Bioinspired thermally conducting packaging for heat management of high performance electronic chips

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Communications Engineering volume 4, Article number: 1 (2025 ) Cite this article exchanger

Conventional electronic chip packaging generates a huge thermal resistance due to the low thermal conductivity of the packaging materials that separate chip dies and coolant. Here we propose and fabricate a closed high-conducting heat chip package based on passive phase change, using silicon carbide which is physically and structurally compatible with chip die materials. Our “chip on vapor chamber” (CoVC) concept realizes rapid diffusion of hot spots, and eliminates the high energy consumption of refrigeration ordinarily required for heat management. Multi-scale wicks and bionic vein structures are applied to CoVC leading to an increase of 164% in heat transfer performance. The thermal resistance of the package was only a third that of traditional packaging systems. This means that the structure of CoVC has a good thermal conducting ability and can reduce energy consumption for heat dissipation.

Heat extraction is fundamentally limited by the thermal resistance between the chip hot spot and packaging1. Conventional chip packages limit the heat extraction between the die hot spots and the coolant, leading to the need for a large heat sink to cool the chip and hinders integration2.

Substantial research efforts have focused on developing novel chip packages with fast thermal paths3,4,5. Bringing the coolant in direct contact with the device may be a way to overcome this limiting factor6, for example, by immersing the chips into the coolant7, by impinging coolant on a bare die8, or by etching micrometer-sized channels directly inside the chip to turn the substrate into a heat sink9. Impinge cooling and etching microchannels are active cooling packages that require additional pumping power10. The design complexity and stringent space requirement for the active cooling packages result in the implementation difficulty for those compact chip packages11. Immersion cooling packages the die hot spots with the coolant, where the hot spots are in direct contact with the coolant9, which is a passive and energy-saving cooling package that can use convection and evaporation of the coolant to cool down the hot spots of the chip12. The immersion cooling which has higher energy efficiency, higher power density and more water saving, would become the mainstream mode for future high-power chips13.

Third-generation semiconductors, silicon carbide (SiC) and gallium nitride (GaN), present wide bandgap and great thermal conductivity14. SiC power devices with high frequency, high voltage, and high thermal conductivity considerably reduce the conduction and switch loss of SiC power modules to endure even higher operational voltages15. The power devices (e.g., SBD (Schottky Barrier Diode), MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor), IGBT (Insulated Gate Bipolar Transistor), etc.) are core components in the fields of the high-speed rail, the electric vehicles, and the national defense16,17. With the advantaged stacking package technology development, the power device microsystem package becomes further compact to endorse the miniaturization of high computing performance, communication technology, and other emerging electronic information systems18,19. The active cooling packages make it difficult to guarantee the junction temperature uniformity for the device array20. In contrast, the coolants inside the passive cooling packages are solitarily driven by the temperature gradient and the capillary force. The typical passive embedded cooling packages contain the vapor chamber (VC, a passive phase change heat transfer device)21,22,23. Associated with the wick structures, the evaporation/condensation circulation in the VC can rapidly remove the heat flux from hotspots and achieve high- temperature uniformity on the hotspot surfaces24,25. Without any external auxiliary power, the passive embedded phase change cooling architecture can be compatibly fabricated in compact packages of power devices.

Ceramic materials, known for their superior mechanical properties and high thermal conductivity, are extensively utilized as packaging and heat dissipation materials for high-power devices like IGBTs26. The high thermal conductivity of ceramic materials facilitates the rapid transfer of heat generated during the operation of IGBTs, while their coefficient of thermal expansion compatibility with IGBT chips helps to reduce thermal stress, thereby enhancing the reliability of the packaging.

In view of this, we propose a high-conducting thermal package called by the chip on vapor chamber structure (CoVC). Specifically, a VC is embedded in the substrate of the SiC power device as a high thermal conductivity substrate with the phase change heat transfer. To further improve its heat transfer performance, a multi-scale porous structure is prepared by using in situ reactive sintering and laser ablation. The preparation process, surface morphology, and thermal performance of the passive embedded cooling package were analyzed in detail.

