Experimental and Numerical Investigations of an Endothermic Fuel Cooling Capacity for Scramjet Application

Abstract

The high heat fluxes generated by hypersonic flight are one of the main issue that should be solved. During the last 40 years, several concepts for the proper thermal management of a high-speed vehicle have been proposed. One of the most promising is the active cooling of the combustion chamber by the fuel itself before the injection inside the engine. In some cases, the physical heat sink provided by the fuel is not sufficient and if the fuel-temperature is high enough (more than 800 K), an endothermic decomposition could start, providing a welcomed additional heat sink. In order to investigate this topic, MBDA-France and ONERA start a collaboration focused on the development of an experimental test-bench called the Promethee Micro-Pilot (MPP, located in ONERA-Palaiseau facility), combined with numerical tools and kinetic models. The MPP reactor zone is a 1 m long stainless-steel tube, high pressure continuous flow, heated by 8 heavyweight copper blocks. The maximum reactor temperature is well above 1100 K with a fluid exit temperature close to 900 K. Thermocouples placed along the reactor enable us to closely monitor the reactor inner (contact with the fuel) and outer (contact with the copper blocks) temperature, as well as the fuel temperature. Test runs were performed with Norpar 12 fuel (simpler fuel composition than a kerosene and cheaper than pure n-dodecane), for different reactor temperatures (between 870 K and 1170 K). Exit fuel temperature was found to be between 730 K and 910 K. To obtain a better understanding of the complex phenomena occurring inside the reactor, those results were compared with computations made with MBDA-France in-house software. For each experimental case two computations were done, one with n-dodecane fuel treated as a chemical inert fluid, and one with a real n-dodecane. Reactor walls temperature and fluid temperature were compared and a reasonable agreement was found between experimental results with Norpar 12 and computations with chemically reactive n- dodecane. The chemical heat sink provided by the fuel pyrolysis was evaluated and compared to the fuel physical heat sink computed in the chemically inert case. The next step of this work will include the addition of a coke deposit model inside the numerical model and some improvement in the way the fluid physical properties are evaluated.