Conference report: SAE 2016 Congress
29 April 2016
The SAE 2016 World Congress was held on April 12-14 in Detroit, MI. The Congress included a three-day schedule of technical sessions, with about 1500 technical papers covering all areas of automotive technology, including a number of papers on engine and emission topics.
Please log in to view the full version of this article (subscription required).
Technical sessions on engine and vehicle emissions were opened with a review of recent regulatory and technical developments in vehicle emissions presented by Tim Johnson of Corning [Paper #2016-01-0919]. Some of the important regulatory developments include the European real driving emissions (RDE) standards for light-duty vehicles, with the conformity factor for NOx being ramped down to 1.5 by 2021. The RDE test, conducted on the road using portable analyzers, shows better correlation with real emissions for both criteria pollutants and for GHGs. There is an increasing difference between RDE and NEDC CO2 emission results—with RDE results 20-50% above NEDC—which indicates that most of the CO2 emission reduction officially reported from the EU light-duty vehicle fleet occurs only in the test lab, with almost no reduction in real world. RDE testing may also reverse the downsizing trend in gasoline direct injection (GDI) engines, as downsized turbocharged engines have higher RDE CO2 emissions than larger engines. In heavy-duty engines, a new round of lower NOx emission limits, in the range of 0.05-0.02 g/bhp-hr, is expected in California. The California ARB indicated that the regulation could be adopted in 2019 and implemented from 2023 to 2027. While this low NOx standard would be rather challenging on its own—after all, it represents up to 90% NOx reduction from the current level—it would also make it significantly more challenging to meet the Phase 2 GHG emission requirements for heavy trucks (expected to be finalized this year). The emission gap between various parts of the world seems to be closing faster, with China and India adopting Euro 6 equivalent emission requirements from 2019 and 2020, respectively. However, the enforcement of emission standards may present an even greater challenge in developing countries than it does in the EU or in the United States. This was illustrated by very interesting data from Brazil, where the total sales of automotive urea are only about 50% of what would be nominally required by SCR vehicles based on their fuel consumption.
Engine and emission control technologies continue to evolve at a fast pace. Gasoline engines are reaching a peak brake thermal efficiency (BTE) of 45%, while heavy-duty diesels reach 50% BTE and target 55%. In diesel aftertreatment technologies, one of the major challenges is the control of N2O emissions from SCR systems, to meet the N2O limits set in US EPA GHG regulations. This creates a new incentive to use vanadium SCR catalysts, which are among the best formulations for low N2O emissions. Due to concerns with vanadium volatility, the use of vanadia SCR catalysts in the US market is currently limited to applications without a DPF (i.e., some nonroad engines). New aftertreatment technologies under development include passive NOx adsorbers and gasoline particulate filters (GPF). The latter technology—already in commercial use in Europe, to control particle number emissions—may be used in the United States to meet the Tier 3/LEV III PM limits of 3 mg/mi and 1 mg/mi, while ensuring better control of PAH emissions that pass through the three-way catalyst in GDI vehicles.
Diesel Emission Aftertreatment. Johnson Matthey [M. Naseri; oral presentation only—no paper publication] discussed their ongoing development of an aftertreatment system for a heavy-duty engine to meet the anticipated California NOx limit of 0.02 g/bhp-hr. A paper presented during last year’s Congress discussed an SCR system that achieved 98% NOx conversion efficiency over the cold start FTP test. The system utilized a passive NOx adsorber (PNA) upstream of the SCR catalyst, thermal management using electric heater, and a pre-saturation of the SCR catalyst with ammonia. While this configuration could achieve the NOx performance target, the N2O emissions were 0.15 g/bhp-hr, or three times the standard. The development work presented this year examined two approaches to reduce N2O: (1) a hybrid Fe + Cu zeolite SCR catalyst and (2) a hybrid V + Cu zeolite catalyst. The SCR system that showed most promise included a PNA, followed by vanadia SCR catalyst coated over a DPF, and a flow-through Cu-zeolite catalyst. With the help of NH3 pre-saturation and thermal management, the system achieved 98% NOx reduction with reduced N2O. However, the NO2/NO ratio and other parameters still require optimization. The vanadia catalyst has a durability of 580°C, which would require passive DPF regeneration.
The use of an electric heater as a means of cold start temperature management in heavy-duty diesel engines was studied by Watlow [2016-01-0927]. The work included laboratory tests using a diesel generator powered by a 6.7 L Cummins engine with SCR aftertreatment, and a simulation study of an engine with DOC + DPF + SCR aftertreatment, targeting Euro VI emissions.
