S. De Schampheleire

An experimental study of natural convection in open-cell aluminum foam

Natural convecton n air-saturated alumnum foam has been measured. A carefully designed experimental setup is built for his ask. The calibraton is done by comparing he results of a flat plate wh literature data, revealing excellent agreement. The nvestigated foams have a pore densiy of 10 and 20 PPI. The bondng of the foam is performed via brazing, or by applying a single epoxy which is enriched wh highly conductve alumna particles. The Rayleigh number is varied between 2500 and 6000, wh he rato of he surface area o he perimeter of he substrate as characteristc length. The foam height is varied between 12 and 25.4 mm. A major difference between both he bondng methods is observed. The brazed samples showed a beter heat ransfer n all cases. Furthermore, when ncreasing he foam height, a clear augmentaton of he heat ransfer is observed. Based on hese results, a correlaton is presented.

Modeling hydrodynamic properties of open-cell metal foams

Modeling the hydrodynamics of open-cell aluminum foam still is a challenging task because of the large range of length-scales, own to the physical phenomena which occur in the complex structure. Upscaling the classical conservation equations is a promising approach, but introduces the problem of modeling closure terms. This is dealt with via the well-known porous properties, i.e., permeability and inertial loss factor. Derivation of these properties is commonly done by linking pressure drop data to velocity via a second order interpolation. This, however, introduces significant deviation between the available data set, up to anorder of magnitude, which in turn results in difficulties to predict pressure drop during desing the desing phase of an applicaiton. As the closure terms have a well-defined physical meaning, it should be possible to compute them with better accuracy. This forms the topic of this paper.