Thermoelectric module, integrated thermoelectric module and closed-loop collaborative design and manufacturing method thereof

Abstract

The application discloses a thermoelectric module, an integrated thermoelectric module and a closed-loop collaborative design method and manufacturing method thereof. The design method is based on a preset system thermal management target, and selects between a first mode of optimizing the temperature uniformity of the module itself and a second mode of optimizing the cooling effect on an external heat source. Through target-oriented thermal analysis, a closed-loop configuration scheme containing a non-uniform reconstruction scheme of the thermoelectric module body and a peripheral structure scheme matched with the non-uniform reconstruction scheme is synchronously generated, and a design leap from passive adaptation to active collaborative management is realized. Based on the method, the application further provides an integrated thermoelectric module in which a thermoelectric unit is directly eutectically welded to a metal heat dissipation substrate and a manufacturing method thereof, and a traditional thermoelectric module optimized by using the method. The application significantly improves the spatial accuracy of thermal management and the overall energy efficiency of the system, and is suitable for a wide range of application scenarios from high reliability to high performance cooling.

Description

Technical Field

[0001] This invention relates to the field of semiconductor thermoelectric technology, specifically to a thermoelectric module, an integrated thermoelectric module, and a closed-loop collaborative design and manufacturing method thereof. Background Technology

[0002] Thermoelectric cooling technology, based on the Peltier effect, boasts advantages such as no moving parts, rapid response, precise temperature control, and high reliability, and has been widely applied in precision temperature management in fields such as optoelectronic devices, biomedical equipment, and high-performance integrated circuits. However, with the continuous increase in chip power density, the surface heat flux distribution exhibits high spatial non-uniformity, with significant local "hot spots," posing a systemic challenge to the design concepts, methodologies, and physical structures of traditional thermoelectric modules.

[0003] Currently, the technological bottlenecks in this field are mainly reflected in the following aspects:

[0004] First, the passive and uniform design philosophy fails to achieve precise thermal management. Mainstream thermoelectric modules generally adopt a "homogenized" design with uniformly arranged thermoelectric units and a globally unified circuit configuration. When faced with actual non-uniform heat loads, this design cannot provide targeted and enhanced cooling to high-heat areas, resulting in low overall system efficiency. It operates in a state of "passively adapting" to the heat source distribution, lacking proactive management capabilities. Furthermore, the "homogenized" design may also cause uneven heating within the thermoelectric module itself, thereby reducing its cooling efficiency and reliability.

[0005] Second, the design paradigm is fragmented and "open-loop," lacking systematic coordination, resulting in unresolved core objectives. Existing technologies generally follow a linear design process of "body first, then periphery": first, the design scheme of the thermoelectric module body (such as particle arrangement and circuitry) is determined independently, and then peripheral structures such as heat sinks are matched independently. This "open-loop" or "segmented" design paradigm creates serious coordination barriers. 1. Lack of process feedback: The design decisions of the ontology are not used as core input to actively and accurately define the design requirements of the peripheral structure; the design of the peripheral structure cannot back-optimize the initial solution of the ontology. The two are merely a simple "patchwork" rather than a deep "coupling".

[0006] 2. Ambiguous Goal Orientation: Due to the aforementioned process fragmentation, a fundamental design contradiction is obscured or ignored at the very beginning of the process: facing a non-uniform thermal field, should priority be given to ensuring the working efficiency and reliability of the thermoelectric module itself ("module efficiency priority"), or to ensuring the overall cooling effect on the external heat source ("system efficiency priority")? Existing methodologies lack a mechanism to clarify this goal at the beginning of the process and to ensure it runs through the entire collaborative design process.

[0007] Third, the control strategy is coarse and makes it difficult to achieve precise thermal management: Traditional drive and control methods usually apply a uniform current or voltage to the entire thermoelectric module, which is a "coarse-grained" control. This cannot perform independent power distribution and fine adjustment of thermoelectric units in different areas inside the module, and therefore it is difficult to effectively track and suppress local temperature fluctuations caused by dynamic changes in heat sources or differences in heat dissipation conditions.

[0008] Fourth, the physical structure has inherent bottlenecks that limit performance. Traditional thermoelectric modules typically employ a multi-layer stacked structure of "ceramic insulating substrate-solder-thermoelectric unit-solder-ceramic substrate," resulting in significant accumulation of interfacial thermal resistance, which becomes a major obstacle to improving cooling efficiency and power density. In addition, the ceramic substrate is brittle and has limited processable dimensions, making it difficult to meet the demands of high-power, large-area applications.

[0009] Furthermore, in traditional structures, the electrical connections at the hot and cold ends are typically symmetrically designed, failing to consider the high thermal resistance at the hot end and the heat loss due to heat radiation from the cold end to the environment. Additionally, the transducer plate at the cold end in traditional structures is usually made of the same material and structure, making it difficult to adapt to the precise requirements of non-uniform heat sources.

[0010] Therefore, there is an urgent need in this field for a technological solution that comprehensively innovates from design concepts and methodologies to physical structures, in order to break through the limitations of "passive uniform response" and "open-loop segmented design" and achieve a technological leap towards "proactive and precise management" and "closed-loop collaborative design". Summary of the Invention

[0011] The thermoelectric module involved in this invention refers to an independent device composed of thermocouples for realizing thermoelectric energy conversion. Its internal structure includes thermoelectric materials, electrodes, substrates, etc., but does not include external structures such as heat dissipation components. The integrated thermoelectric module involved in this invention refers to a thermoelectric module assembly with external structures such as heat dissipation components.

[0012] The technical problem to be solved by the present invention is to overcome the above-mentioned defects of the prior art, and aims to provide a thermoelectric module, an integrated thermoelectric module and its closed-loop collaborative design and manufacturing method.

[0013] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: Firstly, this invention provides a closed-loop collaborative design method for integrated thermoelectric modules. The core of this method lies in overturning the traditional "open-loop" linear design and constructing a systems engineering process of "target setting → analysis → collaborative generation → closed-loop verification." Its key lies in forcibly establishing an inseparable collaborative and data-dependent relationship between the thermoelectric module's main design and its peripheral structural design, specifically including: S1. Target mode selection: Determine system priorities from the outset and choose between "thermoelectric module efficiency priority mode" (prioritizing the uniformity of the module's own operating temperature) and "system cooling efficiency priority mode" (prioritizing the overall cooling effect and temperature uniformity of external heat sources) to provide clear guidance for the entire design process.

[0014] S2. Target-oriented thermal analysis: Based on the selected mode, obtain the expected internal temperature field of the thermoelectric module itself or the heat flux density distribution on the surface of the external heat source.

