On the transient thermal response of thin vapor chamber heat spreaders: Governing mechanisms and performance relative to metal spreaders

https://doi.org/10.1016/j.ijheatmasstransfer.2019.03.058Get rights and content

Highlights

  • Identified key mechanisms that govern transient thermal response of vapor chambers.

  • Experimentally verified these key mechanisms by testing a vapor chamber device.

  • Transient performance of a vapor chamber, relative to copper, is explored.

  • Mapped key thresholds at which transient performance of a vapor chamber is superior to copper.

Abstract

Vapor chambers can offer a passive heat spreading solution for thermal management in electronics applications ranging from mobile devices to high-power servers. The steady-state operation and performance of vapor chambers has been extensively explored. However, most electronic devices have inherently transient operational modes. For such applications, it is critical to understand the transient thermal response of vapor chamber heat spreaders and to benchmark their transient performance relative to the known behavior of metal heat spreaders. This study uses a low-cost, 3D, transient semi-analytical transport model to explore the transient thermal behavior of thin vapor chambers. We identify the three key mechanisms that govern the transient thermal response: (1) the total thermal capacity of the vapor chamber governs the rate of increase of the volume-averaged mean temperature; (2) the effective in-plane diffusivity governs the time required for the spatial temperature profile to initially develop; and (3) the effective in-plane conductance of the vapor core governs the range of the spatial temperature variation, and by extension, the steady-state performance. An experiment is conducted using a commercial vapor chamber sample to confirm the governing mechanisms revealed by the transport model; the model accurately predicts the experimental measurements. Lastly, the transient performance of a vapor chamber relative to a copper heat spreader of the same external dimensions is explored as a function of the heat spreader thickness and input power. The mechanisms governing the transient behavior of vapor chambers are used to explain the appearance of key performance thresholds beyond which performance is superior to the copper heat spreader. This work provides a foundation for understanding the benefits and limitations of vapor chambers relative to metal heat spreaders in transient operation and may inform the design of vapor chambers for improved transient performance.

Introduction

A vapor chamber is a phase-change-driven passive heat spreading device. A typical design consists of a hollow chamber with a liquid-saturated porous wick lining its inner surface enclosing a central vapor core. The operation of a vapor chamber is illustrated in Fig. 1. A localized heat input on the outer surface of the chamber is conducted through the wall causing evaporation at the wick-vapor interface. Localized vapor generation causes vapor to flow away from the heat input and into the vapor core. The vapor condenses onto the opposing (colder) wick-vapor interface, and the heat is rejected from the condenser surface. Capillary forces in the porous wick draw the condensed liquid back towards the heat input region, enabling continuous passive operation.

Heat spreading provides a critical function in the thermal management of electronic devices that has, in part, allowed engineers to develop systems operating at ever higher heat loads and densities. Vapor chambers have been extensively studied for this purpose, with potential applications ranging from the low powers (<10 W) in mobile electronic devices, to the high fluxes (>500 W/cm2) in radar power amplifiers and high-performance computing systems [1]. Tight space constraints and the need for spreading of transient heat loads are common in these applications. For example, mobile electronic devices experience low heat loads during idle operation with intermittent high-power bursts to execute functions such as video recording; the internal thickness allotted for thermal management and heat spreading is less than a millimeter.

Previous work in the design of vapor chambers has focused on improving their steady-state thermal performance [2], [3], [4], [5], [6], [7], [8], [9], [10]. Many studies have identified the important transport mechanisms in a vapor chamber operating at steady state and accordingly proposed designs to improve performance under these conditions. Prasher et al. [11] developed a resistance-network-based model for heat pipes, in which the transport processes in the different sections of the wall, wick, and vapor core are assigned thermal resistances. This model reveals that the resistance across the evaporator wick is most significant for vapor chambers subjected to localized, high-heat-flux inputs. Hence, considerable design efforts [5], [6], [7], [8], [9], [10], [12], [13], [14], [15], [16], [17], [18], [19] have been targeted at achieving a low resistance during evaporation or capillary-fed boiling in this region of the wick. Recent work by Yadavalli et al. [20] revealed the performance-governing mechanisms for thin vapor chambers operating at low powers; at extremely low thicknesses, the thermal resistance in the vapor core was shown to limit the vapor chamber performance relative to metal heat spreader. This understanding was used in our previous studies for designing the condenser-side wick [4] and selecting the working fluid [21] in ultra-thin vapor chambers, for low-power, hand-held applications.

