F.J. Gómez-Moreno

Turbulent transition in impactor jets and its effects on impactor resolution

Abstract A review of a number of widely used impactors suggests that the poorer performance often seen in some of the stages is due to the onset of turbulence in the jet at too large values of either the Reynolds number Re, or the nozzle-to-collector distance L. This phenomenon is studied here by measuring the collection efficiency versus Stokes number curves η(S) of a low-pressure thin-plate-orifice impactor as a function of Re and L (measured in units of the orifice diameter dn). A drastic broadening of the η(S) curve is observed in the vicinity of a critical Reynolds number Re ∗ (L/d n ) . Re ∗ increases at diminishing L/dn, taking values near 800 and 400 for L/dn of 3.1 and 4, respectively. No transition is seen at L/dn=1 or 2, even at Re as high as 2700. This transition in the jet modifies both the high and the low S tails of the η(S) curve. It should be distinguished from a previously studied turbulent transition of the boundary layer near the collector plate, which arises at much larger Reynolds numbers, changes only the low Stokes number tail of the η(S) curves and disappears when using small collector plates. A specialized experimental apparatus is used to provide an initial jet with very low turbulence level, as well as to isolate incipient turbulence effects from other mechanisms leading to broadening of the η(S) curves. The particles are brought very close to the axis via aerodynamic focusing, while particle capture by Brownian diffusion is offset with a repulsive electric field. Free-stream turbulence ahead of the impactor nozzle is eliminated by passing only a small fraction of the flow through the critical orifice and the focusing lenses. The remaining gas required to attain jet Reynolds numbers up to 3700 is introduced laminarly and axisymmetrically as sheath air through an outer porous wall right before the impactor nozzle. At Re in the range of a few tens, a strong increase of the critical Stokes number with increasing L/dn is observed.

Characterization of particulate emissions during pyrolysis and incineration of refuse derived fuel

Abstract Particle emissions generated during different stages of pyrolysis/incineration of refused derived fuel at the outlet of a pilot scale incinerator were characterized for mobility (with SMPS), total mass concentration (by filters) as well as by an optical particle counter and an impactor. Samples were extracted at a gas temperature 600°C and cooled to about 150°C. Total mass reached its peak of about 30 mg / Nm 3 during pyrolysis following the addition of fresh fuel, and dropped to a minimum around 2 mg / Nm 3 during after-combustion, as would be expected. Emissions during all stages were exclusively in the submicron range with number concentrations of 10 5 – 10 6 cm −3 and median diameters between 0.05 and 0.15 μm . The upper ends of those ranges were attained during the transition from pyrolysis to combustion. No particles above 1μm were detected with any of the instruments.

Continuous aerosol size spectrometry with a variable-orifice impactor

An aerosol size spectrometer based on a round-jet impactor of continuously variable geometry is tested using an iris diaphragm as the nozzle. The impactor is run by fixing the nozzle-to-plate distance L, the aerosol mass flow rate m′, and the volumetric pumping capacity Q, which results in a nearly fixed downstream pressure at variable nozzle diameter dn. Tests are carried below 1.3 and 2.5 mm. The size spectrometer has an excellent resolution when the jet Reynolds number and nozzle-to-plate distance are kept within the ranges 175 ⩽ Re ⩽ 700; 0.75 ⩽ Ldn ⩽ 3.33. The instrument may also be run at atmospheric pressure, with an estimated lower size range of 1.7 μm.