Fiber membrane based cooling systems and methods

Abstract

Fiber membrane-based cooling systems and methods are described. An example cooling device includes a fiber membrane, in which a first side of the cooling device is in thermal communication with a heat source, a second side of the cooling device is in thermal communication with a heat sink, and the fiber membrane is positioned between the first side and the second side. The fiber membrane of the cooling device includes a plurality of fibers that form a three-dimensional network of pores, allowing a cooling medium to flow through and transfer heat from the heat source to the heat sink.

Description

FIBER MEMBRANE BASED COOLING SYSTEMS AND METHODSSTATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

[0001] This invention was made with government support under grant number CMMI1762560 awarded by the National Science Foundation (NSF). The U.S. government has certain rights in the invention.CROSS-REFERENCE TO RELATED APPLICATION(S)

[0002] This application claims priority to the provisional application with serial number 63 / 583, 139, titled “PASSIVE EVAPORATOR INCLUDING NANOFIBER MEMBRANES,” filed September 15, 2023. The entire contents of the above noted provisional application are incorporated by reference as part of the disclosure of this document.TECHNICAL FIELD

[0003] This document relates to a cooling system and method involving a fiber membrane.BACKGROUND

[0004] Various sectors have cooling needs for devices that include one or more heat generating components. These sectors encompass electronic devices, wearable and / or foldable technologies, and other devices or systems. Examples include computing devices, data centers, virtual reality (VR) headsets, mobile phones / tablets (including foldable versions), and similar heat producing equipment.SUMMARY

[0005] An aspect of the disclosed technology relates to a cooling device. The cooling device includes a fiber membrane. The first side of the cooling device is in thermal communication with a heat source. The second side of the cooling device is in thermal communication with a heat sink. The fiber membrane is positioned between the first side and the second side. The fiber membrane includes a plurality of fibers forming a three-dimensional network of pores, allowing a cooling medium to flow through andtransfer heat from the heat source to the heat sink when the cooling medium is vaporized in the fiber membrane.

[0006] Another aspect of the disclosed technology relates to a cooling method. The cooling method includes flowing a cooling medium through a fiber membrane of the cooling device from a first side toward a second side of the cooling device, vaporizing during the flowing, the cooling medium in the fiber membrane by absorbing heat from a heat source that is in thermal contact with the cooling device on the first side; and removing heat from the cooling medium to a heat sink on a second side of the cooling device that is in thermal communication with the heat sink on the second side. The fiber membrane includes a plurality of fibers forming a three-dimensional network of pores.

[0007] In some embodiments, the heat removal is achieved by condensing the vaporized cooling medium in a condenser. The cooling medium can be autonomously pumped via capillary pressure formed at the liquid-vapor interfaces inside the small pores in the fiber membranes (i.e., passive-pumping device). Alternatively, the cooling medium can be driven by a pump (i.e., active pumping). In both cases, the vapor is condensed in a condenser and the condensed liquid is returned to the fiber membrane for evaporation via capillary pumping or external pumping, thus forming a closed loop.

[0008] A further aspect of the disclosed technology relates to a wearable or foldable object comprising a cooling device as disclosed herein. A further aspect of the disclosed technology relates to an electronic device including a heat generating component and a cooling device as disclosed herein.BRIEF DESCRIPTION OF THE DRAWINGS

[0001] Figure 1 illustrates schematics of evaporation from different types of membranes and SEM images of an example fiber membrane with three-dimensional (3D) interconnected open pores.

[0002] Figure 2 illustrates a schematic representation of a system according to some embodiments of the disclosed technology.

[0003] Figures 3A and 3B illustrate schematic representations of a passive-pumping and active-pumping cooling device, respectively, according to some embodiments of the disclosed technology.

[0004] Figure 4 illustrates a flowchart of a cooling process according to some embodiments of the disclosed technology.

[0005] Figure 5 shows morphology of glass fiber (GF) membranes (A-D) with 5 different mean pore sizes: (a) 3.2 micrometers; (b) 4.8 micrometers; (c) 5.3 micrometers; (d) 6.1 micrometers; and (e) 11.4 micrometers.

[0006] Figure 6 shows (a) permeability as a function of mean pore size and (b) measured capillary pressure as a function of mean pore size of the GF membranes illustrated in Figure 5.

[0007] Figure 7 shows evaporation experiment and results of the GF membranes illustrated in Figure 5.

[0008] Figure 8 shows comparison of the CHF values between various membranes with isolated pores and fiber membranes with 3D interconnected open pores according to some embodiments of the disclosed technology.

[0009] Figure 9 shows results of long-term stability tests using fiber membranes and morphology of the fiber membranes after the long-term stability tests according to some embodiments of the disclosed technology.

[0010] Figure 10 illustrates GF membrane preparation procedures according to some embodiments of disclosed technology.

[0011] Figure 11 illustrates a setup for testing GF membrane’s maximum tensile strength according to some embodiments of the disclosed technology.

[0012] Figure 12 shows the mechanical strength of the pristine and silane-treated membranes according to some embodiments of the disclosed technology.

[0013] Figure 13 illustrates a permeability test setup according to some embodiments of the disclosed technology.

[0014] Figure 14 illustrates a setup for characterizing GF membrane’s capillary pressure according to some embodiments of the disclosed technology.

[0015] Figure 15 shows a setup for and results of characterization of a temperature coefficient of the resistance (TCR) according to some embodiments of the disclosed technology.

[0016] Figure 16 illustrates an evaporation test setup according to some embodiments of the disclosed technology.

[0017] Figure 17 illustrates thin-film profiles from kinetic modeling at 50 K superheat with 5 different pore sizes according to some embodiments of the disclosed technology.

[0018] Figure 18 shows a schematic representation of a flexible vapor chamber (passive, capillary-driven evaporator device) according to some embodiments of the disclosed technology.

[0019] Figure 19 illustrates an electrospinning setup to fabricate nanofiber membranes according to some embodiments of the disclosed technology.

[0020] Figure 20 illustrates a setup for fabrication and testing of nanofiber membrane evaporators according to some embodiments of the disclosed technology.

[0021] Figure 21 illustrates a schematic of a bonding process to fabricate a hermetically sealed flexible vapor chamber according to some embodiments of the disclosed technology.

[0022] Figure 22 illustrates a schematic of a process to fabricate a flexible vapor chamber according to some embodiments of the disclosed technology.

[0023] Figure 23 illustrates an example fabrication process to make an active evaporation cooling device according to some embodiments of the disclosed technology.

[0024] Figure 24 shows a schematic illustration of an active cooling device working in a closed loop system according to some embodiments of the disclosed technology.

[0025] Figure 25 shows evaporation cooling on a fiber membrane with hydrogel coating on the fibers according to some embodiments of the disclosed technology.

[0026] Figure 26 shows characterization of a hydrogel-coated fiber membrane, (a) and (b) are top view SEM images according to some embodiments of the disclosed technology.

[0027] Figure 27 shows a schematic illustration of an evaporator with hydrogel coating according to some embodiments of the disclosed technology.

[0028] Figure 28 shows a schematic illustration of poly(sodium acrylate) or PSA hydrogel coating covalently bonded to a solid surface via silane chemical anchors according to some embodiments of the disclosed technology.DETAILED DESCRIPTION

[0029] The disclosed embodiments relate to cooling systems and methods involving fiber membranes (or referred to as porous membrane). For example, the cooling system as disclosed herein may include a high flux thermal evaporator made of fiber membranes. An example evaporator can be made flexible. The disclosed cooling systems and methods can address the cooling needs of different sectors, including cooling of electronic device, which is continuously driven by increasing heat flux (over 100 W / cm2and can be as high as 1000 W / cm2) and the prevalent trend towards wearable and foldable heat-generating electronic devices such as virtual reality (VR) headsets and foldable phones / tablets.

[0030] For high heat flux electronic devices (> 100 W / cm2), an attractive thermal management solution is a heat spreader, which may reduce the heat flux to a much lower, manageable level for conventional heat sinks. The heat flux reduction may be achieved by spreading the heat to a larger area. For example, solid materials with high thermal conductivities may be used as thermal spreaders such as Cu and CVD diamond with thermal conductivity of approximately 400 W / m-K and 2000 W / m-K, respectively, but are still inadequate for high heat flux.

[0031] As disclosed herein, capillary-driven thin-film evaporation in fiber membranes (e.g., nanoporous membranes) may provide a thermal management solution for high- power electronic devices due to their high theoretical critical heat flux (CHF) enabled by large effective areas of evaporating regions in the nanopores. However, experimentally observed CHFs in nanoporous membranes are typically limited to be less than 500 W / cm2. The disclosed embodiment, among other features and benefits, describes a fiber membrane with 3D interconnected open pores for thin-film evaporation. By extracting both efficient thin-film evaporation over nano- and micron- sized open pore structures and enhanced liquid permeability from 3D interconnected channels, a high CHF value over a large area may be achieved by capillary-driven evaporation. Merely by way of example, over 600 W / cm2CHF value may be achieved over a relatively large approximately 0.5 cm2heating area via capillary-driven evaporation using deionized water. The disclosed embodiments also demonstrate a long-term stable operation (e.g., over two hours) at high heat flux. The results indicate that the evaporator with micro-sized 3D interconnected structures can reach a higher CHF value among different evaporator strategies while the long-term stability of the device is guaranteed.

[0032] As the trend of miniaturization and increasing power density in electronic devices continues, thermal management is becoming more important for performance, reliability, longevity, and energy environmental sustainability of the computing devices. Many devices possess power dissipation over 100 W / cm2, such as light emitting diode chips (100 W / cm2) and power amplifier chips (500 W / cm2). Furthermore, even higher heat flux exceeding 1 kW / cm2can be expected in many applications such as high- power RF applications (>1 kW / cm2at die level), GaN high electron mobility transistors (HEMTs) (>1 MW / cm2at the transistor level), and pump lasers (>1 MW / cm2in the local facet optical power density). However, traditional cooling techniques, i.e., based on single phase convection or boiling heat transfer, are typically limited to heat fluxes on the order of 100 W / cm2. Therefore, high-performance heat dissipation strategies, e.g., high-flux evaporation on nanoporous membranes, with cooling medium driven by passive capillary pumping (i.e., a vapor chamber) or external active pumping, may be desirable.

[0033] Evaporation heat transfer may be desirable for cooling high-power electronics, due to the large latent heat of liquid-vapor phase change compared to single phase. Compared to boiling heat transfer, thin film evaporation has better stability and exhibits little to no hysteresis, and consumes less pumping power or purely passive capillary driven flow. Thin-film evaporation may occur in a portion of a liquid film where the triple phases, i.e., the (heated) solid, liquid, and vapor, converge. Typically, this thin film evaporating region may be less than 1 micrometer (pm) in thickness, thin enough for small thermal resistance in the liquid layer but not too thin to have too strong adhesive force of the solid to prevent the liquid molecules from evaporating (see, e.g., panel (a) of Figure 1 ). This thin-film meniscus region length may be on the order of 100 nanometers (nm) in a meniscus. Therefore, high-flux evaporation may often only occur on porous or structured surfaces, where the extended surface area leads to a large overall evaporating area. This may be the underpinning mechanism in passive vapor chambers and actively-pump evaporation devices where the evaporators are typically made of porous wicking structures.

[0034] There may be at least two mechanisms leading to the theoretical limit of evaporation heat flux from a porous membrane (e.g., a nanoporous membrane) made of individual (not interconnected) pores (e.g., nanopores) (see, e.g., panel (b) of Figure 1 ). (i) The kinetic limit, corresponding to the maximum vapor flux leaving the evaporating menisci in individual pores (e.g., nanopores). This may be because the kinetic limit of evaporation may fundamentally be governed by how fast the liquid molecules can escape from a liquid surface, which is approximately the speed of sound, and the effective areas of a thin film evaporation menisci, which may be amplified by the nanoscale pores, (ii) In the case of a passive pumping device such as a vapor chamber, the fluid flow may also be limited by the capillary limit, where the maximum liquid mass flux may be achieved due to the balance between the viscous drag and the capillary driven force in the pores (e.g., nanopores). The theoretical CHF may be determined by the lower one of these two limits in a passive device. Some theoretical and numerical studies have shown that these limits to be >5 kW / cm2at the average pore size of 150 nm with 1 pm thickness nanoporous membranes, which corresponds to over 100 kPa, and is larger for larger pores such as the evaporation rate increases from approximately 500 to 10000 fL / s when the nano pore diameter increases from 27 to 255 nm. In an active-pumping device, capillary pressure is not needed to pump the liquid, therefore, the capillary limit may not be applicable. However, capillary force inside a porous membrane is still beneficial for spreading the liquid in the membrane to prevent or delay the formation of hot spots.

