SAE TECHNICAL
`PAPER SERIES
`
`2003 -01 -0778
`
`The Development and Performance of the
`Compact SCR -Trap System: A 4 -Way
`Diesel Emission Control System
`
`Andrew P. Walker, Ronny Allansson,
`Philip G. Blakeman and Mats Lavenius
`Johnson Matthey, Catalytic Systems Division
`
`Sara Erkfeldt and Henrik Landaiv
`Volvo Powertrain Corporation
`
`Bill Ball and Pat Harrod
`Eminox Ltd.
`
`Didier Manning and Leopold Bernegger
`Robert Bosch GmbH
`
`Reprinted From: Diesel Exhaust Emissions Control
`(SP -1754 / SP- 1754CD)
`
`Internationa[
`400 Commonwealth Drive, Warrendale, PA 15096 -0001 U.S.A. Tel: (724) 776 -4841 Fax: (724) 776 -5760 Web: www.sae.org
`
`2003 SAE World Congress
`Detroit, Michigan
`March 3 -6, 2003
`
`BASF-2012.001
`
`

`
`All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
`transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise,
`without the prior written permission of SAE.
`
`For permission and licensing requests contact:
`
`SAE Permissions
`400 Commonwealth Drive
`Warrendale, PA 15096- 0001 -USA
`Email: permissions @sae.org
`724 -772 -4028
`Fax:
`724 -772 -4891
`Tel:
`
`IDA C
`gar W o
`Global Mobility Database®
`All SAE papers, standards, and selected
`books are abstracted and indexed in the
`Global Mobility Database.
`
`For multiple print copies contact:
`
`SAE Customer Service
`877 -606 -7323 (inside USA and Canada)
`Tel:
`724 -776 -4970 (outside USA)
`Tel:
`724 -776 -1615
`Fax:
`Email: CustomerService @sae.org
`
`ISSN 0148 -7191
`Copyright © 2003 SAE International
`
`Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE.
`The author is solely responsible for the content of the paper. A process is available by which discussions
`will be printed with the paper if it is published in SAE Transactions.
`
`Persons wishing to submit papers to be considered for presentation or publication by SAE should send the
`manuscript or a 300 word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE.
`
`Printed in USA
`
`BASF-2012.002
`
`

`
`The Development and Performance of the Compact SCR -Trap
`System: A 4 -Way Diesel Emission Control System
`
`Andrew P. Walker, Ronny Allansson, Philip G. Blakeman and Mats Lavenius
`Johnson Matthey, Catalytic Systems Division
`
`2003 -01 -0778
`
`Sara Erkfeldt and Henrik Landaiv
`Volvo Powertrain Corporation
`
`Bill Ball and Pat Harrod
`Eminox Ltd.
`
`Didier Manning and Leopold Bernegger
`Robert Bosch GmbH
`
`emission control communities. In both Europe and North
`America the major challenges are
`in PM and NOx
`control for the legislation to be introduced in 2005 (Euro
`IV), 2007 (US '07), 2008 (Euro V), and 2010 (US '10).
`1 summarises the evolution of the emissions
`Table
`standards
`in Europe, and Table 2 outlines
`the US
`regulations.
`
`Table 1: Emissions Legislation Limits in Europe (ESC
`Test Cycle; g kW1hr1)
`
`Year
`
`HC
`
`CO
`
`NOx
`
`PM
`
`2000 (Euro III)
`
`2005 (Euro IV)
`
`2008 (Euro V)
`
`0.66
`
`0.46
`
`0.25
`
`2.1
`
`1.5
`
`1.5
`
`5.0
`
`3.5
`
`2.0
`
`0.10
`
`0.02
`
`0.02
`
`Table 2: Emissions Legislation Limits the US (SET Test
`Cycle and US HDT Cycle; g bhp-'hr-1)
`(Note that 1 g bhp-1 hr-1 = 1.341 g kVV1hr1)
`
`CO
`
`NOx
`
`PM
`
`Copyright © 2003 SAE International
`
`ABSTRACT
`
`The tightening of Heavy Duty Diesel (HDD) emissions
`throughout
`the world
`legislation
`leading
`to
`the
`is
`development of emission control devices to enable HDD
`the new standards.
