real-world characterization of particulate matter emissions from

CAFEECenter for Alternative Fuels, Engines, and Emissions

Cambridge Particle Workshop

July 3rd, 2015

Cambridge, UK

CHARACTERIZATION OF PM MASS AND NUMBER EMISSIONS EMITTED BY CURRENT TECHNOLOGY

HEAVY-DUTY ENGINES

REAL-WORLD CHARACTERIZATION OF PARTICULATE MATTER EMISSIONS FROM HEAVY-DUTY TRUCKS

OPERATING IN CALIFORNIA

Arvind Thiruvengadam, Marc Besch, Saroj Pradhan, Daniel Carder

West Virginia University

Mridul Gautam

University of Nevada, Reno

David Quiros, Shaohua Hu, Tao Huai

California Air Resources Board

Adewale Oshinuga, Randall Pasek

South Coast Air Quality Management District

Eon S. Lee, Yifang Zhu

University of California, Los Angeles



July 3rd, 2015

Cambridge, UK

DISCUSSION I - INTRODUCTION

2

• Evaluation of PM emissions from heavy-duty vehicles

powered by Diesel, natural gas or NG/Diesel dual-fueled engines

and,

equipped with advanced after-treatment technology

operated over a variety of real-world driving cycles using a

chassis dynamometer

• Characterization of particle number concentration and

size distributions

• Characterize real-time DPF filtration efficiencies using in-

line particle sensor up/downstream DPF

Effect of regeneration events on DPF filtration efficiency

Restoring rate of high filtration efficiency after regeneration event

=> “soot cake layer” build up



July 3rd, 2015

Cambridge, UK

METHODOLOGY - Test Vehicles

3

Cat Engine/TechnologyTransit

Bus

Refuse

TruckOTR

I Natural Gas Engine with TWC 1 1 3

II HPDI Engine with EGR, DPF at 0.8g NOx 3

III HPDI Engine with EGR, DPF, SCR at 0.2g NOx 1

IV Diesel Engine cert. at 1.2g NOx 1

V Diesel Engine cert. above 0.2g NOx w/o SCR 1 1

VI Diesel Engine cert. at or below 0.2g NOx w/SCR 1 3

Category IVCategory V

Category VI

Category I Category IIICategory II



July 3rd, 2015

Cambridge, UK

METHODOLOGY - PM/PN Sampling System

4

Particle concentrations

and size distributions in

diluted exhaust (CVS)

Gravimetric TPM on

47mm TX40 filter media

(sampling following

recommendations

outlined in 40 CFR 1065)

Total particle number concentrations using double

dilution system with heated first stage and cold

second stage dilution

Gravimetric PM1.0, PM2.5, PM10

47mm TX40 filter media

Parameter 1st Stage 2nd Stage

Diluter TypeEjector Diluter

(AirVac TD110-H)

Ejector Diluter

(AirVac TD110-H)

Dilution Air

Temperature~150°C ~20 - 25°C

Dilution Air

Condition

Dry, HEPA filtered

air

Dry, HEPA filtered

air

Residence Time ~0.6sec Negligible

Dilution Ratio ~7 ~24



July 3rd, 2015

Cambridge, UK

METHODOLOGY - Real-Time PM Sensor

5

• PM detection based on diffusion charging

and escaping current principle

• Flow through device => low maintenance

• Constant dilution air pressure leading to

constant sample inlet flow => constant

internal dilution ratio

• Sensor shows proportional response to

particle surface area

Pre DPF Sampling Location

• Sample extraction via 3ft heated line (@200°C)

• Internal DR ≈ 2.9 @ 22psi dil. pressure

• Sensor heated to 200°C

Post After-treatment Sampling Location

• Directly mounted to exhaust stack

• Internal DR ≈ 3.06 @ 22psi dil. pressure

• Sensor heated to 200°C

Sample ProbePPS

DO

C

DPF

EGR PPSPre-DPF

SCR

Urea Injection

PPSPost-aftertreatment

To CVS

Diesel fueled

DO

C

DPF

EGR PPSPre-DPF PPSPost-aftertreatment

To CVS

Diesel fueled



July 3rd, 2015

Cambridge, UK

METHODOLOGY - Test CyclesVehicle

Vocation

HD-

UDDS

SCAQMD Refuse

Truck Cycle

OCTA

Cycle

Double-length

CBD Cycle

Drayage Port Cycles

i) Neardock, ii) Local, iii) Regional

Refuse Truck X X

Transit Bus X X X

OTR X X

0

20

40

60Neardock Cycle

0

20

40

60Local Cycle

Vehic

le S

peed [m

ph]