[0035] Some experimental studies on thin film evaporation on nanoporous membranes and wicking structures show the efficiency of evaporation may relate to wicking structures, such as a CHF value of 367.9 W / crrr2obtained in multi-layer mesh, a CHF value of approximately 120 W cm-2obtained in bi-porous Si wicks, a CHF value of 153-596.5 W / cm2obtained in different Cu based wicking structure such as the combination of powder, mesh, foam, and microchannel with certain specific evaporator design; a CHF value of 45-144 W / cm2obtained in thin-film evaporation based on anodized aluminum oxide (AAO) membranes in prior studies. These membranes may have isolated vertical nanopores with diameters ranging from 28 to 140 nm. However, none of these has reached a very high CHF value. Among these examples, the highestvalue at the pore-level evaporation of the nanoporous membrane is 502 W / cm2on a 0.884 mm2heater size and a 0.2 mm2total open pore area, and the smaller pore size may enhance the evaporation flux from 54 to 294 W / cm2by decreasing the pore size from 305 down to 28 nm. Additional examples include a 600 W / cm2CHF value obtained from CNT-modified wicking structures via thin-film evaporation, and a CHF value of 143.75 W / cm2with a relatively large size of over 10 mm2, i.e. , commensurate with typical size of electronic chips. In these membranes where the nanopores are isolated, there may be inevitable defects inside nano-sized pores and contaminants from the cooling media (e.g., a liquid) that may clog the individual pores, as shown in panel (b) of Figure 1 . Additionally, inevitable fabrication imperfection can lead to non-uniform pore size where some smaller pore sizes exist. Consequently, when the heat flux is high, the additional flow resistance introduced by the cloggage or small pores may hinder liquid supply or vapor removal, causing local hot spots, which may ultimately lead to device failure and / or premature CHF. This may further be evidenced by the short operation time of nanoporous membranes with isolated pores, such as an operation time of approximately 20 minutes at low heat flux (e.g., <100 W / cm2), and 20 seconds at high heat flux 500 W / cm2.

[0036] Figure 1 illustrates evaporation from different types of membranes. Panel (a) shows a liquid meniscus profile where strong thin-film evaporation only happens within the transition region. Panel (b) shows membranes with isolated pores such as AAO, where inevitable defects and cloggage in certain pores may cause flow constraints that may lead to local hot spots.

[0037] Panel (c) of Figure 1 shows fiber membranes with randomly packed fibers forms 3D interconnected pores for facile fluid (liquid and vapor) transport even in the presence of local cloggage; panels (d) and (e) of Figure 1 show scanning electron microscopy (SEM) images of an example of a fiber membrane with 6.1 pm mean pore size.

[0038] In some embodiments, a cooling device may include an evaporator using a fiber membrane with 3D interconnected pores, rather than isolated pores, as a remedy for the clogging and local hot spot issue. As shown in panels (c)-(e) of Figure 1 , the fiber membrane may have a three-dimensional network of interconnected pores that mayfacilitate the transport of cooling medium, in liquid and / or vapor phases, to and from various locations within the membrane. In the event of a blockage in a particular pathway, which may occur due to contaminants and / or defects, the cooling medium can navigate through alternative routes. Consequently, the presence of a limited number of clogged channels may not necessarily lead to the failure of the entire device.

[0039] Additionally or alternatively, the network of pores may increase the surface area available for heat transfer, enhancing the overall heat transfer efficiency of the cooling device. The fiber membrane may be designed as a thin film, offering flexibility while maintaining sufficient mechanical properties. This characteristic may make it suitable for integration into various applications, such as garments, wearable and / or foldable devices, or cooling systems for electronic devices.

[0040] Figure 2 illustrates a schematic representation of a system according to some embodiments. The system 10 may include a heat source 100, a heat sink 200, and a cooling device 300 configured to cool the heat source 100 by transferring heat from the heat source 100 to the heat sink 200. The arrows in the figure indicate the direction of heat transfer from the heat source 100 to the heat sink 200 through the cooling device 300.

[0041] The heat source 100 may represent the origin of thermal energy that needs to be cooled. The heat source 100 may include a body part of a living object, e.g., a human or another animal. The heat source 100 may include an electronic device, or a portion thereof. Examples of the heat source 100 may include computing devices, mobile and compact devices, specialized processing units, electric vehicles, telecommunications equipment, unmanned aerial vehicles, or other devices that include one or more heat-generating components. Examples of computing devices may include high-performance computers (HPCs), servers, gaming computers, workstations, laptops, and game consoles. Examples of mobile and compact devices may include smartphones and laptops. Examples of specialized processing units may include graphics processing units (GPUs) used in gaming, artificial intelligence (Al) computations, machine learning, and cryptocurrency mining. Examples of the heat source 100 in the context of electric vehicles may include automotive electronic control units (ECUs, including autonomous driving systems and infotainment systems (e.g., anintegrated system that combines information and entertainment functions to enhance the driver and passenger experience)), power converters, inverters, and battery management systems. Examples of the heat source 100 in the context of unmanned aerial vehicles may include drones used for filming, surveying, or agriculture or forestry applications. Examples of the heat source 100 in the context of telecommunications equipment may include network stations, base stations, antennas, routers, switches, and modems in data transmission networks.

[0042] In some embodiments, the heat sink 200 includes a final destination for the thermal energy removed from the heat source 100. The heat sink 200 includes a condenser for cooling medium recirculation or the ambient environment.

[0043] The cooling device 300 may be positioned between the heat source 100 and the heat sink 200. The cooling device 300 may constitute or include an actively-pumped evaporator or a passively-pump evaporator (i.e. , a vapor chamber). The cooling device is in thermal communication with the heat source 100 on its first side and with the heat sink 200 on its second side.

[0044] The cooling device 300, in some embodiments, includes a fiber membrane (e.g., a fiber membrane 320 as illustrated in Figures 3A and 3B). The fiber membrane includes a plurality of fibers forming a three-dimensional network of pores, allowing a cooling medium to flow through and transfer heat from the heat source 100 to the heat sink 200. The fiber membrane may have a mean pore size smaller than 500 micrometers, or 400 micrometers, or 300 micrometers, or 200 micrometers, or 100 micrometers, or 50 micrometers, or 20 micrometers, or 10 micrometers, or 5 micrometers, or 800 nanometers, or 500 nanometers, or 300 nanometers, or 100 nanometers. For example, the fiber membrane may have a mean pore size in a range between 100 nanometers and 20 micrometers, or between 100 nanometers and 500 nanometers, or between 100 nanometers and 200 nanometers. The three-dimensional network of pores may be sized to provide capillary pumping to drive a cooling medium at least partially by capillary pressure in the fibrous membrane.

[0045] The cooling device 300 may be flexible. The fiber membrane may have a thickness less than 3 millimeter, or 500 micrometers, or 400 micrometers, or 300 micrometers, or 200 micrometers, or 100 micrometers. For example, the fibermembrane has a thickness in a range between 200 nanometers and 50 micrometers, or between 400 nanometers and 20 micrometers, or between 500 nanometers and 10 micrometers, or between 500 nanometers and 5 micrometers. As another example, the fiber membrane has a thickness of 500 nanometers, or 1 micrometer, or 2 micrometers, or 5 micrometers.

[0046] The fiber membrane may include fibers of a thermally conductive material. In some embodiments, the fiber material include a glass, ceramic, polymer, metal, or an alloy material. Examples of glass fibers include borosilicate glass or electrical glass (E- glass). Examples of ceramic fibers include aluminum oxide (alumina), aluminum nitride, silicon nitride, or silicon carbide. Examples of polymer fibers include polyethylene or polyimide. Examples of metal fibers include copper, aluminum, titanium, or silver. Examples of alloy fibers include a nickel-chromium alloy or a copper-silver alloy. The material(s) of the fibers may offer various combinations of thermal conductivity, mechanical strength, and chemical resistance, allowing for tailored performance in different cooling applications. The fiber membrane may have a thermal conductivity of at least 0.5 W / m-K, 1 W / m-K, 2 W / m-K, 3 W / m-K, 4 W / m-K, 5 W / m-K, 8 W / m-K, 10 W / m- K, 20 W / m-K, 50 W / m-K, 80 W / m-K, 100 W / m-K, 150 W / m-K, 200 W / m-K, or higher.

[0047] At least a portion of the fibers of the fiber membrane may include a coating to enhance their thermal properties or other characteristics beneficial for cooling applications. Examples of coating materials may include metals, ceramics, polymers, or composite materials. Specific examples include: copper or aluminum for high thermal conductivity; titanium dioxide or zinc oxide for improved durability and heat resistance; thermally conductive polymers such as polyvinylidene fluoride (PVDF) or polyethylene glycol (PEG); and composite coatings like carbon nanotubes dispersed in a polymer matrix for enhanced thermal and mechanical properties. Additionally or alternatively, a hydrogel coating may be used to improve water retention and / or evaporative cooling effects (e.g., by suppressing bubble formation or boiling due to the high affinity to water in hydrogel). A coating material can be selected and applied to improve the fiber membrane's performance in various cooling scenarios.

[0048] In some embodiments, the fiber membrane is manufactured using an electrospinning process. The electrospinning process may allow for precise control overfiber diameter, typically ranging from nanometers to micrometers, by adjusting parameters such as solution concentration, applied voltage, feed rate, and distance between the nozzle and collector. Multiple jets or a moving nozzle may be employed to produce wider membranes. The resulting fiber membrane may exhibit high porosity and large surface area-to-volume ratio, making it suitable for various applications including cooling devices. Post-processing steps such as heat treatment or chemical modification may be applied to enhance the membrane's properties for specific cooling requirements.

[0049] Merely by way of example, and not by limitation, after the fiber membrane is manufactured, it can undergo a coating process as a post-processing step to enhance its properties for cooling applications. Various coating methods may be employed, such as dip coating, spray coating, electroless deposition, electrochemical plating, or vapor deposition. In dip coating, the membrane is immersed in a solution containing the coating material, then withdrawn and dried. Spray coating involves atomizing the coating solution and depositing it onto the membrane surface. Electroless deposition involves the immersion of the membrane in a solution containing metal ions such as gold or silver and the deposition of a thin metal coating onto the fibers through an autocatalytic process. The fibers can be conductive or nonconductive for the electroless deposition. Electrochemical deposition also involves the immersion of the membrane in a solution containing metal ions but an electrical potential is applied between the membrane and the solution to deposit the metal on the fiber surface. It requires the membranes to be electrically conductive. Vapor deposition techniques, like chemical vapor deposition (CVD) or physical vapor deposition (PVD), may be used for applying thin, uniform coatings of metals or ceramics. The choice of coating method may depend on factors such as the coating material, desired thickness, and membrane properties. Post-coating treatments, such as heat curing or UV irradiation, may be applied to improve coating adhesion and performance. The coating process can modify the membrane's surface properties, enhancing its thermal conductivity, hydrophilicity, mechanical strength, corrosion and chemical resistance, or other characteristics relevant to cooling efficiency.

[0050] The fibers of the fiber membrane may have diameters in the range of, e.g., 0.1 micrometers to 10 micrometers, or 50 nanometers to 10 micrometers, or at least 50 nanometers, or at least 100 nanometers, or at least 200 nanometers, or at least 500 nanometers, or at least 1 micrometer, or at least 2 micrometers. The fibers of the fiber membrane may have lengths of at least 5 micrometers, or at least 10 micrometers, or at least 20 micrometers, at least 50 micrometers, at least 100 micrometers, or at least 200 micrometers, or at least 500 micrometers, or in a range of 5 micrometers and 2 centimeter, or 10 micrometers or 1 centimeter.

[0051] In some embodiments, a cooling medium (e.g., a cooling medium 330 as illustrated in Figures 3A and 3B) is employed to facilitate the heat transfer between the heat source 100 and the heat sink 200 via the cooling device 300. In some embodiments, the cooling device 300 includes a closed system with either passive capillary pumping inside the cooling device or at least partially by a pump in which the cooling medium is allowed to circulate and be reused in multiple cooling cycles. This passive closed system is exemplified by the cooling device 300A (Figure 3A), where the cooling medium 330 may be condensed and circulated back to the fiber membrane evaporator 320 by utilizing the capillary pressure from the membrane 320. This actively- pumped closed system is exemplified by the cooling device 300B (Figure 3B), where the cooling medium 330 may be condensed, pumped, and recirculated without leaving the system. The choice between a passive or actively-pumped system may depend on factors such as the specific cooling requirements, environmental considerations, and the nature of the application in which the cooling device is employed.

[0052] The cooling medium may include a phase-change coolant. Examples of the cooling medium may include water, glycol, ethanol, Fluorinert electronic liquid (e.g., FC- 72), or a refrigerant. Merely by way of example, the cooling medium may include deionized water. At least a portion of the cooling medium may undergo a phase change from a first phase to a second phase by absorbing heat from the heat source 100 on the first side of the cooling device 300. The first phase may be a liquid phase. The second phase may be a vapor phase.

[0053] At least a portion of the cooling medium may be allowed to undergo a phase change from a first phase to a second phase by absorbing heat from the heat source100 on the first side of the cooling device 300. The first phase may be a liquid phase, and the second phase may be a vapor phase. After being heated, the cooling medium in the second phase (e.g., the vapor phase), alone or in combination with the cooling medium in the first phase, may flow across the fiber membrane. This flow may be driven, at least in part, by capillary force generated by the pores between the fibers of the fiber membrane.

[0054] In some embodiments, the flow is driven by capillary force without an additional driving force. This process may be referred to as passive evaporation or cooling. For example, passive cooling may occur to cool a person wearing a garment or personal protection equipment (e.g., a firefighter protection jacket and / or helmet, sports clothes) including the cooling device 300, or a mobile, foldable, or compact device (e.g., a smartphone) including the cooling device 300. See, e.g., the cooling device 300A in Figure 3A and the description thereof.

[0055] In some embodiments, the flow is driven by a combination of capillary force and another force, e.g., by a pump (e.g., a pump 360 as illustrated in Figure 3B). This process may be referred to as active evaporation or cooling. For example, active cooling may occur to cool an electronic device (e.g., a high-performance computer, GPUs used in Al or cryptocurrency mining, network or base stations, data centers). See, e.g., the cooling device 300B in Figure 3B and the description thereof.