[0015] S3. Generate a closed-loop collaborative configuration scheme: This step is the essence of this method. Based on the analysis in S2, a holistic scheme comprising the following two parts is generated synchronously and collaboratively: S301 Thermoelectric Module Body Adaptive Reconfiguration Scheme: For different characteristic regions in the thermal field, determine the linkage adjustment strategy of the spatial layout parameters, circuit topology parameters, geometric dimension parameters (cross-sectional area and / or height) and material composition parameters of the thermoelectric unit.

[0016] The S302 peripheral structure matching design scheme: The design input of this scheme is not the initial thermal field of S2, but must be based on the new thermal state of the hot and cold ends of the thermoelectric module expected after implementing the S301 partial reconstruction scheme. This means that the peripheral structure (heat sink, cold end transducer plate) is specially tailored for the "reconstructed module".

[0017] Furthermore, the adjustment of the spatial layout parameters includes changing the distribution density and / or planar geometric arrangement pattern, with the specific pattern selection based on its physical characteristics and the correlation with thermal management objectives; the adjustment of the circuit topology parameters includes selecting a circuit topology between parallel-dominant and series-dominant types.

[0018] Furthermore, when the system is operating in the first operating mode, the adaptive reconfiguration scheme in step S301 is as follows: for the region where the thermoelectric module is expected to have a high temperature, adjust its circuit topology to be series-dominant, and increase the overall resistance of the thermoelectric unit in this region by reducing the cross-sectional area of ​​the thermoelectric particles, increasing the height, and sparse layout, thereby reducing the operating current and Joule heat generation, and optimizing the uniformity of the module's operating temperature.

[0019] Furthermore, when the system is operating in the second operating mode, the adaptive reconfiguration scheme for the region with high heat flux density of the external heat source in step S301 is as follows: adjust its circuit topology to a parallel-dominant type, and reduce the overall resistance of the thermoelectric units in this region by means of denser configuration of thermoelectric units, increasing the cross-sectional area of ​​thermoelectric particles, reducing the height, and selecting thermoelectric units made of materials with lower resistivity, thereby improving the operating current and Peltier cooling capacity and strongly suppressing external hot spots.

[0020] Furthermore, the “matching peripheral structure scheme” mentioned in step S302 refers to: determining the design parameters of the peripheral structure based on the expected changes in the thermal state of the hot and cold ends of the thermoelectric module after executing the adaptive reconfiguration scheme described in step S3(a).

[0021] Furthermore, the peripheral structure scheme is specifically embodied in at least one of the following: (1) A heat dissipation structure designed to have a non-uniform heat dissipation capacity distribution to match the expected heat flux density distribution at the hot end; or (2) A cold-end transducer structure, which is designed as a partitioned structure, wherein the transducer components of at least two partitions differ in at least one of the following aspects: material thermal conductivity, structural thickness, surface treatment process, and the secondary thermal management method coupled thereto.

[0022] In a second aspect, the present invention provides an optimized thermoelectric module comprising the following structure: A thermally conductive metal substrate, wherein the thermally conductive metal substrate constitutes part of a heat dissipation structure; An insulating and thermally conductive layer is formed on at least one main surface of the thermally conductive metal substrate; Multiple hot-end electrical connection portions are disposed on the insulating and heat-conducting layer; Multiple cold-end electrical connections; Multiple P-type and N-type thermoelectric units are provided, with each thermoelectric unit having its two ends directly welded to a hot-end electrical connection and a cold-end electrical connection via a eutectic bonding process. This allows the multiple thermoelectric units to be electrically connected via the hot-end and cold-end electrical connection, forming a thermoelectric functional array.

[0023] The aforementioned thermoelectric module abandons the traditional multi-layered and complex interface of "thermoelectric unit-solder-ceramic insulating substrate-thermal grease-heat sink," innovatively employing a eutectic bonding process to directly weld the thermoelectric units to the electrical connection points on the metal substrate. This thermally conductive metal substrate plays a dual core role in this innovative structure: firstly, it serves as the welding support substrate for the thermoelectric unit array; secondly, it itself constitutes an integrated heat conduction base for the heat sink. This design fundamentally eliminates the additional thermal resistance caused by multiple interfaces in traditional modules, achieving the shortest heat transfer path between the thermoelectric cooling / heat generation components and the final heat dissipation environment. Simultaneously, thanks to the mechanical strength and machinability of the metal substrate, this integrated thermoelectric module can be manufactured as a large-area, monolithic structure, fundamentally eliminating the mechanical splicing interfaces necessary in traditional high-power modules, thus improving heat conduction reliability and temperature uniformity.

[0024] It should be specifically noted that the closed-loop collaborative design method (claims 1-7) and the metal-based thermoelectric module structure (claim 8) proposed in this invention are not independent inventions, but rather an integral technical solution that is interdependent and works synergistically. The metal-based structure, with its ultra-low interfacial thermal resistance and high reliability, provides the necessary hardware foundation for implementing the extreme operating conditions involved in the method, such as high current density and non-uniform heat flow; while the design method provides a systematic implementation path for fully leveraging the performance potential of this hardware platform. Together, they constitute a complete technical chain from design concept to physical implementation.

[0025] Furthermore, in the thermoelectric module, the insulating and thermally conductive layer is a composite layer formed by vacuum thermosetting after slit coating or spraying a mixed slurry containing polyimide and modified boron nitride.

[0026] In a third aspect, the present invention provides an integrated thermoelectric module, the structure of which includes: The thermoelectric module described in the second aspect; and A heat dissipation structure coupled to the hot side of the thermoelectric module; and / or a transducer structure coupled to the cold side of the thermoelectric module; The thermoelectric units in the thermoelectric module have a non-uniform configuration, and the non-uniform configuration includes inter-regional differences of at least one of the following characteristics: (1) Spatial distribution density of thermoelectric units; (2) Circuit topology connection method of thermoelectric unit; (3) Geometric dimensions of the thermoelectric unit; Furthermore, the heat dissipation capacity distribution of the heat dissipation structure matches the heat flux density distribution of the hot end face of the thermoelectric module, and the heat transfer capacity distribution of the transducer structure matches the heat flux density distribution of the external heat load.

[0027] Furthermore, in the integrated thermoelectric module, the non-uniform configuration of the thermoelectric units and the matching design of the heat dissipation structure and the energy transducer structure are determined according to the closed-loop collaborative design method described in any one of the first aspects.

[0028] Furthermore, the aforementioned integrated thermoelectric module also includes an adapter structure for compensating for height differences between different thermoelectric units, the adapter structure being implemented in at least one of the following ways: (a) Ensure that at least two electrical connections have different thicknesses; (b) A non-uniform interface is formed on the surface of the thermally conductive metal substrate that contacts the thermoelectric unit; (c) To form a non-uniform interface on the surface of the transducer structure that contacts the thermoelectric unit.