0 500 1000 1500 2000 2500 3000 3500 4000 45000

20

40

60Regional Cycle

Time [sec]



July 3rd, 2015

Cambridge, UK

7

RESULTS - TPM, NG Vehicles (Cat I)

0

5

10

15

20

25

30

35U

DD

S

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

AQ

MD

RT

Cyc

le

AQ

MD

RC

Cyc

le

UD

DS

OC

TA

2 x

CB

D

OTR 1 (low mileage) OTR 2 (medium mileage) OTR 3 (medium mileage) Refuse Truck Transit Bus

TP

M E

mis

sio

n R

ate

s [m

g/m

ile]

dsPM dsPM Bgcorr

(116,232 mi)(21,465 mi)(63,256 mi)(45,563 mi)(192 mi)

• Increased TPM emissions for refuse truck and transit bus possibly due to increased oil

consumption



July 3rd, 2015

Cambridge, UK

8

RESULTS - TPM, Dual-Fuel Vehicles

(Cat II & III)

0

5

10

15

20

25

30

35

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Cat II - OTR 1 Cat II - OTR 2 Cat II - OTR 3 Cat III - OTR

TP

M E

mis

sio

n R

ate

s [m

g/m

ile]

dsPM dsPM Bgcorr

(45,621 mi)(45,621 mi)(368,080 mi)(196,562 mi)

Local immediately

followed UDDS



July 3rd, 2015

Cambridge, UK

9

RESULTS - TPM, Diesel Vehicles

(Cat IV & V & VI)

0

5

10

15

20

25

30

35

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

AQ

MD

RT

Cyc

le

AQ

MD

RC

Cyc

le

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

UD

DS

AQ

MD

RT

Cyc

le

AQ

MD

RC

Cyc

le

IV - OTR w/ DPF V - OTR w/o SCR V - Refuse Truckw/o SCR

VI - OTR 1 w/ SCR VI - OTR 2 w/ SCR VI - Refuse Truckw/ SCR

TP

M E

mis

sio

n R

ate

s [m

g/m

ile]

dsPM dsPM Bgcorr

(178,564 mi)(36,982 mi)(10,014 mi)(67,373 mi)(80,412 mi) (14,269 mi)

2009 post-2010



July 3rd, 2015

Cambridge, UK

0

5

10

15

20

25

30

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

AQ

MD

RT

Cyc

le

AQ

MD

RC

Cyc

le

Bck

gnd

UD

DS

OC

TA

2 x

CB

D

Bck

gnd

OTR 1 (low mileage) OTR 2 (medium mileage) OTR 3 (medium mileage) Refuse Truck Transit Bus

Em

issio

ns R

ate

s [m

g/m

ile],

Bckg

nd

[m

g/h

]

PM1.0 PM2.5 PM10

RESULTS - PM1, 2.5, 10, NG Vehicles (Cat I)

10

PM10

79.9g/h

(116,232 mi)(21,465 mi)(63,256 mi)(45,563 mi)(192 mi)



July 3rd, 2015

Cambridge, UK

0

20

40

60

80

100

120

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

Cat II - OTR 1 Cat II - OTR 2 Cat II - OTR 3 Cat III - OTR

Em

issio

ns R

ate

s [m

g/m

ile],

Bckgnd [m

g/h

]

PM1.0 PM2.5 PM10

11

RESULTS - PM1, 2.5, 10, Dual-Fuel Vehicles

(Cat II & III)

(45,621 mi)(45,621 mi)(368,080 mi)(196,562 mi)