[0056] On the second side of the cooling device 300, the heated cooling medium in the vapor phase may transfer heat to the heat sink 200. In some embodiments, the heat sink 200 may include a secondary cooling system that may remove heat from the cooling medium 330 via a condenser (e.g., a condenser 340 as illustrated in Figures 3A and 3B). The cooling medium may transfer at least a portion of the absorbed heat to the heat sink 200 by condensing to the liquid phase. The condensed cooling medium may circulate via a cooling medium conduit (e.g., a cooling medium conduit 350 as illustrated in Figure 3B) back to the first side of the cooling device 300 for reuse. This circulation may be facilitated by a pump (e.g., the pump 360 as illustrated in Figure 3B). See, e.g., the cooling device 300B in Figure 3B and the description thereof.

[0057] Figure 3A illustrates a schematic representation of a cooling device according to some embodiments of the present disclosure. The cooling device 300A as illustratedis configured to perform passive cooling. The cooling device 300A may include a substrate 310 and a fiber membrane 320 supported on the substrate 310. The substrate 310 may be positioned on a first side of the cooling device 300A between the heat source 100 (not shown in this figure) and the fiber membrane 320. This arrangement may allow the fiber membrane 320 to be in thermal communication with the heat source 100 via the substrate 310. Although illustrated as part of the cooling device 300A, the substrate 310 may alternatively be part of the heat source 100.

[0058] The substrate 310 may include a thermally conductive material to facilitate efficient heat transfer from the heat source 100 to the fiber membrane 320. Examples of materials for the substrate 310 may include metal (e.g., copper (Cu), aluminum (Al), silicon (Si)), an alloy, thermally conductive ceramics (e.g., aluminum nitride (AIN), boron nitride (BN)), or a composite material (e.g., carbon fiber reinforced polymers).

[0059] The substrate 310 may include features (not visible in this figure) to facilitate the cooling medium 330 to absorb heat from the heat source 100. For example, the substrate 310 may include microchannels, meshes, porous structures, or other structures that form fluidic conduits. The microchannels may be formed by wire mesh or micropillars. These microchannels may have a dimension larger than the pores of the fiber membrane 320. For example, the microchannels may have a channel dimension in the order of 50 micrometers, 100 micrometers, e.g., 200 micrometers, 300 micrometers, 500 micrometers, 1 millimeter, etc. The microchannels may increase the surface area for transferring heat from the heat source 100 to the cooling medium 330, enhancing the overall heat transfer efficiency of the cooling device 300A. The microchannels or similar features of the substrate 310 may also facilitate the flow of the cooling medium to the fiber membrane evaporator 320 by providing flow pathways. In some embodiments, at least a portion of the substrate 310 includes a hydrogel coating. For example, at least a portion of the fluidic conduits including microchannels, meshes, or other porous structures, has a hydrogel coating.

[0060] The fiber membrane 320 may include fibers forming a three-dimensional network of pores that allow the cooling medium 330 to flow through. This membrane structure may be configured to enable capillary action, which drives the flow of thecooling medium 330 without additional forces in this passive cooling setup. See relevant description in connection with Figure 2.

[0061] The fiber membrane 320 may be in fluid communication with the substrate 310, allowing the cooling medium 330 to flow from the substrate 310 to the fiber membrane 320. At least a portion of the cooling medium 330 may undergo a phase change from a liquid phase to a vapor phase after absorbing heat from the heat source 100 at the substrate 310. The cooling medium 330, in vapor form or as a combination of vapor and liquid, may spread over the interface between the substrate 310 and enter the fiber membrane 320. It may then flow toward a second side of the cooling device 300A through the fiber membrane 320, driven by capillary force generated by the network of pores.

[0062] The three-dimensional network of pores in the fiber membrane 320, in contrast to isolated pores, may provide alternative flow paths for the cooling medium 330. This interconnected structure can be less susceptible to interference from clogging or defects in isolated pores. As a result, it may potentially reduce flow resistance that may otherwise hinder liquid supply or vapor removal. Such resistance, if present, may cause local hot spots, which may ultimately lead to device failure and premature critical heat flux (CHF). Additionally or alternatively, the network may increase the surface area available for heat transfer, enhancing the overall heat transfer efficiency of the cooling device 300A.

[0063] In some embodiments, the cooling device 300 includes a casing (or referred to as enclosure) 305 configured to create a fluidly sealed environment from the ambient surroundings. The vapor pressure of the cooling medium 330 inside the casing 305 is maintained at a desirable level by the fluidic sealing provided by the casing 305. The vapor pressure is the saturation of the cooling medium at the the target operation temperature. For example, for methanol, if the target operation temperature is around its melting point of 64.96 °C at 1 atm, the vapor pressure inside the casing 305 is 1 bar. For R134a with an operation temperature of 30 °C, the vapor pressure is approximately 6.69 bar. The cooling medium 330 may circulate within the enclosure 305. The enclosure 305 also shields internal components from environmental factors such as dust, moisture, or physical damage. In some embodiments, the enclosure 305 includesa thermally conductive material to facilitate heat transfer between a component within the enclosure 305 with the ambient surroundings or a system outside the enclosure 305. For example, a component within the enclosure 305 (e.g., condenser 340 described below) may transfer heat to the ambient surroundings or a secondary cooling system via the enclosure 305. Example materials of the enclosure 305 include a metal (e.g., aluminum, copper, stainless steel), thermally conductive plastics (e.g., polyphenylene sulfide (PPS), thermally conductive polyamides), a composite material (e.g., carbon fiber reinforced polymer), a ceramic material, etc.

[0064] In some embodiments, the cooling device 300 includes a condenser 340 configured to condense the vaporized cooling medium to liquid. The condensate (liquid) is returned to the evaporator (fiber membrane 320) as driven by the capillary pumping via a cooling medium conduit 350. In some embodiments, the cooling medium conduit 350 includes porous wicking structures (for example, made of sintered metal powder, or metal meshes). In some embodiments, the cooling medium conduit 350 includes space within a casing or enclosure 305 of the cooling device 300 (e.g., area or layer between the internal components (e.g., 310, 320, 330) and the outer casing 305 of the cooling device 300, serving as a flow path for the cooling medium 330).

[0065] The cooling device 300A may be used in passive cooling applications, working as a vapor chamber, for devices such as cooling garments, protection equipment, wearable or foldable devices (e.g., virtual reality (VR) glasses), or mobile devices, where direct release of the vapor to the environment is feasible and effective.

[0066] Figure 3B illustrates a schematic representation of a cooling device according to some embodiments of the present disclosure. The cooling device 300B as illustrated may be configured to perform active cooling. The cooling device 300B may include a substrate 310 and a fiber membrane 320 supported on the substrate 310, similar to the cooling device 300A described in Figure 3A. The substrate 310 may be positioned on a first side of the cooling device 300B between the heat source 100 (not shown in this figure) and the fiber membrane 320. This arrangement may allow the fiber membrane 320 to be in thermal communication with the heat source 100 via the substrate 310.Although illustrated as part of the cooling device 300B, the substrate 310 may alternatively be part of the heat source 100.

[0067] The substrate 310 may include a thermally conductive material and may have microchannels, meshes, porous structures, or other structures that form fluidic conduits for fluid flow and heat absorption. The fiber membrane 320 may include fibers forming a three-dimensional network of pores that allow the cooling medium 330 to flow through. The substrate 310 and fiber membrane 320 in the cooling device 300B may have the same or similar characteristics as described for the cooling device 300A, the description of which is applicable and not repeated here.

[0068] In contrast to the passive cooling system 300A, the cooling device 300B may include or be operably connected to one or more additional components for active cooling. These may include one or more of a condenser 340, a pump 360, a cooling medium conduit 350, or a compressor 370.

[0069] The condenser 340 may be configured to cool and condense the vapor phase of the cooling medium 330 back into a liquid phase. It may facilitate the transfer of heat from the cooling medium 330 to the heat sink 200 on a second side of the cooling device 300B. The heat sink 200 may be the ambient environment or a secondary cooling system (not shown in this figure). The condenser 340 acts as an interface between the cooling medium 330 and the heat sink 200, enabling efficient heat dissipation from, e.g., the system 10.

[0070] The pump 360 may be configured to circulate the cooling medium 330 through the cooling device 300B. It may drive the flow of the condensed liquid cooling medium 330 from the condenser 340 back to the substrate 310 and fiber membrane 320 for reuse, e.g., in a continuous cooling cycle.

[0071] The cooling medium conduit 350 may provide a pathway for the cooling medium 330 to flow between different components of the cooling device 300B. It may connect the condenser 340, pump 360, and the substrate 310 / fiber membrane 320 assembly. The cooling medium conduit 350 may include one or more porous wicking structures for the cooling medium 330 to flow through. For example, the condensed cooling medium 330 may enter the cooling medium conduit 350 after being condensed at the condenser 340 on the second side of the cooling device 300, and exit the cooling medium conduit 350 at or in the vicinity of the substrate 310 on the first side of the cooling device 300B. The cooling medium conduit 350 and the fiber membrane 320 mayconstitute part of a closed cooling medium circuit through which the cooling medium 330 circulates within the cooling device 300B.

[0072] In some embodiments, the cooling device 300 includes a casing or enclosure 305, thereby fluidly sealed from the ambient surroundings. The cooling medium 330 may be allowed to circulate within the casing or enclosure, driven by, e.g., capillary force. See, e.g., the cooling device 300A as illustrated in Figure 3A. For example, the capillary force may be generated by the pores of the fiber membrane 320, and / or one or more wicking features in the substrate 310 (e.g., 1812 in Figure 18, 2012 in Figure 20, and the wicking features in 2000 in Figure 21) and / or in the condenser 340 (e.g., 21 2 in the condenser in Figure 21 , the Cu mesh in panel (a) of Figure 22).

[0073] The compressor 370, if available, may be configured to increase the pressure of the condensed cooling medium 330. This compression may increase the cooling efficiency of the system by raising the boiling point of the cooling medium 330, allowing it to absorb more heat from the heat source 100 before vaporizing.

[0074] In operation, the cooling medium 330 may absorb heat from the heat source via the substrate 310 and fiber membrane 320, similar to the process described for cooling device 300A. The heated vapor may then flow to the condenser 340, where it can be cooled and condensed. The pump 360 may then circulate the condensed cooling medium 330 back through the system, returning to the substrate 310 and fiber membrane 320. In some embodiments, the cooling medium after the pump 360 may pass through the compressor 370 to increase its pressure before returning to the substrate 310 and fiber membrane 320.

[0075] This active cooling configuration may allow for more efficient heat transfer and may be suitable for applications with higher heat loads. The cooling device 300B may be used in applications such as high-performance computers, data centers, telecommunication equipment, or other electronic devices with significant cooling needs.

[0076] The cooling device 300 including a fiber membrane in some embodiments may possess one or more of the following desirable features. The cooling device 300 may have a heat flux of at least 100 W / cm2, at least 200 W / cm2, at least 300 W / cm2, least 500 W / cm2, at least 600 W / cm2, at least 700 W / cm2, at least 800 W / cm2, at least900 W / cm2, at least 1 kW / cm2, at least 1 .2 kW / cm2, at least 1 .5 kW / cm2, at least 2 kW / cm2, or higher than 2 kW / cm2The fiber membrane's ability to spread heat may reduce the heat flux by at least one order of magnitude, for example, by a factor of over 100. This reduction may bring the heat flux to a manageable level for heat removal by, for instance, a heat sink. The cooling device 300 (e.g., 300A and 300B as illustrated in Figures 3A and 3B) with its capacity for high heat flux may provide efficient cooling for electronic devices, wearable devices, foldable devices, personal cooling garments, etc.

[0077] The cooling device 300 may be thin. The thickness of the cooling device 300 may be smaller than 3 millimeters, 1 millimeter, 0.5 millimeters, 0.2 millimeters, 0.1 millimeters. This minimal thickness may contribute to the flexibility of the cooling device 300, allowing it to conform to various shapes of devices or objects that need cooling.

[0078] The manufacturing process for the cooling device 300 may be cost-effective and scalable. The fiber membrane, a key component of the cooling device, may be produced using electrospinning technology, which can offer advantages in terms of production efficiency and scalability.

[0079] These and other features of the cooling device 300 may facilitate the commercial development of low-cost, lightweight, and efficient cooling solutions. Such solutions may be advantageous in various applications, including lightweight wearable devices and apparel (e.g., firefighter protective gear, sports jackets, VR headsets), portable devices (laptops, mobile phones, gaming consoles), and high-performance computing systems (e.g., data centers, cryptocurrency mining facilities, Al infrastructure).

[0080] Figure 4 illustrates a flowchart of a cooling process according to some embodiments of the present disclosure. The cooling process 400 may be performed using the cooling device 300.

[0081] At 410, a cooling medium (e.g., the cooling medium 330) may flow in the cooling device 300 from a first side toward a second side of the cooling device 300. The fiber membrane may include a plurality of fibers forming a three-dimensional network of pores, similar to the fiber membrane 320 described in Figures 3A and 3B. This flow may be driven by capillary action in the fiber membrane, as described in the passive cooling system 300A, or may be assisted by a pump 360 in an active cooling system 300B.

