Search Results

Studies have been conducted at Michigan Technological University (MTU) for over twenty years on methods for characterizing and controlling particulate emissions from heavy-duty diesel engines and the resulting effects on regulated and unregulated emissions. During that time, control technologies have developed in response to more stringent EPA standards for diesel emissions. This paper is a review of: 1) modern emission control technologies, 2) emissions sampling and chemical, physical and biological characterization methods and 3) summary results from recent studies conducted at MTU on heavy-duty diesel engines with a trap and an oxidation catalytic converter (OCC) operated on three different fuels. Control technology developments discussed are particulate traps, catalysts, advances in engine design, the application of exhaust gas recirculation (EGR), and modifications of fuel formulations.

The effects of a catalyzed particulate filter (CPF) and Exhaust Gas Recirculation (EGR) on heavy-duty diesel engine emissions were studied in this research. EGR is used to reduce the NOx emissions but at the same time it can increase total particulate matter (TPM) emissions. CPF is technology available for retrofitting existing vehicles in the field to reduce the TPM emissions. A conventional low sulfur fuel (371 ppm S) was used in all the engine runs. Steady-state loading and regeneration experiments were performed with CPF I to determine its performance with respect to pressure drop and particulate mass characteristics at different engine operating conditions. From the dilution tunnel emission characterization results for CPF II, at Mode 11 condition (25% load - 311 Nm, 1800 rpm), the TPM, HC and vapor phase emissions (XOC) were decreased by 70%, 62% and 62% respectively downstream of the CPF II.

A high-fidelity multi-step global kinetic Selective Catalytic Reduction (SCR) model which can predict SCR performance in engine exhaust systems is desirable for optimizing the SCR system, designing on-vehicle control systems and on-board diagnostic (OBD) functions. In this study, a Cu-zeolite SCR catalyst in the exhaust of a 2010 Cummins 6.7L ISB diesel engine was experimentally studied under both steady-state and transient conditions. Steady-state engine tests spanned SCR inlet temperatures from 250 to 400°C with a constant space velocity of 60 khr-1. A 1-D Cu-zeolite model originally developed from reactor data was improved and calibrated to the steady-state engine experimental data. The calibrated model is capable of predicting NO/NO₂ reduction, NH₃ slip, and NH₃ storage associated phenomena.

Modeling of diesel exhaust after-treatment devices is a valuable tool in the development and performance evaluation of these devices in a cost effective manner. Results from steady state loading experiments on a catalyzed particulate filter (CPF) in a Johnson Matthey CCRT®, performed with and without the upstream diesel oxidation catalyst (DOC) are described in this paper. The experiments were performed at 20, 40, 60 and 75% of full load (1120 Nm) at rated speed (2100 rpm) on a Cummins ISM 2002 heavy duty diesel engine. The data obtained were used to calibrate one dimensional (1-D) DOC and CPF models developed at Michigan Technological University (MTU). The 1-D 2-layer single channel CPF model helped evaluate the filtration and passive oxidation performance of the CPF. DOC modeling results of the pressure drop and gaseous emission oxidation performance using a previously developed model are also presented.

Automotive exhaust emissions of polynuclear aromatic (C16+) hydrocarbons (PNA) were reduced by 65-70% by current emissions control systems and by about 99% by two experimental advanced emission control systems. At a given level of emission control, PNA emission was primarily controlled by fuel PNA content through the transient storage of PNA in engine deposits and their later emission under more severe engine operating conditions. A relatively minor contribution to PNA emission was made by PNA synthesized from lower molecular weight fuel aromatics, particularly C10-C14 aromatics. Deposit-related PNA emissions were linearly correlated with the PNA content of the deposit formation fuel. In comparison with a fuel of field-average PNA content (0.5 ppm benzo(a)pyrene), a field-maximum fuel (3 ppm) contained 4 to 7 times as much of three major PNA species and caused 3 to 5 times higher emissions of these species.

In 1972 and 1973, the CRC-APRAC Program Group on Diesel Exhaust carried out a fourth program to evaluate techniques for measuring concentration of hydrocarbon in diesel exhaust. The first two programs were conducted in 1967 and 1968. In them, a single cylinder diesel engine was shipped among 13 laboratories and each laboratory measured hydrocarbon emissions by their own method. Agreement among laboratories (instruments) was poor in both programs. The third program was conducted in 1970 at one laboratory on one engine. This time, agreement among instruments was much improved from the earlier programs. The fourth program was conducted to confirm these later results. In it, a multi-cylinder diesel generating set was circulated among 15 participating laboratories, and each laboratory measured exhaust hydrocarbon by methods that complied with SAE Recommended Practice J215, “Continuous Hydrocarbon Analysis of Diesel Exhaust.”

