Composite vapor chamber and manufacturing process thereof

Abstract

The present application relates to the technical field of heat plate, especially to a composite heat plate and its manufacturing process, which comprises stamping step, laser processing step, assembling and welding step and liquid injection packaging step. The present application synchronously forms the shell and the support structure through multi-station continuous die room temperature stamping, synchronously forms the shell, micron-level fractal trunk groove and the support structure through multi-station continuous die room temperature stamping, synchronously forms the shell, micron-level fractal trunk groove and the support structure through multi-station continuous die room temperature stamping, forms the gradient driving function layer through laser surface etching and modification in one step, and realizes the synchronous reliable connection of high-quality sealing and internal support in the smallest heat affected zone through laser-ultrasonic composite welding. The present application ensures the integrity of the internal fine structure and the high reliability of the product, and the process flow is compact and efficient, which is especially suitable for the manufacturing demand of high-performance heat dissipation components in the fields of new energy vehicles, high-end communication equipment and the like.

Description

Technical Field

[0001] This invention relates to the field of heat exchanger technology, and in particular to a composite heat exchanger and its manufacturing process. Background Technology

[0002] With the continuous increase in the power density of electronic devices and the rapid development of new energy vehicles, high-speed communications, and high-end equipment manufacturing, the demand for heat sinks, as efficient two-phase heat transfer devices, has rapidly expanded from the traditional consumer electronics field to industrial and automotive fields with higher requirements for heat dissipation performance, reliability, and production efficiency, such as automotive electric drive systems, power battery thermal management, 5G base stations, servers, and precision machining molds.

[0003] Currently, the manufacture of such high-performance composite capillary heat spreaders typically employs a multi-step discrete process. For example, microgrooves or sintered capillary cores are first fabricated on the inner surface of the shell using methods such as chemical etching, machining, or high-temperature sintering. Subsequently, the internal support structure needs to be fabricated and installed separately. Finally, the shell is sealed using processes such as brazing or traditional laser welding. This manufacturing approach has significant drawbacks: the fabrication and integration processes of each functional structure (macroscopic channels, micro / nano capillary layers, and support) are separate, resulting in lengthy processes, slow production cycles, and high costs. More importantly, during subsequent high-temperature welding or brazing encapsulation, the pre-fabricated precision micro / nano capillary structures are highly susceptible to deformation, blockage, or performance degradation due to heat input. Simultaneously, the connection quality between the support structure and the shell is difficult to guarantee. This severely restricts the application of heat spreaders in industrial applications requiring high reliability and mass production (such as the automotive industry).

[0004] Therefore, how to provide a heat spreader manufacturing process that can efficiently and precisely form complex internal structures, ensure that these delicate functional structures are not damaged by heat during the final welding and packaging process, and achieve highly reliable sealing and integrated internal structure has become the key to meeting the urgent needs of high-performance heat dissipation components in fields such as new energy vehicles, high-end communications and precision manufacturing. It is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by providing a composite heat sink and its manufacturing process. The process involves simultaneously forming the shell and support structure through room temperature stamping using a multi-station continuous die; simultaneously forming the shell, micron-level fractal main grooves, and support structure through room temperature stamping using the same multi-station continuous die; forming a gradient-driven functional layer in one step through laser surface etching and modification; and achieving high-quality sealing with a minimal heat-affected zone and synchronous and reliable connection of the internal support through laser-ultrasonic composite welding. This invention ensures the integrity of the internal fine structure and high product reliability. The process is compact and efficient, making it particularly suitable for manufacturing high-performance heat dissipation components in fields such as new energy vehicles and high-end communication equipment. It solves the problems of existing high-performance heat sink manufacturing processes, such as susceptibility to heat damage to precision micro / nano capillary structures, low production efficiency, and poor reliability of support structure connections due to multi-step discrete processing and traditional high-temperature packaging.

[0006] To achieve the above objectives, the present invention includes the following steps:

[0007] The stamping step utilizes a multi-station continuous die to stamp the layered composite sheet at room temperature to form an upper shell and a lower shell respectively; wherein, at the first station, micron-level fractal main grooves are stamped on the layered composite sheet; at the second station, a support structure is formed in the preset evaporation zone of the lower shell; The laser processing step involves sending the stamped upper and lower shells into a laser processing module, where a laser is used to etch micro-nano capillary structures onto the surface of the micron-level fractal main grooves and the supporting structure, and to modify the surface energy, thereby forming a gradient-driven surface layer extending from the condensation zone to the evaporation zone. The assembly and welding process involves aligning and closing the upper and lower housings, bringing the end of the support structure into contact with the inner wall of the upper housing, and then performing peripheral sealing welding using a laser-ultrasonic composite process. The laser-ultrasonic composite process includes: applying high-frequency ultrasonic vibration to clean and initially connect the welding interface, while using fiber laser to perform circumferential seam scanning welding, and using welding energy to simultaneously connect the end of the support structure to the inner wall of the upper shell. The liquid injection and sealing step involves evacuating the welded shell through a pre-reserved liquid injection port and injecting the working fluid, followed by sealing the liquid injection port.

[0008] On the other hand, a composite heat exchange plate is also provided, manufactured by the above-described process, comprising: The shell is formed by sealing and fitting an upper shell and a lower shell together. The shell is formed by stamping a layered composite plate, and the inner wall of the shell is provided with micron-level fractal main grooves. A support structure is provided in the evaporation zone of the lower shell; A sealing weld is formed at the periphery of the upper and lower shells, and the sealing weld is formed by laser-ultrasonic composite welding; A gradient-driven surface layer covering the surface of the micron-scale fractal main trench and the supporting structure has a surface energy gradient from the condensation zone to the evaporation zone.

[0009] The beneficial effects of this invention are as follows: The shell, micron-level fractal main grooves, and support structure are simultaneously formed by room temperature stamping using a multi-station continuous die; a gradient-driven functional layer is formed in one step through laser surface etching and modification; and high-quality sealing with a minimal heat-affected zone and synchronous and reliable connection of the internal support are achieved through laser-ultrasonic composite welding. This invention ensures the integrity of the internal fine structure and the high reliability of the product. The process is compact and efficient, making it particularly suitable for manufacturing high-performance heat dissipation components in fields such as new energy vehicles and high-end communication equipment. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the manufacturing process of the present invention.

[0011] The reference numerals in the figures include: Detailed Implementation The present invention will now be described in detail with reference to the accompanying drawings.

[0012] like Figure 1 As shown, the manufacturing process of a composite heat exchange plate according to the present invention includes the following steps: Stamping steps: Using a multi-station continuous die, the layered composite sheet is stamped at room temperature to form the upper shell and the lower shell respectively; wherein, at the first station, micron-level fractal main grooves are stamped on the layered composite sheet; at the second station, a support structure is formed in the preset evaporation zone of the lower shell; Multi-station progressive dies achieve precision plastic forming of the upper and lower shells in a single step at room temperature through multi-step plastic deformation. This efficient, synchronous, high-precision, and low-energy forming of the upper and lower shells avoids the multiple positioning errors, inefficiencies, and high energy consumption associated with traditional single-process, multi-step machining. This significantly improves the production cycle time and consistency for large-scale, low-cost manufacturing in the automotive parts and battery components industries.