The chip high-conducting heat packaging structure CoVC realizes the rapid heat transfer through the evaporation-condensation phase change of the coolant, and its heat transfer capacity is dozens or even hundreds of times that of traditional packaging materials22. Figure 1 shows the heat dissipation diagram of the CoVC chip package designed in this work. The silicon carbide vapor chamber as the substrate is directly welded with the IGBT power chip to decrease the interface thermal resistance27. The heat generated by the chip is directly transferred to the evaporator of the SiC VC (SCVC) by heat conduction, and the heat drives the evaporation of the coolant inside the chamber. The gaseous coolant diffuses to the condenser to release the latent heat and condenses into liquid. The capillary force provided by the wick drives the liquid back to the evaporator to achieve the circulation of phase change heat transfer28. The circulation can rapidly remove the heat flux from the hotspots and keep the high-temperature uniformity29.

a Schematic diagram of heat transfer of the CoVC: heat generated by the chips is conducted to the SiC vapor chamber(SCVC), which triggers the evaporation of the coolant and brings the heat to the outer wall of the vapor chamber(VC). b Top view of the CoVC: the chips were bonded directly on the SCVC through the copper metallization. c The multi-scale composite wick with topology groove. The wick can provide sufficient capillary pressure to transport the condensed coolant back to the evaporator. d The integrated model of the chip on the top plate of the SCVC. e The details of the multi-scale composite wick. The wick contains small particles, large particles, and channels. The channels can provide high permeability for the working liquid, which is beneficial for the mass and heat transfer of the SCVC.

The CoVC includes the IGBT chips, the top plate, the inner surface SiC wick of the top plate, the porous SiC column, the inner surface SiC wick of the bottom plate, the bottom plate and the charging tube from the top to the bottom respectively, as shown in Fig. 2. The substrate material is SiC. It has a thermal conductivity of up to 490 W m−1 °C−130, which is a very ideal material for the heat transfer of the semiconductor chips. SiC has low density, which is only about one-third of copper, and a low coefficient of thermal expansion, which is a few tenths of the coefficient of thermal expansion of metals31. The SiC substrates are ground precisely using diamond particles. The wick is fabricated using SiC powder to obtain a higher interfacial compatibility and thermal conductivity. This work focuses on thermal management for semiconductor chips, so deionized water is selected as the working fluid based on the operating temperature of electronic chips.

Insulated Gate Bipolar Transistor(IGBT).

Figure 3 illustrates the details and thermal evaluation of the CoVC. The SCVC worked as the substrate of the chip package. Two kinds of SCVC with different wick structures were fabricated. The single-scale SCVC (SSCVC) has a conventional particle stacking wick. The multi-scale SCVC (MSCVC) has laser ablated wick (Fig. 3c) with microscale and nanoscale pores (Fig. 3d). It can be found that the temperature of the SSCVC is always higher than that of the MSCVC under the same heat load. For example, when the heating power is 80 W, the hotspot temperature of the MSCVC is only 90 °C, while the temperature of the SSCVC is 140 °C. The safe working temperature of the silicon chip is 85 °C, the heat dissipation power of the MSCVC is 100 W, and the heat dissipation power of the SSCVC is 60 W. The heat dissipation power of the MSCVC is 1.67 times that of the SSCVC. As the heat load increases, the temperature of the SSCVC becomes more and more difficult to reach the equilibrium, and a wide range of temperature fluctuation occurs, which indicates that the wick at the evaporator partially dries out32. As the heat load becomes higher, the evaporation rate of the coolant increases33, while the reflux rate of the condensed working liquid is limited by the capillary pressure provided by the wick. When the evaporation rate of the working liquid at the evaporator is greater than the reflux rate of the condensed working liquid, the local dryout will appear at the evaporator34. The junction temperature will continue to rise, and it is difficult to reach a stable state. Temperature distribution on the condenser of SiC vapor chamber is tested, as shown in Supplementary Note 1.

a Photograph of the CoVC: the Insulated Gate Bipolar Transistor(IGBT) chips directly bonded on the SiC vapor chamber(SCVC). b Photograph of the multi-scale composite wick and wick columns. The wick and the columns were sintered on the SiC plate. c The SEM image of the multi-scale composite wick. A continuous CO2 laser was used to fabricate grooves on the surface of the wick. In addition, after laser ablation, the nanoparticles are deposited on the large SiC particles to form nanoporous layers. d The SEM image of the nanoporous structure. The nanoscale pores have strong capillary pressure for coolant according to Darcy’s law. e The start-up characteristics of the CoVC. f The operating temperature curve of CoVC. g The thermal resistance of the CoVC. h The deviation of thermal performance of the CoVC under gravity-assisted and anti-gravity conditions. multi-scale SiC vapor chamber(MSCVC), single-scale SiC vapor chamber(SSCVC), heat transfer deterioration(HTD).