[0082] At 420, the cooling medium may vaporize, when flowing through the fiber membrane 320, in the fiber membrane 320 by absorbing heat from a heat source (e.g., heat source 100) that is in thermal communication with the cooling device 300 on the first side of the cooling device 300. This operation may correspond to the process described in the passive cooling device 300A or the active cooling device 300B, where the cooling medium 330 absorbs heat from the heat source 100 via the substrate 31O.The cooling device 300 may be in thermal communication with a heat sink 200 on the second side.

[0083] At 430, heat may be removed from the cooling medium to the heat sink 200. The heat sink 200 may include a secondary cooling system in thermal communication with the cooling medium via a condenser (e.g., condenser 340 in the cooling device (or referred to as vapor chamber) 300A in the passive cooling system 300A, and in the active cooling system 300B).

[0084] In some embodiments, the heat is removed by condensing the vaporized cooling medium and circulating the condensed cooling medium back toward the first side of the cooling device 300. At least a portion of the vaporized cooling medium is allowed to undergo a phase change from the vapor phase to the liquid phase by the condensing. This circulation may be achieved via a cooling medium conduit 350, driven by a pump 360, by capillary force (e.g., created by the fiber membrane 320, 1816 in Figure 18, nanofiber membrane 2020 in Figure 20, fibrous membrane 2320 in Figure 23), or a combination thereof.

[0085] In certain implementations, the condensed cooling medium may be pressurized by compressing it, using a compressor 370, to enhance cooling efficiency.

[0086] The process 400 may provide a versatile approach to cooling adaptable for various applications, from cooling in wearable devices or objects to high-performance electronic systems.

[0087] Merely for illustration purposes and not intended to be limiting, passive, capillary-driven thin film evaporation experiments were conducted on glass fiber membranes with mean pore sizes (dm) ranging from 3.2 to 11 .4 pm. All the membranes tested possess a high CHF over 200 W / cm2with deionized water at 1 atm, with a relatively large evaporator size of 0.5 cm2. Among the tested samples, the highest CHFwas obtained in membranes with 6.1 m mean pore size, yielding a high value of 643.4±104.6 W / cm2. The pore size dependence of the CHF values can be understood from the competition between the evaporation limit and the balance between the capillary force and permeability. The membranes also demonstrated long-term (> 2 hours) continuous thin film evaporation at high flux. The fiber membranes are low cost and scalable to manufacture, and mechanically flexible and strong, as evidenced from mechanical characterization. The results demonstrate the suitability of using fiber membranes for thin film evaporation and integration into high-flux cooling devices such as evaporators and vapor chambers for wearable objects (e.g., garment, wearable device, foldable device, protection equipment), mobile or compact devices, electronic devices (e.g., high-performance computing devices).

[0088] The morphologies of the membranes were examined by scanning electron microscopy (SEM). As shown in panels (d) and (e) of Figure 1 , the fiber membranes possess 3D interconnected micron-sized pores, close to the structure shown in panel (c) of Figure 1 , which are distinct from one-dimensional (1 D) isolated nanopores in the nanoporous membranes as illustrated in panel (b) of Figure 1. Since the membranes have a broad distribution of fiber diameter and pore size, the mean pore size may be employed to characterize them. Table 1 shows the sample ID and their characteristics. As an example, in panels (d) & (e) of Figure 1 , both sub-micron and > 6 pm open pores can be found in the membrane sample D with a mean pore size of 6.1 pm. More morphological images can be found in Figure 5.Table 1. Sample list, mean pore size, measured capillary pressure and permeability of fiber membranes at different mean pore size.

[0089] The fiber membranes were assessed for their feasibility as thin film evaporators (e.g., the cooling device 300). As mentioned earlier, the maximum liquid mass flux may be determined by the capillary limit, which may be reached when the capillary pressure in the membrane is balanced by the flow resistance across the membranes. Therefore, the capillary pressureand water flow permeability (fc) of the fiber membranes were measured, as shown in panels (a) and (b) of Figure 6, respectively. The values are also included in Table 1 . The permeability value is further used to calculate the mean pore diameter dm, by assuming uniform pores with diameter dmevenly distributed in the membranes, based on Darcy’s law and Darcy-Weisbach equation (1 ):where s is the porosity of the membrane, which is assumed to be 50%, typical for the fiber glass membranes used here.

[0090] The measured permeability was compared to the calculated values for different mean pore sizes based on the Laplace equation (2):where o is the liquid surface tension and 0 is the contact angle between liquid and solid (e.g., the fiber membrane). The range of the contact angles to be 40° and 60° were selected for the hydrophilic glass fibers to estimate the capillary pressure, which led to a range of the calculated capillary pressure, as plotted in panel (b) of Figure 6. The result shows that the measured capillary pressure is within the range of the calculated ones based on equation (2). Therefore, the mean pore size determined by equation (1 ) is a good measure of the membranes for both the permeability and capillary pressure.

[0091] Figure 6 shows (a) permeability as a function of mean pore size and (b) measured capillary pressure as a function of mean pore size of the GF membranes illustrated in Figure 5. Panel (a) of Figure 6 shows permeability as a function of mean pore size. The mean pore size is calculated based on equation (1 ) with the measured permeability data by assuming a uniform pore size. Solid square symbols represent the measured permeability, and the solid line is the calculated one based on equation (1 ). Panel (b) of Figure 6 shows measured capillary pressure as a function of mean poresize. Solid square symbols represent the measured values and the shaded area shows the calculation based on equation (2) with the contact angle 0 ranging from 40° to 60°.

[0092] Figure 7 shows evaporation experiment and results of the GF membranes illustrated in Figure 5. Panel (a) shows a custom-made thin-film evaporation setup. The illustrated evaporation test setup includes two Cu electrodes (II) connect to the outside circuits where a power supply and data acquisition devices were located; inset: as prepared fiber membrane (I) for evaporation test with silver (Ag) paste on the two ends as electrodes, the middle platinum (Pt) film region acts as both a heater and a thermometer. The fiber membrane (I) with the Cu electrodes (II) was held using silicone (III). Water was used as the working fluid (IV).

[0093] Panels (b)-(d) are results based on water as the working fluid. Panel (b) of Figure 7 shows the measured evaporation curve of heat flux as a function of superheat for the five types of fiber membranes with mean pore size (dm) of 3.2-11 .4 pm. The superheat is defined as the difference between the heater temperature (Th) and the saturation temperature (Tsat), which is 100 °C for water at 1 atm. Each curve represents average data over >5 samples tested for each type of the membrane. The error bars are standard deviations of >5 measurements. Each curve and error bar in panel (b) of Figure 7 represents averaged results among >5 tested samples. The evaporation curves display a similar trend among samples B to E, with the mean pore size ranging from 4.8 to 11 .4 pm. When the superheat is above approximately 20 °C, the heat flux starts increasing sharply with little increase in superheat which implies efficient heat transfer, until the samples reach the CHF values. In particular, for sample D with a mean diameter of 6.1 pm, the CHF was the highest, reaching 643.4±104.6 W / cm2. On the other hand, for membrane A, which has the smallest mean pore size (dm=3.2 pm), the evaporation curve deviates from others, with a more gradual increase in the heat flux with the superheat, which implies less efficient heat transfer than the other samples at elevated heat flux level.

[0094] The heat flux and superheat results are converted to heat transfer coefficient (HTC), defined as the heat flux divided by the superheat, as shown in panel (c) of Figure 7. The trend of HTC as a function of superheat is similar among samples B to E. In particular, for samples B to D, the HTC reached over 10 W / cm2 / K with heat fluxesover 600 W / cm2. The HTC for sample E, which has the largest mean pore size of 11 .4 pm, the HTC was slightly lower. Sample A, on the other hand, possessed the lowest HTC.

[0095] Panel (d) of Figure 7 show the CHFs value as a function of the mean pore size. The figure includes a summary of measured CHF values of the fiber membranes with different pore sizes, AAO membranes data from the literature and from this study; and comparison with kinetic and capillary limits. CHF of evaporation from porous membranes may relate to several parameters, including sonic limit, capillary limit, and evaporative limit.

[0096] As illustrated in panel (d) of Figure 7, as the mean pore size increases from 3.2 to 6.1 pm, the CHF initially increased, peaked at 6.1 pm for the sample D (dm= 6.1 pm), and then decreases for larger mean pore size. This non-monotonic trend may be due to the tradeoffs between the capillary pressure and flow resistance as the pore size varies, as well as the clogging effect. The modeled kinetic limit and capillary limit are shown in the same figure for comparison. The shaded area is the calculated capillary limits with different contact angles of 40° - 60° and the dashed line shows the kinetic (or evaporation) limit modeled for single pores with the same diameter as the mean pore size. The calculations of these limits are shown in elsewhere in the present document. See, e.g., the portion regarding the Modeling of Single Pore Kinetic Evaporation Process. When the pore sizes are smaller, the overall thin film evaporating meniscus area is larger, leading to higher thin film evaporation limit. In the meantime, the capillary limit is reduced due to the greater influence of the pore size on the flow resistance compared to the capillary pressure. At the capillary limit, the flow resistance (scales as d~n for a given flow rate) is balanced by the capillary pressure (resistance scales as d^ ). Therefore, a smaller pore size leads to lower capillary limited flow rate and heat flux. Based on these trends, there may exist a desirable pore size, which is about 500 nm - 1 pm based on the theoretical calculation whose results are illustrated in panel (d) of Figure 7. However, this theoretical CHF value, around 4 kW / cm2, is much higher than the experimentally observed CHF. The measured CHF peaked at mean pore size of 6.1 pm and all the measured CHF values are lower than the capillary or evaporation limit. This difference may be caused by the wide distribution of the pore sizes in themembranes (see, e.g., panel (d) of Figure 1 and Figure 5). Even though the permeability and capillary pressure, which are key parameters in the models, are well described by the mean pore size, the actual CHF may be disproportionally dominated by pores at the ends of the distribution spectrum. Local hot spots caused by very small pores due to the capillary limit or very large pores due to the evaporation limit may lead to a lower CHF value than the theoretical prediction based on the mean pore size. Additionally, the smaller pores in the membranes may cause cloggage, thus leading to a lower CHF. One or more of these and other factors may contribute to the observed CHF that are lower than the theoretical limits. The comparison also suggests the potential strategies for further increasing the CHFs, e.g., membranes with uniform fiber diameter and pore size.

[0097] Panels (e) and (f) of Figure 7 show evaporation test results using ethanol as the working fluid (e.g., the cooling medium 330) at 1 atm ambient pressure. Panel (e) shows measured heat flux as a function of superheat, for the five types of fiber membranes with mean pore size (dm) of 3.2-11 .4 pm. Each curve represents average data over 3-4 samples tested for each type of the membrane. The error bars are standard deviations multi-measurements. Panel (f) shows the summarized average CHF value as a function of mean pore size.

[0098] Figure 8 shows comparison of the CHF values between AAO membranes with isolated pores from prior studies and fiber membranes with 3D interconnected open pores according to some embodiments. All the fiber membranes show higher CHF values than AAO membranes (from prior studies) with pore size approximately 100 nanometers. AAO membranes with pore size of approximately 200 nanometers were also measured using the same experimental setup and conditions as the fiber membrane experiments and obtained similar results as the values on the AAO membranes. As discussed earlier, the AAO membranes, having a pore size of approximately 200 nanometers, are expected to have theoretical CHF values of >1500 kW / cm2(see, e.g., panel (d) of Figure 7). However, the experimentally measured values are significantly lower, at around 100 W / cm2. This suggests the higher degree of susceptibility of the AAO membranes to clogging. The difference between the fiber andAAO membranes demonstrates the advantage of the 3D-interconneted open pore structure in the fiber membranes compared to the insolated pores in the AAO.

[0099] CHF value may be influenced by the different heater sizes, where smaller heater size may yield higher CHF due to, e.g., the smaller lateral flow resistance, heat spreading from the smaller heater (e.g., the heat source 100) to the larger evaporator (e.g., the cooling device 300), or one or more other factors. To reduce or eliminate this possible heater size effect, the results of the evaporator including a fiber membrane as disclosed herein were compared to others where >0.4 cm2heater size was applied with limited superheat <80 °C to reduce or eliminate parasitic heat loss. By applying such criteria, the CHF values arise from 45 up to 512 W / cm2from different evaporators from prior studies (including sintered Cu powder & mesh, Si pillars, multi-layer Cu-based wick, and AAO membranes). The fiber membrane exhibits the highest CHF value in the comparison which may be due to one or more factors including the enhanced capillary limit, efficient evaporation kinetic, and lower susceptibility to cloggage in the 3D microsized fibrous structures.

[0100] The lower susceptibility can also lead to stable long-term evaporation of the fiber membranes, compared to AAO or other membranes with isolated nanopores, where the clogging due to contaminant from the testing fluids led to short (e.g., approximately 20 min) evaporation. By introducing interconnected channels in fiber membranes, liquid can be replenished from all directions as shown in panel (c) of Figure 1 to avoid local hot spots at relatively high flux. Therefore, the open pore structure may lead to better long-term stability at elevated heat flux. This situation was tested with two heat fluxes at 150 and 300 W / cm2using sample D that shows the highest CHF value. In the experiments, the heat flux was held at the desired level for an extended period. Panel (a) of Figure 9 shows that the membranes can operate stably over 2 hours at 300 W / cm2and approximately 4 hours at approximately 150 W / cm2. The experiments were stopped due to the depletion of water in the reservoir, not due to the failure of the membranes. The inset plot of nearly constant resistance of the Pt thin film heater (less than 5% change over time) also suggests the stability of the cooling devices configured according to some embodiments. In comparison, the AAO membranes showed less than a stable operation for less than 20 minutes under the heat flux up to 100 W / cm2.

