1st CLEERS Workshop

Emphasized topic areas:

• Addressing the Full-System Context for Lean Exhaust Emissions Control

Objectives:

• Reduce duplication of effort among separate research groups
• Provide up-to-date information on emissions control simulation tools
• Enhance collaboration among industry, government and universities
• Improve recognition of key research priorities

Purpose

The purpose of previous, these and subsequent CLEERS workshops is to solicit input from prominent researchers in Government, Universities, and Industry on the current state-of-the-art in simulation of lean exhaust emissions control. This is the 1st Workshop in a series sponsored by the DOE Office of Heavy Vehicle Technologies (OHVT) and the DOE Diesel Crosscut Team. The results of the workshop discussions, especially the all-participant discussion at the end, will be documented for the DOE Office of Transportation Technology (OTT) for consideration in improving R&D priorities and technical focus in DOE-supported transportation programs.

About Workshop 1

The 1st CLEERS workshop was held at the National Transportation Research Center in Knoxville, Tennessee, on May 7-8, 2001. The central theme was “Addressing the Full-System Context for Lean Exhaust Emissions Control.”  The global issue of system-wide emissions component interactions was selected as the first topic in order to help clarify relationships among the various emissions simulation activities currently funded by DOE. In addition, system-wide simulation capabilities are a major priority for the Diesel Crosscut Team.The workshop agendas and copies of most presentations are available from this site. Workshop discussions have been summarized in a report to DOE, including specific recommendations to the Office of Transportation Technology (OTT) regarding R&D priorities and technical focus in DOE-supported transportation programs.

Planning for second and third workshops is now underway. It is expected that the next workshop will be held in October with a focus on modeling of “NOx adsorbers” and “Catalyzed Particulate Filters.” More details of this meeting will be presented here when information becomes available.

Organizing Subcommittee

R. Blint, General Motors
N. Hakim, Detroit Diesel
G. Singh, U. S.-DOE/OHVT/HQ
H. Kung, Northwestern University
C. Rutland, University of Wisconsin, Madison
S. Daw, Oak Ridge National Laboratory

The first CLEERS workshop was held at the National Transportation Research Center (NTRC) in Knoxville, Tennessee, on May 7- 8, 2001, with the central theme “Addressing the Full-System Context for Lean Exhaust Emissions Control.” The overall goal of this and subsequent CLEERS workshops is to solicit input from prominent researchers in government, universities, and industry on the current state-of-the-art in simulating lean exhaust emissions control. For the first workshop it was decided that the focus should be to better understand how the various components of emissions simulation fit together globally.

Workshop Participants and Agenda

The breakdown according to corporate affiliation was: four from automotive companies, six from diesel engine companies, twenty-two from national labs, five from DOE Headquarters, thirteen from universities, six from consulting and/or software companies, and three from fuel/emissions component suppliers. In all the categories but the last, the number of attendees was close to expected targets. More involvement from the emission control suppliers is still a high priority and will be the focus of increased subcommittee effort for succeeding workshops.

The two-day meeting agenda is posted elsewhere on this web site. Discussion on the first day focused on overview presentations by Gurpreet Singh and Dick Blint explaining the DOE and Crosscut Team perspectives on CLEERS and how those perspectives are connected to the workshop objectives. Following the introductory talks, six prominent representatives from industry (five engine and automotive companies and one emissions control component supplier) described their key concerns and needs in mini-presentations and then fielded questions and comments from the audience. After lunch, Dan Haworth, John Storey, and Kevin Sisken presented invited talks on current capabilities in modeling IC engine out emissions, experimental engine out hydrocarbon speciation, and cycle simulation of emissions, respectively. A fourth invited talk by Se Oh summarized General Motors’ modeling experience with three-way catalyst design and optimization. M.C. Lin also gave a contributed talk on a proposed new mechanism for NO formation that may be relevant for both in-cylinder combustion and post-engine aftertreatment.

The technical presentations on the second day began with an invited talk by Tony Markel on application of PSAT and ADVISOR to global emissions simulations. Seven contributed talks then followed, presented in order by Matthew Thornton, Ismail Celik, Jim Parks, Andre Boehman, Tariq Shamim, Tony McDaniel, Bob Weber, and Bruce Eichinger. These latter talks covered a range of simulation issues from full system analysis, to specific aftertreatment component performance, to heterogeneous surface kinetics, to catalyst design with quantum chemistry. The final talk of the workshop was an overview given by Jim Patten on plans to coordinate the DOE national labs in addressing research needs related to simulation issues for DOE-OTT programs. At the end of the second day, Ron Graves led a group discussion of all the participants to construct a list of recommendations for DOE and these are summarized and sent to DOE headquarters to guide future efforts in this direction.

