Atmospheric Aerosol Particles

1
Atmospheric Aerosol Particles

• Erik Swietlicki
• Professor
• Div. of Nuclear Physics, Physics Department
• Lund University Lund Institute of Technology
• Erik.Swietlicki_at_nuclear.lu.se

2
Denver, Colorado, USA
A clear day Mountains visible in
background. A normal day Hazy atmosphere
3
(No Transcript)
4
(No Transcript)
5
NOAA AVHRR Aerosol Optical Depth
6
Global Aerosol Distribution
Aerosol Optical Depth Extinction of visible light
in the atmosphere. Particle concentrations
Ångström Coefficient Wavelength
dependence Particle size information
Low Conc.
High Particle Concentrations
Large Particles
Small Particles
7
Nimbus-7/TOMS Aerosol Index Detects the presence
of UV-absorbing tropospheric aerosols. Biomass
Burning South America and Africa south of the
equator. Saharan Desert Dust Saharan dust plume
blowing into the Atlantic Ocean north of the
equator.
8
TOMS Aerosol Index
9
TOMS Aerosol Index - Saharan Dust Storm
10
TOMS Aerosol Index - Saharan Dust
11
TOMS Aerosol Index South American Biomass Burning
12
TOMS - Mt. Pinatubo Volcanic Eruption 1991
SO2
Aerosol Index
13
Global Aerosol Optical Thickness MODIS/Terra,
NASA Goddard Space Flight Center
14
Aerosol Optical Depth over Europe
C. Robles-Gonzalez, J.P. Veefkind and G. de
Leeuw, GRL 27(2000)955-959
15
PM2.5 Concentrations over Europe
16
EMEP Modelling of PM10 Secondary Inorganic
Aerosol Primary Particulate Matter
Unit ?g/m3 PM10
17
Aerosol - Definition

• A collection of liquid or solid particles
  suspended in a mixture of gases
• - normally air.

An aerosol is a multi-phase system gas - liquid -
solid
18
Size range of aerosol particles

• The criterion of suspension determines the size
  range of aerosol particles

Smallest particle 1 nm (0.001 µm or 10-9
m) Largest particle 100 µm (10-4 m)
Spanning 5 orders of magnitude in size 15 orders
of magnitude in mass/volume !!
19
One litre of urban air ...
contains ca. 10 million particles (104 cm-3)

• We inhale 10-20 m3 of air per day
• ca. 100 billion particles per day

Mass loading in polluted atmospheres ca. 100 µg/
m3 ca. 1 mg/ day
20
Why are Aerosols important?

• They are vital for the
• biogeochemical cycling of many substances
• (e.g. C, N, S, water, nutrients ... )

They are responsible for a number of negative
effect on the environment and human health
21
Atmospheric Aerosols Negative Effects

• Environmental Effects
• (Critical levels of deposition, Air Quality
  Standards)
• Health Effects
• (Ambient Air Quality Standards)

• - Acidification (and eutrophication)
• - Deposition of toxic substances (HM, POP)
• - Soiling
• - Visibility impairment
• - Global radiation balance (Climate)
• - Increased mortality/morbidity
• - Respiratory disorders
• - Carcinogenic/mutagenic
• - Eye/skin irritants

Reducing Emissions Reducing Effects !
22
Atmospheric Aerosols Negative effects
Climate Direct effect (light scattering)
Indirect effect (clouds) Human health
PM2.5/PM10 Carcinogenic Acidification
Long-range tranport Toxic substanses Heavy
metals
23
Aerosol Formation and Sources
24
Atmospheric Aerosols -Formation

• Primary aerosol particles
• Secondary aerosol particles

- Emitted as particles at the source itself
(e.g. soil dust, sea salt, soot)
- Derived from precursor gases emitted at the
source - Formed in atmospheric processes (e.g.
sulphates, nitrates, secondary organics)
25
Particle Size Distributions Particle
Number Particle Surface Area Particle
Volume or Mass
26
(No Transcript)
27
Particle Number Size Distributions
Linear scale
Logarithmic scale
28
Seasonal variation of particle size
distributions Aspvreten background station south
of Stockholm
Bars show 25-75 percentiles around the median
concentration. Log-normal modes Nucleation,
Aitken and Accumulation modes
29
Particle Volume and Mass Size Distributions
Continental environments
Urban Environments
Urban influenced rural
Near major road
Soil dust
Clean Rural Background
Urban influence
30
Aerosol Particles on a Filter

• PM is very variable.
• What is relevant?
• (mass, number, composition?)
• Tricky to measure and control.

