This paper details the multiscale methodology developed to analyze the formation of nanoparticles in a manner that makes it possible to follow the evolution of the structures in a chemically specific way. The atomistic model for particle inception code that combines the strengths of kinetic Monte Carlo and molecular dynamics is used to study the chemical and physical properties of nanoparticles generated in a premixed fuel-rich benzene flame, providing atomistic scale structures (bonds, bond angles, dihedral angles) as soot precursors evolve into a three-dimensional structure. Morphology, density, porosity, and other physical properties are computed. Two heights corresponding to two different times in the benzene flame, experimentally studied by Bittner and Howard [Proc. Combust. Inst. 18, 1105 (1981)], were chosen to examine the influence of different environments on structural properties of the particles formed. (c) 2006 American Institute of Physics.