Effect of Organometallic Fuel Additives on Nanoparticle Emissions from a Gasoline Passenger Car

Jeremy T. Gidney and Martyn V. TwiggJohnson-Matthey Plc.

David B. KittelsonUniversity of Minnesota

13th ETH Conference onCombustion Generated Particles

Zurich, 22rd – 24th June 2009

Outline

• Introduction• Experimental• Total particle measurements• Solid particle measurements• Conclusions

Introduction

• Paper submitted to EST– Jeremy T. Gidney, Martyn V. Twigg, and David B. Kittelson, 2009. Effect of

Organometallic Fuel Additives on Nanoparticle Emissions from a Gasoline Passenger Car

• Investigates the influence of inorganic fuel additives (Mn, Fe, Pb) on nanoparticle emissions from a PFI gasoline vehicle

• Issues– Health impacts

• Solid nanoparticles• Rapid translocation in body

– Catalyst plugging and engine deposits– Iron and manganese octane boosters still used in Eastern Europe, China, and

developing countries• Mn used at 30, 50, and 55 mg/liter in China, Russia, and Lithuania, respectively• Mn allowed but not used at levels up to 8 and 18 mg/liter in US and Canada,

respectively• Fe allowed at levels up to 37 mg/liter in Russia

Health

• Correlations between fine particles and excess deaths• Increased asthma and respiratory problems in

children living near roadways• Special concerns about ultrafine and nanoparticles

– More surface area and number per unit mass– Increased deep lung deposition efficiency– Very small particles may pass through cell membranes and

along neurons– Effects of solid particles clear, volatile particles uncertain

Outline

• Introduction• Experimental• Total particle measurements• Solid particle measurements• Conclusions

Vehicle and fuels used in study

Vehicle Subcompact carModel year 2000Engine type Gasoline spark ignitionFuel system Multi point injection

Engine 4 cylinder, 1.6 litre displacementCatalyst volume 1.66 litre

Emissions conformance European Stage 3Vehicle age 11,000 milesCatalyst age 4,000 miles

Catalyst cell density 400 cpsi, 6 mil wall thicknessCatalyst formulation Palladium/rhodium

Lubrication oil Fully synthetic 5 - 30 SAE

Table 1: Vehicle, catalyst and other details

Batch Additive Metal concentration Additive concentrationcode mg/l g/20 litreMn-8 (CH3C5H4)Mn(CO)3 8.3 0.658

Mn-18 (CH3C5H4)Mn(CO)3 18 1.429Fe-8 Fe(C6H5)2 8.4 0.562

Pb-30 Pb(C2H5)4 31.3 0.626

Table 2. Details of additive concentrations used in the present study

Particle measurement and sampling

Sample Aerosol Inlet

Aerosol Charger

Sheath air flow

Grounded Electrode Rings

High voltage electrode

Charged Particle Trajectories

Sizing range 5-2000 nmResponse time ~ 1 s

• Cambustion DMS500* used for particle measurements• Samples taken at engine out, catalyst out, and tailpipe using short sampling lines• Integrated DMS diluter used for engine and catalyst out measurements• Dekati ejector dilutor used for tailpipe measurements• Catalytic stripper used to remove volatiles for solid particle measurements

*Reavell, K., T. Hands, and N. Collings, 2002. “A Fast Response Particulate Spectrometer for Combustion Aerosols,” SAE paper 2002-01-2714Kittelson, David, Tim Hands, Chris Nickolaus, Nick Collings, Ville Niemelä, and Martyn Twigg, 2004. “Mass Correlation of Engine Emissions with Spectral Instruments,” JSAE paper number 20045462

Outline

• Introduction• Experimental• Total particle measurements• Solid particle measurements• Conclusions

Cycle average pre and post catalyst size distributions for high speed EUDC test cycle.

0.00E+00

1.00E+08

2.00E+08

3.00E+08

4.00E+08

5.00E+08

6.00E+08

1 10 100 1000Size (nm)

dN/d

logD

p/cc

Mn-8 Fe-8 Pb-30 Gasoline

• Engine out measurements (pre-catalyst) show large accumulation mode– All three metals increase particle emissions– Iron and manganese show clear nucleation mode

• Post-catalyst measurements show large reductions in accumulation mode region indicating efficient removal of organic carbon by three-way catalyst

• Post-catalyst measurements show a distinct nucleation mode for all three metals

Pre-catalyst

0.00E+00

5.00E+07

1.00E+08

1.50E+08

2.00E+08

2.50E+08

3.00E+08

1 10 100 1000Size (nm)

dN/d

logD

p/cc

Mn-18 Mn-8 Fe-8 Pb-30

Post-catalyst

Cycle average post catalyst and tailpipe size distributions for high speed EUDC test cycle.

