Ethanol Effects on Lean-Burn and Stoichiometric GDI Emissions

John M.E. Storey, Teresa L. Barone, John F. Thomas, Shean P. Huff, and Samuel A. Lewis Oak Ridge National Laboratory

2011 Directions In Engine Efficiency and Emissions Research Conference

October 3-6, 2011

DOE Sponsor – Vehicle Technologies

James Eberhardt: Health Impacts

Kevin Stork & Steve Przesmitzki: Fuels

Gurpeet Singh & Ken Howden: Catalysts

2 Managed by UT-Battelle for the Department of Energy Data Source: Heyder (2004)

Nanoparticles can penetrate into alveolar region

Particle Size Influences Location of Deposition in Respiratory System

3 Managed by UT-Battelle for the Department of Energy

Comparison of Stoichiometric and Lean GDI Emissions

Stoichiometric GDI Vehicle Pontiac Solstice

“wall-guided”

Lean GDI Vehicle BMW 120i – Euro spec

“spray guided” • Motivation: 1. Understand potential fuel effects on GDI PM emissions. 2. Health implications and ambient air quality issues •Approach •Test cycles: FTP and US06, transient accelerations plus steady state • Fuels: Gasoline and intermediate ethanol blends (E0, E10, E20) • Measurements:

- Particle mass: collection on Teflon-coated quartz-fiber filters and gravimetric analysis -- Particle composition: organic carbon/elemental carbon (OC/EC) - Particle number concentration and size distributions: analysis by SMPS

Particle Emissions Analysis

Catalyst

Mass

Morphology

EEPS

Steady-State

Number-Size Distributions

Transient

Transmission Electron Microscopy;TEM-Sampler

AVL Micro Soot Sensor

Mettler Microbalance

Organic Fraction

Organic to Elemental

Carbon Ratio

Dilution Systems

Full-flow CVS

Partial Flow BG-3

Microtunnel

European PMP

Microwave Reactor Extraction Organic Speciation

Scanning Mobility Particle Sizer

5 Managed by UT-Battelle for the Department of Energy 0

2

4

6

8

10

PM e

mis

sion

s (m

g/m

ile)

E0 E10 E20

FTP

US06

Stoich GDI Lean GDI Stoich GDI Lean GDI

Stoichiometric GDI Vehicle PM emissions more sensitive to fuel and cycle than lean GDI

Tier 2 LDDV PM limit

6 Managed by UT-Battelle for the Department of Energy

Ethanol leads to lower elemental carbon concentrations in the tailpipe

• Less EC implies less soot formation for E10, E20

- Catalyst temperature too low to oxidize soot

• OC higher for lean burn

• Survives the catalyst

• OC/EC ratios 0.2 – 0.7

• Diesel OC/EC > 5 typically

30 mph

Stoichiometric Lean

EC

Con

cent

ratio

n (u

g/m

3 )

0

200

400

600

800

1000 E0E10E20

30 mph

Stoichiometric Lean

OC

Con

cent

ratio

n (u

g/m

3 )

0

200

400

600

800

1000 E0E10E20

7 Managed by UT-Battelle for the Department of Energy 7

Results – Particle size and number

• Used scanning mobility particle sizer (SMPS ) ~ 1 -2 minutes for size distributions – steady state only

• Particle counter (CPC3025) can be used independently – particle concentration too high! Required 1000:1 dilution

• Investigated three separate sizes for accels: 10, 50, 100 nm

SAE 2010-01-2129

Size distributions consistent at both steady-state points (stoich GDI)

Ethanol content does not change general shape. E20 reduces total number concentration by about 50% at 30 mph and 70% at 80 mph

Geometric mean diameter for all distributions ~ 50 nm Distribution broader than diesel exhaust particles

30 mphPost-catalystDilution Ratio ~21

dp (nm)1 10 100 1000

dN/d

logd

p (#

/cm

3 )

0

2e+5

4e+5

6e+5

8e+5

1e+6E0E10 E20

dp (nm)1 10 100 100

dN/d

logd

p (#

/cm

3 )

0

1e+6

2e+6

3e+6E0E10 E20

80 mphPost-catalystDilution Ratio ~21

10 100

dN/d

logd

p (#

/cm

3 )

0

1e+6

2e+6

3e+6StoichLean GDI Vehicle

Particle Size Distribution for 80 mph (E0 fuel)

Stoich GDI particles biased larger but less abundant than lean particles

Geometric mean ~ 50 nm

dp (nm)10 100

dN/d

logd

p (#

/cm

3 )

0

2e+7

4e+7

6e+7

8e+7LeanStoich

Geometric Mean ~ 40 nm

23 nm European PMP (Particulate Measurement Protocol) counts particles above 23 nm

10 Managed by UT-Battelle for the Department of Energy

Lean GDI total number emissions also higher than stoich GDI – comparable to potential regulation

Cold LA04 Hot LA04 US06Tota

l Par

ticle

Num

ber E

mis

sion

s (#

/mi)

1e+12

1e+13

1e+14Lean DISIStoich DISI

Proposed

CARB

Regulation

E0, Catalyst OutTotal Particle

11 Managed by UT-Battelle for the Department of Energy

Particles per mile decrease with ethanol content and hot cycles

E0 E2050 n

m P

artic

le N

umbe

r Em

issi

ons

(#/m

i)

