Next-generation refrigerants are more environmentally benign than current ones but tend to be mildly flammable with maximum burning velocities in air of less than 10 cm/s. Industry has proposed using the burning velocity as a metric to screen refrigerants for fire risk. This study reports measurements of difluoromethane/air flame propagation for equivalence ratios from 0.9 to 1.4 and difluoromethane/(21%-oxygen/79%-argon) flame propagation for equivalence ratios of 0.7 and 1.3 in a spherical, constant volume device. Experimental burning velocities produced with the aid of an optically thin radiation model are increased by about 18 % compared to an adiabatic model. The influence of flame stretch on burning velocity is characterized by the magnitude of the product of Markstein and Karlovitz numbers. Lean difluoromethane/air flames at small radii are most affected by stretch, which reduces the burning velocity compared to the stretch-free value. Rich flames in air and those with argon are negligibly affected by stretch at small to moderate radii and large radii rich flames in argon are unstable. Using the Markstein-Karlovitz product to define a threshold for stretch-free flames, power law surfaces are fit to the experimental burning velocities in air to extract laminar flame speeds at (298 K, 101 kPa) and (400 K, 304 kPa). Depending on the choice of threshold, reported laminar flame speeds for lean flames in air at (298 K, 101 kPa) can vary by 9 % while those at (400 K, 304 kPa) can vary by 3 %. Greater uncertainty is incurred when extrapolating power law surfaces far from the conditions realized in experiments, e.g., at (298 K,101 kPa) typically 4 %, and is minimized when interpolating between experimental results, e.g., at (400 K, 304 kPa) typically less than 1 %.
Proceedings of the Combustion Institute
R-32, refrigerant flammability, burning velocity, difluoromethane, CH2F2, spherically expanding flame, thermal radiation, flame stretch, constant volume combustion chamber