1.E+04

1.E+05

1.E+06

1.E+07

1.E+08

1 10 100 1000

Cat II - OTR 1 Cat II - OTR 1 Cat II - OTR 2 Cat II - OTR 2

UDDS



July 3rd, 2015

Cambridge, UK

12

0

20

40

60

80

100

120

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

AQ

MD

RT

Cyc

le

AQ

MD

RC

Cyc

le

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

AQ

MD

RT

Cyc

le

AQ

MD

RC

Cyc

le

Bck

gnd

IV - OTR w/ DPF V - OTR w/o SCR V - Refuse Truckw/o SCR

VI - OTR 1 w/ SCR VI - OTR 2 w/ SCR VI - Refuse Truck w/SCR

Em

issio

ns R

ate

s [m

g/m

ile],

Bckg

nd

[m

g/h

]

PM1.0 PM2.5 PM10

RESULTS - PM1, 2.5, 10, Diesel Vehicles

(Cat IV & V & VI)

Regeneration Event

• Increase in size selective PM due to regeneration for Cat IV OTR (2009) and refuse truck

w/o SCR after-treatment system



July 3rd, 2015

Cambridge, UK

13

0

5

10

15

20

25

30

35

40

45

50

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

IV - MY 2009 Diesel w/ DPF VII - MY 2011 Diesel w/o SCR VIII - MY 2011 Diesel w/ SCR

Em

issio

ns R

ate

s [m

g/m

ile],

Bckgn

d [m

g/m

in]

TEC

TOC

0

10

20

30

40

50

60

70

80

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

MY 2009 Vehicle 1 MY 2009 Vehicle 2 MY 2009 Vehicle 3 MY 2011 Vehicle 4

Em

issio

ns R

ate

s [m

g/m

ile],

Bckgn

d [m

g/m

in]

TEC

TOC

0

2

4

6

8

10

12

14

16

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

UD

DS

Loca

l

Reg

ion

al

Nea

r-D

ock

Bck

gnd

Vehicle 1 Vehicle 2 Vehicle 3

Em

issio

ns R

ate

s [m

g/m

ile],

Bckgn

d [m

g/m

in]

TEC

TOC

0

2

4

6

8

10

12

14

16

18

20

UDDS AQMDRefuse

Truck Cycle

AQMDRefuse

CompactionCycle

Bckgnd UDDS AQMDRefuse

Truck Cycle

AQMDRefuse

CompactionCycle

Bckgnd

VII - MY 2011 Refuse Truck w/o SCR VIII - MY 2011 Refuse Truck w/ SCR

Em

issio

ns R

ate

s [m

g/m

ile],

Bckgn

d [m

g/m

in]

TEC

TOC

RESULTS - EC/OC Analysis

Cat I, NG VehiclesCat V & VI, Diesel

Cat IV & V & VI, Diesel Cat II & III, Dual-fuel

OTR 1 OTR 2 OTR 3 Cat V Cat VI

Cat VICat VCat IV

OTR 1 with lowest mileage (192 mi)

=> peak catalytic activity not yet

reached (ageing needed)

DPF regeneration event

DPF regeneration event



July 3rd, 2015

Cambridge, UK

14

101

102

104

105

106

107

108

109

Inte

gra

ted C

oncentr

atio

n d

N/d

log(D

p)

[#/c

m3]

Diameter (Dp) [nm]

Dp [

nm

]

0 100 200 300 400 500 600 700 800 900 1000 11005.6

10

24

42

87

178

365

560

0 0.5 1 1.5 2 2.5 3 3.5

x 106

0 100 200 300 400 500 600 700 800 900 1000 11000

1

2

3

4

5

6

7x 10

5

Tot

Conc.

[#/c

m3]

0 100 200 300 400 500 600 700 800 900 1000 11000

100

200

300

400

Exh.

Tem

pera

ture

s [C

]

Time [sec]

Post SCR

PPS Sample Plane

RESULTS - Particle Size Distribution

• Particle size distribution measured in CVS

(no dilution, no correction for CVS dilution)

• Cat VI Vehicle with SCR operated over

UDDS

• Large release of 17.8nm (CMD) particles

starting at ~650sec

No DPF regeneration event => supported

by Pegasor and CPC measurement in raw

exhaust

• Removing data between ~650 to 840sec

leads to an order of magnitude reduction in

17.8nm particles

1.87x108 vs. 1.58x107

=> UDDS

Dp [nm

]

