ORNL is managed by UT-Battelle for the US Department of Energy

Characterization of GDI PM during start-stop operation with alcohol fuel blends

Presenter: John Storey Melanie DeBusk, Shean Huff, Sam Lewis, Faustine Li, John Thomas, and Mary Eibl

Oak Ridge National Laboratory

This presentation does not contain any proprietary, confidential, or otherwise restricted information.

Effects of Fuel Composition on PM Health Effects Institute Workshop

Chicago, IL December 8, 2016

Kevin Stork and Michael Weismiller Fuel and Lubricant Technologies Vehicle Technologies Office, DOE

2

GDI Vehicle PM Emissions: Impact of cold start, fuels

Will start-stop technology impact GDI PM emissions? Does bio-fuel impact PM? • 2010, 2012: Observed high PM during cold start, ethanol reduced PM • 2014: Detailed HC speciation showed changes in PAHs on PM

(SAE 2010-01-2129; 2012-01-0437; 2014-01-1606)

0.0E+00

5.0E+06

1.0E+07

1.5E+07

2.0E+07

22 26 30 34 38

Resp

onse

, nor

mal

ized

Time (min)

E0E30IB48

1,23

4

5

66

6

7, 8 9

11

10

12

3

GDI Vehicle PM Emissions: Impact of cold start, fuels

Will start-stop technology impact GDI PM emissions? Does bio-fuel impact PM? • 2010, 2012: Observed high PM during cold start, ethanol reduced PM • 2014: Detailed HC speciation showed changes in PAHs on PM

(SAE 2010-01-2129; 2012-01-0437; 2014-01-1606)

• 2014-2015: Obtained and evaluated 2014 Malibu e-Assist vehicle • Bio-fuel may impact both fuel and lube contribution to PM • Focus on Start-Stop effect on PM mass, soot and number

– Tier 3 regulations will lower PM mass standard – PM soot ≈ black carbon, a potent contributor to climate change – Particle number is regulated in Europe currently

4

GDI Vehicle on ORNL’s Chassis Dynamometer

Microsoot Sensor

EEPS ORNL Dilution system

• 1 cold, 8 hot LA4’s each day

• LA4 = FTP Bags 1+2

• Start-stop begins on Hill 2

• PN, PM size with EEPS

• PM Soot with AVL Microsoot

• Zefluor filters for mass, chem

• 3 Cold Bag 1’s/filter X 2 filters

• 9 or 27 Bag 2 runs/filter 0

1

2

3

4

5

Start-Stop No Start-Stop

PM m

ass e

mis

sion

s (m

g/m

ile) E0

E20IBu12

Tier 3 PM limit is 3 mg/mile

Fuels splash- blended • E0 = EEE Tier 2 cert • E20 = EEE + 20% EtOH • iBu12 = EEE + 12% i-BuOH

5

Cold start dominates mass for all three fuels - Filter Mass Measurements

• FTP Composite: weighted average of cold and hot

• Start-stop only increases hot cycle PM for isobutanol

0

2

4

6

8

10

E0 E20 IBu12

PM m

ass e

mis

sion

s (m

g/m

ile

No Start-Stop

Start-Stop

0

2

4

6

8

10

E0 E20 IBu12

PM m

ass e

mis

sion

s (m

g/m

ile

No Start-Stop

Start-Stop

Cold Start Hot start

6

Cold start dominates mass for all three fuels - Filter Mass Measurements

• FTP Composite: weighted average of cold and hot

• Start-stop only increases hot cycle PM for isobutanol

0

0.2

0.4

0.6

0.8

1

E0 E20 IBu12PM

mas

s em

issi

ons

(mg/

mile

No Start-Stop

Start-Stop

0

2

4

6

8

10

E0 E20 IBu12

PM m

ass e

mis

sion

s (m

g/m

ile

No Start-Stop

Start-Stop

Cold Start Hot start

7

Soot emissions show similar trends to PM mass - Micro-Soot Sensor Measurements

• Soot emissions taken second-by-second

• Integrated over cycle and with exhaust flow to get mg/mile

• Wide variability in Hot, despite up to 27 runs

Cold Start Hot start

0

2

4

6

8

10

E0 E30 iBu24

Soot

(mg

/ mi)

No Start StopStop Start

E0 E20 iBu12 0.0

2.0

4.0

6.0

8.0

10.0

E0 E30 iBu24So

ot (m

g / m

i)

No Start StopStop Start

E0 E20 iBu12

8

Soot emissions show similar trends to PM mass - Micro-Soot Sensor Measurements

• Soot emissions taken second-by-second

• Integrated over cycle and with exhaust flow to get mg/mile

• Wide variability in Hot, despite up to 27 runs

Cold Start Hot start

0

2

4

6

8

10

E0 E30 iBu24

Soot

(mg

/ mi)

No Start StopStop Start

E0 E20 iBu12 0.0

0.2

0.4

0.6

0.8

1.0

E0 E30 iBu24So

ot (m

g / m

i)

No Start StopStop Start

E0 E20 iBu12

9

0.0

0.5

1.0

1.5

2.0

2.5

E0 E30 iBu24

Num

ber (

x1014

/ m

i) No Start StopStart Stop

E0 E20 iBu12

Particle number emissions trend lower for Start-Stop - Engine Exhaust Particle Sizer Measurements

• EEPS Total Particle Number includes PM< 23 nm

• (#/cc) taken second-by-second (DR ~ 100)