[0013] In the initial stage of stamping, a micron-level main return channel with a biomimetic fractal structure is directly formed on the sheet metal using the geometry of the die. The micron-level fractal trunk groove optimizes fluid resistance and distribution based on fractal geometry. An efficient macroscopic return path for the working fluid is pre-constructed, and the fractal design optimizes the flow efficiency, laying the foundation for the subsequent construction of micro-nano composite capillary structures. Compared with subsequent etching or sintering processes, this method offers higher geometric accuracy and lower processing costs.

[0014] In subsequent workstations of the same continuous mold, through plastic processing methods such as stamping, extrusion, or embedding, an internal support structure is directly formed in a specific area of ​​the lower shell to resist internal and external pressure differences and prevent shell collapse, while the shell is being formed. This achieves synchronous or integrated forming of the support structure and the shell, eliminating the subsequent separate installation process of the support structure, simplifying the process, avoiding positioning errors and contamination risks caused by secondary assembly, improving product structural reliability, and realizing the process integration and automation sought after in the mold manufacturing field.

[0015] The laser processing step involves sending the stamped upper and lower shells into the laser processing module, where the laser is used to etch micro-nano capillary structures on the surface of the micron-level fractal main grooves and support structures and modify the surface energy, thereby forming a gradient-driven surface layer extending from the condensation zone to the evaporation zone. By utilizing the ablation, melting, and surface modification effects of high-energy-density laser beams, submicron to nanoscale rough structures are directly fabricated on the surface of pre-formed macroscopic structures, while simultaneously altering the surface chemical state (such as oxidation level and wettability). This allows for the in-situ construction of micro / nano composite structures that enhance capillary forces on the macroscopic structure, and by modifying surface energy, they are endowed with directional driving capabilities. This significantly improves the capillary pumping capacity and anti-gravity performance of the heat exchanger, solving the problem of insufficient performance of micro / nano structures in traditional processes and directly enhancing heat transfer efficiency.

[0016] By programmatically controlling laser processing parameters (such as energy, speed, and scanning strategy), surface structures with microscopic morphology and / or chemical composition gradients are created in the condensation and evaporation zones, thereby forming an actively driven layer composed of capillary force gradients. This actively guides the working fluid to flow directionally and efficiently from the condensation zone to the evaporation zone, optimizing the working fluid circulation path and efficiency, making it a core functional layer for improving the heat transfer performance of the vapor chamber.

[0017] Assembly and welding steps: Align and close the upper and lower shells, so that the ends of the supporting structure contact the inner wall of the upper shell, and perform peripheral sealing welding using a laser-ultrasonic composite process; First, precise alignment ensures the connection point of the support structure. Then, ultrasonic vibration is used to break up the oxide film at the interface, clean and activate the surface, while a laser beam provides localized high heat flux density input to achieve metallurgical bonding. This one-step method simultaneously achieves a highly airtight perimeter seal of the shell and a reliable internal connection of the support structure. Its core contribution lies in achieving a low-temperature, high-efficiency, high-quality connection through the synergistic effect of laser and ultrasound. This effectively avoids irreversible thermal damage to the formed precision micro / nano capillary structure caused by the high heat input of traditional single thermal welding (such as brazing), thus ensuring the ultimate performance and long-term reliability of the product and precisely addressing the key defect of "encapsulation thermal damage" in the background technology.

[0018] The laser-ultrasonic composite process includes: applying high-frequency ultrasonic vibration to clean and initially connect the welding interface, while using fiber laser to perform circumferential seam scanning welding, and using welding energy to simultaneously connect the end of the support structure to the inner wall of the upper shell. Ultrasonic vibration removes interfacial contaminants and achieves preliminary solid-state bonding (micro-forging welding) through high-frequency mechanical friction; fiber laser provides precise and controllable localized heat input to achieve final fusion welding. Ultrasonic pretreatment improves the cleanliness and reliability of the weld interface; precise laser heat input results in high-quality, low-deformation welds, ensuring the integrity and strength of the seal.

[0019] The heat from laser welding is conducted through the upper shell to the contact area between the end of the support structure and the inner wall of the upper shell, causing localized melting and metallurgical bonding at this contact area. Without requiring additional processes or heat sources, the residual heat from welding cleverly achieves "in-situ" fixation of the internal support structure while simultaneously sealing the shell, further simplifying the process and strengthening the overall structure.

[0020] Liquid injection and encapsulation steps: Vacuum the welded shell through the reserved liquid injection port and fill it with working fluid, then seal the liquid injection port.

[0021] The internal atmosphere is replaced and the working fluid is filled through a single channel in a completely sealed cavity, and the channel is then sealed to achieve a complete seal. Completing the functional encapsulation of the heat spreader, ensuring the internal vacuum level and working fluid filling volume, is the final guarantee step for realizing the phase change heat transfer function.

[0022] During processing, the layered composite sheet enters a multi-station continuous mold. At the first station, micron-level fractal main grooves are stamped out. At the second station, a support structure is formed in the pre-set evaporation zone of the lower shell, thus efficiently forming the upper and lower shells at room temperature. The formed upper and lower shells are then fed into a laser processing module. The laser etches micro-nano capillary structures on the surface of the micron-level fractal main grooves and the support structure, simultaneously modifying the surface energy to form a gradient-driven surface layer. The upper and lower shells are aligned and joined, with the end of the support structure contacting the inner wall of the upper shell. A laser-ultrasonic composite process is used for peripheral sealing welding. This process simultaneously applies high-frequency ultrasonic vibration to clean and initially connect the welding interface. Fiber laser scanning welding is used for circumferential seam welding, and the welding energy simultaneously connects the end of the support structure to the inner wall of the upper shell. Finally, the welded shell is evacuated through a pre-reserved injection port and filled with working fluid, then the injection port is sealed, completing the fabrication of the composite heat spreader.

[0023] Preferably, the laser processing step further includes: using a laser to perform laser shot peening or laser shock peening on a predetermined non-channel area on the inner wall of the upper and lower shells.

[0024] High-power-density pulsed laser irradiation of the constraint layer covering the inner wall surface of the upper and lower shells induces high-amplitude shock waves. These shock waves penetrate the material interior, inducing plastic deformation at the surface and introducing a beneficial residual compressive stress field, while simultaneously refining the surface grains. This significantly enhances the fatigue resistance of the upper and lower shells under alternating loads such as long-term vibration and thermal cycling in new energy vehicles.

[0025] The laser processing step includes an online morphology detection step: measuring the three-dimensional morphology parameters of the micron-scale fractal trunk grooves; the laser processing step adaptively adjusts the laser processing parameters used for etching the micro-nano capillary structure based on the three-dimensional morphology parameters.

[0026] By employing line laser scanning or structured light 3D measurement technology, key geometric parameters such as depth, width, and sidewall angles of the micron-scale fractal main grooves in the stamped product are acquired non-contactly, forming a digital morphology map. This enables process self-adaptation and closed-loop control, providing real-time and accurate input parameters for subsequent laser processing steps. Laser processing parameters (such as energy, focal length, and scanning path) can be dynamically adjusted based on the measured morphology, compensating for minor dimensional fluctuations that may occur during the stamping process. Ensuring performance consistency and high yield, online full inspection of key macroscopic features allows for the timely detection and rejection of defective products, ensuring high consistency of the shells entering the functional processing stage. This is crucial for the large-scale, high-quality production of automotive parts and battery components, and is a core component of intelligent manufacturing.