`engines
`to meet
`NOx and
`Particulate Matter (PM) are the key pollutants which
`these emission control systems need to address. Diesel
`Particulate Filters (DPFs) are already in use in significant
`numbers to control PM emissions from HDD vehicles,
`and Selective Catalytic Reduction
`is a very
`(SCR)
`promising technology to control NOx emissions.
`
`This paper describes the development and performance
`of the Compact SCR -Trap system - a pollution control
`a DPF -based
`comprising
`device
`system
`(the
`Continuously Regenerating Trap system) upstream of an
`SCR system. The system has been designed to be as
`easy to package as possible, by minimising the total
`volume of the system and by incorporating the SCR
`catalysts on annular substrates placed around
`the
`outside of the DPF -based system. This novel design
`gives rise to an easy -to- package emission control device
`capable of providing very high conversions of all four
`major pollutants, NOx, PM, CO and HC. The design
`the performance of the
`details are discussed, and
`system over both steady state and transient cycles is
`presented. NOx conversions of up to 92% have been
`demonstrated, and the system's emissions of all four
`pollutants are well inside the Euro V, and probably also
`the US 2007 limits (subject to verification of PM).
`
`INTRODUCTION
`
`The progressive tightening of the emissions standards
`for Heavy Duty Diesel vehicles throughout the world
`presents challenges for the engine development and
`
`HC
`
`1.3
`
`0.5
`
`Year
`
`1998
`
`2004
`
`2007
`
`2010
`
`15.5
`
`4.0
`
`15.5
`
`0.10
`
`0.10
`
`0.01
`
`0.01
`
`2.5
`
`1.2
`
`0.2
`
`0.14
`
`15.5
`
`0.14
`
`15.5
`
`BASF-2012.003
`
`

`
`The HC and CO emission limits are not expected to
`cause significant problems, because
`the engine -out
`emissions of these pollutants are already very
`low.
`reductions
`However, major
`in both PM and NOx
`emissions are required in the near future.
`
`One system which has shown great promise
`in
`controlling
`the Continuously
`emissions
`PM
`is
`Regenerating Trap (CRT`, developed and patented by
`Johnson Matthey [1]). Throughout the remainder of this
`paper, this system will be referred to as the Continuously
`Regenerating Diesel Particulate Filter, CR -DPF. There
`are currently over 35,000 CR -DPF systems in operation
`throughout the world, and the long -term field durability of
`the system has been clearly demonstrated [2].
`Indeed,
`systems have been shown to provide very high levels of
`CO, HC and PM reduction even after over 500,000 km
`(and six years) of field operation [3].
`
`The CR -DPF system comprises an oxidation catalyst
`followed by a wall -flow Diesel Particulate Filter (DPF).
`the oxidation catalyst
`traps
`the PM and
`The DPF
`oxidises a portion of the engine -out NO to generate NO2.
`This NO2 combusts the PM at a much lower temperature
`(around 250 °C) than does oxygen (around 550 °C). This
`low temperature combustion of PM by NO2 enables the
`passive operation (with continuous PM removal) of the
`CR -DPF system on a wide range of HDD applications
`(eg buses, trucks, garbage trucks) and is the basis of the
`CR -DPF. The CR -DPF has been used very successfully
`in the retrofit market over a wide range of operating duty
`cycles, since most HDD applications have duty cycles
`to guarantee continuous
`which are sufficiently warm
`regeneration of the CR -DPF system.
`
`the CR -DPF
`The oxidation catalyst used within
`is
`optimised to generate appropriate
`levels of NO2
`to
`regeneration.
`enable continuous, passive PM
`This
`catalyst is also very effective at oxidising CO (into 002)
`and HC (into CO2 and H2O), so the system provides very
`high conversions of PM, HC and CO. However, when
`the NO2 generated by the oxidation catalyst reacts with
`carbon it is converted back into NO, so very little NOx
`conversion is observed over the CR -DPF system.
`
`As discussed above, high conversions of NOx are
`required within the pollution control systems of the
`The most effective on -board NOx reduction
`future.
`for HDD vehicles
`is Selective Catalytic
`strategy
`Reduction (SCR) [4], so this was the approach used to
`SCR systems
`control NOx emissions in
`this work.