Several studies have considered the transient behavior of heat pipes and vapor chambers. El-Genk and Lianmin [22] experimentally studied the heat-up and cool-down of a heat pipe under a range of evaporator-side input powers and condenser-side coolant flow rates, concluding that the transient vapor temperature profiles could be locally represented by an exponential function in the cases investigated. Tournier and El-Genk [23] developed a finite-volume-based model to simulate the mass, momentum and thermal transport in the vapor chamber wick to predict pooling of the liquid phase at the condenser. Zhu and Vafai [24] developed a model for heat spreading from a central heater in disk-shaped and rectangular vapor chambers. The analytical model solved for 1D transient conduction in the vapor chamber wall and wick while the quasi-steady vapor hydrodynamics was modeled using an assumed spatial velocity profile. The model was used to simulate the startup process of a vapor chamber in terms of the transient temperature and velocity fields. Harmand et al. [25] developed a finite-volume-based transport model to predict the transient behavior of rectangular vapor chambers. The model was validated against experiments, and the model capabilities were then demonstrated under several different heating configurations (spatial and temporal). These transient analyses of vapor chambers and others in the literature ([26], [27], [28], [29], [30]) do not attempt to identify the key transport mechanisms that govern the transient response of the vapor chamber. The goal of the current study is to extract an understanding of these transient governing mechanisms in vapor chambers relative to transient conduction in solid metal heat spreaders, so as to facilitate rational design of vapor chambers for improved relative transient performance.

In this work, our time-stepping analytical model for vapor chamber transport [31] is used to simulate the transient behavior of a vapor chamber and a solid copper heat spreader. Comparison of the temporal temperature fields in the two devices is used to identify and understand the key mechanisms that govern the transient behavior and performance of vapor chambers. Experiments are conducted with a commercial vapor chamber and compared to predictions from the model to demonstrate the key governing mechanisms identified. Lastly, the transient performance of a vapor chamber relative to a copper heat spreader of the same external dimensions is explored as a function of two key parameters, namely the heat spreader thickness and input power. Thresholds are identified beyond which the vapor chamber offers improved performance relative to the copper heat spreader. The relationship between the key governing mechanisms and the transient performance thresholds is established.

Section snippets

Time-stepping analytical model for vapor chamber transport

The time-stepping analytical model we previously developed [31] is used for simulating vapor chamber transient behavior. This 3D transient transport model can be used for simulating rectangular geometries of the vapor chamber wall, wick, and vapor core, in a configuration where the wick lines the inner surface of the wall and encloses the vapor core as shown in Fig. 1. The model allows arbitrarily shaped and located time-varying heat inputs to be specified on the faces. The mass, momentum, and

Numerical simulation case details

A vapor chamber and a copper spreader of identical external geometry are simulated to observe their transient response to a step heat input. A comparison of these two cases is used to obtain insight into the mechanisms governing the behavior of the vapor chamber.

Heat spreading in a vapor chamber can occur through vapor spreading in the vapor core, heat diffusion in the wick, and heat diffusion in the solid walls. The heat diffusion in the wick is negligible compared to the other processes and

Experimental facility and procedure

A transient heat spreading experiment is conducted with a thin, commercial vapor chamber. The transient temperature field is characterized in response to a step heat input to demonstrate and confirm the key mechanisms identified in Section 3.2. A photograph of the 150 mm-long, 8.5 mm-wide, and 1.8 mm-thick vapor chamber (Novark Technologies) is shown in Fig. 5a.

Fig. 5c illustrates the experimental test setup, which is designed to isolate the vapor chamber from any object having a significant

Dependence of relative transient performance on time scale

In this section, the performance of a vapor chamber is benchmarked against a copper spreader of the same external geometry. The performance is strongly dependent on time scale, with multiple crossovers in the peak temperature between the two spreaders for the chosen case. The reasons for this complex comparative behavior are discussed based on the relative magnitudes of the governing mechanisms underlying the transient thermal behavior. This discussion serves as a basis for description to

Conclusions

The mechanisms governing the transient thermal response of a vapor chamber are identified using a low-cost, 3D, and transient vapor chamber transport model. Conclusions from this analysis are corroborated with experiments conducted using a commercial vapor chamber. The vapor chamber transport model is used to compare the transient thermal response of a vapor chamber with that of a solid copper spreader. The performance of both heat spreader types is analyzed based on the peak evaporator

Conflict of interest

The authors declared that there is no conflict of interest.

Acknowledgement

Financial support for this work provided by members of the Cooling Technologies Research Center, a graduated National Science Foundation Industry/University Cooperative Research Center at Purdue University, is gratefully acknowledged. Donation of the vapor chamber tested in this work by Novark Technologies is gratefully acknowledged.

References (37)

Cited by (29)

  • Transient recovery from heat pipe dryout by power throttling

    2024, International Journal of Heat and Mass Transfer
  • A study on thermal and hydraulic performance of ultra-thin heat pipe with hybrid mesh-groove wick

    2023, e-Prime - Advances in Electrical Engineering, Electronics and Energy
View all citing articles on Scopus
View full text