations involving more complex systems and analysis of more complex problems, such as systems containing multiple thermal conductive filler particles, multiple types of interfaces, and all organic thermal conductive materials, using finite element analysis methods for prediction will consume a lot of time. In addition, in specific applications, it is necessary to try and screen what type of unit to use, what network density to have, etc. The computational complexity is astonishing, and the accuracy of simulation results will also be reduced. Unlike finite element simulation analysis, the application of machine learning technology in the field of thermal conductive polymers mainly combines materials science with big data to develop functional new advanced thermal management materials. It can also further analyze the potential quantitative relationship between the molecular structure and thermal conductivity of polymers, establish a structure performance relationship model, and reveal the correlation between the thermal conductivity and structure of materials. Machine learning technology can quickly and efficiently predict the thermal conductivity and interfacial thermal resistance of materials, and can also be used to predict the thermal conductivity of various single chain polymers, predict the thermal conductivity of locally sourced materials, such as estimating the thermal conductivity of rocks, concrete, and soil. Machine learning technology requires the establishment of a thermal conductivity database through a large number of data points. These databases are universal for thermal conductivity simulations of the same type of material. Therefore, the establishment and improvement of each database provide guidance and reference for future related research. This simulation technology can perform visual analysis and easily obtain the required data and results. When testing the dataset, its running speed is very fast, but some missing data can make the processing process difficult. At present, the use and development of machine learning emphasizes practical applications and algorithms, and lags behind in theoretical research.
It can be foreseen that in future research, the use of molecular dynamics simulation, finite element simulation, and machine learning techniques to study the relationship between thermal conductivity and structure of composite materials will be greatly expanded, and will develop towards diversified fillers and complex formulations. With the continuous upgrading of methods and the improvement of computer simulation speed, the accuracy of evaluating thermal conductivity and the ability to screen materials will also be greatly improved.
04 Thermal conductivity of composite materials
High thermal conductivity polymers can be divided into intrinsic and filling types according to the preparation process. Intrinsic thermal conductive polymer refers to the process of molecular design, synthesis, and molding, in which certain special structures are obtained by regulating the morphology and structure of molecular chains, inter chain interactions, and promoting ordered orientation, thereby promoting the path of thermal conduction, reducing phonon scattering, and improving the thermal conductivity of materials. Intrinsic thermal conductive polymers are limited in their wide application in the field of thermal conductivity due to their complex preparation process, limited raw material selection, uncontrollable reaction conditions, numerous by-products, and high cost. But some scholars are also committed to developing intrinsic thermal conductive polymers, considering the effects of molecular chain structure, orientation and crystallinity, intermolecular interactions, liquid crystal elements, temperature and pressure on thermal conductivity, in order to develop new thermal conductive materials. The preparation method of filled thermal conductive composite materials is relatively simple, the selection of fillers is wide, and the improvement of thermal conductivity is relatively effective, which has become a focus of research at home and abroad.
At present, the main technical problems faced in the field of filled thermal conductive composite materials include: (1) compatibility and effective dispersion of inorganic thermal conductive fillers in polymer matrices; (2) Interface thermal resistance between fillers/matrix and fillers/fillers; (3) Design and effective construction of thermal conductivity pathways. In response to the above challenges, scholars have proposed many solutions, including surface modification of fillers and matrices, as well as utilizing size effects to improve the dispersion of fillers in matrices; By regulating and designing the interface behavior and interface control between fillers and matrices, as well as the interaction between components, the interface thermal resistance can be reduced; Directional construction of three-dimensional thermal conductivity networks, construction of interpenetrating or two-phase/multi-phase continuous thermal conductivity pathways, learning from nature, biomimetic structures, etc. to construct thermal conductivity pathways.
Next, we will introduce the latest research progress of intrinsic and filled thermal conductive polyimides, summarize the scientific and technological problems currently encountered, and look forward to the opportunities and challenges faced in future development.
4.1 Intrinsic Thermal Conductive Polyimide
Polyimide materials have been widely studied due to their excellent mechanical properties, temperature resistance, solvent resistance, and radiation resistance. However, basic research on the intrinsic polyimide chain structure, molecular chain orientation, and intermolecular forces is still relatively weak. PI also has the characteristics of being insoluble, non melting, and not conducive to processing. Reasonably designing the molecular structure of PI to regulate the chemical properties of materials is one of the technical difficulties in the research of intrinsic thermal conductive polyimides. Scholars have studied the PI molecular chain, structural composition, intermolecular forces, etc. Dong et al. proposed a theoretical model that quantitatively describes the interactions between interchain and intrachain jumps, explains the diameter dependence of polyimide thermal conductivity, and speculates on the upper limit of thermal conductivity of amorphous polymers under quasi one dimensional limit. Morikawa et al. investigated the effects of chemical structure, thickness, viscosity, measurement methods, main chain orientation, and temperature on the thermal diffusion coefficient and mechanism of aromatic polyimides. Next, we will mainly summarize the research status of intrinsic thermal conductive polyimides in recent years from three aspects: chain structure, molecular chain orientation, and intermolecular interactions.