0 100 200 300 400 500 600 700 800 900 1000 11005.6

10

24

42

87

178

365560

0 0.5 1 1.5 2 2.5 3 3.5

x 106

0 100 200 300 400 500 600 700 800 900 1000 11000

1

2

3

4

5

6

7x 10

5

Tot C

onc. [#

/cm

3]

0 100 200 300 400 500 600 700 800 900 1000 11000

100

200

300

400

Exh. T

em

pera

ture

s [C

]

Time [sec]

Post SCR

PPS Sample Plane

Pre DPF

200C



July 3rd, 2015

Cambridge, UK

15

RESULTS - Particle Size Distribution UDDS

101

102

104

105

106

107

108

109

Inte

gra

ted C

oncentr

atio

n d

N/d

log(D

p)

[#/c

m3]

Diameter (Dp) [nm]

Cat VI - OTR 1 w/ SCR (UDDS)

101

102

104

105

106

107

108

109

Inte

gra

ted C

oncentr

atio

n d

N/d

log(D

p)

[#/c

m3]

Diameter (Dp) [nm]

Cat VI - OTR 2 w/ SCR (UDDS)

101

102

104

105

106

107

108

109

Inte

gra

ted C

oncentr

atio

n d

N/d

log(D

p)

[#/c

m3]

Diameter (Dp) [nm]

Cat V - OTR w/o SCR (UDDS)

101

102

104

105

106

107

108

109

Inte

gra

ted C

oncentr

atio

n d

N/d

log(D

p)

[#/c

m3]

Diameter (Dp) [nm]

Cat V - Refuse Truck w/o SCR (UDDS)

101

102

104

105

106

107

108

109

1010

1011

Inte

gra

ted C

oncentr

atio

n d

N/d

log(D

p)

[#/c

m3]

Diameter (Dp) [nm]

Cat VI - Refuse Truck w/ SCR (UDDS)

101

102

104

105

106

107

108

109

1010

1011

Inte

gra

ted C

oncentr

atio

n d

N/d

log(D

p)

[#/c

m3]

Diameter (Dp) [nm]

Cat IV - OTR w/ DPF (UDDS)

Regeneration

Increase in 17.8nm particles not due to DPF

regeneration => low raw concentrations

Increase in 9.31nm particles not due to DPF

regeneration => low raw concentrations



July 3rd, 2015

Cambridge, UK

16

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2x 10

6

Time [sec]

CP

C P

art

icle

Concentr

ation [

#/c

m3]

0 200 400 600 800 10000

2

4

6

8

10

12

14

16

18

20

PP

S p

ost-

aft

ert

reatm

ent

Sig

nal [m

v]

Axle-Work-specific PN [#/kW-hr] 6.22 E+13 2.16 E+13

Distance-specific PN [#/km] 9.92 E+13 3.44 E+13

DPF Filtration Efficiency [%] 99.0 99.59

Dilution

corrected with

DR ≈ 160

Re

d

Blu

e

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2x 10

6

Time [sec]C

PC

Part

icle

Concentr

ation [

#/c

m3]

0 200 400 600 800 10000

2

4

6

8

10

12

14

16

18

20

PP

S p

ost-

aft

ert

reatm

ent

Sig

nal [m

v]

Re

d

Blu

e

Dilution

corrected with

DR ≈ 160

With DPF Regeneration Event Without DPF Regeneration Event

• Three-fold increase in PN during cycle with regeneration event

Total engine-out PN levels increase by ~25%

W/ Regen. W/O Regen.

RESULTS - Diesel, DPF (1.2g/bhp-hr NOx) UDDS



July 3rd, 2015

Cambridge, UK

17

1.E+00

1.E+02

1.E+04

1.E+06

1.E+08

1.E+10

1.E+12

1.E+14

I II III IV V VI

bsPN [#/kW-hr]

dsPN [#/km]

I Diesel, SCR/DPF (< 0.2g/bhp-hr NOx)II Diesel, DPF/heavy EGR (> 0.2g/bhp-hr NOx)

III Diesel, DPF (1.2g/bhp-hr NOx) => no regenIV Diesel, DPF (1.2g/bhp-hr NOx) => with regen

V CNG/Diesel, HPDI 3rd gen/DPF (0.8g/bhp-hr NOx)VI CNG/Diesel, HPDI 4th gen/DPF-SCR (< 0.2g/bhp-hr NOx)

