BACKGROUND
As more calculations are performed on a chip (e.g., a semiconductor chip used in a computer) and such a chip is fabricated with a relatively high degree of integration, an amount of heat per unit area (Power Intensity) or per unit volume (Power Density) is increasing. As the power density increases, a cooling apparatus as well as a cooling method for the chip have been developed to increase cooling efficiency.
SUMMARY
Embodiments of the present disclosure relate to a cooling apparatus, a semiconductor device including the apparatus, and a cooling method capable of cooling heat generated in the chip. More specifically, these embodiments relate to a semiconductor package liquid cooling system structure, material, and cooling method that reduces the thermal resistance of a heat path, increases the flow rate of cooling fluid in a region where the amount of generated heat is relatively large, and reduces leakage and stress.
In an embodiment, a structure may include an inlet hole through which a coolant fluid flows into the structure, an outlet hole through which the coolant fluid flows out from the structure, a plurality of inlet subchannels coupled between the inlet hole and a microchannel structure, and a plurality of outlet subchannels coupled between the microchannel structure and the outlet hole. Each of the inlet subchannels includes a first portion expanding as the coolant fluid flows therethrough, and each of the outlet subchannels includes a second portion contracting as the coolant fluid flows therethrough.
In an embodiment, a cooling apparatus may include a microchannel structure including a plurality of microchannels and a manifold structure disposed over the plurality of microchannels. The manifold structure may include an inlet hole through which a coolant fluid flows into the manifold structure, an outlet hole through which the coolant fluid flows out from the manifold structure, and a plurality of inlet subchannels coupled between the inlet hole and the microchannel structure, each of the inlet subchannels including a first portion that expands as the coolant fluid flows therethrough, and a plurality of outlet subchannels coupled between the microchannel structure and the outlet hole, each of the plurality of outlet subchannels including a second portion that contracts as the coolant fluid flows therethrough.
In an embodiment, a semiconductor device may include a chip and a cooling apparatus disposed over the chip and dissipating heat generated in the chip during an operation of the chip. The cooling apparatus may include a microchannel structure and a manifold structure disposed over the microchannel structure including a plurality of microchannels. The manifold structure may include an inlet hole through which a coolant fluid flows into the manifold structure, an outlet hole through which the coolant fluid flows out from the manifold structure, and a plurality of inlet subchannels coupled between the inlet hole and the microchannel structure, each of the inlet subchannels including a first portion that expands as the coolant fluid flows therethrough, and a plurality of outlet subchannels coupled between the microchannel structure and the outlet hole, each of the plurality of outlet subchannels including a second portion that contracts as the coolant fluid flows therethrough.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1A illustrates a semiconductor device according to an embodiment. FIG. 1B illustrates an enlarged view of a region including a portion of a cover and a manifold of the semiconductor device of FIG. 1A, according to an embodiment. FIG. 1C illustrates a semiconductor device according to another embodiment.
FIG. 2 illustrates a cooling system including a semiconductor device and an external pump disposed outside the semiconductor device, according to an embodiment.
FIG. 3 illustrates a semiconductor device according to an embodiment.
FIG. 4A illustrates a semiconductor device without an intervening layer according to an embodiment. FIG. 4B illustrates a semiconductor device obtained by reducing (or thinning) a thickness of a base and a thickness of a chip of the semiconductor device in FIG. 4A, according to an embodiment. FIG. 4C illustrates a semiconductor device including a base and a back portion of the chip doped with one or more impurities to increase thermal conductivity, according to an embodiment.
FIGS. 5A, 5B, 5C, and 5D illustrate a cooling apparatus including a manifold and a microchannel structure that has a plurality of microchannels, according to an embodiment.
FIGS. 6A, 6B, and 6C illustrate a cooling apparatus including a manifold and a microchannel structure that has a plurality of microchannels, according to another embodiment.
FIG. 7 illustrates a process of operating a flow distribution device according to an embodiment.
FIGS. 8A, 8B, 8C, 8D, and 8E illustrates microchannel structures according to embodiments.
FIGS. 9A and 9B show microchannel structures and manufacturing methods thereof according to embodiments.
FIGS. 10 and 10B show microchannel structures and manufacturing methods thereof according to embodiments.
FIGS. 11A, 11B, 11C, 11D, 11E, 11F, 11G, 11H, and 11I illustrate structures of a manifold according to embodiments.
FIG. 12 illustrates a three-dimensional view of a portion of a cooling apparatus according to an embodiment.
FIGS. 13A, 13B, 13C, 13D, and 13E each illustrate a structure including a manifold according to an embodiment.
FIGS. 14A, 14B, and 14C each illustrate a three-dimensional view of a component of a manifold structure according to an embodiment.
FIG. 15 illustrates a portion of a semiconductor device including a microchannel structure, a heat spreading layer, and an insulating layer, according to an embodiment.
DETAILED DESCRIPTION
Embodiments of the present application relate to a photodiode device, a photodetector including the photodiode device, and a method of forming the photodiode device.
A detailed description of embodiments is provided below along with accompanying figures. The scope of this disclosure is limited by the claims and encompasses numerous alternatives, modifications and equivalents. Although steps of various processes are presented in a given order, embodiments are not necessarily limited to being performed in the listed order. In some embodiments, certain operations may be performed simultaneously, in an order other than the described order, or not performed at all.
Numerous specific details are set forth in the following description. These details are provided to promote a thorough understanding of the scope of this disclosure by way of specific examples, and embodiments may be practiced according to the claims without some of these specific details. Accordingly, the specific embodiments of this disclosure are illustrative, and are not intended to be exclusive or limiting. For the purpose of clarity, technical material that is known in the technical fields related to this disclosure has not been described in detail so that the disclosure is not unnecessarily obscured.
FIG. 1A shows a semiconductor device 102A including a microchannel structure 128 and a chip (or a die) 120 that are in direct contact without an intervening layer (e.g., Thermal Interface Material), according to an embodiment. Specifically, such a device 102A may include a cooling apparatus 106A and a chip 120 that directly contacts a bottom surface of the cooling apparatus 106A, and the cooling apparatus 106A includes a manifold 126 and the microchannel structure 128, the microchannel structure 128 having a plurality of microchannels 124, and a base 122. For example, each of the plurality of microchannels 124 may be defined by a pair of adjacent fins and a portion of the base 122 disposed between the adjacent fins.
In order to minimize the thermal resistance in a cooling path between a coolant flow through the microchannels 124 and the chip 120 with a heat source, a structure for directly bonding the cooling apparatus 106A to the chip 120 is shown in FIG. 1A. In an embodiment, the bonding process temperature is not higher than the melting temperature of the already assembled material. Specifically, the bonding process temperature may not be higher than melting temperatures of materials included in the cooling apparatus 106A and the chip 120. For example, direct bonding between the cooing apparatus 106 to the chip 120 may be performed using one or more of the following methods:
- (1) Fusion Bonding/Direct or molecular bonding;
- (2) Cu—Cu/Oxide Hybrid Bonding at a room temperature (RT); and
- (3) Anodic Bonding.