4.1.1 Chain Structure
By regulating the matrix structure, the thermal conductivity of polymer materials can be effectively improved. The chain structure mainly includes primary process structure and secondary remote structure. The main methods for regulating the primary structure include: introducing flexible groups such as ether bonds, ester groups, amide groups, or rigid units of phthalimide into the main chain of the molecule to regulate the chemical composition of the structural units in the PI molecular chain, introducing phenylalkyne groups at the end and maintaining the connection order of the PI chain through chemical crosslinking, and changing the three-dimensional configuration of the PI molecular chain to improve the orderliness of the molecular chain segment aggregate structure. The main means of regulating the secondary structure are: simulating the conformation of PI molecular chains to study under which states PI molecular chains exhibit regular and ordered arrangement and maintain the lowest energy state; By adding terminating agents and selecting different preparation processes to regulate the molecular weight, the curled molecular chains can be stretched, thereby improving the thermal conductivity of the material. From the perspective of monomer molecular structure, the number and polarization degree of polar groups, molecular regularity, internal tightness, and chain arrangement of monomers all affect their thermal conductivity. Xiao et al. prepared three polyimides containing different polar groups to study the effects of polar groups on the thermal conductivity, transparency, solar absorption, UV radiation resistance, and other properties of the materials. The four structural formulas are shown in Figure 8. The results show that the thermal conductivity of PI-4, PI-3, and PI-2 containing - CF3 groups is 0.40, 0.28, and 0.31 W/(m · K), respectively, which is significantly higher than that of PI-1 without polar groups (0.19 W/(m · K)).
The so-called 'structure determines properties', and the PI molecular chain structure plays a decisive role in thermal conductivity. Lei et al. selected three typical PI materials, namely phthalic anhydride (PMDA)/4,4 ′ - diaminodiphenyl ether (ODA), PMDA/(p-phenylenediamine) PDA, and 3,3 ′, 4,4 ′ - biphenyltetracarboxylic acid dianhydride (BPDA)/PDA. They studied the chemical structure, chain rigidity, and aggregation structure of different polyamide acid (PAA) chains, analyzed the contradiction between PI chain structure and mechanical properties, and analyzed the structure-activity relationship from the perspective of chain conformation. From the given molecular stacking model (Figure 9), ordered orientation and stacking structure contribute to the rapid transmission of phonons.
The different internal structures of molecules correspond to different aggregation patterns, which also result in different thermal conductivity values. In addition, some scholars have found that introducing rigid skeletons and π - π conjugated structures into the main chain of molecules can reduce the rotation of the single chain and improve the thermal conductivity of the material. Therefore, the crystalline, amorphous, liquid crystal and other aggregated structures of polyimide, as well as the rigid structure and π - π conjugation in the molecular chain, will have a significant impact on its thermal conductivity.
4.1.2 Molecular Chain Orientation
In the direction of preferential orientation of polymer chains, the thermal conductivity of materials can be significantly improved. The molecular chains are oriented in the direction of orientation, and the amorphous part of the molecular structure is ordered, resulting in an increase in crystallinity. The crystal structure can provide a high-speed channel for phonon transmission, and chain orientation can be considered the most effective factor in improving the intrinsic thermal conductivity of polymers.
At present, research on the influence of molecular chain orientation on the thermal conductivity of crystalline/semi crystalline polymers mainly focuses on polyethylene, which has a relatively simple structure. Here, we would like to mention a polyethylene material with ultra-high thermal conductivity. Shen et al. prepared ultra stretched polyethylene nanofibers with a thermal conductivity of 104 W/(m · K) using a two-stage heating method. The high thermal conductivity is due to the stretching effect in nanofibers, which facilitates the recombination of polyethylene molecular chains, making it an "ideal" single crystal fiber. For polyethylene that crystallizes itself, needle shaped crystals composed of a large number of oriented molecular chains may form during the stretching orientation process, greatly improving its thermal conductivity value. The molecular structure of polyimide is much more complex than that of polyethylene, and there is currently no research indicating that ultra-high thermal conductivity polyimides like polyethylene can be achieved through stretching. In addition to stretching, electrospinning is a mature and widely used technique for orienting polymer molecular chains, which can achieve directional arrangement of molecular chains. During the electrospinning process, the molecular chains are highly stretched, resulting in an increase in crystallinity. In the filling polyimide section of the following text, some scholars have used electrospinning technology to add thermal conductive fillers to prepare a series of high thermal conductivity polyimide films. The reason for choosing electrospinning is that this technology can control the dispersion of fillers in the matrix while also improving the orientation of molecular chains themselves, thereby increasing the inherent thermal conductivity. Designing specific monomers during the aggregation process is another way to enhance the orientation of molecular chains. Generally speaking, the rigid or semi-rigid segments in liquid crystal polymers give the molecular chains an ordered structure, exhibiting micro periodic orientation and generally exhibiting micro anisotropy and macro isotropy. By introducing liquid crystal elements into polymer molecular chains, the orientation of the molecular chains can be achieved, thereby suppressing interface phonon scattering and improving the degree of freedom of phonon heat transfer. In addition, introducing specific liquid crystal small molecules on the substrate can effectively improve the interfacial compatibility and interaction force between the substrate and the filler, which is expected to prepare new high thermal conductivity polymer materials.