31
Sources of AerosolMeasurement Errors
Original aerosol
Sampling Efficieny
Internal Losses
Aerosol Sampler
Sensor Response
234 cm-3
Data Handling
32
Urban Aerosols

• Number
• Most particles smaller than 0.1 µm
• Surface
• Most surface area between 0.1 - 0.5 µm
• (condensation, coagulation most efficient)
• Mass / Volume
• Bimodal - Fine (lt1 µm) and Coarse
• (roughly equal mass in both modes)

33
Aerosol Particle Separation - Conventions
IPM Inhalable particle fraction (fraction
inhaled through nose and mouth) TPM Thoracic
particle fraction (fraction passing the
larynx) RPM Respirable particle fraction
(fraction reaching the alveoli)
34
Atmospheric Aerosols - Nomenclature

• TSP (Total Suspended Particulate matter)
• PM (Particulate Matter)
• PM-10 Particles lt 10 µm in diameter
• PM-2.5 Particles lt 2.5 µm in diameter
• PM-1 Particles lt 1 µm in diameter
• PM-2.5/PM-10 ratio ? 0.5 (variable)
• Old standard Black Smoke BS
• BS/PM-10 normallylt 1

35
Continuous Aerosol Mass Measurement TEOM

• Measures Aerosol Mass
• Continuous Measurement
• On-line, automatic
• High Sensitivity
• Commercially available
• Various inlets (PM10/PM2.5)
• Expensive!

36
Examples of non-spherical particles
TEM (Transmission Electron Microscopy) pictures
Kerosene lamp (soot agglomerate)
Particle from tire wear
37
Equivalent Particle DiameterRelates to the
sedimentation velocity vTS
Aerodynamic equivalent sphere dae 8.6 µm ?p 1
g/cm3
Volume Equivalent Diameter Shape Factor 1.36 de
5.0 µm ?p 4 g/cm3
Stokes equivalent sphere ds 4.3 µm ?p 4 g/cm3
vTS0.22 cm/s
vTS0.22 cm/s
vTS0.22 cm/s
38
(No Transcript)
39
Number size distributions in Stockholm (various
locations from surburb to busy downtown street
canyon)
Particle Diameter (nm)
40
Volume size distributions in Stockholm (various
locations from surburb to busy downtown street
canyon)
Particle Diameter (nm)
Christer Johansson m.fl., ITM, Stockholm
41
Traffic tunnel in Stockholm SöderledstunnelPar
ticle Number Size Distributions (3-850 nm, 24
hours)
Particle Diameter (nm)
Time of day (h)
Most particles (tail-pipe emissions) are ? 20 nm
42
Differential Mobility Particle SizerDMPS
Aerosol-partiklar
Partikel- räknare
Mäter Partiklarnas storleksfördelning (lt 1 µm,
oftast torrt)
CPC
Monodispers aerosol
dN/dDp
Tork
DMA
Bipolär laddare
Dp
d0
DMA Transfer Function
43
Estimated Ultrafine Particle Concentrations in
Stockholm
Lars Gidhagen SMHI and Stockholm University
Includes aerosol dynamics (MONO32)
Ultrafine Particles Dp lt 100 nm
(Unit /cm3)
44
Number Size Distributions LDV and HDV
LDV
HDV
Mixed Fleet
Tunnel Data Kristensson et al., 2004. 10-20 times
higher emission factors for the Heavy Duty
Vehicles (HDV) compared to the Light Duty
Vehicles (LDV)
45
Traffic Emission Factors (real-world)Particle
Size Distributions
Inverse modelling (Street canyon) Copenhagen
Tunnel Study Stockholm
On-Road Study LIPIKA Helsinki
46
Traffic Emission Factors (real-world)Total
Particle Number
47
Soot Formation during Combustion
Primary soot particles 30 nm C8H
48
Road Simulator VTI, Linköping, Sweden
49
Emission Factors for Particles Originating from
the Tire-Road Interface (Road Simulator)
50 km/h (15-725 nm)
6?1011 particles/(vehicle?km)
Box Model Fit
50
TEM Images of Submicrometer-Sized Particles
Originating from Wear of the Tire-Road Interface
(Road Simulator)
51
Street Canyon Plume Dispersion
Concentrations are highest on the leeward side of
the street
Background
52
Emissionsfaktorer från modellering och mätningar