In an embodiment, align marks may be placed on the backside of the chip 120 as well as the bottom of the microchannel 124 or the bottom of the base 122 to enable bonding not only of the entire wafer but also of individual unit microchannels or chips.
FIG. 1A shows the embodiment in which the cooling apparatus 106A is directly bonded to chip 120. In the absence of the thermal resistances associated with an intervening layer (e.g., the TIM) itself as well as contact surfaces between the TIM and the cooling apparatus 106A and between the TIM and the chip 120 in a conventional cooling system, a total thermal resistance in a cooling system according to an embodiment of the present disclosure can be reduced compared to that of the conventional cooling system.
Moreover, it may be desirable to increase the sealing force so that the liquid or vapor does not substantially leak even after using the liquid cooling device for a relatively long time. In an embodiment, when a mechanical seal (e.g., an O-ring) 130 is used as shown in FIG. 1A, the mechanical seal 130 can allow horizontal deformation in temperature-dependent deformation and maintain the vertical sealing force of a connecting part. Specifically, a cover (e.g., a lid) 140 may be connected to the substrate/PCB 104 with adhesive. Because the lid 140 and the substrate 104 may be relatively strongly bonded, the lid 140 and the manifold 126 can be closely adhered to each other with the O-ring 130 therebetween. Specifically, when an opening of the lid 140 may be coupled to a first portion (e.g., an upper portion) of the manifold 126 and a second portion (e.g., a lower portion) of the manifold 126 may be coupled to a plurality of microchannels 124, the O-ring 130 may be disposed between the lid 140 and the upper portion of the manifold 126 to substantially prevent liquid, gas, or both from leaking. In the embodiment shown in the cross-sectional view of FIG. 1B, two O-rings 130 may be disposed over a top surface of the manifold 126, such that each of the O-rings 130 may be inserted into a recess formed on a bottom surface of the cover 140. The recess may have a closed-loop shape (e.g., a substantially rectangular shape) when seen in a top view.
As another method to reduce the thermal resistance between the cooling apparatus and the chip, a plurality of microchannels may be formed on a back portion of the chip. As shown in the embodiment of FIG. 1C, a plurality of microchannels 124B are formed in a back portion of a chip 120B through semiconductor process to secure a relatively large height of the microchannels 124B. In an embodiment, the minimum thermal resistance can be secured by having the minimum chip thickness and maximum height of the microchannel in the chip 120B. For example, the height of each of the microchannels 124B may be sufficiently large to increase its aspect ratio (e.g., in a range from about 1 to about 60) and surface area for minimizing the thermal resistance and sufficiently short to ensure a proper operation of the chip 120B and the structural integrity of the chip 120B. In other embodiments (e.g., embodiments shown in FIGS. 9A and 9B), a plurality of plates including microchannels may be stacked to form combined microchannels each having an aspect ratio higher than that of the microchannel 124B of FIG. 0.1C.
FIG. 2 shows a cooling system 200 including a semiconductor device 202 (e.g., the semiconductor device 102A in FIG. 1A) and an external pump 250 disposed outside the semiconductor device 202, according to an embodiment.
FIG. 3 illustrates a semiconductor device 302 according to an embodiment. The semiconductor device 302 in FIG. 3 includes elements similar to those of the semiconductor device 102A in FIG. 1, 102B in FIG. 1C, or 202 in FIG. 2, and thus detailed descriptions on these elements may be omitted in the interest of brevity.
The embodiment shown in FIG. 3 differs from the embodiments shown in FIGS. 1A, 1C, and 2 in that the semiconductor device 302 according to the embodiment of FIG. 3 includes a pump 350 inserted between a first port (e.g., an inlet port) 332 and a cooling apparatus 306, the cooling apparatus 306 including a manifold 326. For example, the semiconductor device 302 may include a first cover 340 coupled to a substrate 304 to cover the cooling apparatus 306 and the chip 320, and a second cover 360 disposed over the first cover 340 and including the inlet port 332. The pump 350 may be positioned between the inlet port 332 of the second cover 360 and an opening of the first cover 340 to which the manifold 326 of the cooling apparatus 306 is coupled, thereby directing a coolant flow from the inlet port 332 of the second cover 360 to the manifold 326 through the opening of the first cover 340. In the semiconductor device 302 of FIG. 3, the pump 350 is built-in within each of a plurality of semiconductor devices 302, thereby precisely controlling the flow rate into each of the semiconductor devices 302 and facilitating installation of the pump compared to when a pump is installed outside a corresponding one of the semiconductor devices.
FIG. 4A illustrates a semiconductor device 402A without an intervening layer (e.g., Thermal Interface Material) suitable for use as the semiconductor device 102A shown in FIG. 1A. FIGS. 4B and 4C illustrate semiconductor devices 402B and 402C each having a further reduced thermal resistance between a coolant (e.g., liquid) flowing through a plurality of microchannels and a chip, compared to that of the semiconductor device 402A shown in FIG. 4A, according to embodiments. As a measure to reduce the thermal resistance between the liquid and the chip, it is desirable to make the distance between the liquid and the chip close as possible and to increase the thermal conductivity of each of one or more materials therebetween. Specifically, FIG. 4B illustrates the semiconductor device 402B obtained by reducing (or thinning) a thickness of a base 422A and a thickness of a chip 420A of the semiconductor device 402A in FIG. 4A. FIG. 4C illustrates the semiconductor device 402C including a base 422C and a back portion of the chip 420C that are doped with one or more impurities to increase thermal conductivity.
Referring to FIG. 4B, thinning can be applied to the base 422A and a back portion of the chip 420A of the semiconductor device 402A in FIG. 4A together. However, embodiments of the present disclosure are not limited thereto, and it can be applied to either the base 422A or the back portion of the chip 420A. In an embodiment, the reduced thickness of the base may be sufficiently small in order to minimize the distance between the liquid and the chip, or the semiconductor chip may be ground to have a given thickness, or both. For example, the thickness of the base may be substantially equal to zero, such that a plurality of fins defining the plurality of microchannels 424 are directly bonded to a surface (e.g., a top surface) of the chip 420B to make a coolant directly contact the surface of the chip 420B while flowing through the plurality of microchannels 424. The given thickness of the chip 420B may be determined based on the strength of structure. Specifically, the thickness of the chip 420B may be sufficiently large to substantially prevent an occurrence of fracture during bonding process, and sufficiently small to minimize the thermal resistance associated therewith. For example, the thickness of the chip may be in a range from 3 μm to 10 μm.
