Morphology and burning rates of expanding spherical flames in H2/O2/inert mixtures up to 60 atmospheres

Abstract

Recognizing that previous experimental studies on constant-pressure, outwardly propagating, spherical flames with imaging capability were limited to pressures less than about 5 atm, and that pressures within internal combustion engines are substantially higher, a novel experimental apparatus was designed to extend the environmental pressure to 60 atm. Results substantiate previous observations of the propensity of cell formation over the flame surface due to hydrodynamic and diffusive-thermal instabilities and provide convincing evidence that wrinkled flame is the preferred mode of propagation in hydrogen/air mixtures in environments with pressures above only a few atmospheres. It is further shown that, by using helium as the diluent, and by reducing the oxygen concentration of the combustible, diffusional-thermal instability can be mostly suppressed and the hydrodynamic instability delayed. Stretch-free laminar flame speeds were subsequently determined for such smooth flames up to 20 atm and were compared with the calculated values, allowing for detailed chemistry and transport.

Reduced and Validated Kinetic Mechanisms for Hydrogen-CO-sir Combustion in Gas Turbines

“…Given that the H 2 kinetic models investigated here were all validated against a wide range of (and frequently the same) data from numerous experimental apparatuses including the high pressure flame speeds of Tse et al [25], the disparities noted amongst the predictions and with the present experimental results are noteworthy.…”

“…Compared to Ardiluted flames, however, the temperature dependence is considerably stronger -for example, E a = 80 kcal/mol for CO 2 -diluted flames at 20 atm compared to E a = 53 kcal/mol for Ar-diluted flames at 20 atm over the temperature range from 1500 to 1700K using Eq. (25). Reasons for the stronger temperature and pressure dependence for CO 2 -diluted flames compared to Ar-diluted flames are discussed in the next section.…”

“…The primary focus is on the manifestation of this effect in the flame shape evolution, flame propagation speed, and burned gas speeds along the r-axis in chambers having a length greater than their diameter. These factors and conditions encompass all of the research conducted previously in cylindrical chambers [25,1,35,[37][38][39]. Figure 4 presents the experimental data for flame propagation speed history along the raxis of a closed cylindrical chamber for equivalence ratios of 1.0 and 3.0 and the numerical solution of the flamelet model for a cylindrical chamber of the same aspect ratio as the experiment.…”

A hypothetical burning-velocity formula for very lean hydrogen–air mixtures

“…This may be seen most clearly in Fig. 1, where the predictions of the San Diego mechanism are compared with the most recent accurate experiments [3][4][5][6]. Although the computations, described previously [2], are based on one particular chemical-kinetic mechanism, with other mechanisms the direction of the discrepancies is the same, and the magnitudes of the discrepancies are comparable or larger.…”

“…[13,14], as useful reviews [15,16] describe in greater detail. The papers reporting the burning-velocity measurements specifically comment on observed cell formation for spherical flames at sufficiently lean conditions [3,5,6], but the flames in the steady counterflow experiment were said to be very flat [4] and yet produced velocities in agreement with the other measurements. Although the strain in the counterflow is known to suppress visible cells, it nevertheless may not be sufficient to eliminate transverse preferential diffusion that may result in similar burning velocities.…”