`reduce NOx via reaction with ammonia; this reaction is
`highly selective even in environments containing excess
`oxygen, such as the exhaust of Diesel vehicles:
`
`4 NH3 + 4 NO + 02 4 4 N2 + 6 H2O
`
`(1)
`
`Ammonia is generated on board the vehicle via the rapid
`hydrolysis of urea. One of the keys to the successful
`implementation of SCR technology is the ability to match
`injected to the amount of NOx
`the amount of urea
`emitted from the engine. While it is possible to do this
`
`the more common
`using on -board NOx sensors
`[5],
`approach is to map the NOx emissions of the engine,
`and to inject urea based on the information contained in
`these maps. The latter was the approach used here.
`
`The very effective combination of the CR -DPF (for PM,
`HO and CO control) with an SCR system (for NOx
`control) has previously been demonstrated [6]. This
`system was shown
`SORT M
`very high
`to give
`conversions of all four major pollutants over the ESC: up
`to 85% NOx, 96% PM, 98% HC and 100% CO
`conversion, using a total system volume of 51
`litres on a
`litre engine calibrated
`the Euro
`1 emissions
`to
`12
`standards. This catalyst system will be referred to as the
`SCR -Trap System for the remainder of this paper.
`
`In this earlier work, the combined system was arranged
`with the CR -DPF in front of the SCR system. Within this
`configuration, the CR -DPF removes CO, HC and PM,
`and converts some of the engine -out NO into NO2,
`before the gas passes to the SCR system. This leads to
`temperature NOx
`improvement
`large
`the
`low
`a
`in
`conversion of the SCR catalysts, partly because both
`CO and HC can inhibit the low temperature performance
`of SCR catalysts, but mainly because the presence of
`stream strongly promotes
`the gas
`low
`NO2
`in
`temperature SCR activity [6, 7]. This promotion occurs
`via the very fast reaction:
`
`2 NH3 + NO + NO2 4 2 N2 + 3 H2O
`
`(2)
`
`Furthermore, the removal of the PM upstream of the
`SCR system removes the possibility of PM fouling of the
`SCR catalysts and the urea injection system. So clear
`synergies were demonstrated for this combined system.
`
`In this earlier work, the combined system was arranged
`in a linear configuration. Such an arrangement will be
`for many HDD vehicles, especially buses.
`suitable
`However, for many trucks a more compact arrangement
`would be superior, and therefore the alternative layout
`shown in Figure 1 was developed.
`
`Outle
`
`Urea mixing unit
`
`Slip
`
`SCR
`
`SCR
`
`Oxicat
`
`Slip
`
`SCR
`
`SCR
`
`Urea injection point ---.
`Figure 1: Schematic Representation of the Compact
`SCRTTM Concept
`
`Within this design the gas first passes through the CR-
`DPF system, and is then turned through 180° and flowed
`through
`the SCR catalysts, which are coated onto
`
`BASF-2012.004
`
`

`
`metallic, annular substrates, fitted around the CR -DPF
`This presents a wider but much shorter
`system.
`packaging envelope for the combined system, and this
`system will be referred to as the Compact SCR -Trap
`System for the rest of this paper.
`
`DEVELOPMENT OBJECTIVES
`
`therefore started
`to verify
`A project was
`this
`that
`technology did
`in fact have the potential to meet US
`2007 and Euro V emissions levels, when realized
`in
`practical form. The chosen means of verification was to
`culminate in an extensive field
`test on heavy trucks
`engaged in regional distribution duties in the US. This
`meant that the SCR -Trap systems were to be developed
`typical vehicle
`installation and operating
`to meet
`constraints.
`
`This paper describes the development of the Compact
`SCR -Trap system for this purpose, and outlines
`its
`performance over both the ESC and the US Heavy Duty
`Transient test cycles, on a Volvo D12C, US'00 12 litre
`engine.
`
`COMPACT SYSTEM DESIGN
`
`REQUIREMENTS
`
`The concept of Figure 1 had to be realised in a practical
`design meeting a number of constraints. Key targets
`were:
`
`to fit the unit within the specified space envelope
`(24.4x23.6x24.4in) available
`620x600x620mm
`on the test vehicle, and to match the existing
`inlet and outlet pipe positions of the standard
`muffler
`
`to enable both the CR -DPF filter and the urea
`injector to be easily removed for servicing
`
`to meet a backpressure limit of 20kPa at a flow
`of 0.5 kg /s, 438 °C (2.9 psi @ 1.1 lb/s, 820 °F)
`
`to achieve the structural integrity and durability
`required for a 2 -year 350,000 km (220,000 mile)
`intervals of at
`trial, with service
`field
`least
`50,000km (31,250 miles)
`
`to meet specified noise and emissions limits
`
`In addition, it was the clear intention of all the project
`partners to have a design which was functionally as
`close to a future production system as possible.