• Tailpipe out concentrations roughly a factor of two lower than post catalyst due to diffusion and thermophoresis losses in exhaust system

• Clear, large nucleation mode for Mn and Fe, smaller mode for Pb, no significant mode for standard gasoline

• Why does lead make such a small nucleation mode?

0.00E+00

5.00E+07

1.00E+08

1.50E+08

2.00E+08

2.50E+08

3.00E+08

1 10 100 1000

Size (nm)

dN/d

logD

p/cc

Mn-18 Mn-8 Fe-8 Pb-30

Post-catalyst

0.00E+00

2.00E+07

4.00E+07

6.00E+07

8.00E+07

1.00E+08

1.20E+08

1.40E+08

1.60E+08

1 10 100 1000

Size (nm)

dN/d

logD

p/cc

Mn-18 Mn-8 Fe-8 Pb-30 Gasoline

Tailpipe

Chemical equilibrium calculations suggest lead compounds are more likely to be lost to walls

• Calculations were done using the NASA CEAgui chemical equilibrium program– Gasoline containing sulphur (50 ppm S)– 8.4 and 31.3mg/l iron or lead, respectively– Excess air factor of 0.98– Calculations are based on bulk phase properties, nanoscale and kinetic factors not included

• Iron forms stable solid compound below about 900 C, Mn expected to behave in similar manner• Lead compounds are volatile above about 600 C.

– This temperature is often exceeded at the catalyst inlet– Temperatures at the exhaust valve are even higher

• Volatile compounds are more likely to be lost by mass transfer to the walls of the exhaust system than solid particles because of much higher diffusion coefficients

• Some of the deposited material likely to blow off in larger particles

0.00E+00

1.00E-07

2.00E-07

3.00E-07

4.00E-07

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Temperature (°C)

Mol

e Fr

actio

n FeFeOFe(OH)2Fe3O4(cr)

0.00E+00

1.00E-07

2.00E-07

3.00E-07

4.00E-07

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Temperature (°C)M

ole

Frac

tion

PbPbOPbS(cr)PbSPb(L)

Outline

• Introduction• Experimental• Total particle measurements• Solid particle measurements• Conclusions

A catalytic stripper was used to remove volatile particles with little loss of solid particles

• Recent stripper design– Stripper consists of a 2 substrate catalyst* followed by a cooling coil– The first substrate removes sulfur compounds– The second substrate is an oxidizing catalyst– Diffusion and thermophoretic losses present but well defined

*Catalysts were provided by Johnson-Matthey

Both manganese and iron led to significant solid nanoparticle emissions

• These results are not corrected for diffusion and thermophoresis losses, which are about 60% at 7.5 nm and 30% at 200 nm

• These results show that the nucleation mode consists mainly of solid particles, most likely iron and manganese oxides

• These are the types of particles that have been shown to translocate along the olfactory nervous system to the brain in rats*

0.00E+00

2.00E+07

4.00E+07

6.00E+07

8.00E+07

1.00E+08

1 10 100 1000Size (nm)

dN/d

logD

p/cc

Mn-8 Tail Pipe Average Mn-8 Post Stripper

0.00E+00

2.00E+07

4.00E+07

6.00E+07

8.00E+07

1.00E+08

1 10 100 1000Size (nm)

dN/d

logD

p/cc

Fe-8 Tail Pipe Average Fe-8 Post Stripper

*Translocation of inhaled ultrafine manganese oxide particles on the central nervous system. Alison Elder, Robert Gerlein, Vanessa Silva, Tessa Feikert, Lisa Opanashuk, Janet Carter, Russell Potter, Andrew Maynard, Yasuo Ito, Jacob Finkelstein and Gunter Oberdorster. Environmental Health Perspectives, Volume 114, Number 8, August 2006.