Also, to increase the thermal conductivity of the material, doping with one or more impurity elements (e.g., Boron Arsenide or Boron Phosphorous) may be performed on a base, or a portion (e.g., a back portion) of a chip, or both, thereby making at least a portion of the base, or a portion of the Chip, or both include the impurity elements. Referring to FIG. 4C, the base 422C and the back portion of the chip 420C of the semiconductor device 402C are doped with impurities. In an embodiment, doping may be performed by diffusion and implantation of Boron Arsenide or Boron Phosphor. For example, a diffusion source in solid, liquid, or gaseous state can be brought into contact with silicon included in the base and the chip, and diffused at a relatively high temperature, or impurities can be injected into the base and chip backside and then diffused at a relatively high temperature. Depending on the diffusion thickness and desirable concentration, one or more diffusion methods may be selected. In addition, a method of selectively diffusing one or more necessary parts can be performed. Doping can be applied to the base and the back portion of the chip together, or can be applied to either the base or the back portion of the chip. In an embodiment, doping may be performed to diffuse both into the microchannel base and the chip backside to reduce thermal resistances associated with the microchannel base and the chip together. However, it may be desirable to consider the manufacturing cost and time, check whether the diffusion in a direction from the chip backside to the chip front side does not significantly deteriorate the reliability of the chip, and select the target and depth for diffusion. For example, a depth for diffusion into the chip may be in a given range to sufficiently increase the thermal conductivity of the chip and to substantially prevent one or more operations of circuit elements implemented in the chip from being significantly interrupted. In an embodiment, diffusion may occur in not only the base of the microchannel structure, but also in one or more fins of the microchannel structure.
Although thinning and doping are separately applied to the embodiments shown in FIGS. 4B and 4C, embodiments of the present disclosure are not limited thereto. In another embodiment, thinning and doping may be used in combination. For example, thinning and doping may be applied together to the base, or the back portion of the chip, or both.
FIGS. 5A to 5D illustrate a cooling apparatus 506 including a manifold 526 and a microchannel structure 528 that has a plurality of microchannels 524 according to an embodiment. Specifically, FIG. 5C illustrates the cooling apparatus 506 with a single zone of controlling a flow rate through the manifold 526 and the plurality of microchannels 524. FIG. 5D illustrates a three-dimensional view of a portion (indicated as the dashed box in FIG. 5C) of the cooling apparatus 506 according to an embodiment.
Referring to FIGS. 5A to 5D, a coolant fluid flows into an inlet hole 572 of the manifold 526 and flows through a main inlet channel 594 and a plurality of inlet subchannels 592 in the manifold 526. While flowing through the inlet subchannels 592 in a longitudinal direction thereof, a portion of the coolant fluid flows down to the microchannels 524, flows through a plurality of outlet subchannels 590 and a main outlet channel 596, and then exits through an outlet hole 582 of the main outlet channel 596.
Although the embodiment shown in FIGS. 5A to 5D has a single zone for controlling the entire flow rate through the channels of the manifold 526 and the plurality of microchannels 524, embodiments of the present disclosure are not limited thereto. In other embodiments, a semiconductor device may have two or more zones for controlling respective flow rates through corresponding channels of a manifold (e.g., a manifold 626 in FIG. 6C) and microchannels (e.g., microchannels 624 in FIG. 6C). For example, in order to reduce the total pressure drop and improve the efficiency of heat dissipation, embodiments of the present disclosure divides the manifold into two or more zones so that a relatively large amount of flow can flow where the power generated in the entire chip is relatively high, and thus the flow rates in different zones may vary, or the flow rate in the same zone may vary, or both. By reducing the pressure drop of the cooling apparatus, the operating cost can be reduced by lowering the pumping power required for a given amount of heat dissipation, or more heat can be dissipated while operating with substantially the same pumping power.
FIGS. 6A, 6B, and 6C illustrates a cooling apparatus 602 including a manifold 626 and a microchannel structure 628 that has a plurality of microchannels 624 according to an embodiment. Specifically, FIGS. 6B and 6C illustrate a semiconductor device with a first zone Zone 1 for controlling a first flow rate through a first plurality of microchannels and a second zone Zone 2 for controlling a second flow rate through a second plurality of microchannels.
If there are two or more zones, a flow distribution device can be made for flow distribution. Zones are divided based on their Power Density level and whether they operate independently. In an embodiment, if the chip includes a region with high power density and a region with low power density, the manifold 626 and the plurality of microchannels 624 may be divided into two zones. For example, referring to FIGS. 6B and 6C, a first zone (indicated by an upper dashed box) may be disposed over a first region of a chip where a first amount of heat with relatively high power density is generated during an operation of the chip and include a first plurality of micro channels intersecting with a single path of the manifold. In addition, a second zone (indicated by a lower dashed box) may be disposed over a second region of the chip where a second amount of heat with relatively low power density is generated during the operation of the chip and include a second plurality of microchannels intersecting with two paths of the manifold. In an embodiment, a flow rate in a specific zone may be controlled based on an amount of heat generated in a corresponding area of a chip over which the specific zone is located, or a degree of non-uniformity in a distribution of the generated heat, or both. For example, when a first amount of heat generated in a first area of a chip is greater than a second amount of heat generated in a second area of the chip, a first flow rate in the first zone over the first area to dissipate the first amount of heat may be greater than a second flow rate in the second zone over the second area to dissipate the second amount of heat. When a first distribution of heat generated in a first area of the chip is more non-uniform compared to a second distribution of heat generated in a second area of the chip, a first flow rate in a first zone over the first area of the chip may be greater than a second flow rate in a second zone over the second area of the chip. A first flow rate per unit area of a first zone to dissipate heat generated in a first region of a chip with first power density may be controlled to be greater than a second flow rate per unit area of a second zone to dissipate heat generated in a second region of the chip with second power density, the first power density being greater than the second power density, the area of the first zone and the area of the second zone being defined when seen from a top view of a semiconductor device.
FIG. 7 illustrates a process 700 of operating a flow distribution device according to an embodiment. The flow distribution device may include a sensing (ex. temperature, power) part, a calculation part for control, and an actuator part, and it is characterized by a feedback control.
By making a manifold (e.g., the manifold in FIGS. 6B and 6C) have more than two zones, it is possible to increase the flow rate intensively where necessary, and it includes automatic feedback control of the flow rate by sensing temperature or power. In an embodiment, the flow distribution device may (1) control a flow rate of a zone according to a sensed temperature based on a predetermined relationship (e.g., a table) between the flow rate and the sensed temperature, or (2) using a PID control, and controlling the flow rate according to a temperature margin (e.g., a difference between a sensed temperature and a target temperature), or (3) controlling the flow rate based both of the table and the PID control. For example, at S720, a sensing part (e.g., one or more sensors) may measure one or more of a flow rate of each channel, a temperature, a pressure, and heat flux. At S740, a calculation part for control (e.g., a controller) may perform calculations for a predetermined control method (e.g., PID control). At S760, the controller may control an operation of an actuator part (e.g., one or more actuators) to control a flow rate based on the measurement and calculation results. These steps S720, S740, and S760 may be repeated until one or more predetermined conditions are satisfied, for example, when a difference between a sensed temperature and a target temperature becomes substantially equal to or less than a given threshold.