98.4%

98.6%

98.8%

99.0%

99.2%

99.4%

99.6%

99.8%

100.0%

I II III IV V VI

Tota

l D

PF

Filt

ration E

ffic

iency [%

]

with regeneration

event

• Filtration efficiencies calculated using PPS

measurements up/downstream

aftertreatment system

• For DPF/SCR equipped vehicles => filtration

efficiency corresponds to overall

aftertreatment particle removal efficiency

=> [kW-

hr]

represent

s axle

work

RESULTS - Total PN Comparison



July 3rd, 2015

Cambridge, UK

18

0 200 400 600 800 1000 12000

200

400

600

800

1000

1200

1400

1600

1800

2000

Time [sec]

PP

S S

ignal [m

V]

I - Diesel, SCR/DPF (< 0.2g/bhp-hr NOx)

VI - CNG/Diesel, HPDI 4th gen/DPF-SCR (< 0.2g/bhp-hr NOx)

• Reduced engine-out particle number concentrations for dual-fueled vs. diesel-only

fueled engine during acceleration and high load modes

• Increased engine-out PN during idle portions for dual-fueled engine

RESULTS - Engine-out PN, Diesel vs. Dual-fueled

Engine



July 3rd, 2015

Cambridge, UK

• Real-world emissions from heavy-duty vehicles are of importance

• Operation of SCR (identifying operations leading to higher NOx

emissions)

• DPF filtration efficiencies (Operations that could lead to higher solid

particle emissions)

• In addition to in-use conformity testing using PEMS instrumentation, the

use of a transportable laboratory-grade CVS system offers better

accuracy and versatility in integrating measurement instrumentation

• Regulated pollutants(laboratory grade and PEMS)

• Unregulated pollutants (FTIR)

• PM speciation (EEPS, PMP sampling, soot sensor, etc.)

• European RDE proposal also voices the need for monitoring vehicle

emissions beyond just certification cycles, in particular urban operation.

DISCUSSION II - INTRODUCTION

19



July 3rd, 2015

Cambridge, UK

Day 2: 146 mi

Port of LB/LA Route

Congested Hwy

+ Creep Simulation

Day 1: 437 mi

Central Valley Via

I-5

Open Hwy +

GrapevineDay 5: 439 mi

Central Valley

Via SR-99

Open Hwy +

Grapevine

Day 3: 150 mi

OC Delivery

Congested Hwy +

Suburban Arterial

Day 4: 521 mi

Needles/Blythe

Open Hwy +

Cajon Pass

20



July 3rd, 2015

Cambridge, UK

• Three dilution setups were used

• Mini-Dilution Setup: Ejector-type dilution of raw exhaust (HEPA filtered dry pressurized

air at 25 DegC)

• CVS dilution with HEPA filtered ambient air

• Thermodilution using rotating disk dilutor and TSI Thermodilution (Model 379020A) and

Conditioning system (PMP compliant-Model 379030)

METHODOLOGY - Sampling System

21



July 3rd, 2015

Cambridge, UK

Vehicle Engine Model After Treatment

System

Vehicle 1 (MY 2008) Cummins ISX 525 DPF only

Vehicle 2 (MY 2013) Cummins ISX 15/450 DPF and SCR

Vehicle 3 (MY 2013) Cummins ISX 12G

(Natural gas)

Three-way catalyst

Vehicle 4 (MY 2014) Detroit Diesel DDC15 DPF and SCR

• Vehicle 5 (Volvo D13 with DPF and SCR) and Vehicle 6 (Navistar N13

with DPF and SCR) to be tested.

METHODOLOGY - Test Vehicles

22



July 3rd, 2015

Cambridge, UK

KEY QUESTIONS OR

HYPOTHESESQuestion 1: Does the activity of SCR result in any new

particle formation?

Question 2: How effective are DPFs during real-world

operation?

Hypothesis 1: Idle and low load operation of natural gas

engines produces higher PM emissions. Such PM emissions

are dominated by lubrication oil derivatives.

Hypothesis 2: No significant differences in PM emissions are

observed between different engine manufacturer platforms.

