CLEERS Workshop 7

Seventh DOE Crosscut Workshop
on Lean Emissions Reduction Simulation
June 16th and 17th, 2004
Detroit Diesel, 13400 Outer Drive West, Detroit, MI 48239

Summary Report by Stuart Daw, ORNL

This was the seventh in a series of CLEERS public workshops. The meeting was hosted by Detroit Diesel at their Detroit training facility. Target topics for this workshop included the modeling and simulation of Diesel Particulate Filters (DPF's), Selective Catalytic Reduction (SCR), and Lean NOx Traps. Since the previous workshop had a major emphasis on LNT's, it was agreed that this workshop would focus more heavily on DPF and SCR.

The specific workshop objectives were:

Conduct an in-depth technical discussion on aftertreatment modeling, especially concerning DPF and SCR;
Review the status of development of the standard LNT characterization protocol and how that experience can be adapted for DPF and SCR;
Update the general CLEERS community on the status of the Focus Groups and generate useful feedback to the Focus Groups from that community; and
Identify up-to-date research priorities for DPF's, SCR, and LNT's and feed that information back to Diesel Crosscut Team and DOE.

There were 90 participants, including representatives from software companies, universities, emission controls suppliers, automotive and engine manufacturers, national laboratories, and consulting companies. Altogether, twenty eight presentations were given, with the following general breakdown:

DPF (10 talks)
- Invited
  - Prof. Konstandopoulos (APTL, CERTH/CPERI, Greece )
  - Dr. Merkel (Corning)
  - Dr. Gulati (Corning)
- 7 contributed
LNT (8 talks, all contributed)
SCR (8 talks)
- Invited
  - Prof. Burch (Queens University, Belfast)
- 7 contributed

A detailed agenda and downloadable copies of most of the presentations are available on the CLEERS website (www.cleers.org). Detailed notes about the presentations and discussion taken by John Hoard (Ford) are also available on the website.

Highlights from the technical discussions in each technology area included the following:

Diesel Particulate Filters

Standard characterization and modeling of DPF's are especially challenging because of the large variability and hysteresis in soot properties;
Several groups are involved in detailed modeling of soot cake development and filter media pore plugging, including Michigan Tech University, Pacific Northwest National Lab, and Centre for Research and Technology Hellas (Greece);
Several groups are involved in lower-order/system modeling of DPF performance including University of Wisconsin, University of Michigan, and Michigan Tech University;
The availability of key global rate parameters needed for performance simulation of DPF's is notably lacking; and
There is a widespread belief that although many details of DPF performance are currently hard to quantify in a consistent way, the DPF Focus Group should concentrate on developing some elements of a standard characterization protocol rather than trying to be too comprehensive at the beginning.

In regard to the last point, there was much discussion about the idea for developing a characterization procedure for DPF's that would be analogous to what is currently done for turbochargers. It was widely agreed that this analogy is useful for starting the discussion, but the process for quantifying DPF functionality is more complex than it is for turbochargers. However, engine companies still have a strong need to simplify the process of defining performance parameters that can be used for prediction and system integration.

Lean NOx Traps

GM/Chalmers and ORNL are developing and testing global kinetic models that appear to be useful for interpreting and guiding LNT testing;
Hysteresis is an important factor for LNT's but is less complex than for DPF's;
One of the most complex issues in modeling and simulating LNT's is the impact of different exhaust reductant species on the NOx regeneration efficiency;
An appropriate level of degreening is one of the most difficult issues involved in developing a consistent LNT characterization protocol; and
The current attempt of the LNT Focus Group to develop a standard protocol definition/demonstration is one of the most important CLEERS focus activities, because it can serve as a template for the other groups.

As for the DPF Focus Group, it appears that the most feasible approach is to begin defining at least some of the most important elements of an LNT protocol, while at the same time trying not to be too comprehensive.

Selective Catalytic Reduction

Mechanisms for HC-SCR can be very highly variable from catalyst to catalyst, with only a few reaction steps dominating;
Although it is possible to develop extremely detailed computational models for performance, details should only be added when clearly warranted (Occam's Razor);
High throughput trials at GM have demonstrated the value of this methodology for HC-SCR catalyst exploration and is leading to some promising candidates for further study;
Assessments of the economics for urea-SCR still show that this technology is favorable relative to LNT's, even considering all costs for the urea infrastructure;
One of the most challenging issues for urea-SCR simulation is handling the large number of byproduct species that are produced; and
There are several groups who have constructed global kinetic models for simulating the NOx performance of urea-SCR including the National Renewable Energy Laboratory, Johnson Matthey, and University of Michigan.

Other general feedback coming from the workshop discussions included the following points:

The basic focus group structure being used to address the key simulation issues for the different technologies appears to be successful in meeting the CLEERS objectives;
Interest in the public workshops is still very strong and the general format and timing of the workshops still seems to be appropriate;
Diesel engine companies continue to stress the current lack of accurate rate parameters needed for simulating realistic aftertreatment performance (and the need for CLEERS to assist in providing this information);
The highest priority for obtaining rate parameters information appears to have shifted over the last two years from LNT's to DPF and urea-SCR; and
Perhaps for historical reasons as well as the nature of the technology, the development of standardized procedures for measuring and quantifying the critical rate parameters appears to be furthest along for LNT's.

Back to Top

	Friday, September 10
	Home \| Contact us