In order to lower the temperature of the chip even at the same flow rate, the heat transfer area may be increased, the heat transfer coefficient may be increased, or both. If the surface of the microchannel is roughened, not only the surface area of the microchannel increases, but also the effective heat transfer coefficient can be increased by promoting boiling and turbulence of the coolant flow. To increase the surface area of the microchannel, as shown in FIGS. 8A to 8E, a method of arbitrarily generating scallops by controlling the process conditions during etching, or attaching particles, etc. to the surface may be performed after microchannel formation.
The microchannel structure 828A including microchannels 824A according to the embodiment shown in FIG. 8A may be implemented by forming deep trenches, attaching particles on surfaces of the deep trenches, and performing an etching process. Specifically, the attached particles may function as hard mask patterns to etch exposed portions of the deep trenches during the etching process, thereby forming the structure shown in FIG. 8A. Optionally, the attached particles may be removed after the etching process is complete.
The microchannel structure 828B according to the embodiment shown in FIG. 8B may include a plurality of structures to facilitate formation of a turbulent flow of a coolant flowing through microchannels 824B to increase heat transfer. In an embodiment, the plurality of structures are disposed on sidewalls of fins defining the microchannels 824B, and each of the plurality of structures oscillates toward and away from a corresponding sidewall on which it is disposed when a coolant flows through the microchannel 824B. For example, the structures may be disposed on sidewalls of the microchannels 824B and have a feather shape or a fish scale shape to oscillate in a direction perpendicular to the main flow direction of the coolant.
The microchannel structure 828C including microchannels 824C according to the embodiment shown in FIG. 8C may be implemented by forming deep trenches, forming a passivation layer (e.g., an oxidation layer) over surfaces of the deep trenches, forming openings in the passivation layer to expose portions of bottom surfaces of the deep trenches, and performing an etching process on the exposed portions of the bottom surfaces of the deep trenches to form lower portions of the microchannels each having a cross-sectional area with a substantially trapezoidal shape. When the lower portions of the microchannels 824C each have a cross-sectional area having a trapezoidal shape with an upper edge shorter than a lower edge, as shown in the embodiment of FIG. 8C, the initial vapor generation temperature can be lowered and heat transfer by boiling can be increased. For example, the openings in the passivation layer may be formed to expose center portions of the bottom surfaces of the deep trenches by forming a photoresist layer over the passivation layer, performing a first tilted lithography and a second tilted lithography on portions of the photoresist layer over the center portions of the bottom surfaces of the deep trenches, and developing the photoresist layer to expose the center portions of the bottom surfaces of the deep trenches.
The microchannel structure 828D including microchannels 824D according to the embodiment shown in FIG. 8D may be implemented by forming deep trenches, depositing at least one metal element (e.g., copper) on sidewalls and bottom surfaces of the deep trenches, and forming at least one metal oxide having a flower-like shape using a chemical compound. For example, the metal oxide may include CuO and the chemical compound may include NaOH.
The microchannel structure 828E including microchannels 824E according to the embodiment shown in FIG. 8E may be implemented by forming scalloped deep trenches. For example, forming the scalloped deep trenches may include performing a plurality of sub-cycles, each of the sub-cycles including performing an etching process and removing byproducts (e.g., polymer) that result from the etching process.
FIGS. 9A and 9B each show a microchannel structure and a manufacturing method thereof according to embodiments. Each of the microchannel structure 928A and 928B in FIGS. 9A and 9B may be suitable for use as a microchannel structure (e.g., the microchannel structure 128 shown in FIG. 1) according to an embodiment of the present disclosure.
Referring to FIG. 9A, a plurality of first microchannels 924A are formed in a first plate (e.g., an upper plate) 912A, and a plurality of second microchannels 924B are formed in a second plate (e.g., a lower plate) 914A. The first microchannels 924A may have substantially the same channel width and pitch as those of the second microchannels 924B. For example, each of the first microchannels 924A may be a through-type channel that completely penetrates the upper plate 912A in a vertical direction, and each of the second microchannels 924B may partially penetrate the lower plate 914A in the vertical direction. A bottom surface of the upper plate 912A having the first microchannels 924A may be bonded to an upper surface of the lower plate 914A having the second microchannels 924B, resulting in the microchannel structure 928A having a plurality of microchannels 924C.
Referring to FIG. 9B, a plurality of third microchannels 924D are formed in an upper plate 912B, and a plurality of fourth microchannels 924E are formed in a lower plate 914B. The third microchannels 924D may have substantially the same channel width and pitch as those of the fourth microchannels 924E. Each of the third microchannels 924D may partially penetrate the upper plate 912B in a vertical direction, and each of the fourth microchannels 924E may partially penetrate the lower plate 914B in the vertical direction. After forming the third micro channels 924D without penetrating a lower portion of the upper plate 912B, the upper plate 912B may be turn over and bonded to the lower plate 914B having the fourth micro channels 924E. Subsequently, an upper portion of the resulting structure may be removed through grinding, or a plurality of upper portions respectively corresponding to the third microchannels 924D may be removed through laser drilling. As a result, the microchannel structure 928B having a plurality of microchannels 924F may be formed.
The first method shown in FIG. 9A has fewer process steps compared to the second method shown in FIG. 9B. However, if there are one or more risk factors in handling the upper plate 912A having the first microchannels 924A, the second method may be used to fabricate a microchannel structure. For example, since the first microchannels 912A completely penetrate the upper plate 912A, the upper plate 912A may be susceptible to damage while bonding it to the lower plate 914A depending on the size and/or depth of the first microchannels 924A. Although each of the stack structures 928A and 928B according to the embodiments shown in FIGS. 9A and 9B includes two plates, embodiments of the present disclosure are not limited thereto. For example, a microchannel structure may include three or more plates each including a plurality of microchannels, and may be fabricated by repeating the stacking of layers using the above-described methods.
FIGS. 10A and 10B each show a microchannel structure and a manufacturing method thereof according to embodiments. Microchannel structures 1028A and 1028B in FIGS. 10A and 10B include similar elements to those of the microchannel structures 928A and 928B in FIGS. 9A and 9B, respectively, and thus detailed descriptions of these elements and manufacturing methods thereof may be omitted for the interest of brevity.
The microchannel structure 1028A in FIG. 10A differs from the microchannel structure 928A in FIG. 9A in that the microchannel structure 1028A includes a plurality of first microchannels 1024A having different channel widths and pitches from those of a plurality of second microchannels 1024B. For example, the first microchannels 1024A have channel widths and pitches greater than those of the second microchannels 1024B.
