image
image
image
image
image
image
image
image
image
image
Ultrafine Particles in Urban Air
Dresden 23 to 24/10/2007
Saxon State Agency for Environment and Geology

image
image
UFIPOLNET LIFE04 ENV/DE/000054
Proceedings of UFIPOLNET Final Conference: Ultrafine Particles in Urban Air, Dresden 23 to 24/Oct/2007
Editor:
Dr. Holger Gerwig
Project manager of EU-LIFE Project UFIPOLNET
Saxon State Agency for Environment and Geology
Department 2 - Integrative Environmental Protection, Climate, Air, Radiation
22 Regional Air Quality
Zur Wetterwarte , 01109 Dresden
Holger.Gerwig@smul.sachsen.de
www.ufipolnet.eu
Printed on 100% recycled paper

image
image
UFIPOLNET LIFE04 ENV/DE/000054
Lectures
The Role of Ultrafine Particles in the Urban Atmos-
phere
A. Wiedensohler*, W. Birmili, and B. Wehner
8
Ultrafine Particles Chemical Content
H. Herrmann, E. Brüggemann, T. Gnauk, K. Müller, Y. Iinuma, K.
Beiner, L Poulain
9
UFIPOLNET + LIFE
LIFE and LIFE+ - from program to projects
D. Palenberg
10
UFIPOLNET: Purpose - Partner - Project
H. Gerwig*, G. Löschau, L. Hillemann, B. Wehner, A. Wiedesoh-
ler, A. Zschoppe, C. Peters, A. Rudolph, C. Johansson, J. Cyrys,
M. Pitz, R. Rückerl, J. Novak, H.G. Horn, R. Caldow, G.J. Sem
11
A new particle measurement system for environ-
mental ultrafine particles
A. Zschoppe* L. Hillemann A. Rudolph C. Peters, R. Caldow
12
Characterization of environmental ultrafine parti-
cles with the UFP 330, system calibration and
evaluation
L. Hillemann*, A. Zschoppe, R. Caldow
13
Ultrafine particles: Comparisons UFP 330 DMPS
B. Wehner, T. Tuch, A. Wiedensohler*, A. Zschoppe, L. Hille-
mann, G. Löschau, H. Gerwig
14
Measurements of Ultrafine Particles in Urban
air Part I
Measurements of ultra-fine particles in Europe:
differences and similarities.
R. van Dingenen*, J.P. Putaud, D. Mira-Salama, N.R. Jensen, F.
Raes
15
Origin and features of ultrafine particles in Barce-
lona
S. Rodríguez*, J. Pey, N. Perez, X. Querol, A. Alastuey, R. Van
Dingenen and J.P. Putaud
16
Ultrafine particles in the UK
A. Jones*, R M Harrison
18
Ultrafine particles in Stockholm
C. Johansson, M. Norman, H. Karlsson
20
Ultrafine particles in Prague
J. Novak
22
Particles and Health
Overview: Health and Particles - the Epidemiologic
View
A. Peters
23
Updated WHO Air Quality Guidelines
M. Krzyzanowski
24
Can we use fixed ambient air monitors to estimate
population exposure to ultrafine particles?
J. Cyrys, M. Pitz, J. Heinrich, R. Rückerl and A. Peters
25
(Fine) Particles in Saxony - Exposure and Health
Aspects
U. Franck*, S. Odeh, W.-H. Storch, Th. Tuch, A. Wiedensohler,
B. Wehner, and O. Herbarth
27
Health and Particles: Regulatory Aspects
N. Englert
28
Health effects of inhaled ultrafine particles in the
lungs and other secondary target organs like brain
and heart.
W. G. Kreyling*, M. Semmler-Behnke
30
3

image
image
UFIPOLNET LIFE04 ENV/DE/000054
Modelling Fine and Ultrafine Particles
Ultrafine Particle (UFP) Measurements and Model-
ling
W. Birmili
31
Modelling ultrafine particles in urban environments
M. Ketzel
32
Modelling PMx-Emissions and -Concentrations on
streets for environmental impact assessment and
action plans
I. Düring
34
Measurements of Ultrafine Particles in Urban
air II
Five years ultrafine and fine ambient particles
number concentrations measurements at a traffic-
orientated site in Dresden
G. Löschau*, B. Wehner, A. Wiedensohler
36
Ultrafine particles in NRW - case studies in the
urban background and at an "Autobahn"
T. Kuhlbusch, A. John, U. Quass*
38
Ultrafine and fine particle measurements in Swit-
zerland at various stations and on different roads
A.S.H. Prévôt, E. Weingartner, S. Weimer, J. Sandradewi, M.R.
Alfarra, V. Lanz, C. Hueglin, S. Szidat, U. Baltensperger
40
Temporal and spatial variability of sub-μm aerosol
concentrations in the urban atmosphere of Leipzig
W. Birmili, S. Klose, M. Merkel, B. Wehner, K. König, A. Sonn-
tag, A. Wiedensohler, O. Knoth, D. Hinneburg, Th. Tuch, U.
Franck
41
Chemical composition of aerosol particles including
UFPs in Saxony
T. Gnauk*, E. Brüggemann, H. Gerwig, H. Herrmann, K. Müller,
G. Spindler
43
Air Quality in Directives and their Implementa-
tion
PM abatement from a European perspective - cur-
rent legislation and the CAFE Thematic strategy
P. Bruckmann
44
LIFE-ENVIRONMENT Project KAPA GS Klagen-
furts Anti PM10 Action Programme with Graz and
the South Tyrol
W. Hafner , C. Kurz, G. Bachler, P. Sturm
45
The PM and NOx air pollution in Copenhagen and
assessment of possible measures to reduce the air
pollution
F. Palmgren
47
Implementation of air quality directives in a candi-
date state
D. Gömer
49
Street-Detailed Calculation Of The Air Quality In
Saxony
U. Wolf
51
4

image
image
UFIPOLNET LIFE04 ENV/DE/000054
Poster
Particle number concentration in the urban area of Rome
P. Avino, S. Casciardi, C. Fanizza, M. Manigrasso
54
Using a sampling and monitoring device, a solution for
PM10-PM2.5 assessment? Experience with the Swam 5-A
L. Bertrand, G. Gérard and S. Fays
55
Measurement and analysis in space and time of ultrafine
particle number concentration in ambient air. The case of
Parma
F. Costabile and I. Allegrini
56
Experiences with ultra fine particle monitoring in air quality
monitoring networks in Europe
C. Gerhart, T. Petry, T. Rettenmoser, A. Kranapeter,
HP.Lötscher
58
UFIPOLNET: Concentration of Particle Number Distributions
at 4 Stations in Europe
H. Gerwig, G. Löschau, L. Hillemann, B. Wehner, A.
Wiedensohler, A. Zschoppe, C. Peters, A. Rudolph,
C. Johansson, J. Cyrys, M. Pitz, R. Rückerl, J. No-
vak, H.G. Horn, R. Caldow, G.J. Sem
59
Determination of particle emission factors of individual vehi-
cles under real-life conditions
C.S. Hak, E. Ljungström, M. Hallquist, M. Svane and
J.B.C. Pettersson
60
Particle number size distributions of ambient-state and non-
volatile aerosols in the city of Augsburg, Germany
K. Heinke, W. Birmili, A. Wiedensohler, M. Pitz, J.
Cyrys and A. Peters
62
Determination of the charge distribution of highly charged
aerosols
L. Hillemann, M. Stintz, C. Helsper
63
Quantification of nanoparticle releases from surfaces
L. Hillemann, M.Stintz, M. Heinemann
64
Aerosol mobility spectrometry based on diffusion charging
L. Hillemann, A. Zschoppe and R. Caldow
65
An analysis of traffic-induced particle number emissions
based on long lasting roadside and urban background meas-
urements
S. Klose, W. Birmili, T. Tuch, B. Wehner, A. Wieden-
sohler, U. Franck, M. Ketzel
66
Microscale variations of atmospheric particle number size
distributions in a densely built-up city area
M. Merkel, W. Birmili, A. Wiedensohler, D. Hinne-
burg, O. Knoth, T. Tuch and U. Franck
67
Continuous observations of particle size distributions (wide
range) at the Frohnau Tower in Berlin with an altitude of 320
m
M.Pesch, D. Oderbolz, S. Hartstock
68
Qualitative Characterization of Nanoparticle Emissions from
Office Machines with Printing Function (Poster 2)
S. Seeger, H. Bresch, M. Bücker, O. Hahn, O. Jann,
O. Wilke, W. Böcker
70
Quantitative Characterization of Nanoparticle Emissions from
Office Machines with Printing Function (Poster 1)
S. Seeger, H. Bresch, M. Bücker, O. Hahn, O. Jann,
O. Wilke, W. Böcker
72
PM Measurements with ELPI (Electrical Low Pressure Im-
pactor) and TEOM 1400a in the City of Vienna at the turn of
the year 2004/2005
L. Wind, K. Beckert and T. Kilgus
74
5

image
image
UFIPOLNET LIFE04 ENV/DE/000054
6

image
image
UFIPOLNET LIFE04 ENV/DE/000054
LECTURES
7

The Role of Ultrafine Particles in the Urban Atmosphere
A. Wiedensohler, W. Birmili, and B. Wehner
Leibniz Institute for Troposheric Research, Permoserstr. 15, 04318 Leipzig, Germany
Keywords: air quality, ultrafine particles, soot
The atmospheric aerosol consists of solid and/or
liquid particles covering the size range form few
Nanometers until 10
th
of Micrometers. One can
divide the whole size range into three main
classes; coarse particles (greater than 1 μm), fine
particles (greater 0.1 and smaller 1μm μm), and
ultrafine particles (smaller than 0.1 μm). The
coarse mode contains mainly dust, sea salt,
pollen but also abrasion from tires near streets.
The lifetime is relatively short because they
quickly fall out by sedimentation. Fine mode
particles are generally more representative for
the regional aerosol since their residence time in
atmosphere can be from days until few weeks.
These particles are mainly grown up from
ultrafine particles by condensation of vapours
and cloud processes.
Ultrafine particles in the atmosphere are
generated mainly by two processes:
Homogeneous nucleation from condensable
vapours in the atmosphere (new particle
formation) and combustion of fossil fuel or
biomass.
New particles formation in urban areas can have
two sources; 1) solar radiation and a subsequent
production of condensable gases by photo
oxidation processes, and 2) vehicle emissions
the production of condensable vapours after
cooling. These particles grow in urban areas to
sizes of 10-20 nm and consist mainly of
sulphuric acid and organic carbon.
Incomplete combustion of biomass and fossil
fuel leads to the direct emission of carbonaceous
particles. In urban areas, diesel-driven vehicles
are the main sources of soot particles, which
consist of elemental carbon and adsorbed
organic material. Measurements at the tail-pipe
of vehicles or at a kerb-site (Rose et al., 2006)
showed that the number peak of soot particles
lies in the ultrafine size range between 50 and 80
nm.
Epidemiological studies have shown that aerosol
particles can lead to respiratory tract or cardio-
vascular diseases. Ultrafine particles are
presently under the discussion to be especially
responsible for such diseases (e.g. Peters et al.,
1997).
In the respiratory tract, water- and lipid-soluble
material of deposited aerosol particles is
dissolved and easily removed. Large insoluble
particles are absorbed by macrophages and
removed afterwards. Especially ultrafine
insoluble particles, such freshly emitted soot, are
believed to be associated with the greatest risk
in relation to the above mentioned diseases.
Insoluble ultrafine particles may cross cell
borders and can be distributed in the whole
human body. Fresh ultrafine soot particle are
especially surface-reactive and may additionally
carry carcinogenic substances such as PAHs.
The current legal metric to quantify particulate
air quality are 24 h averages of PM10 particle
mass concentrations (particles smaller than 10
μm). Since a large fraction of this PM10
chemical composition (inorganic salts or acids,
most of the organic carbon) is less relevant in
terms of being responsible for respiratory and
cardio-vascular diseases, it might be necessary
to add other metrics to the existing
measurements. Online soot concentration
measurements, especially in traffic-dominated
urban areas, would provide a direct measure of
surface-reactive soot concentrations and related
toxic and carcinogenic compounds.
References:
Peters , A., Wichmann, H.E., Tuch, T., Heinrich,
J., Heyder, J. (1997) Respiratory effects are
associated with the number of ultrafine particles.
Am.J.Respir.Crit.Care Med. 155, 1376-1383.
Rose D., Wehner B., Ketzel M., Engler C.,
Voigtlander J., Tuch T., Wiedensohler A..
(2006) Atmospheric number size distributions of
soot particles and estimation of emission factors.
ACP Vol. 6, 1021-1031.
8

