A range of low-temperature and isothermal magnetic measurements are used to identify the secondary ferrimagnetic mineral (SFM) grain sizes in 10 representative soil samples from Wales. A comparison of percentage frequency-dependent susceptibility ( zFD percentage) and low-temperature remanence measurements shows that they are sensitive to different ranges of superparamagnetic (SP) grains. The relative distributions of SP grains and stable single domain (SSD) grains are similar in nine of the samples. Typical distributions for soils dominated by SFMs are %20-30 per cent SSD and 70-80 per cent SP. Multidomain (MD) grains were not detected in the samples studied. There is evidence that some soils contain significant numbers of ultrafine SP grains <0.010 pm that are not detected by low-temperature remanence measurements at 20 K and which will have the effect of depressing values of low-field susceptibility (xLF) and zFD percentage. A mixing model suggests that zFD percentage may be used semiquantitatively to estimate the proportion of SP grains in a sample. The positively skewed grain-size distributions strongly suggest a mechanism of SFM formation that is driven by processes at the m scale, thus supporting weathering and fermentation as controlling processes, rather than the degradation of SSD bacterial magnetosomes and primary minerals.

Conventional computational fluid dynamics (CFD) methods for simulating the flow through automotive exhaust catalysts assume a monolith resistance based on one-dimensional laminar flow. This underpredicts the flow maldistribution in the monolith. Incorporation of an additional pressure loss accounting for entrance effects improves predictions for the maximum flow velocity within the substrate.

No abstract

Lean burn after-treatment systems are the current focus for reducing emissions from diesel exhaust. The trend is for commercial CFD packages to use a single channel modelling approach. Due to computational demands, this necessitates specification of representative channels for modelling, implying prior knowledge of the flow field. This paper investigates a methodology for applying the porous medium approach to lean burn after-treatment systems. This approach has proved successful for three-way catalysis modelling and has the advantage that the flow field is predicted. Chemical kinetic rates for NO x trapping and regeneration in the model are based on information available in the open literature. Similarly, filtration information based on mass accumulation and soot combustion kinetics are also readily available. Modification of the source terms in a commercial CFD package enables prediction of trapping and release of NO x . This is an effective way to model a NO x trap after-treatment system and provides simultaneous 3D modelling of the flow field. With diesel, particulate filtration is required. In the case of particulate traps, however, because of channel geometry, some assumptions are necessary for use of the porous medium approach and these are discussed in this paper. Both models produce qualitatively correct output and have parameters that can be tuned to conform to experimental data. Data to validate the NO x trap model is to be measured. The particulate trap model, on the other hand, is a feasibility study for modelling the complete diesel after-treatment system using the porous medium approach.

No abstract