[0101] Panel (b) of Figure 9 shows the morphology of the fiber membranes before and after long-term tests at 300 W / cm2. Two different locations (in the heating region) were selected from the sample after the test as examples to evaluate the membrane’s morphology. In most regions after the test, the morphology was very similar to location #1 , which was almost the same as its pristine morphology (panels (d) and (e) of Figure 1 ). However, there were still some small portions showing signs of cloggage, such as in location #2, which may come from contamination from both environments and the test liquid (e.g., the cooling medium 330). The small portion of the clogged areas similar to location #2 did not affect the long-term stability of the membrane operated at high flux. However, they may have created local hot spots and led to the lower-than-expected CHFs values, as shown in panel (d) of Figure 7. Nevertheless, the long-term stability of the fiber membranes may further validate that the idea that the 3D open pore structures can lead to facile liquid and vapor transport inside the membrane and thus reduce or diminish the likelihood of cloggage.

[0102] Figure 9 shows results of long-term stability tests using fiber membranes and morphology of the fiber membranes after the long-term stability tests according to some embodiments. Panel (a) shows long-term stability test using fiber membranes (Sample D) with a mean pore size of 6.1 pm at two different heat flux levels; inset is the normalized resistance by the resistance at the beginning of target heat flux level over time; shadow area represents AAO membrane device operation time from prior studies included here for comparison purposes. Panel (b) shows the morphology of the fiber membranes after the long-term stability test. The sample comes from #3 in panel (a) of Figure 9; two different locations over the entire heater area were selected.

[0103] The results in Figures 5 through 9 experimentally demonstrate the benefit of using 3D micro-structured fiber membranes as a capillary-driven evaporator in both high CHF value and long-term stability compared to single vertical channel AAO membranes. Among the tested cooling devices including the fiber membranes according to embodiments, the maximum CHF value 643.4±104.6 W / cm2was obtained with a mean pore size of 6.2 pm which compensates for both capillary and evaporation kinetic limits. The characterization of both capillary and liquid transport illustrates an enhancement in capillary limit by facile liquid transport inside an interconnected 3D structured channelcompared to a single vertical channel while a micrometer-sized open pore structure may sustain substantial capillary pressure. An efficient thin-film evaporation process may be obtained due to the relatively small open pore area. The results demonstrate a trend in CHF value as a function of the fiber membrane’s mean pore size. Such a phenomenon may be due to one or more factors including capillary, evaporation kinetic limits, or other factors, or a combination thereof. Furthermore, the results demonstrate a long-term stability of fiber membranes operating in different conditions. Over 2-hour and 4-hour operation times were achieved at 300 and 150 W / cm2. Both morphology and monitored resistance demonstrate good stability during the long-term operation. The 3D structured interconnected channels can effectively reduce or eliminate hot spots due to defects and skinny channels by rapid liquid replenishment from various directions. The cooling efficiency of the cooling device and method as disclosed herein may be further improved by improving the irregular shape and relatively larger size distribution in the fiber membrane's open pore structure, improving or avoiding the atmosphere environment that may introduce additional resistance to vapor transport and potential contamination issues, or the like, or a combination thereof. For example, a closed circuit for the cooling medium may be implemented to reduce or avoid the contamination issues. Using a 3D micron-structured fiber membrane as a capillary-driven evaporator can improve overall performance than single vertical-channel membranes at moderated pore size. Similar 3D micrometer-structured membranes can reach similar or even higher performance depending on membrane materials, pore size, operation conditions, or the like, or a combination thereof.

[0104] The results in Figures 5 through 9 were obtained based on the preparation methods below.

[0105] Preparation of fiber membrane evaporator: glass fiber membranes with various mean pore sizes (Table 1 ) were obtained from commercial vendors (Sterlitech. Co). The membranes were then treated in a silane solution to enhance its mechanical strength as described in connection with Figure 10, which may maintain the morphology of the membranes during the evaporation experiments. The silane solution was mixed with 1 ml of Silane A174 (Sigma-Aldrich), 2 microliters (pL) of Acetic acid (Sigma- Aldrich) and 50 ml of distilled water. The detailed recipe of exemplary saline solutions ismentioned in Figure 10. After drying out the saline solution, an approximately 300 nm thick Pt was deposited onto the middle of the fiber membranes by sputtering through a stencil with a width of 7 mm. Upon the sputtering, the two ends along the length of the rectangular Pt strip were coated with a thick layer of an Ag paste and bonded with Cu sheets serving as electrodes to connect with external wires, leaving a square window with 7 mm size (0.49 cm2area), a Pt film in the middle serving as the heater and resistive thermometer for the evaporation experiment.

[0106] Figure 10 illustrates GF membrane preparation procedures according to some embodiments. Panel (a) illustrates a procedure for making silane solutions: add 1 mL of Silane A174 (3-[(Trimethoxysilyl)propy]trimethoxysilane and 2 microliters (uL) of Acetic acid into 50 mL distilled water; then, the mixed solutions was stirred on a stirring plate with a magnetic stirrer over 12 hours; next, well-prepared silane solutions for the treatment of GF membranes. Panel (b) shows GF membranes preparation before evaporation test: #1 : the well-prepared silane solution was dipped onto pristine GF membranes; the wet membranes were naturally dried out in the fume hood; approximately 300 nm thin-film Pt layer was deposited on the dry membranes; the Ag paste was applied on the two ends of Pt film as electrodes; after approximately 20 minutes curing of the Ag paste at approximately 120 °C, the GF membrane was well- prepared; #2: before evaporation test, the well-prepared GF membrane was immersed into ethanol in 10 mins, then using ethanol to flush out the membrane; finally, the GF membrane was immersed into distilled water over 20 minutes for the evaporation test.

[0107] Measurement of mechanical properties: The mechanical strength of the GF membrane strip was measured as shown in Figure 11 with uniaxial tensile tests using a dynamic mechanical analyzer (RSA-G2, TA Instrument) at room temperature. The sample was prepared by attaching the two ends of the GF membrane strip onto an epoxy-based composite plate (Garolite, McMaster-Carr, USA) using superglue (495, Loctite, USA). A 1 % / s strain rate was maintained, and the sample plates were gripped during the tensile tests. The strength value was taken from the average failure stress value of five replications of each sample.

[0108] Figure 11 illustrates a setup for testing GF membrane’s maximum tensile strength according to some embodiments. The strength value of the GF membrane stripwas measured in a uniaxial tensile test using a dynamic mechanical analyzer (RSA-G2, TA Instrument) at room temperature. The sample was prepared by attaching the two ends of a GF membrane strip onto an epoxy-based composite plate (Garolite, McMaster-Carr, USA) using superglue (495, Loctite, USA). A 1 % / s strain rate was maintained, and the sample plates were gripped during the tensile tests. The strength value was taken from the average failure stress value of five replications of each sample.

[0109] Figure 12 shows the mechanical strength of the pristine (the line with solid squares) and silane-treated (the line with solid circles) membranes according to some embodiments. The results demonstrate that the silane treatment may enhance the mechanical strength. The coating may prevent the sliding between fibers during the evaporation experiment to maintain the morphology of the membranes.

[0110] Permeability test: The permeability of the membranes (fc) was measured according to the Darcy’s law:Q = kA / pL Ap, (3) where Q is the volumetric flow rate, A is the total cross-sectional area of the membranes, L is the thickness of the membrane, Ap is the pressure drop across the membrane, and is the viscosity of water. In the experiment, water was pushed through a membrane with a fixed pressure. The flow rate was measured by the total mass of water passing through the membranes for over 60 seconds using a balance (Electronic Toploader, 1002MP9, Sartorius). More details about permeability test set up can be found in Figure 13, and relevant description thereof.

[0111] Figure 13 illustrates a permeability test setup according to some embodiments. The pressure was applied to a pressurized tank 1310 as a pressure head to drive water flow through the membranes; the tank pressure was monitored via a pressure gauge 1320; by opening a valve V2 1330 connected to the pressurized tank 1310, the water tank 1340 was pressurized from the pressurized tank 1310; during measurements, a valve V1 1350 was kept open for 1 minute, pushing the water to flow from the water tank 1340 to through the membrane 1380; the water flowing out from membrane 1380 was collected in a beaker 1360 on top of an electronic balance 1370;the total mass of water was divided by 1 minute to calculate mass flow rate which may be converted to volumetric flow rate to finally calculate membrane’s permeability.

[0112] Capillary pressure test: The capillary pressure of the membranes was measured by compensating the capillary force of membranes and negative liquid pressure from a mechanical vacuum pump. The fiber membrane was first installed into an evaporation setup. The chamber where the liquid bath was located was then started the vacuum. The vacuum level was recorded by a vacuum gauge in unit kPa. Once liquid receded beneath the evaporator, the corresponding vacuum was considered as membrane capillary pressure.

[0113] Thin film evaporation setup: The membranes deposited with the Pt thin film heater and Cu electrodes were then mounted to a custom-made thin film evaporation setup, as shown in panel (a) of Figure 7. The prepared membrane (I) was mounted onto an acrylic sample holder, with a silicone square-ring (III in panel (a) of Figure 7) in between. The fiber membrane sample (I) was sandwiched between two holders with an open area in the middle for water supply and vapor evacuation. The acrylic holder has a through hole in the center that had (substantially) the same size as the Pt heater of 7x7 mm2and was aligned to the heater for liquid supply. Then, another silicon square ring (III) and an acrylic cover were placed on top of the membrane (I). The silicone and the cover had the same 7x7 mm2opening aligned to the Pt heater on the membrane (I).The acrylic cover was clamped to the holder using four screws (V).

[0114] The bottom of the liquid supply hole in the holder was connected to a liquid bath through a Cu tube (VI). The top level of the water (IV) in the bath was intentionally adjusted to be lower than that of the membrane (I), to guarantee a capillary-driven water supply to the membrane (I). The water bath was placed inside an enclosed chamber with fixed pressure (1 atm in this study) and the water (IV) was kept at its boiling temperature (100 °C) with film heaters wrapped out of the bucket. The Cu tube (VI) was also wrapped with heaters and thermal insulations to ensure that the liquid supplied to the membrane (I) was closed to the saturation temperature of 100 °C (measured to be >80 °C during the measurements).

[0115] Figure 14 illustrates a setup for characterizing GF membrane’s capillary pressure according to some embodiments. The GF membrane 1410 was installed intothe evaporation test setup, held using silicone 1420. The membrane 1410 included a Pt coating. The membrane 1410 was intentionally lifted above the water level in the bath of the working fluid 1430; during the measurements, the valve V1 to the vacuum pump 1440 was slowly opened (0 kPa vacuum pressure inside the chamber before the measurements) and chamber pressure started decreasing by vacuum pumping; the pressure gauge and water level on the membrane 1410 were simultaneously monitored; once water receded from the membrane back to the bath, the pressure gauge reading was recorded as maximum capillary pressure of the membrane 1410.

[0116] Thin film evaporation experiment: The chamber was open to the ambient atmosphere (1 atm). The membrane (I) was placed at a higher position than the water level in the reservoir to guarantee the flow was capillary driven. The liquid was degassed by vacuum and preheating process. Before the measurements, the liquid bath was heated up to ~100 °C with a (substantially) constant heating process. During the measurements, two Cu electrodes (II), which connected to the power supply and data acquisition system, were in direct contact with Ag electrodes on membranes (I). The current and voltage passing through the Pt heater were monitored throughout the entire experiment. Then the collected current and voltage values were converted to Pt film resistance (as RTD) and input power (as heat flux) on the heater. The measurements began at low heat flux. Then, input power was increased to reach a higher heat flux. At each flux level, the reading for both current and voltage were monitored until it reached a stable value (usually less than 10 seconds). We waited for approximately 30 seconds to 1 minute at each heat flux level before we increased the heat flux level. The overall time in one test lasted >30 min.

[0117] During the experiment, the pressures of the liquid bath and the vapor space above the membrane (I) were both fixed at 1 atm. This was different from an earlier thin film boiling experiment where the liquid side temperature was higher than the vapor pressure. The same pressure between the liquid and vapor, combined with the fact that the liquid bath was lower than the membrane (I), ensured that the liquid supply was only driven by the capillary pressure in the membrane (I). Heating power was applied to the central 0.49 cm2Pt heater area through the two Cu leads (II) using a power supply. The Cu and Ag pastes were much thicker than the Pt (>3 urn vs. 300 nm) and are moreconductive (Cu conductivity is 5x of Pt), therefore, the Pt heater was the dominant resistance (over 95%) of the entire circuit. This was further confirmed with four-point vs two-point resistance measurements (in panel (c) of Figure 15). The heating power was increased from zero to the critical heat flux in a stepwise manner, with an interval ~2-10 W (5-20 W / cm2) at each step, by adjusting the current from the power supply. The power applied to the heater was measured from the current and voltage. The voltage was measured by a voltmeter connected to the two Cu leads (II). The current was measured from the voltage reading of a shunt resistor with a constant resistance (FL2- 0.5, 70 V / 20 mA, 3.75 mfi) connected in series with the sample. The temperature of the sample was determined from the resistance reading of the Pt heater which also served as a resistance thermometer. The temperature coefficient of the resistance (TCR) of the Pt thin film heaters for each batch of the samples (made from the same Pt sputtering session) was calibrated by measuring the resistance vs. temperature curve of the samples in a temperature-controlled oven. At least three samples for each type of membranes were measured to get an averaged TCR value. Among the measurements, ± 5% error was expected among each type of membranes. The details about TCR measurements are described in Figure 15. A typical Pt film resistance vs temperature is shown in panel (c) of Figure 15.

