A Combined Experimental and Computational Study of Soot Formation in Normal and Microgravity Conditions

ABSTRACT

In this paper, we consider a combined experimental and computational study of sooting, ethylene-air, laminar, coflow diffusion flames under normal and microgravity conditions. Two different burner configurations (Yale and ACME) are studied. Microgravity experiments are performed aboard the International Space Station. Computed simulations for both configurations under 1 g and 0 g conditions support previous experimental and modeling conclusions that at 0 g, the flame is lengthened, becomes broader and more diffuse, temperatures decrease and soot levels and gas residence times increase. However, in the case of the smaller ACME flame, these effects are significantly diminished. In an effort to assess the cause of the differences between the experimental and computational results at the two gravity levels, we perturbed a variety of model parameters associated with soot evolution to observe their effects on the simulation results. We also investigated the effects of radiation and its re-absorption. The results of the study indicate that it is likely that the model inadequately predicts the thermal field and the use of experimental temperatures imposed onto the solution would provide more accurate simulations of the soot field and heat transfer to the burner lip could alter the boundary conditions of the flames and hence the subsequent solutions.

KEYWORDS:

• Microgravity
• soot
• laminar diffusion flames

Acknowledgments

One of the authors (MDS) first met Paul Libby in 1983 at a Combustion Research Facility (CRF) review meeting at Sandia National Laboratories in Livermore CA. It was at this meeting that Professor Libby talked about the advantages of using the counterflow configuration in studying laminar diffusion flames. He discussed the fact that the flames were one-dimensional and the configuration could be modified to consider nonpremixed, partially premixed as well as fully premixed flames. In addition, the strain rate provided a natural parameter with which to investigate extinction properties of these systems. These discussions helped initiate a counterflow research program in which MDS collaborated with a variety of combustion experimentalists, most notably, Professor K. Seshadri at UCSD. Upon moving to Yale, MDS began a combined experimental and numerical collaboration with Marshall Long (MBL) focused on using multidimensional laser imaging methods with an adaptive Newton-based numerical method to study coflow laminar diffusion flames. Professor Libby was a strong and vocal advocate for this work, not only for the fact that it was the next step in our ability to study multidimensional flames but also because it helped quantify the rapidly advancing field of diagnostic methodologies. His support and guidance of our research programs over the years cannot be underestimated and we will be forever grateful.

XZ and JS acknowledge funding support by FM Global within the framework of the FM Global Strategic Research Program on Fire Modeling.

MBL and MDS acknowledge support from NASA under grants 80NSSC19K1634 and NNX17AD49A.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Aeronautics and Space Administration [80NSSC19K1634 and NNX17AD49A].