[0029] Depending on the actual situation, various methods can be used, such as different electrical connection thicknesses, transducer plates with different interface designs, or thermally conductive metal substrates, to effectively solve the installation problem when the height of the thermoelectric units is inconsistent, thus ensuring the performance of the integrated thermoelectric module.

[0030] Furthermore, in the aforementioned integrated thermoelectric module, the electrical connection portion has asymmetrical geometric features at the hot and cold ends: the electrical connection portion at the hot end has a larger cross-sectional area to reduce thermal resistance, while the electrical connection portion at the cold end has a smaller cross-sectional area and / or a surface shape treated with low radiation to suppress radiative cooling leakage.

[0031] Furthermore, in the aforementioned integrated thermoelectric module, the planar shape and arrangement of the electrical connection portion at the hot end are configured to match the projection area of ​​the heat dissipation unit on the thermally conductive metal substrate, thereby forming the shortest vertical heat conduction path from the thermoelectric unit to the heat dissipation unit. The aforementioned main heat dissipation functional unit refers to the core heat dissipation structure that is directly thermally coupled to the thermally conductive metal substrate and responsible for ultimately dissipating heat to the external environment. Its specific form can vary depending on the heat dissipation method. In air-cooled heat dissipation, the heat dissipation functional unit is the root section of the heat dissipation fins or heat dissipation columns on the substrate; in liquid-cooled heat dissipation, the heat dissipation functional unit is the projection area of ​​the microchannel or cooling channel on the contact surface of the substrate; in heat dissipation using heat pipes or vapor chambers, the heat dissipation functional unit is the contact area of ​​the heat pipe evaporation end or the vapor chamber cavity support structure.

[0032] Regardless of its specific form, the present invention ensures efficient vertical heat transfer by carefully aligning the layout of the electrical connections with the projection areas of these functional units.

[0033] A fourth aspect of the present invention provides a method for manufacturing the above-mentioned integrated thermoelectric module, comprising the following steps: A thermally conductive metal substrate is provided, wherein the thermally conductive metal substrate constitutes part of a heat dissipation structure; An insulating and thermally conductive layer is prepared on at least one main surface of the substrate; Multiple electrically isolated thermally conductive connections are provided on the insulating thermally conductive layer. Provides multiple cold-end electrical connections; The welding area of ​​the electrical connection is cleaned with plasma. In a vacuum or inert gas protected environment, based on the non-uniform configuration scheme determined by the method, P-type and N-type thermoelectric units are directly welded to the corresponding electrical connection parts by eutectic welding process.

[0034] Install the heat dissipation device and cold-end transducer plate determined by the closed-loop collaborative design method based on the integrated thermoelectric module.

[0035] In a fifth aspect, the present invention provides an optimized conventional integrated thermoelectric refrigeration module, comprising: A thermoelectric module, comprising an upper ceramic substrate, a lower ceramic substrate, and multiple P-type and N-type thermoelectric units sandwiched in between; and A heat dissipation structure that is thermally coupled to the thermoelectric module; The thermoelectric units in the thermoelectric module have a non-uniform configuration, and the non-uniform configuration includes inter-regional differences of at least one of the following characteristics: (1) Spatial distribution density of thermoelectric units; (2) Circuit topology connection method of thermoelectric unit; (3) Geometric dimensions of the thermoelectric unit; Furthermore, the heat dissipation capacity distribution of the heat dissipation structure matches the heat flux density distribution of the hot end face of the thermoelectric module.

[0036] Furthermore, in the integrated thermoelectric module, the non-uniform configuration of the thermoelectric units and the matching design of the heat dissipation structure are determined according to the closed-loop collaborative design method of the integrated thermoelectric module described in any one of the first aspects.

[0037] Compared with the prior art, the beneficial effects of the present invention are: 1. Established a closed-loop collaborative design paradigm: It completely changed the open-loop design of "first the core, then the periphery" and achieved precise control of the entire link from system objectives to final product performance through mandatory data dependence and collaborative decision-making.

[0038] 2. Resolved core design contradictions: Through a clear dual-mode framework of "module efficiency first" and "system performance first", it provides clear and different optimization paths for dealing with non-uniform thermal fields.

[0039] 3. Achieved multi-dimensional deep synergy: For the first time, the adjustment of electrical performance parameters of thermoelectric units, especially the adjustment of circuit topology, was elevated to a core position of equal importance with the adjustment of spatial layout, and its linkage design with the peripheral structure was realized, resulting in a system-level performance gain of "1+1>2".

[0040] 4. Provides an innovative, highly integrated physical carrier: The integrated metal-based thermoelectric module structure provides an ideal low thermal resistance and high reliability hardware platform for implementing the above-mentioned advanced design methods, and makes it possible to manufacture large-area monolithic thermoelectric modules.

[0041] 5. Possesses broad industrial compatibility and application prospects: The method and structure described are applicable to both the development of new high-performance integrated modules and the upgrading of traditional products, covering a wide temperature range from electronic cooling to industrial waste heat recovery. Attached Figure Description

[0042] Figure 1 A systematic flowchart of the closed-loop collaborative design method for integrated thermoelectric modules; Figure 2 Schematic diagram of a traditional homogenized thermoelectric module design Figure 3 A schematic diagram of a thermoelectric module designed for the module efficiency-first mode; Figure 4 A schematic diagram of a thermoelectric module designed for a system efficiency-priority mode; Figure 5 This is a schematic diagram of the structure of a metal-based integrated thermoelectric module; Figure 6 A schematic diagram of a heatsink design and a cold-end partitioned transducer plate; Figure 7 This is a schematic diagram of the optimized traditional ceramic-based integrated thermoelectric module. In the figure: 101-thermal conductive metal substrate; 102-insulating thermal conductive layer; 1032a, 1032b, 1032c-hot end electrical connection; 1031a, 1031b, 1031c-cold end electrical connection; 104a, 104b, 104c-thermoelectric unit; 201-upper ceramic substrate; 202-lower ceramic substrate; 203-thermoelectric unit; 204a-heat dissipation unit; 204b-heat dissipation unit; 204c-heat dissipation unit; 205-electrical connection; 206-insulating thermal conductive layer; 207-thermal conductive grease layer; 301-heat dissipation unit; 302-heat dissipation unit; 303-heat dissipation unit; 4011-first transducer plate; 4012-second transducer plate; 4013-third transducer plate. Detailed Implementation