Diesel Particulate Filters. Mixed oxides of ceria with zirconia are often used as catalysts for direct soot oxidation in catalyzed diesel particulate filters. The impact of ceria-zirconia particle size on catalytic oxidation of soot was studied by researchers from CERTH/CPERI and the Aristotle University [2016-01-0968]. Two ceria-zirconia catalyst samples with different zirconia content were subjected to different milling protocols, to shift the catalyst particle size distribution. The produced catalysts were evaluated with respect to their soot oxidation activity. A multi-population kinetics model was used to describe the soot oxidation process and assess the effect of catalyst particle size. Three populations of soot were identified with different activation energy, likely due to different structures of the surface oxygen complexes.
New cordierite wall-flow DPF substrates for heavy-duty engines were developed by Corning [2016-01-0940]. The new DPFs feature asymmetric cell technology (ACT) and thin walls, to provide low thermal inertia and fast SCR catalyst light-off in SCR-on-filter applications. Through the use of asymmetric cells, the filters can provide up to 50% lower ash pressure drop. New aluminum titanate DPF substrates were also introduced by Sumitomo [Andrzej Sieminski]. The new configuration, dubbed ‘Micro Gear’, features a wavy inlet channel wall to increase the filtration area by 40%. This design can provide a low pressure drop and higher regeneration efficiency.
An interesting presentation by Daimler India [2016-01-0959] discussed DPF regeneration problems on Mercedes cars when using Indian BS III diesel fuel of 350 ppm sulfur limit. A thick plume of white smoke is often seen during regeneration, due to the release of sulfates from the catalyst washcoat. A laboratory investigation showed that DPF cracking can be caused by uncontrolled regenerations due to high oxygen levels and HC slip from the upstream DOC that is deactivated by sulfur. New engine calibrations were developed in order to control the A/F ratio during drop to idle and to desulfate the DOC+DPF system.
Paccar published a paper [2016-01-0928; no oral presentation] about an investigation into ash from DPFs returned from the field. The DPFs were tested on an engine dynamometer to determine the loss of performance, cleaned using various cleaning techniques, and tested again to evaluate the recovery of performance. Ash samples were analyzed to determine the ash composition and origin.
CTS provided an update on their (formerly FST) radio frequency (RF) sensor for the measurement of soot load in particulate filters [2016-01-0943]. A fast response version of the RF sensor was developed and evaluated [2016-01-0918]. Measurements of the instantaneous change in the filter PM loading state using the fast response RF sensor showed an impressive correlation with real time measurements of engine-out PM emissions using the tapered element oscillating micro-balance (TEOM).
Emission Characterization and Measurement. Researchers from the University of Minnesota studied particle emissions from light-duty vehicles during cold start [2016-01-0997]. The study characterized solid and semi-volatile particle number, mass, and size distributions during cold engine start-up, after an overnight soak, with an average ambient temperature of -8 ± 7°C. The average PN emitted during 180 s by GDI and PFI vehicles were 3.09×1013 and 2.12×1013 particles, respectively. Comparing to the 2017 Euro 6 NEDC limit on cumulative particles emitted over the entire test cycle (6.0×1011), most PFI and GDI vehicles exceeded this limit in 6-12 s after a cold start. In addition, US EPA Tier 3 particle mass requirements were exceeded for tested GDI vehicles due to their characteristically high concentration of accumulation mode particles. In comparison, diesel vehicles with DPF’s were the cleanest, with particle concentrations close to background levels.
The US EPA presented a study that characterized and quantified PM emissions from a GDI vehicle [2016-01-0992]. The study examined particulate emissions during transient operation in a recent model year Ford Escape vehicle equipped with a GDI engine. The tests were conducted over the FTP, the HWFET, and the US06 cycles. Measurements included particle size distributions using an EEPS instrument, as well as real-time soot emissions from an AVL MSS soot sensor.
The effect of ethanol on PM emissions from conventional (Tier 2) gasoline engines was analyzed by Thomas Darlington [2016-01-0996]. According to models developed by the US EPA, ethanol blends cause a slight increase in PM emissions. In their model, the EPA used the T50 and T90 distillation temperatures to characterize the fuel properties. However, the current study found that when the T70 temperature is added as a variable to the analysis, ethanol has no effect on PM emissions, suggesting that the EPA model may be deficient in that regard. A somewhat agitated discussion of the issue took place during the Q&A period, with the participation of EPA staff who attended the presentation.
The SwRI studied the emission impacts of their Dedicated EGR® (D-EGR) technology [2016-01-1006]. A number of regulated and unregulated emissions were measured from a light-duty vehicle equipped with D-EGR and compared with emissions from an identical production GDI vehicle without externally cooled EGR. HC emissions were observed to increase with the D-EGR strategy, and a shift in species was observed, with the less reactive fuel components dominating the emission spectrum. Changes in particulate emissions involved an increase in particle mass and number for lightly loaded drive cycles, and a net reduction for highly loaded cycles.