The microchannel structure 1028B in FIG. 10B differs from the microchannel structure 928B in FIG. 9B in that the microchannel structure 1028B includes a plurality of third microchannels 1024D having different channel widths and pitches from those of a plurality of fourth microchannels 1024E. For example, the third microchannels 1024D have channel widths and pitches greater than those of the fourth microchannels 1024E.
Stacking microchannels with different channel widths and pitches as shown in the embodiments of FIGS. 10A and 10B may increase the surface area while more uniformly distributing the flow, compared to the embodiments shown in FIGS. 9A and 9B.
Although each of the stack structures 1028A and 1028B according to the embodiments shown in FIGS. 10A and 10B includes two plates, embodiments of the present disclosure are not limited thereto. For example, a microchannel structure may include three or more plates each including a plurality of microchannels, and may be fabricated by repeating the stacking of more layers using stacking methods similar to the above-described methods with reference to FIGS. 9A and 9B. It may be desirable to have relatively wide channel widths in terms of flow distribution, but various channel widths can be stacked in various combinations for other purposes (e.g., turbulence formation) to increase cooling efficiency.
FIG. 11A to 11I illustrate manifold structures and methods for adjusting flow distribution according to embodiments of the present disclosure.
FIG. 11A is a plan view of a manifold 1126A according to an embodiment of the present disclosure. The manifold 1126A in FIG. 11A includes an inlet main channel 1194A, a plurality of inlet subchannels 1192A, an outlet main channel 1196A, and a plurality of outlet subchannels 1190A. The inlet main channel 1194A may include an inlet hole (e.g., the inlet hole 572 in FIG. 5B) and be coupled to the inlet subchannels 1192A. The outlet main channel 1196A may include an outlet hole (e.g., the outlet hole 582 in FIG. 5B) and be coupled to the outlet subchannels 1190A. Each of the inlet subchannels 1192A may have a substantially constant cross-sectional area along a specific direction (e.g., a longitudinal/second direction with respect to FIG. 11A) through which a coolant flows. For example, the cross-sectional area of the inlet subchannel 1192A may have a width W1A in a first direction and a height in a third direction orthogonal to a plane defined by the first direction and the second direction, and the width W1A and the height may be substantially constant along the second direction. Similarly, the cross-sectional area of the outlet subchannel 1190A may have a width W2A in the first direction and a height in the third direction, and the width W2A and the height may be substantially constant along the second direction.
FIG. 11B is a plan view of a manifold 1126B according to another embodiment of the present disclosure. The manifold 1126B in FIG. 11B includes an inlet main channel 1194B, a plurality of inlet subchannels 1192B, an outlet main channel 1196B, and a plurality of outlet subchannels 1190B. The inlet main channel 1194A may extend in a first direction and receive a coolant fluid, and each of the inlet subchannels 1192B may be coupled to the inlet main channel 1194A and extend in a second direction, the second direction being perpendicular to the first direction. The outlet main channel 1196A may extend in the first direction and discharge the coolant fluid, and each of the outlet subchannels 1190A may be coupled to the outlet main channel 1196A and extend in the second direction. Each of the inlet subchannels 1192B may have a width W1B in the first direction that decreases along the second direction (or the longitudinal direction) through which the coolant flows, thereby obtaining a substantially uniform flow distribution. Specifically, when a coolant fluid flows through the inlet subchannel 1192B in the longitudinal direction, a portion of the coolant fluid flows down to microchannels (e.g., the microchannels 524 in FIG. 5A) to reduce a flow rate of the coolant fluid in the longitudinal direction through a cross-sectional area of the inlet subchannel 1192B. Since the width W1B of the cross-sectional area also decreases in the longitudinal direction, a velocity of the coolant fluid flowing through the cross-sectional area may be maintained substantially uniform along the longitudinal direction. The uniform flow distribution provided by the manifold 1126B makes the temperature of the chip surface uniform. Each of the outlet subchannels 1190B in FIG. 11B may have a width W2B in the first direction that increases along the second direction and be coupled to the outlet main channel 1196B, thereby making vapors generated in a microchannel heat sink including the microchannels under the manifold 1126B move smoothly toward the outlet main channel 1196B. The smooth movement of bubbles toward the outlet main channel 1196B may lower the pressure drop in the chip cooling system and reduce the pumping power supplied to drive the cooling system, thereby reducing energy use and improving cooling efficiency.
FIG. 11C illustrates a cross-sectional area along the line A-A′ of the manifold 1126A in FIG. 11A according to an embodiment. For example, each of the outlet subchannel 1190-1 and the inlet subchannel 1192-1 may have a substantially rectangular cross-sectional area.
FIG. 11D illustrates a cross-sectional area along the line A-A′ of the manifold 1126A in FIG. 11A according to an embodiment. For example, when the manifold 1126A is disposed over a microstructure (e.g., the microchannel structure 528 in FIG. 5C) including microchannels (e.g., the microchannels 524 in FIG. 5C) in a vertically upward direction, a horizontal width of the inlet subchannel 1192-2 may decrease as it approaches the microchannels in a vertically downward direction. As the width of the horizontal width of the inlet subchannel 1192-2 decreases as it gets closer to the microchannels, it is possible to supply fluid with a relatively high fluid velocity to the microchannels through a narrowing passage, thereby improving cooling performance. In addition, a horizontal width of the outlet subchannel 1190-2 increases as it approaches the microchannels in the vertically downward direction, thereby making vapors generated in the microchannel heat sink under the manifold 1126A move smoothly to the outlet subchannel 1190-2 and an outlet main channel (e.g., the outlet main channel 1196A in FIG. 11A) including the exit outlet. The smooth movement of vapors to the outlet may lower the pressure drop in the chip cooling system and reduce the use of supplied pumping power.
FIG. 11E illustrates a cross-sectional area along the line A-A′ of the manifold 1126A in FIG. 11A according to an embodiment. For example, two or more inlet subchannels 1192-3 each have a cross-sectional area that is tilted vertically to direct the coolant fluid toward a specific region of a microchannel heat sink (e.g., the microchannel structure 528 in FIG. 5C). As a result, a flow rate directed to the specific region may be increased to effectively reduce the temperature of a hot spot generated in the specific region of the microchannel heat sink disposed under the manifold 1126A.