[0043] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0044] Example 1: Systematic Implementation of Closed-Loop Collaborative Design Method for Integrated Thermoelectric Modules Reference Figure 1 The closed-loop collaborative design method for thermoelectric modules of the present invention includes the following steps: S1. Determine the system thermal management objective: Select either the first mode that optimizes the temperature uniformity of the thermoelectric module's own hot end face, or the second mode that optimizes the cooling effect on an external heat source. S2. Target-oriented thermal analysis: Analyzes the heat flux density distribution or expected temperature field corresponding to the selected model; S3. Generate a closed-loop non-uniform configuration scheme: Based on the thermal field distribution information obtained in step S2, derive an overall non-uniform configuration scheme; the scheme must be a closed-loop design result that includes the following two aspects, both of which are collaboratively determined based on the thermal field distribution information: S301 is an adaptive reconfiguration scheme for the thermoelectric module body, used to make differentiated configurations for different characteristic regions in the thermal field distribution information; S302 A peripheral structure scheme that matches the adaptive reconfiguration scheme in thermal and / or electrical performance is used to achieve efficient heat removal or precise cooling distribution. The "matching" mentioned here is not a simple selection or adaptation, but a reverse design process based on the coupling of thermal, fluid, and solid-state multiphysics fields. Specifically, after determining the non-uniform reconfiguration scheme of the thermoelectric module body in step S301, the expected heat flux density distribution of the hot end face of the thermoelectric module under this reconfiguration scheme is first obtained through simulation or analytical calculation. and the expected heat load distribution at the cold end face Subsequently, with the goal of minimizing the hot-end temperature or maximizing the cold-end heat transfer efficiency, the flow channel layout, fin density, heat transfer medium flow rate, and other parameters of the heat dissipation structure were designed in reverse to achieve a spatial distribution of heat dissipation capacity. ( , )and There is a positive correlation; similarly, the design of the cold-end transducer structure is based on The heat flow distribution from external heat sources can be spatially differentiated by using methods such as zoned material selection, variable thickness design, or integrated microchannels. This design process ensures that the peripheral structure is not a simple stacking of general-purpose components, but a customized solution deeply coupled with the reconstructed module.

[0045] The adaptive reconfiguration scheme of the main body is achieved through a linkage adjustment strategy between the circuit topology and the spatial layout parameters of the thermoelectric units (such as distribution density and / or planar geometric arrangement pattern) and / or the size / material of the thermoelectric units.

[0046] Changing the circuit topology involves switching between parallel-dominant and series-dominant configurations. In the first mode (module efficiency priority), the tendency is to maintain or adjust local circuitry to a series-dominant configuration, combined with sparse arrangement of thermoelectric units and high internal resistance design of thermoelectric particles in that area. This increases the overall resistance of the thermoelectric units in that area, thereby reducing operating current and Joule heating. This is a fundamental electrical strategy for optimizing the uniformity of the module's operating temperature and improving reliability. In the second mode (system performance priority), the tendency is to adjust the circuitry from a series-dominant to a parallel-dominant configuration in critical areas, and to configure thermoelectric units with lower internal resistance. The parallel structure provides multiple independent paths for current, and combined with low internal resistance units, it maximizes the current allowed to flow through the thermoelectric units in the target area, thereby stimulating the maximum Peltier cooling power in that area to powerfully suppress external hot spots.

[0047] Planar geometric layout patterns include square arrays, hexagonal close-packed patterns, gradient distributions, and directional hotspot clustering patterns. The specific pattern selection is based on its correlation with the thermal management objectives. Square arrays, as a basic and easy-to-design-and-manufacture regular layout, are particularly suitable for scenarios where the heat source distribution is relatively regular or where high consistency in manufacturing processes is required; hexagonal close-packed patterns help minimize thermal resistance and maximize local density; gradient distributions can naturally match heat flux gradients; and directional hotspot clustering is the optimal layout for achieving precise cooling.

[0048] Changing the geometry of a thermoelectric element involves altering its cross-sectional area and / or height, directly and linearly altering its resistance and thermal resistance. Increasing the cross-sectional area and decreasing the height of the thermoelectric particles reduces the overall resistance of the thermoelectric element, achieving a low internal resistance design; conversely, it allows for a high internal resistance design. In the second mode (system efficiency priority), reducing the height of the thermoelectric element is a key optimization for hot spots of external heat sources. This directly reduces the resistance of the thermoelectric element, enabling it to carry higher currents in parallel circuit topologies, thus significantly improving Peltier cooling power. It also reduces its axial thermal resistance, allowing the cooling generated at the cold end to be transferred to the external heat load with less resistance, achieving efficient and precise cooling delivery. This synergizes with the electrical effect of reducing resistance by increasing the cross-sectional area, while providing greater design flexibility in the thermal path. Coordinated design of thermoelectric element geometry allows for precise matching of the current carrying capacity and heat conduction requirements of different regions, serving as an important means of optimizing local current density distribution, interfacial thermal resistance, and even the overall mechanical structure.

[0049] Changing the material composition refers to selecting materials with suitable thermoelectric properties for different regions or the overall thermoelectric unit based on system thermal management objectives (such as target operating temperature range, temperature difference requirements, and efficiency requirements). For example, Bi2Te3-based materials can be used in near-room temperature refrigeration applications (such as electronic cooling); PbTe-based materials can be used in medium-temperature applications (such as industrial waste heat recovery); and SiGe-based materials can be used in high-temperature applications (such as spacecraft thermoelectric power generation). The core of this invention lies in providing a design framework that can be adapted to the above-mentioned wide range of material systems, rather than being limited to specific materials.

[0050] Another indispensable part of the closed-loop non-uniform arrangement scheme is to determine the design parameters of the peripheral structure based on the expected changes in the thermal state of the hot and cold ends of the thermoelectric module after the adaptive reconfiguration scheme. This includes: a heat dissipation structure designed with a non-uniform heat dissipation capacity distribution to match the expected heat flux density distribution at the hot end; and / or a cold end transducer structure designed as a partitioned structure, wherein the transducer components of at least two partitions differ in at least one of the following aspects: material thermal conductivity, structural thickness, surface treatment process, and coupled secondary thermal management method, to achieve precise cooling distribution. It should be noted that the 'secondary thermal management method' of the cold end transducer structure in this invention does not refer to 'heat dissipation' in the traditional sense of heat dissipation to the environment, but rather to auxiliary thermal control measures taken to optimize cooling distribution along the heat transfer path between the cold end transducer plate and the external heat load. Its purpose is to enhance the directional transfer of cold energy to the target area or reduce the loss of cold energy to non-target areas. Specifically, it may include, but is not limited to, integrating a heat exchange plate to enhance the lateral temperature uniformity, adding a heat insulation layer to reduce edge cold leakage, or combining micro thermoelectric components for local fine-tuning of cold energy.

[0051] To more intuitively demonstrate the application process and technical effects of the closed-loop collaborative design method of this invention, this embodiment uses the heat dissipation scenario of the same high-performance computing chip as an example, and uses quantitative calculations to demonstrate the application of the closed-loop collaborative design method of this invention in the first mode (module efficiency priority) and the second mode (system performance priority). For clarity, this embodiment sets the initial thermoelectric module as a traditional uniform design (such as...). Figure 2 (and uses a common circuit topology as a comparison benchmark).