The comparison of single-scale and multi-scale shows that the MSCVC has a faster temperature response and faster start-up, unlike the SSCVC which takes a longer time to reach the equilibrium state. The faster the temperature response and faster start-up it has, the faster it can dissipate heat flux caused by the sudden increase in power of chips, which will ensure the chip’s stability and reliability. In addition, the operating temperature of the MSCVC is smooth under a specific heat load, while the temperature fluctuation of the SSCVC is more obvious, and the higher the heat load, the greater the fluctuation. The temperature of the evaporator of the SCVC increased with the increase of power, and the temperature changed in a stepped manner (Fig. 3f).

When the heat load is 30 W, the thermal resistance of the SSCVC drops to the lowest, 0.36 °C W−1. Then, the thermal resistance of the SSCVC increases sharply, and at the heat load of 70 W, the thermal resistance is 0.59 °C W−1. The reason is that the liquid film in the wick at the evaporator keeps thinning, and when the heat load exceeds 30 W, liquid film at the center of the evaporator gradually recessed, and the phenomenon of heat transfer deterioration (HTD) appears35. The heat load continues to increase, the dry-out area further expands, and the HTD becomes more serious, so the thermal resistance increases rapidly (Fig. 3g). The thermal resistance of the MSCVC did not show such a rapid trend of change36. Due to its ultra-high capillary performance, the multiscale of laser-ablated wick can quickly transport the liquid mass to the evaporator through the laser-ablated grooves and multiscale pores. The effect of increasing heat load on the thickness of the liquid film of the wick at the evaporator is less strong than that on the SSCVC. Therefore, the thermal resistance of the MSCVC slowly decreases until it reaches a minimum value of 0.33 °C W−1 at a thermal load of 140 W. The effect of working liquid filling ratio on the performance of SCVCs is studied as shown in Supplementary Note 2. Supplementary Fig. 4 is the Experimental platform of Fig. 3h. Also, the thermal resistance of VC is measured through the thermocouple placed in the top plate and bottom plate, as shown in Supplementary Note 3.

Based on common plant leaf veins, we designed four different evaporator wicks as shown in Fig. 4. Pattern A is a Y-shaped plant vein bionic structure. Pattern B is a radial wick structure. Pattern C is an H-shaped plant leaf vein bionic structure. Pattern D is a double-leaf bionic structure in which Y-shaped veins and H-shaped veins are combined. In this paper, CoVCs with four different wick designs were prepared. In order to control the variables, the wicks at the condenser of the four vapor chambers are all bidirectional laser ablation multiscale wicks. Taking the wick designs as the only variable, the heat transfer performance of the silicon carbide vapor chamber was investigated.

a Different plant veins: Y-shaped ginkgo vein, radial palm vein, and H-shaped banana vein. b Bionic wick structures of the CoVC with different vein patterns: pattern A (P-A) with Y-shaped ginkgo vein, pattern B (P-B) with radial palm vein, pattern C (P-C) with H-shaped ginkgo vein, and pattern D (P-D) with Y-shaped ginkgo vein H-shaped ginkgo vein. c The chip temperature of the CoVC. d The thermal resistance of the CoVC. e The laser ablation surface ratio and minimum thermal resistance of the CoVC. The CoVC with P-B bionic wick structure had the lowest laser ablation ratio and minimum thermal resistance.

