Urban and Rural Sources of Particulate Matter

John Wenger

Centre for Research into Atmospheric ChemistryEnvironmental Research Institute

University College CorkIreland

Email: j.wenger@ucc.ie web: http://www.ucc.ie/en/crac/

Outline

• Properties of Particulate Matter (PM)

• Linking chemical composition and sources of PM

• Case Studies- Cork City- Killarney

• Summary and Perspectives

Particulate MatterPM10 - Particulate Matter with diameter less than 10 microns

PM2.5 - Particulate Matter with diameter less than 2.5 microns

All Shapes and Sizes

• Large number of particles < 0.1 microns• Majority of mass in range 0.1-10 microns

Natural AnthropogenicPrimary Sources

• Formation and growth of particles in the atmosphere

• Ammonium, sulfate, nitrate, secondary organic aerosol

Gas +scavenging

Particle Particle

Particle Particle+coagulation

Particle

Gas Gas+nucleation

Particle

Gas condensationParticle

Secondary Sources

Fine fraction (PM2.5) Coarse fraction (PM2.5-PM10)

• Approximate composition of PM in Ireland determined by off-line analysis of filter samples

Chemical Composition

Elemental andOrganic CarbonSulphate

Nitrate

Ammonium

Chloride

Insoluble minerals

Na, K, Mg, Ca

Chemical species SourcesElemental/Black Carbon (EC or BC)

Fuel combustion (automobiles,industry, coal/wood burning)

Organic Carbon Fuel combustion, secondary organic aerosols from VOC oxidation processes

Nitrate / Sulfate Gas-particle conversion of NO2 / SO2produced from combustion processes

Ammonium Gas-particle conversion of NH3produced from agriculture

Chloride Sea sprayMinerals (Oxides of Ca, Mg, Si, Al, Fe)

Resuspension of dust/soil

Metals (K, V and Ni, Pb, Zn, Cd, Hg etc.)

Industry, combustion, often specific sources

Linking Composition and Sources of PM

Reducing PM levelsWe need to know AND quantify the sources

• How much PM is from traffic?• How much PM is from solid fuel burning?• How much PM is from other sources?• How do the emissions from these sources vary

during the day and by season?

Detailed measurements of the PM are required

• Size, concentration and chemical composition at a HIGH-TIME resolution

• Source Apportionment Modelling

Case Study: Cork Harbour

• Long-term (1 year) monitoring campaigns

• Intensive (1 month) measurement campaigns

XBB B

Tivoli Docks August 2008 and February 2009

Intensive Measurement Campaign

Healy et al., Atmospheric Chemistry and Physics 2010

Tivoli Docks August 2008 and February 2009

Intensive Measurement Campaign

A range of state-of-the-art instruments deployed for On-line monitoring of particle mass, size, number and chemical composition in real-time

Aerosol Time-of-Flight Mass Spectrometer• Detects elemental

carbon, organic carbon, metals, inorganic ions in single particles

• Provides size-resolved chemical composition(0.1-3.0 micron)

• Operates in real-time → big advantage over filter collection and off-line analysis approach

Sea-Salt Particle Mass Spectrum (Na and Cl are markers of interest)

Biomass Burning Particle Mass Spectrum (K is major marker for biomass)

Single Particle Mass Spectra

Sources in Cork Harbour: 3 Vehicular Traffic

0

500

1000

1500

2000

2500

ATO

FMS

coun

ts (h

-1)

Time

Ca-traffic EC-traffic EC-phos

Sources in Cork Harbour: Vehicular Traffic

Healy et al., Atmospheric Chemistry and Physics 2010

0

50

100

150

200

250

0

100

200

300

400

500

600

700

800

ATOFM

S counts (h- 1)ATO

FMS

coun

ts (h

-1)

Time

coal peat wood

Sources in Cork Harbour: Solid Fuel Combustion

Healy et al., Atmospheric Chemistry and Physics 2010

• State-of-the-art analytical techniques used to apportion PM mass

Source Apportionment of PM

Healy et al., Atmospheric Chemistry and Physics, 2010

PM2.5average (µg/m3)

Solid FuelBurning

%

Traffic %

Other Local

Sources %

Regional Sources

%

August 2008

9.7 5 23 24 26

February 2009

16.2 50 19 21 10

Kourtchev et al., Science of the Total Environment, 2011

Dall’Osto et al., Atmospheric Chemistry and Physics, 2013

Extent of Bituminous Coal Ban 2015

• What is the contribution of residential solid fuel burningto PM levels in towns where the Ban on Bituminous Coalis not in place?

The Burning Question

What is the contribution of each fuel type?

