0
Research Papers

Flexible Thermal Ground Planes Fabricated With Printed Circuit Board Technology

[+] Author and Article Information
Li-Anne Liew

Department of Mechanical Engineering,
University of Colorado at Boulder,
Boulder, CO 80309
e-mail: Li-Anne.Liew@colorado.edu

Ching-Yi Lin, Ryan Lewis, Susan Song, Qian Li

Department of Mechanical Engineering,
University of Colorado at Boulder,
Boulder, CO 80309

Ronggui Yang

Department of Mechanical Engineering,
University of Colorado at Boulder,
Boulder, CO 80309
e-mail: Ronggui.Yang@colorado.edu

Y. C. Lee

Department of Mechanical Engineering,
University of Colorado at Boulder,
Boulder, CO 80309
e-mail: Yung-Cheng.Lee@colorado.edu

Contributed by the Electronic and Photonic Packaging Division of ASME for publication in the JOURNAL OF ELECTRONIC PACKAGING. Manuscript received August 30, 2016; final manuscript received November 13, 2016; published online December 7, 2016. Assoc. Editor: Xiaobing Luo.

J. Electron. Packag 139(1), 011003 (Dec 07, 2016) (10 pages) Paper No: EP-16-1103; doi: 10.1115/1.4035241 History: Received August 30, 2016; Revised November 13, 2016

Thermal ground planes (TGPs) are passive thermal management devices that utilize the phase-change of a working fluid to achieve high thermal conductivity and low thermal resistance. TGPs are flat, two-dimensional heat pipes—similar to vapor chambers—in which liquid is held within a capillary wick, and vapor is held in a sealed vapor layer. Heat is absorbed at an evaporator region, causing the liquid to evaporate. The heated vapor in the vapor core is carried via convection to a condenser region where it condenses as the heat is expelled from the TGP to an external heat sink. The condensed liquid is then pulled back to the evaporator via capillary forces in the wick. In numerous applications, mechanical flexibility of the TGP is required, as is low-cost manufacturing and viable integration routes with electronics. This work describes a flexible TGP (FTGP) fabricated using printed circuit board (PCB) technology, in which commercially available copper-cladded polyimide sheets are used as the casing material. The wick is composed of three layers of fine copper mesh electroplated or sintered together and coated with atomic layer deposited TiO2. A coarse nylon or polyether ether ketone (PEEK) mesh defines the vapor transport layer, and water is used as the working fluid. The perimeter of the device is heat-sealed with flouroethylene propylene (FEP), which has been found to provide a near-hermetic seal for several months and is suitable for flexible applications. This architecture allows the TGP to function with minimal reduction in heat transfer performance while bent by 90 deg, and full functionality is returned when the device is returned to its flat configuration. The FTGP's measured thermal resistance is about half that of an equivalent copper reference for input heat fluxes of 3–6 W/cm2. More than 30 copper-cladded polyimide FTGPs were fabricated and characterized using both simple qualitative and more involved quantitative test setups. The results show that the fabrication and assembly processes developed in this work are repeatable and the devices are durable.

FIGURES IN THIS ARTICLE
<>
Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 2

(a) Cross section of a copper-cladded polyimide FTGP and (b) exploded schematic where the thicknesses are exaggerated for clarity

Grahic Jump Location
Fig. 3

Photograph of the material layers used to construct the FTGP

Grahic Jump Location
Fig. 4

(a) Scanning electron micrograph of an electroplated triple-layer copper mesh. The surfaces are roughened after the electroplating process and (b) close-up of the mesh surface.

Grahic Jump Location
Fig. 5

Comparison of wicking speeds between triple-layer electroplated-bonded meshes and triple-layer sintered meshes, showing that electroplating-bonding is a feasible alternative to sintering for fabricating the wicks. “E” = electroplated, “S” = sintered. All meshes have an area of 5 cm × 9.5 cm and did not have ALD hydrophilic coating applied, but were subjected to O2 plasma treatment before the measurement to make the surfaces hydrophilic.