Ultrafine Particles Chemical Content
H Herrmann,
E. Brüggemann, T. Gnauk, K. Müller, Y. Iinuma, K. Beiner, L Poulain
Leibniz-Institut für Troposphärenforschung, Chemistry Dept., Permoserstr. 15, 04318 Leipzig
Keywords: UFP, Ions, OC/EC, Organics
In the first part of this presentation the current
knowledge on particle chemical composition and its
size-dependence will be summarized. The main
goups of particle constituents are reviewed and it will
be outlined what can be learned from the quantitative
chemical analysis of particle constituents
The IfT chemistry department has performed
a number of particle characterisation field campaigns
with extensive size-resolved sampling and chemical
characterisation over the last decade and mostly in
the urban environment. Sampling was mostly done
with the so-called six stage Berner impactor, the first
stage of which samples particles with aerodynamic
diameters in the range 50-140 nm and hence covering
the larger fraction of ultrafine particles. Results from
different campaigns centered on the chemical
composition of the first Berner impactor stage will be
reviewed. The substance groups of interest cover OC,
EC and TOC, the inorganic anions and cations and
single organic species. In selected experimental
campaigns also metals were determined.
Methods for the determination of single
organic compound are summarized and their
potential for different applications such as mass
closure or source apportionment is elucidated.
Unfortunately, many currently available methods for
the determination of single organics require input
masses of material which could only be produced by
sampling ultrafine particles in unrealistically long
sampling efforts.
Efforts have been started to extend particle
sampling and subsequent chemical analytics into the
range beyond 50nm as aerodynamic. Several
approaches are being described such as an extended
cross comparison of the nano-Moudi sampler with
DMPS (i.e. particle size distribution) measurements
which lead to considerable discrepancies especially
for very small particles. An outlook is given on
ongoing and future sampling development.
As an alternative to sampling followed by
offline chemical analysis, the application of particle
realtime analysis with the Aerodyne Aerosol Mass
spectrometer (AMS) will also be discussed.
A summary will be given on the current state
of the art for chemical sampling and analysis of
paricles and ultrafine paricles together with an
outlook.
9

LIFE and LIFE + …….from program to projects
D. Palenberg
1
1
Department blue! advancing european projects GbR, Freising, Germany
Keywords: LIFEIII, LIFE+ .
With the preparation of the new structural funds
period 2007 - 2013, other EU-funding
programmes have parallel redesigned their
overall scope as well as their framework
regulations. The LIFEIII - programme, the
funding source of UFIPOLNET, has been
redesigned to the LIFE+-Programme.
Whilst some company-oriented key-contents are
now covered by other programmes, LIFE+
remains an important and viable instrument for
the support of innovations in the environmental
sector and for the realisation and promotion of
nature protection and biodiversity.
The presentation explains the way from
LIFE III to LIFE+ by highlighting the main
criteria for the new shape of the programme. It
also gives an overview on how the different
topics covered by the former programme are
now covered by other funding sources.
Furthermore, the concrete possibilities for
project initiators are listed up and
implementation examples are proposed.
Furthermore, funding rules as well as deadlines
are presented.
10

image
11
UFIPOLNET: Purpose - Partner - Project
H. Gerwig
1
, G. Löschau
1
, L. Hillemann
2
, B. Wehner
3
, A. Wiedensohler
3
, A. Zschoppe
4
, C. Peters
4
, A. Rudolph
4
,
C. Johansson
5
, J. Cyrys
6
, M. Pitz
6
, R. Rückerl
6
, J. Novak
7
, H.G. Horn
8
, R. Caldow
9
, G.J. Sem
9
1
LfUG - Section Air Quality, Saxon State Agency for Environment and Geology, 01109 Dresden, Germany
2
UBG – Staatliche Umweltbetriebsgesellschaft, 01445 Radebeul, Germany
3
Leibniz-Institute for Tropospheric Research, 04318 Leipzig, Germany
4
Topas GmbH, 01279 Dresden, Germany
5
ITM – Department of Applied Environmental Science, Stockholm University, 106 91 Stockholm, Sweden
6
GSF National Research Centre for Environment and Health, 85764 Neuherberg, Germany
7
CHMI – Czech Hydrometeorological Institute, 14306 Prague, Czech Republic
8
TSI Gmbh, 52068 Aachen, Germany
9
TSI Inc., Shoreview, Minnesota, 55126, USA
Keywords: atmospheric aerosols, instrument development, number concentration, number size distribution
The objective of the project UFIPOLNET
(Ultrafine Particle Size Distributions in Air Pollution
Monitoring Networks) is to demonstrate that the
newly developed Ultrafine Particle Monitor is able to
perform adequately in routine air monitoring
networks.
Epidemiological studies show a relationship
between high concentrations of UFP and adverse
health effects. However, there are only a limited
number of long-term ultrafine particle (=UFP)
measurements in Europe.
Therefore the European Commission
addressed its needs to get more information about
UFP concentrations within the CAFE process and the
Thematic Strategy on Air Pollution.
For these reasons a group of different
organisations, such as research institutes,
universities, monitoring networks, as well as big and
small companies, were used to develop the project
idea from 2003 to 2004. The goal was to create an
easy to use and reasonably priced Ultra-fine Particle
Monitor. The idea was supported by the LIFE-
program of the European Commission DG
Environment.
The resulting instrument produces a number
size distribution in 6 size classes between 20 – 800
nm. In addition the sampling conditions were
harmonised at all sampling stations.
The instruments have run since February 2007
in Dresden, Augsburg, Stockholm and Prague. All
sites are near busy roads; Augsburg is an urban
background site. The data of all sites are transferred
to the different databases of the central measurement
network station. The data of the 4 stations are sent
monthly to the central database where it is evaluated.
The different size classed number concentrations are
correlated with other airborne pollutants and traffic
numbers. Comparable results will allow absolute
differences between ultrafine aerosol size
distributions at many polluted sites to be analyzed
over long periods.
It is planned to run the 4 instruments for a
longer period over several years.
First comparisons with a DMPS for ambient
aerosols show a good correlation. Also NOx and
other traffic caused pollutants correlate good.
Comparison between the 4 different stations
showed different factors between particle number
concentrations and to NOx in August 2007.
Figure 1: Ultrafine Particle Monitor at 4 stations in
EU
Acknowledgements
UFIPOLNET
(www.ufipolnet.eu)
is financed by the LIFE
financial instrument of the European Community under No.
LIFE04 ENV/D/000054. The authors wish to thank Mrs. D.
Palenberg and Mr. T. Kipp from blue! advancing european
projects, Mrs. D. Pardo-Lopez, Mr. A. Kaschl, F. Vassen and P.
Grzesikowski from the European Commission as well as Mr. M.
Reisenberger from LIFE Monitoring Team –Astrale GEIE –
Particip GmbH, H.-G. Kath, M. Lohberger and S. Sorkalle from
UBG, Mrs. J. Naacke, Mrs. U. Lindner, Mrs. S. Burkhardt, Mrs. C.
Beasley, Mrs. A. Leon, Mr. W. Sommer, Mrs. A. Hausmann, Mr.
H. Gräfe, Mr. H. Schwarze, Mr. A. Bobeth and Mr. M. Böttger
from Saxon State Agency for Environment and Geology, for
helpful discussions to organise a European project like this.
Thanks goes to Mr. C. Helsper from FH Aachen for his very
important technical help during the instrument development.
Moreover the authors thanks several co-workers of the project
partners not named here, who helped a lot in making this project
possible, like i.e. in the financial administration and in technical
fields.

image
image
A new particle measurement system for environmental ultrafine particles
A. Zschoppe
1
, L. Hillemann
2
, A. Rudolph
1
, C. Peters
1
, R. Caldow
3
1
Topas GmbH, Wilischstr. 1, D-01279 Dresden, Germany
2
Umweltbetriebsgesellschaft, Altwahnsdorf 12, D-01445 Radebeul, Germany
3
TSI Incorporated, 500 Cardigan, MN 55126 Shoreview, USA
Keywords: ultrafine particles, instrumentation, measurement
It is well known that ultrafine particles have
an impact on human health. Consequently it is neces-
sary to monitor the exposure in cities and urban ar-
eas. There are some instruments commercially avail-
able for this task like SMPS, DMPS or FMPS but
their original field of application is contradictory to
the utilization in air pollution monitoring networks.
These sensitive devices had been developed
primarily for scientific purposes in the lab environ-
ment, they are servicing-intensive and delicate to
handle. The use of radioactive sources in the charger
complicates the transport of the devices. Furthermore
a compromise has to be found between the necessary
accuracy in the field of air monitoring and the costs
of the devices. The high accuracy of these devices is
not needed within the scope of air pollution monitor-
ing and lead to too high costs.
Based on the requirements of monitoring net-
works a new particle measurement system was de-
veloped and tested in the project UFIPOLNET – us-
ing of a diffusion charger instead of a radioactive
source and an electrometer instead a CPC:
Figure 1. Schematic of the new particle measurement
system for ultrafine particles.
Additional components are a long DMA (TSI 3081)
with sheath air circuit, a complete control unit (PCB)
and a single board computer for data processing and
a database driven data storage.
The data inversion delivers the particle num-
ber concentration in the following size classes:
CH1
(nm)
CH2
(nm)
CH3
(nm)
CH4
(nm)
CH5
(nm)
CH6
(nm)
20…30 30…50 50…70 70…100 100…200 >200
In the field of environmental aerosols particu-
lar attention has to be paid to the sampling system..
Basically it consists of a PM1-inlet, a membrane
dryer and an equalizing tank. The membrane dryer
requires no maintenance and induces only minimal
particle losses.
In comparison measurements between the new
spectrometer and a DMPS at a street canyon site in
Leipzig a good correlation was found:
Figure 2. Comparison measurements between the
new spectrometer and a DMPS (30 – 50 nm).
Within the project UFIPOLNET at four meas-
urement stations in Europe prototypes were installed
with identical sampling systems. The instruments are
running over a longer period and experiences of the
users will also be presented.
UFIPOLNET
(www.ufipolnet.eu)
is financed
by the LIFE financial instrument of the European
Community under No. LIFE04 ENV/DE/000054.
12

image
image
image
image
Characterization of environmental ultrafine particles with the UFP 330,
system calibration and evaluation
L. Hillemann
1
, A. Zschoppe
2
and R. Caldow
3
1
Institute of Process Engineering and Environmental Technology, TU Dresden, D-01062 Dresden, Germany
2
Topas GmbH, Wilischstr. 1, D-01279 Dresden, Germany
3
TSI Incorporated, 500 Cardigan, MN 55126 Shoreview, USA
Keywords: electrical effects, charged particles, instrumentation, measurement
The UFP 330 is an aerosol spectrometer basing its
measurement on the classification of charged
particles in an electric field. Accurate particle size
detection requires a well-defined charge status for the
aerosol which is achieved by diffusion charging. This
process generates a high charge level of the particles
making the data inversion difficult (Hillemann,
2007).
The aim of this work is to describe the calibration of
the UFP 330 and to give some results of the
evaluation process during the project.
SYSTEM CALIBRATION
The aerosol spectrometer UFP 330 combines an
electrostatic classifier and an electrometer to measure
the mobility distribution of an aerosol. A
measurement cycle delivers the mobility distribution
of the particle-bound electrical charge. The inversion
problem to calculate the size distribution f(x) from
the measured current distribution g(y) is described by
the Fredholm-equation.
The kernel data K(x,y) of the UFP-system is recorded
by a parallel quantification of a monodisperse aerosol
by SMPS, CPC and UFP.
The inversion algorithm employed in the UFP 330
uses constraints like the typical shape of an
environmental particle size distribution and
boundaries for the particle size.
Figure 1. Experimental setup for the system calibration
DATA EVALUATION
Since January 2007 a prototype of UFP 330 was
installed in an air-quality monitoring station in
Dresden. Once per month it has been dismounted and
transferred to an aerosol lab to measure the counting
efficiency with several test aerosols. This comparison
of the UFP 330 to SMPS and CPC delivered well
comparable results.
Figure 2. Measured counting efficiency for different
particle sizes
Figure 3. Comparison of the UFP 330 to SMPS
UFIPOLNET
(www.ufipolnet.eu)
is financed by the
LIFE financial instrument of the European
Community under No. LIFE04 ENV/D/000054.
L. Hillemann, A. Zschoppe and R. Caldow (2007),
Aerosol mobility spectrometry based on diffusion
charging, European Aerosol Conference 2007,
Salzburg, Abstract T05A008.
13