Ruan et al. prepared a novel intrinsic high thermal conductivity liquid crystal polyimide (LC-PI) film using phthalimide groups as rigid mesounits and ether bonds as flexible groups. Within the temperature range of liquid crystals, molecular chains can maintain a perfectly ordered structure. The in-plane and out of plane thermal conductivity of the film at room temperature are 2.11 W/(m · K) and 0.3 W/(m · K), respectively, significantly higher than traditional polyimide films. LC-PI also retains excellent mechanical and thermal properties, and has potential applications in highly integrated electronic fields. Due to the fact that the precursor of PI is PAA solution, which has the characteristic of processability, liquid crystal polyimide has certain advantages as a thermal conductive material. Scholars have studied the liquid crystal behavior and properties of polyimide. Shoji et al. investigated the relationship between the liquid crystal morphology and thermal diffusivity of PI films, and prepared cross-linked liquid crystal polyimide films containing siloxane spacer units by introducing acetylene end groups at the chain end. The results indicate that the thermal diffusivity of the film in the thickness direction increases with the degree of crosslinking; The polymer chains are vertically arranged in the thickness direction of the film, which plays an important role in phonon conduction. Yu synthesized a novel semi fatty liquid crystal polyimide containing siloxane with the aim of reducing the transition temperature of liquid crystal polyimide by determining the type and position of substituents. The results indicate that chlorinated and fluorinated substituents have a promoting effect on the formation of liquid crystal phase, but significantly reduce the transition temperature of liquid crystal, while methyl substituents severely disrupt the stability of liquid crystal phase. Ewa et al. synthesized an aliphatic aromatic diamine containing naphthalimide (DA) and a liquid crystal polyimide containing naphthalene and perylene diimide groups, and studied their thermally induced liquid crystal behavior. In the field of liquid crystal thermal conductive polyimide, how to prepare intrinsic liquid crystal thermal conductive PI materials, achieve controllable curing within the liquid crystal range, and how the liquid crystal structure involved affects the thermal conductivity of PI are currently facing another important technical challenge.
In addition, introducing liquid crystal elements into PI molecular chains or directly synthesizing PI with liquid crystal structure by selecting appropriate dianhydride and diamine, the molecular chains generally only maintain an ordered state at the microscopic level, but exhibit an isotropic disordered state at the macroscopic level, thus limiting the improvement of thermal conductivity. In future research on intrinsic thermal conductive polyimides, we can start from improving the micro and macro order of PI chains, and purposefully regulate the chain structure of PI through molecular structure design to prepare new high thermal conductive materials with greater development potential.
4.1.3 intermolecular interactions
The intermolecular interactions mainly include electrostatic forces, van der Waals forces, and hydrogen bonds. These non covalent interactions affect the interactions and spatial configurations between molecular chains, helping to construct long-range ordered structures and promoting the formation of phonon transmission pathways. Covalent atoms in polymer structures can provide channels for phonon transmission. Hydrogen bonding and van der Waals forces mainly enhance the thermal conductivity of materials by promoting crystallization and limiting the twisting of molecular chains. There are imide groups in polyimide, which can easily introduce hydrogen bonds, improve the interaction between the matrix and filler, and also enhance the mechanical properties of polyimide based composites. In addition, hydrogen bonds can also act as soft handles, limiting the twisting motion of molecular chains and improving the rigidity of polymers. Nicholls et al. used thermoplastic polyimide polyurea as the research system and incorporated urea bonds into PI to enhance hydrogen bonding interactions. Kim et al. studied the influence of several different hydrogen bonds on thermal conductivity from the perspective of intermolecular interactions. To achieve high thermal conductivity of polymers, strong intermolecular bonds must replace weak interactions, intermolecular bonds must be tightly connected to the polymer main chain, and the concentration of certain hydrogen bonds must exceed the percolation threshold.
In summary, based on the above analysis, in order to establish hydrogen bonding between polyimide molecular chains, it is necessary to consider factors such as the molecular weight of the initial raw materials, the structure of the main and branch chains, and the radius of rotation, find suitable hydrogen bond donors and acceptors, and further optimize the synthesis and processing of the materials to prepare intrinsic thermal conductive polyimide films.
At present, the introduction of hydrogen bonds in polyimide to improve the thermal conductivity of materials is mainly focused on the research of filled thermal conductive polyimides. The active functional groups on the surface of thermal conductive fillers are prone to form hydrogen bonds with the carbonyl groups of polyimide, which helps with heat transfer and the construction of thermal conductive networks. At the same time, it can also enhance the interaction between fillers and matrices, which will be introduced in the following sections.
The above pioneering works indicate that the thermal conductivity of polyimide is highly correlated with its microstructure, and the improvement of thermal conductivity is limited by multiple factors such as polyimide chemical structure, molecular chain orientation, crystallinity, and intermolecular interactions.