[0027] The laser processing step is followed by an online wettability detection step: measuring the wettability distribution of the gradient-driven surface layer; and adaptively compensating the laser processing parameters of subsequent products based on the measurement results.

[0028] By employing image-based contact angle measurement instruments or high-throughput surface energy rapid measurement technology, the wettability, such as the water droplet angle, of gradient-driven surface layers in different regions, especially the evaporation and condensation zones, can be measured online and non-destructively, quantifying their surface energy gradient.

[0029] This system enables online monitoring and feedback optimization of functional performance, directly monitoring the performance of the gradient-driven surface layer—the core functional layer that determines the heat transfer efficiency of the vapor chamber—in real time. The monitoring results are then fed back to the laser processing system, forming an intelligent closed loop of "processing-monitoring-compensation," achieving proactive control over the product's ultimate performance.

[0030] To ensure consistent and high-quality product performance, direct monitoring of the chemical properties of the functional layers ensures that the capillary drive force and working fluid circulation efficiency of each heat sink meet design requirements, greatly improving product consistency and reliability and meeting the stringent "high-performance" standards for heat dissipation components in high-end communication equipment, servers, and other applications.

[0031] Preferably, the assembly and welding process integrates an online monitoring and adaptive control system for weld quality. This system monitors the welding temperature field in real time using an infrared thermal imager and dynamically adjusts the laser welding parameters to keep the welding heat input within a preset ideal range.

[0032] This method utilizes an infrared thermal imager to capture the two-dimensional temperature distribution of the welding area in real-time, non-contact, and full-field manner, forming a dynamic temperature field image to accurately identify the temperature gradient and peak value of the weld and the surrounding heat-affected zone. The real-time monitored temperature field data (such as the highest temperature and the width of the heat-affected zone) is compared with a preset "ideal temperature range" model. Based on the comparison results, core parameters such as laser welding power, scanning speed, or defocusing amount are dynamically adjusted through a feedback control algorithm (such as a PID algorithm), forming a closed-loop control circuit of "sensing-judgment-execution." This achieves real-time, closed-loop, and precise control of the welding heat input, strictly limiting the welding heat-affected zone within a preset safe range. It eliminates the risk of overheating, deformation, or performance degradation of micro / nano capillary structures caused by heat input fluctuations or concentrations during laser welding, providing a definitive technical solution that directly and thoroughly addresses the key defect in the prior art of "precise internal micro / nano structures being susceptible to heat damage." Through dynamic control, the impact of slight unevenness of the sheet metal, minor differences in surface condition, or fluctuations in equipment power on welding quality is eliminated, ensuring that the sealing welds of each product have uniform penetration depth, dense structure, and high airtightness. This is crucial for heat dissipation components used in high-pressure, high-reliability environments such as electric drive systems and battery packs in new energy vehicles, significantly improving the service life of products under long-term vibration and thermal cycling.

[0033] The layered composite material in this embodiment is a three-layer composite structure of copper-intermediate layer-copper, wherein two copper layers are bonded to both sides of the intermediate layer through a solid-state composite process; the intermediate layer is a metal layer.

[0034] The design employs a layered metal composite material, utilizing solid-state composite processes (such as rolling composite) to achieve a metallurgical bond between the copper layer and the intermediate layer. This combines the high thermal conductivity of copper with the high strength and low thermal expansion characteristics of the intermediate layer. The material design optimizes the overall performance of the shell, ensuring excellent heat conduction paths while leveraging the intermediate layer to enhance the shell's structural strength and rigidity and adjust the overall coefficient of thermal expansion. This makes the product more adaptable to the thermal cycling and mechanical stresses encountered in scenarios such as automotive electric drives and battery packs.

[0035] At least one of the two copper layers is a functionally graded structure (FJD). The composition or structure of the FJD changes continuously or stepwise from its working surface to the interface with the intermediate layer. The thermal conductivity of the working surface is higher than that of the area near the interface, while the strength and hardness of the area near the interface are higher than those of the working surface. This achieves a performance gradient transition within the copper layer from the working surface (inner wall, requiring high thermal conductivity) to the interface (requiring high bonding strength). This further optimizes material performance matching; the high thermal conductivity of the working surface facilitates heat dissipation, and the high strength of the interface ensures strong interlayer bonding, preventing delamination failure during stamping or thermal cycling, and improving process stability and product lifespan.

[0036] In this embodiment, the metal layer is a stainless steel layer; the stainless steel layer has a perforated structure with a regular array of micropores. The use of stainless steel provides structural strength and corrosion resistance, while the perforated structure achieves lightweighting while maintaining mechanical properties and may form auxiliary capillary channels. This achieves a lightweight casing, suitable for weight-sensitive applications in new energy vehicles. Simultaneously, the perforated structure may help enhance the distribution of the working fluid or provide additional mechanical support points.

[0037] The metal layer in this embodiment is a composite functional intermediate layer; The composite functional interlayer includes a molybdenum-based functional layer, which is adjacent to a stainless steel layer and forms a metallurgical bonding interface between them. The coefficient of thermal expansion of the molybdenum-based functional layer is lower than that of the stainless steel layer. A low coefficient of thermal expansion (CTE) molybdenum base layer is introduced as a functional interlayer and bonded to the high coefficient of thermal expansion stainless steel layer to form a gradient transition structure of thermal expansion coefficient. This significantly improves the thermal matching of the heat sink when soldering high-power chips such as IGBTs and CPUs, reduces thermal stress caused by thermal expansion coefficient mismatch, improves the long-term reliability of the solder joint, and solves the heat dissipation problem of high-power devices.

[0038] The molybdenum-based functional layer contains micropores, or it can be a copper-based composite material layer reinforced with molybdenum fibers. By introducing porous or fiber-reinforcing phases, the density, thermal conductivity, and coefficient of thermal expansion of the molybdenum-based functional layer can be further controlled. This provides a more flexible space for material performance design; the micropores can assist in liquid storage or adjust the coefficient of thermal expansion, while molybdenum fiber reinforcement can improve mechanical properties while maintaining a low coefficient of thermal expansion, providing a material basis for meeting the needs of specific extreme applications (such as aerospace and high-end servers).

[0039] In this embodiment, at the second station of the stamping step, pre-reserved micropores are formed in the preset evaporation zone of the lower shell through stamping, and ceramic support columns are cold-extruded and embedded into the pre-reserved micropores to form a support structure. High-hardness, high-insulation, and low-thermal-conductivity ceramic is used as the support column, and the plastic deformation generated by cold extrusion achieves an interference fit and mechanical interlock with the metal shell. The support structure possesses excellent compressive strength and high electrical insulation, avoiding the electrochemical corrosion or short-circuit risks that may be caused by metal supports, making it particularly suitable for power battery module heat dissipation scenarios with stringent electrical safety requirements. Specifically, the ends of the ceramic support columns are metallized to facilitate welding of the ceramic support columns to the upper and lower shells.