Protocols for sampling and analysis of particulate and gaseous-phase diesel emissions were developed to characterize the chemical and biological effects of using ceramic traps as particulate control devices. A stainless-steel sampler was designed, constructed, and tested with XAD-2 sorbent for the collection of volatile organic compounds (VOC). Raw exhaust levels of TPM and SOF and mutagenicity of the SOF and VOC were all reduced when the traps were used. Hydrocarbon mass balances indicated that some hydrocarbons were not collected by the sampling system and that the proportions of collected SOF and VOC were altered by the use of the traps. SOF hydrocarbons appeared to be derived mainly from engine lubricating oil; VOC hydrocarbons were apparently fuel-derived. There was no apparent effect on SOF mutagenicity due to either sampling time or reexposure of particulate to exhaust gases.

The CRC-APRAC CAPI-1-64 Odor Panel was formed in 1973 to assess an instrumental measurement system for diesel exhaust odor (DOAS) developed under CRC-APRAC CAPE-7-68 by Arthur D. Little, Inc. Four cooperative studies were conducted by nine participating laboratories using common samples. The objectives of these studies were to define the DOAS system variables and to validate and improve the sampling and collection procedures. A fifth study, serving as a review of each analysis step, showed that analysis of common derived odorant samples could be conducted within acceptable limits by the participating laboratories. Three in-house sampling system design and operating parameter studies were conducted simultaneously with the cooperative work. The combined findings from the in-house and cooperative studies led to a tentative recommended procedure for measuring diesel exhaust odor.

Methods available for measuring hydrocarbons in diesel exhaust were evaluated by the CRC-APRAC Program Group on Diesel Exhaust Composition during 1967-1970. Early tests showed distressingly large variations from instrument to instrument and undesirably large variations among repeated measurements by one instrument. Instrument quality and operator competence were better in later tests and agreement among instruments was relatively good and errors within instruments were small. Current techniques appear acceptable for engineering measurements. No further cooperative work is planned by CRC at present, but techniques for measuring hydrocarbons in diesel exhaust will be reappraised periodically.

This is the fourth in a series of tests conducted as a Coordinating Research Council cooperative program to evaluate the measurement methods used to analyze diesel exhaust gas constituents. A multi-cylinder engine was circulated to 15 participants who measured emissions at three engine conditions. All 15 participants measured nitric oxide and carbon monoxide with several laboratories measuring nitric oxide by both NDIR (Non-Dispersive Infrared) and CHEMI (Chemiluminescence). Some participants also measured carbon dioxide, nitrogen dioxide, oxygen, and unknown span gases. The test results are compared with the Phase III cooperative tests which involved simultaneous measurement of emissions by participants. The precision of the results was poorer in Phase IV than Phase III.

A Coordinating Research Council cooperative program was conducted to evaluate the measurement methods used to analyze nitric oxide and carbon monoxide in diesel exhaust. Initially, a single-cylinder test engine was circulated among participants with poor results. Tests were then conducted at one site using a multicylinder diesel engine. Six organizations participated in the program. Exhaust analyses were conducted at steady-state engine conditions and on a 3 min cycle test. Span gases of unknown concentration were also analyzed. The participants results varied but averaged less than ±5% standard deviation both within (repeatability) and among (reproducibility) the instruments. The short cycle test was in good agreement with the steady-state measurements. No significant difference in the use of Drierite, nonindicating Drierite, or Aquasorb desiccants was evident in sampling system tests.

A cooperative test program was conducted by the CRC-APRAC CAPI-1-64 Composition of Diesel Exhaust Program Group to evaluate the technical aspects of a proposed EPA recommended Heavy Duty Diesel Emission Measurement and Test Procedure. The proposed changes affected the sampling configurations and the types of instruments used. Six participants studied the effects of a number of variables on the proposed changes and evaluated some alternative systems that included both CHEMI and NDIR instruments. The tests were conducted at one site using a multi-cylinder engine operating on the 13-Mode Cycle. Equivalency of systems was demonstrated and the best performance was obtained with a special NDIR system.

A microprocessor controlled lubricating oil cooling system of truck diesel engine was designed to minimize the sump oil temperature fluctuation during start-up and nonsteady engine operations. Model reference adaptive control method is utilized in the control system design. The analysis involved in the design of the microprocessor controlled oil cooling system, and the applications of a special vehicle-engine-cooling system (VEC) computer simulation code in the implementation and testing of the model reference adaptive control strategy are described. Using the VEC simulation code, the performance of the microprocessor controlled oil cooling system and the conventionally controlled oil cooling systems were compared for the ATB, temperature disturbances, and cold weather transient tests. An explanation of each test, as well as a review of the results of comparison tests are presented.