4.2 Filled thermal conductive polyimide
The thermal conductivity of filled composite materials is the result of the combined action of polymer matrix and filler. Thermal conductive filler is one of the important factors affecting the thermal conductivity of composite materials. However, the introduction of filler will form a contact interface with the polymer matrix, and the overlap between fillers will also form a new interface. When heat passes through the interface, interface thermal resistance is generated, which will greatly reduce the thermal conductivity efficiency. So, the microstructure of the filler is the key to constructing a thermal conduction path and reducing the interfacial thermal resistance between the polymer/filler to achieve high thermal conductivity. Therefore, the selection and modification of fillers, as well as the control of interfaces, are key issues that need to be considered to improve the thermal conductivity of composite materials. In addition, the preparation process of composite materials affects the dispersion and distribution of fillers in the polymer matrix, and simple blending of thermally conductive fillers with polymers may also introduce defects, which have a significant impact on the thermal conductivity value of the material. In this section, we mainly discuss the effects of thermal conductive fillers, interface compatibility, and molding processes on the thermal conductivity of polyimide films, and summarize the research progress of filled thermal conductive polyimides in recent years.
4.2.1 Thermal Conductive Filler
The commonly used thermal conductive fillers can be divided into metal, ceramic, and carbon based materials. Metal materials mainly include gold, silver, copper, aluminum and their alloys, etc. Their thermal conductivity is several hundred or even thousands. The thermal conductivity of metals is mainly electronic conductivity, and using metals as fillers can effectively improve the thermal conductivity of materials, but it cannot meet the requirements of insulation. Therefore, it is suitable for thermal conductivity scenarios with low insulation requirements. Carbon based materials mainly include graphene, carbon nanotubes, graphene nanosheets, carbon fibers, etc. Graphene and its derivatives with a planar thermal conductivity higher than 5000 W/(m · K) have extraordinary thermal conductivity and are considered the most promising fillers for preparing polymer based thermal conductive composites and applying them in thermal management. In recent years, thermal conductive polyimides based on carbon based materials have been widely studied.
Wang has prepared a novel flexible printed circuit board (G-FPC), which includes a layer of graphene film with a sandwich structure (PI/M-GF/PI), as shown in Figure 10. The highest thermal conductivity can reach 739.56W/(m · K), and G-FPC also has good heat dissipation ability, which can significantly reduce chip temperature. It has potential applications in future flexible and wearable electronic products. Guo et al. used in-situ polymerization, electrospinning, and hot pressing to prepare a polyimide composite material with all carbon based thermal conductive fillers. Functionalized multi walled carbon nanotubes and graphene oxide (f-MWCNT-g-rGO) were used as thermal conductive fillers, achieving a joint improvement in thermal conductivity and mechanical properties at relatively low filler loads.
The π - π interactions between graphene molecules make graphene prone to aggregation, resulting in poor dispersion in the polymer matrix and limiting the enhancement of material thermal conductivity. Research has shown that introducing metal nanoparticles on the surface of graphene can effectively limit its aggregation. Guo et al. used electrospinning technology combined with in-situ polymerization to prepare high thermal conductivity silver/reduced graphene oxide/polyimide (Ag/rGO/PI) nanocomposites. Among them, silver nanoparticles serve as "spacers" that can maintain a high specific surface area of rGO and improve the cross plane thermal conductivity of the composite material. When the total mass fraction of Ag/rGO filler is 15%, the maximum thermal conductivity value of the composite material reaches 2.12 W/(m · K). Ceramic fillers have good applications in thermal and electrical insulation composite materials due to their inherent thermal conductivity and insulation properties. It mainly includes oxides such as magnesium oxide (MgO), aluminum oxide (Al2O3), and silicon dioxide (SiO2), as well as non oxides such as boron nitride (BN), silicon carbide (SiC), carbon nitride (CN), and aluminum nitride (AlN). There are no freely moving electrons inside these fillers, and they mainly conduct heat through phonons. Among them, BN has been widely studied and applied in the field of thermal conductivity due to its excellent thermal conductivity, good mechanical properties, stable molecular structure, and low coefficient of thermal expansion.
Researchers have found that although a single filler polymer can theoretically form a three-dimensional thermal conductivity network, it is also difficult to achieve high thermal conductivity. The reason is that the defects of the filler itself and the existence of interfaces can cause phonon scattering. Excessive addition of a single filler can also make it difficult for the filler to disperse in the matrix, affecting the processing and forming of the material, and thus affecting the mechanical properties of the composite material. Usually, two or more different shapes or types of thermal conductive fillers are added to the polymer matrix. The mixed filler system can establish a bridge between the fillers, achieve the maximum filling density of the fillers, and construct an effective thermal conductive network. Fillers of different shapes or sizes can fill the gaps between fillers or between fillers and the matrix, helping to form a more complete heat conduction path. Another advantage of mixed fillers is that they help reduce the total filling amount of fillers in the polymer matrix, lower the viscosity of the system, and thus improve the dispersion of fillers in the matrix. The effective construction of thermal conductivity network has a direct impact on the thermal conductivity of filled composite materials. According to the different network structures constructed by mixed fillers in the system, they can be divided into the following four situations: (1) the filling amount of the two fillers is low, and they can be uniformly dispersed in the polymer matrix, but no thermal conductivity network is formed, nor is there any synergistic effect; (2) One type of filler forms a thermal conductive network, while the other maintains a uniformly dispersed state, forming a single thermal conductive network, but there is no synergistic effect; (3) One type of filler continues to maintain uniform dispersion, while the other type of filler overlaps the two together to form a single thermal conductivity network, resulting in a synergistic effect; (4) When the filler is increased to a certain extent, both types of fillers form a thermal conductive network, and the dual networks overlap with each other, exhibiting a synergistic effect. Scenarios (3) and (4) form a single/double thermal conductivity network with synergistic effects, which has important reference significance for studying new methods to improve the thermal conductivity of composite materials.