[0040] An inwardly protruding flow-guiding structure is integrally stamped onto the inner wall of the upper shell. After the assembly and welding steps are completed, the flow-guiding structure is located on the side of the ceramic support pillar to distribute the working fluid to the evaporation zone around the ceramic support pillar. The flow-guiding structure is integrally formed as part of the upper shell, forming a specific liquid flow guiding channel after closure. This optimizes the liquid distribution in the evaporation zone, ensuring sufficient working fluid supply to areas with concentrated heat (such as below the chip hotspot), and improving the local heat dissipation capacity and overall temperature uniformity of the heat spreader.

[0041] In this embodiment, at the second station of the stamping step, a localized area of ​​the layered composite sheet is stamped to form a metal micro-lattice structure integrally formed with the lower shell, serving as a support structure. Precision stamping induces localized three-dimensional plastic deformation in the material, forming an integrated micro-truss or grid lattice structure. This achieves complete integration and interface-free manufacturing of the support structure and shell, resulting in the highest structural stiffness and stability, and extremely high manufacturing efficiency. The micro-lattice structure itself possesses a high specific surface area and porosity, which can further enhance capillary action and evaporation area.

[0042] During the stamping process, an inwardly protruding flow-guiding structure is integrally stamped onto the inner wall of the upper shell. After the assembly and welding steps are completed, the flow-guiding structure is at least partially interwoven into the metal micro-lattice structure to distribute the working fluid to the evaporation zone inside the metal micro-lattice structure. The flow-guiding structure and the micro-lattice structure form a composite flow channel network that is interlocked and interpenetrating in three-dimensional space. This greatly enhances the penetration and uniformity of the working fluid within the complex porous structure of the evaporation zone. The flow-guiding structure directly guides the condensate into the depths of the micro-lattice, fully utilizing the high specific surface area of ​​the micro-lattice to achieve ultra-efficient evaporative heat dissipation, making it particularly suitable for high-power-density chips.

[0043] In practical use, the flow guiding structure is selected from at least one of the following: radial ribs extending from the inner wall of the upper shell to the center of the evaporation zone, spiral grooves, and multi-level fractal groove network structure. The radial rib structure consists of straight or slightly curved ribs radiating from the condensation zone (periphery) to the center of the evaporation zone, providing the shortest and most direct macroscopic flow path for the working fluid from the edge to the center.

[0044] The spiral groove structure is a continuous spiral line surrounding the evaporation zone, which forces the working fluid to move towards the center along the spiral path, effectively extending the flow path and the contact time with the shell wall, and enhancing the uniformity of condensate distribution and preheating effect.

[0045] The multi-level fractal groove network mimics the veins of a leaf or the blood circulation system, constructing a tree-like or mesh-like flow channel system formed by the hierarchical branches of the main grooves. Based on fractal geometry, it optimizes the transport efficiency and space filling rate, realizing the three-dimensional, efficient, and uniform penetration of the working fluid in the evaporation zone.

[0046] These directional flow channel structures, especially the multi-level fractal network, systematically solve the problem of insufficient working fluid supply below the "chip hotspot" caused by uneven liquid distribution in the evaporation zone of traditional vapor chambers. By constructing a biomimetic high-efficiency transport network, it ensures that the condensed and refluxed working fluid is actively, quickly, and evenly transported to the entire evaporation zone, especially the core area with high heat flux density, thereby completely preventing localized drying and greatly improving the temperature uniformity and heat dissipation reliability of the vapor chamber. Different structural forms can be flexibly matched with heat sources of different shapes and layouts (such as multiple CPUs or long strip IGBT modules), reflecting the demand for customized and highly adaptable solutions in the high-end equipment manufacturing field.

[0047] The flow-guiding structure possesses a surface energy gradient, with its surface energy increasing from the condensation zone towards the center of the evaporation zone. Through laser modification and coating deposition techniques, continuous changes in chemical composition or microstructure are created on the surface of the flow-guiding structure (especially the inner wall of the flow channel), causing its surface energy (hydrophilicity) to gradually increase from the initial end (near the condensation zone) to the terminal end (the core of the evaporation zone). This creates a capillary force gradient or Marangoni effect (surface tension gradient) from low surface energy regions to high surface energy regions. The surface energy gradient, in conjunction with the gradient-driven surface layer from laser treatment, forms a powerful directional driving force. This not only more efficiently guides the basic working fluid to high heat flux regions but also synergizes with magnetic nanoparticle composite working fluids: high surface energy regions exhibit stronger adsorption or guiding effects on magnetic nanoparticles. Combined with an external controllable magnetic field, precise heat management and dynamic enhancement can be achieved. This directly solves the problem of passive heat dissipation capabilities and inability to cope with transient high heat flux impacts in existing technologies.

[0048] Creating a controllable surface energy gradient on a metal surface represents a fusion of advanced inorganic non-metallic materials (surface modification technology) and intelligent manufacturing (precision processing and control). This upgrades the vapor chamber from a passive heat dissipation component into an intelligent thermal management unit with sensing and responsive capabilities, aligning with the core technological requirements of intelligent thermal management systems in new energy vehicles and data centers.

[0049] In the laser treatment step of this embodiment, by adjusting the laser processing parameters applied to different regions of the vapor chamber shell, the etching of the micro-nano capillary structure on the shell surface and the surface chemical modification are synchronously completed in a single-pass processing process, so as to form a gradient-driven surface layer extending from the condensation region to the evaporation region; Wherein, the laser processing parameters at least include pulse energy E and scanning speed V. By controlling the first laser fluence F1 applied to the evaporation region to be greater than the second laser fluence F2 applied to the condensation region, the average characteristic size d1 of the micro-nano capillary structure formed in the evaporation region is smaller than the average characteristic size d2 of the condensation region, and the surface oxidation degree O1 of the evaporation region is higher than the surface oxidation degree O2 of the condensation region; Wherein, the laser fluence F is jointly determined by the pulse energy E and the scanning speed V, and is positively correlated with E and negatively correlated with V; the average characteristic size d is negatively correlated with the laser fluence F applied to the corresponding region, and the surface oxidation degree O is positively correlated with the laser fluence F applied to the corresponding region.

[0050] Using the composite effect of laser-material interaction, in a single processing process, the physical micro-nano structure (i.e., realizing the etching function) is directly formed by means of laser ablation, and meanwhile, the laser-induced in-situ thermo-chemical reaction is utilized to change the surface chemical composition (such as an oxidation reaction). This process is completed before the sealing welding, which can completely avoid the thermal damage caused by the subsequent welding heat input to the formed precise capillary structure, and fundamentally eliminate the inherent defects of the traditional method from the process sequence.

[0051] By controlling the first laser fluence (F1) applied to the evaporation region to be greater than the second laser fluence (F2) applied to the condensation region, that is, setting F1>F2, so as to coordinately regulate two key surface characteristics: Since the average characteristic size (d) of the micro-nano capillary structure is negatively correlated with the laser fluence (F), that is, d∝1 / F, a higher F1 will lead to the formation of a denser capillary structure with a smaller characteristic size (d1<d2) in the evaporation region, thereby generating a stronger capillary pumping force.

[0052] At the same time, the surface oxidation degree (O) is positively correlated with the laser fluence (F), that is, O∝F. A higher F1 enables a more sufficient oxidation reaction on the surface of the evaporation region to obtain a higher oxidation degree (O1>O2). The surface energy of metal oxides is usually higher than that of the metal matrix.

