Characterisation of Exhaust Particulate Emissions from Road Vehicles

Transcription

1 PARTICULATES Characterisation of Exhaust Particulate Emissions from Road Vehicles Deliverable 5: Definition of the detailed measurement matrix for the production of representative emissions factors Final Draft - Version /2001 A project sponsored by: In the framework of: EUROPEAN COMMISSION Directorate General Transport and Environment Fifth Framework Programme Competitive and Sustainable Growth Sustainable Mobility and Intermodality

2 Contractors LAT/AUTh: Aristotle University of Thessaloniki, Laboratory of Applied Thermodynamics - EL CONCAWE: CONCAWE, the oil companies' European organisation for environment, health and safety - B VOLVO: AB Volvo - S AVL: AVL List GmbH - A EMPA: Swiss Federal Laboratories for Material Testing and Research - CH MTC: MTC AB - S TUT: Tampere University of Technology - FIN TUG: Institute for Internal Combustion Engines and Thermodynamics, Tech. University Graz - A IFP: Institut Français du Pétrole - F AEAT: AEA Technology plc - UK JRC: European Commission Joint Research Centre - NL REGIENOV: REGIENOV - RENAULT Recherche Innovation - F INRETS: Institut National de Recherche sur les Transports et leur Securité - F DEKATI: DEKATI Oy - FIN SU: Department of Analytical Chemistry, Stockholm University - S DHEUAMS: Department of Hygiene and Epidemiology, University of Athens Medical School - EL INERIS: Institut National de l Environment Industriel et des Risques - F LWA: Les White Associates - UK TRL: Transport Research Laboratory - UK VKA: Institute for Internal Combustion Engines, Aachen University of Technology D VTT VTT ENERGY Engine Technology and Energy in Transportation - FI II

3 Publication data form 1. Framework Programme European Commission DG TrEn, 5 th Framework Programme Competitive and Sustainable Growth Sustainable Mobility and Intermodality 3. Project Title Characterisation of Exhaust Particulate Emissions from Road Vehicles (PARTICULATES) 5. Deliverable Title Definition of the detailed measurement matrix for the production of representative emissions factors 7. Deliverable Responsible CONCAWE 10. Author(s) Neville Thompson 12. Summary 8. Language English 2. Contract No 2000-RD Coordinator LAT/AUTh 6. Deliverable No 5 9. Publication Date April Affiliation CONCAWE The Particulates programme aims to improve understanding of particulate emissions and to develop representative emissions factors for road vehicles. In the first stages, a harmonised exhaust particulate sampling and testing protocol is being developed. This protocol should then be used to evaluate particulate emissions with a range of engine and vehicle technologies, exhaust aftertreatment technologies and fuels. This report defines the overall matrix of engines, vehicles, fuels and test cycles to be used in the measurement campaign. This version 1.1 of deliverable 5 is considered as a live working document and will be further refined on completion of Work Package 300, prior to the start of the measurement campaign. 13. Notes This is the final draft version to be published on the PARTICULATES web-site. 14. Internet reference Key Words 16. Distribution statement 17. No of Pages Price FREE FREE 19. Declassification date 20. Bibliography YES III

4 Table of Contents Table of Contents Summary Introduction Background and objectives of Particulates Objectives of Work Package Test fuel selection (Task 420) Background Fuel Matrices Lubricant Selection Selection of engines, vehicles and after treatment technologies (Tasks 410 and 430) Overall programme plan VOLVO INRETS CONCAWE AVL VKA TUG MTC AEAT EMPA IFP LAT PSA VTT ENERGY Stockholm University JRC Petten Definition of drive cycles and test conditions (Task 440) Light duty vehicle procedures NEDC CADC (Common Artemis Driving Cycles) or Real World Cycles (RWC) Additional transient light duty cycles Heavy duty engine procedures ECE R49 Heavy Duty Engine Exhaust Emission Test Procedure ESC (European Steady Cycle) ETC (European Transient Cycle) Heavy Duty Vehicle procedures Test Sequences, Vehicle Conditioning and Fuel Change Procedures Light Duty Vehicle Test Sequence Heavy Duty Engine Test Sequence General test design, number of repeats Procedures for testing systems with regenerative devices Roadside particulates measurements (Task 450) Methodology for roadside measurements of particulate mass and size distribution Measurement Location Roadside measurements in ambient air

5 Measurements in road tunnels Measurement dates Requirements for the equipment Particulate mass (TSP, PM 10, PM 2.5 ) Particle size distribution with SMPS < 1µm Non-exhaust particulates measurements (Task 460) Background Experimental Methodology Gravimetric determination of tyre and brake wear rates Chemical analysis of vehicle components and wear dust Sampling and analysis of airborne particles and road dust REFERENCES...41 Appendix 1 Additional Transient light duty cycles...42 Appendix 2 Test Fuel Handling Gasoline and Storage Procedure...44 Appendix 3 Sampling and analysis of PAH

6 1. Summary The Particulates programme aims to improve understanding of particulate emissions and to develop representative emissions factors for road vehicles. In the first stages, a harmonised exhaust particulate sampling and testing protocol is being developed. This protocol should then be used to evaluate particulate emissions with a range of engine and vehicle technologies, exhaust after-treatment technologies and fuels. This report defines the overall matrix of engines, vehicles, fuels and test cycles to be used in the measurement campaign. This version 1 of deliverable 5 is considered as a live working document and will be further refined on completion of Work Package 300, prior to the start of the measurement campaign. 3

7 2. Introduction 2.1. Background and objectives of Particulates The effect of particulate emissions on health has been of great interest for many years. To date, particulate emissions from vehicles have been controlled via legislation based on particulate mass. Recent studies have however suggested that adverse health effects may not only be dependent on total particulate mass. Smaller particles have been claimed by some to cause more adverse effects than large particles. This has led to revision of the particulates Air Quality Standard in the U.S.A to include measurement of finer particulates (PM 10 PM 2.5 ) and to further evaluation of the best metric for air quality standards worldwide. In Europe, further tightening of controls on particulate emissions from vehicles is being effected through the Euro III and IV standards for light duty vehicles and Euro III, IV and V standards for heavy duty vehicles. Along with the reduction in particulate mass emissions from vehicles, measurement methodologies have been enhanced. In the automotive arena there has been much emphasis on the development of methodologies to measure particulate size and number emissions as well as mass. Techniques for evaluating particulate composition have also been improved. Much of the current interest in further characterisation of particulate emissions arose from Bagley et al, 1996 (1) who described an increase in the formation of nanoparticles in a new technology engine, where particle mass emissions were being reduced. Further research has indicated that there are two distinct types of particles emitted from vehicles: Accumulation mode particles, mainly carbonaceous in nature and greater than ca. 30 nm in particle size Nucleation mode particles, generally below 30 nm particle size, predominantly liquid and comprised mainly of sulphate and heavy hydrocarbons In view of the health concerns and uncertainties on particle size and number emissions from motor vehicles, further characterisation of particle emissions is needed. The Particulates project was established to develop this knowledge and aims : to increase knowledge and understanding of particulate emissions from motor vehicles, to provide a harmonised particulate sampling and measurement methodology, to develop representative emissions factors for particulates to enhance air quality modelling tools and help explain health effects, to assess the effectiveness of technical measures for reducing particulate emissions. A critical first step in the Particulates programme is the definition of the exhaust aerosol properties to be examined and the identification of suitable instruments and measurement techniques to be used. It was decided that both accumulation and nucleation mode particles should be measured and the nucleation mode particles present a major challenge as they have been observed to be highly sensitive to test conditions. A review of the currently available instrumentation and sampling techniques was carried out and has been reported as deliverable 2 from the project (2). The next step was to develop a suitable sampling and testing methodology for the measurement of automotive particulate emissions. Work is currently in progress to develop this harmonised sampling and testing protocol, which is expected to be completed in the next 3 months. 4

8 The protocol developed should then be applied in a detailed measurement campaign to study particulate emissions from the current and near future vehicle technologies with a range of fuels Objectives of Work Package 400 The Particulates programme has been arranged under a series of Work Packages. Work Package 200 covered the definition of the exhaust aerosol properties to be examined and the identification of suitable instruments and measurement techniques to be used. Work Package 300 is currently finalising the harmonised sampling and testing protocol for the measurement of exhaust particulate emissions. Work Package 400 has the task to refine the overall programme for the measurement campaign to be carried out in Work Package 500. The plans developed by Work Package 400 have been defined in order to achieve the overall programme objectives : - To understand particle emissions from the current light and heavy duty fleet - To assess the potential of likely future emissions reduction technologies The overall objectives and tasks of Work Package 400, including the task leaders involved in the production of the Work Package 400 report are shown in Table 1. Table Organisation of a detailed measurement matrix for the production of representative emission factors Concawe 410 Selection of vehicle technologies to be tested LAT/AUTh 420 Selection of fuels to be used Concawe Gasoline LDV Diesel LDV GDI LDV HDV Conventional diesel fuels Conventional gasoline fuels Alternative fuels 430 Test of emission reduction techniques IFP 440 Testing conditions INRETS Test cycles Ambient temperature conditions Specific test protocols 450 Definition of roadside measurements TUG 460 Definition of non exhaust particulate measurements TRL 5