A number of presentations discussed emission measurement instruments and methods. Horiba presented an on-board PM analyzer potentially applicable to real driving emission testing [2016-01-0993]. The analyzer is a combination of a partial flow dilution system (PFDS) particulate sampler and a diffusion charge (DC) sensor for real-time PM signals. Acceleration of the vehicle caused uncertainty of flow measurement in the PM sampler. In another paper, a team from TNO and Horiba [2016-01-0975] reported on laboratory ammonia measurements from SCR catalyst systems. The experimental setup utilized simultaneously three Horiba FTIR analyzers. Both steady-state and dynamic responses were evaluated.
PEMS Panel. A panel of experts—moderated by Tim Johnson (Corning), with the participation of Les Hill (Horiba), Joe Kubsh (MECA) and Charlie Roberts (SwRI)—discussed in-use emission measurement using PEMS equipment. While the panel discussion was inspired by the VW emission scandal and the increasing focus on in-use emissions, the discussion centered around the European RDE testing requirements.
The panel opened with a summary of the RDE testing requirements, illustrated by the Horiba modular PEMS and presented by Les Hill. The RDE test must last from 90 to 120 minutes, and satisfy a number of conditions to qualify as a valid test. Therefore, RDE testing tends to be time consuming and costly. The measured species include NOx and CO (HC emissions are not measured), as well as particulates, including particle number. The conversion from concentrations to mass values must be based on a measurement of the exhaust gas flow rate (as opposed to using the OBD interface). The analyzer must be battery powered. The final RDE conformity factor for NOx, 1.5 effective from 2020, represents a very stringent limit, because the 0.5 margin is the accuracy of current PEMS equipment.
It was emphasized during the discussion that the problem of high real driving emissions in Europe is linked to weak enforcement mechanisms in the EU emission regulations. In particular, no regulatory compliance testing had been conducted in the EU until the recent Euro VI PEMS program for heavy-duty vehicles. In contrast, the US EPA and California ARB have had a long tradition of in use compliance testing programs.
While the RDE regulation may be effective in reducing in-use vehicle emissions, the complexity of the RDE testing may present a problem with RDE implementation in developing countries. The RDE approach is also unlikely to be adopted by the regulatory agencies in the United States.
Panel: Sub-23 nm Particles. Another panel discussion—with the participation of David Kittelson (UMN), Matti Mariq (Ford), Imad Khalek (SwRI), and Barouch Giechaskiel (JRC; over an internet connection)—focused on issues surrounding the measurement of particle number emissions. European emission regulations use the PMP procedure, where PN emissions are limited to solid particles with a diameter above 23 nm. This definition has been attracting criticism, because real emissions include both solid and liquid particles, and may include a significant proportion of particles below 23 nm.
The PMP procedure has been adopted due to a mix of pragmatic, as well as health effect reasons. The measurement of particles below 23 nm represents a challenge and includes a growing number of artifact particles as the particle size decreases. The measurement of liquid particles is highly uncertain—the total particle measurement can be changed by more than one order of magnitude by changing dilution parameters, which makes it difficult to measure the total particle number without over-restricting the sampling methodology. The engineering feasibility argument is further supported by the suspected health risk from solid particles, as opposed to liquid material.
However, many engine technologies and/or operating conditions can produce sub-23 nm particles. Examples include PFI and GDI gasoline engines, or natural gas engines (where sub-23 nm particles can be formed from the lube oil). The only class of engines that does not produce sub-23 nm particles are diesel engines equipped with particulate filters (provided the DPF is the last component in the exhaust system). Even DPF-equipped diesels can produce sub-23 nm particles during DPF regeneration.
According to European studies, the ratio of the number of 10-23 nm particles (i.e., particles not counted under the PMP procedure) to the number of particles that are counted (> 23 nm) amounts to 150% in mopeds, 50% in motorcycles, 100% in PFI engines, 50% in GDI engines and 10% in diesels with DPF. To address these issues, modifications of the PMP procedure are planned to lower the cut-off diameter to 10 nm. A number of open issues still need to be addressed, including the use a catalytic stripper in place of the volatile particle remover, calibration procedures for 10-15 nm particles, or a PN-PEMS investigation.
It was emphasized that, notwithstanding its shortcomings, the current PMP method is very effective in detecting high emitting vehicles powered by any type of engine. One of the practical aspects of the EU PN standards is that from 2017, particle emissions from GDI vehicles in the EU will be controlled to much lower levels compared to those in the United States.
Conference website: sae.org/congress