23-07-2014, 03:26 PM
DIESEL ENGINE NOX REDUCTION VIA NITROGEN-ENRICHED AIR
DIESEL ENGINE NOX.doc (Size: 451 KB / Downloads: 33)
Abstract
It has long been known that NOx emissions from diesel engines can be reduced by reducing the peak temperatures of combustion. Dilution of the cylinder charge with inert gases is one method of lowering the peak cylinder temperatures. Nitrogen is an obvious diluent; however until recently its use was limited to stationary laboratory engines. Compact, high productivity, air separation membranes have recently been developed. These membrane modules provide means for generating nitrogen-enriched air (NEA) at the point of use, for example, under the hood of a diesel truck. NEA offers an attractive, clean alternative to dilution with exhaust gases.
NEA is generated by feeding the cooled, turbocharged air to the bore side of a hollow fiber membrane device. A pressure differential across the wall of the hollow fiber causes oxygen to permeate preferentially through the polymeric wall.
Thus as air flows along the length of the hollow fibers, it becomes slightly depleted of oxygen and enriched in nitrogen. The resulting NEA is fed to the intake manifold of the engine at only slightly lower pressure than the turbocharged air. The oxygen-enriched co-product (OEA) is simply vented to the atmosphere. The effectiveness of the NEA for NOx emissions reduction is also related to the composition of the NEA. Only slight enrichment is needed and NEA compositions in the 80% to 82 % nitrogen range prove to be very effective in NOx reduction.
Developments of the NEA technology have progressed beyond the laboratory engine scale. NEA is now being studied on a number of commercial engine platforms with good success. NOx emission reductions as high as 50% are being achieved on diesel engines
Introduction
Air separation membranes have become a practical means for on-site generation of nitrogen during the last 20 years. Major industrial gas companies utilize highly selective polymeric membranes for the removal of oxygen from compressed air to produce and deliver on-site
pressurized, high purity nitrogen gas. In the screening of polymers for such membranes it was recognized that there is a trade-off between selectivity (the relative rates at which gas species permeate across the membrane) and the permeability or the gas flux across the membrane.
Robeson observed that there appears to be an upper limit to polymeric membrane selectivity, the ability to distinguish between two gas species, that declines with increasing permeability, the rate at which a gas will transport through the membrane barrier. Figure 1 shows an empirical fit to the highest oxygen-nitrogen selectivity observed for polymers with increasing permeability for oxygen. As such, Robeson's Line seems to represent the natural or fundamental limits of air separation in polymeric membranes. When producing high purity nitrogen, it is desired that the membranes have the highest selectivity in order to maximize the amount of nitrogen produced from the air feed. Such membranes fall within the upper left region in Figure 1. To compensate for their low membrane flux, high purity nitrogen generation units are typically operated at high pressures (100 to 200 psig). Even so, the surface area of such membrane required would be such that they could not be fitted under the hood of a truck! By contrast, CMS membrane polymers lie in the mid-range and though less selective are more productive, having oxygen fluxes that are over two orders of magnitude higher than those
MEMBRANE CONSIDERATIONS
The basic principle of membrane operation and various designs and operating characteristics are described in Nemser et al [14]. The hollow fiber (HF) membranes and devices being employed in NEA applications are described in Figures 3 and 4. The HF membranes employed for NEA operate at substantially lower pressure differentials and much higher volumetric gas rates than those applied for industrial nitrogen generation. Thus they can have relatively thin walls, as can be seen in Figure 3. The perfluoropolymer layer on the outer surface of the porous polymer HF wall forms the separating layer across which oxygen permeates preferentially to nitrogen. The high flow rates for the NEA application necessitate large inside diameters and short fiber lengths in comparison to industrial membrane devices to minimize the pressure loss between the air feed and the NEA.
Figure 4 shows a schematic of an NEA membrane device. The pressurized air feed passes along the inside of the hollow fibers that are surrounded by atmospheric pressure on their outside
DEVELOPMENT PROGRESS
Application of NEA membranes for NOx reduction on diesel engines has progressed from the research phase into the field demonstration stage. It is presently receiving increasing scrutiny as a viable NOx reduction option in its march toward commercialization. Thus far, four engine platforms have been employed in the evaluation and demonstration of the NEA membrane technology. The first was a 3kW Ferryman single cylinder diesel research engine. The second involved a Volkswagen 81kW 1.9 liter TDI engine, on which it was also compared against exhaust gas recycle. The third was a Lister Peter diesel electricity generator on which a variety of NEA membrane schemes were investigated. The fourth level involves recent Caterpillar C-12
Conclusions
On the basis of steady-state engine tests on a number of engine platforms, the following observations are made:
• NEA supplied to four widely different diesel engines with membrane devices resulted in substantial reductions in NOx emissions in all cases.
• Like EGR, NEA results in some increase in the particulate emissions.