[0053] The morphology gradient (d1<d2) and the chemical gradient (O1>O2) act together and in the same direction, both leading to the surface energy (γ1) of the evaporation region being significantly higher than the surface energy (γ2) of the condensation region. This "composite gradient" obtained synchronously by regulating a single physical quantity (F) jointly constructs a strong self-driven surface energy gradient field pointing from the condensation region to the evaporation region.

[0054] The constructed surface energy gradient becomes an additional powerful driving force for the circulation of the working fluid, effectively improving the heat transfer limit and start-up performance of the heat spreader, and is especially beneficial for resisting gravity or long-distance heat transfer.

[0055] The micro-nano capillary structure is integrally formed with the substrate, eliminating the thermal resistance at the bonding interface. Furthermore, its performance is fully maintained during subsequent welding processes, ensuring the final performance and long-term reliability of the product and meeting the requirements of demanding applications such as automotive electronics.

[0056] By integrating microstructure processing and surface modification into a single process, the process flow is significantly shortened, and the equipment and space requirements are reduced.

[0057] Laser processing is fast, parameters are digitally controlled, it is easy to integrate into continuous production lines, and it can quickly adapt to the changing needs of different products for gradient design, thus achieving efficient and flexible production.

[0058] Dry laser processing avoids the waste liquid treatment problems caused by traditional wet chemical etching, which is more in line with the concept of green manufacturing.

[0059] The clearly defined quantitative relationship between "laser parameters (E, V) → fluence (F) → structural / chemical characteristics (d, O) → surface energy (γ)" transforms the process from experience-dependent to model-driven, improving the predictability and stability of the process window. This clear causal chain provides an ideal technical framework for constructing digital twins of processes, implementing AI-based global parameter optimization, and online quality monitoring, laying the foundation for intelligent manufacturing.

[0060] The laser processing technology, through the gradient control of laser flux, innovatively constructs a composite gradient of morphology and chemical composition in a single process. This not only fundamentally solves the contradiction between performance and reliability in traditional processes, but also achieves significant progress in manufacturing efficiency, controllability, and intelligent potential.

[0061] The working fluid in this embodiment is a composite phase change functional working fluid, which includes a base liquid phase, phase change material microcapsules dispersed in the base liquid phase, and magnetic nanoparticles loaded inside or on the surface of the phase change material microcapsules; the phase change material microcapsules absorb latent heat through a solid-liquid phase change when absorbing heat; the magnetic nanoparticles can generate motion or heat in response to an external magnetic field.

[0062] It significantly improves the equivalent specific heat capacity and thermal management capability of the working fluid, and realizes the intelligent thermal management function of "passive phase change thermal storage + active magnetocaloric regulation", which can effectively cope with transient pulse thermal loads in scenarios such as new energy vehicle batteries and server CPUs.

[0063] For example: The base liquid phase is deionized water, a liquid with high thermal conductivity.

[0064] Phase change material microcapsules, using paraffin (phase change temperature approximately 45-55℃) as the core material and urea-formaldehyde resin as the wall material, are prepared via in-situ polymerization. The average particle size is 10 micrometers. The latent heat of phase change in the microcapsules is used to absorb and release concentrated heat in the evaporation zone.

[0065] Magnetic nanoparticles, specifically iron(III) oxide (Fe3O4) nanoparticles with high saturation magnetization and an average particle size of 20 nanometers, are loaded onto the outer surface of urea-formaldehyde resin wall materials through surface modification (such as the use of silane coupling agents).

[0066] Basic fluidity is provided by deionized water. Paraffin microcapsules, acting as latent heat storage and transport units, melt endothermally in the evaporation zone and solidify exothermally in the condensation zone. Surface-loaded iron oxide nanoparticles endow the microcapsules with magnetic responsiveness, allowing them to preferentially accumulate and optimize their distribution in the evaporation zone under the guidance of an external magnetic field or driven by the magnetic force of a gradient-driven surface layer. This enhances heat transfer and is particularly suitable for applications requiring high thermal conductivity and rapid magnetic response.

[0067] For example: The base liquid phase uses a fluorinated liquid with a high boiling point and low saturated vapor pressure (such as FC-72).

[0068] Phase change material microcapsules are prepared by a sol-gel method using a mixture of polyols (such as a eutectic mixture of erythritol and mannitol with a phase change temperature of about 80-100℃) as the core material and silicon dioxide (SiO2) as the wall material. The average particle size is 5 micrometers.

[0069] Magnetic nanoparticles, specifically chemically stable nickel (Ni) nanoparticles with an average particle size of 50 nanometers, are coated within the silica wall material via a co-precipitation method, either between the core material and the wall material or embedded within the wall material.

[0070] The fluorinated liquid exhibits excellent chemical inertness and a wide operating temperature range. The polyol microcapsules are suitable for medium- and high-temperature applications. The encapsulated Ni nanoparticles provide magnetic responsiveness while avoiding direct contact with the working fluid, thus improving long-term stability. Similarly, the distribution of the microcapsules can be controlled via a magnetic field, synergistically driving the surface layer through a gradient to optimize thermal management performance, focusing on applications with wide temperature ranges and antioxidant properties.

[0071] The surface energy gradient distribution of the gradient-driven surface layer synergizes with the magnetic response characteristics of the magnetic nanoparticles. Specifically, the surface energy of the gradient-driven surface layer increases from the condensation zone to the evaporation zone, and within the evaporation zone, the surface energy at points with high heat flux density is higher than that of their surrounding areas. The surface energy gradient guides magnetic nanoparticles to regions with higher surface energy, such as the evaporation zone, particularly where hotspots accumulate. These accumulated magnetic nanoparticles can generate additional Joule heating or induce micro-convection under the influence of an external or internal alternating magnetic field. This achieves intelligent targeted enhanced heat dissipation for localized hotspots. The surface energy gradient guides the working fluid and functional particles to actively flow to the areas most in need of heat dissipation, and the response characteristics of the magnetic particles further enhance heat transfer in these areas, enabling the heat spreader to have adaptive hotspot management capabilities. This represents a forward-looking technology for future intelligent thermal management systems.

[0072] The manufacturing process in this embodiment also includes a global process optimization step based on a digital twin model. This involves real-time acquisition of production line data and using reinforcement learning algorithms to iteratively optimize the process parameters for each step within the digital twin model, feeding the optimization results back to the physical production line for execution. A digital twin that interacts with and is dynamically updated in real-time with the physical production line is constructed, and artificial intelligence algorithms (reinforcement learning) are used to optimize and verify massive parameter combinations in a virtual space. This achieves intelligent manufacturing and adaptive, self-optimizing processes. It can dynamically compensate for batch-to-batch material differences and equipment status fluctuations, continuously improving product yield, performance consistency, and production efficiency. This represents a key technology for the intelligent and digital transformation of the high-end equipment manufacturing industry, significantly reducing R&D and trial-and-error costs and improving production resilience and quality levels.