0

2e+10

4e+10

6e+10

8e+10Stoich DISICold LA04Engine Out

50 nm

Cold LA04 Hot LA04

0.0

5.0e+9

1.0e+10

1.5e+10

2.0e+10

2.5e+10Stoich DISIEngine Out

E2050 nm

• All fuels show majority of particles from cold start • Ethanol reduces overall particle count

12 Managed by UT-Battelle for the Department of Energy

Reduction by E-blends E10 & E20 Similar for lean DI vehicle

Cold LA04 Hot LA04 US06

Par

ticle

Num

ber E

mis

sion

s (#

/mi)

3.0e+12

6.0e+12

9.0e+12

1.2e+13

1.5e+13

1.8e+13E0E10E20

ProposedCARBRegulation

EuropeanRegulation

8.1 e+11

Lean DISICatalyst OutTotal Particle

< 23 nm particles included in our measurements

13 Managed by UT-Battelle for the Department of Energy

Time (sec)0 100 200 300 400 500 600

Num

ber C

once

ntra

tion

(par

t./cm

3 )

0.0

2.0e+4

4.0e+4

6.0e+4

8.0e+4

1.0e+510 nm50 nm100 nm

Stoich GDI Wide Open

Throttle 0 – 80 mph

Time (sec)0 200 400 600

Num

ber C

once

ntra

tion

(par

t./cm

3 )

0

2e+4

4e+4

6e+4

8e+4

1e+510 nm50 nm100 nm

E0 Post-catalyst

E20 Post-catalyst

0

20

40

60

80

100

Spe

ed

MP

H

Speed Trace

14 Managed by UT-Battelle for the Department of Energy

Time (sec)0 100 200 300 400 500 600Pa

rtic

le N

umbe

r Con

cent

ratio

n (#

/cm

3 )

0

10000

20000

30000

40000

5000010 nm50 nm100 nm

Time (sec)0 100 200 300 400 500 600Pa

rtic

le N

umbe

r Con

cent

ratio

n (#

/cm

3 )

0

10000

20000

30000

40000

5000010 nm50 nm100 nm

E20

E0 Lean GDI vehicle

Wide Open Throttle

0 – 80 mph

0

20

40

60

80

100

Spe

ed

MP

H

Speed Trace

Distinct spikes disappear for E20; not much reduction in 100 nm particles

Conclusions: Fuel Effects for the stoich GDI Vehicle • Use of E20 resulted in a 40 to 60% reduction in PM mass emissions

• >50% reduction in total particle number concentration for 30 and 80 mph; no change in size distribution

• Reduction of 50 and 100 nm particle emissions during acceleration

• Cold start significant contributor to overall PM number

Fuel Effects for the lean GDI Vehicle • E20 resulted in 30-40% drop in cycle-based PM mass emissions

• Lean vehicle had smaller mean size, larger number of particles

• Ethanol slightly reduced number based emissions

• Implies fewer larger particles

• Less difference between cold start and warm cycles.

Implications for GDI PM control

• Need for exhaust particulate filter (GPF) depends on success of in-cylinder control – Cold start worse for wall-guided injectors – Oil entrainment? – Avoidance of sooting conditions

• Particulate filter issues for GDI – High EC content implies refractory particles – High temperatures will help – Ash content would have to be addressed

17 Managed by UT-Battelle for the Department of Energy

Extra slides

Contact Information

John Storey storeyjm@ornl.gov 865-946-1232 Teresa Barone baronetl@ornl.gov 865-946-1338

19 Managed by UT-Battelle for the Department of Energy

Stoich Wide Open Throttle 0 – 80 mph (E20 detail)

Time (sec)0 200 400 600

Num

ber C

once

ntra

tion

(par

t./cm

3 )

0

2000

4000

6000

8000

10000

12000

14000 10 nm50 nm100 nm

E20 Post-catalyst

Stoich vehicle moderate acceleration - 0 to 50 mph

E0 Post-catalyst

E20 Post-catalyst

• Particles may be generated in in-cylinder fuel rich zones

• Ethanol may reduces amount of carbon available for soot formation through CO formation pathway (Wu et. al, 2006)

Time (sec)0 100 200 300 400 500 600

Num

ber C

once

ntra

tion

(par

t./cm

3 )

0

2e+4

4e+4

6e+4

8e+4

1e+510 nm50 nm100 nm

Time (sec)0 100 200 300 400 500 600

Num

ber C

once

ntra

tion

(par

t./cm

3 )

0

2e+4

4e+4

6e+4

8e+4

1e+510 nm50 nm100 nm

E0 Post-catalyst

E20 Post-catalyst

0

20

40

60

80

100

Sp

ee

d (

MP

H)

Speed Trace As for the WOT accelerations, 50 and 100 nm particle emissions reduced by E20

(FTP acceleration)

21 Managed by UT-Battelle for the Department of Energy

The cycling behavior of the lean GDI vehicle results in a variety of PM size distributions

dp (nm)10 100

dN/d

logd

p (#

/cm

3 )

0.0

2.0e+6

4.0e+6

6.0e+6

8.0e+6

1.0e+7

1.2e+7

1.4e+7LeanSlightly RichStoich

30 mph