The advancement of technologies for the realization of rocket thrust chambers allows the selection of new design and materials as well as new liquid propellants as coolants. One of the keys for the success of a new technology is the capability to realize safe and efficient designs of the cooling systems. To this goal, an important support is provided by numerical simulations, which must be reliable and efficient. The present research aims at studying the role of the material selection on the coolant behaviour, especially in presence of high heat loads, as those occurring in liquid rocket engines. The study is carried out with a conjugate heat transfer numerical approach, which has been already validated for specific test cases. It is based on in-house solvers for the solution of both steady Reynolds Averaged Navier Stokes (RANS) equations in the channel and heat conduction equation in the material. An additional analysis with Direct Numerical Simulations (DNS) will strengthen the value of the RANS solution and further validate the averaged approach. The in-house DNS code solves for the three-dimensional, unsteady Navier-Stokes equations down to the dissipative scale. The solver relies on the immersed-boundary method to handle complex geometries, and it is capable of handling conjugate heat transfer problems through local suppression of convective terms and change of the heat conduction coefficient within solid zones. The reference geometry for the study is a straight channel with rectangular cross section. After a comparison of results on thermally developed water test cases, simulations will be carried out including two metallic materials one featuring high thermal conductivity and one low thermal conductivity. The different thermal conductivity affects the distribution along the channel perimeter of the heat flux entering the coolant flow and may enhance or discourage thermal stratification and the possible occurrence of heat transfer deterioration.
The novelty of the present study lies in the potential results that will allow to strengthen the knowledge of the role of heat flux distribution around the channel on the overall coolant behaviour. To our knowledge, no studies are available in the open literature which discuss this aspect that is of key importance when approaching new designs, especially if they are based on classical one-dimensional modelling of flow in the cooling channels. The latter works well in average but it does not consider the possible occurrence of local phenomena which could actually depend on the thermal behaviour of the surrounding material. Although RANS and DNS simulations have been both used in the literature for specific cooling channel problems, their combined use in the framework of the same study of conjugate heat transfer uniquely characterizes the present research. In fact, the use of both RANS and DNS provides results of maximum significance for tests representing sizes and Reynolds numbers of practical interest and at the same time offers a detailed validation, which cannot be otherwise obtained by the comparison of the "integral" values typically provided by experimental measurements. The combined approach aims to increase the knowledge on the reliability of RANS simulations with respect to conjugate heat transfer problems with high heat-fluxes and high temperature/density gradients. Tests and assessment of different calibrations and corrections like that relevant to the occurrence of high gradients of density will constitute an important set of information for the future studies on the present and similar subjects.
Finally, as one of the most important expected innovations, the present research has the potential to increase the understanding of the level of heat loads that, for coolants like methane or propane, would lead to heat transfer deterioration as depending on the selected material for cooling channel walls. The results would provide a starting point for the selection of appropriate analysis tools in both development and verification phases of new engine designs.