Sod Peat (Turf)

“Smokeless” Coal

Wood

Bituminous (Smoky) Coal

Peat Briquettes

Source Apportionment of Particulate Matter in

Urban and Rural Residential Areas of Ireland

(SAPPHIRE)

1 April 2014 – 31 March 2016

http://www.ucc.ie/en/crac/research/sapphire/

• Outside the Smoky CoalBan Area (pop. < 15,000)

• No natural gas supply

• High usage of solid fuels(coal, peat/turf & wood)

Monitoring Locations

• Killarney, Co. Kerry (Nov & Dec 2014)• Enniscorthy, Co. Wexford (Jan & Feb 2015)

K

E

• Site is located on the western side of the town, in the grounds of the Community Hospital in a residential area

Monitoring Location: Killarney

• Site is located on the western side of the town, in the grounds of the Community Hospital in a residential area

Monitoring Location: Killarney

TEOM

PM2.5 mass concentration

• PM2.5 up to 10 times higher during evening hours

PM2.5 mass concentration

• Strong diurnal pattern

Aerosol Time-of-Flight Mass Spectrometer• Detects elemental

carbon, organic carbon, metals, inorganic ions in single particles

• Provides size-resolved chemical composition(0.1-3.0 micron)

• Operates in real-time → big advantage over filter collection and off-line analysis approach

PEAT

PEAT

WOOD

COAL

COAL

ECSulfate

Potassium

Assigned on the basis of combustion experiments COAL → EC & some potassium, sulfate dominates negative spectraPEAT → EC & OC fragments, some potassiumWOOD → Potassium dominates positive spectra

EC

OC

WOOD

Mass Spectra: Solid Fuel Combustion

SEA SALT

TRAFFIC

AMINE/AMMONIUM

NaCl

NaCl2Na2Cl3 Sea salt characteristics:

→ sodium & chloride peaks, no EC

Traffic characteristics: → calcium & phosphate (lubricating oil), some EC

Phosphate

Calcium

Ammonium/amine characteristics:→ ammonium, trimethylamine, OC, large sulfate peak in negative spectra

Ammonium

Mass Spectra: Other Particle Types

Transportedsea salt

ATOFMS Particle Number

• Low wind speed – local emissions dominate• High wind speed – regional sources dominate

Particles from solid fuel burning80% of PM2.5

Particle Numbers

Particles from solid fuel burning77% of PM2.5

Particle Mass

ATOFMS: Source Contribution to PM2.5

Mass

Scaling

• Local sources account for 70-90% of PM2.5 in Cork City. Traffic accounts for ~20%; solid fuel burning 50% in winter.

• Residential solid fuel burning contributes 70-80% of PM2.5in Killarney in winter

• Similar results for Enniscorthy: also likely replicated in tens of small towns across Ireland.

• Peat, coal and wood all contribute: Extending the smoky coal ban may not be enough to deliver improvements in air quality

• No source apportionment study yet performed in Dublin!

Summary and Perspectives

Acknowledgements

John Sodeau

Ian O’Connor Eoin McGillicuddy Jovanna Arndt

Paul BuckleyStig Hellebust

Extra Slides

• Missing mass due to regional sources – organic aerosol, ammonium sulfate?

ATOFMS Particle Mass vs TEOM

Particles directly emitted from solid fuel combustion = 66% of measured PM2.5

Source contributions (% of TEOM PM2.5)

Killarney: ATOFMS Mass: Diurnal

• Particle numbers for entire sampling period averaged to 1 day → clear evening peak shows influence of solid fuel burning on total particle numbers.

• Clear evening peak in averaged mass concentration: ‒ peat (16 μg/m3)‒ wood (12 μg/m3)‒ coal (10 μg/m3)

(combined 38 μg/m3 average per night)

29/09/15 SAPPHIRE Meeting: ATOFMS (Jovanna Arndt) 37

Unidentified30%

Marine and Aged

6%Local Aged

4%Traffic3%Amines

5%

Burning52%

Preliminary Source Apportionment

• 5 factors identified• Primary emissions from

solid fuel burning = 52% of PM2.5 …….but no separation by fuel type

• PMF ME-2 using ATOFMS particle classes, EC-OC, SMPS, OPS, NOx, Aethalometer

Instrument Parameter(s) measured Temporal resolution

Aerosol time-of-flight mass spectrometer (TSI model 3800)

Single particle chemical composition (100-3000 nm)

1 min

Scanning mobility particle sizer (TSI model 3081)

Particle number concentration (10-800 nm)

3 min

Optical Particle Sizer (TSI model 3330)

Particle number concentration (300-10000 nm)

3 min

TEOM (Thermo Electron model RP 1400a)

PM2.5 mass concentration 30 min

Thermal-optical carbon analyser (Sunset Inc. model 3rd generation)

Elemental and organic carbon mass concentrations

2 hr

7-Wavelength Aethalometer(Model AE33, Magee Scientific)

Black Carbon concentration 1 min

High volume sampler (Digitel model DHA 80)

Collection of particulate matter (PM2.5)

6 hr

Key Instrumentation

TEOM Sunset ECOC

Elemental and Organic Carbon (EC/OC)

• Majority of PM2.5 during night-time pollution events is carbonaceous aerosol

Low winds – Local sources High winds – Regional sources

Influence of Meteorology

• Low wind speed – local emissions dominate• High wind speed – regional sources dominate

Enniscorthy: ATOFMS Mass: Diurnal

• Averaged ATOFMS mass concentration per day maximum = ~ 60 μg/m3

• Peak averaged daily mass concentrations: − peat (16 μg/m3)− wood (14 μg/m3)− coal (12 μg/m3)− PAH-containing (9 μg/m3)

29/09/15 SAPPHIRE Meeting: ATOFMS (Jovanna Arndt) 42

Enniscorthy: ATOFMS Mass Breakdown

29/09/15 SAPPHIRE Meeting: ATOFMS (Jovanna Arndt) 43

Solid fuel combustion particles = 89% of ATOFMS PM2.5 mass

(no comparison with TEOM yet so this % will probably decrease a bit)

PAH-containing particles associated with all three fuel types.