Mixing and doping two different types of fillers may result in higher thermal conductivity than composite materials prepared by filling either filler alone. Some scholars have doped carbon based materials and ceramic materials to prepare PI films, achieving higher thermal conductivity. He et al. prepared flexible PI composite films with high thermal conductivity and excellent thermoelectric insulation properties using graphene oxide (GO) nanosheets and hexagonal boron nitride (BN) binary fillers. Compared with individual filling, two-component fillers have a stronger improvement in the thermal conductivity of PI composite films, and theoretical models indicate that this can be attributed to the synergistic effect between fillers. Under the separate filling of 1 wt% GO and 20 wt% BN, the thermal conductivity of the film was 6.118 W/(m · K) and 6.391 W/(m · K), respectively. However, under the two fillers of 1 wt% GO and 20 wt% BN, the maximum thermal conductivity was 11.203 W/(m · K), as shown in Figure 11. Wang et al. doped carbon nitride with oxidation-reduction graphene（ rGO@CN ）As a thermal conductive filler, when rGO@CN When the content is 10 wt%, the thermal conductivity can reach 6.08 W/(m · K), achieving high thermal conductivity under low filler loading. After introducing the filler, the PI composite film still maintains excellent mechanical flexibility, thermal stability, and low coefficient of thermal expansion. In practical production and application, a certain amount of non thermal conductive fillers such as glass fiber, boron powder, etc. are also added to the polyimide matrix to improve the mechanical properties and stability of the material. However, different types of fillers are not added arbitrarily, and the influence of geometric properties such as filler dosage, filler shape, and size on the construction of the thermal conductive network needs to be considered. In addition, mixing fillers with the same shape but different particle sizes can optimize the filling density of the fillers. Small sized fillers can be filled into the gaps of large-sized fillers, increasing the possibility of forming a thermal conductivity path. Fillers of the same type but with different shapes and sizes can also exert synergistic effects between fillers.
Combining fillers of different dimensions can also achieve efficient heat conduction pathways. Commonly used zero dimensional (0D) materials include spherical particle fillers such as Al2O3 and Ag microspheres; One dimensional (1D) materials mainly include tubular, rod-shaped, or linear materials such as carbon nanotubes (CNT), carbon fibers, and silver nanowires (AgNWs); Two dimensional (2D) materials include boron nitride nanosheets (BNNS), graphene, and other sheet-like planar materials. A single network with synergistic effects is mainly composed of mixed fillers of 0D+1D, 0D+2D, and 1D+2D. The schematic diagram of the single network is shown in Figure 12. In a single thermal conductive network, one filler is uniformly dispersed in the polymer matrix, while another bridges them together to form a single network with synergistic effects. Liu et al. proposed a two-step synergistic strategy using Al2O3 microspheres and BN as fillers to improve the completeness of the three-dimensional thermal conductivity network in PI membranes. Firstly, BNNS coated Al2O3 microspheres are formed Al2O3@BN Spherical fillers are the first step in synergy, followed by BN and Al2O3@BN Doping forms the second step synergy, and its preparation process is shown in Figure 13. The BN coated on the surface of Al2O3 microspheres plays a role in bridging the thermal conductivity network and improving the interface between Al2O3 and PI. Al2O3@BN Further hybridization preparation with BN Al2O3@BN &BN/PI composite materials, including Al2O3@BN Assist BN in forming a thermal conductive network with isolated structures in the PI matrix. exist Al2O3@BN Under the condition of a 2:1 mass ratio to BN, a total filling amount of 35 wt% was prepared Al2O3@BN &The thermal conductivity of BN/PI composite film is 3.35 W/(m · K), which is 1664% higher than that of pure PI.
The dual thermal conductive network with synergistic effect is mainly constructed by mixed 1D+2D fillers. Similarly, the overlapping interconnection of two-dimensional sheet-like fillers can achieve extremely low interfacial thermal resistance, while one-dimensional fillers play a role in increasing network density and bridging. It is worth noting that the formation of a dual thermal conductivity network usually requires a higher filling ratio, and the synergistic effect is more significant at high concentrations. This dual network can provide a more effective thermal conductivity path. For some dual filler systems, once the dual thermal conductivity network is formed, the thermal conductivity of the composite material will increase sharply, and the change in thermal conductivity can be explained by percolation theory.
