Diesel Exhaust System Influences on Transient Particulate Emissions and Particle Size Distribution

The behaviour of particulate matter emissions from a Ford XLD 4T, passengercar diesel engine through a practical exhaust system in place was investigated during transient conditions, namely cold start and fast acceleration. Particulate emissions were measured at four sampling points through the exhaust system and the changes in particulate total mass concentration, total number concentration, particle size distribution and Carbon/SOF fractions were determined for various engine operating conditions.
Each cold start test consisted of a step-change cold-start with fast acceleration, reaching one of the following target operation conditions: Idle, 1500rpm - 15kW,
2250rpm - 15kW, 2250rpm-35kW and 3500rpm - 15kW. Two preconditioning procedures were designed to provide repeatable cold start tests. These consisted of a)
Idle operation for 4 hours the day before the test, followed by overnight soak; and b)10 minutes at high engine speed operation. Fast acceleration tests consisted of idle
preconditioning followed by a step change to the target operation conditions. The particulate matter changes through the exhaust system were shown to be
dependent on the previous operational history of the engine, idle conditions being effective at forming particle deposits. The amount of particulate deposited or blown
out from the exhaust system constituted a significant fraction of the total engine exhaust emissions in a significant number of cold start and acceleration tests. The
changes in particle concentration did not occur throughout the system in the same fashion. The catalyst produced predominantly particle number and mass reduction
and the second silencer was a more efficient particle collector than the first silencer.
From the first silencer, particles were resuspended more easily and in many more cases.
Regarding the chemical composition, high-load conditions produced lower Solvent Organic Fraction (SOF) than their low load counterpart. However, the SOF did not change significantly through the exhaust system.
Part of this work consisted of examining the use of the Electrical Low Pressure Impactor (ELP1) to estimate particulate mass emissions. It was observed that the
ELP1 tended to overestimate the number of particles in the large (>0.1 (im) size range. This greatly affected the conversion from total particle number concentration
to particle mass concentration. A correction based on comparison between the electrical and gravimetric methods (ELPI vs. Andersen Impactors) in the common size range for both techniques was proposed.

Transient and steady-state tailpipe emission factors, expressed as grams of particulate per unit of engine work in kWh, were calculated from the test results and
used to estimate the effect of transients on total cycle emissions in cycles with a different design from those followed in this work. The ELPI proved to be useful, yet limited for particle collection on Transmission Electron Microscope (TEM) grids in several size ranges. TEM images
of particles were analysed and their fractal dimension determined.