High heat flux cooling of accelerator targets with micro-channels

https://doi.org/10.1016/j.nima.2006.02.104Get rights and content

Abstract

Accelerator targets, both for radioisotope production and for neutron sources generate very high thermal density in the target material, which absorbs the particles beam. Total power is in the order of 10–100 kW. Generally, the geometric size of the targets is very small so the heat flux is high. The design of these targets requires efficient heat removal techniques in order to preserve the integrity of the target. The average heat fluxes from these targets can be up to 1 kW/cm2. Few techniques exist to deal with such high heat fluxes. One of them is mini- or micro-channels heat exchanger. In order to evaluate the performance of mini- and micro-channels heat exchangers for high heat flux cooling, experimental cooling loop has been designed and built at Soreq. Initial experiments have been preformed. These experiments demonstrated a cooling capacity over 1 kW/cm2. This cooling capacity can be obtained with relatively little cost as compared to results of earlier studies.

Introduction

In accelerator systems, the interaction of the particles beam with the target, both for radioisotope production and for neutron sources, generates very high density of thermal energy. Total power is in the order of 10–100 kW. Generally, the geometric size of the targets is very small and the heat flux is high. Normal average heat fluxes from these targets are around 1 kW/cm2. The design of these targets requires efficient heat removal techniques in order to preserve the integrity of the target.

Few techniques exist to deal with such high heat fluxes. Two of these techniques are mini/micro-channels and jet-impingement. Lienhard [1] provides a detailed review of the research work that has been done on jet-impingement cooling. Silverman et al. [2], [3] demonstrated that this technique can be used to build cooling systems capable of dealing with heat fluxes of 2 kW/cm2 and even more. However, this technique provides only external cooling of the target and requires a complex cooling head to generate the high-velocity jet and collect the cooling fluid.

A second technique is micro-channels cooling that have the advantage that it can be integrated into the target and provide internal cooling. The initial work by Tuckerman and Pease [4], directed to cool high-power electronic devices, demonstrated a cooling capability of up to 790 W/cm2. Later, Vidmar and Barker [5] suggested using micro-channels to cool windows for particle beams and RF transmission as well as VLSI chips. They have been able to increase the cooling capacity, using micrometer tubes, to a level of 2.7 kW/cm2 at the expense of using ultra-high pressure drop, up to 306 atm. Here the heat flux is calculated based on the outer diameter of the tubes. Kendall et al. [6] presented an assessment of the effect of channel diameter on cooling capability of heat sinks. One of the conclusions is that although CHF increases as the channel diameter decreases the system efficiency decreases too. Hence, it is true that one can increase the power dissipation but he would pay much more for this capability. The results of Vidmar and Barker demonstrate this conclusion.

Section snippets

Experimental setup

In order to evaluate the actual potential of mini/micro-channels heat sinks for high heat flux cooling, two prototype mini/micro-channels heat sinks have been built and tested. The heat source for these tests has been an electron gun that can provide up to 20 kW heating power with heat fluxes above 5 kW/cm2.

Two channel geometries have been studied: (1) A single micro-channel with a flow area of 0.2×32 mm2 (see Fig. 1). (2) A set of 25 parallel mini-channels with flow area of 1×3 mm2 each (see Fig. 2

Results

The experimental data for the single micro-channel heat sink is summarized in Fig. 3. The figure presents the target temperature measured by TC #1 located at the center of the heated area and 2.5 mm from the cooled surface (the total wall thickness is 5 mm). The available pump differential pressure limits the flow rate. The total pressure drop across the whole heat sink for the highest flow rate, 4.2 l/min, is about 3.75 bar. Results for two different flow rates are presented. As can be expected,

Conclusions

Two mini/micro-channels heat sink configurations have been tested. Initial experiments demonstrate that both configurations can achieve high heat flux cooling in the range of 1 kW/cm2 and above. These high cooling capacities have been achieved by relatively little cost in required pump power.

Based on these preliminary results, a mini/micro-channels cooled beam dump and beryllium target for neutron productions are being designed. The beam dump is designed on the concept of heat sink #2 to stop a

References (6)

  • J.H. Lienhard

    Liquid jet impingement

  • I. Silverman, A. Nagler, in: Proceedinngs of the 2004 ASME Heat Transfer/Fluids Engineering Summer Conference,...
  • I. Silverman, A. Arenshtam, D. Kijel, A. Nagler, High heat flux accelerator targets cooling with liquid metal jet...
There are more references available in the full text version of this article.

Cited by (24)

  • A unified model for Digitized Heat Transfer in a microchannel

    2014, International Journal of Heat and Mass Transfer
    Citation Excerpt :

    While some of these methods have achieved extremely high heat transfer rates, they sacrifice cost and reliability. For example, Silverman et al. reported 2000 W/cm2 in microjet devices using liquid metals [9] and Hirshfeld et al. reported 1500 W/cm2 in microchannels [10]. While effective for high heat flux cooling, these methods are limited by high pumping pressure, susceptibility to mechanical failure, and lack of adaptability.

View all citing articles on Scopus
View full text