Zhang et al. constructed a network structure of PI/BNNS membrane CNT @ α PVA (carbonized polyvinyl alcohol). The CNT @ α PVA network, as a continuous thermal conductive network, can connect the gaps between BNNS and help construct the thermal conductive network. It can also serve as an interpenetrating thermal conductive network for PI/BNNS networks. Due to the formation of a dual thermal conductivity network, when the addition of BNNS is 30 wt% and the addition of CNT is 0.3 wt%, the planar thermal conductivity reaches 8.4 W/(m · K). The schematic diagram of the thermal conductivity network is shown in Figure 14.
It is worth noting that by directly blending two or more fillers together, the spatial matching effect of thermal conductive fillers of different sizes is mainly utilized. The hybridization methods of fillers include physical adsorption and chemical bonding. Physical adsorption usually uses electrostatic interactions or π - π interactions to directly form hybrid thermal conductive fillers with specific structures during chemical reactions. The chemical bonding method involves first attaching reactive groups to different thermal conductive fillers through pretreatment, and then preparing hybrid fillers through chemical reactions between the fillers. These hybrid fillers can be firmly bonded due to chemical reactions, but excessive pretreatment may damage the inherent properties of the fillers and fail to achieve the expected results.
4.2.2 Interface compatibility
As mentioned in the previous section, the introduction of thermal conductive fillers can significantly improve the thermal conductivity of polymer materials, but it also generates a large number of polymer/filler interfaces. The interface between polymer and matrix is one of the main reasons limiting the thermal conductivity improvement of filled thermal conductive composite materials. In the process of heat conduction, the mismatch between the chemical environment at the interface and the phonon vibration mode results in severe phonon scattering, leading to a sharp decrease in the mean free path of phonons. When the heat flow passes through the interface, it is often hindered to a certain extent, causing serious heat loss and resulting in the thermal conductivity of the composite material being much lower than the theoretical value. In addition, due to the different polarities of the polymer matrix and filler, the interface compatibility between the two is also poor during the composite process. Surface modification of fillers is an effective strategy to reduce interfacial thermal resistance and improve thermal conductivity. Common methods to reduce interfacial thermal resistance include chemical modification of the filler surface, building covalent bonds between fillers/polymers or fillers/fillers, coating polymer layers on the surface of thermally conductive fillers, and using nanoparticles to construct "bridges" between fillers/polymers or fillers/fillers. The schematic diagram of these strategies is shown in Figure 15.
Many studies have shown that chemical modification of fillers can activate the surface of inert fillers, introduce functional groups on the filler surface, improve the interfacial compatibility between fillers/polymers, and promote the dispersion of fillers in the matrix. Silane coupling agents or surfactants are usually used to activate the surface of fillers, effectively connecting the phonon vibration spectrum of polymers with thermally conductive fillers. Yang modified boron nitride (m-BN) with 3-glycidoxypropyltrimethoxysilane (γ - MPS) and used it as a thermal conductive filler to prepare PI/m-BN composite materials. When the content of m-BN is 40 wt%, the thermal conductivity of the composite material is 0.748 W/(m · K). Yang Xi prepared hexagonal boron nitride (h-BN)/PI composite films modified with titanium ester coupling agents. The modification effect was evaluated by the change in contact angle before and after h-BN modification. The thermal conductivity of the composite film with a filler content of 40 wt% h-BN after modification was 0.7032 W/(m · K), which was 3.73 times that of the pure PI film. The formation of covalent bonds between polymers and thermal conductive fillers is beneficial for phonon coupling and heat transfer. Although it improves the dispersion of thermal conductive fillers in the polymer matrix, chemical modification may also lead to the formation of defects, damage the perfect structure of the filler surface, and thus reduce the average free path of phonons. The thermal conductivity value of the resulting composite material is not as high as expected. Therefore, the key to chemical modification of fillers is to introduce the required functional groups while preserving the integrity of the thermally conductive fillers.
As mentioned earlier, non covalent interactions such as electrostatic bonding, van der Waals forces, and hydrogen bonding can promote the formation of phonon transmission pathways, effectively improving the thermal conductivity of composite materials. π - π stacking is also a non-contact chemical force close to covalent bond strength, which plays a positive role in improving the thermal conductivity of composite materials. Cao used BNNS as a thermal conductive filler to prepare self-assembled high thermal conductivity PI/BNNS composite materials through van der Waals interactions between PI matrix and filler. BNNS achieved directional arrangement during the deformation of PI microspheres, establishing an effective heat transfer pathway, with a maximum in-plane thermal conductivity of 4.25 W/(m · K). During the ultrasonic assisted liquid-phase exfoliation process, h-BN powder undergoes edge hydroxylation, forming O-H polar groups and adhering to the surface of BNNS. The abundant carbonyl groups in the imide ring form hydrogen bonds with the polar O-H in BNNS, which is beneficial for improving the compatibility between the filler and the polyimide chain. Meanwhile, the polar B-N bonds in BNNS can form strong adsorption through dipole dipole interactions with carbonyl and imide groups in PI. The prepared nano PI composite material also has excellent electrical insulation performance and good thermal stability, and can be used as a thermal conductive material in modern electronic devices. Chen et al. utilized hydrogen bonding and π - π stacking interactions to achieve simple self-assembly of PI matrix and r-GO filler, and prepared core-shell structured PI/r-GO composite materials, constructing highly ordered three-dimensional graphene networks. Graphene oxide has amphiphilicity, with hydrophilic functional groups such as hydroxyl, carbon, and carboxyl groups. It can form hydrogen bonds with the carbon group of PI, enhancing the intermolecular interaction between PI microspheres and graphene oxide. When the content of r-GO is 2 wt%, the thermal conductivity of the composite material is 0.26W/(m · K).