[0073] Specifically, the global process optimization steps based on the digital twin model include: S51. Digital twin model construction: Establish a digital twin model that is synchronously mapped to the physical production line; A digital twin model includes at least: a geometric model of the production line equipment, a mechanism model reflecting the physical laws of each process from stamping to laser processing to welding to liquid injection, and a proxy model for characterizing the quality of intermediate products; S52. Real-time data synchronization and feature extraction: During the operation of the physical production line, process parameter data from the stamping unit, laser processing unit, assembly and welding unit, and liquid injection and packaging unit, intermediate product quality data from the online optical inspection device and online contact angle measurement device, and welding process data from the high-speed infrared thermal imager are collected in real time; the collected multi-source heterogeneous data are time-series aligned, cleaned, and feature extracted to form a unified format process-quality feature vector, and synchronously updated to the digital twin model; S53. Reinforcement learning optimization iteration: In the digital twin model, the process-quality feature vector is used as the state input, the adjustment command of the key process parameters of each unit is used as the action output, and the thermal resistance value or thermal response speed of the final composite heat spreader predicted by simulation is used as the reward signal. Iterative simulation is carried out through reinforcement learning algorithm. The reinforcement learning algorithm finds the process parameter adjustment strategy that optimizes the reward signal by exploring and utilizing. S54. Optimize instruction issuance and verification: Dynamically issue the optimal set of process parameters or parameter adjustment strategies obtained in step S53 to the control system of the corresponding workstation on the physical production line for execution; and after execution, collect a new round of actual product quality data and feed it back to the digital twin model for online correction and updating of the surrogate model and reinforcement learning strategy.

[0074] By constructing a digital twin model, a virtual mapping is established that is fully synchronized with the physical production line in terms of geometry, physics, and behavioral rules. The geometric model describes the three-dimensional structure of equipment and products; the mechanistic model simulates the physical processes of stamping, laser modification, welding, and liquid injection based on physical laws such as fluid mechanics, heat transfer, and materials science; and the surrogate model (such as a neural network) learns from historical data to establish rapid predictive relationships from process parameters to intermediate product quality (such as groove morphology, surface energy, and weld quality). A high-fidelity virtual production environment is constructed, providing an accurate "digital sandbox" for subsequent simulation optimization. The mechanistic model ensures the physical credibility of the optimization, while the surrogate model enables rapid evaluation of complex nonlinear relationships. Together, they constitute the core foundation of intelligent manufacturing—a calculable, predictable, and optimizable digital image of a production system.

[0075] By synchronizing data in real time and extracting features, multi-source data (process parameters, quality data, and process data) from the entire physical production line are captured in real time through a sensor network (optical detection, contact angle measurement, infrared thermography, etc.). Data fusion, cleaning, and feature engineering are then performed to transform the raw data into key feature vectors characterizing process status and product quality. This achieves real-time and accurate information synchronization between the physical and digital worlds. Heterogeneous data from different stages such as stamping, laser processing, assembly and welding, and liquid injection packaging are unified into machine-readable and algorithmically processable process-quality feature vectors. This provides high-quality real-time input for the continuous calibration and optimization of digital twin models, serving as the data foundation for building an intelligent "perception-decision" closed loop.

[0076] By using reinforcement learning for iterative optimization, the manufacturing process is abstracted into a sequential decision-making problem. The process-quality feature vector generated by S52 serves as the environmental "state," the adjustment instructions for key parameters of equipment (such as stamping presses, lasers, and welding machines) (such as pressure, power, and speed) serve as the agent's "actions," and the final product performance indicators (thermal resistance, thermal response speed) predicted by the digital twin model serve as the "reward." Reinforcement learning algorithms (such as Deep Deterministic Policy Gradient (DDPG)) autonomously optimize the process parameter strategy that maximizes the "reward" through extensive trial and error (exploration and utilization) in a virtual environment. This replaces the traditional trial-and-error method and offline optimization that rely on engineer experience, achieving autonomous, online, and global optimization of process parameters. This step can dynamically adapt to material batch fluctuations and equipment status changes, automatically finding the globally optimal solution under multi-process collaboration (such as the best match between stamping depth, laser energy, and welding heat input), thereby systematically improving the performance (such as lower thermal resistance) and consistency of the final product. It is a concentrated manifestation of intelligent decision-making capabilities in the high-end equipment manufacturing industry.

[0077] By optimizing instruction issuance and verification, the optimal parameter set or dynamic adjustment strategy obtained from virtual space optimization in step S53 is issued to the programmable logic controller or CNC system of the corresponding equipment on the physical production line via the industrial network for execution. After execution, actual product performance data is acquired again through the data acquisition system in step S52 and compared with the prediction results of the digital twin model. The deviation data is used to train and correct the surrogate model and reinforcement learning strategy network online. This forms a complete intelligent closed-loop control system of "perception-optimization-execution-feedback". This not only realizes the implementation of optimization strategies, but more importantly, through the feedback of actual production data, it continuously iterates and improves the accuracy and adaptability of the digital twin model and optimization algorithm, enabling the entire manufacturing system to have self-learning and self-evolutionary intelligent characteristics, fundamentally ensuring the ultimate stability and continuous improvement of product performance in large-scale production.

[0078] Implementation of Global Process Optimization Based on Digital Twin Model This embodiment provides a specific implementation of a global process optimization step based on a digital twin model, which achieves dynamic optimization and continuous improvement of process parameters through the following closed-loop process.

[0079] 1. System Architecture and Data Foundation The central control platform deploys digital twin models and reinforcement learning algorithms. Each unit on the physical production line (stamping, laser processing, assembly and welding, liquid injection and packaging) is interconnected with the platform via industrial Ethernet, uploading data in real time. Stamping unit: stamping force (P_stamping, typical value 100kN, range 90-110kN), die stroke.

[0080] Laser processing unit: laser pulse energy (E_laser, typical value 3mJ, range 1-10mJ), scanning speed (V_laser, typical value 500mm / s, range 100-2000mm / s); online detection of groove depth (D_groove, target value 50±5μm), and contact angle between evaporation zone and condensation zone (θ_evap, θ_cond).

[0081] Assembly and welding unit: welding laser power (P_weld, typical value 300W, range 200-400W); peak temperature of the molten pool monitored by infrared thermal imager (T_peak).

[0082] Liquid injection and packaging unit: working fluid injection volume, cavity vacuum degree.

[0083] 2. Digital Twin Model Construction Establish a virtual model synchronized with the physical production line and integrate it: Mechanism models: a stamping model based on finite element analysis (FEA), a laser surface modification model based on thermal-fluid coupling, and a welding pool dynamics model are used to simulate the physical processes of the process.

[0084] Proxy Model: A Gaussian process regression (GPR) model is adopted. This model takes real-time acquired process parameters (such as P_stamping, E_laser) and intermediate quality characteristics (such as D_groove, θ_evap) as inputs to quickly predict the final thermal resistance value (R_th_pred) of the heat spreader, and performs real-time optimization instead of the computationally time-consuming full-order mechanistic model.

[0085] 3. Strengthen learning and optimize execution processes Optimization in a digital twin environment is carried out iteratively, with the following core elements: State (s_t): Defined as the "process-quality feature vector" of the current product.

[0086] For example: s_t=[P_stamping=102kN, D_groove=48.5μm, θ_evap=10°, T_peak=1410°C,...].

