skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Extinction and reignition in direct numerical simulations of CO/H 2 temporal plane jet flames

Abstract

Direct numerical simulations of three-dimensional turbulent temporally-evolving plane CO/H2 jet flames have been performed with skeletal chemistry at Reynolds numbers of up to 9,000 and with up to 500 million grid points (Hawkes, E.R., Sankaran, R., Sutherland, J.C., Chen, J.H., Proc. Combust. Inst. 31 (2007) 1633-1640). In the present paper, the data are analyzed to understand the processes of extinction and reignition observed in the simulations. A measure of extinction based on the amount of stoichiometric surface area having a reacting scalar less than a threshold value is used to characterize extinction. Employing this characterization leads naturally to the appearance of a local displacement speed of 'flame edges' as the primary quantity of interest. Flame edges are defined as the boundaries on the stoichiometric surface between areas having a reacting scalar less than the threshold (extinguished) and those above it (burning). The displacement speed is the speed at which these boundaries move relative to the local flow. The motion of flames edges is studied using a massively parallel analysis tool. The joint probability density function of the local edge flame speed and scalar dissipation rate has been extracted and reveals a transition in character as the simulation progresses. The transitionmore » is interpreted in the context of the physical mechanisms of extinction and reignition. Along with evidence of the alignment of the scalar and mixture fraction normal vectors, it indicates that the mechanism of folding by turbulence of burning regions onto extinguished ones is the dominant reignition mechanism for the simulated conditions.« less

Authors:
 [1];  [2];  [1]
  1. Sandia National Laboratories (SNL)
  2. ORNL
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States); Center for Computational Sciences
Sponsoring Org.:
USDOE Office of Science (SC)
OSTI Identifier:
930891
DOE Contract Number:
DE-AC05-00OR22725
Resource Type:
Conference
Resource Relation:
Conference: 5th US National Combustion Meeting, San Diego, CA, CA, USA, 20070325, 20070328
Country of Publication:
United States
Language:
English
Subject:
37 INORGANIC, ORGANIC, PHYSICAL AND ANALYTICAL CHEMISTRY; CARBON MONOXIDE; HYDROGEN; FLAMES; COMBUSTION KINETICS; PROBABILITY DENSITY FUNCTIONS; REYNOLDS NUMBER; SCALARS; INHIBITION; IGNITION; COMPUTERIZED SIMULATION