A thermal management system for heavy duty diesel engines is presented for maintaining acceptable and constant engine temperatures over a wide range of operational conditions. It consists of a computer controlled variable speed coolant pump, a position controlled thermostat, and a model-based control strategy. An experimentally validated, diesel engine cooling system simulation was used to demonstrate the thermal management system's capability to reduce power consumption. The controller was evaluated using a variety of operating scenarios across a wide range of loads, vehicle speeds, and ambient temperatures. Three metrics were used to assess the effects of the computer controlled system: engine temperature, energy savings, and cab temperature. The proposed control system provided very good control over the engine coolant temperatures while maintaining engine metal temperatures within a desired range.

A computer simulation program was developed to simulate the thermal responses of an on-highway, heavy duty diesel powered truck in transient operation for evaluation of cooling system performance. Mathematical models of the engine, heat exchangers, lubricating oil system, thermal control sensors (thermostat and shutterstat), auxiliary components, and the cab were formulated and calibrated to laboratory experimental data. The component models were assembled into the vehicle engine cooling system model and used to predict air-to-boil temperatures. The model has the capability to predict real time coolant, oil and cab temperatures using vehicle simulation input data over various routes.

Optimizing the performance of the aftertreatment system used on heavy duty diesel engines requires a thorough understanding of the operational characteristics of the individual components. Within this, understanding the performance of the catalyzed particulate filter (CPF), and the development of an accurate CPF model, requires knowledge of the particulate matter (PM) distribution throughout the substrate. Experimental measurements of the PM distribution provide the detailed interactions of PM loading, passive oxidation, and active regeneration. Recently, a terahertz wave scanner has been developed that can non-destructively measure the three dimensional (3D) PM distribution. To enable quantitative comparisons of the PM distributions collected under different operational conditions, it is beneficial if the results can be discussed in terms of the axial, radial, and angular directions.

In this paper, a model-based linear estimator and a non-linear control law for an Fe-zeolite urea-selective catalytic reduction (SCR) catalyst for heavy duty diesel engine applications is presented. The novel aspect of this work is that the relevant species, NO, NO2 and NH3 are estimated and controlled independently. The ability to target NH3 slip is important not only to minimize urea consumption, but also to reduce this unregulated emission. Being able to discriminate between NO and NO2 is important for two reasons. First, recent Fe-zeolite catalyst studies suggest that NOx reduction is highly favored by the NO 2 based reactions. Second, NO2 is more toxic than NO to both the environment and human health. The estimator and control law are based on a 4-state model of the urea-SCR plant. A linearized version of the model is used for state estimation while the full nonlinear model is used for control design.

A study was conducted to assess the effects of a water-diesel fuel emulsion with and without an oxidation catalytic converter (OCC) on steady-state heavy-duty diesel engine emissions. Two OCCs with different metal loading levels were used in this study. A 1988 Cummins L10-300 heavy-duty diesel engine was operated at the rated speed of 1900 rpm and at 75% and 25% load conditions (EPA modes 9 and 11 respectively) of the 13 mode steady-state test as well as at idle. Raw exhaust emissions' measurements included total hydrocarbons (HC), oxides of nitrogen (NOx) and nitric oxide (NO). Diluted exhaust measurements included total particulate matter (TPM) and its primary constituents, the soluble organic (SOF), sulfate (SO42-) and the carbonaceous solids (SOL) fractions. Vapor phase organic compounds (XOC) were also analyzed. The SOF and XOC samples were analyzed for selected polynuclear aromatic hydrocarbons (PAHs).

Diesel particulate matter in both the diluted and undiluted state is subject to the processes of coagulation, condensation or evaporation, and nucleation which causes continuous changes in its physical characteristics. The Electrical Aerosol Analyzer (EAA) is used to measure the diesel particle size distribution in the MTU dilution tunnel for a naturally aspirated direct-injection diesel engine operated on the EPA 13 mode cycle. The design and development of accurate and repeatable sampling methods using the EAA are presented. These methods involve both steady-state tunnel and bag measurements. The data indicate a bimodal nature within the 0.001 to 1 μm range. The first mode termed the “embroynic mode” has a saddle point between 0.005 to 0.015 μm and the second mode termed the “aggregation mode” lies between .08 to .15 μm for the number distribution.

The diesel engine emits exhaust particles that pose a unique set of measurement requirements. To document the state-of-the-art of measurement technology and to improve measurement quality, the Smoke and Particulate Panel of the Diesel Exhaust Composition group of the Coordinating Research Council reviewed published literature and particulate-sampling data generated by panel members to identify (1) the effects of key sampling parameters on measured particulate mass, (2) the causes of measurement variability, (3) the effects of dilution system design on particulate mass measurement, and (4) promising real-time mass measurement methods. The panel found greater measurement difficulty associated with particulates than for gaseous pollutants because of engine-produced variations, the sensitivity of measured particulate mass to dilution parameters, and random errors in the independent measurements which comprise a particulate measurement.