Ignition delay times, laminar flame speeds, and mechanism validation for natural gas/hydrogen blends at elevated pressures

Abstract

New experimental ignition delay time data measured in both a shock tube and in a rapid compression machine were taken to determine the increase in reactivity due to the addition of hydrogen to mixtures of methane and natural gas. Test conditions were determined using a statistical design of experiments approach which allows the experimenter to probe a wide range of variable factors with a comparatively low number of experimental trials. Experiments were performed at 1, 10, and 30 atm in the temperature range 850-1800 K, at equivalence ratios of 0.3, 0.5, and 1.0 and with dilutions ranging from 72% to 90% by volume. Pure methane- and hydrogen-fueled mixtures were prepared in addition to two synthetic 'natural gas'-fueled mixtures comprising methane, ethane, propane, n-butane and n-pentane, one comprising 81.25/10/5/2.511.25% while the other consisted of 62.5/20/10/5/2.5% C-1/C-2/C-3/C-4/Cs components to encompass a wide range of possible natural gas compositions. A heated, constant-volume combustion vessel was also utilized to experimentally determine laminar flame speed for the same baseline range of fuels. In this test, a parametric sweep of equivalence ratio, 0.7-1.3, was conducted at each condition, and the hydrogen content was varied from 50% to 90% by volume. The initial temperature and pressure varied from 300 to 450 K and 1 to 5 atm, respectively. Flame speed experiments conducted above atmospheric pressure utilized a 1:6 oxygen-to-helium ratio to curb the hydrodynamic and thermal instabilities that arise when conducting laminar flame speed experiments. All experiments were simulated using a detailed chemical kinetic model. Overall good agreement is observed between the simulations and the experimental results. A discussion of the important reactions promoting and inhibiting reactivity is included. (C) 2013 The Combustion Institute. Published by Elsevier Inc. All rights reserved.