`
`OVERALL LAYOUT
`
`Designs with a range of alternative catalyst and filter
`sizes which could be fitted within the available space
`the best compromise between
`for
`were assessed
`backpressure (with filter clean and after accumulating
`
`thermal mass and weight while achieving the
`ash),
`space velocities estimated to be required for satisfactory
`catalyst performance. This resulted in
`the choice of
`nominal sizes as follows:
`
`CR -DPF catalyst (diameter x length): 381 x 74.6 mm (15
`x 2.94 in)
`
`Filter (diameter x length): 381 x 305 mm (15 x 12 in)
`
`SCR catalysts (2 off) + ammonia slip catalyst (1 off):
`each of inner diameter 408.5 mm (16.08
`in), outer
`diameter 575 mm (22.64 in), and length 74.6 mm (2.93
`in),
`
`So within a total length of 620mm there had to be
`accommodated CR -DPF elements with a combined
`length of 380mm and SCR /ammonia slip catalysts of
`combined length 224mm (all before canning), as well as
`room for the inlet and outlet flows, the urea mixing
`thermally -insulated endcaps.
`section, and
`A major
`design challenge was therefore achieving the length
`constraint while minimising pressure losses within the
`unit.
`
`KEY DESIGN FEATURES
`
`Unlike the schematic representation of Figure 1, both
`inlet and outlet pipes had to be oriented in a radial
`direction. The length constraint was met by overlapping
`the pipes in an axial direction, and separating the inlet
`and outlet flows by means of a partition across the outer
`annulus [8]. The inlet and outlet are separated angularly
`by only 66 degrees, which adds to the difficulty of
`achieving low pressure losses and uniform flow through
`the catalysts. The CR -DPF is supported by a central
`tube, which locates on both endcaps. The tube contains
`apertures which allow the flow through it from the inlet to
`the CR -DPF and from the CR -DPF to the SCR section.
`Loads are transferred to this tube from the mountings via
`the partition and a support strut assembly. The latter
`avoids significant loads being transferred through the
`SCR catalysts, which for the field trial units have been
`made removable. This removability also necessitated
`the incorporation of inner and outer seals to prevent flow
`bypassing these catalysts, which is particularly important
`to minimise ammonia slip and
`to maximise NOx
`conversion.
`It also required careful design to overcome
`problems of radial tolerance build -up. The filter and the
`CR -DPF catalyst are retained axially by a removable
`retention device inside the central tube [8]. To clean out
`incombustible residues from the filter, it can be removed
`by first removing the mixer assembly and then the
`retention device, each requiring the release of only a
`single fastener; the filter
`is then withdrawn using a
`special tool.
`
`The urea is injected from a centrally -positioned injector.
`Mixing of the urea with the exhaust gas stream occurs in
`a mixing unit downstream of the
`injector [8].
`The
`distance from the end of the filter to the inside of the
`in), but the mixer achieves
`endcap is only 127mm (5
`
`BASF-2012.005
`
`

`
`effective mixing of the urea sprays in both radial and
`circumferential directions before the flow enters the SCR
`catalyst, as will be seen from the high NOx conversion
`efficiencies described below. The mixer was designed
`using mixed energy and spray trajectory calculations, so
`the required mixing with a minimum
`to achieve
`as
`pressure loss. CFD calculations predicted a coefficient
`of variation of urea concentration at SCR catalyst entry
`of less than 10 %, and a velocity profile uniformity index
`the maximum
`flow condition.
`above 0.95 at
`Urea
`deposits are a potential issue with SCR systems. The
`mixer design seeks to minimize the area of solid surface
`which the urea might contact, and to easily re- entrain
`into the flow any liquid that is deposited. The urea
`injection strategy is also vital in avoiding deposits. No
`deposit problems occurred during the prototype pass -off
`testing, but this remains an area to be monitored during
`the upcoming field trials.
`
`Both endcaps are double- skinned, to provide rigidity and
`thermal insulation.
`
`Figure 2 shows the completed unit alongside a CR -DPF
`unit. The latter is sized for a truck with an engine of one
`third less power, and is the same size as the standard
`muffler for that vehicle.