FIG. 11F illustrates a cross-sectional area along the line A-A′ of the manifold 1126A in FIG. 11A according to an embodiment. For example, a horizontal width of the inlet subchannel 1192-4 increases as it approaches the microchannels in the vertically downward direction, and the inlet subchannel 1192-4 has two portions (e.g., branches) through which the flow is divided and injected into the microchannels, thereby reducing the flow resistance and increasing the flow rate. In addition, by making the inlet subchannel 1192-4 have two branches, vapors generated in the microchannel heat sink under the manifold 1126A may be moved smoothly to outlet subchannels 1190-1 located on both sides of the inlet subchannel 1192-4. For example, by making the inlet subchannel 1192-4 have two lower portions that are respectively tilted toward a pair of the outlet subchannels 1190-1 located adjacent to the inlet subchannel 1192-4, vapors generated in the microchannel heat sink may be moved smoothly to the outlet subchannels 1190-1 and an outlet main channel (e.g., the outlet main channel 1196A in FIG. 11A).
FIG. 11G illustrates a cross-sectional area along the line B-B′ of the manifold 1126A in FIG. 11A according to an embodiment. For example, the outlet main channel 1196A may have a substantially rectangular cross-sectional area.
FIG. 11H illustrates a cross-sectional area along the line C-C′ of the manifold 1126A shown in FIG. 11A according to an embodiment. The outlet subchannel 1190A may have a height H in the vertical direction that increases along the second/longitudinal direction, and uses the buoyancy of the vapors generated in the microchannel heat sink under the manifold 1126A to make the vapors move smoothly toward the outlet main channel 1196A and substantially prevent the backflow of vapors. In addition, by preventing the backflow of vapors from the outlet main channel 1196A, the pressure drop in the chip cooling system may be reduced and the pumping power supplied for driving the cooling system may be reduced, thereby reducing energy use and improving cooling efficiency.
Embodiments of the present disclosure relate to a chip cooling apparatus that reduces a thermal resistance of a cooling path. In some embodiments, such a chip cooling apparatus reduces a conduction thermal resistance in the cooling path by one or more of directly bonding a portion (e.g., the microchannel structure 128 in FIG. 1A) of the cooling apparatus to a chip (e.g., the chip 120 in FIG. 1A), reducing one or both of a thickness of a base (e.g., substantially zero thickness as shown in FIG. 4B) of the microchannel structure and a thickness of the chip (e.g., the chip 420B in FIG. 4B), and doping one or both of the base (e.g., the base 422C in FIG. 4C) of the microchannel structure and the chip (e.g., the chip 420C in FIG. 4C). In some embodiments, the chip cooling apparatus reduces a convection thermal resistance in the cooling path by increasing a surface area of the microchannels, or promoting boiling/turbulence of a coolant flow, or both. For example, rough surfaces may be formed as shown in the embodiments of FIGS. 8A to 8E, microchannels each having a relatively high aspect ratio may be formed as shown in the embodiments of FIGS. 9A and 9B, and/or microchannels having different widths and pitches in a stacking direction may be formed as shown in the embodiments of FIGS. 10A and 10B.
FIG. 12 illustrates a three-dimensional view of a portion of a cooling apparatus according to an embodiment. Specifically, FIG. 12 is a schematic view illustrating a manifold 1226 and a microchannel structure 1228.
The manifold 1226 in FIG. 12 includes a plurality of inlet subchannels 1292 and a plurality of outlet subchannels 1290. The inlet subchannels 1292 and the outlet subchannels 1290 are alternately arranged in a specific direction (e.g., a first direction I in FIG. 12).
Referring to FIG. 12, a coolant fluid flows into the inlet subchannels 1292 of the manifold 1226. Each of the inlet subchannels 1292 includes a portion that expands as the coolant fluid flows therethrough. For example, when the coolant fluid flows into the inlet subchannels 1292 in a first direction I (e.g., a vertically downward direction in FIG. 12), each of the inlet subchannels 1292 includes a portion having a length in a second direction II that increases along the first direction I. In the embodiment shown in FIG. 12, the second direction II may be substantially orthogonal to the first direction I.
The microstructure 1228 in FIG. 12 includes a base 1222 and a plurality of microchannels 1224. The coolant fluid having passed through the inlet subchannels 1292 flows into the microchannels 1224, and then flows from the microchannels 1224 into the outlet subchannels 1290 of the manifold 1226.
Each of the outlet subchannels 1290 includes a portion that contracts as the coolant fluid flows therethrough. For example, when the coolant fluid flows out from the outlet subchannels 1290 in a third direction III (e.g., a vertically upward direction in FIG. 12), each of the outlet subchannels 1290 includes a portion having a length in the second direction II that decreases along the third direction III.
A conventional cooling apparatus may have a relatively narrow flow passage from an inlet to microchannels and from the microchannels to an outlet, and the flow passage may cause a coolant fluid to abruptly change its flow direction as it flows therethrough. In contrast, in a cooling apparatus including a manifold (e.g., the manifold 1226 in FIG. 12) according to an embodiment of the present disclosure, a coolant fluid gradually expands when it flows through a plurality of first subchannels (e.g., the inlet subchannels 1292 in FIG. 12) and gradually contracts when it flows through a plurality of second subchannels (e.g., the outlet subchannels 1290 in FIG. 12). As a result, a pressure drop in a cooling apparatus according to an embodiment of the present disclosure may be significantly reduced compared to that of the conventional cooling apparatus. Accordingly, in a cooling apparatus according to an embodiment of the present disclosure, pumping power required for a given amount of heat dissipation can be significantly reduced, or more heat can be dissipated while operating with substantially the same pumping power compared to the conventional cooling apparatus. In other words, cooling efficiency of a cooling apparatus according to an embodiment of the present disclosure can be significantly improved compared to the conventional cooling apparatus.
Also, a conventional cooling apparatus may require an additional structure to achieve uniform flow distribution in a manifold therein, leading to an increase in area and volume of the cooling apparatus. In contrast, in a cooling apparatus according to an embodiment of the present disclosure, a manifold (e.g., the manifold 1226 in FIG. 12) includes a plurality of first channels (e.g., the plurality of inlet subchannels 1292) and a plurality of second channels (e.g., the plurality of outlet subchannels 1290 in FIG. 12), each of which includes a specific portion that gradually changes a cross-sectional area through which a coolant fluid flows to achieve uniform flow distribution, obviating the need of an additional structure in the conventional cooling apparatus. In addition, since the plurality of first subchannels (e.g., the inlet subchannels 1292 in FIG. 12) and the plurality of second subchannels (e.g., the plurality of outlet subchannels 1290 in FIG. 12) are alternately arranged in a specific direction (e.g., a direction orthogonal to a plane defined by the first direction I and the second direction II in FIG. 12), a cooling apparatus including the manifold according to an embodiment of the present disclosure can significantly reduce a total area and a total volume compared to those of the conventional cooling apparatus.
