Particle size distribution functions (PSDF) and mean particle sizes have been determined in a laminar premixed ethylene/air flame with three different experimental approaches: photo-ionization mass spectrometry (PIMS), scanning mobility particle sizing (SMPS), and laser-induced incandescence (LII). The main goal of this investigation was the crossvalidation of these three methods used at our institute for the determination of particle sizes in a great variety of flames or exhaust gases. We found good agreement between the three methods in the ranges where they are comparable as well as a complementary behavior for the different size ranges. PIMS and SMPS are able to measure the particle size distribution functions with good resolution. PIMS is favorable in detecting the smallest particles (<6 nm) and thereby able to detect even bimodal distributions of the soot precursor particles. SMPS and LII are suitable in the mid-and upper range of the particle sizes (>2 nm and >3 nm, respectively). LII offers the particular advantage of being a non-intrusive method. This makes it applicable in extreme environments, such as high pressure flames, as well as in very sensitive flames because no probe is needed.

Study of carbonaceous nanoparticles in premixed C 2 H 4 -air flames and behind a spark ignition engine, Combust. Flame 156 (2009) 791-800.The original publication is available at www.elsevier.com http://dx. AbstractNanoparticle size distributions and their concentrations were studied in atmospheric premixed ethylene/air flames using photo ionization mass spectrometry (PIMS) and total organic carbon (TOC) calibration supplemented by differential mobility analysis (DMA). Focus of this study is the evolution of nanoparticles as a function of height above burner (HAB) and of the C/O ratio of the unburned gases. It was found that especially particles of the cluster type exhibit a sharp concentration drop by more than two orders of magnitude within a narrow C/O window which is close to the sooting threshold. Using DMA a decline by two orders of magnitude was found. These results suggest that at best only small concentrations of nanoparticles should be formed significantly below the sooting threshold. As these conditions prevail in a homogeneously charged IC engine no or only very small nanoparticle emissions are expected in the exhaust gas. This was indeed found for a small Otto engine driving a power generator unit. Using flame nanoparticle profiles as standard, absolute concentrations for their emissions could be deduced. These data were supported by additional DMA measurements. The calibration using TOC did not completely match the one based on the condensation particle counter of the DMA apparatus.Contrary to this clear evidence of nanoparticles it is much more under debate whether a minimum C/O ratio comparable to the sooting threshold is required for the formation of these particles. In a series of papers by D'Alessio and coworkers [3, 16 -19] premixed ethylene/air flames (C/O = 0.4 to 0.9) were studied. For these flames the yellowish luminosity occurs for C/O >~ 0.6. By contrast, nanometer-sized particles, called transparent soot, were reported to