23



July 3rd, 2015

Cambridge, UK

24

0 1000 2000 3000 4000 5000 6000 7000 80000

20

40

60

80

Vehicle Speed [mph]

0 1000 2000 3000 4000 5000 6000 7000 80000

0.1

0.2

0.3

0 1000 2000 3000 4000 5000 6000 7000 80000

2

4

6NOx [g/s]

NH3 [ppm]

0 1000 2000 3000 4000 5000 6000 7000 8000

200

400

600

avg SCR Temp [degC]

avg DPF Temp [degC]

0 1000 2000 3000 4000 5000 6000 7000 800010

7

109

1011

1013

Time [sec]

PM # Mini-dilution [#/sec]

PM # CVS-dilution [#/sec]

PM # Thermo-dilution [#/sec]

Periodic SCR

activity and

increased PN

count

Observed particles

are believed to be

liquid in nature and

magnitude is

lowered upon heat

treatment of

sample

RESULTS: Question 1

SCR Activity and Particle Formation



July 3rd, 2015

Cambridge, UK

Question 1 Cont’d

25

*Impact of Selective Catalytic Reduction on Exhaust Particle Formation over Excess Ammonia Events,

Environ. Science and Tech, 2014, 48, 11527−11534

Accumulation mode particles

coinciding with SCR activity

• Further questions: Ammonium nitrate? Ammonium sulfate

or ammonium chloride????????

• Aggressive urea dosing could increase the particle count

Vehicle 2: DPF-SCR

Vehicle 4: DPF-SCR

=> Results derived from extended highway operation

Thermodilution (PMP)



July 3rd, 2015

Cambridge, UK

26

• For measurements from the CVS no significant differences are observed in PM emissions between

different power levels

• In the case of thermodilution of sample

Close to an order of magnitude higher PN count for power levels between idle and 20% load in

comparison to higher load operation

• Previous studies on chassis dynamometer have indicated possible inorganic nucleation mode particles

derived from lube oil.

• Presence of nucleation mode particles after evaporator stage of thermodilutor indicates the possible

presence of solid particles (lube oil ash, re-nucleated lube oil additives)

< 10 nm particles

RESULTS: Hypothesis 1

Natural Gas Engine PM vs. Engine Load



July 3rd, 2015

Cambridge, UK

27

• Distance-specific PN count are

well below the to EU-PN

regulation of 6 X 1011 #/km

• Higher PN count with

thermodilutor sampling with a

2.5 nm cut-point CPC,

indicates particles below 23 nm

exists even after thermal

conditioning

• Differences in after-treatment

design and urea dosing

strategy between vehicles 2

and 4 do not seem to influence

the PN count significantly

• In the case of natural gas

engines the PMP measurement

might be ignoring the inorganic

particles below 23 nmVehicle 2 was equipped with a conventional DPF and SCR while Vehicle 4 was

equipped with an integrated DPF and SCR design with an Ammonia Oxi. Catalyst

Above chart represents PN count for hill climb route

RESULTS: Hypothesis 2

Differences in PM Emissions Between Engine Platforms



July 3rd, 2015

Cambridge, UK

28

• Close to 100% filtration

efficiency during all

operating conditions

• Above chart shows

filtration efficiency for hill

climb route with maximum

gravimetric filter loading

• No regeneration events

observed during the 1700

mi test campaign on each

vehicle

ON A SEPARATE NOTE

Need of the hour: Robust OBD strategy to detect DPF failures (tailpipe

sensors????)DPF failures caused as a cascading effect originating from a engine durability issues-

EGR related issues invariably result in DPF failures undetected by engine

0 1 2 3 4 5 660

70

80

90

100

110

Time [sec]

DPF Efficiency [%]

RESULTS: Question 2

Real-World DPF Efficiency



July 3rd, 2015

Cambridge, UK

29

• Measuring particulate matter emissions from Diesel, natural gas, and

NG/Diesel dual-fueled engines equipped different after-treatment system

configurations.