Figure 4c shows the temperature of four kinds of biomimetic CoVCs. In order to more intuitively compare the effect of the bionic structure wick on the heat transfer performance of the vapor chamber, the bidirectional cross groove (CG)(Fig. 3c) wick vapor chamber is added as a comparison in the figure. It can be seen that the temperature at the evaporator of the vapor chamber increases with the increase of the heat load, and the structure of the bionic wick has a significant effect on the chip junction temperature of the CoVC (Fig. 4c). Under the same heat load, the P-B vapor chamber with radial wick structure has the lowest junction temperature. When the heat load is lower than 100 W, the chip junction temperature of the bidirectional CG is the highest. Although the fluid diffusion speed in the two-dimensional plane is enhanced, in the unidirectional transport of the fluid, the cross groove will shunt the fluid, resulting in a single reduced transport property to liquid. For the bionic wicks of the other four vapor chambers, the arterial structures of the laser-ablated grooves converge at the center of the evaporator, and the one-way fluid transport speed is faster than that of the CG wick, so the junction temperature is lower than that of the CG CoVC. For the four vapor chambers with bionic channels and wicks, the junction temperature of the CoVC with a double-leaf biomimetic structure is the highest. When the heat load is 120 W, the junction temperature of VCP-D is 135.13 °C, and the evaporator temperatures of P-A, P-B, and P-C were 105.62 °C, 105.14 °C, and 111.80 °C, respectively.

Figure 4d shows the thermal resistance of the biomimetic CoVCs and the bidirectional cross groove wick vapor chamber. Except for P-D, the thermal resistance of all biomimetic CoVCs is much lower than that of the bidirectional cross groove wick vapor chamber. The thermal resistance of all five vapor chambers exhibits a trend of first decreasing and then increasing with increasing heat load. The thermal resistance of P-A reaches its minimum value of 0.23 °C W−1 at a heat load of 110 W, undergoing a transition where the decreasing trend shifts to an increasing trend. P-B achieves the minimum thermal resistance of 0.20 °C W−1 at a heat load of 100 W, which is also the lowest thermal resistance among the five vapor chambers. Similarly, P-C reaches the minimum thermal resistance of 0.24 °C W−1 at a heat load of 100 W. P-D experiences a thermal resistance inflection point at a heat load of 80 W, where the heat transfer performance begins to deteriorate, and the thermal resistance starts to increase, with its minimum thermal resistance being 0.35 °C W−1. Among the biomimetic CoVCs, P-D exhibits the earliest occurrence of HTD. This is attributed to the working fluid refluxing through porous silicon carbide columns internally, which, in the bionic structure, needs to pass through multiple curved vein-like channels before reaching the center of the evaporator. The reflux path for the working fluid is longer, and the bends introduce additional local resistance, reducing the flow velocity of the liquid. Therefore, the thermal resistance of P-D is higher than that of other biomimetic CoVCs.

Compared to the bionic vapor chamber, the bidirectional cross groove wick vapor chamber exhibits significantly higher thermal resistance. Although the intersecting ablation channels on the wick structure aid in the two-dimensional diffusion of the liquid, the numerous 90° bends along the reflux path, when the fluid returns towards the center of the wick, create local resistance, affecting the reflux velocity. Due to the substantial differences in shape between the bionic structure and bidirectional cross groove, this work adopts the laser ablation surface ratio as a comparative parameter for the wick structure of the five vapor chambers. The laser ablation surface ratio is defined as the ratio of the scanned area of the laser beam spot to the surface area of the wick.

Figure 4e displays the laser ablation surface ratio and minimum thermal resistance for the five vapor chambers. P-B exhibits the lowest thermal resistance and the smallest laser ablation surface ratio, indicating that P-B achieved the best heat transfer performance with only 16.59% of the laser ablation surface ratio. CG has the largest laser ablation surface ratio, reaching 51%, and its heat transfer performance is higher than that of P-D but lower than that of P-A and P-B. A larger laser ablation surface ratio implies higher laser processing energy consumption. In conclusion, the radial vein and Y-shaped bionic veins can significantly improve the heat transfer performance of the vapor chambers, reducing thermal resistance by 39% and 30%, respectively, while also lowering the energy consumption of laser processing.