`
`testing of the inlet and CR -DPF sections, to validate the
`main structural elements and the filter retention device.
`This unit survived 413,000 cycles of amplitude 90kPa
`with some internal cracking but no loss of integrity. By
`comparison
`to previous experience using
`test
`this
`procedure, this again
`indicates a service life well
`in
`excess of the planned 2 year field trial.
`
`DESIGN SUMMARY
`
`The final design contains 71.5 litres of catalyst and filter
`within a total volume of approximately 164 litres. As far
`as can be known before field testing, it meets or betters
`all the key targets. Although containing several features
`appropriate only to prototypes, a very similar design
`would be suitable for mass production with little change.
`
`UREA DOSING UNIT
`
`Figure 3 shows the interfaces of the reactant agent
`dosing system with the control unit, exhaust gas system
`with catalyst, urea water solution tank, air reservoir and
`the vehicle CAN -bus.
`Figure 4 shows a schematic
`layout of the dosing system.
`
`engine. vehitde
`engine -CAN
`inj. quantity speed
`roofing water temp
`
`ECU
`
`E
`
`air reservoir
`
`dosing system
`
`catalyst
`
`Figure 3: Schematic of the Interfaces within the Urea
`Dosing Unit
`
`Figure 2: Size Comparison between Compact SCR -
`Trap Sized for 347kW Engine (left) and CR -DPF Sized
`for 231 kW (right)
`
`MECHANICAL DURABILITY
`
`The muffler unit is constructed from AISI 304 austenitic
`stainless steel. The field trial units use design features
`and manufacturing methods well -proven from Eminox's
`heavy -duty CR -DPF systems where possible, in order to
`minimise development
`time, cost and
`This
`risk.
`approach has been vindicated by the results of two
`mechanical durability rig tests. A complete unit mounted
`as in the vehicle underwent shake testing.
`It passed full
`life expectancy according to Volvo's standard procedure.
`Another unit was subjected to pressure pulse fatigue
`
`AC©ae
`bathibw
`
`AtlBae- +..
`};'
`irYet
`
`is
`
`AffYip b
`
`1'.Dx
`
`CorMre,<trtl
`ea n1Fi
`
`>{ì(feg.tl:/ÁNr
`val¡-l9_-7rlt---.,y
`
`'RLLtT
`
`,......._
`
`Tank level sensor
`......... _ .................................
`
`Tarxtenraxre:sn:or
`
`...
`
`Engxre-CAp
`
`AdE6P
`
`pr :.
`
`n3g(
`
`--;í{ l
`
`Adtlae.
`fén petátiió tCrG3r
`
`Vont
`vakÁ
`
`t.
`
`I
`
`AIrDrrs:ué..
`
`Firpid}a
`
`Figure 4: Schematic Layout of the Urea Dosing System
`
`BASF-2012.006
`
`

`
`With the help of a feed pump, a urea /water solution is
`delivered from the urea tank to the dosing valve. The
`system pressure is controlled via a pressure sensor, to
`secure flawless system functionality. So
`that if,
`for
`example, the pressure build -up was disturbed due to air
`or gas bubbles,
`the system would be automatically
`purged.
`The
`injection pressure used was 4.6 bar
`absolute, to ensure good urea spray characteristics.
`
`The dosing strategy is matched to the NOx emissions of
`the engine via a self- contained electronic control system.
`The urea is then injected into a predefined air mass flow
`delivered from the vehicle air reservoir or compressor in
`a subsequent mixing chamber. A back -flow valve is
`integrated
`the mixing chamber,
`into
`to avoid any
`intrusion of urea into the air supply. The air mass flow
`transports the urea through piping and then injects it into
`the exhaust gas flow through a spray tip. Under these
`conditions a fine atomisation and even urea distribution
`the exhaust flow are pre- requisites for high NOx
`in
`conversions.
`
`A focal point in the component development is
`the
`mixing chamber which distinguishes itself functionally
`with the following characteristics:
`
`prevention of urea entering the system
`
`prevention of urea crystalline products in
`system
`
`the
`
`equal mixture of the urea water solution and air
`
`Figure 5 shows the compact dosing unit used in this
`work. A modular dosing system is being developed at
`present, which will allow
`further vehicle system
`installation flexibility though separate pump and dosing
`units.