[0052] I. Scenario Description and Common Premise Application target: High-performance computing chips.

[0053] Initial thermoelectric module: The thermoelectric units are evenly arranged in 10 rows × 10 columns. For ease of calculation and comparison, the initial circuit topology is set as follows: 10 thermoelectric units in each row (horizontal) are connected in series to form a branch, for a total of 10 such series branches. These 10 branches are then electrically connected in parallel. See [link to relevant documentation]. Figure 2 .

[0054] Basic parameters: Unit reference resistance R0 = 0.1Ω, Seebeck coefficient α = 0.2 mV / K, operating voltage V = 12V, unit cross-sectional area 1.0mm², thermoelectric particle height 1.5mm. Ambient temperature 30°C, maximum allowable chip temperature 85°C.

[0055] It should be noted that the thermal scenarios of the first and second modes described below are completely independent assumptions, used only for the purpose of comparative analysis.

[0056] II. First Mode (Module Efficiency Priority) Optimization Mode Selection and Objectives: Select Mode 1. The core objective is to identify and optimize the internal temperature unevenness issues generated by the thermoelectric module during operation, in order to significantly improve its long-term reliability and lifespan.

[0057] Target-oriented thermal analysis (for the module itself): Analyzing the expected internal temperature field of the initial module under standard operating conditions. Simulations show that due to uneven heat generation, the upper region of the module (region A, corresponding to the upper three rows of the thermoelectric module) is expected to have a heat generation power per unit area (mainly Joule heating) as high as 150 W / cm², with its thermoelectric unit hot end temperature expected to reach 95°C. In contrast, the lower region (region B) has a heat generation power per unit area (mainly Joule heating) of 50 W / cm², with a temperature of 65°C, resulting in an internal temperature difference of 30°C. This "hot spot" region A is the optimization target.

[0058] Closed-loop collaborative configuration scheme generation: Body reconstruction solution (for module's own high-temperature zone A): See Figure 3 a) Spatial layout optimization (sparserization): The number of thermoelectric units in area A is reduced from 3 rows of 30 units to 2 rows of 20 units (a reduction of 33.3%).

[0059] b) Geometry optimization (high resistance design): The cross-sectional area of ​​the thermoelectric unit is reduced from 1.0 mm² to 0.8 mm², and the height is increased from 1.5 mm to 1.8 mm, thereby increasing the unit resistance from 0.1 Ω to 0.15 Ω.

[0060] c) Circuit topology adjustment (series-dominated): Connect all 20 thermoelectric units in region A in series to form a single high-resistance branch.

[0061] Peripheral structure matching scheme: Based on the expectation that the heat flow at the hot end of the module will tend to be uniform after reconstruction, the heat dissipation structure can adopt a heat sink with uniform design, and the cold end transducer structure can adopt a copper plate with uniform thickness.

[0062] Implementation results and calculation verification: Calculations show that after optimization, the current in branch A of region A dropped significantly from the initial 12A to 4A, and the total net heat generation in region A dropped from 411W to 43.5W (a reduction of 89.4%).

[0063] The highest temperature at the hot end of the module was reduced from 95°C to 73°C, and the temperature difference was reduced from 30°C to 12°C, fundamentally improving the reliability of the module itself.

[0064] This solution has extremely low current requirements (total current of approximately 16A), a simplified system, and is particularly suitable for high-reliability scenarios such as aerospace and industry.

[0065] III. Second Mode (System Performance Priority) Optimization Mode Selection and Objective: Select the second mode. The core objective is to maximize the cooling effect of the thermoelectric module on the known non-uniform external heat source (chip), ensuring chip performance and reliability.

[0066] Goal-oriented thermal analysis (for external heat sources): This analysis examines the heat flux density distribution on the chip surface. The analysis clarifies the non-uniformity of the external heat source: the heat flux density in the upper core region of the chip (corresponding to thermoelectric module region A) is 150 W / cm², while in the lower region (corresponding to thermoelectric module region B) it is 50 W / cm². Region A requires approximately three times the cooling capacity of region B. This analysis serves as a direct input for the module redesign.

[0067] Closed-loop collaborative configuration scheme generation: Body reconstruction scheme (for chip high heat flux region A): See Figure 4 a) Spatial layout optimization (densification): Increase the number of thermoelectric units in area A from 30 to 45 (an increase of 50%), and reduce the spacing between units.

[0068] b) Geometry optimization (low resistance design): The cross-sectional area of ​​the thermoelectric unit is increased to 1.5 mm² and the height is reduced to 1.2 mm, reducing the unit resistance to approximately 0.053 Ω.

[0069] c) Circuit topology adjustment (parallel-dominant): Configure the 45 thermoelectric units in region A as 3 parallel branches, with 15 units in series in each branch, to provide higher total current capacity.

[0070] Peripheral structure matching scheme: Based on the expectation that the hot end of the reconstructed module will generate extremely high heat flux density in region A: a) Heat dissipation structure: Microchannel liquid cooling is used in the lower part of area A, and the flow channel density is designed to be 2.5 times that of the background area; conventional air cooling is used in area B.

[0071] b) Cold-end transducer structure: Region A uses a high thermal conductivity diamond-copper composite plate with integrated microchannels; Region B uses a conventional copper plate.

[0072] Implementation results and calculation verification: Calculations show that the total cooling current in area A increased significantly from the initial 36A to 67.5A, resulting in a net cooling capacity increase of approximately 179%.

[0073] The chip's maximum temperature was reduced from 92°C (under conventional cooling design) to 78°C, the surface temperature difference was significantly reduced from 35°C to 8°C, and the overall system efficiency (COP) was improved from 0.45 to 0.62.

[0074] This solution achieves superior chip cooling performance at the cost of higher system complexity and current requirements (total current of approximately 67.5A), making it suitable for high-performance computing, optoelectronic device cooling, and other scenarios.

[0075] IV. Comparison and Summary This embodiment clearly demonstrates that, within the same application scenario, based on different top-level objectives (prioritizing module efficiency or system cooling performance), the closed-loop collaborative design method of this invention can generate optimization schemes with drastically different physical implementations and performance characteristics. This reflects the goal-oriented and systematic nature of the method of this invention.

[0076] One application scenario of the second mode is to use the heat and power distribution generated by the heterogeneous integrated chip layout as a reference, and to resist the heat dissipation surface of the heterogeneous integrated chip and the cold end (transducer substrate) of the thermoelectric module. The transducer substrate is divided into zones according to the heat and power generated by the heterogeneous integrated chip layout. Different transducer substrate materials are selected for each zone according to the amount of heat and power generated by the heterogeneous integrated chip layout. The higher the power consumption, the higher the thermal conductivity of the material. The lower the power consumption, the lower the thermal conductivity of the material. This truly realizes the use of materials on demand and effectively solves the problem of transferring heat generated by high power consumption to low power consumption areas. Thermal isolation is generated between zones to avoid mutual thermal interference.