Traditional solid heat sink (THS) requires the introduction of direct bonded copper (DBC) for chip encapsulation because of electrical conductivity and high coefficient of thermal expansion37. This structure increases the thermal resistance between the hot spot of the chip and the coolant, and the heat transfer efficiency is low, which makes the chip temperature remain high. Figure 5a, b show the COMSOL simulated temperature contours of the SCVC and THS at a heat flow density of 100 W cm−2. The maximum temperature of the chip with SCVC was 109 °C, and the maximum temperature of the chip with THS was 159 °C. Figure 5c, d show the schematic diagrams of the heat sink package structure with SCVC and THS. The THS package structure requires the introduction of additional DBCs, resulting in the thermal resistance of the packaged IGBT module being much higher than that of the heat sink package structure with SCVC. Figure 5e, f show a comparison of chip temperature and thermal resistance for the two package structures. The thermal resistance of the THS is constant, while the thermal resistance of the SCVC decreases with the increase of heat flux. When the heat flux is 160 W cm−2, the thermal resistance of the SCVC is 0.34 °C W−1, which is only one-third of the thermal resistance of the THS.

a The thermal simulation of the SCVC. b The thermal simulation of the chips with traditional solid heat sink (THS). c The schematic diagrams of the heat sink package structure with SCVC. d The schematic diagrams of the heat sink package structure with THS. e The simulated chip temperature of the SCVC and THS. f The simulated thermal resistance of the SCVC and THS.

In this work, a chip packaging structure called CoVC is proposed, and SCVC with different wick structures were fabricated. The thermal performance of SCVC with single-scale wick and multi-scale wick is investigated, and based on common plant leaf veins, bionic structure wicks are proposed and also investigated. What’s more, the thermal simulation of SCVC and THS is run to make a comparison of their thermal performance. The main conclusions are as follows:

The MSCVC has better thermal performance than SSCVC. When the heat load is 30 W, the thermal resistance of the SSCVC drops to the lowest, 0.36 °C W−1. Then, the thermal resistance of the SSCVC increases sharply. As for MSCVC, its thermal resistance slowly decreases until it reaches 0.33 °C W−1 at the heat load of 140 W.

During the heating process, CoVC with radial structure wick shows the lowest thermal resistance among five vapor chambers, and the junction temperature of it is the lowest at the heat load of 120 W, 105.14 °C.

In the thermal simulation of SCVC and THS, the maximum temperature of the chip with SCVC is 109 °C, and the maximum temperature of the chip with THS is 159 °C when the heat flow density is 100 W cm−2. When the heat flux is 160 W cm−2, the thermal resistance of SCVC is 0.34 °C W−1, only one-third of the thermal resistance of THS.

This means that with CoVC package structure, the global computing power will be multiplied without changing the current power consumption. In 2030, the global data center traffic will reach 60000 EB. The electricity consumption of global data centers will reach 1800 TWh. The proposed integrating phase-change cooling package of CoVC will be able to dramatically decrease chip thermal resistance and reduce energy consumption in data centers, saving approximately 1200 TWh of electricity.

The SCVC of the CoVC has a low coefficient of thermal expansion and good insulation38, allowing it to be packaged directly with the IGBT chip. The measurements of SCVC are shown in Table 1.

The fabrication process of SCVC is shown in Fig. 6, which mainly includes the following processes: the precision grinding and the molding of top and bottom SiC plates, the fabrication of the multi-scale composite wick, and the powder sintering column, the sealing and the leak detection of the SCVC, filling, evacuation, secondary degassing and sealing39. The SiC slurry was prepared with SiC powder with a particle size of 300 mesh, Al2O3 powder with a particle size of 8000 mesh, and a polyvinyl alcohol solution. The mass ratio of SiC powder to Al2O3 powder is 10:1. After thorough mixing, the mixed powder is added to a polyvinyl alcohol solution with a mass fraction of 4%. The mass ratio of the powder to the polyvinyl alcohol solution is 3:1. Then the SiC slurry was placed on the inner surface of the SiC plate, using a rectangular silicone mold to define the shape of the SiC slurry. After the slurry is cured and dried, the porous silicone molds are placed on the top SiC plate, and the slurry continues to be placed to fabricate a wick column that is fully connected to the wick layer, as shown in Fig. 7. Subsequently, reaction sintering is performed at a sintering temperature of 1500 °C and held for 60 min. The top and bottom SiC plates are tightly laminated together, and then the glazed SiC shell is heated using a silicate glaze applied to the joint of the two SiC plates. The glaze is held at 1150 °C for 0.5 h. The glaze will form a continuous liquid phase vitreous body to seal the SCVC at this temperature. The charging tube is inserted into the charging hole on the side of the shell and then the tube is secured by applying a silicone sealant around the perimeter. Deionized water is injected into the plate through the charging tube, and the air inside the plate is evacuated using a vacuum pump for 400 s. Finally, the pressure sensor shows a pressure of 7 × 10-4 Torr at the sealing interface40. The injection tube is clamped flatly using a clamping die to seal the tube. The laser ablation of a SiC wick is shown in Supplementary Note 4.