[0087] Action (a_t): Defined as the vector of instructions for adjusting key controllable parameters, i.e., a_t=[ΔP_stamping,ΔE_laser,ΔV_laser,ΔP_weld].

[0088] For example, a specific action output could be [+1kN, -0.1mJ, +20mm / s, -5W].

[0089] Reward (r_t): The reward function is designed with r_t = -1.0. R_th_pred-0.1 ||a_t|2. Where R_th_pred is the thermal resistance value predicted by the surrogate model based on the adjusted parameters; ||a_t||2 is the L2 norm of the action vector, used to quantify the adjustment magnitude and penalize excessive adjustments to maintain stable production.

[0090] Policy learning: The Proximal Policy Optimization (PPO) algorithm is employed. The agent undergoes millions of simulation trials in a digital twin environment to learn a mapping policy from any state s_t to the optimal action a_t, in order to maximize cumulative reward.

[0091] 4. Online deployment, verification, and continuous optimization Command issuance and execution: The optimal actions (i.e., the process parameter adjustment set) recommended by the trained policy network for the real-time production status s_t_real are issued to the production line for execution. For example: "Adjust the stamping pressure to 103kN, set the laser energy to 2.95mJ, set the scanning speed to 520mm / s, and set the welding power to 295W."

[0092] Closed-loop verification and model update: After production with new parameters, the actual thermal resistance R_th_real of the finished product is measured. The actual data pair {s_t_real, a_t, R_th_real} is fed back to the system for periodic fine-tuning of the surrogate model (GPR) and updating of the reinforcement learning strategy, enabling the system to adapt to drifting production conditions (such as equipment wear) and achieve continuous self-optimization.

[0093] This embodiment concretizes digital twins and reinforcement learning into an executable closed-loop optimization system. This system dynamically optimizes core process parameters such as stamping, laser processing, and welding, autonomously finding the optimal process window to improve product thermal performance while ensuring production stability. This realizes the practical application of intelligent manufacturing in the production of high-end heat spreaders.

[0094] On the other hand, a composite heat exchange plate is also provided, manufactured by the above-mentioned process, including: The shell is formed by sealing and fitting an upper shell and a lower shell together. The shell is formed by stamping a layered composite plate, and the inner wall of the shell is provided with micron-level fractal main grooves. A support structure is installed in the evaporation zone of the lower shell; The sealing welds formed at the periphery of the upper and lower shells are formed by laser-ultrasonic composite welding. A gradient-driven surface layer covers the micron-scale fractal trunk trenches and the surface of the supporting structure. The gradient-driven surface layer has a surface energy gradient from the condensation zone to the evaporation zone.

[0095] The aforementioned manufacturing process produces a high-performance composite vapor chamber product that integrates structure, function, and performance. This product not only possesses a precise internal structure resulting from its unique manufacturing process (micron-level fractal grooves, micro / nano capillary layers, gradient-driven surfaces, and integrated / embedded supports), but its manufacturing process also ensures high reliability, high consistency, and potential customizable and intelligent thermal management capabilities. This product is the culmination of the aforementioned advanced manufacturing processes and the embodiment of their commercial value. Its performance advantages directly address the urgent needs of fields such as new energy vehicle electric drive systems, 5G communication equipment, and high-end servers for efficient, reliable, and lightweight heat dissipation solutions.

[0096] The layered composite plate has a three-layer composite structure of copper-intermediate-copper, with at least one of the two copper layers having a functionally graded structure, and the intermediate layer being a stainless steel layer. Different combinations of structures and materials allow the functionally graded structure to adapt to different thermal and mechanical requirements, while the stainless steel provides stable intermediate support. This enhances the overall performance of the layered composite plate and meets the complex operating conditions required by heat exchangers.

[0097] Specifically, the metal layer is a composite functional interlayer; the composite functional interlayer includes a molybdenum-based functional layer, which is adjacent to the stainless steel layer and forms a metallurgical bonding interface between them. The thermal expansion coefficient of the molybdenum-based functional layer is lower than that of the stainless steel layer; the metallurgical bonding interface enhances the bonding force, and the difference in thermal expansion coefficients reduces interface stress when heated. This improves the structural stability of the interlayer and prevents problems such as delamination caused by thermal stress.

[0098] The molybdenum-based functional layer contains micropores, or it may be a copper-based composite material layer reinforced with molybdenum fibers. The micropores increase surface area and improve wettability; the molybdenum fiber reinforcement enhances strength and other properties. Optimizing the performance of the molybdenum-based functional layer enhances the function of the intermediate layer.

[0099] Preferably, the stainless steel layer has a perforated structure with a regular array of micropores. This perforated structure increases the surface area, facilitating fluid flow and heat exchange, thus improving the heat transfer efficiency of the heat spreader.

[0100] Specifically, the support structure is a ceramic support column that is cold-extruded and embedded in the lower shell, and the inner wall of the upper shell is provided with an inwardly protruding flow guide structure located on the side of the ceramic support column.

[0101] In another embodiment, the support structure is a metal micro-lattice structure integrally stamped with the lower shell. The inner wall of the upper shell is provided with an inwardly protruding flow-guiding structure, which at least partially penetrates the metal micro-lattice structure.

[0102] Both cold-extruded ceramic support columns integrated into the lower shell and integrally stamped metal micro-lattice structures with the lower shell can ensure shell strength, while flow-guiding structures guide fluid flow. This enhances the stability of the heat spreader structure, optimizes fluid flow paths, and improves heat dissipation efficiency.

[0103] The flow-guiding structure is selected from at least one of the following: radial ribs extending from the inner wall of the upper shell towards the center of the evaporation zone, spiral grooves, and a multi-level fractal groove network structure. The flow-guiding structure has a surface energy gradient, with the surface energy increasing in the direction from the condensation zone towards the center of the evaporation zone. This specific structure guides the directional flow of fluid, and the surface energy gradient utilizes capillary forces to promote fluid movement from the condensation zone to the evaporation zone. This improves fluid flow efficiency and enhances heat transfer capacity.

[0104] Specifically, the gradient-driven surface layer is a composite gradient surface layer that combines microscopic morphological gradients and surface chemical gradients. The working fluid injected into the shell is a composite phase change working fluid, which includes a base liquid phase, phase change material microcapsules dispersed in the base liquid phase, and magnetic nanoparticles loaded in or on the surface of the phase change material microcapsules. Furthermore, the surface energy of the gradient-driven surface layer increases from the condensation zone to the evaporation zone. In the evaporation zone, the surface energy at the high heat flux density points is higher than the surface energy of the surrounding areas at the high heat flux density points.

[0105] Microscopic morphology gradients influence fluid flow and wettability, while surface chemical gradients alter surface energy and other properties. Optimizing surface properties enhances fluid driving ability and improves heat dissipation.

[0106] Composite phase change functional working fluids utilize the endothermic properties of phase change to guide fluid accumulation in high heat flux density regions through surface energy gradients. This improves the heat dissipation efficiency of vapor chambers, especially enhancing their heat dissipation capacity in high heat flux density regions, thus meeting the heat dissipation requirements of high-power equipment.