`
`EXPERIMENTAL
`
`CATALYST AND FILTER DETAILS
`
`The Pt -based catalyst within the CR -DPF part of the
`system was coated onto a metallic monolith substrate
`with a cell density of 54 cells.cm -2 (350 cells.in-2) and a
`wall thickness of 0.15 mm (0.006 in) using standard
`processing techniques. The CR -DPF catalyst used for
`the engine evaluations had a catalyst volume of 8.44
`litres. The cordierite particulate filter used in the CR-
`DPF section of the system was a Corning production
`standard type, with 15 cells.cm -2 (100 cells.in "2) and a
`wall thickness of 0.43 mm (0.017 in). The volume of the
`filter was 34.5 litres.
`
`The SCR catalysts were coated onto metallic, annular
`substrates, with a cell density of 54 cells.cm"2 (350
`cells.in "2) using standard processing techniques. Each
`annular substrate had a volume of 9.52 litres. For some
`experiments an ammonia slip (clean -up) catalyst coated
`onto a single annular substrate was used.
`In other
`experiments, the ammonia slip catalyst was coated onto
`a 266.7 x 152.4 mm ceramic substrate with a cell density
`(400 cells.in "2), and was
`located
`downstream of the compact system.
`
`functional robustness
`
`BENCH ENGINE EVALUATION
`
`The self- contained control unit is mounted to the base
`plate and connected to the internal wire harness via a
`separate connector system.
`The connection of the
`control unit to the engine and vehicle is secured through
`a separate connector. Communication between
`the
`dosing control unit and
`the engine control unit
`is
`maintained through the standardised CAN -bus according
`to J 1939 -71.
`
`Figure 5: Compact Urea Dosing Unit
`
`The engine used for these studies was a Volvo D12C 12
`litre, 347 kW (465 hp) turbocharged, aftercooled engine,
`for US'00 and with slightly modified engine
`certified
`calibration due to the pollution control system.
`
`The European Steady -state Cycle (ESC) is a 13 mode
`cycle which probes the whole range of the engine map.
`The ESC evaluations were carried out using standard
`procedures. The engine was operated for the prescribed
`time in each mode, and completed the engine speed and
`load changes in the first 20 seconds at each mode. The
`sampling
`times were at
`least 4 seconds per 0.01
`weighting factor. One PM sample was taken over the
`complete test cycle by means of a dilution tunnel. The
`filter
`papers were
`stabilised
`under
`controlled
`temperature and humidity conditions for 24 hours prior to
`the testing and then installed for the test. After the test
`cycle had been completed, the filter papers, with the
`collected PM, were again stabilised before being
`weighed.
`
`The US Heavy Duty Transient (HDT) cycle is a transient
`cycle which simulates real world driving.
`The HDT
`evaluations were carried out on the same engine using
`standard procedures according
`to EPA guidelines.
`Regulated emissions
`(except PM) were measured
`before and after
`the Compact SCR -Trap System.
`
`BASF-2012.007
`
`

`
`Second -by- second measurements of gaseous emissions
`and engine parameters were made during the test cycle.
`
`The fuel used during all the tests was Swedish MK1 fuel
`(sulfur level < 10 ppm) from a standard refinery source.
`The engine oil used was Shell Myrina 15W 40. The urea
`used was a commercially available formulation of 32.5 %
`in water. Ammonia to NOx
`urea
`ratios (ANRs) of
`between 0.85 and 1.00 (by moles) were used in this
`work; the ANR was controlled via the urea injection rate.
`An ANR of 1 corresponds to the stoichiometric ratio
`which potentially enables 100% NOx conversion. The
`volumetric flow rate of the urea solution to the injector
`nozzle was measured periodically to check that the urea
`dosing rate was correct.
`
`Gaseous emissions were measured raw before and after
`the catalyst system using a
`range of analysers:
`chemiluminescence for NO and NO2, heated FID (Flame
`Ionisation Detector) for HC, NDIR (Non- Dispersive Infra
`Red) analysers
`for CO, CO2 and N20, and a
`paramagnetic detector for 02. Ammonia was detected
`using a laser -based AltOptronic LD 3000 analyser which
`measures the sample in situ.
`
`RESULTS AND DISCUSSION
`
`ESC EVALUATIONS
`
`The performance of the system was initially measured
`over the ESC test cycle. The urea injection strategy was
`carefully calibrated to provide very high NOx conversion
`with very little ammonia slip. The urea was injected to
`provide an ammonia- to -NOx (ANR) ratio of 0.95, so the
`maximum possible NOx conversion in these experiments
`was 95 %.