9 3. Test fuel selection (Task 420) 3.1. Background The core test fuels have been selected in view of the objectives to develop representative emissions factors for the current and future vehicle fleets as well as to enhance understanding of fuel effects. Existing knowledge implicates fuel sulphur as the key fuel effect on nano-particle emissions. However, few studies have separated the effect of sulphur from other fuel variables. Fuel sulphur has been recognised as the key fuel effect on exhaust after-treatment systems and hence the EU Commission is currently finalising proposals on fuel specifications which are likely to see fuel sulphur reductions from today s levels to 50 and then 10 ppm max, with all European gasoline and diesel fuels having 10 ppm max S by The EU Commission are not proposing any changes to other fuel specifications in this period. On this basis, the programme should include a sulphur matrix based on fuels with other properties held as close as possible to average year 2000/05 levels. For diesel fuel it is proposed to include an additional fuel at the 10 ppm max S level Swedish Class 1 diesel fuel in order to quantify any further potential benefits from extreme changes in other fuel properties. An old diesel fuel will also be included in order to quantify the air quality impact of older heavy duty diesel engines Fuel Matrices The following fuel matrices are proposed for the programme. Tables 2 and 3 show target properties for the fuels, based on estimates of average market fuel properties and hence which would provide a firm basis for the development of emissions factors. These target properties have been derived from work on the development of reference fuel specifications for 2005 and beyond. 6

10 Table 2 Diesel Fuels Fuel Code D-1 D-2 D-3 D-4 D-5 Fuel Description Units EN 590 : pre-1996 EN 590 : 2000 EN 590 : 50 ppm S EN 590 : 10 ppm S Swedish Class 1 Characteristic Cetane 53 ± 2 53 ± 2 53 ± 2 53 ± 2 53 ± 2 Density kg/m ± ± ± ± ± 5 T50 C 265 ± ± ± ± ± 10 T95 C 350 ± ± ± ± ± 5 FBP C 363 ± ± ± ± ± 7 Flash point C 56 min 56 min 56 min 56 min 56 min CFPP C - 5 max - 5 max - 5 max - 5 max - 10 max 40 C mm 2 /s 2.8 ± ± ± ± ± 0.4 PAH % m/m 5.5 ± ± ± ± max Sulphur mg/kg 1500 ± ± ± 10 8 ± 2 10 max Cu corrosion Class 1 max Class 1 max Class 1 max Class 1 max Class 1 max Conradson % m/m 0.2 max 0.2 max 0.2 max 0.2 max 0.2 max Carbon Ash % m/m 0.01 max 0.01 max 0.01 max 0.01 max 0.01 max Water % m/m 0.02 max 0.02 max 0.02 max 0.02 max 0.02 max Acid No. mg KOH/g 0.02 max 0.02 max 0.02 max 0.02 max 0.02 max Ox Stab mg/ml max max max max max HFRR µm 400 max 400 max 400 max 400 max 400 max FAME Nil Nil Nil Nil Nil Note that the feasibility of blending fuel D-4 still has to be confirmed as this represents a future fuel quality which will have to be produced from components available now. Assuming fuel D-4 can be prepared, fuels D-1 to D-3 will be blended by doping D-4 with sulphur compounds. 7

11 Table 3 Gasolines Fuel Code G-1 G-2 G-3 Fuel Description Units EN 228 : 2000 EN 228 : 50 ppm S EN 228 : 10 ppm S Characteristic RON MON Density kg/m ± ± ± 5 RVP kpa 58 ± 2 58 ± 2 58 ± 2 E70 % v/v 32 ± 8 32 ± 8 32 ± 8 E100 % v/v 54 ± 4 54 ± 4 54 ± 4 E150 % v/v 86 ± 4 86 ± 4 86 ± 4 FBP C 200 ± ± ± 10 Residue % v/v 2.0 max 2.0 max 2.0 max Olefins % v/v 11 ± 3 11 ± 3 9 ± 3 Aromatics % v/v 39 ± 3 32 ± 3 32 ± 3 Benzene % v/v 0.8 ± ± ± 0.2 Sulphur mg/kg 125 ± ± 10 8 ± 2 Oxygen % m/m 1.0 max 1.0 max 1.0 max Induction time minutes 480 max 480 max 480 max Existent gum mg/100ml 4 max 4 max 4 max Cu Corrosion Class 1 max Class 1 max Class 1 max Lead mg/l 5 max 5 max 5 max Phosphorus mg/l 1.3 max 1.3 max 1.3 max Not all partners will test all fuels. For example, INRETS have the objective to establish the performance of the current light duty fleet and hence should run this task with current fuels only. Other partners are to provide understanding of different vehicle and fuel technologies and hence should select relevant test fuels from the matrix. The overall programme plan is given in section 4 (vehicle technology selection). The above fuels will be available from CONCAWE (at agreed prices) and should be used by all partners in order to provide a common fuel set and common database for the programme. Some partners are also planning to evaluate alternative fuels, including RSME, oxygenated fuels and natural gas. As these are one-off tests, these fuels will not be available from CONCAWE but will be obtained by the partners running the tests 8

12 3.3. Lubricant Selection A common batch of lubricant should be used for the programme in order to minimise effects from the lubricant. The lubricant should be representative of current typical European lubricant quality. On this basis, a good quality, high volume, conventional mineral oil formulation should be selected to the following specification : - 10W-40 - ACEA Class A3 / D3 for light duty - Also meeting ACEA Class E3 for heavy duty - Sulphur content : % m/m This quality oil will be suitable for use in both light duty gasoline and diesel engines and in heavy duty diesel engines. CONCAWE are willing to co-ordinate supply of this oil. Prior to commencing emissions testing, the lubricant should be aged by running for at least 500 km in vehicles, or 500 km equivalent on engines. 9

13 4. Selection of engines, vehicles and after treatment technologies (Tasks 410 and 430) 4.1. Overall programme plan Selection of the engines, vehicles and exhaust after-treatment technologies to be tested has been made in view of the objectives to develop representative emissions factors for the current and future vehicle fleets and to identify the potential benefits available from the likely technologies to be used to meet the emissions control requirements of Euro-IV and beyond. To accomplish these targets, a range of vehicle technologies from Euro I through to Euro IV-V have to be evaluated, including a range of engine and combustion system types and a range of exhaust after-treatment technologies. Thus the programme will include : Light duty Diesel vehicles Range of engine technologies representative of Euro I to IV IDI and DI engines High/low pressure injection and common rail Heavy duty Diesel engines Range of engine technologies representative of Euro I through to Euro V Light duty gasoline vehicles Range of engine technologies representative of Euro I to IV Conventional MPI with TWC New G-DI - stoichiometric and lean burn Exhaust after-treatment A comparison of the likely technologies to meet Euro IV/V will be carried out, including oxidation catalysts, SCR, CRT, SiNOx, plasma system and particulate traps on diesel. For Diesel vehicles and engines, particulate trap technologies are likely to be one of the key routes to achieving future particulate emissions control standards. Hence specific tests will be included in some laboratories to evaluate vehicles / engines both with and without the traps in place in order to better understand the impacts of fitting particulate traps. Moreover, different types of PM traps will be tested: Cordierite, SiC, 3M fibre, For gasoline engines: TWC and NOx traps will be evaluated. All vehicles and engines will be tested over their respective legislative test cycles but in addition, further tests will be run to assess other off-cycle conditions, including steady states and transient real world driving cycles. The details of the test cycles to be used are given in section 5. Table 4 provides an overview of the overall test programme. This is followed by a summary of the individual programmes planned to be run by each partner. 10

