Modeling of Expanding Metal FoamsB. Chinè[1,2], M. Monno
Laboratorio MUSP, Macchine Utensili e Sistemi di Produzione, Piacenza, Italy
School of Materials Science and Engineering, Costa Rica Institute of Technology, Cartago, Costa Rica
Politecnico di Milano, Dipartimento di Meccanica, Milano, Italy
Metal foams are interesting materials with many potential applications. They are characterized by a cellular structure represented by a metal or metal alloy and gas voids inside (Fig.1). A common metallic cellular material is aluminum foam which can be produced metallurgically by heating a precursor, made of aluminum alloy and TiH2 as foaming agent, in a furnace. In this case, the foaming process involves the heating of the solid precursor containing the embedded gas source that, upon temperature increasing, releases H2 gas and drives the foaming process.
The complexity of the foaming process of a metal and the number of parameters to control simultaneously demand a preliminary and hugely wide experimental activity to manufacture foamed components with a good quality. The development of computational models could help to reduce experimental works and costs, although the task is very challenging.
The complex physical phenomena arising during the foaming process can be modeled by recurring to the mechanics principles, particularly to the simultaneous conservation laws for the energy, momentum and mass transport. Attention is demanded by the presence of a dynamic interface between the gas bubbles continuously originating and the surrounding liquid. Other phenomena are represented by the growth, coalescence and collapse of the gas bubbles.
In this work we use COMSOL Multiphysics® to model a simplified metal foam expansion during mold filling. Flow, heat and mass transport phenomena with surface tension effects are considered in the model, which is assumed to be two-dimensional. We start modeling and simulating a foaming process by considering an isothermal bubble expansion. Then, we model a more realistic non-isothermal metal foaming process inside a furnace, represented by a number of H2 bubbles expanding in liquid aluminum. In both cases the flow is assumed to be laminar and compressible. A Stokes flow regime is considered. Fig. 2 shows a model of a metal foam made by seven gas bubbles and aluminum metal. To control coalescence phenomena, the diffusion of species in the system is also taken into account by using the Fick's law. Phase field techniques are set up to describe the movement of the interfaces in the foam.
The numerical findings of the simulations show that the computational models, based on COMSOL Multiphysics® can be effective for modeling the foaming process of a metal. Other physical mechanisms such as heating and cooling rates, drainage, disjoining pressure and final solidification of the foam could be included for more comprehensive models. But in the latter case, when developing more complete models for a foaming process, computational requirements should be also considered.
To date it is not known whether the supercritical power plant dynamic responses can satisfy the GB Grid Code requirement , therefore simulation-based research using COMSOL Multiphysics® will be focused specifically on the forced convection of pseudo-critical water inside a heated pipe (figure 1). The heat transfer coefficient will be evaluated in response to sudden changes in water conditions such as pressure, heat flux or mass flux, and thermal efficiencies. These results will be then verified by designing and constructing an experimental apparatus based on initial calculations. The apparatus consists of a test element wherein water will be pumped and heated through the critical point, in order to recreate the conditions experienced in supercritical water cycle. Subsequently, energy calculations based on real data will be conducted. In addition, a novel approach to energy storage as an alternative to the drum in conventional power plants will be investigated in an attempt to adhere to the constraints of the GB Code.
As a result, the best working conditions from a thermodynamic point of view will be identified, and more accurate control models and efficient boiler designs will be possibly produced.