FIGS. 13A-13E illustrate a structure including a manifold suitable for use as the manifold 1226 shown in FIG. 12, according to an embodiment. Specifically, FIG. 13A illustrates a perspective view of the structure, FIG. 13B illustrates a three-dimensional view obtained by cutting the structure along a line A-A′ in FIG. 13A, FIG. 13C illustrates a three-dimensional view obtained by cutting the structure along a line C-C′ in FIG. 13A, FIG. 13D illustrates a three-dimensional view obtained by cutting the structure along a line B-B′ in FIG. 13A, and FIG. 13E illustrates a cross-sectional view obtained by cutting the structure along a line D-D′ in FIG. 13A.
FIGS. 13A-13E illustrates a structure including a cover layer (e.g., a lid layer) 1331, an intermediate layer 1333, and a manifold 1326. The cover layer 1331 includes an inlet hole 1372 and an outlet hole 1382. The intermediate layer 1333 includes a main inlet channel 1394 and a main outlet channel 1396. The manifold 1326 includes a plurality of inlet subchannels 1392 and a plurality of outlet subchannels 1390.
Referring to FIGS. 13B and 13C, the inlet hole 1372 is coupled to the main inlet channel 1394, and the main inlet channel 1394 is coupled to the plurality of inlet subchannels 1392. A coolant fluid flows through the inlet hole 1372, the main inlet channel 1394, and the plurality of inlet subchannels 1392.
An end portion (e.g., an upper end portion) of each of the plurality of inlet subchannels 1392 is disposed proximate to the main inlet channel 1394 and has a streamlined structure to reduce flow resistance when the coolant fluid flows from the main inlet channel 1394 to the inlet subchannels 1392. For example, the end portion may have a substantially triangular cross-section with a rounded tip.
Each of the plurality of inlet subchannels 1392 has a first portion (e.g., an upper portion) 1392U and a second portion (e.g., a lower portion) 1392L. The upper portion 1392U gradually expands as the coolant fluid flows therethrough to reduce flow resistance and a pressure drop. In an embodiment, when the coolant fluid flows into the inlet hole 1372 in a first direction (e.g., a vertically downward direction), the upper portion 1392U has a length in a second direction (e.g., along the line A-A′) that increases along the vertically downward direction. For example, the upper portion 1392U may have a first sidewall inclined with respect to the vertically downward direction by a first angle and a second sidewall inclined with respect to the vertically downward direction by a second angle, such that the sum of the first angle and the second angle is in a range from 10° to 70°. The first angle may be an angle of rotation in a first rotation direction from the vertically downward direction, and the second angle may be an angle of rotation in a second rotation direction from the vertically downward direction, where the second rotation direction is opposite to the first rotation direction to make the first sidewall and the second sidewall diverge along the vertically downward direction. In the embodiment of FIG. 13B, the upper portion 1392U may have a left sidewall inclined with respect to the vertically downward direction by the first angle of 0° (that is, extending in the vertically downward direction) and a right sidewall inclined with respect to the vertically downward direction by the second angle between 10° and 70°. The lower portion 1392L in FIG. 13B has a pair of opposite sidewalls, each of which extends in the vertical direction. In some embodiments, the upper portion 1392U may have a first sidewall inclined with respect to the vertically downward direction by a third angle, and a second sidewall included with respect to the vertically downward direction by a fourth angle, where the third and fourth angles are in the same rotation direction from the vertically downward direction to make the first sidewall and the second sidewall diverge along the vertically downward direction. In some embodiments, the upper portion 1392U may have a first sidewall extending in the vertically downward direction and a second sidewall having a curved profile to make the first sidewall and the second sidewall diverge along the vertically downward direction. In some embodiments, the upper portion 1392U may have a first sidewall extending in the vertically downward direction and a second sidewall having a stepped profile to make the first sidewall and the second sidewall diverge along the vertically downward direction. The coolant fluid having passed through the upper portion and lower portion 1392U and 1392L of each of the inlet subchannels 1392 flows into a plurality of microchannels (e.g., the microchannels 1224 in FIG. 12) of a microchannel structure (e.g., the microstructure 1228 in FIG. 12), and then flows from the microchannels into the plurality of outlet subchannels 1390.
Referring to FIGS. 13D and E, the outlet hole 1382 of the cover layer 1331 is coupled to the main outlet channel 1396, and the main outlet channel 1396 is coupled to the plurality of outlet subchannels 1390. The coolant fluid flows through the plurality of outlet subchannels 1390, the main outlet channel 1396, and the outlet hole 1382.
Each of the plurality of outlet subchannels 1390 has a first portion (e.g., an upper portion) 1390U and a second portion (e.g., a lower portion) 1390L. The upper portion 1390U gradually contracts as the coolant fluid flows therethrough to reduce flow resistance and a pressure drop. In an embodiment, when the coolant fluid flows out from the outlet hole 1382 in a first direction (e.g., a vertically upward direction), the upper portion 1390U has a length in a second direction (e.g., along the line B-B′) that decreases along the vertically upward direction. In the embodiment of FIG. 13D, the upper portion 1390U may have a first sidewall inclined with respect to a third direction (e.g., a vertically downward direction) opposite to the first direction by a first angle and a second sidewall inclined with respect to the third direction by a second angle, such that the sum of the first angle and the second angle is in a range from 10° to 70°. The first angle may be an angle of rotation in a first rotation direction from the vertically downward direction, and the second angle may be an angle of rotation in a second rotation direction from the vertically downward direction, where the second rotation direction is opposite to the first rotation direction to make the first sidewall and the second sidewall diverge along the vertically downward direction. In the embodiment of FIG. 13D, the upper portion 1390U may have a right sidewall inclined with respect to the vertically downward direction by the first angle of 0° (or extending in the vertically downward direction) and a left sidewall inclined with respect to the vertically downward direction by the second angle between 10° and 70°. In some embodiments, the upper portion 1390U may have a first sidewall inclined with respect to the vertically downward direction by a third angle, and a second sidewall included with respect to the vertically downward direction by a fourth angle, where the third and fourth angles are in the same rotation direction from the vertically downward direction to make the first sidewall and the second sidewall diverge along the vertically downward direction. In some embodiments, the upper portion 1390U may have a first sidewall extending in the vertically downward direction and a second sidewall having a curved profile to make the first sidewall and the second sidewall diverge along the vertically downward direction. In some embodiments, the upper portion 1390U may have a first sidewall extending in the vertically downward direction and a second sidewall having a stepped profile to make the first sidewall and the second sidewall diverge along the vertically downward direction. The lower portion 1390L in FIG. 13D has a pair of opposite sidewalls, each of which extends in the substantially vertical direction. The coolant fluid having passed through the lower portion 1390L and the upper portion 1390U of each of the outlet subchannels 1390 flows into the main outlet channel 1396, and then flows from the main outlet channel 1396 into the outlet hole 1382 of the cover layer 1331.
An end portion (e.g., an upper end portion) of each of the outlet subchannels 1390 is disposed proximate to the main outlet channel 1396 and has a streamlined structure to reduce flow resistance when the coolant fluid flows from the outlet subchannels 1390 to the main outlet channel 1396. For example, the end portion may have a substantially triangular cross-section with a rounded tip.