14 Table 4 Overview of Particulates test plans Gasoline up to Euro II LD VOLVO INRETS CONCAWE AVL VKA TUG MTC AEAT (1) EMPA IFP LAT VTT 4V*1F*7C Gasoline Euro III LD 2V*1F*7C 2V*5C*3F 2V*5C*1F 1V*2C*3F 3V*4C*2F Diesel up to Euro II LD 6V*1F*7C Diesel Euro III LD 3V*1F*7C 2V*5C*4F 1V*5C*2F 1V*2C*4F 3V*4C*4F Diesel up to Euro II HD 1Euro-I*3F*2C 2V*3C*3F 1V*2C*4F Diesel Euro III HD 1V*3F*2C 2V*2C*4F 1V*3C*2F 1V*2C*4F Alternative fuels LD 11 1V*5C*2F (alcohol blends) Alternative fuels HD 1V*3C*RME 1V*5C*NG Engine Management LD 1V*3F*3 modes Aftertreatment LD 1V*2Traps 1V*5C*2F 2V*5C*1F Aftertreatment Cold Start (-7 or -15C) Cycle HD LD 1V*2F*2C [CRT] ETC + ESC NEDC+6RWC (2Cold+4Hot) NEDC +RWC+SS 2V*2C*3F [CRT vs. SCR] ESC+ETC SS U+R+H +ETC sim 4V (Normal gas, G-DI, DI diesel, common rail) *5C*2F*3T NEDC+RWC +3SS 2V*1C*1F*2T NEDC+RWC +3SS 4V*1F*3T NEDC+RWC 1*G-DI *2C*3F 1*Dies.*2C*4F 1*Dies.*1C*1F (3) NEDC+RWC + specific trap loading / regen cycles 1V*4Traps *4F (4) NEDC+3RWC +SS Size distribution b b b b b b b b b b b b Gaseous Regulated b b b b b b b b b b b b Non-Regulated SOF/EC, PAH CH 4, PAH, aldehydes NH 3, PAH, Aldehydes, SOF/EC NH 3, PAH, butadiene, aldehydes, aromatics, S compounds (2) metals PAH + SOF + metals Particle Morphology b b (2) b b b PAH, SOF/EC, metals 1V*1C*CNG 1V*1C*LPG 3V*2C*3F [CRT & SiNOx] Abbreviations: Notes: V = Vehicle / engine ESC = European Steady Cycle (1) Also investigating relationship between lube oil consumption and particulate emissions C = Cycle ETC = European Transient Cycle (2) To a certain extent on key tests F = Fuel ETC sim = ETC simulated (3) 1 gasoline G-DI *2C*3F (TWC + NOx Trap) T = Temperature U+R+H = Urban+rural+highway 1 Commercial Common rail Diesel*2C*4F (Oxidation catalyst + PM trap) NG = Natural Gas RWC = Real world cycle 1 Prototype Common rail Diesel*1C*1F (PM trap + NOx trap) RME = Biodiesel TWC = 3-way catalyst (4) Total number of cycles equivalent to 200 UDC (Urban Drive Cycles) per trap and fuel SS = Steady state CRT = Continuously regenerative trap NEDC = New European Drive Cycle SCR / SiNOx = Selective catalytic reduction (urea) LD = Light Duty HD = Heavy Duty ESC + ETC on engines Transient on vehicles

15 4.2. VOLVO Volvo will study particle emissions from heavy-duty engines tested under realistic conditions on a test bench equipped with the necessary sampling devices (Tasks 520 and 540). Work Plans 1) Implement the proposed sampling strategy from work package 300 in the Volvo engine test rig for heavy-duty engines. 2) Study the aerosol size distribution, based on numbers as well as mass, of exhaust particulate matter from modern engines as well as engines from previous generations. The sampling protocol derived from WP 300 will be used and the experiments will be performed in an engine test rig according to the following study design: Table 5 Volvo Test Plans Engine Fuel Test cycles Euro I D-1 EN ppm S D-4 EN590 8 ppm S D-5 Swedish Class 1 Stationary (ESC) Transient (ETC) Euro III D-2 EN ppm S D- 4 EN590 8 ppm S D-5 Swedish Class 1 Stationary (ESC) Transient (ETC) 3) Study the influence of CRT-based filter technology on particle emissions from an Euro III engine. Swedish Class 1 fuel (D-5) and two driving cycles (ESC and ETC) will be used. Fuel D-4 will also be tested if budget allows once the final sampling and testing protocol has been defined by WP-300. Measurements Regulated emissions (gaseous and particulate mass) Particulate size and number distribution Non-regulated emissions such as soot (AVL 415 meter), SOF/EC, and to supply filters to Stockholm University for PAH analysis. Deliverables Knowledge on: sampling conditions for particulate characterisation in full-scale test rigs particulate emissions from current and previous generation of heavy-duty engines effects of fuel quality on particulate emissions effects of exhaust filtration on particulate emissions 4.3. INRETS The task of INRETS is to map out the performance of the existing light duty fleet of gasoline and diesel vehicles, Euro I to III technology level, on current (year 2000) specification fuels (G-1 gasoline and D-2 diesel). (Task 510). Vehicle tests were originally planned to be performed on a chassis dynamometer using a fleet comprising approximately 30 passenger cars. A range of driving cycles would be included, including the legislated cold cycle (NEDC), 2 cold real-world cycles and 4 hot real-world cycles. 12

16 Due to the time shift between ARTEMIS and PARTICULATES projects, INRETS is now unable to measure particulates sizing according to a new methodology, on 30 LDVs as initially planned. Measurements for ARTEMIS have started by end March and should last 1.5 year. New equipment for particulate sampling and sizing might be operating by the end of year This implies new investment for the laboratory, so that we can reasonably plan the new particulate measurement on half of the initial fleet, i.e. 15 LDV. The distribution in the various vehicle categories could be as followed: Table 6 INRETS Measurement Programme Cycle (standard temp.) Starting conditions hot cold Regulated pollutants Particulates sizing Non-Reg. Fuel Nb * Veh. type INRETS Technology and cycle effects NEDC * * INRETS specific urban * * INRETS specific road * * INRETS specific motorway * * INRETS urban short * * INRETS road short * * CADC urban * * CADC rural * * CADC highway * * ELPI PAH, Aldehydes 1 Fuel * 2 Gas. Euro I 2 Gas. Euro II 2 Gas. Euro III 3 Dies. IDI 3 Dies. DI Euro II 3 Dies. DI Euro III 1 Fuel 5 LDV (in the 15 above) * Details of the driving cycles are given in section 5. Measurements : Regulated emissions (gaseous and particulate mass) Particulate size and number distribution CO 2, FC, methane, PAH and aldehydes Deliverables : Mapping of current light duty fleet on current gasoline and diesel fuels 4.4. CONCAWE CONCAWE plans to investigate advanced (Euro III-IV) light duty vehicles with current and potential future fuels over a range of test conditions (Task 510). CONCAWE will also cooperate with AVL on their heavy duty engine tests. LD Vehicles (Euro III - IV) Two G-DI gasoline cars Two DI diesel cars (small car without PM trap and large car with PM trap) Fuel matrix Gasolines: G-1 to G-3 Diesel fuels: D-2 to D-5 Test cycles NEDC, hot start CADC and several steady state conditions Statistical test design with sufficient replicates to establish vehicle and fuel effects Standard WP 300 sampling and testing protocols on transient tests Steady state tests will be used to complement transient data by providing particle size distribution by SMPS for a range of fuels and sampling conditions Measurements : Regulated emissions (gaseous and particulate mass) Particulate size and number distributions 13

17 Deliverables : Knowledge on performance of advanced LD vehicles and fuels AVL AVL's contribution will involve testwork on advanced heavy duty engines and evaluation of two exhaust after-treatment technologies which are likely to be employed to meet Euro IV or Euro V emissions limits (Tasks 520 and 540) : Engines - Prototype Medium HD (1 l/cyl) engine - Prototype Heavy HD (2 l/cyl) engine - Both engines meeting Euro-III without exhaust after-treatment and Euro IV-V with aftertreatment After-treatment systems - One engine to be tested with EGR and a CRT system - One engine to be tested with a SCR system Fuels - Current diesel fuel versus potential future qualities (D-2 to D-5) Cycles - ESC and ETC on heavy duty - Steady states including selected modes with extended sampling/testing Measurements - Regulated emissions (gaseous and particulate mass) - Particulate size and number distribution - Non-regulated emissions, including aldehydes, PAH, ammonia Deliverables : Knowledge on performance of advanced HD engines and fuels VKA VKA's contribution will involve the evaluation of the effect of advanced Diesel engine technology on particulates and particle emissions character (Tasks 530 and 540). The effect of advanced Diesel engine combustion technology shall be investigated under well defined boundary conditions taking into account latest knowledge about particle characterisation technology. * Tests will be carried out on an advanced (Euro IV?) passenger car common rail Diesel engine. * Access to the engine ECU will enable the study of the effects of relevant engine control parameters. * Testing will be carried out under steady-state conditions on a test bench. Three mapping points: (2000 rpm / 2bar) and (2000rpm/6bar) and (3500rpm/6bar) will be covered. Combustion parameters subject to testing: * injection pressure * EGR rate * boost pressure * pre-injection * post-injection 14