Ultrafine particles: Comparisons UFP 330/ DMPS
B. Wehner
1
, T. Tuch
1
, A. Wiedensohler
1
,
A. Zschoppe
2
, L. Hillemann
3
, G. Löschau
4
, H. Gerwig
4
1
Leibniz-Institute for Tropospheric Research, 04318 Leipzig, Germany
2
Topas GmbH, 01279 Dresden, Germany
3
UBG – Staatliche Umweltbetriebsgesellschaft, Radebeul, Germany
4
LfUG - Saxon State Agency for Environment and Geology, Section Air Quality, 01109 Dresden, Germany
Keywords: number concentration, instrument development, comparison, network operation
Several epidemiological studies have shown a
relationship between high number concentrations of
ultrafine particles (< 100 nm) and adverse health
effects. However, most routine measurements of
particulate matter are limited to the mass
concentration, e.g. PM10 or PM2.5. One major
reason for this is that commercially available
measurement technique is relatively expensive and
needs more maintenance than in the routine network
operation can be provided. Within the frame of the
project UFIPOLNET a new instrument to measure
ultrafine particle number concentrations has been
developed which is easy to handle and needs less
maintenance than e.g. available SMPS systems.
The new instrument (Ultrafine Particle
Monitor, UFP330) consists of a Corona Charger, a
DMA, and an electrometer. The measured current is
online transferred to a number size distribution (20 –
500 nm) and locally stored. Within routine networks
the number of measured parameters which might be
saved continuously is limited. Thus, number size
distributions are usually replaced by integral
concentrations within certain size classes. For the
UFP330 the size classes have been defined as
follows:
name
range
N1
20 - 30 nm
N2
30 – 50 nm
N3
50 – 70 nm
N4
70 – 100 nm
N5
100 – 200 nm
N6
> 200 nm
Within the frame of UFIPOLNET 4
prototypes of the instrument have been built and are
operated at 4 stations in Europe. To ensure the data
quality and comparability the UFPs have been
operated in parallel to a DMPS system in a street
canyon, representing a typical measurement site
within an urban network. Number size distributions
obtained by DMPS have been converted to size
classes according to N1 – N6.
24
48
72
96
120
144
0
2000
4000
24
48
72
96
120
144
0
1000
2000
24
48
72
96
120
144
0
500
1000
1500
UFP #3
UFP #4
DMPS
N2
N4
number concentration [cm
-3
]
N6
hour
Figure 1: Number concentrations measured in
parallel with two new UFP and one DMPS in the
street canyon site
Figure 1 shows the results of three selected
size channels for parallel measurements between two
UFP prototypes and one IfT-operated DMPS. In the
first view they show a good correlation, only a few
outliers have been registered. In general, the
correlation for the size classes N2 – N5 is higher than
for the largest and smallest one. The result is nearly
independent on the concentration at this site.
In February 2007 the instruments have been
set up at four measurement sites within Europe.
Three of them measure there parallel to another size
spectrometer such as SMPS or DMPS. These results
will give more information about the data quality at
differently polluted sites and also over a longer
period.
This project (
www.ufipolnet.eu
) is financed by the LIFE
financial instrument of the European Community under No.
LIFE04 ENV/D/000054.
14

Measurements of ultra-fine particles in Europe: differences and similarities
Rita Van Dingenen, Jean-Philippe Putaud, Daniel Mira-Salama, Niels R. Jensen, and Frank Raes
European Commission, DG JRC, Institute for Environment and Sustainability, I-21020 Ispra, Italy
Keywords: PM, Europe, size distributions
Epidemiological studies have shown a close
relationship between particulate matter (PM) mass
concentrations and mortality and / or morbidity
(e.g. Dockery et al., 1993). However, they have
also pointed out spatial differences in the health
impact of PM mass concentrations or mass
concentration increments (e.g. Le Tertre et al.,
2005). One of the reasons for this might be that
PM characteristics (chemical composition, size
distribution) do vary from place to place, both at a
large scale (across a continent) and at a more
regional/local scale (from rural background to
kerbside sites). To test this hypothesis, the results
of the epidemiological studies should be
confronted to PM characterisation data, beyond the
only PM mass concentrations reported by the air
quality monitoring networks.
Such parameters (PM mass concentration,
chemical composition and particle number
concentration and / or size distribution) have been
measured at various locations in Europe for many
years. However, most of them rarely reach
possible users such as epidemiologists, modellers,
and policy makers. Here we present a compilation
of European PM data sets that are more
comprehensive than the traditional PM-mass
oriented monitoring network sites (EMEP,
AIRBASE).
These data are complemented with a
dedicated winter time study in the Milan area,
where the urban and regional background particles
were characterized under extremely clean
conditions, i.e. in absence of a regional
background (see Figure 1). This allows for an in-
situ characterisation of locally emitted particles in
the urban environment.
Following observations have been made
from the available data sets:
At all sites, most of the particles (70 – 80% of the
number) have a diameter < 100 nm. At polluted
sites, those particles consist mainly of
carbonaceous material (BC + organic matter).
Volatility measurements indicate that freshly
emitted elemental carbon is not internally mixed
with the volatile (organic carbon and sulfate)
particles. Particles originating in the urban area
come mainly from combustion processes,
especially direct traffic emissions, domestic
heating and industrial activities, whereas the
regionally emitted particles are different with
much less traffic contribution. Nitrate, after POM
the major secondary component during stagnant
conditions, can be entirely apportioned to the
regional background aerosol.
Particle number (> 10 nm) increases more than
proportionally to PM mass, due to an increasing
contribution of small (< 100 nm) particles in
polluted sites.
At polluted sites, PM2.5, PM10 as well as
number concentration are highest during winter.
At polluted sites, the relative contribution of
nitrate is highest during winter due to semi-
volatile nature of ammonium nitrate.
Physical and chemical characterisation of the
urban background aerosol indicate consistently
that local urban emissions account for 50 – 70%
of the PM mass in the Milan urban background
during winter time, whereas the regional
background accounts for 30 – 50% of the mass.
0
5000
10000
15000
20000
25000
30000
35000
0.001
0.01
0.1
1
Dp (μm)
dN/dlogDp #/cm³
Urban BG, day, normal
Urban BG, day, clean
Figure 1
: Urban background daytime aerosol
number size distribution under normal (stagnant)
winter time conditions and under extremely clean
conditions. The former is a superposition of locally
emitted + long range transported particulate matter.
In the latter case, only locally emitted particles are
observed.
15

Origin and features of ultrafine particles in Barcelona
S. Rodríguez
1,2
, J. Pey
2
, N. Perez
2
, X. Querol
2
, A. Alastuey
2
, R. Van Dingenen
3
and J.P. Putaud
3
1
Izaña Atmospheric Observatory, University of Huelva, Santa Cruz de Tenerife, 38071, Canary Islands, Spain
2
Institute of Earth Science ‘Jaume Almera’, CSIC, Place, 08028, Barcelona, Spain
3
Institute for Environment and Sustainability, European Commission– DG Joint Research Centre, Ispra, T.P.
290, (VA), 21020, Italy
Keywords: ultrafine particles, road traffic, air quality, number size distribution, aerosol chemistry
Concentrations of the aerosol number (N),
PM10, PM2.5 and PM1, and the chemical
composition of PM10 and PM2.5 are being
monitored in Barcelona since 2003. The main
features of the ultrafine particles number size
distributions are summarised in this presentation.
In this Mediterranean city, the aerosol N10,
N5 (N of particles >10nm and >5nm, respectively)
and PM2.5 concentrations exhibits mean values of
15400cm
-3
(2003-2005), 22800cm
-3
(2005-2007) and
27μg/m
3
, respectively. The size distribution of N10-
800 (N of particles with size into the range 10-
800nm) presents a maximum in the particle diameter
34nm (DpN: diameter where the dN/dlogD reaches
the maximum). Ultrafine particles (<100nm)
accounts for 86% of the number concentration with
the range 10-800nm.
The aerosol N concentrations typically exhibit
strongly marked weekly cycles. During weekdays,
the number concentration reaches the highest values
during the morning and evening rush hours, in such a
way that the ultrafine particles concentration exhibits
a morning-to-night concentration ratio of 2.5, being
the highest morning-to-night concentration ratio
(~3.0) observed in the 30-50nm size bin. During
weekends these cycles are not observed owing to the
strong change in the daily pattern of the road traffic
intensity. The DpN diameter exhibits lower values
during daylight (~37nm) than at night (~50nm)
during working days and a relatively ‘flat’ daily
patter during weekends (~40nm). The results
obtained points that the ‘road traffic intensity’
(number of vehicles/hour) and the processes affecting
the ‘new particle formation during the dilution and
cooling of the vehicle exhaust emissions’ are the
main factors inducing the observed mean daily
patterns in the ultrafine particles concentrations.
The processes affecting the day-to-day
variations in the ‘number concentration and number
size distribution’ and their relationship with fine
particle mass concentration, has also been studied.
The main results are summarised as follow:
1.
The time series of the aerosol DpN
diameter and fine particle mass concentrations
(PM2.5 and PM1) tend to exhibit: i) correlated
increases at night, and ii) day-to-day correlated
variations. Moreover, during events of strong
increases in the DpN diameter, ammonium-nitrate is
the aerosol specie exhibiting the highest increase.
These observations points that condensation of
ammonium-nitrate onto pre-existing particles is
highly involved in the growth events associated with
increases in the fine particle mass. These correlations
between DpN and ‘fine particle mass’ are more
clearly observed in winter because of the enhanced
condensation rates at low temperatures.
2.
The number concentration in all size ranges
is negatively correlated with wind speed. However,
the correlation coefficient between wind speed and
the number size distribution dN/dogD decreases
when decreasing the particle size (e.g. r= -0.32 for
90nm particles and r= -0.16 for 13nm particles;
Figure 1). This is caused by the fact that very small
ultrafine particles are formed under ‘windy clean air
conditions’ favoured by the low surface area
available for condensation. The occurrence of these
events has also been documented in other European
cities (Charron and Harrison, 2003). The influence of
wind speed in the particle size >20nm is higher in
winter than in summer owing to the high background
aerosol concentrations in this season.
3.
The time series of the N and PM2.5
concentrations tend to exhibit frequent day-to-day
correlated variations. However, events of high total
number concentration (N) associated with high N10-
20 and low PM2.5 concentrations are frequently
recorded. These high N10-20 (>6000cm
-3
) and low
PM2.5 (<20μg/m
3
) episodes were mostly recorded
during the morning and evening rush hours in
summer and autumn being these events associated
with an enhancement of the new particle formation
during the dilution and cooling of the vehicle
exhausts.
4.
Events of new particle formation in ambient
air (so-called ‘nucleation events’, as described by
Kulmala
et al.
, 2004 or Hamed
et al.
, 2007) are
rather infrequent. About 2 to 4 events per year have
been observed, being these episodes recorded in
summer. These events also contributed to the above
cited high N10-20 (>6000cm
-3
) and low PM2.5
(<20μg/m
3
) episodes.
An analysis of the PM2.5 chemistry and the
aerosol size distribution dN/dlogD shows that
organic matter and black carbon are the only two
PM2.5 chemical components showing a significant
correlation with the ultrafine particles. The other
16

aerosol species included in the fine fraction, such as
sulphate, nitrate, ammonium, sea salt and mineral
dust, only shows a significant correlation with the
number of particles >0.1μm.
The number concentration and the DpN
diameter exhibit a higher day-to-day variability and
reaches higher concentrations in the cold season.
This is caused by the more frequent day-to-day
changes in the synoptic scale meteorology and by the
lower temperatures which favour both condensation
and nucleation processes during the dilution and
cooling of the vehicle exhaust emissions.
Some of the features observed in the ultrafine
particles in Barcelona are common to other cities of
the European Union (see as examples Charron &
Harrison, 2003; Rodríguez
et al.
, 2007; Rodríguez &
Cuevas, 2007).
-0.35
-0.30
-0.25
-0.20
-0.15
10
100
1000
Particle diameter, nm
correlation coefficient
-0.35
-0.30
-0.25
-0.20
-0.15
10
100
1000
Particle diameter, nm
correlation coefficient
Figure 1. Correlation coefficient between wind speed
and aerosol size distribution at Barcelona.
Part of this work was supported by the Ministry of
Environment of Spain (project “Evaluación integral
del impacto de las emisiones de partículas de los
automóviles en la calidad del aire urbano”).
Charron, A., & Harrison, R.M. (2003).
Atmos.
Environ.
, 37, 4109–4119.
Hamed, A., Joutsensaari, J., Mikkonen, S.,
Sogacheva, L., Dal Maso, M., Kulmala, M.,
Cavalli, F., Fuzzi, S., Facchini, M.C., Decesari,
S., Mircea, M., Lehtinen, K. E. J., & Laaksonen,
A. (2007).
Atmos. Chem. Phys.
, 7, 355–376.
Kulmala, M., Vehkamaki, H., Petäjä, T., Dal Maso,
M., Lauria, A., Kerminenb, V.M., Birmili, W., &
McMurry, P.H. (2004).
J. Aerosol Science
, 35,
143–176.
Rodríguez, S., Van Dingenen, R., Putaud, J.P.,
Dell’Acqua, A., Pey, J., Querol, X., Alastuey, A.,
Chenery, S., Ho, K.-F., Harrison, R.M., Tardito,
R., Scarnato, B., Gemelli, V. (2007).
Atmos.
Chem. Phys.
, 7, 2217–2232.
Rodríguez, S., & Cuevas E. (2007).
J. Aerosol
Science
, in press. DOI:
10.1016/j.jaerosci.2007.09.001.
17

