AmericanChemicalSociety.com

Research activity in the year 2008-09 year was very fruitful and lead to three presentations at the 6^th Annual Combustion Meeting (Ann Arbor, MI, May 2009) with graduate student, Bruno Coriton, presenting two of them and the PI, the third one. In addition, one of the presentations already evolved into a first article that was published in the peer-reviewed literature (Coppola et al., Combust. Flame, 156, pp. 1834-1844, 2009). Mr. Coriton made considerable progress towards completion of his doctoral thesis that will be finalized in the Spring of 2010. His work also attracted the attention of a colleague at Sandia National Laboratories, Dr. Jonathan Frank, who invited Bruno over for a few weeks to take data on our burner in the state-of-the-art facility at Lawrence Livermore, Ca. Mr. Coriton is currently there performing measurements of OH and CO concentration, as well as of the forward reaction rate CO+OH—CO₂+H, using the Sandia set up. We anticipate that a minimum of three articles will be completed for submission to peer-reviewed journals at his return. As a result, the ACS/PRF funded research will have met the anticipated goals of supporting one graduate student, putting him in a position to publish in a number of premiere journals, as he contemplates the next step in his professional career.

In the article, we showed that a turbulent counterflow flame (TCF) is an ideal benchmark of complexity intermediate between laminar flames and practical systems. We demonstrated the phenomenology of turbulent counterflow flames operating under conditions spanning unpremixed, partially premixed and premixed regimes. By operating in a turbulent Reynolds number regime of relevance to practical systems, TCF retains the interaction of turbulence and chemistry of such environments, but offers several advantages including: a) the achievement of high Reynolds numbers, comparable to those of gas turbines and IC engines, without pilot flames, which is particularly advantageous from a modeling standpoint; b) control of the transition from stable flames to local extinction/reignition conditions; c) compactness of the domain by comparison with jet flames, with obvious advantages from both a diagnostic and, especially, a  computational viewpoint; and d) the reduction or, altogether, elimination of soot formation, thanks to the high strain rates and low residence times of such a system, and the establishment of conditions of large stoichiometric mixture fraction, as required for robust flame stabilization. The system lends itself to the validation of DNS and other computational models. It is also well-suited for the examination of different fuel blends- a need that is becoming more and more pressing in view of the anticipated diversification of the future fossil fuel supply. The PI expects to submit a new proposal for follow-on research in collaboration with a colleague with computational combustion expertise in the Spring of 2010, as a result of the exciting outcome of the ACS/PRF sponsored research and the interest that this publication generated.

Specific to premixed flames, in the spirit of a flamelet approach to turbulent combustion, we initially focused on computational work to shed light on the observed behavior in the experimental turbulent work, even though the latter is under turbulent conditions. We studied the extinction of laminar premixed CH₄/O₂/N₂ flames counterflowing against a jet of combustion products at equilibrium. The configuration is identical to our experimental work, except that conditions are laminar rather than turbulent. Still, important information on flame structure can be gathered in these types of studies. The work relied on detailed chemistry and transport mechanisms. The composition of the hot stream was determined by computing the chemical equilibrium of combustion products of lean methane/air flames at a fixed temperature of 1800 K. Contrary to similar studies of this type that were focused on heat loss/gain from the hot product stream, the emphasis here was on the hitherto unexplored role of the combustion product composition on the extinction of lean-to-stoichiometric premixed flames. It was found that premixed flames do not extinguish when there is an excess of oxygen in the counterflowing combustion products. At elevated strain rates, the flame, rather than stabilizing on the fresh mixture side of the stagnation plane, is forced to cross the stagnation plane in the so-called partially extinguished regime. As the strain rate is increased further, the heat release tends to be independent of the dilution and of the equivalence ratio of the premixed flame, and is affected primarily by the oxygen concentration in the counterflowing hot gases. The mechanisms involved in the different burning regimes and the flame structures were analyzed and implications for turbulent combustion regimes were considered.

In the experimental work, on the other hand, we focused on the phenomenology of premixed CH₄/O₂/N₂ flames by counterflowing a jet of turbulent CH₄/O₂/N₂ premixed flames against a jet of products of combustion in chemical equilibrium at 1800 K by OH-PLIF and PIV.  The focus was on characterizing conditions of departure from the flamelet regime in a broad range of turbulent Reynolds numbers from 200 to 700 and turbulent Karlovitz numbers from 1 to 400.  Seemingly paradoxical observations include local extinction at Karlovitz numbers of order unity were made.  These findings can be rationalized in terms of the effects of strain, heat losses, and the composition of the hot combustion product stream, all of which are present in practical systems, and the computational study discussed earlier can help shed light on the observed behavior. The present combustion system allows for a systematic examination of real flame effects to improve on the highly idealized Borghi diagram with which turbulent premixed combustion regimes have been characterized to date.