• En andra metod Numeriska modeller (CFD).
• Försök från vattenkanal och CFD-modell (Caton et
  al., 2003)

53
Development and validation of a model for
estimating PM10 in urban environmentsFinanced
by the Swedish National Road Administration
(Vägverket)Site Vallstanäs (E4 north of
Stockholm)
54
Model Comparison
55
PM10 Emission Factors Estimated from
Source-Receptor Modelling
56
Development and Validation of a Model for
Estimating Urban PM10
57
Source-receptor modelling VavihillNatural
sources (sea, soil) contribute significantly to
particle mass
58
Number Size Distribution (Cold period)
Lycksele, Jan-March 2002
Wood Combustion (Residential area)
Traffic (centre)
Particle Diameter (nm)
59
Number size distributions (in the chamber) during
different kinds of wood pellets combustion
Semi-poor comb.
Good (optimized) comb.
Poor comb.
60
Exchange between phases
HNO3
NH3
Organic compounds with low vapour pressures
NO3-
NH4
HCl
Cl-
OC
H
H2O
SO4
H2O
SO2
61
Vapour Pressures of Organic Compounds
62
High boiling temp. ? High aerosol Fraction Less
in gas phase
Various PAH
63
Ground-level ozone Photochemical smog
In polluted environments (e.g. large cities),
concentrations of hydrocarbons, nitrogen oxides,
ozone and aerosol particles often follow a
certain diurnal pattern.
Hydrocarbons
Oxidized and nitrated hydrocarbons
Aerosol particles

NO
NO2
O3
64
Urban Aerosols

• Fine particle mode
• Combustion processes
• Gas-to-particle conversion
• Coarse particle mode
• Mechanical disintegration processes
• Traffic - soil dust, tire wear
• Incomplete combustion - fly ash
• Bioaerosols - plant debris, bacteria, fungi...
• Sea salt