Ultrafine particles in the UK
A. M. Jones and R. M. Harrison
Department of Environmental Health and Risk Management, University of Birmingham,
Edgbaston, Birmingham B15 2TT, United Kingdom
Keywords: particles, ultrafine, wind, emission
Particle number has been counted at eight
predominately urban sites in the UK (Figure 1) since 2000
using TSI 3022A condensation particle counters (50%
detection at 7 nm). Measurements of particle size spectra
have also been made in the 12 – 450 nm at three sites
since 1998 using scanning mobility particle sizers (TSI
3071A and TSI 3022A) size range. Annual data capture
efficiencies were up to 92.7% for particle number, and up
to 79.3% for particle size spectra.
49
50
51
52
53
54
55
56
57
58
59
-11
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
Belfast
Glasgow
Manchester
Birmingham
Port Talbot
London
Harwell
LONDON SITES:
North Kensington
Marylebone Road
Bloomsbury
Figure 1: Site locations
These data are now available from
www.airquality.co.uk/archive/particle_data.php
while
information about the locations of the sites is available
from
www.bv-aurnsiteinfo.co.uk.
Data has been updated since first reported
(Harrison and Jones, 2005). Except at Glasgow, where
data was affected by a small number of days with
particularly high values, particle number concentrations
were lowest in summer months (July, August) and highest
in the winter (December, January) at all sites.
Concentrations were similar at the urban centre and urban
background sites (typically around 20,000 cm
-3
), with
lower concentrations at a (coastal) industrial site, and
substantially higher concentrations at an urban kerbside
site.
On weekdays at urban sites there was a clear
diurnal profile in the particle number concentration with
the change in time of the peak concentration between
winter and summer time indicating an anthropogenic
source for the particulate.
Directional profiles of particle number, PM
10
and
NOx at Birmingham show higher concentrations when the
wind is from the south east, mainly related to the lower
wind speeds associated with this wind direction. Mean
particle number concentrations were generally related to
wind speed by the relationship;
Χ = c.(-ln(U/2) + 3),
where U is the wind speed in knots. This relationship is
similar to the results obtained for airborne bacteria at
inland sites and with a standard Gaussian dispersion
model (ADMS3) with a large area source and a neutral
atmosphere over a wind speed range of 2 to 20 knots
(Harrison et al, 2003). At Port Talbot, higher
concentrations of particle number and PM
10
occur when
the wind is from a local steel works, and when wind speed
increases, and not from the direction of a nearby
motorway which was associated with the highest
concentrations of NOx. In contrast, at the kerb of a
heavily trafficked road in an urban street canyon
(Marylebone Road), the directional concentration profile
of particle number was more closely related to that of
NOx, than to those of PM
10
or PM
2.5
.
Emission factors for particle number in three size
ranges (11-30; 30-100 and >100 nm) were estimated
separately for heavy and light-duty vehicles in a heavily
trafficked street canyon where the traffic speed vary
considerably over short distances (Table 1) (Jones and
Emission factor [number veh
-1
hr
-1
]
Particle size
range [nm]
Heavy duty
Light duty
11-30
2.14 x 10
14
2.30 x 10
13
30-100
3.19 x 10
14
2.84 x 10
13
>100
1.03 x 10
14
0.70 x 10
13
Table 1: Calculated emission factors
Harrison, 2006). The calculation of emission factors
assumes that particulate disperses from the on-road source
in a similar way to the dispersion of NOx.
This work was funded by the UK Department for the
Environment, Food and Rural Affairs under contracts
EPG 1/3/184 and CPEA 28. Particulate data was collected
by Bureau Veritas (formerly Casella) and the National
Physical Laboratory and King’s College, London.
Meteorological data is from the UK Meteorological Office
via the British Atmospheric Data Centre.
18

Harrison, R. M., Jones. A. M., Biggins, P. D. E.,
Pomeroy, N., Cox, C. S., Kidd, S. P., Hobman, J.
L., Brown, N. L., & Beswick, A. (2003).
Climate
factors influencing bacterial count in background
air samples
. Int. J. Biometeorology, 49, 167-178.
Harrison, R. M. & Jones, A. M. (2005).
Multisite study of
particle number concentrations in Urban Air
.
Environ. Sci. Technol. 39, 6063-6070.
Jones, A. M. & Harrison, R. M. (2006).
Estimation of the
emission factors of particle number and mass
fractions from traffic at a site where mean vehicle
speeds vary over short distances
. Atmos. Environ.
40, 7125-7137.
19

image
Ultrafine particles in Stockholm
C. Johansson
1,2
,M. Norman
2
& H. Karlsson
1
1
Department of Applied Environmental Science, Stockholm University, SE-10691 Stockholm, Sweden
2
Environment and Health Administration, SLB-analysis, P. O. Box 8136, SE-10420 Stockholm, Sweden
Keywords: vehicle exhaust, NOx, road traffic, PM10
Introduction
In Stockholm, most measurements of particulate
matter are for PM10 and the levels exceed the EU-
directives along many streets in the city centre
(Norman and Johansson, 2006). But substantial
differences in the temporal and spatial distribution of
PM10 and particle number concentrations (referred
to as PNC) are observed due to differences in
emissions, background concentrations and
meteorology (Johansson et al., 2007). While high
PM10 levels are mainly due to non-exhaust
emissions of road wear, consisting of soil mineral,
ultrafine particles (UFP, diameter < 100 nm) are due
to vehicle exhaust particles, consisting of soot and
organic compounds. UFP contribute with less than
10% to PM10 and is suspected to be more important
for mortality than the coarse particles. Nevertheless,
the EU directive regulates the total mass of all
particles less than 10 μm irrespective of size,
morphology and chemistry and also irrespective of
their health effects. Measures aimed at reducing the
negative health impact of particles must necessary
build on an understanding of the controlling factors
not only for PM10 concentrations but also for other
particle size fractions. Routine monitoring of particle
size distribution has now started at Hornsgatan in
Stockholm in the frame of the UFIPOLNET project.
Measurement sites & methods
Ultrafine particles have been measured using the
instrument developed within the UFIPOLNET
project (hereafter called UFIPOL). During a period
of 2 months parallel measurements were made using
a commercial SMPS (TSI 3071A) and CPC (TSI
3010) (called SMPS). Total number of particles (> 7
nm aerodynamic diameter) was measured using a
CPC3022 (TSI Inc.). Other measurements include
NOx, CO, BC, PM2.5 and PM10. The site,
Hornsgatan, is a densely trafficked site in central
Stockholm (see Johansson et al., 2007). In this paper
we also present data from a rural background site,
Aspvreten, 70 km south of Stockholm. At this site
PM10, BC and particle size distribution is measured.
Comparison of UFIPOL and SMPS
The comparison of total particle number
concentrations from UFIPOL and SMPS show good
agreement with a correlation of 0.88. The median
number concentrations were 9620 and 10100 cm
-3
for
UFIPOL and SMPS, respectively (Table 1).
However, for the smallest size fraction (20 – 30 nm)
the UFIPOL showed 34% lower concentrations and
the correlation was only 0.70, indicating that the
UFIPOL gives to low concentrations. Other size
fractions did not differ significantly (Figure 1).
Figure 1. Mean size distributions measured with
SPMP and UFPS.
Table 1. Comparison of UFIPOL and SPMS
measurements during May - August, 2007 at
Hornsgatan.
Size
fraction
UFIPOL
(nm)
Size
fraction
SMPS
(nm)
UFIPOL
(N/cm
3
)
SMPS
(N/cm
3
)
Slope
1
Inter-
cept
1
Cor-
rela-
tion
>20 >20.5 9620 10100 0.98
±0.01
-370
±150
0.88
20 - 30
20.5-
31.6
1510 2280 0.77
±0.02
-230
±40
0.70
30 - 50
31.6-
48.8
2870 2710 1.17
±0.02
-320
±70
0.75
50 - 70
48.8-
74.8
1930 2430 0.84
±0.01
-120
±40
0.80
70 – 100
74.8-
99.8
1570 1200 1.35
±0.02
-70
±30
0.82
100 –
200
99.8-
205.4
1530 1470 1.17
±0.02
-170
±33
0.80
1
UFIPOL=k*SMPS+m, orthogonal regression, where k=slope and
m=intercept.
Comparison with PNC, NOx, BC and
PM10
Figure 2 shows the correlation coefficients of particle
concentrations of different sizes with total number
concentrations >7 nm (PNC), NOx, BC and PM10 as
measured at Hornsgatan 2007. Particle
concentrations of all sizes except the largest size (100
– 200) show rather high correlations for PNC, BC
and NOx, between 0.65 and 0.82. For all there is
slightly lower correlation with the smallest size range
(20 – 30 nm). For PM10 there is very low correlation
with all particle sizes (<0.5).
20

0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
20 -30
30 -50
50 - 70
70 - 100
100 - 200
>200
PNC
Nox
BC
PM10
Figure 2. Correlation between different particle size
fractions and PNC (total particle number, > 7nm),
NOx, BC and PM10 (Hornsgatan, Feb. – Jul., 2007).
A large fraction of the total number of
particles at Hornsgatan is smaller than 20 nm.
Comparison with the total number as measured by
the UFIPOL, with a cut-off around 20 nm with the
total number of particles measured with a CPC3022
shows that 60% of all particles < 7 nm are between 7
and 20 nm (Hornsgatan during February-July, 2007).
A scatter plot of total number of particles >7 nm and
the sum of all measured by the UFIPOL (>20 nm) is
shown in Figure 3.
-10000
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
0
50000
100000
150000
200000
CPC3022 (>7nm) (cm
-3
)
UFIPOL (>20nm) (cm
-3
)
UFIPOL=CPC*0.39 - 291
R = 0.82
Figure 3. Orthogonal regression of total particle
number concentrations (>7 nm, CPC3022) versus
total particle number as measured with the UFIPOL
(>20 nm).
Contribution from non-local sources
Most of the particles larger than 200 nm originate
from sources outside of Stockholm. Total number
concentrations (> 10 nm) at a rural background site
70 km south of Stockholm, is around 2000 cm
-3
, as
compared to around 30 000 cm
-3
at Hornsgatan. At
Hornsgatan the maximum number is seen for 20 nm
sized particles, whereas at Aspvreten particles are
around 100 nm during most of the time due to the
aged aerosol (Tunved et al., 2004).
This aerosol is clearly seen to have a major influence
on the variability of accumulation mode particles also
at a densely traffic site in central Stockholm. This is
illustrated by comparing the variability of the particle
number concentrations of different sizes at
Hornsgatan with that observed at a rural background
site (Aspvreten 70 km south of Stockholm) (Figure
5). Very high correlation is seen between particles at
Hornsgatan and Aspvreten of sizes larger than 200
nm, whereas the correlation is low for particles
smaller than 70 nm (<0.3). The diurnal variations of
particles with diameters larger than 200 nm is totally
different as compared to the variation of particles
<100 nm (Figure 4).
-0,4
-0,2
0
0,2
0,4
0,6
0,8
1
10 - 21
21 - 47
47 - 71
71 - 98
98 - 198
198 - 390
>390
C o rrelatio n c o effic ien t
Particle size at rural background site
20 nm
30 nm
50 nm
70 nm
100 nm
200 nm
Figure 5. Correlation between particle number
concentrations of different sizes at Hornsgatan in
Stockholm (UFIPOL) with concentrations of
different sizes at a rural background site, Aspvreten,
70 km south of Stockholm.
This work was supported by the EU/LIFE project
UFIPOLNET.
References
Norman, M. & Johansson, C. (2006).
Atmospheric
Environment, 40, 6154-6164.
Johansson, C., Norman, M. & Gidhagen, L. (2007).
Environmental Monitoring and Assessment, 127,
477-487.
0
1000
2000
3000
4000
5000
6000
7000
8000
1_0
1_4
1_8
1_12
1_16
1_20
2_0
2_4
2_8
2_12
2_16
2_20
3_0
3_4
3_8
3_12
3_16
3_20
4_0
4_4
4_8
4_12
4_16
4_20
5_0
5_4
5_8
5_12
5_16
5_20
6_0
6_4
6_8
6_12
6_16
6_20
7_0
7_4
7_8
7_12
7_16
7_20
20-30
30-50
50-70
70-100
100-200
>200
cm
-3
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday
0
1000
2000
3000
4000
5000
6000
7000
8000
1_0
1_4
1_8
1_12
1_16
1_20
2_0
2_4
2_8
2_12
2_16
2_20
3_0
3_4
3_8
3_12
3_16
3_20
4_0
4_4
4_8
4_12
4_16
4_20
5_0
5_4
5_8
5_12
5_16
5_20
6_0
6_4
6_8
6_12
6_16
6_20
7_0
7_4
7_8
7_12
7_16
7_20
20-30
30-50
50-70
70-100
100-200
>200
cm
-3
0
1000
2000
3000
4000
5000
6000
7000
8000
1_0
1_4
1_8
1_12
1_16
1_20
2_0
2_4
2_8
2_12
2_16
2_20
3_0
3_4
3_8
3_12
3_16
3_20
4_0
4_4
4_8
4_12
4_16
4_20
5_0
5_4
5_8
5_12
5_16
5_20
6_0
6_4
6_8
6_12
6_16
6_20
7_0
7_4
7_8
7_12
7_16
7_20
20-30
30-50
50-70
70-100
100-200
>200
cm
-3
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday
Tunved, P., Ström, J. &
Hansson, HC (2004).
Atmos. Chem. Phys., 4,
2581-2592.
Figure 4. Mean diurnal variations of the different particle sizes based on UFIPOL.
21