• Integrated over cycle and with exhaust flow to get #/mile

Cold Start Hot start

0

0.5

1

1.5

2

2.5

E0 E30 iBu24

Num

ber (

x1014

/ mi) No Start Stop

Start Stop

E0 E20 iBu12

10

0.0

0.1

0.2

0.3

0.4

0.5

E0 E30 iBu24

Num

ber (

x1014

/ m

i) No Start StopStart Stop

E0 E20 iBu12

Particle number emissions trend lower for Start-Stop - Engine Exhaust Particle Sizer Measurements

• EEPS Total Particle Number includes PM< 23 nm

• (#/cc) taken second-by-second (DR ~ 100)

• Integrated over cycle and with exhaust flow to get #/mile

Cold Start Hot start

0

0.5

1

1.5

2

2.5

E0 E30 iBu24

Num

ber (

x1014

/ mi) No Start Stop

Start Stop

E0 E20 iBu12

ANOVA (Analysis Of Variance)

Soot F p

Fuel (E0, E20, IB12) 5.19 0.0072

Mode (SS, no SS) 19.18 0 Fuel * Mode 14.54 0

Particle Number F p

Fuel (E0, E20, IB12) 1.31 0.273

Mode (SS, no SS) 1.78 0.1837 Fuel * Mode 56.86 0

o Null hypothesis: there is no

difference between fuels or start-stop modes o p< 0.05 means you reject the null

hypothesis o p< 0.05 is statistically significant

o For soot production, Fuel, Mode,

and their interaction produced a significant difference in soot.

o For particle number, Fuel and Mode did not produce a significant effect. But their interaction did.

12

Time (s) Size (nm)

~60

10 100 20 80

Rat

e (#

/min

) EEPS shows variability for same time intervals

• Variability between hot cycles observed (5 shown above)

• Wide bands, even with 9 cycles

• Data analysis ongoing to look at specific transients

13

E0 E20 iBu12

Par

ticle

Dia

m (n

m)

Spe

ed (m

ph)

Rate (1013 particles/min)

EEPS maps relate size, number to soot production

14

Chemistry of GDI PM HCs

15

Collection and direct thermal desorption/pyrolysis of soot

• GDI PM collected from filter – Light suction on glass capillary – ~0.5 mg needed (70 or 90 mm)

• Transfer to pre-cleaned thimble • TDP-GC-MS (2 chromatograms)

– 1st Step Desorption to 325 °C – 2nd Step Pyrolysis direct to 500 °C

15

16

Does bio-fuel impact PM HCs? • 2014: Collected soot under rich conditions with 3 fuels, E0, E30, iBu48 • Detailed HC speciation showed changes in PAHs on PM

(SAE 2014-01-1606)

GDI Vehicle PM Emissions: Impact of cold start, fuels

0.0E+00

5.0E+06

1.0E+07

1.5E+07

2.0E+07

22 26 30 34 38

Resp

onse

, nor

mal

ized

Time (min)

E0E30IB48

1,23

4

5

66

6

7, 8 9

11

10

12

0.0E+00

5.0E+06

1.0E+07

1.5E+07

2.0E+07

2.5E+07 E0E30IB48

17

Injection matters: Differences in adsorbed HCs apparent for two different platforms

0

2000000

4000000

6000000

8000000

10000000

12000000

14000000

16000000

10 15 20 25 30 35 40

Resp

onse

nor

mal

ized

to 1

mg

of so

ot

Time (min)

Cell 7 engineCell 2 engine

PAHs So What? Both engines: 2 L, GDI, λ=0.9 Difference: - Side vs. Center

injectors - Different intake

geometry

Side GDI

Center mount GDI

SAE 2012-01-0437

18

Start-Stop Study: E20 fuel has lowest measured PM and PAH • FTP cold-hot weighted mass data for start-stop • Ethanol appears to reduce PAH formation in the soot

0

1

2

3

4

5

Start-Stop No Start-Stop

PM m

ass e

miss

ions

(mg/

mile

) E0E20IBu12

Tier 3 PM limit is 3 mg/mile

0

200000

400000

600000

800000

1000000

1200000

3-ringPAH

4-ringPAH

4-ringPAH*

5-ringPAH

6-ringPAH

Rela

tive

Resp

onse

E0 E30E20 E0

19

Start-stop study: GC-MS didn’t detect lube HCs on filter

0.0E+00

1.0E+06

2.0E+06

3.0E+06

4.0E+06

5.0E+06

5 10 15 20 25 30 35 40 45

Amou

nt (a

rbitr

ary

units

)

Time (min)

Hot startCold start

Lube HCs appear here as large peaks if present

20

Summary: GDI vehicle PM depends on fuel and mode

Lowest value: No Start-Stop Start-Stop PM Mass E0 ≈ E20 E20 Soot Mass iBu12 E0 Particle # iBu12 E0

• Largest particles contribute to soot emissions in first 300 s • PAHs affected by alcohols • Injection technology has improved both mass and chemistry • “Uncontrolled” burns may be associated with PAH • How about lubricant contribution?

• Not conclusive, no lubricant found in PM organic fraction by GC-MS • Why is this research important?

• Start-up has highest PM emissions for GDI • Start-stop could impact particulate filter operation if GPF needed in 2025

• Takeaway: Operation and fuel both have to be considered for PM control strategies

• Lowest Cold Start PM mass, soot and number = E20 • Hot start PM affected differently

21

Acknowledgements

• Kevin Stork and Michael Weismiller - DOE Vehicle Technologies Office, Fuel and Lubricant Technologies Program

• “This research was supported in part by an appointment to the Higher Education Research Experiences Program at Oak Ridge National Laboratory” (Faustine Li, Mary Eibl)

• Much helpful discussion with Matti Maricq, Imad Khalek, Dave Kittelson, Alla Zelenyuk

• Contact: John Storey storeyjm@ornl.gov 865-946-1232