[0077] On the other plane of the cold end (transducer substrate) of the thermoelectric module, there is a first electrical circuit layer for connecting to the P / N thermoelectric unit. The thickness and cross-sectional shape of the electrical circuit layer are determined by simulation analysis based on the intensity matching of the heat dissipation partition of the heterogeneous integrated chip, the power of the P / N thermoelectric unit, and the heat power generated by the power of the P / N thermoelectric unit. A disordered power layout of the P / N thermoelectric unit is obtained. The cross-sectional shape of the first electrical circuit layer is 5%-10% larger than the cross-section of the P / N thermoelectric unit end face. Its thickness is based on the maximum current of the P / N thermoelectric unit during operation, and a total transmission resistance of less than 0.005 ohms is obtained. A reference thickness is calculated based on its resistance value. Then, combined with the height of the P / N thermocouples in the partition, an optimal first circuit layer thickness is obtained that is in a plane with each area of ​​the transducer substrate. This ensures that the P / N thermocouples generate extremely low Joule heat during operation, further improving the cooling efficiency. The layout of its electrical circuit is suitable for the disordered power layout of the P / N thermocouples.

[0078] The P / N thermoelectric unit features a disordered power layout. Each zone is designed based on the heat dissipation generated by the heterogeneous integrated chip layout. Thermocouple pairs with higher heat dissipation are arranged more densely, while those with relatively lower heat dissipation are arranged more sparsely. Furthermore, the cross-sectional and height dimensions of the thermocouple pairs are inconsistent between zones. The heat generated by the thermocouple pairs in each zone is simulated, and an optimal disordered arrangement is evaluated based on the simulation results. Through the disordered arrangement and the inconsistent cross-sectional and height dimensions of the thermocouple pairs, hot spots are eliminated at the hot ends of the thermocouple pairs, achieving temperature uniformity. This allows heat to be evenly transferred to the outside through the second electrical circuit layer on the other side of the P / N thermoelectric unit for exchange.

[0079] The second electrical circuit layer on the other side of the P / N thermoelectric unit is arranged in a disordered layout with the power of the P / N thermoelectric unit. The thickness of the second electrical circuit layer is consistent with that of each zone, and its cross-sectional shape is different. The cross-sectional area is 15%-22% larger than that of the corresponding first electrical circuit layer, thereby further reducing the thermal resistance of heat conduction generated when the P / N thermoelectric unit is working. Since the P / N thermoelectric unit absorbs and releases heat on one side while working, a certain temperature difference will be generated between the two. If the cross-section of the first electrical circuit layer and the second electrical circuit layer is the same, they will interfere with each other through radiation and cause cold leakage, resulting in a reduction in the working efficiency of the P / N thermoelectric unit.

[0080] Example 2: Structure and fabrication of a non-uniformly configured metal-based integrated thermoelectric module.

[0081] This embodiment fully demonstrates an integrated thermoelectric module structure and preparation method designed according to the present invention, targeting an external heat source with high heat density in the middle and low heat density on both sides.

[0082] like Figure 5As shown, this integrated thermoelectric module employs a closed-loop collaborative design with a non-uniform configuration scheme. The module's main structure includes a thermally conductive metal substrate 101. An insulating thermally conductive layer 102 is fabricated on one main surface of the thermally conductive metal substrate 101. The insulating thermally conductive layer 102 is a composite layer formed by vacuum thermosetting after slit coating or spraying of a mixed slurry containing polyimide and modified boron nitride. Multiple hot-end electrical connection portions 1032a, 1032b, and 1032c are disposed on the insulating thermally conductive layer 102. Multiple different P-type and N-type thermoelectric units 104a, 104b, and 104c are also present. 104b corresponds to the high external heat density of the middle zone (i.e., the central high-power area), exhibiting a higher thermoelectric unit distribution density, a larger thermoelectric unit cross-section, and a appropriately reduced thermoelectric unit height, and employing a parallel circuit topology. 104a and 104c correspond to the lower heat density of the two sides (i.e., the relatively low-power area), with even lower density and corresponding circuit topologies. These thermoelectric units are welded at one end to the corresponding hot-end electrical connection part by eutectic welding process, and at the other end to the corresponding cold-end electrical connection parts 1031a, 1031b, 1031c by eutectic welding process, thereby forming an integral thermoelectric unit array.

[0083] For the thermoelectric unit 104b with an appropriately reduced height, a thicker cold-end electrical connection portion 1031b is designed to match it, ensuring that the contact surface between 1031b and the external heat source is on the same plane as that of 1031a and 1031c. The thickness of the electrical connection portion can be optimized based on the electrical performance of the circuit it belongs to, in order to achieve extremely low parasitic resistance, thereby significantly reducing Joule heat loss during operation. The determination of its thickness requires comprehensive consideration of the conductivity of the connection portion material, the required current to be carried, the height of the thermoelectric unit connected to it, and the partitioning plane requirements of the cold-end transducer substrate, and is matched and optimized through electro-thermal coupling analysis. This differentiated combination of 'connection portion thickness - thermoelectric unit height' provides additional and effective design freedom for precisely controlling local thermal resistance in the module thickness direction (Z-axis), optimizing heat transfer paths, and adapting to the thermal interface of complex three-dimensional packaging.

[0084] The electrical connection at the cold end and the electrical connection at the hot end have asymmetrical geometric features: the electrical connection at the hot end has a larger cross-sectional area to reduce thermal resistance; the electrical connection at the cold end has a smaller cross-sectional area to suppress radiative cooling leakage.

[0085] The planar shape and arrangement of the hot-end electrical connection portion are configured to match the projection area of ​​the heat dissipation unit on the thermally conductive metal substrate, so as to form the shortest vertical heat conduction path from the thermoelectric unit to the heat dissipation unit. The contact area between the projection area and the corresponding electrical connection portion is not less than 85% of the area of ​​the electrical connection portion.

[0086] like Figure 6As shown, the matching design of the heat dissipation structure of the hot end completed according to the reconstruction scheme of the thermoelectric module is as follows: a denser heat dissipation fin (heat dissipation unit 302) is designed for the central high power consumption area, and corresponding heat dissipation structures (heat dissipation unit 301, heat dissipation unit 303) are designed for the two sides with relatively low power consumption areas.

[0087] At the cold end of the module, transducer plates are arranged in zones. The second transducer plate 4012, made of a high thermal conductivity material (such as diamond copper), is used for the central core area; while conventional copper transducer plates (first transducer plate 4011 and third transducer plate 4013) are used for the peripheral areas. The three types of transducer plates are precisely integrated into the same cold end base.

