Nanoparticle Emissions from Internal Combustion...

Nanoparticle Emissions from Internal Combustion Engines

Professor David B. KittelsonDepartment of Mechanical Engineering

University of MinnesotaMPLS, MN 55455 USA

presented at

The Royal Society Discussion MeetingUltra Fine Particles in the Atmosphere

15 March 2000 Engine Exhaust Particle Emissions:

Some Current Issues

Outline

• Background

• Formation and character of exhaust particles

• Nuclei mode formation and growth

• Role of carbonaceous agglomerates• Summary and Conclusions

Most of the recent U of M experimental work reported here was performed by Dr. Imad Abdul-Khalek and Mr. Qiang Wei with funding from Perkins Engine Company

Emissions of Ultrafine and Nanoparticles from Engines• Current emission standards are mass based. Recently interest

in other measures, i.e, size, number, surface, has increased.

• Concerns about particle size– New ambient standards on fine particles– Special concerns about ultrafine and nanoparticles– Indications that reductions in mass emissions may increase number

emissions

• Difficulties associated with measurement of ultrafine and nanoparticles– Often more than 90% of particle number are formed during exhaust

dilution– Particle dynamics during sampling and dilution are highly nonlinear -

large changes in of particle number may result from small changes dilution and sampling conditions

Typical Diesel Particle Size Distribution - Log Scale

Typical Engine Exhaust Size DistributionBoth Mass and Number Weightings are Shown

0.00001

0.0001

0.001

0.01

0.1

1

0.001 0.010 0.100 1.000 10.000

Diameter (µµ m)

No

rmal

ized

Co

nce

ntr

atio

n, d

C/C

tota

l/dlo

gD

p

Mass Weighting Number Weighting

Fine ParticlesDp < 2.5 µm

Ultrafine ParticlesDp < 100 nm

NanoparticlesDp < 50 nm

NucleiMode

AccumulationMode

CoarseMode

PM10Dp < 10 µm

Number Size Distribution Data from HEI Report and 1979 CRC Roadway Study

1.0E+06

1.0E+07

1.0E+08

1.0E+09

1.0E+10

0.0010 0.0100 0.1000 1.0000

Particle Diameter ( µµm)

dN

/d(l

og

(Dp

)) (

par

t./c

m3 )

1991 engine, low S fuel 1988 engine low S fuel 1979 Roadway study

The new engine increased number emissions 10 to 30 fold!

HOWEVER, 1979 roadway measurments made on behind a truck powered by anengine of the same family showed highnanoparticle emissions! Other roadsideand on road measurements made since the late 60's have shown high nanoparticle emissions

The new engine sharply reduced mass emissions, by about a factor of 3.

Normalized Particle Size Distributions from 1999 Tuscarora Tunnel Study, Courtesy Alan Gertler, Desert Research Institute

1

10

100

1000

1 10 100 1000

Diameter (nm)

d N

/d lo

gD

13% HD

65% HD

78%HD

Mainly Light-DutySpark Ignition

Mainly Heavy-Duty Diesel

Nanoparticles in the atmosphere appear to be associated with both spark ignition and diesel engines.

Outline

• Background




Typical Composition of Diesel Particulate Matter: Lube oil contributes to SOF, ash, sulfate

60%10%

10%

13%7%

Carbon

Ash

Sulfate and Water

Lube Oil SOF

Fuel SOF

Particles consist mainly of highly agglomerated solid carbonaceousmaterial and ash and volatile organic and sulfur compounds.