18 Variables: * speed/load * dilution ratio * fuel (future Diesel qualities) Fuels * Different fuels will be tested to evaluate the influence of sulphur content (D-2 to D-4) * VKA plan to perform measurements with 3 different dilution ratios as well as 3 different temperatures of the probe before entering into the SMPS. With the variation of probe inlet temperature into the SMPS we will separate small liquid particles from solid particles, thus eliminating the influence of sulphate production on the small particles. After-treatment In the final stage, an investigations of the impact of a particulate trap with different materials (SiC, Cordierite) will be included. Measurements - Regulated emissions (gaseous and particulate mass) - Particulate size and number distribution (soot and volatile particles) - particle morphology - SOF/soot ratio Deliverables Particulate mass and size distribution as a function of engine combustion parameters 4.7. TUG TU-Graz will carry out measurements on a chassis dynamometer for 2 heavy duty vehicles, including a test of RME (Tasks 520 and 530). In addition, one engine will be tested on the engine test bed in steady state and transient cycles. This engine will be used for evaluation of the DEKATI dilution system and detailed cross-checks on the most suitable set-up for Heavy Duty measurements in cooperation with AVL. Test plans include : - 1 HD engine on the transient engine test bed with the measurement systems from DEKATI, Prof. Reischl and an AVL secondary dilution system running in parallel. Different options of dilution (heated, not heated, residence time,..) will be tested on the effects on particulate number and size distribution. - 2 HD vehicles in 3 Cycles with 3 diesel qualities, 2 measurements per cycle and fuel - 1 of the HD vehicles will also be tested with RME TUG will also carry out an analysis of particulate morphology. Soot particles from Diesel engines can be characterized by transmission electron microscopy (TEM). The particle size of single soot particles is in the range of nanometers and these particles are often agglomerated to bigger structures. By employing TEM it is possible to get information on the morphology of the soot (e.g. particle shape, particle size) at a resolution of about 0.2 nm. The soot particles are collected directly unto copper grids which are covered with a thin electron transparent polymer film. In order to get a representative information the specimens have to be investigated in several viewing fields and at different magnifications MTC MTC will carry out tests on one Euro III diesel passenger car over 5 cycles (NEDC, RWC and 3 steady states) with two fuels - current diesel and Swedish Class 1 (Task 510). They will also evaluate alcohol fuel blends over the same cycles (Task 530). 15

19 MTC will test current Diesel fuel versus Swedish Class 1 in one LDV with exhaust aftertreatment (Task 540). This will involve the particle filter system developed by Peugeot- Citroen, and will be tested over the same test cycles. MTC will also evaluate 4 light duty vehicles under cold temperature test conditions, including normal gasoline, G-DI, DI and common rail diesel. Performance will be measured over the 5 test cycles, with two fuels and 3 temperatures (Task 550). MTC will also evaluate a heavy duty CNG engine (Task 540) and oil consumption effects (Task 560). MTC plan to measure other non-regulated components, i.e. PAH analysis in co-operation with Stockholm University, ammonia, butadiene, aldehydes, aromatics and sulphur compounds, on selected key tests from their programme AEAT AEAT will carry out testing on light duty engines (gasoline and diesel) in the areas of emerging after-treatment concepts, cold start temperature effects and the influence of oil consumption (Tasks 540, 550, 560). AEAT s activities within WP 500 are generally concerned with the investigation of emerging vehicle technology effects which may have significance for future regulatory policy planning but which are not covered in the project s main technology/fuels test matrix. The selection of technologies for study is based on : - requirements for regulatory policy planning - availability and practicality of emerging emissions abatement technology - compatibility with the work programmes of other partners. Particle characterisation will be carried out on the basis of the recommendations to emerge from WP300. The planned work programme includes measurements on two emerging after-treatment systems on LD diesel vehicles. One of these systems is plasma based and the other has yet to be decided upon. Measurements will be made pre- and post-device where possible. For the post-device measurements, cycles will include standard NEDC, the project s hot CADC real world cycles and three steady states. We also aim to investigate the characteristics of particulate emissions from gasoline fuelled vehicles of topical relevance to regulatory developments. It is likely that the vehicle technologies will be drawn either from emerging G-DI systems or from motorcycles relevant to the 2003 regulatory stage. Two vehicles representative of major components of the current UK parc will be selected for testing. G-DI system selection would take account of the activities of other partners in the project. In the event of motorcycles being selected it would complement the current activity within ARTEMIS wherein a large matrix of motorcycles is being emissions tested but without particulate emissions characterisation. In either case the aim would be to investigate the most recent technologies thereby filling gaps not covered by other recent projects. Cycles will include standard NEDC, appropriate real world cycles and three steady states. Measurements will be made at regulatory standard temperature and also the effect of cold start temperatures down to -7 C. Oil consumption effects will be investigated with a view to determining whether or not they are important in future regulatory policy planning. This work will be centred on a LD gasoline engine. The aim is to investigate the characteristic particulate emissions under operating conditions that generate proportionally higher and lower rates of oil consumption. AEAT will also attempt to indicate which characteristic drive cycle events are most prone to oil-sourced contributions to particulate emissions. This is an ambitious objective for which the research techniques have yet to be established and therefore the detail of the experimental approach 16

20 will evolve as the task activity progresses. It is anticipated that this task will involve experiments on AEAT s LD transient engine test facility although this may need to be backed up with tests on a high oil burning vehicle. AEAT anticipate co-operation with JRC Petten on trace metal analysis in support of this task. In general the activities to be undertaken by AEAT are of an exploratory rather than routine measurement nature. At this stage they are not planning to investigate fuel effects per se. However, they retain the flexibility to include fuel effects at the expense of other planned variables should the evolving needs of the project so demand. Measurements Regulatory gaseous emissions and particulate filter mass Particle number and size distributions. Particle number emissions versus individual events in transient cycles Deliverables Knowledge on advanced after-treatment devices, including plasma Temperature effects on particulate emissions from gasoline vehicles (G-DI or motorcycles) Understanding of the impact of oil consumption EMPA EMPA will investigate the effects of cold temperature testing on light duty vehicles (Task 550). Tests are planned to be carried out consistent with the test definitions of the ARTEMIS project (Task 323) - Cold Start LDV (ambient temperature, -7 C, -15 C) - 4 vehicles - 1 test cycle - 1 fuel (standard commercial) Measurements Regulated gaseous emissions and particulate filter mass. Particle size and number distributions and particle morphology. Deliverables Understanding of the influence of start temperature on particle emissions IFP IFP will investigate the influence of after-treatment technologies on particulates emissions (Tasks 510 and 540). Engine and after-treatment technologies have been chosen to be representative of the present and future European market (Euro III and Euro IV). Vehicles and after-treatment systems Five vehicles will be tested with both gasoline and Diesel engines and different aftertreatment systems: Gasoline Euro III (MPI with a 3 way catalyst) Diesel Euro III (DI with an oxidation catalyst) Commercial vehicle with gasoline direct injection engine and NOx trap and 3 way catalyst (VW Lupo FSi or PSA with HPI engine) Commercial vehicle with Common-Rail Diesel engine and particle trap (PSA 607) Prototype vehicle with particle trap and/or NOx trap 17

21 The prototype vehicle, based on a Peugeot 406 with a common-rail engine has been modified by IFP to allow the control of engine parameters. Using this system, a specific algorithm will be developed and tested to evaluate the influence of after-treatment system on particulate emissions. Test cycles Tests will be performed following the NEDC cycle and the Real World Cycles (RWC), consistent with the Artemis programme. Furthermore, IFP will carry out tests to assess the effect of loading and regeneration of traps on emissions. For that purpose, specific loading and regeneration cycles will be used. The test matrix will be the as shown in Table 7: Table 7 IFP Test Matrix Vehicle Fuels NEDC RWC Loading / regeneration cycles Gasoline Euro III G1 to G3 X X Diesel Euro III D2 to D5 X X Commercial Gasoline DI engine G1 to G3 X X X Commercial Diesel Common rail D2 to D5 X X X Prototype Diesel Common rail D3 X Measurements Regulated PM and gaseous emissions (continuous and cumulated) Particulate size and number Unregulated: PAH (By Stockholm University), SOF, trace metal (by JRC) Deliverables Results of emissions on Euro III vehicle Effect of fuel properties (mainly sulphur content) on particle trap and NOx trap performance. Emissions before, during and after regeneration phases. Influence of NOx and PM trap implementation and management on gaseous and particles emissions, including size distribution LAT LAT will assess current gasoline and diesel cars on current fuels and to evaluate the impact of potential future particle trap technologies and fuel qualities on diesel emissions (Tasks 510 and 540). The following programme is planned : Gasoline Euro III Diesel Euro II-III Diesel after-treatment 3 vehicles, 4 transient cycles, 2 gasolines (G-1 and G-3) 3 vehicles, 4 transient cycles, 4 fuels (D-2 to D-5) 1 vehicle, 4 traps, up to 4 fuels (D-2 to D-5) Test cycles: NEDC + 3RWC + steady states Vehicles : 1996 Golf TDI - Euro II (when equipped with oxicat) Common Rail Renault Scenic Euro III 18