Modifying and coating polymers on the surface of thermal conductive fillers can effectively improve their surface properties. These fillers usually have a core-shell structure and exhibit enhanced physical and chemical properties. Polydopamine (PDA) is a widely used coating polymer material.
Ding et al. modified hexagonal boron nitride with PDA（ h-BN@PDA ）For thermal conductive fillers, research was conducted h-BN@PDA /The thermal conductivity of PI composite materials. The PDA modified h-BN surface has sufficient hydroxyl and amino groups to form hydrogen bonds with the PI matrix, enhancing the affinity between the PI matrix and h-BN, and improving the interfacial adhesion. The filler and matrix also form an aligned structure, constructing a two-dimensional oriented structure. At a filler content of 20 vol% (PDA modification time for h-BN is 6 h), the highest planar thermal conductivity of PI composite material can reach 3.01 W/(m · K), and the presence of PDA coating also improves the thermal stability of the composite material. In addition to coating polymers on the surface of inorganic fillers, inorganic materials can also be coated on the surface of inorganic fillers to form core-shell structured fillers.
Wang Yuting's core-shell structure formed by coating AgNWs on SiO2 surface AgNWs@SiO2 A thermal conductive and insulating polyimide film was prepared using electrospinning technology as a thermal conductive filler. When the filler content is 25 wt%, the thermal conductivity of the film reaches 2.80 W/(m · K).
The random dispersion of thermal conductive fillers in the polymer matrix makes it difficult to form a continuous thermal conductive path and increases the contact area between the fillers and the matrix, thereby increasing the interfacial thermal resistance.
Building a "thermal bridge" between fillers/fillers or fillers/substrates is another effective strategy to reduce interfacial thermal resistance and increase heat transfer area. Duan has prepared a novel high thermal conductivity polyimide dielectric composite material, in which the filler particles (F-BA) are spherical alumina coated with nano boron nitride (n-BN) and polydopamine（ PDA@Al2O3 ）In order to improve the interface compatibility between the filler and the matrix, 1,6-diisocyanate (HDI) was innovatively used as a "bridging agent" to connect n-BN and PDA@Al2O3 Form filler particles with a core-shell structure. The thermal conductivity of the PI composite film containing 25 wt% F-BA filler in the planar and cross planar directions is 6.41 W/(m · K) and 1.01 W/(m · K), respectively, which is 36 times and 6 times higher than that of pure polyimide at 0.18 W/(m · K). The specific thermal conductivity data is shown in Figure 16.
Dong et al. added AgNWs as a "thermal bridge" to vertically arranged anisotropic BNNS/PI materials to form a interconnected thermal network. When the filling amount of BNNS AgNWs was 20 wt%, the in-plane thermal conductivity of the composite material reached 4.75 W/(m · K). Compared with the material without AgNWs, the thermal conductivity of PI composite film with AgNWs added shows a more prominent trend of increasing with the increase of filler content.
4.2.3 Molding process
The preparation process and molding method of composite materials are closely related to the thermal conductivity of the materials. The molding process affects the dispersion of thermal conductive fillers in the matrix and may also introduce defects in the system. Choosing an appropriate molding method can help construct a thermal conductive network and prepare materials with higher thermal conductivity. The main preparation method of composite materials is blending, but simple melt blending, mechanical blending, and solution blending of the matrix and filler may cause larger interfaces, and the improvement of thermal conductivity may not achieve the desired effect. Therefore, the molding process of composite materials is of great significance for improving thermal conductivity. Researchers have improved the dispersion of thermal conductive fillers in the matrix and constructed an effective thermal conductivity network to enhance the thermal conductivity of materials. Electrospinning technology was also mentioned earlier, which can improve the agglomeration of fillers in the matrix and achieve effective alignment of inorganic fillers in the polymer matrix. This technology can also disperse nanoparticles in polymer matrices to prepare fiber materials with large aspect ratios, high specific surface areas, and controllable orientations. Many scholars have conducted extensive research on thermally conductive polyimides using electrospinning technology. The PI film with a dual network structure, PAA/BNNS fibers, and PVA/CNT fibers mentioned in the thermal conductive filler section were prepared by co electrospinning technology. These two precursor fibers were uniformly interwoven in the film, and the precursor film was subjected to high-temperature treatment followed by PVA carbonization and PAA thermal imidization to prepare the PI composite film.