Hydrocarbon / Sulfate Particles

Sulfuric Acid Particles

Solid Carbonaceous / Ash Particles withAdsorbed Hydrocarbon / Sulfate Layer

0.3 µm

Particle Composition and Structure

Carbon formation/oxidationt = 2 ms, p = 150 atm.,

T = 2500 K

Ash Condensationt = 10 ms, p = 20 atm.,

T = 1500 K

Exit Tailpipet = 0.5 s, p = 1 atm.,

T = 600 KSulfate/SOF

Nucleation and Growtht = 0.6 s, p = 1 atm.,D = 10, T = 330 K

Fresh Aerosol overRoadway–Inhalation/Aging

t = 2 s, p = 1 atm.,D = 1000 T = 300 K

IncreasingTime

Formation

Atmospheric AgingExposure

Particle Formation History: From the Start of Combustion to the Nose

University of Minnesota

This is where most of the nanoparticles emitted by engines

usually form.

Outline

• Background




Studies of Diesel Nanoparticle Formation Using a Variable Residence Time Dilution System

Engine

CPC

PrimaryEjectorPump

Insulated Variable Residence Time

section

Diluted NOx

SMPS

OverflowBypass

Three Way Valve

SecondaryOrifice

SecondaryEjector Pump

OverflowBypass

PressureGauge

PrimaryOrifice

ExhaustPipe

Tygon Flexible

Tube

Air InletThermocouple

Humidity and temperature

Sensor

PressureGauge

Emission Rack, CO2,

NOx

Three Way Valve

SamplingProbe

Thermo-couple

Retractable tubing

Heated or Cooled Compressed Air Compressed

Air

Influence of Residence Time onNumber Weighted Size Distributions

1.00E+05

1.00E+06

1.00E+07

1.00E+08

1.00E+09

1.00E+10

1 10 100 1000

Dp (nm)

DN

/DL

og

Dp

(P

art.

/cm

³)

Residence time = 1000 ms 100 ms 230 ms

Tdilution = 32 °C, Primary DR ~ 12

1600 rpm, 50% load

Influence of Dilution Temperature on Number Weighted Size Distributions

1.00E+05

1.00E+06

1.00E+07

1.00E+08

1.00E+09

1.00E+10

1 10 100 1000

Dp (nm)

DN

/DL

og

Dp

(P

art.

/cm

³)

Tdil (32 °C) Tdil (48 °C) Tdil (65 °C)

Residence Time ~ 400 ms, Primary DR ~ 12Humidity Ratio ~ 0.0016

1600 rpm, 50% load

Influence of Primary Dilution Ratio on Number Weighted Size Distributions

1.00E+05

1.00E+06

1.00E+07

1.00E+08

1.00E+09

1.00E+10

1 10 100 1000

Dp (nm)

DN

/DL

og

Dp

(P

art.

/cm

³)

Primary Dilution Ratio (PDR) ~12

PDR ~ 25

PDR ~ 40

Primary Dilution Temperature (PDT) ~ 48 °C,

Residence time ~1000 ms, HR ≤≤ 0.0016

1600 rpm, 50% load

Our results suggest nanoparticles are mainly volatile, we have demonstrated this with a catalytic stripper

Typically more than 99% of the nanoparticles disappear by 300 C (except with metal additives)

1.000E+06

1.000E+07

1.000E+08

1.000E+09

1.000E+10

1 10 100 1000

Dp (nm)

DN

/DL

og

Dp

(Par

t./c

m³)

Engine Out-1000 ms Engine Out-40 ms

Stripper-160 °C-1000 ms Stripper-295 °C-1000 ms

Engine Out (Res. 1000 ms)

Stripper - 295 °C

Stripper– 160 °C

Engine Out (Res. 1000 ms)

Engine Out (40 ms)

Which Volatile Materials in the Exhaust Are Likely Precursors to Nuclei Mode Formation and Growth?

Influence of Dilution Upon Extractables

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

1 10 100

Dilution Ratio

Sat

ura

tio

n r

atio

Sat. Ratio Adiab. Sat. Ratio

Assumes extractable mass of

8 mg/m3 is nonodecaneTex = 240 C, Ta = 27 C

These levels are too low for nucleation but would drive absorption and adsorption

For typical diesel exhaust conditions, saturation ratio of hydrocarbons comprising the SOF may not be high enough for nucleation, S > ~ 3

• An extreme dependence of the nucleation rate upon temperature, H2SO4, and H2O concentrations makes theoretical predictions of nucleation difficult.