image
image
Ultrafine Particles in Urban Air of Prague
J. Novák
1
1
Air Quality Protection Division, Czech Hydrometeorological Institute, Na Šabatce 17, 143 06 Prague 4,
Czech Republic
Keywords: particle, particulate matter, gases, transport, local furnaces, OC, EC
The question of air quality and the respective
problems have been recently more and more
connected with particulate matter and the size
fractions of these particles in connection with the
gaseous components emitted mainly by passenger
and freight transport and by local furnaces. The
dominant role is played by transport, both in the
Czech Republic and in Europe, as concerns its impact
on human health.
Although the EU adopted the limit for PM10 and the
limits for PM2.5 are being discussed, the WHO
requires, with regard to research outputs, not only
PM2.5 and smaller PM1 particles measurements but
also monitoring of even smaller particles with size
fractions of tens and hundreds of nanometers.
Therefore the Project EU-UFIPOLNET was launched
within which new instruments for monitoring the
number of ultrafine particles were developed and
tested in the measuring networks in the Czech
Republic, Germany and Sweden.
For the Czech Republic the measuring station Praha–
Smíchov Tunnel was selected, where air quality is
influenced mainly by traffic – ca 30 000–50 000 cars
per day (see Figure 1).
To demonstrate the results of the six-month
measurements the months August and September
2007 were selected, and to show the relations
between individual measured components
correlations were carried out (see Table 1 and Table
2).
The presented tables show high correlation
between fractions of particles CL2-CL4 both in NO
and NO2 and CO, and in fact also with MPXY(meta-
para xylene), which clearly confirms the influence of
traffic on creation of ultrafine particles. Almost zero
correlations for SO2, PM10 and O3, which is
decomposed by reaction with NO at traffic stations,
are not surprising.
In the future, after obtaining further data from
the measurements, statistical processing for
individual fractions will surely be useful, connected
with the analysis of particles for the contents of
OC/EC.
Figure 1 a, b: Smichov – Tunnel, Praha.
Table 1 Smichov – Tunnel, Praha Correlation by CORREL/MSEXCEL August 2007
PM10 NO NO2 NOx CO O3 SO2 BZN TLN EBZN MPXY OXY
CL0
0.17 0.42 0.33 0.42 0.42 0.12 0.21 0.06 0.19 0.22 0.18 0.18
CL1
0.29 0.65 0.61 0.69 0.65 -0.10 0.11 0.21 0.21 0.44 0.47 0.34
CL2
0.35 0.75 0.74 0.81 0.75 -0.29 0.01 0.31 0.20 0.57 0.67 0.46
CL3
0.38 0.80 0.78 0.86 0.79 -0.32 -0.03 0.33 0.20 0.58 0.71 0.49
CL4
0.44 0.84 0.74 0.88 0.78 -0.18 0.01 0.29 0.18 0.49 0.57 0.42
CL5
0.57 0.36 0.43 0.42 0.44 -0.16 -0.04 0.54 0.26 0.37 0.41 0.33
Table 2 Smichov – Tunnel, Praha Correlation by CORREL/MSEXCEL September 2007
PM10 NO NO2 NOx CO O3 SO2 BZN TLN EBZN MPXY OXY
CL0
0.46 0.86 0.77 0.87 0.73 -0.39 0.46 0.64 0.47 0.59 0.58 0.57
CL1
0.53 0.92 0.84 0.93 0.81 -0.39 0.44 0.72 0.53 0.67 0.65 0.64
CL2
0.54 0.92 0.87 0.94 0.84 -0.39 0.42 0.76 0.55 0.71 0.70 0.68
CL3
0.57 0.91 0.88 0.94 0.87 -0.40 0.41 0.80 0.56 0.75 0.73 0.72
CL4
0.62 0.90 0.88 0.93 0.86 -0.34 0.42 0.80 0.53 0.74 0.73 0.72
CL5
0.61 0.71 0.68 0.73 0.74 -0.46 0.33 0.78 0.49 0.72 0.71 0.67
22

Health and Particles: The epidemiological view
Annette Peters
1
1
GSF-National Research Center for Environment and Health, Institute of Epidemiology, Neuherberg, Germany,
email: peters@gsf.de
Keywords: Air Pollution, Fine Particulate Matter, Ultrafine Particles, Environmental Medicine, Public Health.
Ambient air pollution in general has improved
substantially over the past decades due to innovative
changes in combustion processes and overall
pollution mitigation. However, environmental
research has implicated over the last decade that the
mass of ambient fine particles with a diameter
smaller than 2.5 μm which penetrate into the deep
lung and ultrafine particles defined as particles
smaller than 100 nm persist to cause significant
disease burden to European populations. Specifically,
systemic effects of ambient particles outside the lung
have called attention and lead to exacerbation of
cardiovascular diseases. Research in current years
has established the biological plausibility of the
health effects observed and provides evidence on
several mechanisms contributing to the exacerbation
of cardiovascular disease by ambient particles. No
evidence for a threshold of the health effects of
ambient particles has been detected. In urban areas,
traffic related pollution as well as regionally
transported particles persist to be associated with
adverse health effects, and further research
establishing the best ways to protect vulnerable
populations such as children, the elderly and the
chronically ill is needed.
23

Updated WHO Air Quality Guidelines
M. Krzyzanowski
1
1
European Centre for Environment and Health, World Health Organization,
Hermann-Ehlers-Strasse 10, 53113 Bonn, Germany
Keywords: Air pollution – adverse effects, Risk assessment.
The results of research conducted in the last
decade have significantly increased the evidence on
health impacts of the most common air pollutants,
including particulate matter and ozone. This newly
accumulated evidence was used to review and update
the WHO Air Quality Guidelines, completed in 2006
and resulting in a set of guidelines for PM10, PM2.5,
O3, NO2 and SO2 (WHO 2006). The research data
for O
3
and particulate matter PM indicate that there
are risks to health at concentrations currently found
in many cities in developing and developed
countries. Moreover, the research has not identified
thresholds below which adverse effects do not occur.
The guideline values summarized in Table 1 indicate
the air quality required for a significant reduction of
health risks.
Table 1. Summary of air quality guideline values
Pollutant Averaging time
AQG value
Particulate
matter
PM
2.5
PM
10
1 year
24 hour
1
1 year
24 hour
1
10 μg/m
3
25 μg/m
3
20 μg/m
3
50 μg/m
3
Ozone, O
3
8 hour, daily
maximum
100 μg/m
3
Nitrogen
dioxide, NO
2
1 year
1 hour
40 μg/m
3
200 μg/m
3
Sulfur dioxide,
SO
2
24 hour
10 minute
20 μg/m
3
500 μg/m
3
1
99
th
percentile (3 days / year)
In addition to guideline values, interim targets
are given for each pollutant
.
These are proposed as
incremental steps in a progressive reduction of air
pollution and are intended for use in areas where
pollution is high. These targets aim to promote a shift
from high air pollutant concentrations, which have
acute and serious health consequences, to lower air
pollutant concentrations. If these targets were to be
achieved, one could expect significant reductions in
risks for acute and chronic health effects from air
pollution. Progress towards the guideline values
should, however, be the ultimate objective of air
quality management and health risk reduction in all
areas.
In addition to PM
2.5
and PM
10
, ultrafine
particles have recently attracted significant scientific
and medical attention. The Guidelines conclude that,
while there is considerable toxicological evidence of
potential detrimental effects of ultrafine particles on
human health, the existing body of epidemiological
evidence is insufficient to reach a conclusion on the
exposure–response relationship to ultrafine particles.
Therefore no recommendations can be provided at
present as to guideline concentrations of ultrafine
particles.
Air Quality Guidelines, Global update 2005. World
Health Organization 2006.
http://www.euro.who.int/Document/E90038.pdf
24

Can we use fixed ambient air monitors to estimate exposure to ultrafine particles?
J.Cyrys
1,2
, M. Pitz
1,2
, J. Heinrich
1
, R. Rückerl
1
and A. Peters
1
1
GSF, National Research Center for Environment and Health, 85758 Neuherberg, Germany
2
WZU, Center for Science and Environment, University of Augsburg, 86159 Augsburg, Germany
Keywords: ultrafine particles, outdoor, indoor, personal exposure
Nearly all epidemiological ambient air-
pollution studies have used ambient air monitoring
data as surrogate for the exposure of the population
of interest. The reasons for the use of fixed ambient
air monitors to estimate short- and long-term
exposures to population are that this approach has
been successful, and most epidemiological studies
require large populations and long exposure/effect
follow-up times. Monitoring of individual exposures
is too inconvenient and expensive to be applied for
such purposes. However, since people spend up to
90% of their time indoors, the validity of ambient
concentrations as an accurate estimate of exposure
has raised concerns because exposure
misclassification could bias epidemiologic results
.
Ultrafine particles (smaller than 0.1 μm in
diameter) are more variable in space and time than
fine particles (PM
2.5
or PM
10
) as they have a higher
dependence on the particle sources and their
variability due to faster removal from the
atmosphere. Consequently, the exposure assessment
of ultrafine particles is believed to be more error-
prone than the exposure assessment for PM
2.5
or
PM
10
(Pekkanen and Kulmala, 2004). In the present
review, we provide a discussion of three issues: (a)
the representativeness of an ambient monitoring
station for an urban area, (b) the relationship between
outdoor and indoor and (c) between outdoor and
personal ultrafine particle concentrations.
Representativeness of an ambient monitoring
station for an urban area
In the framework of our study on spatial and
temporal variation of particle number concentration
(NC) in Augsburg, Germany we measured NC at four
locations during December 2003 (winter period), and
at three locations from April to May 2004 (spring
period). One monitor (MON) was located at an urban
background site, two monitors (FH and BOU) were
located at traffic influenced background sites,
approximately 100 m away from the nearest major
road, and the fourth monitor (UNI) was located in the
outskirts of the city.
Figures 1 and 2 present the time series of
hourly NC averages for the winter and the summer
period, respectively. Although the particle number
concentrations at all sites followed the same pattern,
the range of the day-to-day variation differs for
different monitoring sites.
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
2-Dec-03
4-Dec-03
6-Dec-03
8-Dec-03
10-Dec-03
12-Dec-03
Particle number concentration (cm
-3
)
MON
FH
BOU
UNI
Figure 1: NC at different monitoring sites during the
winter measurement period in Augsburg, Germany.
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
05-Apr-04
12-Apr-04
19-Apr-04
26-Apr-04
03-May-04
10-May-04
Particle number concentration (cm
-3
)
MON
FH
BOU
Figure 2: NC at different monitoring sites during the
spring measurement period in Augsburg, Germany.
In the spring period the average NC levels at
the two background sites with traffic impact were
16,943 cm
-3
and 20,702 cm
-3
respectively, compared
to 11,656 cm
-3
at the background site. The inter-site
correlations between the monitoring sites were high
for both monitoring periods (r>0.80).
The absolute NC levels differed significantly
implying that cross-sectional studies should enhance
the spatial resolution of exposure estimates for NC to
attribute more accurate exposure levels to study
subjects. The high temporal correlations of NC
implicate that in epidemiological time-series studies
the use of one single ambient monitoring station is an
adequate approach for characterizing exposure to
ultrafine particles.
Relationship between outdoor and indoor
ultrafine particle concentrations
People spend typically 80-90% of their time
indoors. Therefore, particle concentrations indoors
25