65
Aerosol Size Distributions Urban Roof-top
Malmö, Sweden
April 2005 April 2006 Average Size Distributions
Workdays
?50 nm
National holidays have been omitted.
66
Measurements sites Southern Sweden
55? 36' 23'' N, 13? 0' 9'' E
67
Urban Roof-top Regional Background Malmö
Vavihill
Urban site (Malmö)
Backgrund (Vavihill)
68
Ship Traffic Contribution
69
Ship plumes from Malmö City Harbour
SO2
70
Processes affecting the Atmospheric Aerosol
Raes et al. Atm. Env. 34(2000)4215
71
Modelling Urban Aerosol Dynamics
Regional background, Urban Traffic Emissions and
Dilution, No Aerosol Removal process
5 h
3 h
1 h
0 h
72
Modelling Urban Aerosol Dynamics
Regional background, Traffic Emissions, Dilution,
Coagulation, Deposition with u1.33 m/s, No
condensation
Regional background, Urban Traffic Emissions and
Dilution, No Aerosol Removal process
5 h
3 h
1 h
0 h
73
Modelling Urban Aerosol Dynamics
Regional background, Traffic Emissions, Dilution,
Coagulation Deposition with u1.33 m/s, No
condensation
Regional background, Traffic Emissions, Dilution,
Coagulation, Deposition Condensation, Growth
rate6 nm/h, Condensing vapour conc. 1.85 x 108
molec cm-3
Regional background, Urban Traffic Emissions and
Dilution, No Aerosol Removal process
74
Modelling of the urban background
75
Gas phase chemical reactions
Hot vapours
Low volatility gases
Homogenous nucleation and condensation
Condensation
Wind blown dust Emissions Sea
spray Volcanoes Plant Particles
Condensation nuclei
Primary particles
Droplets
Coagu- lation
Agglomerates
Coagulation
Activation
Coagu- lation
Rainout and Washout
Diffusion
Sedimentation
Coarse Particles
Aitken mode
Accumulation mode
Nucleation mode
76
Nucleation
Homogeneous homomolecular Self-nucleation of a
single species. No foreign nuclei or surfaces
involved. Homogeneous heteromomolecular Self-nu
cleation of two or more species. No foreign
nuclei or surfaces involved. Heterogeneous
homomolecular Nucleation of a single species on a
foreign substance. Heterogeneous
heteromomolecular Nucleation of two or more
species on a foreign substance.
77
Particle Number Size DistributionsVavihill
(south Sweden)
78
Atmospheric Aerosol Processes Effect on
Concentrations
Particle Number
Sea Spray
Traffic (tail-pipe)
Homogeneous Nucleation
Primary Emissions
Original Concentrations
Soil Dust
Dry Deposition
Condensation
Coagulation
Wet Deposition
Particle Mass
79
Jet and Film drops form large and small sea salt
particles respectively
80
Bubble Bursting and Sea Spray Production Jet
Droplets
Double-click to start moviesLargest Source of
Aerosol Mass Globally
81
Sea Salt Particle Production
82
Sea Salt Particle Production Upward Particle
Number Fluxes vs. Wind Speed Dependence
Nilsson et al. JGR 2001 (AOE96)

• Eddy Correlation technique
• Good correlation between particle number flux and
  local wind speed!
• Open Sea
• Leads in Pack-Ice
• Both

83
Particle Number Concentrations Wind Speed
Dependence
Leck et al. JGR 2001 (AOE96)
84
Sea Salt Particle Production (kg km-2 hr-1)
85
Soil dust emissions(kg km-2 hr-1)
86
Anthropogenic Black Carbon Emissions(kg km-2
hr-1)
87
Suplhate Aerosol Sources(kg km-2 hr-1)
Fossil fuel burning, , oxidation of anthropogenic
SO2
DMS Oxidation (phytoplankton)
88
Organic aerosol sources(kg km-2 hr-1)
Biomass burning, fossil fuel, oxidation of
anthropogenic VOC
Oxidation of biogenic VOC
89
Estimated Global Emission Rates of Particles into
the Atmosphere (Tg yr-1) Raes et al. Atm. Env.
34(2000)4215
90
Ambient Aerosols Summary
http//www.epa.gov/ncea/pdfs/partmatt/VOL_I_AQCD_P
M_3rd_Review_Draft.pdf
91
(No Transcript)
92
Aerosols and Health
93
Health Effects of Aerosols
Aerosol Dynamics, Dispersion Modelling
Dispersion Modelling, Indoor-Outdoor
Aerosol Characterization
Emission Factors
Lung Deposition
Emission
Exposure
Dose
Health effect
Concentration
Epidemiological studies
94
Photographer Lennart Nilsson
Soot particles (yellow) deposited in the alveoli.
95
A macrophage attacks the soot particle and tries
to engulf it.
Photographer Lennart Nilsson
96
WHO - Health Effects of PM

• WHO Health Risk Estimations for Particulate
  Matter

97
Short Term Mortality
Change in Daily Mortality
Estimated per cent changes in daily mortality
associated with a 10 mg/m3 increase in PM10 (with
95 confidence intervals) for a number of cities.
(Pope et al. 1995)
98
Health Effects from Short Term PM10
ExposureCompilation of various studies
Percent change (increased risk) for a
PM10-concentration increase (day to day) of 10
?g/m3
99
Health Effects from Chronic PM Exposure
American Cancer Society Study, (Pope et al. 1995)
Harvard Six Cities Study, (Dockery et al. 1993)
Both figures show how the risk for mortality
increases in relation to PM2.5-concentrations
taken as an average over several years (chronic
exposure).

