Decreased particle concentrations in vehicle exhaust due to the use of particulate traps pose new challenges to particle measurement methods in terms of sensitivity and repeatability. In the present paper nanoparticle measurement methods are compared to gravimetric and coulometric analysis of filter samples from CVS sampling with special attention to the consequences for the measurement of DPF efficiency. Repeatability and sensitivity of the methods are assessed, and the influence of dilution and sampling is investigated.
It is shown that the repeatability of nanoparticle measurement methods is of the same order as that of PM measurement. In terms of sensitivity, nanoparticle methods are by three orders of magnitude better than today's PM and EC measurement. This provides a much better signal-to-noise ratio, especially when particle emissions are very low. While a gravimetric “zero” measurement may still contain a lot of nanoparticles, a “zero” with a nanoparticle method is by a factor 1000 closer to particle free air. Therefore, the distinction of high-efficiency DPF from other aftertreatment systems hardly deserving the label “filter”, is inseparably tied to the application of sensitive particle measurement methods. In order to improve the repeatability of those methods, existing calibration concepts have to be thoroughly developed.
Addressing the issue of dilution and sampling it is once more demonstrated that condensation of vapors may lead to wrong conclusions about trap efficiency. Such erroneous results can be avoided by applying thermo-dilution, post-dilution thermo-conditioning or direct sampling, or by using material specific particle detection. A new problem occurs when testing trap equipped vehicles with standard CVS technique: In CVS systems the particle background of the dilution air is sometimes higher than the tailpipe particle emissions, thus making proper evaluation of the exhaust levels impossible.
Many nanoparticle instruments have a compact, rugged design at low cost, and they provide on-line response. As opposed to a CVS system, they do not require a roller dynamometer setup but can be mounted on a vehicle or truck. Due to these properties they are ideal candidates for in-use-compliance testing and even OBC instrumentation.