burning modes and oxidation rates of soot: relevance to ... · burning modes and oxidation rates of...

www.usra.edu

Burning Modes and Oxidation Rates of Soot:Relevance to Diesel Particulate Traps

Randy L. Vander Wal(*)(1), Aleksey Yezerets(2), Krishna Kamasamudram(2), Neal W. Currier(2), Do Heui Kim(3),Chong Min Wang(3)

(1) - USRA at the NASA-Glenn Research Center Cleveland, OH44135, USA

(2) - Cummins Inc., Columbus, IN, 47201, USA (3) - Pacific Northwest National Laboratory, Richland, WA, USA

DEER 2007 Conference, Detroit, MI Aug. 13th - 16th

Acknowledgements:Support through contract IND61957 with Cummins Inc.Soot samples supplied by Cummins Inc. and Cabot Corp.

Outline

I. Current Oxidation Status

II. Alternative Burning Modes via HRTEM

III. Nanostructure Quantification via Image Analyses

IV. Current Efforts

V. Conclusions

ic

What is Diesel Particulate Matter ?

• Composition:–“Dry carbon”

• turbostratic graphite • initial evidence of fulleren

structures in some cases –Adsorbed HCs–Inorganic materials

• Lube oil ash, H2SO4, HNO3, H2O

• Nanostructure - Amorphous, fullerenic, graphitic

• Morphology: –Primary particles: ~20-40 nm –Agglomerates: 0.1-1 micron

DieselNet

Dieselnet.com

Reactivity evolution over a life cycle of a soot particle

Printex

A: High reactivity due toambient agingB: Steady-state oxidationC: Steep increase

Diesel Soot

A: High reactivity due to: – Adsorbed HC; ambient aging

0.00E+00

2.00E-04

4.00E-04

6.00E-04

8.00E-04

1.00E-03

1.20E-03

0 20 40 60 80 100

soot oxidized, %

Norm

aliz

ed r

ate

[m

in-1

] Printex,380°C,10%O2

0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

0.010

0 10 20 30 40 50 60 70 80 90 100

soot consumed, %

No

rmal

ized

rat

e, [m

in -1

] Diesel soot 450°C,10%O2

A B

C

B

C A

B: “Steady-state” oxidation C: Increased reactivity at later stages of oxidation

r = A · exp(-Ea/RT) · [C]a · [O2] b · [H2O]c

Progressive oxidation

Reactivity is increasing with degree of oxidation:

-9

-8

-7

-6

-5

-4

-3

-2

1.2 1.25 1.3 1.35 1.4 1.45 1.5 1.55

1000/T (K-1)

Ln

(K,

min

-1)

8.4-19.4%19.4-28.7%28.7-34.6%34.6-42.9%43.3-49.6%49.6-56.5%67.4-72.8%72.8-78.1%88.2-92.2%Linear (8.4-19.4%)Linear (19.4-28.7%)Linear (43.3-49.6%)Linear (28.7-34.6%)Linear (34.6-42.9%)Linear (49.6-56.5%)Linear (67.4-72.8%)Linear (72.8-78.1%)Linear (88.2-92.2%)

IncreasingOxidation

– No measurable changes of Ea, or reaction order in O2* Reaction chemistry appears to be independent of the degree of

carbon oxidation.* Number (density) of reactive sites (A) appears to be near constant!

Specific Surface Area*

200

300

400

500

600

700

800

0 20 40 60 80Degree of carbon oxidation, %

Sp

ecific

S.A

. m

2/g

Printex

Soot A

Soot B

A. Yezerets, et al. Applied catalysis B: 61 (2005), 134-143.

–BET surface area measured in-situ at different stages of oxidation–samples pre-treated by thermal desorption

* Development of the reactivity does not appear to correlate directly with thespecific surface area

* Need a different parameter which would correlate with the number of active sites *Courtesy: Dr. Do Heui Kim, (PNNL)

Puzzles (thus far):

* Comparative changes in reactivity

* Comparative evolution of surface areas

Advantages of electron microscopy (HRTEM)

* Direct observation without property assumptions

* Potential to reveal changes in nanostructure (during oxidation)

* Correlate oxidation characteristics (rate) with nanostructure

………What changes in nanostructure?

Nanostructure and Implications: Reactivity

0

3

6

9

12

15

18

21

0 0.93 1.9 2.8 3.7 4.7 5.6 6.5

% o

f F

rin

ge

s

Fringe Length (nm)

Acetylene Derived Soot

0

3

6

9

12

15

18

21

0 0.93 1.9 2.8 3.7 4.7 5.6 6.5

% o

f F

rin

ge

s

Fringe Length (nm)

Benzene Derived Soot

0

3

6

9

12

15

18

21

0 0.93 1.9 2.8 3.7 4.7 5.6 6.5

% o

f F

rin

ge

s

Fringe Length (nm)

Ethanol Derived Soot

Amorphous Fullerenic Graphitic

(g/c

m-s

) x 1

0

Soot Burnout Rates

0

0.5

1

1.5

2

2.5 Bu

rnou

t Rat

e 2

-5

Ethanol Benzene AcetyleneNSC

w

0 5 10 15 20 25 30 35

Percent (%) Excess Oxygen

ENG-A - Original from Trap

ENG-A - Post partial Oxidation (TGA, 50%)

-15 degree tilt

No tilt ENG-A 75% Burnout

Hollow particlesviewed thoughSuccessive tilts

+15 degree tilt

Eng-A: Oxidized at 450 °C

Eng-A: As-received

“Solid particles” “Hollow particles”

PNNL

Printex - as supplied

Printex - Post Partial Oxidation50% Burnout, via TGA

Summary of Results

Sample

designation

Ash contents,

wt %

SOF contents,

wt%

Observations

ENG(A) 6.5 ± 0.5% 9 ± 1% No shells/capsules

Printex-U™ <0.5% 4 ± 1% No shells

Does Internal burning correlate with either ash or soluble organicfraction (SOF) content??

A Carbon Black

Statistical Properties Extracted FromHRTEM Images (of soot nanostructure)

A T = L/A

* Other inputs –Maximum join distance –Minimum fringe length

PositionPosition

LengthLength

FringeFringeSeparationSeparation

OrientationOrientation

TortuosityTortuosity

FringeFringeDensityDensity

Fringe Length (nm)

Tortuosity

Interpretation(s)

A. Densification - a pseudonym for graphitization

1. Thermally induced densificationCreation of radicals by thermal evolution of volatiles or loss of H-atoms permits lamella growth

2. Oxidation Creation of radical sites by oxidative removal of amorphouscarbon or loss of H-atoms

B. Internal Burning - disordered carbon and/or trapped volatiles preferentially burnout

Trap conditions could promote both densification and/or volatileevolution.

Phenols

XPS-Characterization of Carbon Nanostructure

C-C sp2

CarbonylsCarboxylic

C-C sp3

Conclusions * Burning mode dependent upon nanostructure and oxidation conditions

(ash and SOF are not unique predictors)* Diesel soot and Printex U exhibit nearly identical activation energies

and burning rates and even similar active site numbers BUT vastly different surface area evolution! * Measures other than surface area are needed for modeling

burning mode and rate.(A key feature will link the distribution of active sites to the

nanostructure)* Convolved with initial nanostructure are the change(s) enabled by

oxidation.

Implications:* Latter stage burnout will strongly depend upon burnout mode* Soot burning mode(s) could affect regeneration efficiency and models.* DPF regeneration costs fuel and each cycle limits lifetime.

Costs money!