[0118] Figure 15 shows a setup for and results of characterization of a temperature coefficient of the resistance (TCR) according to some embodiments. Panel (a) shows 4- point measurements (4pp) inside the temperature-controlled chamber; two P-type thermal couples were located on the top and beneath the membrane for accurate temperature readings; the total applied power on Pt film was less than 50 mW to avoid joule heating effect; the two thermal couples temperature, current and voltage readings were recorded after >30 minutes at each temperature set points to ensure the temperature stability of membranes. The arrow in panel (a) shows the direction of the current applied. Panel (b) shows the results of comparing Pt film resistance from 2-point measurements (2pp) during the evaporation tests and 4-point measurements; <5 % difference in 2pp measured resistance indicates that majority electronic resistance comes from the Pt film itself. Panel (c) shows the results of TCR measurements on sampled #D with mean pore size 6.1 pm, > 0.999 R2value indicates good linearity usingthe resistance temperature detector (RTD) as the temperature sensor. Panel (d) shows fitted TCR values for all 5 types of GF membranes (>3 measurements on each type); <5 % error in TCR values for all 5 types of GF membranes.

[0119] Figure 16 illustrates an evaporation test setup according to some embodiments. The vapor chamber 1610 and sample part 1620 are described with respect to Figure 7. The data acquisition system includes two multimeters 1630 and 1640. The top one 1630 in Figure 16 may be configured to monitor the voltage (VR) across the shunt resistance 1650 (Rshunt = 3.75 m£l), and the bottom one 1640 may be configured to record the total voltage (Vo) applied to the entire circuit from the AC transformer 1660.

[0120] Before the test, the distilled water was filled into a bucket in the chamber 1610. Then the water was degassed at ~0.1 bar in the chamber 1610 over 40 minutes. After the vacuum degassing procedure, the entire bucket was heated up to the boiling point of water (~100 °C at 1 atm) using two thin-film heaters wrapped outside the bucket. After water boils inside the bucket, the heating power is decreased and well- controlled to keep water hot (>95 °C) during the entire measurement.

[0121] The prepared GF membrane sample (as described with reference to Figure 10) was installed into the evaporation test setup. Before the measurements, hot water (e.g., (>95 °C) was allowed to flow through the membrane 1620 for a while (e.g., >1 minute) to flush out debris or contaminations inside the setup and samples. Next, the membrane 1620 was lifted to the level above the water level inside the water bath 1670 where the water was unable to flow out through the membrane 1620 by adjusting the height of the table 1680 where the chamber 1610 and the membrane 1620 were placed. This step was to ensure the occurrence of the capillary-driven evaporation process.

[0122] Once the membrane sample 1620 was installed, the measurements started. By dividing the shunt resistance value, the current (I = Vp / Rshunt) passing-through of the sample 1620 was obtained. Using the total voltage applied to the entire circuit, the total input power (Qin = Vol) was calculated. Meanwhile, the Pt film resistance was extracted by dividing the current from the total voltage (Rpt = Vo / I) as RTD. Finally, by dividing the Pt heater size (~0.49 cm2), the heat flux was extracted.

[0123] During the test, the power level (heat flux) was incrementally increased by a controlling alternating current (AC) power transformer, with each step resulting in an increase of approximately 5-20 W / cm2heat flux. To ensure the stability of the system and samples, the two voltage values at each step were recorded after 30 s - 1 min waiting when the multimeter readings become stable. The input power continued to increase following the same procedure until its value reached the corresponding CHF value. To ensure the repeatability of the test, >5 samples were tested for each types of GF membranes.

[0124] Long-term evaporation stability test: For long-term stability test, the heat flux was incrementally increased from zero to the target heat flux of approximately 300 W / cm2within 30 minutes and then kept at this level for over 2 hours. The experiments were stopped right before the water in the reservoir was depleted. The water in the reservoir had an effective volume of about 0.6 - 0.8 L. At evaporation rate of 3.5x1 O’5kg / s corresponding to ~150 W of power (300 W / cm2over 0.49 cm2), the expected holding time is 4.2-5.7 hours.

[0125] Characterization of the membranes: The morphologies of the fiber membranes were examined using scanning electron microscopy (SEM) (Zeiss Sigma 500). Both the membranes before the evaporation test (panels (d) and (e) of Figure 1 ) and those after the long-term evaporation tests (Figure 9) were imaged to identify potential contaminants and cloggages in the membranes.

[0126] Modeling of the capillary limit of heat flux: The capillary lim it was calculated by balancing the pressure drop (Ap) due to the flow resistance across membranes with the capillary pressure (pcap). Based on Darcy’s law and the Laplace equation, the volumetric flux can be expressed as:where p is the viscosity and L is the distance of the fluid flow (or the membrane thickness), p.is liquid density, and Hfgis vaporization latent heat. The capillary limit of the heat flux is the volumetric flux multiplied by the evaporation latent heat.

[0127] Modeling of the kinetic limit of heat flux: The evaporation heat flux was calculated from a single pore by solving the meniscus profile (5) inside open pore area along the length ( / ) via the following governing equation (5):Equation (5) is derived from the following modified Young-Laplace equation Pv- Pi = Pc+ Pd. (6)In equation (6), the pressure difference between vapor (Pv) and liquid (Pi) is balanced at the liquid-vapor interface by both capillary (Pc) and disjoining Pd) pressures. By applying the boundary conditions, the evaporation mass flux can be calculated at different locations along the meniscus within a thin-film region following similar solving procedures. As a result, the local evaporation heat flux (q(x)) can be predicted. The heat flux over the thin-film region (x = 0-L) is defined by averaging the local heat flux via equation (7). Eventually, the overall thin-film evaporation heat flux (qave) over the entire open pore structure with pore size d was extracted by Equation 7 More details of the evaporation model can be found below:

[0128] Modeling of Single Pore Kinetic Evaporation Process: During the capillary- driven evaporation process, besides the liquid transport across the membranes, evaporation kinetic at the liquid-vapor interface is another relevant factor. To quantify the evaporation kinetic, the average heat flux over single pores may be calculated with various pore sizes. The profile of the liquid film in the pore may be modeled and then the local evaporation heat flux may be determined, which is then averaged over the entire thin-film regime to obtain the average heat flux, as shown by the dash-dotted line in panel (d) of Figure 7. The thin-film evaporation process over a single pore channel may be investigated at different pore sizes by numerically solving evaporation meniscus profile. Solving procedure of evaporation meniscus governing equation by combing HK equation for mass flux boundary conditions, and other boundary conditions (based on pore size, superheat, liquid properties, and vapor side pressure assumed 1 atm here).The code may be based on, e.g., the MATLAB ode45 package. Other software tool may be used.

[0129] Governing equation for thin-film profile: With respect to an evaporation thin- film profile, a modified Young-Laplace Equation, equation (8) below, may be used to relate the pressure difference between the vapor pressure and the liquid pressure at the liquid-vapor interface, which is due to both capillary and disjoining pressures:Pv- Pi = Pc+ Pd, (8) where Pvis the vapor pressure, Pi is the liquid pressure, and Pcand Pd are capillary and disjoin pressure, respectively. The disjoin pressure (Pd) can be expressed using equation (9) aswhere 8 is the film-thickness, and A is the dispersion constant. The capillary pressure can be formulated as the product of interfacial curvature K, and surface tension a of the fluid using equation (10):Pc= rr / <, (10) where K = 5"(1 + <5'2)-1-5, and 8' and 8" are the first and second derivatives of the film thickness with respect to length x. By combining equations (8), (9), and (10), and performing differentiation with respect to x, the thin-film profile (<5) can be derived as equation (11 ):

[0130] Equation (11 ) is a third-order differential equation, assuming vapor pressure Pvis uniform along the meniscus profile. The fluid flow may be considered to be steady, incompressible, and the fluid is Newtonian. The thin film may have a very large aspect ratio. It may also be a very low Reynolds number flow where viscous forces dominate. Hence, lubrication theory may be employed to obtain the pressure gradient. Using a noslip boundary condition at the wall and a no-shear boundary condition at the liquid-vapor interface, the mass flow rate '(x) and the liquid pressure gradient may relate as a function of the net evaporation flux from the initial position of the film to its final position, as equation (12):By substituting equation (12) into equation (11 ), and differentiating with respect to x, a fourth-order ordinary differential equation (13) may be obtained to compute the thin-film profile:where v is liquid kinetic viscosity. Wayner’s evaporation model may be used to calculate the evaporation flux (m").

[0131] Solving the ODE: A fourth-order Runge-Kutta method may be used to solve the fourth-order, non-linear ODE in equation (13). The problem can be treated as an initial-value problem, where the properties of the fluid may be defined at the beginning of the thin-film, at x = 0 and four initial conditions are employed, 5(0), 5'(0), 5"(0) and 6"' (0). The initial thickness at 5(0) may be set as the characteristic non-evaporating film thickness and can be calculated using equation (14):

[0132] 5'(0) is the initial slope of the film and is close to zero. A value of 5'(0) = le-11is used in the present study for illustration purposes and not intended to be limiting. 5"(0) may be determined by using a far-field boundary conditions where the slope of the film approaches the intrinsic meniscus radius, which is approximately half of the channel height when the liquid is completely wetting. The fourth condition of 5"'(0) may be obtained by solving equation (5) using the results of 5(0), 5'(0), 5"(0) coupled with equation (15): f*00m" dx = 0. (15)

[0133] Computing the heat transfer along the meniscus: Once the meniscus profile along the length has been obtained (5(x)), the local heat flux can be derived using equation (16):where B =klVl(Tw Tv)and C =k v'Jv2. To obtain thin-film region evaporation heat flux, a Ahfg Pv fg well-established rule as expressed by equation (17) to determine the thin-film region:

[0134] Equation (17) indicates that the evaporation thin-film region merges into the intrinsic meniscus region at x = L where the disjoining pressure (Pd) is sufficiently small compared to the value where the thin-film region originates. Therefore, the evaporation thin-film region may be defined within the length 0 < x < L and thickness 80< 8 < 8X. As a result, the local evaporation heat flux (q"(x) in W / cm2) may be obtained along the length of the entire thin-film region from equation (16). By integrating the heat flux over the thin-film region and normalizing by the entire open pore area, the averaged heat flux (qave) can be predicted via equations (18) and (19):where d is the pore diameter. The dash-dotted curve in panel (d) of Figure 7 shows the evaporation kinetic limits versus different pore sizes. The curve indicates that as pore size increases, the overall average heat flux decreases due to, e.g., a relatively small thin-film region compared to the total open pore area in equation (19).

[0135] Thin-film evaporation modeling results: By numerically solving the meniscus profile from the governing equation (13), the meniscus thickness (6) along the length can be visualized. Panel (a) of Figure 17 shows meniscus thickness versus length. As mean pore size increases, the curvature of the meniscus decreases which develops a longer meniscus region. Based on the equation (17), the thin-film region is specified in panels (b) and (d) of Figure 17. Due to the extended meniscus region, the length of the thin-film region may also increase as the mean pore size increases. As a result, the thin-film region length may fall between approximately 800 nm and approximately 1600 nm.

[0136] To quantify the heat transfer inside the open pore structure, panel (c) ofFigure 17 illustrates local heat flux along the film length. It is clearly shown that the heat flux significantly increases within the first hundred nanometers due to the small enoughthin-film meniscus thickness for efficient heat transfer. As illustrated, the local heat flux may reach its maximum value within 200 nm. Next, the local heat flux may dramatically drop to at least one order of magnitude smaller within a few hundred nanometers due to the increase of meniscus thickness which imposes an additional thermal resistance between the wall and liquid-vapor interface. Therefore, efficient heat transfer may occur within a thin-film evaporation region which may be selected as a reference kinetic limit shown in panel (d) in Figure 7. By solely analyzing thin-film region evaporation heat transfer, the integrated total thin-film flux via equation (18) may increase as pore size increases due to, e.g., the longer thin-film region length from the extended meniscus profile as shown in panel (d) of Figure 17. By normalizing of total open pore area to get an effective evaporation heat flux over the entire open pore structure, the overall heat transfer may be governed by the ratio between the thin film length ( / ) and the pore size (d) as shown in equation (19). However, the thin-film region length may be at least a few times smaller than the pore size as shown in panel (d) of Figure 17. Therefore, the overall open pore structure effective evaporation heat flux may decrease as pore size increases as plotted in Figure 7.

[0137] The cooling device 300 as disclosed herein may have high heat flux and / or high flexibility, due to the use of fiber membranes that may be manufactured with low cost and scalable. For example, a flexible cooling device (or referred to as an evaporator or vapor chamber) including a fiber membrane may be produced by bonding one or more nanofiber membranes onto Cu meshes (e.g., the substrate 310), unlike those in other vapor chambers that are made of a thick layer (> 1 mm) of sintered Cu powder and thus are not flexible and have large thermal resistance. The fiber membranes as disclosed herein may include a three-dimensional network of interconnected pores (e.g., nanopores) and large surface areas for evaporation with high heat flux rivaling that of dedicated ceramic membranes but are also scalable to manufacturing with the electrospinning process, flexible, low-cost, and durable.