• Seinfeld and Pandis give an empirical expression to predict the onset of nucleation in this system:

Ccrit = 0.16exp(0.1T - 3.5RH - 27.7)where Ccrit is the threshold H2SO4 concentration in µg/m3, T is temperature, RH is relative humidity (0 to 1).

• I have used this expression and mass and energy balances applied to the dilution process to predict the ratio of the actual H2SO4 concentration to the critical one, C/Ccrit. When this critical concentration ratio exceeds one, nucleation is likely.

What does nucleate? Consider the threshold for Binary H2SO4-H2O nucleation

Influence of Fuel and Exhaust Conditions on Sulfuric Acid Nucleation

0.01

0.1

1

10

100

1 10 100 1000 10000

Dilution ratio

Cri

tica

l Co

nce

ntr

atio

n R

atio

C/C

crit

Tex = 150 C

240 C

330 C

Tex = 150 C

240 C

330 C

Nucleation likely for

C/Ccrit > 1

500 ppm Fuel Sulfur

5 ppm Fuel Sulfur

Fuel sulfur conversion to sulfate = 3%,

Air-fuel ratio = 30, Tair = 27 C, RH = 40%

Influence of Fuel and Lube Oil Sulfur (courtesy C.Barnes, Perkins Engine Company)

1.00E+05

1.00E+06

1.00E+07

1.00E+08

1.00E+09

1.00E+10

1 10 100 1000

Dp (nm)

dN

/dlo

g(D

p)

par

t/cm

3) 1

07 Jan 2000, 450 ppm Sulphur, Normal Lube Oil

11 Nov 1999, 3 ppm Sulphur, Normal Lube Oil

16/18 Nov 1999, 3 ppm Sulfur Fuel + Low Sulphur Lube Oil

Medium-Duty Engine, 1600 rpm, 10% Load

Sulfuric Acid Appears To Trigger Nucleation But Typical Concentrations Are Too Low To Produce Observed Particle Growth Rates

0

100

200

300

400

500

600

700

0 0.2 0.4 0.6 0.8 1

Time (s)

Vap

or

Den

sity

( µµg

/m3)

0

5

10

15

20

25

30

35

40

45

Dro

ple

t D

iam

eter

(n

m)

Acid vapor density Hydrocarbon vapor densityDp acid/water Dp hydrocarbon

Maximum rates of particle growth and vapor depletion in free molecularregime. 400 PPM sulfur fuel, 3% sulfur to sulfate conversion, 15% RH, 0.4 m2/m3 soot surface area, 8 mg/m3 C19H40 as hydrocarbon, Dilution ratio 12, T dilution 32 C

Growth rates of about 20 nm/s are observed for these fuel and dilution conditions

Nanoparticle Nucleation and Growth

• It appears that with current engines binary sulfuric acid - water nucleation triggers the process.

• The initial size of these nuclei is about 1 nm.

• In most cases there is not enough sulfuric acid present in the exhaust to explain the observed rates of particle growth.

• Hydrocarbons normally associated with the soluble organic fraction apparently are absorbed by the concentrated sulfuric acid nuclei leading to the observed growth rates.

Outline

• Background



• Role of carbonaceous agglomerates

• Summary and Conclusions

Carbonaceous Agglomerates Adsorb Volatile Particle Precursors and Suppress Nanoparticle Nucleation and Growth

• A time constant for molecular adsorption may be determined from the molecular collision frequency and the particle surface area available for adsorption.

• Under typical roadway conditions the dilution ratio, D, of an exhaust plume varies roughly linearly in time so that a time constant for dilution can be defined as follows.

• Then a dimensionless adsorption rate can be defined.

