Particle Size and Number Emissions

from Dual-Fuel Reactivity Controlled

Compression Ignition

Christopher Kolodziej, Jesús Benajes

CMT – Motores Térmicos

Universitat Politècnica de València

Martin Wissink, Reed Hanson,

Derek Splitter, and Rolf D. Reitz

Engine Research Center

University of Wisconsin

Cambridge Particle Meeting

18 May, 2012

University of Cambridge, UK

2

Outline

• Introduction

• Effects of Gasoline SOI

• Effects of Gasoline/Diesel Proportion

• Conclusions

3

What is Dual-Fuel RCCI?

• Premixed Combustion– Very Low PM and NOx emissions through low local

equivalence ratios during combustion

• In-Cylinder Blending of Higher and Lower Reactivity Fuels– Increased control of ignition (longer premixing possible)

– Stratification of fuel reactivity and local equivalence ratio for lower rate of combustion

– Greatly reduced EGR dependence compared to premixed diesel LTC concepts

• Increased Brake Thermal Efficiency– Through reduced heat transfer losses and optimized

combustion phasing

– Decreased specific fuel consumption

– Decreased specific CO2 emissions

4

Objectives

• Measure particle size and number emissions

from heavy-duty dual-fuel RCCI with both

fuels injected in-cylinder

• Study effects of in-cylinder gasoline injection

timing on exhaust particle emissions (while

fixing in-cylinder diesel injection timings)

• Investigate effects of gasoline/diesel

proportioning on exhaust particle emissions

5

Conv. Diesel Particle Size Distribution

dN/ d log Dp

0 10 30 100

Log D

Typical Distribution Engine Exhaust GasTypical Distribution Engine Exhaust Gas

p

10 nm

30 nm

10 nm

100 nm

Distribution ofDistribution of

CCoreore Distribution ofDistribution of

SSolid olid PParticlesarticles

22

Distribution of Distribution of VVolatile olatile

FFraction (Hraction (H O vapor + HCO vapor + HC））））））））22

Distribution of Distribution of VVolatile olatile

FFraction (Hraction (H O vapor + HCO vapor + HC））））））））

Kittelson Model

“The nuclei mode typically contains 1-20% of the particle mass and more than 90% of the particle number.”

Kittelson, J. Aerosol Sci., 1998

Kawai Model

“Hypothetical model for Diesel nano-particle distribution.”

Montajir, Kawai, Goto, Odaka, SAE 2005-01-0187

6

Premixed Diesel LTC PM - SOI

SAE 2010-01-1121

(Benajes, Novella, Arthozoul, Kolodziej)

• Mono-modal shaped size

distribution

• Decreased number of

larger particles was

accompanied by

increased number of

smaller particles1.E+04

1.E+05

1.E+06

1.E+07

1.E+08

1.E+09

1 10 100 1000Diameter [nm]

Concentration [#/cc]

-24-21-18-15-12-9

-9

-12

-18

-21

-15

-24

1.E+04

1.E+05

1.E+06

1.E+07

1.E+08

1.E+09

1 10 100 1000Diameter [nm]

Concentration [#/cc]

-24-21-18-15-12-9

-9

-12

-18

-21

-15

-24

Pivot Point

1.35 bar Pint

7

Premixed Diesel LTC PM - Impingement

• Liquid fuel impingement

during earlier injections

• Initially only increased

number of larger particles

• Bi-modal size distribution

when number of smaller

particles increased as well

(by 2 orders of magnitude)

-25

-20

-15

-10

-5

0

5

10

15

0 10 20 30 40 50 60

Radial Position [mm]

Injector Vertical Position [mm]

Piston-33-30-27-24

Advanced Injection

Benajes, Garcia-Oliver, Novella, Kolodziej, Fuel, 2011

1.E+04

1.E+05

1.E+06

1.E+07

1.E+08

1.E+09

1 10 100 1000Diameter [nm]

Concentration [#/cc]

-33-30-27-24

-33

-30

-27

-24

1.E+04

1.E+05

1.E+06

1.E+07

1.E+08

1.E+09

1 10 100 1000Diameter [nm]

Concentration [#/cc]

-33-30-27-24

-33

-30

-27

-24

1.35 bar Pint

8

Premixed Diesel LTC PM – Intake O2

• Decreased intake O2 caused a general increase in mono-modal size distributions, though PM mass was similar

• Much fewer particles larger than 100 nm compared to conventional diesel combustion (higher Pint than previous slide)

• Diameter of peak number concentration smaller than 60 nm

SAE 2011-01-1355

(Payri, Benajes, Novella, Kolodziej)

10 100Diameter [nm]

105

106

107

108

109

dN/dlogDp [#/cm

3]

9% O2

10% O2

11% O2

12% O2

13% O2

Intake O2 Conc.