[0138] Figure 18 depicts a schematic representation of a flexible vapor chamber (fVC) according to some embodiments of the present disclosure. Panel (a) illustrates a schematic diagram of the flexible vapor chamber (fVC) 1800. The f C 1800 may spread the high heat flux from the heat source 1810 (e.g., an electronic chip) via evaporationand vapor spreading. This process may involve low-resistance and near-isothermal vapor flow upon evaporation of a cooling medium 1830 from an evaporator 1820, leading to a reduction in heat flux by a factor of, e.g., more than 100 at the condenser 1840. This reduced heat flux may be manageable by a conventional heat sink.

[0139] The low-resistance vapor flow may be achieved, at least in part, by capillary force generated by pores in the fiber membrane of the evaporator 1820, in combination with a network of pathways for the vapor via the interconnected pores in the fiber membrane of the evaporator 1820. The effective lateral thermal conductivity of the vapor chamber 1800 may exceed 10,000 W / m-K (approximately 25 times that of copper).

[0140] When the heat from the heat source 1810 is spread from the evaporator 1820 to the condenser 1840 with a much larger area (more than 100 times), the heat flux may be reduced to a manageable level in the condenser 1840. This reduced heat flux may be cooled with a routine heat sink such as a fan or even natural convection without a fan. The condenser 1840 may include one or more features configured to enhance heat removal to a heat sink. Merely by way of example, the condenser 1840 may include fins 1850, as illustrated in panel (a) of Figure 18, to enhance heat removal to the heat sink. In some embodiments, heat from the vapor may be dissipated far away from the evaporator 1820 (e.g., in heat pipes). The condensed cooling medium 1830 may be then driven back to the evaporator 1820 by the capillary force of the porous wick made of the fiber membrane 1816 to form a closed loop. See also, e.g., Figure 3A and relevant description thereof.

[0141] Panel (b) of Figure 18 illustrates a schematic representation of the evaporator 1820. The evaporator 1820 may be produced using by bonding an electrospun nanofiber membrane 1816 (e.g., the fiber membrane 320) onto a Cu wire mesh 1812 (e.g., the substrate 310) as illustrated in panel (b) of Figure 18. The entire vapor chamber may include the fibrous membranes, the wire meshes in the evaporator 1820 and condenser 1840, and a casing (e.g., Cu / polymer laminated). The vapor chamber may be flexible and scalable to manufacture.

[0142] Some embodiments relate to the manufacture of the fiber membranes as disclosed herein. Merely for illustration purposes and not intended to be limiting, amethod is described for making a fiber membrane including ceramic nanofibers with high thermal conductivity, such as SiC (k > 100 W / m-K), TiO (k = 8 W / m-K), etc., with fiber diameter and pore size ranging from 100~200 nm to 500 nm using an electrospinning technique.

[0143] Figure 19 illustrates an electrospinning setup to fabricate nanofiber membranes according to some embodiments. To electrospin ceramic fibers, a ceramic precursor solution may be prepared. A TiO2 and SiC precursor solution may be made to fabricate TiCh and SiC nanofiber membranes, respectively. The titania precursor solution may include titanium teraisopropoxide, alcohol, high-molecular-weight poly(vinyl pyrrolidone), and acetic acid for hydrolysis control. The SiC precursor solution may include a polymer and the corresponding precursor such as polycarbosilane.

[0144] As shown in Figure 19, the precursor solution may be loaded into a syringe pump 1910 with a metallic needle 1915, which may be connected to a high-voltage supply 1920. A copper mesh 1930 may be placed a few centimeters (e.g., 5 centimeters, 10 centimeters, 15 centimeters) away from the needle 1915 as a fiber collector. After electrospinning, the nanofibers 1940 may be left in the air to allow the completion of the hydrolysis of the precursor. Calcination may be performed to remove the polymer content from the fibers 1940.

[0145] The resultant nanofiber membranes (including fibers 1940 bonded onto the mesh 1930) may be characterized for microstructures (using, e.g., scanning electron microscopy), material composition (e.g., by electron diffraction), and / or thermal conductivity (using, e.g., laser flash analysis or modulated photothermal radiometry).

[0146] The size of electrospun fibers and the morphology of the fibrous membrane may depend on the electrospinning parameters, including one or more of applied voltage, flow rate, the concentration of the electrospinning polymer solution, the distance between the syringe 1910 and the collector 1930. Accordingly, the manufacturing process may be adjusted to produce fiber membranes with desirable fiber sizes and pore sizes. For example, fiber and pore sizes may be tailored within ranges of 20 nm to 200 nm, 100-200 nm, or approximately 500 nm. These adjustments may be achieved by tuning one or more of the aforementioned parameters, e.g., the concentration of metal oxide in the precursor solution.

[0147] Figure 20 illustrates a setup for fabrication and testing of nanofiber membrane evaporators according to some embodiments. Panel (a) shows diffusion bonding via hot pressing to bond nanofiber membranes 2120 onto Cu meshes 2112 to make an evaporator 2000. The evaporator 2000 may be produced by bonding nanofiber membranes 2020 onto a substrate 2010. In some embodiments, the substrate 2010 may include a wire mesh 2012 and a base 2016. The wire mesh 2012 may form microchannels on the substrate 2010 configured to facilitate the heat transfer to a cooling medium. For example, the wire mesh 2012 may increase the surface area for heat transfer, e.g., when a cooling medium flows through the substrate 2010 and absorbs heat, via the substrate 2010, from a heat source in thermal communication with the substrate 2010. In some embodiments, the microchannels may be formed by micropillars on the base 2016. Micropillars may be fabricated by, e.g., laser milling.

[0148] A Cu wire mesh may be used because it is readily available with different wire and pitch sizes, and is flexible. Hydrophilic treatment of Cu meshes may be carried out to further increase capillary force by using an alkaline erosion method, where the water contact angle is near 0 degrees. The wire mesh 2012 may in turn be bonded to a solid base 2016, e.g., a thin solid Cu base. The wire mesh 2012 and the solid base 2016 may include a same material or different materials. Merely by way of example, a diffusion bonding process may be performed to achieve the bonding, in which the different layers (the fiber membrane 2020, the wire mesh 2012, and the solid base 2016) may be stacked together under a certain uniaxial pressure, and the entire setup may be placed in a vacuum or Ar-filled furnace for heating. The bonding of the membranes 2020 and the Cu meshes 2012 and bases 2016 may be carried out in a hot-pressing system under high uniaxial pressure (e.g., from a hydraulic pressure, up to, e.g., 30 MPa), temperature, and in an inert Ar or vacuum environment.

[0149] In some embodiments, the thermal performance of the evaporator 2000 using an open-loop setup as shown in panel (b) of Figure 20 illustrating a thermos-fluid testing. The liquid reservoir 2050 may be located outside of the evaporator 2000. The heater 2030 as the heating source may be positioned underneath the evaporator 2000. This setup may mimic a component layout of a real electronic device. In this testing, the liquid (e.g., water) may be fed through Cu meshes 2012 (e.g., from the edges asil lustrated by the two arrows in panel (b)) and to the nanofiber membrane 2020 by capillary pumping. Heat may be supplied from the bottom and conducted to the top of the membrane 2020 where evaporation occurs, hence high thermal conductivity of the membrane is needed. This is different from the joule heating in Pt thin film deposited on top of the membranes as illustrated in the setup in panel (a) of Figure 7. The temperature and heat flux between the heater 2030 and the Cu base 2016 may be monitored with a temperature monitor 2040. Merely by way of example, the temperature monitor 2040 may include a thin film heat flux sensor and thermocouples. The heat flux vs. superheat under different membrane and mesh parameters (e.g., membrane material, pore size & thickness, permeability, and Cu mesh size and porosity) may be tested and used to guide future design and optimization to achieve desirable functions or parameters, e.g., higher CHF over 500 W / cm2. One or more mechanical parameter of the evaporator 2000, or a portion thereof (e.g., the fiber membrane 2020, the substrate 2010) may be tested. For example, the bending modulus, flexural strength, and fatigue property of the evaporator 2000 may be tested at different temperatures (room temperature to 100 °C) to assess its flexibility, durability, and strength using an Instron universal testing machine.

[0150] Figure 21 illustrates a schematic of a process to make a hermetically sealed flexible vapor chamber (fVC) according to some embodiments. The fVC 2100 may include an evaporator, a condenser, and a casing. The process may include the vapor chamber fabrication and working fluid charging process. The evaporator 2000 may be made as illustrated in Figure 20. The condenser 2140 may include a wire mesh 2142 (e.g., a Cu wire mesh), with surface treated to enhance condenser performance. Then, the condenser block 2140 may be bonded onto the evaporator side block 2000 by, e.g., soldering / brazing in an enclosed environment with a desired bonding temperature in, e.g., an environmentally controlled oven. After sealing the vapor chamber block 2000, a working fluid (e.g., water) may be injected into the vapor chamber 2100 with a target partial pressure (and hence the saturation temperature) through a filling tube, followed by the sealing of the filling tube within a casing 2160.

[0151] The manufacturing of the fVC 2100 may be done by improving or optimizing the bonding between the condenser and evaporator with hermetic sealing: the temperature, pressure, and duration of the diffusion bonding process will be optimized.

[0152] The fabricated fVC 2100 may be tested with a high-power electronic device with heat flux over 500 W / cm2(such as a gaming GPU or a laser). The flexibility and durability of the fVC 2100 may be tested with accelerated bending and thermal cycling tests.

[0153] The fVC 2100 or other cooling device 300 as disclosed herein may be used to satisfy the cooling needs for laptops, high-performance computers (gaming and graphicintensive computers, data centers and supercomputers), optoelectronic devices (LEDs, lasers), electrical vehicles and batteries, inverters and other power electronics, motors, garments, personal protection equipment, mobile devices, compact devices, foldable devices, etc.

[0154] Figure 22 illustrates a schematic of a process to fabricate a flexible vapor chamber according to some embodiments. Panel (a) shows the fabrication of a condenser with pre-cleaned Cu substrates and mesh with certain mesh size. The Cu mesh may be sintered with Cu substrates by hot pressing in a vacuum ambient. Post surface treatment may be applied to enhance condenser performance.

[0155] Panel (b) shows the condenser block produced in (a) may be bonded onto an evaporator side block (e.g., evaporator 2000) by soldering / brazing procedure at desired bonding temperature in an environmentally controlled oven.

[0156] Panel (c) shows after the vapor chamber block is sealed as illustrated in (b), a working fluid may be injected via a filling tube into the sealed vapor chamber block at a controlling inject rate and the chamber vacuum degree with desired filling ratio and the ambient pressure. A heating degassing process may be performed to remove noncondensable gas inside the chamber. A welding process may be performed to seal the filling tube.

[0157] Figure 23 shows an example fabrication process to make an active evaporation cooling device (corresponding to Figure 3B). First, as illustrated in panel (a), a shell housing unit 2305 with fluidic manifolds and a microchannel substrate 2310 are fabricated, e.g., using micromachining. The material can be Cu. The substrate 2310has microchannels 2312 supported on a base 2316. Second, as illustrated in panel (b), a fiber membrane 2320 is bonded to the microchannels 2312 by, e.g., soldering, sintering. The bonding may be performed, e.g., in an inert environment with a compressive pressure applied to the membrane and the microchannel. As illustrated in panel (c), after the bonding, the shell housing 2305 and the microchannel substrate 2310 can be bonded using a similar process as in (b) including, e.g., soldering, sintering, in an inert environment with a compressive pressure. The manifolds can then be connected to a fluid conduit (e.g., conduit 350 in device 300B in Figure 3B) in an active evaporation cooling system.

[0158] Figure 24 shows a schematic illustration of an example active cooling device working in a closed loop system, similar to Figure 3B. A cooling medium flows within a fiber membrane 2420 from a first side toward a second side of the cooling device 2400. During the flowing, the cooling medium may be vaporized by heat from the heat source 100 via the substrate 2410. The substrate 2410 includes microchannels to facilitate the heat transfer. The heat source 100 is in thermal communication with the cooling device 2400 on the first side. Heat is removed by condensing the vaporized cooling medium by a condenser 2440 on the second side of the cooling device 2400. The removed heat is transferred to a heat sink, e.g., a secondary cooling system or the ambient environment. The condensed cooling medium flows towards the first side of the cooling device 2400 for reuse. The flow may be driven by a pump 2460. In some embodiments, the condensed cooling medium may be pressurized by a compressor 2470. The compressor 2470 may be positioned upstream (as illustrated in Figure 24) or downstream of the pump 2460.

[0159] Figure 25 shows evaporation cooling on a fiber membrane 2520 with a hydrogel coating on the fibers. Panel (a) shows a schematic of the fiber membrane 2520. The fiber membrane 2520 is substantially the same as or similar to those shown elsewhere in the present document (such as panels (d-e) in Figure 1 ), but with the fibers 2522 coated with a thin layer of hydrogel 2526. The hydrogel coating can be of thickness ranging from 100 nanometers to 10 micrometers. The role of the hydrogel coating is to promote the hydrophilicity and capillary pumping of the membranes and to suppress bubble formation or boiling in the liquid phase of the cooling medium.

[0160] Panel (b) of Figure 25 shows experimental results of a fiber membrane with a mean pore size of 6.1 micrometers (substantially the same as or similar to sample D in Figure 7) with hydrogel coating (curve I) and without hydrogel coating (curve II). The sample with the hydrogel coating shows evaporation at a much higher superheat without boiling or bubble formation, indicating that the hydrogel coating promotes evaporation and suppresses bubble formation due to its strong affinity to water.

