A proof-of-concept study on characterization of primary particle size distribution and thermal accommodation coefficient of soot aggregate by using conduction-dominated laser-induced incandescence

• Published in: Optics and lasers in engineering 2021-06, Vol.141, p.106572, Article 106572
• Main Authors: Zhang, Jun-You, Qi, Hong, Gao, Bao-Hai, Ren, Ya-Tao, Zhou, Lei
• Format: Article
• Language: English
• Subjects: conduction-dominated LII; laser-induced incandescence; primary particle size distribution; soot aggregate; thermal accommodation coefficient
• ISSN: 0143-8166; 1873-0302
• DOI: 10.1016/j.optlaseng.2021.106572

•By defining the fractal-dependent thermal accommodation coefficient, a simplified LII model for conduction-dominated LII is proposed•The feasibility of inferring primary particle size distribution and fractal-dependent thermal accommodation coefficient is studied•The inversion is independence of most model parameters required in the general LII model•The trends in the variation of two of the parameters being fitted strongly depend on prior parameter uncertainty Laser-induced incandescence (LII) is a powerful tool for in-situ measurement of soot aggregate properties, including the primary particle size distribution (PPSD). When integrating the shielding factor in the LII model to quantify the aggregate shielding effect, thermal accommodation coefficient (TAC) and shielding factor appear as a product. It is impossible to determine them independently without their accurate prior values. To overcome this problem, this study defines a new parameter fractal-dependent TAC (αT, f) which combines the overall impact of both TAC and shielding factor on the all detected aggregates. Given the ill-posed inverse problem is sensitive to the uncertainty from model parameters, inferring αT, f simultaneously with μdp and σdp (two characteristics parameters of PPSD) from conduction-dominated LII has the remarkable advantage of being independent of most LII model parameters. To evaluate the feasibility of the inversion from the respective of model parameter uncertainty, three relatively simple two-dimensional (2D) invers problems are investigated: (1) inferring μdp and σdp when prior αT, f is biased; (2) inferring μdp and αT, f when prior σdp is biased; (3) inferring σdp and αT, f when prior μdp is biased. The results and analysis presented in this study provide a theoretical basis for inferring μdp, σdp, and αT, f from conduction-dominated LII.