Industrial Mini-Presentations and Panel Discussion

There were a number of themes consistently stressed by the industrial participants (including engine companies, automotive companies, and emission control suppliers) in their individual presentations and panel discussions. Among these themes were:

• Physical models for aftertreatment components (e.g., pressure drop, heat transfer) are more developed than chemical reaction models.
• Reasonably good models for sensors and exhaust conditioning components are already available commercially and internally to OEM’s.
• Commercial and in-house system integration models are extensively used by OEM’s, but their access to aftertreatment component models is significantly lacking.
• Many pre-competitive issues remain in the definition of aftertreatment component models. Good pre-competitive models will be strong on first principles, have easily accessible/adaptable code, and will readily interface with other models.
• Critical needs for fundamental information include chemical mechanisms and kinetics, adsorption/desorption, plasma chemistry, thermophysical properties, fuel effects, particulate formation, soot regeneration, aging, and poisoning.
• Better interactions are needed between OEM’s and aftertreatment component suppliers. In particular, there is a disconnect between the specifications typically given by OEM’s to suppliers and what they actually need for simulation purposes.
• Once aftertreatment component models are available, a formal sensitivity analysis apparatus/methodology would be helpful in identifying performance bottlenecks.
• Adequate software support is a critical feature for OEM users of component models.
• High near-term priorities include sulfur effects, aging effects, chemistry and kinetics for NOx adsorbers, PM trap regeneration, faster screening protocols for new candidate devices, shorter development time between bench testing and full system integration, models with easy to use interfaces, parallel development of component and system integration models, and standard data sets and system integration models for benchmarking aftertreatment component models.
• Engine-out conditions and speciation are important for developing and validating aftertreatment device models and this information will most likely have to be provided by engine manufactures.

Regarding the need for balance between low-order and detailed component models, Se Oh suggested a useful paradigm that resonated among the other industrial participants. Referred to as “Prater’s principle of optimum sloppiness”, the basic idea is that there is always an optimum degree of modeling detail in terms producing the maximum engineering value. More than the optimum level of detail produces a rising cost that cannot be compensated by the additional information produced. Less than optimal detail reduces cost but at a disproportionate reduction in usefulness. One potential benefit of CLEERS would be to help identify quantitative approaches for identifying where this optimal point is located.

University, Consultant, And National Lab Presentations

The university presentations included a wide range of topics from in-cylinder emissions generation to aftertreatment performance. From these presentations and subsequent discussions, it is clear that the universities have strong interest and capabilities in fundamental emissions modeling, experimental measurements (both engine scale and bench top), and assessment of novel aftertreatment technologies. The Diesel Crosscut Team concluded that participation is needed from universities in the CLEERS activity, either as direct collaborators with national labs and/or the industries or as independently supported sources of data and models. The previous experience of universities in engine modeling and experimental catalysis could certainly be an important resource.

The presentations by consulting companies revealed that there has already been considerable interest on their part in developing specialty software for fundamental modeling of relevant catalyst chemistry. Because of the proprietary nature of this software, however, it is difficult to judge how it might be applied to the type of pre-competitive R&D proposed for CLEERS. Key questions that remain to be answered include:

• The availability of consultant generated software for independent evaluations in dynamic and/or full-system simulations
• The applicability of such software for developing information about rate controlling steps and chemical mechanisms that can be directly applied to aftertreatment design and/or control
• The willingness of consulting companies to share their expertise in improving the value of the CLEERS database and benchmark component models

The potential benefits to CLEERS could certainly be substantial if mechanisms can be found for including consulting company tools and/or data in our resource libraries.

National lab presentations also confirmed the Crosscut Team’s assumption that the labs can offer a wide range of experimental and computational capabilities. The labs are obviously unique in their ability to focus large facilities and diverse technical disciplines on difficult problems. Based on the discussions, there have already been a number of tools developed at the labs that could be applied directly to the CLEERS activity, but it is insufficient to the needs. The presentation by Jim Patten provided a summary of how the capabilities of the various labs might be coordinated to address some of the issues identified for CLEERS. Specific lab capabilities to be exploited for these targets include fundamental experimentation and modeling, materials micro-characterization, high-performance computing, and development and implementation of advanced instrumentation.