[0161] Figure 26 shows characterization of hydrogel-coated fiber membrane, (a) and (b) are top view SEM images, (c) is a zoomed-in SEM image showing fibers with coating, (d) is an EDAX image showing the uniformly distributed carbon element in the hydrogel.

[0162] Figure 27 shows a schematic illustration of an evaporator with hydrogel coating. The evaporator is made of fiber membrane 2520 bonded onto a microchannel, the same as 1816 in Figure 18 and 2000 in Figure 20, but with hydrogel coating 2526 on the fibers 2522 as illustrated in Figure 25 in the fiber membranes 2520 and hydrogel coating 2716 on the channel walls of the microchannels 2712. The role of the hydrogel coating 2526 and 2716 is to promote hydrophilicity and capillary pumping and to suppress bubble formation or boiling.

[0163] As an example, the coating of the hydrogel can be done via dip coating of a substrate in the monomer solution prior to the polymerization step. A thin hydrogel coating will be formed on the surface of the fibers upon the polymerization. The surface of the fibers will be treated with a functional silane solution [e.g., 3-(trimethoxysilyl) propyl methacrylate] to form strong covalent bonds with the hydrogel (Figure 28). The thickness of the gel coating will be controlled by the viscosity of the monomer solution and the duration and number of dip coating processes.

[0164] As an example, the synthesis of the hydrogel can be done through free- radical polymerization of monomers using a thermal-initiated method with ammonium persulfate (APS) as an initiator. Examples of monomers include sodium acrylate (SA) and vinyl alcohol (VA). The monomers are then dissolved in deionized water to obtain a homogenous solution. Afterwards, N,N'-methylenebisacrylamide (MBA) crosslinker and APS initiator are added into the solution with a fixed molar ratio of monomers, crosslinker and initiator, such as 50:1 :1. In some embodiments, sodium alginate (Alg)can be added to create a double-network structure for enhanced toughness. After being fully dissolved, the homogeneous solution is degassed for 10 min before being put into the oven of 70 °C, where the hydrogel is copolymerized for 10 min. The hydrogel is kept in the oven overnight to ensure complete co-polymerization.

[0165] The above description and drawings are illustrative and are not to be construed as limiting. Numerous specific details are described to provide a thorough understanding of the disclosure. However, in some instances, well-known details are not described in order to avoid obscuring the description. Further, various modifications may be made without deviating from the scope of the embodiments.

[0166] The use of examples anywhere in this specification including examples of any term discussed herein is illustrative only, and is not intended to further limit the scope and meaning of the disclosure or of any exemplified term. Likewise, the disclosure is not limited to various embodiments given in this specification.

[0167] While this patent document contains many specifics, these should not be construed as limitations on the scope of an invention that is claimed or of what may be claimed, but rather as descriptions of features specific to particular embodiments. Certain features that are described in this document in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a sub-combination or a variation of a sub-combination. Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results.

[0168] Only a few examples and implementations are disclosed. Variations, modifications, and enhancements to the described examples and implementations and other implementations can be made based on what is disclosed.EXAMPLES

[0169] Example 1 . A cooling device, comprising a fiber membrane comprising a plurality of fibers forming a three-dimensional network of pores, wherein: a first side of the cooling device is in thermal communication with a heat source, a second side of the cooling device is in thermal communication with a heat sink, the fiber membrane is positioned between the first side and the second side, and the three-dimensional network of pores allows a cooling medium to flow through and transfer heat from the heat source to the heat sink.

[0170] Example 2. The cooling device of any one or more examples disclosed herein, further comprising a substrate on which the fiber membrane is supported.

[0171] Example 3. The cooling device of any one or more examples disclosed herein, wherein: the substrate is located between the heat source and the fiber membrane; and the fiber membrane is in thermal communication with the heat source via the substrate.

[0172] Example 4. The cooling device of any one or more examples disclosed herein, wherein the substrate includes a thermally conductive material.

[0173] Example 5. The cooling device of any one or more examples disclosed herein, wherein the substrate comprises microchannels configured to facilitate the cooling medium’s heat absorption from the heat source. In some embodiments, at least a portion of the substrate, e.g., at least a portion of the microchannels, includes a hydrogel coating.

[0174] Example 6. The cooling device of any one or more examples disclosed herein, wherein the fiber membrane has a mean pore size smaller than 100 micrometers.

[0175] Example 7. The cooling device of any one or more examples disclosed herein, wherein the fiber membrane has a mean pore size in a range between 100 nanometers and 20 micrometers.

[0176] Example 8. The cooling device of any one or more examples disclosed herein, wherein the fiber membrane has a thickness less than 10 micrometers.

[0177] Example 9. The cooling device of any one or more examples disclosed herein, wherein at least a portion of the plurality of fibers of the fiber membrane include a thermally conductive material. The coating may have a different material than the fibers of the fiber membrane.

[0178] Example 10. The cooling device of any one or more examples disclosed herein, wherein at least a portion of the plurality of fibers of the fiber membrane include a glass, ceramic, polymer, metal, or an alloy material.

[0179] Example 11 . The cooling device of any one or more examples disclosed herein, wherein at least a portion of the plurality of fibers include a coating of a thermally conductive material.

[0180] Example 12. The cooling device of any one or more examples disclosed herein, wherein at least a portion of the plurality of fibers include a coating of hydrogel.

[0181] Example 13. The cooling device of any one or more examples disclosed herein , wherein the cooling medium comprises water, glycol, Fluorinert electronic liquid, or a refrigerant.

[0182] Example 14. The cooling device of any one or more examples disclosed herein, wherein at least a portion of the cooling medium is allowed to undergo a first phase change from a first phase to a second phase by absorbing heat from the heat source.

[0183] Example 15. The cooling device of any one or more examples disclosed herein, wherein the first phase is a liquid phase, and the second phase is a vapor phase.

[0184] Example 16. The cooling device of any one or more examples disclosed herein, wherein the heated cooling medium is configured to flow from the first sidetoward the second side through the fiber membrane driven at least in part by capillary force generated by the pores between the fibers of the fiber membrane.

[0185] Example 17. The cooling device of any one or more examples disclosed herein, wherein: the heat sink comprises an ambient environment of the cooling device.

[0186] Example 18. The cooling device of any one or more examples disclosed herein, wherein at least a portion of the cooling medium is allowed to undergo a second phase change from the second phase to the first phase by transferring heat to the heat sink.

[0187] Example 19. The cooling device of any one or more examples disclosed herein, further comprising a cooling medium conduit through which the cooling medium flows from the second side of the cooling device toward the first side of the cooling device. In some embodiments, the cooling medium conduit comprises a porous wicking structure.

[0188] Example 20. The cooling device of any one or more examples disclosed herein, wherein the cooling medium conduit and the fiber membrane constitute part of a closed cooling medium circuit through which the cooling medium circulates within the cooling device.

[0189] Example 21 . The cooling device of any one or more examples disclosed herein, further comprising a casing within which the cooling medium circulates.

[0190] Example 22. The cooling device of any one or more examples disclosed herein, further comprising a pump configured to drive the cooling medium to flow within the cooling device.

[0191] Example 23. The cooling device of any one or more examples disclosed herein, wherein the heat source includes an electronic device, or a portion thereof.

[0192] Example 24. The cooling device of any one or more examples disclosed herein, wherein the cooling device is a garment, protection equipment, or a portion thereof.

[0193] Example 25. The cooling device of any one or more examples disclosed herein, further comprising a condenser configured to condense the cooling medium by facilitating heat transfer from the cooling medium to the heat sink.

[0194] Example 26. The cooling device of any one or more examples disclosed herein, further comprising a compressor configured to increase pressure of the condensed cooling medium.

[0195] Example 27. The cooling device of any one or more examples disclosed herein, wherein the heat sink comprises an ambient environment of the cooling device or a secondary cooling system.

[0196] Example 28. A cooling method, comprising: flowing a cooling medium through a fiber membrane of a cooling device from a first side toward a second side of the cooling device; vaporizing, during the flowing, the cooling medium in the fiber membrane by absorbing heat from a heat source that is in thermal contact with the cooling device on the first side; and removing heat from the cooling medium to a heat sink that is in thermal contact with the cooling device on the second side .

[0197] Example 29. The cooling method of any one or more examples disclosed herein, further comprising: condensing the vaporized cooling medium; and circulating, via a cooling medium conduit, the condensed cooling medium toward the first side of the cooling device.

[0198] Example 30. The cooling method of any one or more examples disclosed herein, wherein circulating the condensed cooling medium toward the first side of the cooling device comprises: driving, using a pump, the condensed cooling medium from the second side toward the first side of the cooling device via a cooling medium conduit of the cooling device; or pulling the cooling medium from the first side toward the second side of the cooling device via the fiber membrane.

[0199] Example 31 . The cooling method of any one or more examples disclosed herein, further comprising: pressurizing the condensed cooling medium by compressing the condensed cooling medium.

[0200] Example 32. The cooling method of any one or more examples disclosed herein, wherein the heat sink is an ambient environment or a secondary cooling system.

[0201] Example 33. A wearable object comprising a cooling device of any one or more examples disclosed herein.

[0202] Example 34. The wearable object of any one or more examples disclosed herein, wherein the wearable object comprises a garment, protection equipment, or a portion thereof.

[0203] Example 35. An electronic device comprising: a heat generating component, and a cooling device of any one or more examples disclosed herein, the cooling device being configured to cool the heat generating component.

[0204] Example 36. The electronic device of any one or more examples disclosed herein, wherein the heat generating component comprises at least one of a semiconductor chip or an antenna.

Claims

CLAIMSWhat is claimed is:1 . A cooling device, comprising: a fiber membrane comprising a plurality of fibers forming a three-dimensional network of pores, wherein: a first side of the cooling device is in thermal communication with a heat source, a second side of the cooling device is in thermal communication with a heat sink, the fiber membrane is positioned between the first side and the second side, and the three-dimensional network of pores allows a cooling medium to flow through and transfer heat from the heat source to the heat sink.

2. The cooling device of claim 1 , further comprising a substrate on which the fiber membrane is supported.

3. The cooling device of claim 2, wherein: the substrate is located between the heat source and the fiber membrane; and the fiber membrane is in thermal communication with the heat source via the substrate.

4. The cooling device of claim 2, wherein the substrate includes a thermally conductive material.

5. The cooling device of claim 2, wherein the substrate comprises microchannels configured to facilitate the cooling medium’s heat absorption from the heat source.

6. The cooling device of claim 2, wherein at least a portion of the substrate includes a hydrogel coating.

7. The cooling device of claim 1 , wherein: the fiber membrane has a mean pore size in a range between 100 nanometers and 20 micrometers, or the fiber membrane has a thickness less than 10 micrometers.

8. The cooling device of claim 1 , wherein the plurality of fibers of the fiber membrane includes a first thermally conductive material, or a coating of a second thermally conductive material.

9. The cooling device of claim 1 , wherein: the plurality of fibers of the fiber membrane includes a glass, ceramic, polymer, metal, or an alloy material, or the plurality of fibers includes a coating of hydrogel.

10. The cooling device of claim 1 , wherein the cooling medium comprises water, glycol, Fluorinert electronic liquid, or a refrigerant.11 . The cooling device of claim 1 , wherein at least a portion of the cooling medium is allowed to undergo a first phase change from a first phase to a second phase by absorbing heat from the heat source.

12. The cooling device of claim 11 , wherein the first phase is a liquid phase, and the second phase is a vapor phase.

13. The cooling device of claim 11 , wherein the heated cooling medium is configured to flow from the first side toward the second side through the fiber membrane driven at least in part by capillary force generated by the pores between the fibers of the fiber membrane.

14. The cooling device of claim 11 , wherein at least a portion of the cooling medium is allowed to undergo a second phase change from the second phase to the first phase by transferring heat to the heat sink.

15. The cooling device of claim 1 , further comprising a cooling medium conduit through which the cooling medium flows from the second side of the cooling device toward the first side of the cooling device.

16. The cooling device of claim 15, wherein the cooling medium conduit and the fiber membrane constitute part of a closed cooling medium circuit through which the cooling medium circulates within the cooling device.

17. The cooling device of claim 1 , further comprising a pump configured to drive the cooling medium to flow within the cooling device.

18. The cooling device of claim 1 , wherein the heat source includes an electronic device, or a portion thereof.

19. The cooling device of claim 1 , wherein the cooling device is a garment, personal protection equipment, or a portion thereof.

20. The cooling device of claim 1 , further comprising: a condenser configured to condense the cooling medium by facilitating heat transfer from the cooling medium to the heat sink, or a compressor configured to increase pressure of the condensed cooling medium.21 . A cooling method, comprising: flowing a cooling medium through a fiber membrane of a cooling device from a first side toward a second side of the cooling device;vaporizing, during the flowing, the cooling medium in the fiber membrane by absorbing heat from a heat source that is in thermal contact with the cooling device on the first side; and removing heat from the cooling medium to a heat sink that is in thermal communication with the cooling device on the second side.

22. The cooling method of claim 21 , further comprising: condensing the vaporized cooling medium; and circulating, via a cooling medium conduit, the condensed cooling medium toward the first side of the cooling device.

23. The cooling method of claim 22, wherein circulating the condensed cooling medium toward the first side of the cooling device comprises: driving, using a pump, the condensed cooling medium from the second side toward the first side of the cooling device via a cooling medium conduit of the cooling device; or pulling the cooling medium from the first side toward the second side of the cooling device via the fiber membrane.

Images