• NG fueled engines showed lowest overall gravimetric TPM emissions

Increased TPM emissions observed for NG engines; i) with high mileage

(transit bus) and; ii) aggressive operation => frequent accel./decel. (refuse

hauler)

Due to increased oil consumption leading to ash particle formation

• DPF regeneration events increase TPM emissions

• PM1,2.5,10 emissions results show large variability in filter weights => low signal-

to-noise ratio

• PM for all engine technologies, fuels and after-treatment packages are

predominantly of organic nature (OC). EC is negligible for all engines unless

during DPF regeneration events that show a slight increase in EC.

• Particle size distributions and number concentrations measured in CVS were

found to be dependent on engine technology

DISCUSSION I - CONCLUSION



July 3rd, 2015

Cambridge, UK

30

1. SCR activity results in particles both in nucleation (e.g. sulfates) and

accumulation mode particles

2. Composition and the environmental impact (if any) from these accumulation

mode particles is worth investigating

3. Solid particles in the nucleation mode are observed in exhaust of natural

gas vehicles

a) Data from this study suggests that lubrication oil contribution is

prevalent even in engines with less miles

b) Engine component ageing could potentially increase the PN count from

natural gas engines

4. No significant differences in PM emissions (both mass and number) were

observed between different engine manufacturers

1. It would be interesting to observe the impact of optional regulation such

as ARB 0.02 g/bhp-hr Nox standard on PM emissions

5. OBD related development of PM sensors or fault detection algorithms are

the need of the hour to monitor DPF efficiency

DISCUSSION II - CONCLUSION



July 3rd, 2015

Cambridge, UK

31

OUTLOOK - Effects of Dilution Conditions on Particulate Matter Emitted from HD Engines: In-Plume, Windtunnel and

In-Laboratory Measurements

SCRDPF

Engine

Tpre-DPF

Ppre-DPF

Part ial-Flow System

Dry, Ambient

Temperature,

Compressed

HEPA Filtered Air

Supply

ECU Data

via J1939

Horiba OBS-2200

+

Horiba OBS-TRPM

Tpost-SCR

Pitot-tube

flowmeter

Ppost-SCR

TSI

EEPSMFC

TSI

EEPS

TEM Grid

Holder

Probe

FTIR

Analyzer

Adjustable

Probe Holder

Solenoid

Valves

Manifold

• Assess the likelihood of reproducing real-world PM size distributions collected from the

exhaust plume, by using two methodologies: CVS and Partial-Flow sampling

Differences in dilution rates occurring in dispersing plume => enhancement/suppression

of nucleation

• Compare PM development in real-world exhaust plume vs. full-scale windtunnel vs. full-flow

CVS vs. partial flow sampling

• Compare PM morphology and composition via TEM/EDX analysis of PM samples obtained

through plume and laboratory sampling



July 3rd, 2015

Cambridge, UK

32

• Provide spatial and temporal coverage of ambient air PM pollution measurement through a network of mobile low-cost sensors

• Monitor real-time tailpipe PM/PN emissions rates from heavy-duty vehicles

• Neural Network learning algorithms to be used as an advanced engine fault diagnosis and high emissions alert (Sensor Failure Detection and Accommodation) Monitoring real-time DPF filtration efficiency

AirCom Mobile

Sensor Platforms

Community Data

Usage and Interface

App

Mobile Application

Web-interface

Community Organizations

AirCom Data QA/QC, Processing and Storage

Sensor

Models

Pollutant

Gradient Map

Modeling

Stationary Air

Monitoring Data

CO2

CO NO2

O3

PM2.5

Concentration

Calculation

Drift

CorrectionQA/QC

OUTLOOK - Increase in special resolution of PM data via mobile, miniaturized measurement platforms

115ft



July 3rd, 2015

Cambridge, UK

33

THANK YOU FOR YOUR ATTENTION

Arvind Thiruvengadam - [email protected]

Marc C. Besch - [email protected]

Saroj Pradhan – [email protected]

Daniel K. Carder - [email protected]

Mridul Gautam - [email protected]

CAFEE Center for Alternative Fuels,

Engines and Emissions

ACKNOWLEDGEMENT

• ARB and SCAQMD for funding the study

• Don Chernich, Mark Burnitzki, Wayne Sobeiralski, Robert Ianni from ARB

Depot Park facility for their support and providing instrumentation

• Eon S. Lee, Yifang Zhu from UCLA for providing PM instrumentation

real-world characterization of particulate matter emissions from

Documents