Solid nucleation mode emissions with metal doped gasoline are remarkably similar to idling Diesel

• All measurements made downstream of a catalytic stripper• Diesel results are for a heavy-duty engine without after treatment complying with US 2004

standards– Measurements made with an SMPS– BP50 fuel– Solid ash particles mainly from additives in the lube oil, Ca, Zn, Mg, P

0.00E+00

1.00E+07

2.00E+07

3.00E+07

4.00E+07

5.00E+07

6.00E+07

1 10 100 1000Size (nm)

dN/d

logD

p/cc

Mn-8 EUDC post stripper

Mn-8 50 kph post stripper

Mn-8 Idle post stripper

0.00E+00

1.00E+07

2.00E+07

3.00E+07

4.00E+07

5.00E+07

6.00E+07

1 10 100 1000Size (nm)

dN/d

logD

p/cc

Mode 1 solid

Mode 2 solid

Mode 5 solid

Carbonaceous soot

Ash

Here the results for the EUDC measurements of solid number emissions and the EU number standard

0.0E+00

2.0E+12

4.0E+12

6.0E+12

8.0E+12

1.0E+13

1.2E+13

1.4E+13

1.6E+13

1.8E+13

1.00 10.0 100 1000

Diameter (nm)

Cum

mul

ativ

e N

umbe

r (pa

rtic

le/k

m)

Mn-8Fe-8EU solid number standardEU lower size limit

79% below 23 nm

79% below 23 nm

Outline

• Introduction• Experimental• Total particle measurements• Solid particle measurements• Conclusions

Conclusions

• Mn and Fe additives are used many areas outside the EU, Japan, and US

• Mn and Fe additives resulted in the formation of a distinct nucleation mode– The particles in this mode were nearly all solid, likely metal oxides– Particles in this mode were nearly all below the current lower size limit

of EU number regulations– With Mn-8 and Fe-8

• Total solid number emissions were 25-30 times current EU standard• Only about 20% of the solid particle emissions were in the EU

measurement range above 23 nm– Dosing rates of up to 55 mg/liter used in some countries would lead

much higher emissions

Health – Ultrafine particles are deposited on the deep lung regions more efficiently than fine particles

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

1 10 100 1,000 10,000

Diameter (nm)

Nor

mal

ized

Con

cent

ratio

n (1

/Cto

tal)d

C/d

logD

p

0

0.2

0.4

0.6

0.8

1

Dep

ositi

on

Number Surface Mass Deposition (Alveolar + Tracheo-Bronchial, ICRP 1994)

Fine ParticlesDp < 2.5 µm

Ultrafine ParticlesDp < 100 nm

NanoparticlesDp < 50 nm

PM10Dp < 10 µm

Deposition

Which is the relevant measure of biological impact of particles, number, surface, or mass?

Percent of Neutrophils in BAL 24 hrs after Instillation of TiO2 in Rats

0 500 1000 1500 20000

10

20

30

40

50

ultrafine TiO2 (~20nm)fine TiO2 (~250nm)saline

Correlation with Particle Mass

Particle Mass, μg

% N

eutr

ophi

ls

0 50 100 150 200 2500

10

20

30

40

50

ultrafine TiO2 (~20nm)fine TiO2 (~250nm)saline

Percent of Neutrophils in BAL 24 hrs after Instillation of TiO2 in RatsCorrelation with Particle Surface Area

Particle Surface Area, cm2

% N

eutr

ophi

ls

Oberdorster’s work suggests best correlation of biological impact is with surface area Oberdorster, G., Pulmonary effects of inhaled ultrafine particles, Int. Arch. Occup. Environ. Health 74:1-8 (2001).

Proposed PMP number measurement system

A specially designed CPC with a lower size cutoff of 23 nm is used

Effect of Organometallic Fuel Additives on Nanoparticle Emissions from a Gasoline Passenger CarOutlineIntroductionHealthOutlineVehicle and fuels used in studyParticle measurement and samplingOutlineCycle average pre and post catalyst size distributions for high speed EUDC test cycle.Cycle average post catalyst and tailpipe size distributions for high speed EUDC test cycle.Chemical equilibrium calculations suggest lead compounds are more likely to be lost to wallsOutlineA catalytic stripper was used to remove volatile particles with little loss of solid particlesBoth manganese and iron led to significant solid nanoparticle emissionsSolid nucleation mode emissions with metal doped gasoline are remarkably similar to idling DieselHere the results for the EUDC measurements of solid number emissions and the EU number standardOutlineConclusionsHealth – Ultrafine particles are deposited on the deep lung regions more efficiently than fine particlesWhich is the relevant measure of biological impact of particles, number, surface, or mass?Proposed PMP number measurement system