`Two Compact SCR -Trap Systems were
`tested: one containing
`two annular SCR catalysts
`followed by an annular ammonia slip catalyst, and the
`other containing three annular SCR catalysts followed by
`an ammonia slip catalyst located downstream of the
`compact
`the emissions
`both
`unit.
`cases
`In
`measurements were made after
`the ammonia slip
`catalyst. The ESC emissions of these two compact
`systems are summarised in Table 3.
`
`Table 3: Emissions of the Compact SCR -Trap System
`over the ESC (g kW' hr')
`
`System
`
`HC
`
`CO
`
`NOx
`
`PM
`
`Engine Out
`
`0.123
`
`0.324
`
`6.926
`
`0.022
`
`2 SCR Catalysts
`
`0.002
`
`0.000
`
`1.097
`
`0.006
`
`3 SCR Catalysts
`
`0.002
`
`0.000
`
`0.562
`
`0.008
`
`Euro V Limits
`
`0.460
`
`1.500
`
`2.000
`
`0.020
`
`It can be seen that the system with the lower catalyst
`volume
`(the one containing only 2 annular SCR
`catalysts) is very comfortably inside the Euro V emission
`legislation
`for all four pollutants.
`Indeed,
`the NOx
`emissions of the system are approximately half the
`proposed Euro V NOx standard of 2 g kW' hr'. When
`the additional SCR catalyst is added to the system, the
`NOx emissions are
`reduced
`to approximately one
`quarter of the proposed Euro V NOx level. As expected,
`the addition of the third SCR catalyst does not lead to
`any significant changes in the other emissions, which all
`remain far below the Euro V legislated levels.
`
`The HC and CO conversions obtained using
`the
`compact systems are extremely high, with both systems
`providing 98% HC and 100% CO conversion. The NOx
`conversion obtained using two annular SCR catalysts is
`rising to 92% when the third SCR catalyst
`84 %,
`is
`added.
`(Since urea is
`injected to provide an ANR of
`0.95, 95% NOx conversion is the maximum possible in
`these experiments, so this 92% is a remarkably high
`conversion figure). The PM conversion from the system
`is around 70 %, but
`relatively
`this
`figure
`low
`is a
`consequence of the very low engine -out PM emissions,
`which, at 0.022 g kW' hr', almost meet the Euro V PM
`standard of 0.020 g kW' hr' without the CR -DPF filter
`system. Note that the PM emissions from the compact
`systems are less than half of the Euro V legislated level.
`
`This very high NOx conversion can be compared with
`previous data generated using the same urea injection
`system as used here. ESC NOx conversions of up to
`83% were reported using an SCR system with no oxicat,
`DPF or ammonia slip catalyst [9]. Much larger SCR
`volumes were used in the earlier work (2.7 to 2.8 times
`engine swept volume, compared to 1.6 here).
`
`Figure 6 shows the NOx conversion and ammonia slip
`level from the compact system containing three annular
`SCR catalysts at each mode of the ESC (except Mode 1,
`which is the idle mode). Note that the NOx conversion is
`measured over the complete system (ie downstream of
`the ammonia slip catalyst) while the ammonia slip is
`recorded
`in front of the ammonia slip catalyst. The
`ammonia slip was measured here to show how much
`ammonia breaks
`through
`the SCR catalysts, and
`therefore to provide an accurate assessment of the
`precision of the urea injection calibration.
`
`Figure 6 shows
`that the NOx conversion
`remains
`extremely high
`in each ESC mode, never
`falling
`significantly below 90 %.
`The ammonia slip
`levels
`measured before the slip catalyst are very low, never
`rising above 10ppm, which testifies to the excellent urea
`injection calibration developed
`this work.
`As
`in
`expected, the ammonia slip level measured downstream
`of the slip catalyst was almost undetectable.