a Schematic diagram of the solid SiC plate. b The bottom plate of the SCVC after precision grinding. c The SiC plates of the SCVC with wick layers sintered on their inner surface. d Schematic diagram of the laser ablation of the wick layers. e Schematic diagram of assembly of the SiC plates, wick layers, and charging tube. f Schematic diagram of degassing, charging of DI water, and sealing.

a Photograph of the SiC block. b Photograph of the SiC block after precision grinding. c Photograph of the SiC plates with dried SiC slurry on their inner surface.

To precisely test the heat transfer performance of the SCVC comprehensively, a specified test platform was designed and manufactured as shown in Fig. 8. The test platform mainly consists of four parts, which are a power heating system, temperature data acquisition unit, convection cooling unit and sample fixing device. The power supply heating system includes a programmable DC power supply, a copper heating block with two cylindrical heating rods embedded in it. The heating power is controlled by the output power of the DC power supply. To embed two heating rods at the bottom, the overall structure of the copper heating block is pyramidal, with the uppermost square measuring 10 mm × 10 mm and the bottom square measuring 25 mm × 25 mm. The entire heating block is wrapped in phenolic plastic with good insulation properties to reduce heat loss and improve the accuracy of the experiment. The data acquisition unit includes an Agilent data acquisition instrument (Agilent 34970A, Keysight, USA), T-type thermocouples and a laptop computer to accurately record the temperature of the heat source and the various parts of the SCVC. The convection cooling system includes a thermostatic water bath (F25-ME, Julabo, Germany), a rotameter, and a copper liquid cooling plate. The temperature of the bath is set at 25 °C. The copper cooling plate is in close contact with the condenser of the SCVC, and the contact surface is made of thermally conductive silicone grease to reduce the contact thermal resistance.

The test platform mainly consists of four parts, which are the power heating system, temperature data acquisition unit, convection cooling unit, and sample fixing device. The right side shows the exploded view of the inside of the resin fixing block, which mainly includes the metal heat source, phenolic resin insulation block, SCVC, copper liquid cooling plate, and elastic gasket. Integrating and fixing the various components can realize the switching of anti-gravity working conditions, pro-gravity working condition, and horizontal working condition.

All the data needed to evaluate the conclusions in the paper are present in the paper and in the Supplementary Information.

Cui, Y., Qin, Z., Wu, H., Li, M. & Hu, Y. Flexible thermal interface based on self-assembled boron arsenide for high-performance thermal management. Nat. Commun. 12, 1284 (2021).

Wang, Y. et al. A high-performance solid-state electrocaloric cooling system. Science 370, 129–133 (2020).

Qin, Y. et al. Thermal management and packaging of wide and ultra‐wide bandgap power devices: a review and perspective. J. Phys. D: Appl. Phys. 56, 093001 (2023).

Wen, Y. et al. Advances on Thermally Conductive Epoxy‐Based Composites as Electronic Packaging Underfill Materials—A Review. Adv. Mater. 34, 2201023 (2022).

Otaki, D., Nonaka, H. & Yamada, N. Thermal design optimization of electronic circuit board layout with transient heating chips by using Bayesian optimization and thermal network model. Int. J. Heat. Mass Transf. 184, 122263 (2022).

Sikarwar, S., Kumar, R., Yadav, A. & Gwalwanshi, M. Battery thermal management system for the cooling of Li-Ion batteries, used in electric vehicles. Materials Today: Proceedings. (Elsevier, Amsterdam, 2023) https://doi.org/10.1016/j.matpr.2023.02.293.

Yang, B. et al. Cooling and entangling ultracold atoms in optical lattices. Science 369, 550–553 (2020).

Xu, H. et al. Electrospray characteristics and cooling performance of dielectric fluid HFE-7100. Energy 259, 125072 (2022).