22 Others to be confirmed Particulate traps to be tested : It is planned that one diesel car will be tested with 4 different types of traps and/or regeneration systems and with up to 4 different fuels. The car selected will be equipped with different traps in order to comparatively evaluate their performance, on the basis of a test protocol specifically developed for trap equipped cars (incl. loading and regeneration cycles). The different types of traps currently available are: * Corning EX80 and NGK C558 (i.e. design currently in the market - cordierite) * Ibiden SiC traps (i.e. design currently in the market - cordierite) * New design NGK traps (both cordierite and SiC) * 3M fiber traps As regards regeneration initiation systems: It is planned to compare additive supported regeneration (Cerium or Manganese based) with initiation of regeneration by different means (such as turbocharger bypassing). Measurements Regulated PM and gaseous emissions Particulate size and number Unregulated : PAH, SOF/EC, TEM/SEM Deliverables Understanding of performance of current light duty vehicles, impacts of particle traps and fuel quality PSA PSA have recently requested to join the programme and their contribution has yet to be defined VTT ENERGY VTT propose to evaluate heavy duty diesel engines with a range of after-treatment devices, including CRT and SiNOx systems. VTT is also planning to run 4 city buses, on a chassis dynamometer. It has not yet been specified which individual buses would be tested, but it is planned to include standard diesel, CRT equipped diesel, CNG and LPG. Tests should compare current versus potential future diesel fuels (D-2 to D-5). On engines/vehicles with the after-treatment devices, only the ultra low sulphur fuels, D-3 to D5 should be tested. The programme outline is shown in Table 8. 19

23 Table 8 Proposal for the heavy-duty test matrix at VTT Engine/vehicle Aftertreatment Fuel Test cycle Volvo DH10 no D2, D3, D4, D5 ESC and ETC Volvo DH10 CRT D3, D4, D5 ESC and ETC Scania DC11 no D2, D3, D4, D5 ESC and ETC Scania DC11 SINOX D3, D4, D5 ESC and ETC 2 diesel buses* with & without CRT D3, D4, D5** transient on chassis 1 natural gas bus* no CNG transient on chassis 1 LPG bus* no LPG transient on chassis * individual buses are not yet specified ** for buses equipped with CRTs, only ultra low sulphur fuels will be tested Table 9 Characteristics of the diesel engines. VOLVO DH10A-285 SCANIA DC Application bus engine truck engine Displacement, dm Number of cylinders 6, in-line 6, in-line Maximum power min min -1 Maximum torque min min -1 Injection system direct-injection, in-line pump with EDC electrically controlled pumpinjector system turbo, water/air cooling turbo, air/air cooling Aftertreatment no/crt no/sinox Emission level Euro 2 Euro 3 Aftertreatment devices For heavy-duty engines, current technology, i.e. the combination of modifications to the engine and traditional aftertreatment devices will not be sufficient to reduce nitrogen oxides (NOx) to the limit values for Euro IV and beyond. Oxidation catalysts, currently used for converting HC and CO and to reduce particulate matter (SOF) also convert small quantities of NOx. However, conversion levels above 20 % NOx cannot be achieved. The most promising technology for NOx reduction is the use of selective catalytic reduction (SCR), which is used in the SiNOx system developed by Siemens. This technology is based on the addition of urea, which reacts highly selective to the nitrogen oxides. The catalyst used is a bulk extrudate based on titania (TiO 2 ). An aqueous solution (32.5%) of urea is injected into the exhaust gas stream at temperatures above 200 C and reacts with water to form ammonia and CO2. The formed ammonia adsorbs to the SCR catalyst and reacts with the nitrogen oxides to form nitrogen and water. The SiNOx system meets the stringent Euro IV limits. The SiNOx system, that VTT has comprises of the following elements: urea tank needed for storing the urea solution, dosing unit for bringing the urea into the exhaust gas stream and the SCR catalyst. The continuously regenerating trap (CRT) system is a concept for effective particulate treatment. It consists of a ceramic particulate filter and an upstream oxidation catalyst. The catalyst converts unburned hydrocarbons and CO and oxidises NO to NO2. The NO2 is used to oxidise the diesel particulate matter deposited on the particulate filter. In this way a continuous regeneration of the filter in assured. The CRT requires practically sulfur-free fuel 20

24 for a faultless operation. Diesel fuel with a sulfur content of less than 50 ppm is required, because SO 2 disturbs the formation of NO 2 in the exhaust gas and generates sulphates in the oxidation catalyst which block the filter permanently. The CRT system that VTT possesses has been sized to be used with the Volvo engine assembly Stockholm University Stockholm University will support the programme with detailed compositional analysis of solid particles versus size distribution, including PAH content. A copy of Stockholm University s draft protocol for PAH analysis is attached as Appendix JRC Petten The contribution of the JRC is to perform analysis of trace elements. JRC Petten plan to provide detailed trace metal analysis on particulate filters using an activation technique, including a co-operation with AEA Technology on the impact of lube oil consumption on particulate composition. Neutron activation analysis is a powerful tool to determine a great variety of elements in environmental, biological and geological samples. By neutron capture, a compound nucleus is formed in an excited state, which decays almost instantaneously to the ground state, by gamma-ray emission. During the process a signal is induced when the nuclei of individual atoms capture neutrons to produce excited states that immediately decay by the emission of photons (prompt gamma rays). In most cases a radio-nuclide is produced that continues to decay according to its individual decay scheme and half life through one or more excited states that also emit photons (delayed gamma-rays). For example the measurement of manganese by neutron activation would use the following nuclear reaction: 55 Mn 25 + [ 56 Mn 25 ] + Y prompt 0 ß -1 + Y delayed + 56 Fe 26 (stable) The prompt gamma rays are emitted in coincidence with the capture of the neutron and must be measured in real time. The delayed gamma rays are emitted according to the half-life (2.58 hrs.) of the induced Mn-56 radionuclide. The net effect is that a very small fraction (on the order of 10-6 %) of the Mn-55 target atoms are converted to radioactive Mn-56 product atoms and ultimately stable Fe-56 atoms. Gamma-ray energies are discrete and can be accurately measured by pulse-height analysis using high-purity germanium (HPGe) semiconductor detectors. The gamma-ray energy is used to ascertain the presence of an element (qualitative analysis) and the intensity is used to find its concentration (qualitative analysis). The detailed procedures to be used for the Particulates programme are still under development due to low amount of particulate mass, likely to be available. Hence the detailed protocol will be included at a later stage, prior to the commencement of WP

25 5. Definition of drive cycles and test conditions (Task 440) The specific detailed sampling and testing protocol for the measurement of exhaust gas particulate emissions will be described in the report of Work Package 300. This section provides a description of the various engine and vehicle test cycles to be used, as well as the proposed test sequence and minimum test requirement for both light and heavy duty tests Light duty vehicle procedures NEDC The standard NEDC (or MVEG) cycle is the current year 2000 cycle for emissions certification of light duty vehicles and should be used as the core transient test for both gasoline and Diesel light duty vehicle tests. Tests should be carried out according to the standard NEDC procedure (ref. 3). A short description of the test is given below : The full NEDC cycle is a combination of the urban ECE-15 from key-on, with the higher speed EUDC cycle. The first part of the NEDC cycle is the ECE-15 which defines an urban test cycle, as shown in Figure 1. The cycle was devised to be representative of city-centre driving and thus has a maximum speed of only 50 km/h. Prior to testing the vehicle must be preconditioned and then allowed to soak for at least 6 hours at a test temperature of between 20 and 30 C. It is then started, emissions sampling commences from key-on and the ECE-15 mode driving cycle is repeated four times without interruption. Emissions are measured by the "Constant Volume Sampling" (CVS) technique. After 4 repeats of the ECE-15 cycle, the higher speed, Extra Urban Driving Cycle (EUDC) (Figure 2) is run. Figure 1 ECE 15 Test Cycle (repeated four times)

