Ignition of hydrogen-air mixing layer in turbulent flows

H. G. Im, J. H. Chen, Chung King Law

Research output: Contribution to journalConference article

89 Citations (Scopus)

Abstract

Autoignition of a hydrogen-air scalar mixing layer in homogeneous turbulence is studied using direct numerical simulation (DNS). An initial counterflow of unmixed nitrogen-diluted hydrogen and heated air is perturbed by two-dimensional homogeneous turbulence. The temperature of the heated airstream is chosen to be 1100 K, which is substantially higher than the crossover temperature at which the rates of the chain-branching and termination reactions are equal. Three different turbulence intensities are tested in order to assess the effect of the characteristic flow time on the ignition delay. For each condition, a simulation without heat release is also performed. The ignition delay determined with and without heat release is shown to be almost identical up to the point of ignition for all of the turbulence intensities tested, and the predicted ignition delays agree well within a consistent error band. It is also observed that the ignition kernel always occurs where hydrogen is focused, and the peak concentration of HO2 is aligned well with the scalar dissipation rate. The dependence of the ignition delay on turbulence intensity is found to be nonmonotonic. For weak to moderate turbulence, the ignition is facilitated by turbulence via enhanced mixing, while for stronger turbulence, whose timescale is substantially smaller than the ignition delay, the ignition is retarded due to excessive scalar dissipation, and hence diffusive loss, at the ignition location. However, for the wide range of initial turbulence fields studied, the variation in ignition delay due to the corresponding variation in turbulence intensity appears to be quite small.

Original languageEnglish (US)
Pages (from-to)1047-1056
Number of pages10
JournalSymposium (International) on Combustion
Volume27
Issue number1
DOIs
StatePublished - Jan 1 1998
Event27th International Symposium on Combustion - Boulder, CO, United States
Duration: Aug 2 1998Aug 7 1998

Fingerprint

turbulent flow
ignition
Turbulent flow
Ignition
Hydrogen
Turbulence
air
turbulence
hydrogen
Air
homogeneous turbulence
scalars
dissipation
heat
spontaneous combustion
counterflow
flow characteristics
Direct numerical simulation
direct numerical simulation
crossovers

All Science Journal Classification (ASJC) codes

  • Mechanical Engineering
  • Energy Engineering and Power Technology
  • Chemical Engineering(all)
  • Fluid Flow and Transfer Processes
  • Fuel Technology
  • Physical and Theoretical Chemistry

Cite this

@article{ad88228ac5b24a5b8b68ef331bda34af,
title = "Ignition of hydrogen-air mixing layer in turbulent flows",
abstract = "Autoignition of a hydrogen-air scalar mixing layer in homogeneous turbulence is studied using direct numerical simulation (DNS). An initial counterflow of unmixed nitrogen-diluted hydrogen and heated air is perturbed by two-dimensional homogeneous turbulence. The temperature of the heated airstream is chosen to be 1100 K, which is substantially higher than the crossover temperature at which the rates of the chain-branching and termination reactions are equal. Three different turbulence intensities are tested in order to assess the effect of the characteristic flow time on the ignition delay. For each condition, a simulation without heat release is also performed. The ignition delay determined with and without heat release is shown to be almost identical up to the point of ignition for all of the turbulence intensities tested, and the predicted ignition delays agree well within a consistent error band. It is also observed that the ignition kernel always occurs where hydrogen is focused, and the peak concentration of HO2 is aligned well with the scalar dissipation rate. The dependence of the ignition delay on turbulence intensity is found to be nonmonotonic. For weak to moderate turbulence, the ignition is facilitated by turbulence via enhanced mixing, while for stronger turbulence, whose timescale is substantially smaller than the ignition delay, the ignition is retarded due to excessive scalar dissipation, and hence diffusive loss, at the ignition location. However, for the wide range of initial turbulence fields studied, the variation in ignition delay due to the corresponding variation in turbulence intensity appears to be quite small.",
author = "Im, {H. G.} and Chen, {J. H.} and Law, {Chung King}",
year = "1998",
month = "1",
day = "1",
doi = "https://doi.org/10.1016/S0082-0784(98)80505-5",
language = "English (US)",
volume = "27",
pages = "1047--1056",
journal = "Proceedings of the Combustion Institute",
issn = "1540-7489",
publisher = "Elsevier Limited",
number = "1",

}

Ignition of hydrogen-air mixing layer in turbulent flows. / Im, H. G.; Chen, J. H.; Law, Chung King.