Van, Erp, R., Soleimanzadeh, R., Nela, L., Kampitsis, G. & Matioli, E. Co-designing electronics with microfluidics for more sustainable cooling. Nature 585, 211–216 (2020).

Rangarajan, S., Schiffres, S. N. & Sammakia, B. A Review of Recent Developments in “On-Chip” Embedded Cooling Technologies for Heterogeneous Integrated Applications. Engineering 26, 185–197 (2023).

Kang, Y. et al. Optimizing the Passive House Planning Package simulation tool: A bottom-up dynamic approach to reduce building performance gap. Energy Build. 276, 112512 (2022).

Qian, S. et al. High-performance multimode elastocaloric cooling system. Science 380, 722–727 (2023).

Masanet, E., Shehabi, A., Lei, N., Smith, S. & Koomet, J. Recalibrating global data center energy-use estimates. Science 367, 984–986 (2020).

Parker, M. A gallium nitride HEMT that enhances. Nat. Electron. 4, 858 (2021).

Lukin, D. M. et al. 4H-silicon-carbide-on-insulator for integrated quantum and nonlinear photonics. Nat. Photonics 14, 330–334 (2020).

Sampayan, S. E. et al. Characterization of carrier behavior in photonically excited 6H silicon carbide exhibiting fast, high voltage, bulk transconductance properties. Sci. Rep. 11, 6859 (2021).

Lentzsch, T., Schwabe, C., Lutz, J. & Basler, T. Calibration methods and power cycling of double-side cooled SiC MOSFET power modules. Microelectron. Reliab. 150, 115077 (2023).

Ma, Y. et al. Investigation of Manifold Microchannel Cooler for High-Heat-Flux Electronic Cooling of 1000 W/cm2 with Varying Manifold and Microchannel Height. In Proceedings of the Eighth Asia International Symposium on Mechatronics. Singapore, 1261–1269 (Springer Nature, 2022).

Shulaker, M. M. et al. Three-dimensional integration of nanotechnologies for computing and data storage on a single chip. Nature 547, 74–78 (2017).

Cong, B. et al. A combined solution of thermoelectric coolers and microchannels for multi-chip heat dissipation with precise temperature uniformity control. Appl. Therm. Eng. 219, 119370 (2023).

Peterson, G. A., Johnson, C. L. & Jellinek, A. M. Thermal evolution of Mercury with a volcanic heat-pipe flux: Reconciling early volcanism, tectonism, and magnetism. Sci. Adv. 7, eabh2482 (2021).

Yan, C. et al. A novel ultra-thin vapor chamber with composite wick for portable electronics cooling. Appl. Therm. Eng. 226, 120340 (2023).

Kang, S. Y., Kim, S. H., Park, J. & Jang, S. P. Thermal performance of a thin flat vapor chamber with a multiscale wick fabricated by SAC305/Sn58Bi micro-sized particles. Int. J. Heat. Mass Transf. 212, 124305 (2023).

Jiang, G. et al. Laser microstructuring of extremely-thin vapor chamber with hybrid configuration for excellent heat dissipation. Energy Convers. Manag. 290, 117214 (2023).

Yuan, X. et al. Experimental study of large-area ultra-thin vapor chamber for microelectronic heat dissipation. J. Energy Storage 72, 108219 (2023).

Kim, H. T. High thermal conductivity ceramics and their composites for thermal management of integrated electronic packaging. In Heat Transfer - Models, Methods and Applications, 1–27 (Intech, 2018).

Zhang, X. et al. Ultra-thermostable embedded liquid cooling in SiC 3D packaging power modules of electric vehicles. Energy Convers. Manag. 276, 116499 (2023).

Zhou, F., Zhou, G., Zhou, J., Jia, X. & Huai, X. Performance comparative evaluation of three thin vapor chambers with different wick structures. Appl. Therm. Eng. 230, 120749 (2023).

He, Z., Yan, Y. & Zhang, Z. Thermal management and temperature uniformity enhancement of electronic devices by micro heat sinks: A review. Energy 216, 119223 (2021).

Yamaoka, S. et al. Directly modulated membrane lasers with 108 GHz bandwidth on a high-thermal-conductivity silicon carbide substrate. Nat. Photonics 15, 28–35 (2021).

