Investigation on PM Emissions of a Light Duty Diesel Engine with 10% RME and GTL Blends

Hongming Xu Jun ZhangUniversity of Birmingham

Philipp Price Ford Motor Company

International Particle Meeting, Cambridge University 21 May 2010

Presentation Outline

1. Introduction

2. Experimental System

3. Results and Discussion• Effect of 10% RME and GTL blends

• Effect of Injection strategy

4. Conclusions

1. Introduction

Background

Diesel Engine Emissions

ParticlesNOx HC & CO Others

SN/Mass

Human Health & Environment

Number

size

Tier Date CO NOx HC+NOx PM PN*

Euro 4Jan 2005

0.50 0.25 0.30 0.025 -

Euro 5Sept 2009

0.50 0.18 0.23 0.005 6.0 x 1011

Euro 6Jan 2014

0.50 0.08 0.17 0.0045 6.0 x 1011

Development of legislations g/km

• Compared with Euro 4, Euro 5 confines the emissions further for carbon monoxide (CO), hydrocarbons (HC), oxides of nitrogen (NOx) and Particulates Matter (PM) and

the latter two had a 28% and 80% reduction respectively.• Particle number

*#/km

Regulations (EC) No 715/2007 of the European Parliament and the Council, "Emissions-Light Duty Vehicles", Jul, 2009.

Objective of the present study

Emission Reduction

10% Fuel Blends with Diesel

2020 target

Particlecharacteristics

Alternative Fuels(Biodiesel, GTL Diesel, and etc)

Injection Strategy

Pilot Injection

A

B

NOx control

2. Experimental System

Test cell and the Ford Puma engine

Bore 86mm, 4 cylindersStroke 94.6mm

Compression Ratio 16.6

Engine Capacity 2198cc

Max Power 96KW (±5%)@3500rpm

Max Torque310.0NM(±5%)@1600-2500rpm

Injector type Common Rail, Direct Injection

Engine test rig layout

Inter-cooler

Heat Exchanger

Exhaust Measurement

Horiba MEXA 7100 DEGR

TSI SMPS 3936

AVL Smoke meter 415SG002

PROPERTY UNIT Diesel RME GTL DieselEster content % (m/m) / 99.44 /Density @ 15°C kg/m3 834.9 883.3 781Viscosity @ 40°C mm2/s 2.87 4.441 3.1Flash point °C 68.5 171.5 91Sulphur content mg/kg 8.6 <3.0 <3.0Carbon residue % (m/m) 0.13 <0.1 <0.3Cetane number 51.1 51 77Total contamination (particulate)

mg/kg 6.01.6 1.6

Lubricity μm 402 / 612Distillation (Initial Boiling Point)

°C 181.3/ 204

Aromatics %,m / / <0.1

Fuel properties

Engine Test Modes (A)

ModeEngine Speed

(rpm)Torque

(Nm)Load (%)

EGR Valve Opening (%)

1 800 2.1 0.68 0

2 1800 30 9.68 30.91

3 1800 30 9.68 15.45

4 1800 30 9.68 0

5 1800 134 43.23 0

6 3100 35 11.29 18.18

7 3100 138 44.52 0

8 3100 230 74.19 0

Engine Test Modes (B)

ModeEngine Speed

(rpm)Main SOI (BTDC)

BMEP (bar)

EGR

Idle 800 -1 0.68 NO

Middle Speed/Load 1800 -2.69 5.2 YES

High Speed/Load 2500 -2.69 7.0 NO

Pilot Injection0

(mm3/stroke)

1.5mm (mm3/stroke

)3(mm3/stroke)

+5º CA

(a)

(b) (e)

Base (c) (f)-5º CA (d) (g)

3. Test Results

A. With 10% RME and GTL blends

Particulate number and mean diameter

Mode 1Mode 2Mode 3Mode 4Mode 5Mode 6Mode 7Mode 8103

104

105

Tot

al C

once

ntra

tion(

Par

t/cm

3 )

3100rpm1800rpm800rpm

Engine Mode

Diesel RME10 GTL10

RME10/ GTL10 Increase of engine load

Increase of speed

Less EGR

Fewer particles

Fewer particles, larger size

More particles, larger size

Fewer particles, smaller size

Mode 2 Mode 3 Mode 4 Mode 5 Mode 6 Mode 7 Mode 850

60

70

3100rpm1800rpm

Mea

n D

iam

eter

(nm

)Engine Mode

Diesel RME10 GTL10

Smoke

Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6 Mode 7 Mode 80.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

3100rpm1800rpm800rpm

FS

N

Engine Mode

Diesel RME10 GTL10

•The trend of variation of smoke is not quite the same as the particle numbers

Non-volatile particles

Mode 1 Mode 2 Mode 5 Mode 6103

104

105

106

Engine Test Mode

3100 RPM1800 RPM800 RPM

Tot

al C

once

ntra

tion(

Par

t/cm

3 )

Diesel RME10 GTL10

Mode 2 Mode 5 Mode 650

60

70

3100 RPM1800 RPM

Mea

n D

iam

eter

(nm

)

Engine Test Mode

Diesel RME10 GTL10

-39%

-42%

-33%-43%

-53%

-34%

-28%

-32%

-28%-32%

-34%

-29%

• The non-volatiles number reduction by the alternative fuel blends were all higher than the rates when thermo-dilution was not used

10 100102

103

104

D

N/D

LogD

p(P

art./

cm3 )

Diameter (nm)

Diesel RME10 GTL10

Particulate Size Distribution at 800 rpm, Idle

• Bimodal mode• A general reduction of

particles in different sizes

• Small peak at 20nm when RME10 was used

10 100102

103

104

Diameter (nm)