In: Symposium (International) on Combustion, Vol. 27, No. 1, 01.01.1998, p. 1047-1056.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Ignition of hydrogen-air mixing layer in turbulent flows

AU - Im, H. G.

AU - Chen, J. H.

AU - Law, Chung King

PY - 1998/1/1

Y1 - 1998/1/1

N2 - Autoignition of a hydrogen-air scalar mixing layer in homogeneous turbulence is studied using direct numerical simulation (DNS). An initial counterflow of unmixed nitrogen-diluted hydrogen and heated air is perturbed by two-dimensional homogeneous turbulence. The temperature of the heated airstream is chosen to be 1100 K, which is substantially higher than the crossover temperature at which the rates of the chain-branching and termination reactions are equal. Three different turbulence intensities are tested in order to assess the effect of the characteristic flow time on the ignition delay. For each condition, a simulation without heat release is also performed. The ignition delay determined with and without heat release is shown to be almost identical up to the point of ignition for all of the turbulence intensities tested, and the predicted ignition delays agree well within a consistent error band. It is also observed that the ignition kernel always occurs where hydrogen is focused, and the peak concentration of HO2 is aligned well with the scalar dissipation rate. The dependence of the ignition delay on turbulence intensity is found to be nonmonotonic. For weak to moderate turbulence, the ignition is facilitated by turbulence via enhanced mixing, while for stronger turbulence, whose timescale is substantially smaller than the ignition delay, the ignition is retarded due to excessive scalar dissipation, and hence diffusive loss, at the ignition location. However, for the wide range of initial turbulence fields studied, the variation in ignition delay due to the corresponding variation in turbulence intensity appears to be quite small.

AB - Autoignition of a hydrogen-air scalar mixing layer in homogeneous turbulence is studied using direct numerical simulation (DNS). An initial counterflow of unmixed nitrogen-diluted hydrogen and heated air is perturbed by two-dimensional homogeneous turbulence. The temperature of the heated airstream is chosen to be 1100 K, which is substantially higher than the crossover temperature at which the rates of the chain-branching and termination reactions are equal. Three different turbulence intensities are tested in order to assess the effect of the characteristic flow time on the ignition delay. For each condition, a simulation without heat release is also performed. The ignition delay determined with and without heat release is shown to be almost identical up to the point of ignition for all of the turbulence intensities tested, and the predicted ignition delays agree well within a consistent error band. It is also observed that the ignition kernel always occurs where hydrogen is focused, and the peak concentration of HO2 is aligned well with the scalar dissipation rate. The dependence of the ignition delay on turbulence intensity is found to be nonmonotonic. For weak to moderate turbulence, the ignition is facilitated by turbulence via enhanced mixing, while for stronger turbulence, whose timescale is substantially smaller than the ignition delay, the ignition is retarded due to excessive scalar dissipation, and hence diffusive loss, at the ignition location. However, for the wide range of initial turbulence fields studied, the variation in ignition delay due to the corresponding variation in turbulence intensity appears to be quite small.

UR - http://www.scopus.com/inward/record.url?scp=0032277646&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032277646&partnerID=8YFLogxK

U2 - https://doi.org/10.1016/S0082-0784(98)80505-5

DO - https://doi.org/10.1016/S0082-0784(98)80505-5

M3 - Conference article

VL - 27

SP - 1047

EP - 1056

JO - Proceedings of the Combustion Institute

JF - Proceedings of the Combustion Institute

SN - 1540-7489

IS - 1

ER -