rather than those outdoors can have a large effect on
personal exposure. Particles generated indoors can
confound the association of health with ambient
particles only if the particles from the two origins
have similar toxicity and their concentrations
correlate. As there are indoor sources of ultrafine
particles, such as smoking, cooking and cleaning, it
can be assumed that the concentration of ultrafine
particles generated indoors are not correlated with the
ones that penetrate into indoor environments from
outdoors.
Only few studies have evaluated the
correlation between outdoor and indoor concentration
of ultrafine particles. When major sources indoor
sources were excluded, a high correlation (r=0.92)
was observed in a room with closed windows (air
exchange rate a=0.91h
-1
). The indoor/outdoor ratio
was 0.33 for closed windows and increased to 0.78
for tilted windows (a=3.44h
-1
) (Cyrys et al., 2004).
Windows closed
0
4000
8000
12000
16000
20000
24000
0 4 8 12 16 20 24
Hours
NC (cm-3)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
I/O ratio
Indoor
Outdoor
I/O Ratio
Windows opened twice a day
0
4000
8000
12000
16000
20000
24000
0 4 8 12 16 20 24
Hours
NC (cm-3)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
I/O ratio
Ventilation at 8 am
Ventilation at 2 pm
Windows tilted
0
4000
8000
12000
16000
20000
24000
0
4
8
12
16
20
24
Hours
NC (cm-3)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
I/O ratio
Figure 3: Diurnal pattern of indoor and outdoor
average particle NC as well as indoor/outdoor ratios
for different ventilation modes
Figure 3 shows the diurnal patterns of 1-hour
averages of indoor and outdoor NC for different
ventilation conditions: closed windows, opened
windows twice a day and tilted windows all day long.
It could be seen that rapid changes of the air
exchange rates during the day may lead to lower
correlations between indoor and outdoor NC
concentrations.
Relationship between outdoor and personal
ultrafine particle concentrations
The main requirement for a time-series study
of health effects of ambient particulate matter (PM)
is that the daily variation in the PM concentration
measured at a central monitoring site correlates with
the variation in average personal exposure (Zeger et
al., 2000). Many studies have demonstrated that
individual personal exposures to PM are poorly
correlated with ambient concentrations. Wilson and
Suh (1997) and Mage et al. (1999) argued that the
composition and properties of ambient particles
differ substantially from those generated in other
microenvironments and that epidemiological studies
use central-site ambient PM as a surrogate for
exposure to “PM of ambient origin”, not as a
surrogate for total PM exposure.
Nowadays a few studies have reported the
attempt to estimate the contribution to personal
exposures from particles originating outdoors, to
quantify exposure errors arising from the use of a
central site surrogate, and to understand the effect of
such errors on epidemiological conclusions. These
results demonstrate the usefulness of separating total
personal particle exposures into their ambient and
non-ambient components. The results support
previous epidemiologic findings using ambient
concentrations by demonstrating an association
between health outcomes and ambient (outdoor
origin) particle exposures but not with non-ambient
(indoor origin) particle exposures.
So far, there are no studies on the correlation
between the ambient-air and personal-exposure
concentrations of ultrafine particles. In our
presentation we will show the first results of a pilot
study on the relationship between the personal and
outdoor ultrafine particle concentrations.
Cyrys J., Pitz M., Bischof W., Wichmann H.-Erich &
Heinrich J. (2004)
J. Expo. Anal. Environ.
Epidemiol.
14, 275-283.
Mage D., Wilson W., Hasselblad V. & Grant L.
(1999)
J. Air Waste Manage. Assoc.
49, 1280-
1291.
Pekkanen J. & Kulmala M. (2004).
Scand. J. Work.
Environ. Health
; 30, 9-18.
Wilson W. E. & Suh H. H. (1997)
J. Air Waste
Manage. Assoc.
47, 1238-1249.
Zeger SL., Thomas D., Dominici F., Samet JM.,
Schwartz J., Dockery D. & Cohen A. (2004)
Environ. Health Perspect.
108, 419–42
26

Cardiovascular Emergency Calls Associated to Urban Submicron Aerosol Fractions
U. Franck
1
, S. Odeh
1
, W.-H. Storch
2
, Th. Tuch
1
, A. Wiedensohler
3
, B. Wehner
3
, and O. Herbarth
1, 4
1
Department Human Exposure and Epidemiology, Helmholtz Centre for Environmental Research – UFZ,
Permoserstraße 15, 04318 Leipzig, Germany
2
Ärztlicher Leiter Rettungsdienst - Brandschutzamt der Stadt Leipzig, Goerdelerring 7, 04092 Leipzig, Germany
3
Leibniz-Institute for Tropospheric Research, Permoserstraße 15, 04318 Leipzig, Germany
4
Department of Environmental Hygiene and Epidemiology, Medical Faculty, University of Leipzig, Liebigstraße
27, 04103 Leipzig, Germany
Keywords: Health effects of aerosols, submicron particles, size distribution, urban areas
Introduction:
It is well known that high
concentrations of airborne particles are associated
with the development of various environment related
diseases and the exacerbation of various diseases.
Since middle of the nineties epidemiologists found
more and more indications that in addition to
respiratory illnesses cardiovascular illnesses seem to
be associated to airborne particulates. Recently,
ultrafine particles have come under special scientific
scrutiny. Usually, these particles do not contribute
significantly to the mass concentration PM10 but
they dominate particle number concentration.
Especially, cardiovascular diseases are under
suspicion to be evoked and exacerbated by particles
significantly smaller than one micrometer.
This study is primarily aiming at two
questions:
One aim is to investigate the influence of “rather
low” common urban concentrations on the health
state of city dwellers, who are not occupationally
exposed.
Secondly, this study is quantifying risk differences
for selected cardiovascular diseases associated with
different size fractions of submicrometer particles,
PM2.5 and PM10.
Material and Methods: The study was carried
out in the City of Leipzig. This city is located in the
Leipzig basin with no significant elevations in and
around the city. Leipzig has approx. 500,000
inhabitants. There is no significant pollution by
industry. In the City of Leipzig urban traffic is a very
important source of airborne particles.
Aerosol measurements were carried out at the
Leibniz Institute for Tropospheric Research using
TDMPS (twin differential mobility sizer system)
working detecting particles with diameters from 3 to
800 nm. The measuring site can be regarded as urban
background. Additionally, PM10 and PM2.5 data of
public authorities were used for comparison of the
health effects of these coarser particles with the
effects of the smaller ones.
Cardiovascular emergency calls were selected
from the total number of emergency calls for a time
period of 12 month within the City of Leipzig.
Therefore, there is no bias produced by the selection
of areas within the city.
Results:
Table 1 lists the urban particle
concentrations of different size fractions found
during the measuring period.
Size Fractions
(Dp nm; PM
μg/m³)
Mean Median Min Max
Dp<100 1209410893
1487 34650
100<Dp<500 1919
1723 334 18668
Dp>500 28.54
17.91 2.107 280
Dp<800 1404313111 2450 35338
PM10 32.48
28.56 6.829 109.7
PM2.5 20.61 18.18 1.375 84.06
Table 1. Statistics of daily averages of particle
number concentrations during the measuring period.
In total 5326 cardiovascular emergency calls
were used for epidemiologic analysis. Generally,
there is no significant difference in incidence
between the weekdays and weekends. There are
22.74% and 22.44 % of cardiovascular emergency
calls, respectively.
We found:
a significant positive correlation between the risk
for cardiovascular emergency calls and the particle
number concentrations,
a time lag of 1 to 8 days for the health effect of the
particles,
differences in effect for different particle size
fractions,
differences in effect on different cardiovascular
diseases.
Significant effects could be found despite of the great
inner urban differences in the concentrations of some
particle fractions (Tuch
et al.
.2006).
Tuch, Th., Herbarth, O., Franck, U., Peters, A.,
Wehner, B., Wiedensohler, W., Heintzenberg, J. (2006).
J
Expo Sci Environ Epidemiol
, 16, 486-490.
27

Health and Particles - Regulatory Aspects
N. Englert
Federal Environment Agency (Umweltbundesamt), 14195 Berlin, Germany
Keywords:
Air Quality Framework Directive, Daughter Directives, national legislation
Science and regulation have different
approaches to problems. The scientist finds an open
question and will try to find out the facts and the
mechanisms behind them in full detail. To fill gaps
in knowledge is a welcome task.
The regulator finds a possible problem with
many uncertainties. Gaps in knowledge are unwel-
come. The questions to be answered are:
Is there a problem, and do I have enough
time to find out whether or not?
Is there a need for (immediate) action?
If yes, which kind of action?
Does this action solve the problem, or does
it cause more problems (for me or for
others)?
With respect to particles and health, the
answer to the first two questions is YES. Time-series
studies and cohort studies could demonstrate health
effects of particles: An increase in total mortality
and hospital admissions, respiratory diseases and
symptoms. The data indicate possible carcino-
genicity, too. Thus, there is a problem and action is
necessary.
Which kind of action is adequate to reach the
aim of lowering immission concentrations of
particles? Which particles?
The
COUNCIL DIRECTIVE of 15 July 1980
on air quality limit values and guide values for
sulphur dioxide and suspended particulates
(80/779/EEC)
refers to the determination of black
smoke and its conversion into gravimetric units in
Annex I. Alternatively, Annex IV refers to gravi-
metric measurements without specifying the size
fraction to be sampled.
During the following years, more data on
health effects emerged suggesting smaller particles
to be more important. The
COUNCIL DIRECTIVE
96/62/EC of 27 September 1996 on ambient air
quality assessment and management
(Framework
Directive) listed:
Fine particulate matter such as soot
(including PM
10
) and
Suspended particulate matter
among the atmospheric pollutants to be taken into
consideration.
The
COUNCIL DIRECTIVE 1999/30/EC of
22 April 1999 relating to limit values for sulphur
dioxide, nitrogen dioxide and oxides of nitrogen,
particulate matter and lead in ambient air
(First
Daughter Directive) defined limit values for PM
10
with a
Stage 1
to be met by January 2005 and
indicative limit values in
Stage 2
forseen to be met
by 2010, but to be reviewed in the light of further
information on health and environmental effects,
technical feasibility and experience in the appli-
cation of
Stage 1
limit values in the Member States.
The scientific background was described in
the Position Paper
Ambient Air Pollution by
Particulate Matter (1997)
. It stated:
“There is increasing evidence that health
effects occur at very low levels of PM, and without
an apparent threshold. This evidence arises from
studies, initially in the US, but which have more
recently been carried out in Europe and elsewhere
with similar conclusions. These studies have, in
general, used different measures of particles, but the
majority have used PM
10
and the Group felt that this
was the most appropriate measure for a limit value
for particles at the present time.”
“Good reasons might be given for
considering other fractions rather than just PM
10
,
e.g., PM
2.5
, and there might be an increasing need of
such kind of measurements in the future. At present
knowledge on health effects of particle fractions is
insufficient, and sufficiently standardised measure-
ment methods are not available to provide a sound
basis for limit values for particle fractions smaller
than PM
10
. As this will probably change in the
future, limit values set for PM
10
now may have to
undergo revision at a later stage.”
It remains difficult to decide if PM
10
or
PM
2.5
is the more adequate fraction because only
few studies with parallel measurement of both frac-
tions exist (Englert, 2004). WHO (2004) summarises
the results of the
Systematic Review of Health
Aspects of Air Pollution in Europe
as follows:
“Nevertheless, there is sufficient concern to
consider reducing exposure to coarse particles as
well as to fine particles. Up to now, coarse and fine
particles have been evaluated and regulated together,
as the focus has been on PM
10
. However, the two
types have different sources and may have different
effects, and tend to be poorly correlated in the air.
The systematic review therefore recommended that
consideration be given to assessing and controlling
coarse as well as fine PM. Similarly, ultrafine
particles are different in composition, and probably
to some extent in effect, from fine and coarse
28