26 Figure 2 Extra Urban Driving Cycle Characteristics ECE 15 Cycle EUDC Cycle Distance km 4 x = Time sec 4 x 195 = Average Speed km/h Maximum Speed km/h Acceleration % Time Acceleration m/s 2 (max.) Deceleration % Time 13.8 Deceleration m/s 2 (max.) Idle % Time Steady Speed % Time CADC (Common Artemis Driving Cycles) or Real World Cycles (RWC). The CADC (Common Artemis Driving Cycles) are the ARTEMIS reference cycles and have also been named as RWC (Real World Cycles). From large databases and a strong experience in the field, INRETS has built a set of reference driving cycles (speed profile vs. time), which are to be used within the framework of national and European projects aiming at the assessment of emission factors from light-duty vehicles. Built initially for the ARTEMIS project, these cycles have been called CADC. They provide for all partners an harmonised and common test procedure when measuring emissions from light-duty vehicles. Methodology of setting driving schedules is based on the analysis of the real-world conditions of use and operation of the vehicles and on the analysis of the traffic, through the speed profile features. Correspondence analyses and classification of the cross distribution speeds and modal accelerations are part of the data processing. The databases consist of km monitored on board 80 passenger cars in France, Germany, Great Britain and Greece. These data are supplemented by a few km obtained in Switzerland and Italy under controlled traffic conditions. These specific well-described data in terms of road characteristics allow the classification of the first group of data according to the established relations between speed profile features and road types. Their analysis makes it possible to establish a typology of the urban, road and motorway conditions of traffic in 12 standard classes, from a great diversity of observed cases. The structure of the trips and their distribution in these various traffic types is then used for the development of 3 driving cycles representing the urban, the road and the motorway driving behaviours. These 3 cycles, with a whole duration of 40 minutes, describe the various and most current driving conditions encountered in Europe (figures below). In addition, an algorithm of determination of the gear to be used each second of these cycles has been 23

27 developed and applied on a typology of European cars. Thus 10 laboratories taking part in the ARTEMIS project are measuring the actual emissions of cars, under the most representative driving behaviour. The complete procedure to perform these test cycles will be provided by INRETS. Figures 3-5 The CADC cycles: urban, road and motorway types and their structure according to the traffic types. Figure 3 Figure 4 24

28 Figure Additional transient light duty cycles INRETS plan to carry out a comparison of several other cycles with the standard NEDC and CADC cycles These other cycles include: INRETS cycles by vehicle category (called INRETS specific in Table 6) These cycles were developed from the same database used for CADC cycles and with the same statistical tools but according to a split of the monitored vehicles into 2 categories of specific power: those having a power/mass ratio under 0.06 kw/kg and those above this limit. The cycle structure (acceleration) and the speed levels are then more adapted to the various tested vehicles. The selected sequences which form the driving cycles are part of 2 different data sets. INRETS short cycles These 2 cycles were specially designed to assess the influence of cold start on emissions. By the repetition of a well identified sequence, typical of the urban or the road driving behaviour, the evolution of emissions can be followed precisely according to time or mileage and engine temperature. Further details on the INRETS cycles by vehicle category and the INRETS short cycles are given in Appendix 1. 25

29 5.2. Heavy duty engine procedures Relevant heavy duty engine emissions test cycles should be used. In general the ESC and ETC tests will be used together with selected extended steady state modes. The ECE R-49 test may also be used for older heavy duty engines. Tests should be carried out in accordance with the standard procedures (ref.4). Brief descriptions of the cycles are given below ECE R49 Heavy Duty Engine Exhaust Emission Test Procedure This procedure, superseded in the EU in 2000, was developed as a test for medium and heavy duty diesel engines operating in Europe. It is an engine rather than a vehicle test and a summary of the modes employed are as follows: Table 10 ECE R49 Test Modes Mode Speed Load % Weighting Factor 1 Idle Intermediate Intermediate Intermediate Intermediate Intermediate Idle Rated Rated Rated Rated Rated Idle ESC (European Steady Cycle) The test speeds for the ESC cycle were chosen following an investigation of the power curves of a number of modern engines, which showed that the usable speed range of an engine lies between 50% and 70% of rated power, before and beyond the peak in the power curve respectively. The three test speeds are determined by dividing this speed range into four equal sectors (see lower diagram of Figure 6). In addition to idle, the other 12 modes are set at a combination of the three speeds established above and at 25%, 50%, 75% and 100% load. Weighting factors are assigned to each mode as shown in Figure 6. To ensure that there are no rogue operating conditions which give abnormally high emissions, three more modes can be selected within this operating envelope by the personnel certifying the engine. For acceptance, the values measured in these modes must correspond with the emission characteristics measured at the fixed cycle points within established tolerances. The test sequence is selected to test engines with exhaust gas treatment systems at realistic temperatures, it being considered sufficient to hold the engine under each condition for no longer than one minute to reproduce actual operating conditions. For the dynamic load response test, the engine is preconditioned and then accelerated from 10% load to full throttle at maximum acceleration. In this way the engine runs through all the fuel/air mixture conditions defined by the management system (Figure 7). 26

30 Figure 6 European ESC/ELR heavy duty exhaust emissions procedure - 13-mode cycle Additional modes determined by certification personnel Mode Weighting Factors, Speed Definition 27

31 Figure 7 European ELR dynamic response test for smoke emissions 100 Torque (%) 50 Opacity 10 0 Time A B C 2000 Idle Speed (rpm) ETC (European Transient Cycle) This cycle was developed from data of road speed/power curves of typical driving patterns collected in a collaborative project by German and Swiss authorities for the purpose of updating emissions factors. The data were analysed in terms of road type (motorway, rural and urban), traffic density, road gradients and vehicle weight, distances between congestion or stopping points and vehicle type (lorries, coaches and local buses). The time curve of vehicle speed was normalised for engine speed and torque (Figure 8) assuming a vehicle weight of 28 tonnes. The normalised figures were at first integrated into three sub-cycles of 15 minutes, but were finally reduced to ten minutes each. The frequency pattern of the cycle is in good agreement with the average data for local and long-distance lorries. For establishing limits, emissions from the three sub-cycles may be measured separately and combined using weighting factors. 28

32 Figure 8 European ETC heavy duty exhaust emissions procedure - Transient cycle 5.3. Heavy Duty Vehicle procedures Transient test cycles to be used in PARTICULATES for Heavy Duty vehicle testing will be chosen from the ARTEMIS cycles according to the measurement results in ARTEMIS to cover the most interesting engine operating regimes. Urban, rural and highway cycles will be included, but the initial ARTEMIS results have to be studied in more detail in order to finally define the cycles to be used for PARTICULATES. This further definition on the test cycles will be provided by TUG. The range of HDV transient cycles used in ARTEMIS is shown in Figure 9. VTT have recently joined the project and plan to contribute with measurements on heavy duty bench engines and several city buses. For the city bus tests, specific transient bus cycles will be employed on a chassis dynamometer, details to be defined by VTT. 29

33 Figure 9 Artemis Cycles Artemiszyklen speed [km/h] _ _ _ time [sec] 5.4. Test Sequences, Vehicle Conditioning and Fuel Change Procedures Conditioning procedures need to be defined in order to bring vehicles and engines to standard conditions prior to testing. In general, several repeats of the relevant warmed up transient cycles are expected to be required for light duty vehicle tests and steady state preconditioning for the heavy duty engine tests. The details of the conditioning procedures will be finalised following completion of WP 300 s protocol. Specific fuel change and handling procedures will be required to avoid contamination from previous fuel tests and to ensure sulphur purging from the engine/vehicle systems when changing sulphur levels. Proposed test sequences for light and heavy duty testing are given in Figure 10 and Table 11 below. A detailed fuel handling procedure for gasolines is given in Appendix 2. 30

34 Light Duty Vehicle Test Sequence Figure 10. Flow chart for testing Light Duty Vehicles TEST SEQUENCE START SAME TEST FUEL? NO YES DRAIN FUEL CONDITIONING CYCLES Gasoline - 1 ECE + 2 EUDC Diesel - 3 EUDC COLD SOAK HOURS 10 LITRE FILL 5 MIN IDLE DRAIN FUEL NEDC (COLD) EMISSIONS TEST 25 LITRE FILL CADC TESTS (HOT) STEADY STATE TESTS - 50, 90, 120 km/h - road load NO FUEL MATRIX COMPLETE? YES STOP Heavy Duty Engine Test Sequence For heavy duty engine and vehicle tests, in general these will be run from auxiliary fuel tanks, so draining the fuel system for fuel change is not required. It is sufficient to purge the fuel system and condition by steady state running. If the vehicle s fuel tank is used for any heavy duty vehicle tests, a similar protocol to that for the light duty vehicles should be followed with a suitable fuel flushing volume. A common rigorous test sequence will be required in order to obtain comparable results from different labs and for different fuel/engine combinations. A proposed daily test sequence is given in Table 11, based on the format used in the DETR/CONCAWE/SMMT programme (ref. 5). This includes ESC and ETC tests as well as selected extended steady state modes. In the DETR/CONCAWE/SMMT programme, it was found that the first ESC of the day showed poorer day to day repeatability than the second. On this basis, 3 ESC tests are included in the proposed test sequence, with the intent to use the data from the second and third tests only. 31