[0107] The gradient setting of the surface energy of the gradient-driven surface layer promotes directional fluid flow, reduces disordered fluid flow within the vapor chamber, and accelerates the fluid circulation speed between the condensation and evaporation zones, thereby improving heat transfer efficiency and enabling the vapor chamber to dissipate heat from the heat source more quickly. Within the evaporation zone, the surface energy at high heat flux density points is higher than the surrounding areas, causing more fluid to concentrate in these high-heat areas, effectively removing a large amount of heat and preventing localized overheating. This enhances the vapor chamber's heat dissipation capacity for high-power-density heat sources, meeting the demands for efficient heat dissipation in fields such as electronic devices with increasing power density and new energy vehicles and high-speed communications. The overall surface energy gradient design and the special settings for high heat flux density points within the evaporation zone result in a more rational fluid distribution and more uniform heat transfer within the vapor chamber, optimizing the overall heat dissipation performance, improving its reliability and stability, and facilitating its application in industrial scenarios requiring high reliability and high-volume production (such as the automotive industry).

[0108] The above description is only a preferred embodiment of the present invention. For those skilled in the art, there will be changes in the specific implementation and application scope based on the ideas of the present invention. The content of this specification should not be construed as a limitation of the present invention.

Claims

1. A manufacturing process for a composite heat spreader, characterized in that, Includes the following steps: In the stamping step, a multi-station continuous die is used to stamp the layered composite sheet at room temperature to form an upper shell and a lower shell respectively; wherein, in the first station, micron-level fractal main grooves are stamped on the layered composite sheet; in the second station, a support structure is formed in the preset evaporation zone of the lower shell; In the laser processing step, the stamped upper and lower shells are sent into the laser processing module. The laser is used to etch micro-nano capillary structures on the surface of the micron-level fractal main groove and the support structure and to modify the surface energy, thereby forming a gradient-driven surface layer extending from the condensation zone to the evaporation zone. The assembly and welding process involves aligning and closing the upper and lower housings, bringing the end of the support structure into contact with the inner wall of the upper housing, and then performing peripheral sealing welding using a laser-ultrasonic composite process. The laser-ultrasonic composite process includes: applying high-frequency ultrasonic vibration to clean and initially connect the welding interface, while using fiber laser to perform circumferential seam scanning welding, and using welding energy to simultaneously connect the end of the support structure to the inner wall of the upper shell. The liquid injection and sealing step involves evacuating the welded shell through a pre-reserved liquid injection port and injecting the working fluid, followed by sealing the liquid injection port.

2. The manufacturing process of a composite heat spreader according to claim 1, characterized in that, The layered composite material is a three-layer composite structure of copper-intermediate layer-copper, wherein two copper layers are bonded to both sides of the intermediate layer through a solid-state composite process; At least one of the two copper layers is a functionally graded structure, wherein the composition or structure of the functionally graded structure changes continuously or stepwise from its working surface to the interface with the intermediate layer, and the thermal conductivity of the working surface is higher than that of the region near the interface, and the strength and hardness of the region near the interface are higher than those of the working surface. The intermediate layer is a metal layer.

3. The manufacturing process of a composite heat spreader according to claim 2, characterized in that, The metal layer is a stainless steel layer; the stainless steel layer has a hollow structure with a regular array of micropores.

4. The manufacturing process of a composite heat spreader according to claim 3, characterized in that, The metal layer is a composite functional intermediate layer; The composite functional intermediate layer includes a molybdenum-based functional layer, which is disposed adjacent to the stainless steel layer and forms a metallurgical bonding interface between them. The coefficient of thermal expansion of the molybdenum-based functional layer is lower than that of the stainless steel layer. The molybdenum-based functional layer has micropores distributed within it, or the molybdenum-based functional layer is a copper-based composite material layer reinforced with molybdenum fibers.

5. The manufacturing process of a composite heat spreader according to claim 1, characterized in that, In the second station of the stamping step, a reserved micropore is formed in the preset evaporation zone of the lower shell by stamping, and the ceramic support column is cold-extruded and embedded into the reserved micropore to form the support structure; An inwardly protruding flow guiding structure is integrally stamped on the inner wall of the upper shell; after the assembly and welding steps are completed, the flow guiding structure is located on the side of the ceramic support column to distribute the working fluid to the evaporation zone around the ceramic support column.

6. The manufacturing process of a composite heat spreader according to claim 1, characterized in that, In the second station of the stamping step, a local area of ​​the layered composite plate is stamped to form a metal micro-lattice structure integrally formed with the lower shell, which serves as the support structure. In the stamping step, an inwardly protruding flow guiding structure is integrally stamped on the inner wall of the upper shell; after the assembly and welding step is completed, the flow guiding structure is at least partially inserted into the metal micro-lattice structure to distribute the working fluid to the evaporation zone inside the metal micro-lattice structure.

7. The manufacturing process of a composite heat spreader according to claim 1, characterized in that, In the laser processing step, by adjusting the laser processing parameters applied to different areas of the heat spreader shell, the etching of the micro-nano capillary structure on the shell surface and the surface chemical modification are completed simultaneously during a single processing step, so as to form a gradient-driven surface layer extending from the condensation zone to the evaporation zone. The laser processing parameters include at least pulse energy E and scanning speed V. By controlling the first laser flux F1 acting on the evaporation zone to be greater than the second laser flux F2 acting on the condensation zone, the average characteristic size d1 of the micro-nano capillary structure formed in the evaporation zone is smaller than the average characteristic size d2 of the condensation zone, and the surface oxidation degree O1 of the evaporation zone is higher than the surface oxidation degree O2 of the condensation zone. The laser flux F is determined by both pulse energy E and scanning speed V, and is positively correlated with E and negatively correlated with V; the average feature size d is negatively correlated with the laser flux F acting on the corresponding region, and the surface oxidation degree O is positively correlated with the laser flux F acting on the corresponding region.

8. The manufacturing process of a composite heat spreader according to claim 1, characterized in that, The working fluid is a composite phase change functional working fluid, which includes a base liquid phase, phase change material microcapsules dispersed in the base liquid phase, and magnetic nanoparticles loaded in or on the surface of the phase change material microcapsules. The surface energy gradient distribution of the gradient-driven surface layer is coordinated with the magnetic response characteristics of the magnetic nanoparticles; wherein, from the condensation zone to the evaporation zone, the surface energy of the gradient-driven surface layer increases, and in the evaporation zone, the surface energy of the corresponding high heat flux density points is higher than the surface energy of their surrounding areas.

9. The manufacturing process of a composite heat spreader according to claim 1, characterized in that, The manufacturing process also includes a global process optimization step based on a digital twin model. This involves collecting production line data in real time and using reinforcement learning algorithms to iteratively optimize the process parameters of each step in the digital twin model, and then feeding the optimization results back to the physical production line for execution.

10. A composite heat spreader, manufactured by the process described in any one of claims 1 to 9, characterized in that, include: The shell is formed by sealing and fitting an upper shell and a lower shell together. The shell is formed by stamping a layered composite plate, and the inner wall of the shell is provided with micron-level fractal main grooves. A support structure is provided in the evaporation zone of the lower shell; A sealing weld is formed at the periphery of the upper and lower shells, and the sealing weld is formed by laser-ultrasonic composite welding; A gradient-driven surface layer covering the surface of the micron-scale fractal main trench and the supporting structure has a surface energy gradient from the condensation zone to the evaporation zone.

Images