particles. Nevertheless, their effect on human health
has been insufficiently studied to permit a quanti-
tative evaluation of the risks to health of exposure to
such particles.”
Regulations should be simple and cost-
effective. They have to be proportional and not
excessive. It is necessary that compliance can be
checked.
A system with limit values for PM
2.5
and
PM
10
or PM
2.5
and coarse particles (PM
10-2.5
, CP) in
parallel would not be really simple, and a change
from PM
10
to another metric only a few years after
implementation of PM
10
measurements might not be
cost-effective. Another aspect is that the smaller
fractions are part of the larger ones, i.e., ultrafine
particles (UF) are part of PM
2.5
, and PM
2.5
is part of
PM
10
(see Figure 1). Under scientific aspects, a
change from PM
10
to PM
2.5
might be justified, but
for regulatory purposes this is not necessarily the
case considering the great contribution PM
2.5
is
making to PM
10
.
UF
PM
10
PM
2.5
PM
10-2.5,
CP
Figure 1. Particulate matter fractions. The ratios of
the areas of the squares roughly correspond to the
ratios of the concentrations in European cities
(adapted from Englert (2004)).
The scientific basis for defining limit values
is sufficient for PM
10
and for PM
2.5
, too, but at
present not for CP and even less for ultrafine
particles. The situation with respect to measurement
equipment and intercomparisons is similar.
However, a special problem for regulators
is the fact that no “safe” PM level can be defined.
Which size of remaining risk has to be tolerated is
difficult to decide, and cost-benefit analyses can
only shed light on some of the aspects but cannot
provide really “objective” criteria, one reason being
the comparison of real costs and monetised values of
statistical life.
Scientists usually favour differentiation, but
a system based on several limit values in parallel is
not easily understood by the public (and maybe not
only by the public). The proposal made by the 1997
Position Paper Group was a 24-hour PM
10
limit
value of 50 μg/m³ as a 98-Percentile (i.e. a maxi-
mum of 7 exceedances) combined with an annual
PM
10
mean of 20 μg/m³. In this case, short-term and
long-term values are roughly equivalent. This
proposal was adopted as
Stage 2
. For
Stage 1
,
however, the 24-hour PM
10
limit value of 50 μg/m³
with a maximum of 35 exceedances, which would be
equivalent to an annual mean of 30 μg/m³, was
linked with an annual mean of 40 μg/m³. The result
was as could be expected: The annual mean can be
met much easier than the 24-hour mean.
Unfortunately, the noncompliance with the 24-hour
limit value may give the misleading message that
health problems are mainly due to short-term
changes in air pollution.
Keeping in mind future developments with
the possible inclusion of ultrafine particles, the way
how to make the system simple and cost-effective is
far from being found. A Directive combining
Framework Directive and Daughter Directives is
currently in preparation and will probably
supplement PM
10
Stage 1
values with PM
2.5
targets.
Inclusion of ultrafine particles (number con-
centration) will be reserved to future revisions.
And future revisions will come. Current air
pollution legislation is based on European
Directives. It is a long way from a first draft to the
final adoption of a Directive, and the process of
transfer to national legislation also needs time,
usually at least two years. Meanwhile, new scientific
results emerge, inducing a call for revision. This
process probably will continue.
Englert, N. (2004).
Toxicology Letter,
149, 235-242.
Position Paper (1997)
Ambient Air Pollution by Par-
ticulate Matter
http://ec.europa.eu/environment/air/pdf/pp_pm.pdf
WHO (2004).
Health Aspects of Air Pollution.
Results from the WHO Project "Systematic
Review of Health Aspects of Air Pollution in
Europe”.
WHO Regional Office for Europe,
Copenhagen, Denmark.
29

Health effects of inhaled ultrafine particles in the lungs and other secondary target
organs like brain and heart.
W. G. Kreyling, M. Semmler-Behnke
GSF-Research Center for Environment and Health
D-85764 Neuherberg / Munich, Germany
Keywords: ultrafine particle, nanoparticle, translocation into circulation, nanoparticle biokinetics, nanoparticle
accumulation in secondary target organs.
Nanoparticles are increasingly used in a
wide range of applications in science,
technology and medicine. Since they are
produced for specific purposes which cannot be
met by larger particles and bulk material they
are likely to be highly reactive, in particular,
with biological systems. On the other hand a
large body of know-how in environmental
sciences is available from toxicological effects
of ultrafine particles after inhalation. Since
nanoparticles feature similar reactivity as
ultrafine particles a sustainable development of
new emerging nanoparticles is required.
Cardio-vascular effects observed in
epidemiological studies triggered the discussion
on enhanced translocation of ultrafine particles
from the respiratory epithelium towards
circulation and subsequent target organs, like
heart, liver, spleen and brain, eventually causing
adverse effects on cardiac function and blood
coagulation, as well as on functions of the
central nervous system. There is clear evidence
that nanoparticles can cross body membranes
and reach and accumulate in the above
mentioned secondary target organs.
To quantitatively determine accumulated
fractions in such organs the ultimate aim is to
quantitatively balance the fractions of
nanoparticles in all interesting organs and
tissues of the body including the remainder body
and total excretion collected between application
and autopsy. Substantial uncertainty remains if
only selected organs are analyzed. Furthermore,
in case of analysis based on a label (radioactive,
fluorescent, magnetic, etc.) firm fixation of the
label to the nanoparticles need to be
demonstrated. Since these gross determinations
of nanoparticle contents in organs and tissues do
not provide microscopic information on the
anatomical and cellular location of nanoparticles
such studies are recommended to be
complemented by electron microscopy analysis.
In addition, the role of particle parameters
determining this translocation dynamics remains
to be not fully understood. Nanoparticle
parameters such as size, hydro- / lipophilicity,
surface charge, surface ligands and their
possible exchange in various body fluids need to
be considered. The current knowledge on
systemic translocation of ultrafine particles in
man and animal models and an estimate of
accumulating particle number, surface area and
mass in secondary target organs during short-
term and chronic exposure will be discussed in
order to demonstrate the relevance of
translocated fractions of nanoparticles.
This work was supported by the CEC (FIGD-
CT-2000-00053), EU FP6 PARTICLE_RISK
012912 (NEST), and U.S. NIH Grant
HL070542.
Kreyling WG, Semmler M, Erbe F, Mayer P,
Takenaka S, Schulz H, et al. (2002)
J Toxicol
Env Health A
65(20):1513-1530.
Semmler M, Seitz J, Erbe F, Mayer P, Heyder J,
Oberdorster G, Kreyling WG. (2004)
Inhal
Toxicol
16(6-7):453-459.
Kreyling, W., Semmler-Behnke, M., and Möller,
W. (2006).
J. Nanopart. Res.
8
, 543-562.
Semmler-Behnke, M. Takenaka, S. Fertsch, S.
Wenk, A. Seitz, J. Mayer, P. Oberdörster G.,
Kreyling, W. G. (2007)
Environ Health
Perspect
115(5):
9685.
30

Ultrafine Particle (UFP) Measurements and Modelling
Wolfram Birmili
Leibniz Institute for Tropospheric Research, Permoserstrasse 15, 04318 Leipzig
Session “Modelling Fine and Ultrafine Particles” on October 24, 2007.
Keywords: UFP measurement, UFP modelling, transport, aerosol dynamical processes
During the past years, there has been a growing
interest in ultrafine particles (UFP) that are
encountered in our environment. According to the
most used convention, the class of UFPs includes
airborne particles with diameters less than 100 nm.
Several properties of evironmental UFPs have
attracted the concern of environmental health
research: Their small size and therefore deep
entrance into the human lung; their capability to
travel from the respiratory system into practically all
organs of the body including the brain; their
increased fraction of relatively insoluble material and
therefore increased resistance against clearance; their
relatively high abundance in terms of particle number
concentration especially in urban areas; their specific
surface area compared to a mass-equivalent of larger
particles; their increased fraction of soot and its
associated toxic micropollutants; their relatively long
life-time in the atmosphere due to inefficient wash-
out and sedimentation.
It is of a natural consequence that the sources,
the sinks as well as the entire lifecycle of
environmental UFPs need to be explored in order to
better assess the human exposure to these particles
and to evaluate the associated risks.
Important contributions of environmental
UFPs originate from primary sources, such as high
temperature combustion, as well as secondary
sources, such as gas-to-particle conversion. It appears
that within cities and larger conurbations, the
background of primary emissions, such as from
industry and traffic, dominates the average
concentrations of UFPs while in the continental
background atmosphere, their mean concentrations
are overwhelmingly controlled by secondary particle
formation. In the background atmosphere, UFPs have
been shown to originate at large numbers at
diameters < 10 nm from gas-to-particle formation.
Stable particles at sizes between 1 and 2 nm can be
formed from charged or neutral molecular clusters
(Kulmala et al. 2007).
UFP measurements in Europe have been
summarised for rural and urban locations by van
Dingenen et al. (2004), whereas for Germany a
survey has been given by Birmili et al. (2006).
“Modelling ultrafine particles” can refer to, either the
simulation of particle transport near their sources
(Vardoulakis et al., 2003), the evolution of the
aerosol population as a result of dynamic processes,
or the combination of both.
Aerosol dynamical processes are relevant for
the evolution of new particles resulting from
secondary particle formation, but also for the primary
emission processes. The most important atmospheric
processes include the nucleation (formation) of new
particles, condensation of vapours onto pre-existing
particles, coagulation between particles, and particle
deposition (e.g., Friedlander 2000). Downstream
primary particle sources such as traffic, particles may
undergo coagulation and also chemical processes.
This presentation will summarize some main findings
during UFP observations, and outline parallels to the
achievements of model simulations:
UFP observations
Primary UFPs, mainly from motor traffic
Secondary UFPs
Aerosol dynamical processes in the
atmosphere
The lifetime of UFPs in the atmosphere
Indoor UFPs
Birmili, W. and 11 others (2006)
Räumlich-zeitliche
Verteilung, Eigenschaften und Verhalten ultrafeiner
Aerosolpartikel (< 100 nm) in der Atmosphäre, sowie
die Entwicklung von Empfehlungen zu ihrer
sys¬tem¬atischen Überwachung in Deut¬sch¬land.
93 p. UBA-Texte, 26, on-line at
http://www.umwelt-
daten.de.
Friedlander, S. K. (2000)
Smoke, Dust, and Haze
, 2nd ed.,
Oxford University Press, Oxford.
Kulmala, M. and 16 others (2007), Toward Direct
Measurement of Atmospheric Nucleation.
Science
,
www.sciencexpress.org,
30 August 2007.
Van Dingenen, R and others (2004) A European aerosol
phenomenology 1: physical characteristics of
particulate matter at kerbside, urban, rural and
background sites in Europe.
Atmos. Env
. 38:2561-2577.
Vardoulakis, S., B. E. A. Fisher, K. Pericleous, N.
Gonzalez-Flesca, 2003: Modelling air quality in street
canyons: a review. Atmos. Env., 37, 155-182.
31

Modelling ultrafine particles in urban environments
Matthias Ketzel
Department of Atmospheric Environment, National Environmental Research Institute,
Aarhus University, 4000 Roskilde, Denmark
Keywords: ultrafine particles, traffic emissions, aerosol modelling
INTROCUCTION
Ultrafine particles (UFP) and their sources
and fate in the atmosphere continue to be key
subjects in the atmospheric research due to concerns
about the effects on human health and global climate.
Over the recent years intensive measuring activities
were conducted to characterise the particle size
distribution and composition at different ambient and
indoor locations, including kerbside, urban
background, near-city and rural level. Traffic
emissions could be identified as a dominating source
of UFP (aerodynamic diameter less than 0.1μm).
This can be seen by the large gradient in particle
concentration between rural and kerbside locations
shown in figure 1. Particle number emission factors
and size distribution could be estimated for real
world driving conditions (Ketzel et al., 2003) and for
soot particles (Rose et al. 2006).
0
5000
10000
15000
20000
25000
30000
1
10
100
1000
Diameter [nm]
dN/dlogDp [#/cm
3
]
kerbside (JGTV)
urban bg. (HCOE)
near-city (LVBY)
rural (VVHL)
Figure 1. Measured particle size distribution at
several locations in Denmark and Southern Sweden.
(Based on Ketzel et al. 2004)
METHODS
Here we apply these emission factors together
with a street pollution model OSPM (Berkowicz,
2000) to predict time series of total particle number
and NOx at street level. Moreover we identify the
relevant particle dynamic processes acting at
different time scales. We describe a coupling of a
plume dispersion model and a sectional aerosol
dynamic model that is able to describe the evolution
of the particle size distribution on the way from the
exhaust to the urban background (Ketzel et al. 2005).
CONCLUSIONS
The time scales for particle dynamic processes
(e.g. coagulation, deposition, dilution, condensation)
are estimated for urban street and background
environment and discussed in their relevance for
influencing the particle size distribution. In
agreement with the literature it can be concluded that
coagulation is too slow to alter the size distribution in
the exhaust plume and dilution is the dominating
process. A similar conclusion can be drawn for the
street scale. In a more confined environment as e.g. a
road tunnel the removal processes coagulation and
deposition might play an important role.
We show that for model predictions at street
level the total particle number can be treated as ‘inert
tracer’, i.e. particle transformation has not to be
modelled in detail and rather considered as part of the
emission process. Using OSPM we obtain high
correlation (R=0.8) between measured and modelled
total particle number assuming constant emission
factors (Figure 2). If we account for a temperature
dependence of the particle emissions the model
results improve slightly (R=0.82).
-10000
0
10000
20000
30000
40000
50000
60000
70000
80000
18-06-2001
25-06-2001
02-07-2001
09-07-2001
ToN_mod
ToNdiff_obs
Figure 2. Time series of OSPM model predictions
(grey) and measurements (black line) for total
number concentration ToN [cm-3] at a kerbside
(Jagtvej) for a period of 3 weeks.
32