35 The extended steady state modes and multiple ETC tests provide opportunity for generation of more data on size distribution profiles and/or testing of more than one exhaust sampling/dilution condition depending on the final outcome of WP 300. Table 11 Heavy Duty Test Sequence Test Steady State Transient Order Warm-up 1h + power curve 1h + power curve 1 ESC* ETC set-up 2 ESC ETC set-up 3 ESC ETC 4 Mode 2 ETC 5 Mode 4 0.5h cond. 6 Mode 7 ETC 7 Mode 10 ETC 8 Mode h cond. 9 ETC 10 ETC * R-49 may be used for older engines 5.5. General test design, number of repeats In order to produce representative test results, several repeat tests will be required for each vehicle/fuel/cycle combination. Light Duty Where laboratories are testing only one fuel, 3 repeats of each test condition should be the minimum requirement. Where laboratories are testing more than one fuel in an engine/vehicle, a randomised block design of the type shown below should be employed. Three blocks is expected to be the minimum test requirement to produce statistically valid results. However, this should be finalised once further information on the test precision is known from WP 300. An example of a test fuel sequence for one vehicle/engine is given below. Test Order (Fuels D2 D5) Fuel: Test Block 1 D3, D5, D2, D4 Test Block 2 D5, D4, D3, D2 Test Block 3 D4, D2, D5, D3 The order of fuel testing should be randomised across all vehicles and engines to be tested. The final test design, definition of expected precision range and number of repeats should be finalised once the expected precision is better known from the work of WP 300. Repeated test results on the same fuels should be checked against the expected precision range. If outliers are detected, the engineering cause should be investigated. If necessary, an additional test block should be run. In general, outliers should only be rejected from the final analysis if a good engineering reason for the outlying results can be identified. 32

36 Heavy duty In general the same test design principles should be applied to heavy duty tests. However, since the core heavy duty tests involve warmed-up cycles, repeat results can already be obtained within each days testing. Day to day repeatability data if of course still required since different fuels are tested on different days. Two test blocks may be sufficient for heavy duty but again this should be confirmed once knowledge is available from WP 300. The planning base should be two test blocks with a third block to be run if adequate precision is not achieved Procedures for testing systems with regenerative devices For regenerative exhaust after-treatment devices, e.g. PM traps, specific procedures will be required in order to evaluate performance under their typical operation and produce meaningful emissions factors. For each vehicle/trap/fuel combination, the test protocol is expected to consist of three main parts: (a) The loading phase. It is proposed to use a repetition of a limited number of urban drive cycles, e.g. 25 UDCs, equivalent to about 100 km, starting with a clean (regenerated) filter. Thus in principle the same quantity and quality of particulates is accumulated in the trap. Trap loading should be followed by monitoring the differential pressure across the trap (b) The particulate emissions tests (c) The regeneration phase. Regeneration is induced by running the engine under high speed and load conditions, to clean the trap by reaching a sufficient exhaust gas temperature for particulate combustion. Typical procedures have been described in the literature (refs. 6,7). However for the Particulates programme, specific sequences will need to be developed for both light and heavy duty tests at the start of the test programme. These are likely to be specific to the technologies to be tested and hence will be developed on a case by case basis at the start of the evaluation of each technology. 33

37 6. Roadside particulates measurements (Task 450) 6.1. Methodology for roadside measurements of particulate mass and size distribution Exhaust aerosols are largely modified upon emission to the atmosphere as a result of reactions between the particles and atmospheric gases. In order to investigate these effects roadside measurements are planned, combined with PM measurements in mechanically ventilated road tunnels. The measurements will be performed inside the tunnel (roadside), at the beginning of the exhaust air duct and at the end of the exhaust air duct as well as in ambient air. The following describes the methodology for roadside measurements of particulate matter mass and size distribution. The roadside measurements will be made in strong connection to the validation measurements undertaken in WP 1200 of ARTEMIS Measurement Location Roadside measurements in ambient air Roadside measurements will be performed at different distances from highways. The measurements will last for several weeks in order to cover different emission and meteorological situations. These measurements will be coupled with PM 2.5 mass measurements Measurements in road tunnels Road tunnels have the big advantage that they can be considered as big laboratories, because boundary conditions like air volume flow, traffic, etc. can be determined relatively easily. Experiments in two or three different tunnels are planned. One tunnel has a longitudinal ventilation, the second one a transverse ventilation and up to 250 m high stacks (mountain tunnel). The measurement in the transverse ventilated mountain tunnel allows for a clear differentiation between direct emissions under real world dilution ratios (roadside measurements) and aged PM at different locations from roadside. As the measurements of the aged PM can be performed in up to 250 m high ventilation stacks the ageing of the PM can be examined without disturbing influences from outside. Figure 11 shows a sketch of a measurement arrangement in a transverse ventilated tunnel. 34

38 PM2.5 + SMPS measurements Figure 11: Sketch of a transverse ventilated road tunnel. It is foreseen to undertake the measurements in the tunnel (roadside) and at the foot and the top of one exhaust air stack In the longitudinal ventilated tunnel it is planned to make parallel measurements at the tunnel inlet and exit. In this case two effects are coupled. The measurement at the inlet is influenced mainly by the background concentration and size distribution of the background air (of course an influence of the near by emissions from the passing road vehicle exists). The exit location monitors the PM emission inside the tunnel but already as a mixture of fresh PM emissions and aged emission. The location of the equipment can be seen in Figure

39 PM2.5 + SMPS measurements Figure 12: Measurement location in a longitudinal ventilated stack 6.3. Measurement dates It is foreseen to run the roadside measurements throughout the year 2001 in or close to the city of Graz, Austria. The tunnel experiments are planned for spring 2001 in Sweden, Summer 2001 in England and Winter 2001 in Austria Requirements for the equipment Particulate mass (TSP, PM 10, PM 2.5 ) It is suggested to focus on PM 2.5 to minimize the influence of re-suspended particles. The inlet has to be as short as possible and should consist of a vertical line only (no or very short horizontal lines) 1 day intercalibration in tunnel with the instruments next to each other Documentation has to include - device type - filter type - sample inlet head - temperature in case of heated filters - flow rate through filter - measurement time (sample time) - measurement location Particle size distribution with SMPS < 1µm Parallel measurements for half a day Calibration with Latex aerosols with known size Slow scan times (suggested 300 seconds up-scan and 100 seconds down-scan) Flows must be constant 36

40 7. Non-exhaust particulates measurements (Task 460) 7.1. Background The study of non-exhaust particle emissions from road vehicles is being conducted by TRL Ltd., in collaboration with the University of Hertfordshire (UH) and the Centre for Ecology and Hydrology (CEH). A literature review conducted by Warner et al. (ref. 8) revealed that non-exhaust particle emissions originate from a number of processes, including: Tyre wear Brake wear Clutch wear Road surface wear Corrosion of chassis, bodywork, and other vehicle components Corrosion of street furniture and structures Suspension or resuspension of road dust However, non-exhaust particles have received relatively little attention in the context of air pollution, and the data relating to the emission rate, size, and composition of particles arising from such sources are far from comprehensive, even though they may contribute significantly to atmospheric concentrations. Furthermore, as exhaust emission are subject to tighter control, the relative contribution of non-exhaust particulate matter is likely to increase. Task 580 involves a programme of experimental work which is designed to address some of the gaps in the understanding of non-exhaust particle emissions. This experimental work will concentrate on tyre and brake emissions, while the loss of material due to the corrosion and road surface wear will be estimated, where possible, from the literature. The experimental methodology is described below. This incorporates the recommendations presented in the literature review Experimental Methodology There will be four main strands to the experimental work: The gravimetric determination of tyre and brake wear rates The compositional analysis of brake and tyre materials, and brake and tyre dust The sampling and analysis of airborne particles The sampling and analysis of road dust The results from the various different tests will be combined in order to derive emission factors for brake wear and tyre wear. The analytical methods used may also allow for the determination of resuspension rates for road dust. The main objective of the gravimetric test work is to obtain an estimate of the total amount of component material being released to the environment during use. Initially, this will be equated to emissions of total particulate matter, though this will clearly need to be refined later as, for example, some of the weight loss due to abrasion will be in the form of gaseous combustion products, and some larger chunks of material will be lost. At present, it is not clear what proportion of the component material lost to the environment is emitted as airborne PM, and this proportion may vary with location. The study is designed so that the partitioning of non-exhaust particles between the road surface and the atmosphere can be determined via the use of molecular tracers. 37