`
`BASF-2012.008
`
`

`
`100
`
`90
`
`80
`
`70
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`0
`
`50
`
`45
`
`40
`
`y T
`
`ca
`
`35
`
`f NOx Conversion
`
`-C) -NH3 Slip
`
`U
`_
`30 R.-,
`á E
`U
`cu
`o
`m
`.o
`a
`
`25
`
`sa-
`
`20
`
`15
`
`\- -
`
`10
`
`5
`
`_ z
`
`2
`
`3
`
`4
`
`5
`
`9
`6
`7
`8
`ESC Mode Number
`
`10
`
`11
`
`12
`
`13
`
`Figure
`6: Mode -by -Mode NOx Conversion and
`Ammonia Slip Levels using 3 SCR Catalysts over the
`ESC Test Cycle
`
`US HEAVY DUTY TRANSIENT EVALUATIONS
`
`During the calibration of the urea injection strategy a
`great deal of emphasis was placed on the transient
`performance of the system. This is extremely important
`since real -world driving is often highly transient in nature.
`To assess the efficiency of the urea injection calibration,
`the Compact SCR -Trap System was also tested over the
`US Heavy Duty Transient test cycle. For these tests the
`SCR system comprised two annular SCR catalysts
`followed by an annular ammonia slip catalyst.
`
`The US Heavy Duty Transient test starts with the engine
`traditionally present
`tests
`cold;
`such cold
`start
`challenges for emission control devices. With
`this
`system, cold start PM control is not an issue, since the
`filter will
`remove soot very efficiently at
`particulate
`ambient
`temperatures.
`But NOx control
`is more
`challenging, since urea cannot be injected at very low
`temperatures, because
`it will not hydrolyse and
`ammonia will not be generated at temperatures below
`around 160 °C. This sets a practical limit for the start of
`the urea injection in SCR systems in such cold start
`tests, and therefore limits the NOx conversion that can
`be obtained.
`
`tailpipe NOx
`the engine out and
`Figure 7 shows
`emissions during a cold start US Heavy Duty Transient
`test. The temperature of the SCR system is also shown.
`takes around 600 seconds for the SCR
`Note that it
`temperature
`the critical 160 °C
`catalyst
`rise
`to
`to
`temperature level. This is principally because the CR-
`DPF system has a large thermal mass, so it takes a long
`time for the engine heat to reach the downstream SCR
`catalysts. Urea injection was started after approximately
`625 seconds, at which point the tailpipe NOx level falls
`well below the engine -out NOx level.
`
`It can be seen that substantial NOx conversion is then
`achieved. However, because no urea injection (and
`therefore no significant NOx conversion) occurred during
`the first half of the test, the overall NOx conversion
`
`during this test was relatively low, at 56.1%. Note that
`this is similar to the 55.6% conversion reported by Miller
`et al [10], although in that case the SCR catalyst volume
`was much larger and there was no upstream CR -DPF to
`increase the SCR warm -up time.
`
`2000
`
`1800
`
`--1600
`E
`R1400
`ó 1200
`
`Ç 1000
`
`C 800
`o
`U
`600
`o
`z 400
`
`200
`
`-Engine -Out NOx
`Tailpipe NOx
`-SCR Catalyst Temperature
`
`400
`
`350
`
`300
`
`250
`
`V
`m
`
`200 R
`m Q
`m
`
`150
`
`100
`
`50
`
`o
`
`200
`
`400
`
`600
`Time (s)
`
`800
`
`1000
`
`1200
`
`Figure
`7: NOx Emissions and SCR Catalyst
`Temperature During a Cold Start US Heavy Duty
`Transient Test
`
`The emission and conversion
`levels of the gaseous
`pollutants during this cold start test are given in Table 4.
`
`Table 4: Gaseous Emissions From the Compact SCR -
`Trap System in the Cold Start US Heavy Duty Transient
`Test (g bhp -1hr 1)
`
`System
`
`HC
`
`CO
`
`NOx
`
`Engine Out
`
`0.184
`
`1.311
`
`5.418
`
`Tailpipe
`
`0.013
`
`0.355
`
`2.378
`
`Conversion
`
`93%
`
`73%
`
`56%
`
`The tailpipe ammonia slip was very low during this cold
`start test, with an average of 6ppm, and a maximum of
`20ppm. These low levels confirm the accuracy and
`control of the urea injection strategy.
`
`the engine -out and
`Figure 8 shows
`tailpipe NOx
`emissions, as well as the SCR catalyst temperature,
`during a hot start US Heavy Duty Transient test.
`
`Much better NOx conversion is seen during hot start US
`Heavy Duty Transient tests, since in such
`tests the
`temperature is always high enough
`to permit urea
`injection.
`Table 5 summarises the emissions and
`conversio