Search

International audience; In-flame optical characterization of soot is of vital importance to understand soot formation mechanisms as well as to develop and validate accurate soot models. The present work introduces an unconventional methodology adapted to laminar axisymmetric flames that avoids the issue of variable measurement vol- ume with varying scattering angle in the existing light scattering techniques and thus enables the deter- mination of aggregate size with a higher spatial resolution. Coupled with multi-wavelength line-of-sight attenuation measurements, the proposed Horizontal Planar Angular Light Scattering at 532 nm was found able to provide radial profiles of aggregate size, number and diameter of primary spheres, soot volume fraction, and number density in a laminar axisymmetric coflow ethylene/air diffusion flame established over a Gülder burner. The spatial variation of soot optical properties associated with soot maturity was considered in data analysis.

Optical characterization of soot (e.g., Laser Induced Incandescence, LII) and its impact on climate largely depend on light absorption. As soot grows, its morphology changes affecting its optical or radiative properties. Here, the impact of soot maturity on its light absorption is investigated by coupling Discrete Element Modeling (DEM) with Discrete Dipole Approximation (DDA) during soot surface growth and agglomeration. The Mass Absorption Cross-section, MAC, of nascent and mature soot agglomerates is estimated by DDA and validated against atomistic point dipole interactions and mesoscale DDA calculations. Using a refractive index, RI, for mature soot yields constant average and absorption function overestimating the nascent soot light absorption up to 75%. The RI is interpolated between those of nascent and mature soot for wavelengths, λ = 532 and 1064 nm to account for quantum confinement and evolving number of clustered sp2-bonded rings that affect the optical band gap, Eg. This results in excellent agreement of the DEM-derived evolutions of , and ratio R = E ( λ = 532 nm ) > E ( λ = 1064 nm ) > with the corresponding LII measurements in methane and ethylene premixed flames. The nascent soot Eg decreases during aging and agglomeration, increasing and by 65%. The R decreases from 1.34 to 0.95 by aging and coagulation and slowly converges to the asymptotic 0.89 of mature soot measured in diffusion flames. The good agreement between DEM and LII data confirms that soot dynamics by surface growth and agglomeration strongly correlate with soot maturity, composition and RI that are essential for quantifying soot light absorption and scattering.

A novel diagnostic is proposed to characterize the maturity of soot particles in a laminar axisymmetric coflow ethylene diffusion flame in terms of the spectral dependence of soot absorption function. The method relies on the combination of line-of-sight attenuation (LOSA) and emission measurements at four wavelengths (500, 532, 660 and 810 nm). The analysis of the measured signals enables the determination of soot temperature, soot volume fraction, soot maturity and the contribution of soot scattering to extinction. The analysis of extinction and emission measurements considers the spatial variation of soot optical properties. The introduction of a maturity index allows the evaluation of soot maturity based on the spectral variation of the soot absorption function. The maturity index is correlated with the organic or the mature soot content and finally in terms of the absolute value of absorption function at 810 nm. The methodology is validated using a set of synthetic spectral LOSA and emission signals representing experimental measurements based on numerical results obtained using the CoFlame code. A sensitivity analysis of the Abel inversion is also performed to properly address the effect of deconvolution procedure. Finally, the proposed method is applied to analyze the experimental data of spectrally-resolved LOSA and emission acquired in a laminar axisymmetric coflow ethylene diffusion flame established on a Gulder burner. The two-dimensional distributions of soot temperature, soot volume fraction, soot maturity, and the ratio of total scattering to absorption are determined. Mature soot particles are found on the top of the flame in the centerline region and also in the outer edge of the flame wing displaying strong gradients.

The impact of organic carbon (OC) on the light absorption of soot is determined by discrete element modeling coupled with the discrete dipole approximation for computing the scattering of radiation by soot particles. The mass absorption cross-section (MAC) of soot is used widely to determine its light absorption. Typically MAC is obtained from the mass average refractive index of OC and elemental carbon (EC) with large C/H that make up mature soot. As such, MAC can be overestimated by a factor of 3 in fuel-rich flames where newly-formed young soot contains EC with small C/H and OC that predominantly scatters light reducing its absorption by soot. Here a relation for the soot refractive index is derived accounting for soot morphology, maturity and OC content through its band gap at wavelength, λ = 266–1064 nm. Using this relation, the MAC of soot containing OC (up to 50 wt%) is in excellent agreement with carbon black, graphene and soot data. This confirms that soot morphology, maturity and OC content greatly influence light absorption during characterization of in-flame and freshly-emitted soot by laser induced incandescence and light extinction, especially in fuel-rich flames, and need to be properly accounted for in the soot refractive index., Carbon, 172, ISSN:0008-6223