Grahic Jump Location
Fig. 7

Detailed schematic of the cross section of the TGP, view from evaporator side to condenser side. Fluid flow is into the plane of the figure. Each circle is the cross section of a wire in the corresponding meshes. The wires extend into the plane of the page.

Grahic Jump Location
Fig. 6

FTGP assembly process: (a) The copper-cladded polyimide sheets are heat-sealed along three edges to form a pouch, into which (b) the wick (if not already bonded to one of the polymide sheets) and vapor spacer layer and charging tube are inserted. (c) The fourth edge is heat-sealed over the charging tube, and vacuum epoxy is applied around the interface around the tube to seal any leaks. (d) The TGP is evacuated and charged with water through the tube. (e) A final heat seal is made in front of the charging tube. (f) The outer section of the TGP containing the charging tube is then cut away.

Grahic Jump Location
Fig. 1

Schematic representation of the working principle of a TGP, including evaporation of the working liquid (water) at the evaporator region, vapor transport in the vapor core, condensation of vapor to liquid at the condenser region, and liquid transport back from the condenser to evaporator in the wick

Grahic Jump Location
Fig. 12

(a) Schematic and photograph of a FTGP in 90-deg bend configuration, in this case with a 1 cm bending radius. (b) Thermal resistance of FTGP B from Fig. 10, when bent at 90-deg with 5 mm bend radius compared to when unbent. The evaporator and condenser areas are 1 in.2. The data shown are discontinuous, and the lines connecting the points are for illustration purposes only. We calculate the uncertainty in the Rth to be 0.07 K/W, and the uncertainty in the input power to be 0.2 W.

Grahic Jump Location
Fig. 13

Simplified test setup for rapid qualitative thermal characterization of FTGP devices to show flexibility and long-term durability: (a) schematic of testing apparatus, (b) photograph of testing apparatus with FTGP in the flat configuration, and (c) photograph of FTGP tested with a 90-deg bend and a bend radius of 5 mm

Grahic Jump Location
Fig. 14

Typical performance curve of one FTGP over the course of 38 days, using a simplified test set up where Rth is calculated from the input power. The evaporator and condenser areas are 1 in.2. The data shown are discontinuous, and the lines connecting the points are for the purposes of guiding the eyes only. Note that in this 38-day test, variation of up to 22% can be seen, but these variations do not correspond with degradation over time and are more likely due to operator variations when fixturing the TGP from one day to another. This TGP had a sintered mesh as the wick.

Grahic Jump Location
Fig. 8

(a) Schematic of the FEP heat-sealing approach, showing the five linear sections that are bonded during the assembly process. In addition, photographs of the completed FTGP are shown before (b) and after (c) the charging tube is removed.

Grahic Jump Location
Fig. 9

Schematic of the test set-up to measure the thermal resistance of the FTGP. Not shown is the insulation covering the test sample. The sample shown in this schematic is a 5 cm × 9.5 cm × 1 mm thick copper reference.

Grahic Jump Location
Fig. 10

Thermal resistance measured for two copper-cladded polyimide FTGPs and a copper reference sample, for evaporator and condenser areas of 1 in.2. The active thermal transport area in both FTGPs and the copper surface are 5 cm × 9.5 cm, and the thickness of the samples is 1 mm. The two TGPs differed slightly in construction, as summarized in Table 1, accounting for their different performance curves. The data shown are discontinuous, and the lines connecting the points are for the purposes of guiding the eyes only. We calculate the uncertainty in the Rth to be 0.07 K/W, and the uncertainty in the input power to be 0.2 W.

Grahic Jump Location
Fig. 11

Evaporator and condenser temperatures for FTGPs A and B. The data shown are discontinuous, and the lines connecting the points are for the purposes of guiding the eyes only. The uncertainty in the temperature is 0.1 °C due to the resolution of the temperature sensors used, and the uncertainty in the input power is calculated to be 0.2 W. Note that these temperatures are measured by the thermocouples embedded in the heater and condenser blocks at a point located about 0.5 mm below their surfaces at the center of the contact area.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In