Wang, R., Huang, Y., Yang, D., Pi, X. Impurities and defects in 4H silicon carbide. Appl. Phys. Lett. 122(18), 180501 (2023)

Baraya, K., Weibel, J. A. & Garimella, S. V. Wetting hysteresis as the mechanism of heat pipe post-dryout thermal hysteresis and recovery. Int. J. Heat. Mass Transf. 204, 123875 (2023).

Shoeibi, S., Saemian, M., Khiadani, M., Kargarsharifabad, H. & Mirjalily, S. A. A. Influence of PV/T waste heat on water productivity and electricity generation of solar stills using heat pipes and thermoelectric generator: An experimental study and environmental analysis. Energy Convers. Manag. 276, 116504 (2023).

Li, S. et al. Experimental study on the thermo-hydrodynamic characteristics of a nitrogen pulsating heat pipe. Int. Commun. Heat. Mass Transf. 146, 106920 (2023).

Deng, W. et al. Performance enhancement of nanoscale heat pipe with hydrophilic/hydrophobic pattern. Int. Commun. Heat. Mass Transf. 144, 106767 (2023).

Zhang, K., Lu, M., Sun, Y., Wu, Q. & Wei, W. Study of the heat transfer deterioration of supercritical CO2 in vertical pipes using a hybrid RANS/LES method. Int. J. Therm. Sci. 185, 108101 (2023).

Moon, S. et al. High‐Performance Thin‐Film VCSELs Integrated with a Copper‐Plated Heatsink. Adv. Mater. Interfaces, 10, 2300191 (2023).

Wang, P. et al. Negative thermal expansion of Sc2W3O12 interlayer with three-dimensional interpenetrating network structure for brazing C/SiC composites and GH3536. Carbon 201, 765–775 (2023).

Lee, Y. J. & Kim, S. J. Experimental investigation on the effect of transverse grooves in a channel on the thermal performance of a flat-plate pulsating heat pipe. Int. J. Heat. Mass Transf. 218, 124759 (2024).

Wang, H. et al. A hybrid vapor chamber heat sink incorporating a vapor chamber and liquid cooling channel with outstanding thermal performance and hydraulic characteristics. Energy Convers. Manag. 244, 114499 (2021).

This work was supported by Science and Technology Program of Guangzhou (No. 2019050001), the Program for Guangdong Innovative and Enterpreneurial Teams (No. 2019BT02C241), Guangdong Province Fund (No. 2023A1515012675).

Guangdong Provincial Key Laboratory of Optical Information Materials and Technology & Institute of Electronic Paper Displays, South China Academy of Advanced Optoelectronics, South China Normal University, Guangzhou, 510006, PR China

Huawei Wang, Pengfei Bai, He Cui, Xiaotong Zhang, Yifan Tang, Shaoyu Liang, Shixiao Li & Guofu Zhou

Shenzhen Guohua Optoelectronics Tech. Co., Ltd., Shenzhen, 518110, PR China

Academy of Shenzhen Guohua Optoelectronics, Shenzhen, 518110, PR China

You can also search for this author in PubMed Google Scholar

Huawei Wang: Conceptualization, Methodology, Validation, Data curation, Writing-original draft. Pengfei Bai: Conceptualization, Project administration, Funding acquisition. He Cui: Validation, Writing-review & editing. Xiaotong Zhang: Data curation, Formal analysis. Yifan Tang: Data curation, Writing-original draft. Shaoyu Liang: Data curation, Methodology. Shixiao Li: Methodology, Formal analysis. Guofu Zhou: Funding acquisition.

The authors declare no competing interests.

Communications Engineering thanks Shwin-Chung Wong and the other, anonymous, reviewers for their contribution to the peer review of this work. Primary Handling Editors: [Rosamund Daw and Saleem Denholme].

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Wang, H., Bai, P., Cui, H. et al. Bioinspired thermally conducting packaging for heat management of high performance electronic chips. Commun Eng 4, 1 (2025). https://doi.org/10.1038/s44172-024-00338-6

DOI: https://doi.org/10.1038/s44172-024-00338-6

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Communications Engineering (Commun Eng) ISSN 2731-3395 (online)

fermentation tank Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Bioinspired thermally conducting packaging for heat management of high performance electronic chips | Communications Engineering