41 7.3. Gravimetric determination of tyre and brake wear rates Although tyre and brake wear rates have been obtained in the past, these have usually been determined in accelerated laboratory-based tests. It appears that there are few measurements of in-service wear. The loss of material from the tyres and brakes of four inservice cars is therefore being monitored by weighing components at regular time intervals. The test cars and owners have been selected to reflect different types of driving conditions. For example, two of the owners drive predominantly on motorways, whereas the other two drive mainly in urban areas. For each vehicle component, the amount of material lost will be determined as a function of the distance driven and the type of driving. Every two months during a 14-month period, the components will be removed, cleaned thoroughly, dried, weighed and replaced (the removal and re-fitting of components will be carried out by qualified mechanics). The vehicle owners have also been asked to complete a brief log book for each journey undertaken. The log book is designed to provide information which may subsequently be used to determine how road type and average speed influence component wear. The test matrix for this part of the work programme is presented in Table12. The first two test phases were completed in February and April of Table 12 Test matrix for gravimetric tests Component Number of test vehicles Number of components per vehicle Number of test phases Total number of measurements Tyres Brake shoes/pads Chemical analysis of vehicle components and wear dust For tyre tread and brake lining material, the chemical composition of unused and used material, as well as wear particles where possible, will be determined in the laboratory. Also where possible, the sample of materials tested will include duplicates of the types of component fitted to the vehicles used in the gravimetric tests. The primary objective of the chemical analysis of components and wear dust will be to identify unique molecular tracer species for each particle source. Particular attention will be paid to the tracer species identified in the review by Warner et al. (ref. 8). For example, the following species have already been identified as potentially useful molecular tracers for tyre wear particles in the environment: 2-(4-Morpholinyl)benzothiazole N-Cyclohexyl-2-benzothiazolamine Styrene-butadiene rubber The n-alkanes C 35 Organic zinc Although reasonably detailed information on the composition of brake linings and brake wear particles has been reported, no strong evidence was found in the literature of reliable tracer species for brake wear particles, although polyalkylene glycol ethers present in brake fluid have been found in relatively high concentrations in brake wear particles. Studies show that a number of metals are present in significant quantities in brake pad material, but most of these are not likely to be unique tracers for brake wear. It has also been noted that the various cycles that take place in an urban environment can seldom be traced with a single species. More commonly, multi-element tracers will be required to characterise the different sources of trace elements and PM. Also, receptor modelling provides a powerful source apportionment tool; multivariate statistical techniques, such as multilinear regression, cluster analysis, 38

42 discriminant analysis, and principal component analysis, can be used to identify common patterns in atmospheric data. The complex and highly variable nature of road dust means that the investigation of resuspension has progressed through the interpretation of physical effects (e.g. examination of size distributions), rather than via the use of chemical tracers. Further information to aid the understanding of resuspension should be generated in the experimental work presented here. Table 13 summarises the proposed analyses. A mixture of unused and used tyre tread and brake lining will be included. If tyre dust can also be obtained at low cost, then such samples may be added to the analysis. Table 13 Chemical analyses of vehicle components Component Samples Species Techniques Tyre rubber 10 Brake pads 10 Brake dust 10 Zn Organics inc. n-alkanes C 35 Zn/Sr/Ti/Ba/Ca/Fe/Cu/Va/Na/Sb PAHs & glycol ethers Zn/Sr/Ti/Ba/Ca/Fe/Cu/Va/Na/Sb PAHs & glycol ethers AAS GC-MS or HPLC AAS/ICP-MS GC-MS or HPLC AAS/ICP-MS GC-MS or HPLC 7.5. Sampling and analysis of airborne particles and road dust Airborne particles and road dust will be collected at a total of 10 sites (5 paired roadside and background locations) in the UK. The sites will probably include existing TRL, UH, or AUN sites, as well as ad hoc locations. At each site the parameters measured will include: Size distribution and chemical composition of airborne and road dust particles. Ambient temperature, wind speed and direction, humidity Traffic volume and composition Traffic speed and acceleration At some sites, information on gaseous pollutants may also be available. A number of ambient monitoring instruments are available for use, including TEOM, GRIMM, MOUDI, Partisol StarNet, and Andersen Impactor. However, the availability and usability of equipment for the non-exhaust work will vary at established sites. Also, because of the different response characteristics of the different sampling instruments, it is considered that the use of a range of instruments would generate results which would be difficult to interpret. Therefore, the same experimental method will, as far as possible, be employed at each sampling location. At this stage, it is anticipated that portable particle samplers (such as the Partisol StarNet gravimetric high-volume system with size selective inlets - PM 10, PM 2.5, PM 1 ) will be deployed at each location for a period of between one and two months. The overall monitoring campaign will be conducted over a 12-month period, which should be sufficient to cover a range of traffic and weather conditions. Road dust will be collected from the road surface by dry suction over a pre-defined area and after a pre-defined period of dry weather. Road dust which is on the verge and at the background site may be collected using deposition gauges. Again, the chemical composition and size distribution of the particulate matter in the road dust complex will be determined, with particular attention being paid to the tracer species. An indication of the size distribution of 39

43 wear particles will be gained by examining the concentrations of tracer species in various size fractions of road dust and airborne particles. This information will subsequently be used in combination with the material loss estimates to derive emission factors for brake and tyre wear. 40

44 8. REFERENCES 1. Bagley, S. T., K. J. Baumgard, L. D. Gratz, J. H. Johnson, and D. G. Leddy Characterization of Fuel and Aftertreatment Device Effects on Diesel Emissions. Health Effects Institute Research Report Number 76, 88 pp. 2. DG TREN PARTICULATES Consortium. Characterisation of exhaust particulate emissions from road vehicles. Deliverable 2, Version 1, April Vehicle exhaust particulates characterisation, properties, instrumentation and sampling requirements 3. EU (1998) Directive 98/69/EC of the European Parliament and of the Council of 13 October 1998 relating to measures to be taken against air pollution by emissions from motor vehicles and amending Council Directive 70/220/EEC. Official Journal of the European Communities No. L 350/1, Jon D. Andersson et al. DETR/CONCAWE/SMMT Particle Research Programme: Sampling and Measurement Experiences. SAE EU (1996) Directive 1999/96/EC of the European Parliament and of the Council of 13 December 1999 on the approximation of the laws of the Member States relating to measures to be taken against the emission of gaseous and particulate pollutants from compression ignition engines for use in vehicles, and amending Council Directive 88/77/EEC. Official Journal of the European Communities No. L044, K. Pattas, N. Kyriakis, Z. Samaras, P. Pistikopoulos and L. Ntziachristos. Dept. of Mechanical Engineering, Aristotle University, Greece. Effect of DPF on Particulate Size Distribution using an Electrical Low Pressure Impactor. SAE K. Pattas, Z. Samaras, N. Kyriakis, P. Pistikopoulos and T. Manikas. Dept. of Mechanical Engineering, Aristotle University, Greece. Investigation of the Operation of a DPF Installed on a Direct Injection Turbo Charged Vehicle with the Use of a Cerium Based Catalytic Additive. SAE Warner L R, Sokhi R, Luhana L and Boulter PG (2001). Non-exhaust particle emissions from road transport: a literature review. TRL Unpublished Report PR/SE/213/2000. Transport Research Laboratory, Crowthorne. 41

45 Appendix 1 ADDITIONAL TRANSIENT LIGHT DUTY CYCLES INRETS cycles by vehicle category (called "INRETS specific" in table 6) These cycles were developed from the same database used for CADC cycles and with the same statistical tools but according to a split of the monitored vehicles into 2 categories of specific power: those having a power/mass ratio under 0.06 kw/kg and those above this limit. The cycle structure (acceleration) and the speed levels are then more adapted to the various tested vehicles. The selected sequences which form the driving cycles are part of 2 different data sets. The "INRETS specific" cycles developed for the urban, road and motorway driving conditions and for 2 types of vehicle specific power. 42

46 INRETS short cycles These 2 cycles were specially designed to assess the influence of cold start on emissions. By the repetition of a well identified sequence, typical of the urban or the road driving behaviour, the evolution of emissions can be followed precisely according to time or mileage and engine temperature. Full details are available from INRETS 43