AmericanChemicalSociety.com

Turbulent lean-to-stoichiometric premixed flames were experimentally studied in two counterflow configurations at turbulent Reynolds numbers as large as one thousand. The primary objectives were to examine conditions of departure from the flamelet regime, characterized by laminar-like, singly-connected turbulent flame fronts, and analyze the turbulent premixed flame structure in the non-flamelet regime, for which disrupted and locally-extinguished flame fronts are expected.

The first burner configuration consisted of two highly turbulent opposed jets fed with the same fresh reactant mixtures. In that configuration, turbulent twin-premixed flames could be stabilized at turbulent Karlovitz numbers, Ka_t, of at most order unity. The envelope of turbulent flame regime was limited to the flamelet regime because of the inevitable annihilation of the turbulent flames by the large bulk strain rate when Ka_t was increased. The interaction of the turbulent flames with the coherent structures was analyzed using simultaneous PIV/OH-LIF measurements.

In the second counterflow burner configuration, a turbulent stream of fresh premixed reactants was opposed to a second stream of hot products of combustion. This approach let us study the effects of heat losses and flame dilution by combustion products. These effects, which are not accounted by the current diagrams of turbulent premixed combustion, were revealed to lower the boundary of the flamelet regime significantly to Ka_t values of nearly unity. The structure of the oxidation layer analyzed using simultaneous CO/OH-LIF measurements was found to be sensitive to the turbulence intensity and the hot product composition. The quenching of the oxidation layer, which is not currently accounted for in turbulent combustion theory, appeared to be a critical element of departure from the flamelet regime.

The interpretations of the experimental results were aided by numerical calculations of the extinction of strained laminar premixed flames using a one-dimensional arc-length continuation code with detailed chemistry and transport mechanisms. Two distinct extinction modes were observed: an abrupt one and a smooth one, the latter being favored by an excess of oxidizing species in the combustion product stream. This study provided valuable insights in the mechanisms by which premixed flames extinguish.

Mr. Bruno Coriton was supported by this grant during the last three years of doctoral studies and successfully defended in September 2010. A number of publications and presentations stemmed out of this work, as detailed below. We anticipate two additional peer-reviewed articles to be submitted in the next few months. This tour de force opened up new vistas in the challenging area of turbulent combustion that will keep us busy for years to come. The work lead to a collaboration with Dr. Andreas Kempf at Imperial College for concurrent computational modeling, and with Dr. Jonathan Frank from Sandia National Laboratories to apply state-of-the-art diagnostics. The work laid the foundation for a collaborative proposal with Professor Steve Pope at Cornell that was one of the few funded in this year round of submissions to NSF. With this new grant, we will continue to study interesting combustion regimes experimentally, while Professor Pope will model the flames computationally.

The above is evidence that the relatively modest investment of ACS/PRF has been very successful on a number of levels, including: the discovery of new fundamental aspects of turbulent combustion on which 85% of energy conversion is based worldwide, student support, publications and promoting additional sponsored research.