Effective temperatures of pulsed-laser-heated soot particles were derived from their thermal emission intensities using optical pyrometry in a laminar ethylene coflow diffusion flame. The present study concerns conditions of relatively low laser fluences under which soot particles are heated to temperatures below 3500 K to avoid complications of soot particle vaporization in both the experiment and the numerical calculations. The current nanoscale heat transfer model for laser-induced incandescence (LII) of soot was improved to account for the effect of the fractal structure of soot aggregates on the rate of heat loss to the surrounding gas. Mean primary soot particle diameter and mean aggregate size at the location of measurement were determined using the technique of thermophoretic sampling/transmission electron microscopy analysis. Numerical calculations based on the improved LII model were conducted to predict the soot particle temperature with known gas temperature, the heat conduction coefficient, the primary particle diameter, and the mean aggregate size, as well as values of assumed soot absorption function E(m) and the thermal accommodation coefficient of soot alpha. The experimentally observed soot temperature history, characterized by the peak value and the temporal decay rate, cannot be reproduced numerically using the values of E(m) and alpha found in the literature. By utilizing the experimental peak temperature and temporal decay rate new values of E(m) at 1064 nm and a were determined. Uncertainties in the derived values of E(m) and alpha caused by the uncertainty in the primary soot particle diameter and the mean aggregate size were analyzed. A novel method to determine the values of the soot absorption function E(m) and the thermal accommodation coefficient alpha was developed in the present study. Crown Copyright (C) 2003 Published by Elsevier Inc. on behalf of The Combustion Institute. All rights reserved.