= pa A

MRT

5.0

2/1

πτ

dtD τ/1+=

adadsR ττ /=

H2SO4 Nucleation and Growth Is Suppressed by Adsorption onto Carbon Agglomerates

1.00E-02

1.00E-01

1.00E+00

1.00E+01

1.00E+02

1 10 100 1000 10000

Dilution Ratio

Co

nce

ntr

atio

n R

atio

C/C

crit

Rads = 0 0.01 0.02 0.08 0.27 0.92 3.20

Nucleation suppressed by increasing carbon concentration or slower dilution

Increasing dimensionless adsorption rate

Low exhaust carbon emissions and/or fast (roadway) dilution.More potential to form nanoparticles.

High exhaust carbon emission and/orslow (laboratory, congested traffic) dilution.Less potential to form nanoparticles.

H2SO4 / H2O nucleation

500 PPM S fuel, 3% conversion

Texh =330 C, Tair = 27 C, RH =40%

Outline

• Background




Summary Diesel - 1

• A significant amount of particulate matter (e.g. 90 % of the number and 30% of the mass) is formed during exhaust dilution from particle precursors that are in the vapor phase inthe tailpipe (e.g., sulfuric acid, fuel and oil residues). – New particles are formed by nucleation. This is likely to be the source

of most of the ultrafine and nanoparticles (and particle number)associated with engine exhaust.

– Preexisting particles grow by adsorption or condensation.

– Nucleation and adsorption are competing processes. Carbon agglomerates provide a large surface area for adsorption that suppresses nucleation.

– Thus, engines with low carbon mass emissions may have high number emissions of volatile, but NOT solid particles.

– It should be possible to reduce nanoparticle emissions by reducing emissions of sulfuric acid and hydrocarbons.

Summary Diesel - 2

• Nanoparticle measurements are very strongly influenced by the sampling and dilution techniques employed.– Nucleation, adsorption, absorption, and coagulation during sampling

and dilution depend upon many variables, including dilution rate, (or residence time at intermediate dilution ratio), humidity, temperature, and relative concentrations of carbon and volatile matter.

– Changes of more than two orders of magnitude in nanoparticle concentration may occur as dilution conditions are varied over the range that might be expected for normal ambient dilution, e.g., 0.1 to 2 s dilution time scales. Even larger changes may occur with engines that emit very little carbon.

– Carbon in the exhaust suppresses nanoparticle formation less effectively when the exhaust is diluted rapidly over a roadway than when it is diluted slowly.

• Sampling systems should mimic atmospheric dilution to obtain samples representative of the tailpipe to nose process.

The most difficult problem associated with exhaust particle measurement is understanding the dilution process from tailpipe to nose -- this is being examined in the CRC E-43 program

This illustration shows a typical application of the University of Minnesota Mobile Aerosol laboratory developed for the CRC E-43 Program

Nanoparticles are produced by new engines, but concentrations may be much lower than feared

1.0E+02

1.0E+03

1.0E+04

1.0E+05

1.0E+06

1 10 100 1000

Particle Diameter (nm)

dN

/d(l

og

(Dp

)) (

par

t./c

m3 )

Cruise

Hard Acceleration

Summary Diesel - 3

• Currently most of the particles in the nanoparticle size range are volatile. However, as engines become cleaner, metallic ash particles from the lubricating oil (or fuel if metallic additives are present) may become more important.– Solid particles raise new issues…

– But they are easy to control with exhaust filters

• Spark ignition engines typically emit smaller particles than diesel engines and are an important source of fine particles andnanoparticles.– A recent study in Colorado concluded that up to 2/3 of the fine particle

mass emitted by vehicles was from spark ignition engines– New gasoline direct injection engines emit much higher particle

concentrations than conventional engines and may approach diesellevels under some conditions

Nanoparticle Emissions from Internal Combustion...

Documents

Transcript of Nanoparticle Emissions from Internal Combustion...