• PM2.5 gives a stronger relationship than PM10
  (exposure-health effect)

100
Atmospheric Aerosols -Air Quality Guidelines

• USA - National Ambient Air Quality Standards
• Annual Arithmetic Mean 24-h Average
• PM-2.5 (1997) 15 µg/m3 65 µg/m3
• PM-10 (1997) 50 µg/m3 150 µg/m3
• (http//www.epa.gov/airs/criteria.html)
• EU countries - Air Quality Limit Values
• Annual Arithmetic Mean 24-h Average
• PM-10 (2005) 40 µg/m3 50 µg/m3 (35/yr)
• PM-10 (2010) 20 µg/m3 50 µg/m3 (7/yr)

101
European PM10 Concentrations
102
European PM2.5 Concentrations
103
European PM Concentrations
PM2.5 and PM10 correlated PM2.5 / PM10 ratio
0.5 to 0.9
104
European PM Concentrations
PM2.5 and PM10 correlated PM2.5 / PM10 ratio
0.5 to 0.9
105
European PM10 Composition
106
European Coarse Mode CompositionPM10-PM2.5
107
European PM2.5 Composition
108
ICRP Deposition Model International Commission
on Radiological Protection
109
Particle hygroscopic properties Importance for
deposition in the lungs
Hygroscopic particles shift the minimum in the
deposition curve to smaller sizes.
Lung Deposition and Hygroscopic Growth (at
RH99.5)

• Hygroscopic particles affect deposition
• More particle mass (gt200 nm) is deposited in the
  upper airways.
• Fewer very small (lt100 nm) particles are
  deposited in the lower airways (number).

Hygroscopic Growth Factor (at RH99.5)
Deposited Fraction
Deposition Humidified Deposition Dry Growth
Factor
Deposition increases
Deposition decreases
Dry Particle Diameter (nm)
110
Particle hygroscopic propertiesLung deposition
(Forsdala)
Number Deposited Number Surface Deposited
Surface Volume Deposited Volume
Dry Particle Diameter (nm)
111
Soluble particles
The response from the body depends on the
particle mass, composition and number
The particles lose their original shape and
physical properties after deposition
Number of deposited particles can affect the
physiological response
Epitelial cell in alveoli
112
Insoluble particles
The response from the body depends on the
particle surface properties and number
The particles keep their original shape and
physical properties even after deposition
Number of deposited particles can affect the
physiological response
Epitelial cell in alveoli
113
Insoluble particles may enter the blood
Epithelial lining fluid
Alveolar epithelium
Blood vessel
114
Aerosols and Climate
115
Cloud droplets
CCN
Short-wave solar radiation
Gases
Particles
Gases
Particles
Biota
Human Activities
Biota
116
The Earths Annual and Global Mean Energy Balance
31 Reflection
69
100
22
48
12
20
Absorption 69
102
9
95
23
7
49
114
Kiehl and Trenberth, 1997 Earths Annual Global
Mean Energy Budget, Bull. Am. Met. Soc. 78,
197-208.
117
UN Intergovernmental Panel for Climate Change
(IPCC) The Global Mean Radiative Forcing of the
Climate System 1750 to 2000
Greenhouse gases
Aerosol particles
Radiative Forcing (W/m2) Cooling
Heating

