Morphological characterization of diesel soot agglomerates based on the Beer–Lambert law

A new method is proposed for the determination of the number of primary particles composing soot agglomerates emitted from diesel engines as well as their individual fractal dimension. The method is based on the Beer–Lambert law and it is applied to micro-photographs taken in high resolution transmission electron microscopy. Differences in the grey levels of the images lead to a more accurate estimation of the geometry of the agglomerate (in this case radius of gyration) than other methods based exclusively on the planar projections of the agglomerates. The method was validated by applying it to different images of the same agglomerate observed from different angles of incidence, and proving that the effect of the angle of incidence is minor, contrary to other methods. Finally, the comparisons with other methods showed that the size, number of primary particles and fractal dimension (the latter depending on the particle size) are usually underestimated when only planar projections of the agglomerates are considered.