Guo et al. prepared multilayer polyimide composite films with high thermal conductivity. Among them, graphene oxide/expanded graphite (GO/EG) is used as the top thermal conductivity and electromagnetic interference shielding layer, Fe3O4/PI is used as the middle shielding reinforcement layer, and electrospun PI fibers are used as the base layer to improve the thermal conductivity and mechanical properties of the PI composite film. The in-plane thermal conductivity of the PI composite material containing 61.0% GO/EG and 23.8% Fe3O4/PI is 95.40 W/(m · K), and the electromagnetic interference shielding coefficient reaches 34.0 dB. This composite film has broad application prospects in the fields of optical fields and miniaturized electronic devices.
In addition, some progress has been made in the preparation of thermally conductive polyimide films using ice template method, freeze-drying technology, and magnetic field assisted orientation technology. Wei prepared rGO/PI composite films with a three-dimensional rGO structure using freeze-drying technology. When the amount of rGO filler added is 8 wt%, the thermal conductivity of the PI film can reach 2.78 W/(m · K), while the thermal conductivity of the thin film prepared by conventional methods containing 8 wt% rGO is only about 0.5 W/(m · K). Liu prepared GF-BN/PI (MF) composite films using PDA modified oxidation-reduction graphene and magnetic needle shaped iron oxide hybrid nanoparticles as fillers (GF). The ordered orientation of filler particles in the composite film was achieved through a mobile magnetic field induction strategy. The mechanism of magnetic field induction and thermal conduction in the composite film is shown in Figure 17. When the filler content is 30 wt%, the in-plane and out of plane thermal conductivity of the composite film are 2.532 and 0.425 W/(m · K), respectively. After the above analysis and discussion, the molecular chain structure of polyimide, the selection and modification of fillers, interface compatibility, and molding processes all have a certain impact on the thermal conductivity of the material. Each factor has a different effect on the thermal conductivity, but from the perspective of thermal conductivity mechanism, there is a certain correlation.
The carbonyl group in the PI molecular structure can form hydrogen bonds with the polar hydroxyl group on the hydroxylated filler, and the presence of hydrogen bonds can improve the compatibility between inorganic fillers and PI. Modifying the surface of fillers is one of the effective strategies to reduce interfacial thermal resistance and improve interfacial compatibility. The ordered arrangement of molecular chains along the orientation direction can provide a high-speed transmission channel for phonons, while processes such as electrospinning, high-temperature hot pressing, and freeze-drying can orient the molecular chains to have a more ordered molecular structure. The construction of a thermal conductive network requires the orderly dispersion and arrangement of fillers in the matrix, and the selection and content of fillers directly affect the formation of the thermal conductive network. The preparation process will have a certain impact on the dispersion of thermal conductive fillers in the matrix, and will also strengthen or weaken the formation of thermal conductive networks.
Therefore, when discussing the influence of a certain factor on the thermal conductivity of composite materials, it is not limited to a single influencing factor, and the correlation between various factors should be fully considered in order to obtain more authoritative theoretical analysis results.
05 Conclusion
In recent years, with the increasing application of thermal conductive dielectric polymer materials, their research has also become more in-depth. Based on the current research status and development trends of thermal conductive polyimide films, it is expected that more research work will be carried out in the following areas:
(1) It is necessary to conduct in-depth research on the effects of the ordering, aggregation morphology, molecular chain orientation, and intermolecular forces of polyimide structure on the thermal conductivity of PI, identify the intrinsic relationships, and deeply elucidate the thermal conductivity mechanism. Further exploration is needed to investigate liquid crystal thermal conductive polyimides from aspects such as chain structure, selection of dianhydride and diamine monomers, introduction of functional groups and capping groups.
(2) It is necessary to fully consider factors such as the structure, degree of crystallization, defects, impurities, etc. of thermal conductive fillers, and search for and develop new high-quality thermal conductive fillers to improve the dispersion of inorganic filler particles in polyimide matrix, reduce the interfacial thermal resistance between fillers/fillers and fillers/matrices; Further research on how to construct a thermal conductivity network that fully utilizes the synergistic effect between fillers, so that PI composite materials have high thermal conductivity at low filler content.
(3) Introducing inorganic filler particles into polyimide matrix will increase the number of interfaces, seriously affecting the thermal conductivity and mechanical properties of PI. Therefore, it is important to focus on and study all organic thermal conductive composite materials to avoid interface compatibility issues caused by fillers.
(4) We should fully utilize computer software combined with mathematical models to dynamically simulate the thermal conductivity or other properties of composite materials, and gain a deeper understanding of the factors that affect the thermal conductivity of composite materials. This will provide theoretical support for studying the thermal conductivity mechanism of PI and developing new high thermal conductivity PI materials.
(5) Considering the complexity of future electronic product application environments, whether in the field of electronics/microelectronics under low voltage conditions, or in the field of electronics under high voltage and ultra-high voltage, developing PI products with high thermal conductivity, low dielectric constant, low coefficient of expansion, flame retardancy, and electromagnetic shielding has become an important direction for future research.