[0088] This embodiment illustrates that, to achieve precise suppression of external heat source 'hot spots,' the spatial layout of the thermoelectric module can be refined into a 'power density-adapted non-uniform layout.' Here, 'non-uniform' means that the distribution density, geometric dimensions, and other parameters of the thermoelectric unit pairs break the constraints of any regular array, and are mapped one-to-one, highly customized entirely based on the specific power density (heat flux density) of each point on the surface of the external heat source. The goal is to achieve an optimal match between the module's cooling capacity distribution in space and the heat distribution of the object being cooled.

[0089] A direct physical consequence of this deeply adapted layout is that the heat generation and temperature distribution at the hot end of the thermoelectric module itself will also exhibit a corresponding non-uniform state—in the region corresponding to the external hot spot, the heat flux density at the hot end is significantly higher. This is precisely the manifestation of the module deploying stronger cooling power in specific areas.

[0090] Therefore, in the closed-loop design framework described in this invention, this 'non-uniform layout' scheme directly determines and outputs a non-uniform expected heat flow distribution at the hot end. This distribution then serves as a mandatory input, driving step S302 to simultaneously design a non-uniform heat dissipation structure (such as densifying heat dissipation fins or flow channels in the high-temperature region) that precisely matches its heat dissipation capacity distribution. Through this closed-loop data-driven and collaborative design—from non-uniform cooling layout to non-uniform hot end heat flow, and then to non-uniform heat dissipation capacity—the ultimate goal of suppressing external heat sources is ensured to be achieved in the most efficient way.

[0091] The fabrication process of the non-uniformly configured integrated thermoelectric module structure is as follows: 1. Provide a thermally conductive metal substrate 101.

[0092] 2. Preparation of insulating and thermally conductive layer 102: Using a slot coater, polyimide and an effective amount of surface-modified boron nitride mixed slurry are uniformly coated, and then cured in a vacuum environment through a stepped temperature program to form a dense and highly thermally conductive insulating and thermally conductive layer 102. This composite layer has both excellent electrical insulation and anisotropic thermal conductivity.

[0093] 3. Fabrication of Electrical Connections: Patterned cold-end electrical connections 1031a, 1031b, and 1031c, and hot-end electrical connections 1032a, 1032b, and 1032c are fabricated on the insulating and thermally conductive layer 102. The electrical connections located at the hot end of the module employ a larger cross-sectional area design to reduce thermal resistance; the electrical connections located at the cold end of the module employ a smaller cross-sectional area and optimized shape design to reduce surface area, thereby suppressing radiative heat leakage. The planar shape and arrangement of the electrical connections are configured to match the projection area of ​​the main heat dissipation functional units on the thermally conductive metal substrate to form the shortest vertical heat conduction path from the thermoelectric unit to the heat dissipation functional unit.

[0094] 4. Cleaning treatment: Perform plasma cleaning on the welding areas of electrical connections to thoroughly remove oxides and contaminants and ensure welding quality.

[0095] 5. Eutectic welding: In a vacuum environment, based on the determined non-uniform configuration scheme, thermoelectric units 104a, 104b, and 104c are precisely welded to the corresponding electrical connection parts through eutectic welding process to form a large-area, single, and continuous thermoelectric unit array.

[0096] 6. Install the heat dissipation device and cold-end transducer plate determined based on the closed-loop collaborative design method. Example 3: Optimized conventional integrated thermoelectric module.

[0097] like Figure 7 As shown, for an external heat source with high heat density in the middle and low heat density on both sides, an optimized traditional integrated thermoelectric module is designed and manufactured using a non-uniform configuration scheme based on the closed-loop collaborative design method described in claims 1 to 8. Its structure includes a lower ceramic substrate 202, an upper ceramic substrate 201, and P and N thermoelectric units 203 welded together via upper and lower electrical connection parts 205. Ceramic substrates 201 and 202 are connected to the electrical connection parts via an insulating and thermally conductive layer 206. The upper ceramic substrate 201 is connected to the outer heat dissipation unit via a thermally conductive grease layer 207. The heat dissipation unit adopts a non-uniform heat sink structure: a denser heat sink fin design (heat dissipation unit 204b) is used for the central high-power area, and corresponding cooling structures (heat dissipation unit 204a, heat dissipation unit 204c) are designed for the relatively low-power areas on both sides. The projection area of ​​the heat dissipation unit on the thermally conductive metal substrate and the contact surface of the corresponding hot-end electrical connection part are not less than 85% of the area of ​​the hot-end electrical connection part, so that the heat generated by the thermoelectric unit during operation can reach the heat dissipation unit with the shortest path and the lowest thermal resistance, effectively reducing heat accumulation.

[0098] This design significantly improves the cooling effect of traditional ceramic-based integrated thermoelectric modules, demonstrating the highest value of the systematic design of this invention.

[0099] In the description of this application, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0100] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A closed loop co-design method for integrated thermoelectric modules, characterized in that, Includes the following steps: S1. Determine the system thermal management objective: Select either the first mode that optimizes the temperature uniformity of the thermoelectric module's own hot end face, or the second mode that optimizes the cooling effect on an external heat source. S2. Target-oriented thermal analysis: Analyze the heat flux density distribution or expected temperature field corresponding to the selected model to obtain thermal field distribution information; S3. Generate a closed-loop collaborative configuration scheme: Based on the thermal field distribution information obtained in step S2, a collaborative configuration scheme for the integrated thermoelectric module is generated synchronously; the collaborative configuration scheme includes the following two aspects, and the two aspects are collaboratively determined based on the thermal field distribution information to form a closed-loop design result: S301. An adaptive reconfiguration scheme for the body of the integrated thermoelectric module, used to perform differential configuration for different characteristic regions in the thermal field distribution information; S302. An external structural scheme that matches the adaptive reconfiguration scheme in terms of thermal and / or electrical performance; wherein the external structural scheme is determined based on the expected thermal state of the hot and cold ends of the thermoelectric module after the implementation of the adaptive reconfiguration scheme.

2. The method according to claim 1, characterized in that, In step S301, the adaptive reconfiguration scheme for the integrated thermoelectric module body is achieved by coordinating the adjustment of the following parameters: (i) The circuit topology of the thermoelectric units in the body, wherein the adjustment of the topology involves selection and conversion between parallel-dominant and series-dominant types; and (ii) At least one of the following types of parameters: (a) Spatial layout parameters of thermoelectric units in the body, including distribution density and / or planar geometric arrangement pattern; (b) Geometric dimensional parameters of the thermoelectric units in the body, including cross-sectional area and / or height; The adjustment of the circuit topology and the adjustment of the at least one type of parameter are determined in a coordinated manner based on the thermal field distribution information and the selected mode.