In the embodiment shown in FIGS. 13A-13E has a single zone with a single inlet hole 1372 and a single outlet hole 1382 for controlling the entire flow rate through the main inlet channel 1394, the inlet subchannels 1392, the outlet subchannels 1390, and the main outlet channel 1396. However, embodiments of the present disclosure are not limited thereto. For example, a semiconductor device may have two or more zones for controlling respective flow rates through corresponding channels of a manifold, where each of the two or more zones includes a single inlet hole and a single outlet hole. Controlling these respective flow rates through the two or more zones to achieve uniform cooling in a semiconductor device has been described above with reference to FIGS. 6A-6C, and thus detailed descriptions may be omitted herein for the interest of brevity.
FIGS. 14A, 14B, and 14C each illustrate a three-dimensional view of a component (e.g., a layer) of a manifold structure (e.g., the structure shown in FIG. 13A) according to an embodiment. Specifically, FIG. 14A illustrates a cover layer 1431, FIG. 14B illustrates an intermediate layer 1433, and FIG. 14C illustrates a manifold 1426.
One or more of the cover layer 1431, the intermediate layer 1433, and the manifold 1426 each may be fabricated through machining or microfabrication processes. In an embodiment, one or more of the cover layer 1431, the intermediate layer 1433, and the manifold 1426 have a thermal expansion coefficient sufficient to be compatible with a microchannel structure (e.g., the microchannel structure 1228 in FIG. 12) when the manifold structure is combined with the microchannel structure. In an embodiment, one or more of the cover layer 1431, the intermediate layer 1433, and the manifold 1426 have a thermal conductivity sufficiently low to significantly reduce heat transfer with a coolant fluid flowing therethrough, thereby improving cooling efficiency. For example, one or more of the cover layer 1431, the intermediate layer 1433, and the manifold 1426 may include metal (e.g., Al, Cu), plastic (e.g., Polycarbonate, Polyether ether ketone), or glass (e.g., Borosilicate glass).
The cover layer 1431, the intermediate layer 1433, and the manifold 1426 may be combined with each other using adhesive(s), or sealant(s), or both. For example, the cover layer 1431, the intermediate layer 1433, and the manifold 1426 may be combined with each other using any one of O-rings, silicone sealant, epoxy adhesive, and a combination thereof.
FIG. 15 illustrates a portion of a semiconductor device including a microchannel structure 1528, a heat spreading layer 1523, and an insulating layer (e.g., a dielectric layer) 1525, according to an embodiment. The microchannel structure 1528 includes a plurality of microchannels 1522 and a base 1524.
The insulating layer 1525 in FIG. 15 includes a heat source 1527. In an embodiment, the heat source 1527 is a heating element (e.g., CPU, GPU, Neural Engine, ASIC, FPGA, laser diode) or a hot spot serving as a localized heat source. Although not shown in FIG. 15, in other embodiments, the insulating layer 1525 may be a printed circuit board (PCB) that is spaced apart from the heat spreading layer 1523, and the heat source 1527 is disposed between the insulating layer 1525 and the heat spreading layer 1523.
The heat spreading layer 1523 in FIG. 15 is disposed between the microchannel structure 1528 and the insulating layer 1525. In an embodiment, the heat spreading layer 1523 directly contacts with a bottom surface of the microchannel structure 1528. For example, the heat spreading layer 1523 directly contacts a bottom surface of the base 1524 of the microchannel structure 1528. In other embodiments, the microchannel structure 1528 omits the base 1524, and the heat spreading layer 1523 directly contacts bottom surfaces of the microchannels 1522 of the microchannel structure 1528.
The heat spreading layer 1523 in FIG. 15 has a relatively high thermal conductivity. In an embodiment, the heat spreading layer 1523 may have a thermal conductivity higher than that of the microchannel base 1524 (e.g., silicon, silicon carbide, diamond-like carbon). In some embodiments, the heat spreading layer 1523 may have a thermal conductivity higher than that of the heat source 1527. For example, the heat spreading layer 1523 may have a thermal conductivity not less than 100 W/mK.
The heat spreading layer 1523 in FIG. 15 may include aluminum nitride, boron nitride, silicon carbide, or diamond. For example, a thin film formed of aluminum nitride, boron nitride, silicon carbide, or diamond may be deposited directly on a bottom surface of the microchannel structure 1528 using a deposition method (e.g., CVD, sputtering, ALD).
The heat spreading layer 1523 in FIG. 15 has a width Wspread and a height Hspread. In an embodiment, the width Wspread of the heat spreading layer 1523 is greater than a width WHS of the heat source 1527 and smaller than a width of the base 1524 of the microchannel structure 1528. For example, the width Wspread of the heat spreading layer 1523 is in a range from three times to ten times as great as the width W H S of the heat source 1527. When the width Wspread of the heat spreading layer 1523 is smaller than three times the width WHS of the heat source 1527, the hot spreading layer 1523 may not sufficiently support the base 1524 to ensure structural stability. When the width Wspread of the heat spreading layer 1523 is greater than ten times the width W H S of the heat source 1527, a thermal resistance in a horizontal direction of the heat spreading layer 1523 may be excessively large to significantly deteriorate cooling efficiency for removing heat generated from the heat source 1527. In an embodiment, the height H spread of the heat spreading layer 1523 may be less than 10 mm. When the height Hspread of the heat spreading layer 1523 is greater than 10 mm, a thermal resistance in a vertical direction of the heat spreading layer 1523 may be excessively large to significantly deteriorate cooling efficiency for removing heat generated from the heat source 1527.
In a conventional chip cooling apparatus, a thermal interface material may be disposed under between a base of a microchannel structure, and a heat spread layer may be disposed under the thermal interface material. As a result, a total thermal resistance of the heat spread layer and the thermal interface material may be excessively large to significantly deteriorate cooling efficiency for removing heat generated from a hot spot occurred in a chip. In contrast, a heat spreading layer (e.g., the heat spreading layer 1523 in FIG. 15) in a chip cooling apparatus according to an embodiment of the present disclosure directly contacts a microchannel structure (e.g., the microchannel structure 1528 in FIG. 15) and has a relatively high thermal conductivity. Accordingly, thermal resistances in both vertical and horizontal directions of a heat spreading layer according to an embodiment of the present disclosure are reduced compared to those of the conventional heat spread layer and thermal interface material, leading to increased cooling efficiency.
Aspects of the present disclosure have been described in conjunction with the specific embodiments thereof that are proposed as examples. Numerous alternatives, modifications, and variations to the embodiments as set forth herein may be made without departing from the scope of the claims set forth below. Accordingly, embodiments as set forth herein are intended to be illustrative and not limiting.
Patent Application
20240096742