DN

/DLo

gDp(

Par

t./cm

3 )

DieselMode4 DieselMode5 RME10Mode4 RME10Mode5 GTL10Mode4 GTL10Mode5

Particulate Size Distribution at 1800 rpm

• Mono-modal feature• A general reduction of particles

in different sizes• The increase of load leads to

less nucleation mode particles

30NM, No EGR

134NM, No EGR

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

DieselMode6 RME10Mode6 GTL10Mode6

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

DieselMode7 RME10Mode7 GTL10Mode7

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

DieselMode8 RME10Mode8 GTL10Mode8

• Particles numbers reduced by RME10 or GTL10

• Larger particles with the increase of the load

Particulate Size Distribution at 3100rpm

35NM, With EGR 138NM, No EGR

230NM, No EGR

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

DieselMode2 DieselMode3 DieselMode4

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

RME10Mode2 RME10Mode3 RME10Mode4

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

GTL10Mode2 GTL10Mode3 GTL10Mode4

• More EGR, more nucleation particles • some nucleation particles around 10 nm might be reduced

1800 RPM

Particulate Size Distribution with different EGR

10 100102

103

104

Diameter (nm)

Con

cent

ratio

n D

N/D

LogD

p(P

art./

cm3 )

1 2 3 4 5 6

Non-volatiles during Warming-up

(1) large amount of soot from incomplete combustion.

(2)The mutated hydrocarbons during the combustion were likely to quench.

RME10

(a) Diesel magnification of 10000(b) Diesel magnification of 65000 (c) RME 10 magnification of 10000 (d) RME magnification of 65000 (e) GTL10 magnification of 10000 (f) GTL10 magnification of 65000

Particle morphology (1800rpm, 30Nm)

Summary and conclusions (A)• RME10 and GTL10 can lead to a similar reduction in total particle numbers

under various engine conditions but their influences to the particle mean diameters are not clear

• RME10 and GTL10 reduce the accumulation mode particles and some nucleation particles in the larger size range (>30nm); however, RME 10 could also increase those in the small size range under certain cases (<20nm).

• Particles from diesel combustion have more clusters than those from the RME10 or GTL10 and the primary particle size of all the three fuels is around 20-50 nm.

• At 1800rpm, the increase of engine load results in an increase of particle mean diameter and the reduction of particle numbers (differently at 3000rpm); the increase of either engine speed or EGR increases the particle numbers as well as the mean diameters

• Cold starts could result in much higher non-volatile particles in the nucleation mode.

B. Particles influenced by pilot injection

Effect of pilot injection

Idle Middle High0

1x104

2x104

3x104

4x104

5x104

Engine Mode

Tot

al C

once

ntra

tion(

Par

t/cm

3 )

withoutpilot withpilot

Middle High0

20

40

60

80

100

Engine Mode

Mea

n D

iam

eter

(nm

)

withoutpilot withpilot

Particle number reduction at idle and high speed/load mode, increased by EGR at medium load

Idle mode Middle mode High mode

Larger particles using pilot injection in idle/middle mode

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter(nm)

withoutpilot withpilot

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

withoutpilot withpilot

10 100

102

103

104

DN

/DLo

gDp(

Par

t./cm

3 )Diameter(nm)

withoutpilot withpilot

Idle Middle High0

20

40

60

80

100

Engine Mode

Mea

n D

iam

eter

(nm

)

Base+5BTDC Base Base-5BTDC

Effects of pilot injection timing

• The advance of the pilot injection leads to a reduction of the particle numbers and mean diameters

Idle Middle High0

1x104

2x104

3x104

4x104

5x104

Engine Mode

Tot

al C

once

ntra

tion(

Par

t/cm

3 )

Base+5BTDC Base Base-5BTDC

Idle Middle High0

1x104

2x104

3x104

4x104

5x104

Tot

al C

once

ntra

tion(

Par

t/cm

3 )

Engine Mode

pilot 1.5mm 3/str pilot 3mm 3/str

Middle High0

20

40

60

80

100

Engine Mode

Mea

n D

iam

eter

(nm

)

pilot 1.5mm 3/str pilot 3mm 3/str

Effect of pilot injection quantity

Idle mode middle mode high mode

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

12,1.5mm3/str 12,3mm3/str

17,1.5mm3/str 17,3mm3/str

10 100102

103

104

105

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

18,1.5mm3/str

18,3mm3/str 23,1.5mm3/str 23,3mm3/str

10 100

102

103

104

DN

/DLo

gDp(

Par

t./cm

3 )

Diameter (nm)

21,1.5mm3/str 21,3mm3/str 26,1.5mm3/str

26,3mm3/str

SOI (BTDC) SOI (BTDC)SOI (BTDC)

• In the absence of EGR, pilot injection seems to help reduce particle numbers, as in the idle and high speeds. When EGR is applied, as in medium speed/load, the introduction and increase of pilot injection quantities increases both the particle number and mean diameter.

• The advanced timing of a higher radio of pilot injection tends to reduce the number and diameter of particles.

• The strategy of pilot injection influences the PM emissions from the pilot combustion and at the same time the main combustion through ignition delay. This effect is less significant with the increase of the engine load and with the advance of the pilot injection timing.

• It is expected that when the strategy of pilot injection is used for NOx and NVH reduction as for biodiesel, attention will be required to minimise its impact on PM emissions.

Summary and conclusions (B)

Acknowledgement

The authors gratefully acknowledge research funding from EPSRC under the grant EP/F061692/1 and TSB under the grant M0597H, industrial support from Jaguar Land Rover, Ford Motor Company and Shell Global Solutions, and contributions to the related research from the Birmingham University, Oxford University including colleagues and students who have worked with us.

Thank you!