13 to 9%vol

1.6 bar Pint

9

Test Methodology

1.151.131.11.09Intake Pressure [bar]

0

27

35

65

0

37

26

74

0

47

20

80 84Gasoline Proportion [%]

16Diesel Proportion [%]

57Intake Temperature [°aTDC]

0EGR Rate [%]

-38Second Diesel In-Cylinder

Injection Timing [°aTDC]

-58First Diesel In-Cylinder

Injection Timing [°aTDC]

-360, -340, -320, -300Gasoline In-Cylinder

Injection Timing [°aTDC]

• Sweep gasoline injection timing from -300 to -360 °aTDC at each fuel reactivity test condition

• Vary intake temperature to maintain constant CA50

• Vary intake pressure to maintain overall equivalence ratio

10

RCCI Exhaust Dilution Conditions

1.E+05

1.E+06

1.E+07

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

TDR 26

TDR 38

TDR 52.5

TDR 105

TDR 142

• Varied heated

primary dilution air

ratio

• Fixed ambient-

temperature

secondary dilution

ratio at maximum of

Dekati FPS-4000

diluter

• Total dilution ratio

(TDR) of 130-135:1

was used for testing

11

Outline

• Introduction

• Effects of Gasoline SOI

• Effects of Gasoline/Diesel Proportion

• Conclusions

12

Effects of Gasoline SOI

0.00

0.05

0.10

0.15

0.20

0.25

-370 -360 -350 -340 -330 -320 -310 -300 -290GDI SOI [°aTDC]

PM Opacity Measurement [FSN]

65%/35%

74%/26%

80%/20%

84%/16%

1.E+05

2.E+06

4.E+06

6.E+06

8.E+06

1.E+07

1.E+07

-370 -360 -350 -340 -330 -320 -310 -300 -290GDI SOI [°aTDC]

dN/dlogDp [#/cc]

65%/35%

74%/26%

80%/20%

84%/16%

• Slight decrease in total

particle numbers from -360

to -300°aTDC

• Two-fold higher particle

number emissions from

lowest gasoline proportion

• Advanced gasoline

injection timing caused

slight PM mass increase

from -300 to -340°aTDC

• More noticeable increase

from -340 to -360°aTDC

13

Effects of Gasoline SOI

1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

-360°aTDC

-340°aTDC

-320°aTDC

-300°aTDC

65% Gas : 35% Diesel1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

-360°aTDC

-340°aTDC

-320°aTDC

-300°aTDC

74% Gas : 26% Diesel

1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

-360°aTDC

-340°aTDC

-320°aTDC

-300°aTDC

80% Gas : 20% Diesel1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

-360°aTDC

-340°aTDC

-320°aTDC

-300°aTDC

84% Gas : 16% Diesel

14

Conclusions (Gasoline SOI)

• Advancing gasoline SOI increased PM mass and number emissions for all gasoline cases, especially from -340 to -360°aTDC

• Change from -340 to -360°aTDC was characterized by a sharp increase in accumulation mode and simultaneous decrease in nucleation mode (similar to diesel LTC SOI study)

• Decreased number of larger particles with increased number of smaller particles produced less change in total particle numbers than in PM mass

• Further work is needed to determine if decreased number of larger particles (and PM mass) was due to decreased formation or increased oxidation effects

15

Outline

• Introduction

• Effects of Gasoline SOI

• Effects of Gasoline/Diesel Proportion

• Conclusions

16

Effects of Gasoline Proportion

• Increased gasoline proportion, decreased PM mass

• Gasoline -360°aTDC SOI had higher PM for all fuel blends

• Increased gasoline proportion also decreased total particle numbers

0.00

0.05

0.10

0.15

0.20

0.25

60 65 70 75 80 85 90Fuel Composition [% Gasoline]

PM Opacity Measurement [FSN]

-360°aTDC

-340°aTDC

-320°aTDC

-300°aTDC

0.E+00

2.E+06

4.E+06

6.E+06

8.E+06

1.E+07

1.E+07

60 65 70 75 80 85 90Fuel Composition [% Gasoline]

dN/dlogDp [#/cc]

-360°aTDC

-340°aTDC

-320°aTDC

-300°aTDC

17

Effects of Gasoline SOI

1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

65%/35%

74%/26%

80%/20%

84%/16%

-300°aTDC1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

65%/35%

74%/26%

80%/20%

84%/16%

-320°aTDC

1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

65%/35%

74%/26%

80%/20%

84%/16%

-340°aTDC1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

65%/35%

74%/26%

80%/20%

84%/16%

-360°aTDC

18

Conclusions (Fuels)

• Increasing gasoline proportion caused a simultaneous

decrease in numbers of larger and smaller sized

particles (similar to diesel LTC intake O2 study)

• Further work needed to understand driver of the

simultaneous decrease in both the smaller and larger

sized particles with increased gasoline proportion

19

Outline

• Introduction

• Effects of Gasoline SOI

• Effects of Gasoline/Diesel Proportion

• Conclusions

20

Conclusions (General)

• PM mass emissions were reduced below the Smokemeter minimum detection limit (<0.05FSN) in lowest engine operating conditions

• Particle number emissions were reduced to a similar order of magnitude as the “best” premixed diesel LTC cases (12-13% Intake O2 with 1.6 bar Pint)

• Although RCCI is a form of premixed combustion, its particle size distributions were bi-modal (unlike the mono-modal size distributions typical of premixed diesel LTC)

1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000Mobility Diameter [nm]

dN/dlogDp [#/cc]

65%/35%

74%/26%

80%/20%

84%/16%

10 100Diameter [nm]

105

106

107

108

109

dN/dlogDp [#/cm

3]

9% O2

10% O2

11% O2

12% O2

13% O2

Intake O2 Conc.

13 to 9%vol

1.6 bar Pint

RCCI

LTC

21

Thank you for your attention.

Questions?

Christopher Kolodziej

chko@mot.upv.es

ckolodziej@hotmail.com