Low
High
Medium
?--------------- Very Low ------------?
Level of Scientific Understanding
118
Direct aerosol effect
Aerosol plume extending from Himalayas to Indian
Ocean MODIS true-color Satellite Image
(29/04/02) (From http//visibleearth.nasa.gov)
119
Radiative forcing by aerosol types
Dust warms and cools since it is also absorbing,
even at visible ?. Estimates for forcing 0.09
W m2 to - 0.46 W m2
Fossil fuel Soot carbon forcing is 1.6 W m2 for
external mix with sulphate and 0.42 W m2 for an
internal mix (organic lt -0.04 W m2).
Biomass burning soot/organic carbon forcing is
1.6 W m2 for external mix with sulphate and
0.42 W m2 for an internal mix.
120
RH Hysteresis Effect
121
Every cloud drop contains at least one aerosol
particle.
Humidified particle RH90
Dry particle
Cloud drop RHgt100
Salt
Water solution
The more water-soluble material the particle
contains, the more likely it is that it will form
a cloud drop (at RH gt 100).
122
Köhler theory for cloud droplet formation
Raoults term
Kelvin term
Kelvin term
Raoults term
123
Indirect Aerosol Radiative Forcing (Cloud effects)
Ship tracks Increased albedo of stratocumulus
clouds formed in plumes with elevated particle
concentrations in an otherwise clean marine air
mass. Evidence of the first indirect (Twomey)
effect.
124
Indirect aerosol effect
Ship tracks over the North Atlantic MODIS
true-color Satellite Image (29/04/02) (From
http//visibleearth.nasa.gov)
125
Indirect aerosol effect
Clean
Polluted
126
(No Transcript)
127
Biomass burning perturb regional and global
climate
The Amazon, BrazilExtensive biomass burning
during the dry period (September-November)
Fires in South America during the dry period 2001
Pyrocumulus cloud
Rain forest recently burnt(Rondonia, Brazil,
Oct. 2002)
128
(No Transcript)
129
Land Use and Cover Change in Amazonia
Rondônia 1986
Rondônia 1992
Local changes Global effects?
Rondônia 1996
130
Hypothesis Interaction betweenAerosols Clouds
- Precipitation
Input of more aerosol particles from
anthropogenic sources lead to smaller and more
numerous cloud drops. Coalescence is
suppressed, preventing warm precipitation to
form.Clouds will rise to higher altitudes, were
freezing of cloud drops occurs.More latent heat
will be transported to high altitudes.Vigorous
thunderstorms will form (lightning, heavy rain).
131
The Amazon, BrazilDry period September-November
2002
More than ten-fold increase in the cloud droplet
number due to biomass burning.
132
Acidification
133
Aqueous Phase Oxidation of Sulphur

• S(IV) is oxidized in aqueous solution
• Via hydrogen peroxide H2O2
• HSO3- H2O2(aq) H ?
• 2H SO42- H2O
• Via ozone O3
• S(IV) O3(aq) ?
• S(VI) 2H O2
• Catalyzed by transition metals
• (e.g. Mn, Fe) at night, winter-time.
• (?SO2 1 week).

134
Emissions of sulphur dioxide 1997 (EMEP)
Unit tonnes of SO2
135
EMEP Eulerian Acid Deposition model - Sulphur
(1997)
SO2
SO42-
Sulphur deposition to Sweden
Sulphur deposition
136
Critical load Acidification, eutrophication
Definition (Nilsson and Grennfelt, 1988) The
threshold below which significant harmful effects
on specified sensitive elements of the
environment do not occur according to present
knowledge is called the critical load. Critical
load (acidification, eutrophication) Represents
a deposition that is sutainable in the long
term. Can be expressed in various ways. E.g.
2-percentile for critical load is the deposition
of acidifying (or eutrophying) compounds at which
98 of all ecosystems are protected in the long
term.
137
Critical Load
The critical load for acidification given as
the 2- percentile (protects 98 of all
ecosystems) (equivalents / ha / year) Nordic
soils are very sensitive. Mediterranean soils are
quite insensitive to acid deposition.
138
Exceedance of Critical Load
(Unit of ecosystems, Scale 0-100)
For emissions according to the 1999 Gothenburg
protocol
For emissions 1990