Citation Formats

Hawkes, Evatt R, Sankaran, Ramanan, and Chen, Jacqueline H. Extinction and reignition in direct numerical simulations of CO/H2 temporal plane jet flames. United States: N. p., 2007. Web.
Hawkes, Evatt R, Sankaran, Ramanan, & Chen, Jacqueline H. Extinction and reignition in direct numerical simulations of CO/H2 temporal plane jet flames. United States.
Hawkes, Evatt R, Sankaran, Ramanan, and Chen, Jacqueline H. Mon . "Extinction and reignition in direct numerical simulations of CO/H2 temporal plane jet flames". United States. doi:.
@article{osti_930891,
title = {Extinction and reignition in direct numerical simulations of CO/H2 temporal plane jet flames},
author = {Hawkes, Evatt R and Sankaran, Ramanan and Chen, Jacqueline H},
abstractNote = {Direct numerical simulations of three-dimensional turbulent temporally-evolving plane CO/H2 jet flames have been performed with skeletal chemistry at Reynolds numbers of up to 9,000 and with up to 500 million grid points (Hawkes, E.R., Sankaran, R., Sutherland, J.C., Chen, J.H., Proc. Combust. Inst. 31 (2007) 1633-1640). In the present paper, the data are analyzed to understand the processes of extinction and reignition observed in the simulations. A measure of extinction based on the amount of stoichiometric surface area having a reacting scalar less than a threshold value is used to characterize extinction. Employing this characterization leads naturally to the appearance of a local displacement speed of 'flame edges' as the primary quantity of interest. Flame edges are defined as the boundaries on the stoichiometric surface between areas having a reacting scalar less than the threshold (extinguished) and those above it (burning). The displacement speed is the speed at which these boundaries move relative to the local flow. The motion of flames edges is studied using a massively parallel analysis tool. The joint probability density function of the local edge flame speed and scalar dissipation rate has been extracted and reveals a transition in character as the simulation progresses. The transition is interpreted in the context of the physical mechanisms of extinction and reignition. Along with evidence of the alignment of the scalar and mixture fraction normal vectors, it indicates that the mechanism of folding by turbulence of burning regions onto extinguished ones is the dominant reignition mechanism for the simulated conditions.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Mon Jan 01 00:00:00 EST 2007},
month = {Mon Jan 01 00:00:00 EST 2007}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Abstract not provided.
  • Abstract not provided.
  • In this paper the authors explore the effects of differential diffusion in nonpremixed turbulent jet flames. Pulsed Raman scattering spectroscopy is used to measure temperature and species concentrations in chemically reacting jets of H[sub 2]/CO[sub 2] into air, over a range of jet Reynolds numbers from 1,000 to 30,000 based on cold jet fluid properties. Results show significant effects of differential diffusion at all jet Reynolds numbers considered. Differential diffusion between H[sub 2] and CO[sub 2] produces differences between the hydrogen element mixture fraction (f[sub H]) and the carbon element mixture reaction (f[sub C]). The greatest effects occur on themore » rich side of stoichiometric, where f[sub H] is observed to be smaller than f[sub C] at all Reynolds numbers. Differential diffusion between H[sub 2] and H[sub 2]O creates a net flux of hydrogen element toward the stoichiometric contour and causes a local maximum in f[sub H] that occurs near the stoichiometric condition. A differential diffusion variable z[sub H] is defined at the difference between f[sub H] and f[sub C]. The variance of z[sub H] conditional on f[sub C] also shows that differential diffusion effects are greatest on the rich side of the flame. Conditional variances of z[sub H] are largest at intermediate Reynolds numbers.« less
  • This paper explores effects of differential diffusion in nonpremixed turbulent jet flames. Pulsed Raman scattering spectroscopy is used to measure temperature and species concentrations in chemically reacting jets of H{sub 2}/CO{sub 2} into air, over a range of jet Reynolds numbers from 1,000 to 30,000 based on cold jet fluid properties. Results show significant effects of differential diffusion at all jet Reynolds numbers considered. Differential diffusion between H{sub 2} and C0{sub 2} produces differences between the hydrogen element mixture fraction ({xi}{sub H}) and the carbon element mixture fraction ({xi}{sub c}). The greatest effects occur on the rich side of stoichiometric,more » where {xi}{sub H} is observed to be smaller than {xi}{sub C} at all Reynolds numbers. Differential diffusion between H{sub 2} and H{sub 2}O creates a net flux of hydrogen element toward the stoichiometric contour and causes a local maximum in {xi}H that occurs near the stoichiometric condition. A differential diffusion variable {sup Z}H is defined as the difference between {xi}{sub H} and {xi}{sub C}. The variance Of {sup Z}H conditional on {xi}{sub C} also shows that differential diffusion effects are greatest on the rich side of the flame. Conditional variances of {sup Z}H are largest at intermediate Reynolds numbers.« less
  • The difficulty of experimental measurements of the scalar dissipation rate in turbulent flames has required researchers to estimate the true three-dimensional (3D) scalar dissipation rate from one-dimensional (1D) or two-dimensional (2D) gradient measurements. In doing so, some relationship must be assumed between the true values and their lower dimensional approximations. We develop these relationships by assuming a form for the statistics of the gradient vector orientation, which enables several new results to be obtained and the true 3D scalar dissipation PDF to be reconstructed from the lower-dimensional approximations. We use direct numerical simulations (DNS) of turbulent plane jet flames tomore » examine the orientation statistics, and verify our assumptions and final results. We develop and validate new theoretical relationships between the lower-dimensional and true moments of the scalar dissipation PDF assuming a log-normal true PDF. We compare PDFs reconstructed from lower-dimensional gradient projections with the true values and find an excellent agreement for a 2D simulated measurement and also for a 1D simulated measurement perpendicular to the mean flow variations. Comparisons of PDFs of thermal dissipation from DNS with those obtained via reconstruction from 2D experimental measurements show a very close match, indicating this PDF is not unique to a particular flame configuration. We develop a technique to reconstruct the joint PDF of the scalar dissipation and any other scalar, such as chemical species or temperature. Reconstructed conditional means of the hydroxyl mass fraction are compared with the true values and an excellent agreement is obtained.« less