3. The method according to claim 1, characterized in that, In step S301, the adaptive reconfiguration scheme is achieved by coordinating the adjustment of the following parameters: (i) The circuit topology of the thermoelectric units in the body, wherein the adjustment of the topology involves selection and conversion between parallel-dominant and series-dominant types; and (ii) Select material composition parameters with different thermoelectric properties; The adjustment of the circuit topology and the selection of material composition parameters are determined in a coordinated manner based on the thermal field distribution information and the selected mode.

4. The method according to claim 2 or 3, characterized in that, When the system is operating in the first mode, step S301 includes: For regions where the thermoelectric module is expected to have high temperatures, its circuit topology is adjusted to a series-dominated type, and at least one of the following is performed in conjunction: (1) Configure sparser thermoelectric units; (2) Adjust the geometry of the thermoelectric unit, including reducing the cross-sectional area and / or increasing the height.

5. The method according to claim 2 or 3, characterized in that, When the system is operating in the second mode, in step S301, for the region of high heat flux density of the external heat source, the circuit topology is adjusted to a parallel-dominated type, and at least one of the following is executed in a coordinated manner: (1) Increase the distribution density of thermoelectric units; (2) Adjust the geometry of the thermoelectric unit, including increasing the cross-sectional area and / or reducing the height.

6. (3) Select thermoelectric units made of materials with lower resistivity.

7. The method according to claim 1, characterized in that, The matching peripheral structure scheme in step S302 refers to: determining the design parameters of the peripheral structure based on the expected thermal state changes of the hot and cold ends of the thermoelectric module caused by the adaptive reconfiguration scheme obtained after step S301.

8. The method according to claim 6, characterized in that, The peripheral structure scheme is specifically embodied in at least one of the following: The heat dissipation structure is designed to have a non-uniform heat dissipation capacity distribution to match the expected heat flux density distribution at the hot end. The cold-end transducer structure is designed as a partitioned structure, wherein the transducer components of at least two partitions differ in at least one of the following aspects: material thermal conductivity, structural thickness, surface treatment process, and coupled secondary thermal management method.

9. A thermoelectric module for implementing the closed-loop collaborative design method as described in any one of claims 1 to 7, characterized in that, include: A thermally conductive metal substrate, wherein the thermally conductive metal substrate constitutes part of a heat dissipation structure; An insulating and thermally conductive layer is formed on at least one main surface of the thermally conductive metal substrate; Multiple hot-end electrical connection portions are disposed on the insulating and heat-conducting layer; Multiple cold-end electrical connections; Multiple P-type and N-type thermoelectric units are provided, with each thermoelectric unit having its two ends directly welded to a hot-end electrical connection and a cold-end electrical connection via a eutectic bonding process. This allows the multiple thermoelectric units to be electrically connected via the hot-end and cold-end electrical connection, forming a thermoelectric functional array.

10. The thermoelectric module according to claim 8, characterized in that, The insulating and thermally conductive layer is a composite layer formed by vacuum thermosetting after applying or spraying a mixed slurry containing polyimide and modified boron nitride through slits.

11. An integrated thermoelectric module, characterized in that, include: The thermoelectric module as described in claim 8 or 9; A heat dissipation structure coupled to the thermal side of the thermoelectric module; A transducer structure coupled to the cold side of the thermoelectric module; The thermoelectric units in the thermoelectric module have a non-uniform configuration, and the non-uniform configuration includes inter-regional differences of at least one of the following characteristics: (1) Spatial distribution density of thermoelectric units; (2) Circuit topology connection method of thermoelectric unit; (3) Geometric dimensions of the thermoelectric unit; Furthermore, the heat dissipation capacity distribution of the heat dissipation structure matches the heat flux density distribution of the hot end face of the thermoelectric module, and the heat transfer capacity distribution of the transducer structure matches the heat flux density distribution of the external heat load.

12. The integrated thermoelectric module according to claim 10, characterized in that, The non-uniform configuration of the thermoelectric units and the matching design of the heat dissipation structure and the transducer structure are determined by the closed-loop collaborative design method according to any one of claims 1 to 7.

13. The integrated thermoelectric module according to claim 10 or 11, characterized in that: The integrated thermoelectric module also includes an adapter structure for compensating for height differences between different thermoelectric units, the adapter structure being implemented in at least one of the following ways: (a) Ensure that at least two electrical connections have different thicknesses; (b) A non-uniform interface is formed on the surface of the thermally conductive metal substrate that contacts the thermoelectric unit; (c) To form a non-uniform interface on the surface of the transducer structure that contacts the thermoelectric unit.

14. The integrated thermoelectric module according to claim 10 or 11, characterized in that, The electrical connection has asymmetrical geometric features at the hot and cold ends: the electrical connection at the hot end has a larger cross-sectional area, and the electrical connection at the cold end has a smaller cross-sectional area and / or a surface shape treated with low radiation.

15. The integrated thermoelectric module according to claim 10 or 11, characterized in that, The planar shape and arrangement of the hot-end electrical connection portion are configured to match the projection area of ​​the heat dissipation unit on the thermally conductive metal substrate, so as to form the shortest vertical heat conduction path from the thermoelectric unit to the heat dissipation unit.

16. A method for manufacturing an integrated thermoelectric module as described in any one of claims 10-14, characterized in that, Includes the following steps: A thermally conductive metal substrate is provided, wherein the thermally conductive metal substrate constitutes part of a heat dissipation structure; An insulating and thermally conductive layer is prepared on at least one main surface of the substrate; Multiple electrically isolated thermally conductive connections are provided on the insulating thermally conductive layer. Provides multiple cold-end electrical connections; Clean the welding area of ​​the electrical connection; In a vacuum or inert gas protected environment, based on a determined non-uniform configuration scheme, P-type and N-type thermoelectric units are directly welded to the corresponding electrical connection parts by eutectic welding process; Install matching heat dissipation devices and cold-end transducer plates.

17. An integrated thermoelectric module, characterized in that, include: A thermoelectric module includes an upper ceramic substrate, a lower ceramic substrate, and multiple P-type and N-type thermoelectric units sandwiched in between. as well as A heat dissipation structure that is thermally coupled to the thermoelectric module; The thermoelectric units in the thermoelectric module have a non-uniform configuration, and the non-uniform configuration includes inter-regional differences of at least one of the following characteristics: (1) Spatial distribution density of thermoelectric units; (2) Circuit topology connection method of thermoelectric unit; (3) Geometric dimensions of the thermoelectric unit; Furthermore, the heat dissipation capacity distribution of the heat dissipation structure matches the heat flux density distribution of the hot end face of the thermoelectric module.

18. The integrated thermoelectric module according to claim 16, characterized in that, The non-uniform configuration of the thermoelectric units and the matching design of the heat dissipation structure are determined by the closed-loop collaborative design method of the integrated thermoelectric module according to any one of claims 1 to 7.

Images