The modelling of the full particle size
distribution at urban scale shows that 1) the
coagulation and deposition account for ca. 25% loss
of particles in the urban background compared to
street level and 2) the size dependent removal lead to
a slight increase in 'peak' - diameter of the particle
size distribution from ca. 25nm to 30nm, see figure 2.
Both effects are in agreement with the observation.
0.01
0.1
1
Dp [μm]
0
4000
8000
12000
16000
dN / dlogDp [cm
-3
]
LVBY near-city
no removal
+Coagulation
+Coa.+Deposition
+Coa.+Dep.+Cond.
Growth
Removal ca. 25%
Figure 3. Modelled particle size distribution using the
measured near-city distribution as background.
(Based on Ketzel et al. 2005)
ACKNOWLEDEMENTS
This work was supported by the Danish
Research Agency and the Danish EPA under the
Danish Ministry of Environment. The authors thank
Prof. Erik Swietlicki and Dr. Adam Kristensson for
sharing their experimental data with us.
.
REFERENCES
Berkowicz, R. (2000): OSPM - A parameterised
street pollution model.
Environmental Monitoring
and Assessment
65, 323-331. (see also:
http://ospm.dmu.dk
)
Ketzel, M., P. Wåhlin, R. Berkowicz, and F.
Palmgren (2003) Particle and trace gas emission
factors under urban driving conditions in
Copenhagen based on street and roof-level
observations,
Atmos. Environ.
, 37, 2735–2749,
2003.
Ketzel, M., Wåhlin, P., Kristensson, A., Swietlicki,
E., Berkowicz, R., Nielsen, O. J. and Palmgren, F.
(2004): Particle size distribution and particle mass
measuremnts at urban, near-city and rural levl in
the Copenhagen area and Southern Sweden.
Atmos.Chem.Phys. 4, 281-292.
Ketzel, M. and Berkowicz, R. (2005): Multi-plume
aerosol dynamics and transport model for urban
scale particle pollution.
Atmospheric Environment
39
, 3407-3420.
Rose, D., B.Wehner, M. Ketzel, C. Engler, J.
Voigtländer, T. Tuch, and A.Wiedensohler
(2006), Atmospheric number size distributions of
soot particles and estimation of emission factors,
Atmos. Chem. Phys
.,
6
, 1021–1031.
33

Modelling PMx-Emissions and –Concentrations of streets for environmental impact
assessments and action plans
Dr. rer. nat. I. Düring, Ingenieurbüro Lohmeyer GmbH & Co. KG, Radebeul, Germany
Keywords: Particular matter, vehicle emission, action plan, re-suspension.
The field measurements in the vicinity of
roads and in streets canyons in Germany show at
some locations an exceedance of the limit values for
PM10 of the EU Directive 1999/30/EU respective
22. BImSchV (2007). Therefore action plans and
PM10-modelling for environmental impact
assessments are needed.
There exist a large number of modelling tools
for the determination of PM concentrations. In the
internet a user specific selection on the basis of a
catalogue of criteria is possible for example using the
"Model Documentation System (MDS)" of the
European Topic Centre on Air and Climate Change
1
.
Usually, in the frame of action plans and
environmental impact assessments for the
determination of PMx concentrations, models are
used, which do not directly account for the formation
or interaction of particles. In the near field of roads,
this is acceptable because of the large time scales of
these effects and the restriction to PM10. These
simplified and cost efficient procedures use the
information about the large- and regional scale
background concentration from the results of
appropriate monitoring stations in the open
countryside. The concentrations caused by the
emissions of the city are either also taken from
monitoring stations or are calculated on the basis of
the cities emissions (traffic, industrial and domestic
sources). Traffic induced hot-spot concentrations can
be detected on the basis of screening or more
sophisticated models, in Germany frequently applied
are LASAT, PROKAS, IMMIS, MISKAM,
AUSTAL.
PM emissions from traffic are a main source
of ambient concentrations especially at hot spots in
urban environment. Traffic PM emissions can be
subdivided into three main groups:
A) direct exhaust emissions,
B) direct emissions other than exhaust as e.g.
from brakes and clutches, and
C) indirect or re-suspended PM emissions from
the tyre/road interface.
A) Direct exhaust emissions are predominantly found
in the fine fraction (PM2.5) and are basically a
function of the vehicle engine and driving patterns,
but not weather or road conditions
(EMEP/CORINAIR, 2003, UBA 2004). They are
measured in laboratories and documented in different
emission databases (e.g. COPERT 2006, UBA 2004,
1
http://air-
climate.eionet.eu.int/databases/MDS/index_html
EMEP/CORINAIR 2003). In Germany is used the
emission database of UBA (2004).
B) Emissions from wear of brakes and clutches
consist to about equal parts of the fine and coarse
(PM10-PM2.5) fraction (Garg et al., 2000,
EMEP/CORINAIR, 2003) and correlate well with the
direct emissions and other vehicle emissions e.g.
NOx (Wåhlin et al., 2006). For these emissions there
are less measurements available and they are for
example not included in the UBA emission databases
(UBA, 2004).
C) Emissions from road abrasion, tyre wear and road
dust re-suspension are found partly in the fine but
mostly in the coarse fraction. This PM source is often
less correlated with the exhaust emission due to an
influence from ‘external factors’ such as tyre type,
vehicle induced turbulence, road and weather
condition (Gustafsson et al., 2005; Kupiainen et al.,
2005; Norman and Johansson, 2006; Johansson et al.,
2006). These external factors provide a major
challenge for the estimation of this type of emissions
and presently much research is undertaken to
elucidate this PM source.
The findings of some research projects (e.g.
Lohmeyer et al., 2004) together with results of
Gehrig et al. (2003) led to the procedure suggested in
Table 1. This method to display the results follows
the general procedure of the INFRAS emission factor
handbook (UBA, 2004) for all exhaust emissions and
the procedure of Gehrig et al. (2003), which is to
provide the emission factors as a function of the so
called “traffic situations”, e.g. motorway, city main
roads, city slow traffic etc.. For a description of these
traffic situations see Table 1 or UBA (2004).
The data indicate that non-exhaust emissions
are also increased for low average travelling speed
due to more unsteady driving conditions and the
more frequent acceleration / deceleration cycles.
At most of the locations in the studies of
Lohmeyer et al. (2004) and Gehrig et al. (2003) the
vehicle speed is 40-50 km/h only one location with
slightly higher and one with slightly lower speed.
Due to the limited range of vehicle speeds in our data
34

and the many other influencing factors this issue is
not discussed any further within this studies.
Finally it can be concluded, that in fact the
presently available tools for the quantification of
emission and dispersion allow a determination of
PM-concentrations for assessments and action plans
but the uncertainties are still large and important
physical dependencies are insufficiently determined
or absolutely unknown. In this respect, a large need
of research and development exists.
COPERT 4, 2006: Computer Programme to calculate
Emissions from Road Transport. Emission
Inventory Guidebook - Road transport. September
2006 See also
http://lat.eng.auth.gr/copert/
CORINAIR, 2003: Automobile and brake wear.
Web-site supporting the development of chapter
B770 (SNAP 0707) of the EMEP/Corinair
Emission Inventory Guidebook.
http://vergina.eng.auth.gr/mech/lat/PM10/title.htm
Garg, B. D., Cadle, S. H., Mulawa, P. A., and
Groblicki, P. J., 2000: Brake wear particulate
matter emissions. Environmental Science &
Technology, 34, 4463-4469.
Gehrig, R., Hill, M., Buchmann, B., Imhof, D.,
Weingartner, E., Baltensperger, U., 2003: Ve-
rifikation von PM10-Emissionsfaktoren des
Straßenverkehrs. Abschlussbericht der
Eidgenössischen Materialprüfungs- und
Forschungsanstalt (EMPA) und des Paul Scherrer
Institutes (PSI) zum Forschungsprojekt
ASTRA 2000/415.
Juli 2003.
www.empa.ch/plugin/template/empa/700/5750/---
/l=1.
Gustafsson, M., Blomqvist, G., Dahl, A.,
Gudmundsson, A., Ljungman, A., Lindbom, J.,
Rudell, B, and Swietlicki, E., 2005: Inhalable
particles from the interaction between tyres, road
pavement and friction materials. VTI publication
No. 521. See also
http://www.vti.se/EPiBrowser/Publikationer/Engli
sh/R521.pdf
Ketzel, M, Omstedt, G., Johansson, G., Düring, I.,
Pohjola, M., Oettl, D., Gidhagen, L., Wåhlin
,
P.,
Lohmeyer, A., Haakana, M., Berkowicz, R., 2007:
Estimation and Validation of PM2.5/PM10-
exhaust and non exhaust emission factors for
practical street pollution modeling: Atmospheric
Environment, article in press.
Kupiainen, K. K., Tervahattu, H., Räisänen, M.,
Mäkelä, T., Aurela, M., and Hillamo, R., 2005:
Size and composition af airborne particles from
pavement wear, tires, and tractor sanding.,
Environ. Sci. Technol., 39, 699-706.
Lohmeyer, 2004: Berechnung der Kfz-bedingten
Feinstaubemissionen infolge Aufwirbelung und
Abrieb für das Emissionskataster Sachsen.
Ingenieurbüro Dr.-Ing. Achim Lohmeyer,
Radebeul unter Mitarbeit der IFEU Heidelberg
GmbH und der TU Dresden, Institut für
Verkehrsökologie. Projekt 2546, November 2004.
Gutachten im Auftrag von: Sächsischen
Landesamt für Umwelt und Geologie, Dresden.
Herunterladbar unter
http://www.lohmeyer.de/Literatur.htm.
Norman, M. and Johansson, C., 2006: Studies of
some measures to reduce road dust emissions
from paved roads in Scandinavia. Atmospheric
Environment 40, 6154-6164.
UBA, 2004: Handbuch Emissionsfaktoren des
Straßenverkehrs, Version 2.1/April 2004.
Dokumentation zur Version Deutschland
erarbeitet durch INFRAS AG Bern/Schweiz in
Zusammenarbeit mit IFEU Heidelberg. Hrsg:
Umweltbundesamt Berlin. Herunterladbar unter
http://www.hbefa.net/
.
Wåhlin, P., Berkowicz, R. and Palmgren, F., 2006:
Characterisation of traffic-generated particulate
matter in Copenhagen. Atmospheric Environment
40, 2151-2159.
Table 1: Simplified version of the proposed German PM
10
emission factors for non-exhaust emissions (last two
columns) in dependence of the traffic situation. Values for a fleet mix containing 4% heavy duty traffic and
exhaust emissions according to (UBA, 2004) are given. Table adapted from Ketzel et al. (2007).
non exhaust
emission factor*
[mg/km veh]
Traffic situation
average
Speed
[km/h]
Share of
constant
speed
driving[%]
exhaust
emiss. factor
(fleet-mix)
[mg/km veh]
non-exhaust
emission factor
(fleet-mix)
[mg/km veh]
cars / vans
trucks
motorways or outside cities
60-130
22
200
Tunnel 60-100 10 200
City main road (HVS1)**
56
46
19
29
22
200
City main road (HVS2)**
44
52
20
41
30
300
City main road (HVS3)**
34
44
22
54
40
380
City main road (HVS4)**
28
37
26
66
50
450
City traffic lights (LSA2)**
24
32
28
82
60
600
City slow traffic (IO_Kern)**
17
23
32
118
90
800
* Values for good quality of the road surface, flat terrain and conditions of rain as usual in Germany.
** Speed
limit = 50 km/h;
35

Five years ultrafine and fine ambient particles number concentrations measurements
at a traffic-orientated site in Dresden
G. Löschau
1
, B. Wehner
2
and A. Wiedensohler
2
1
Saxon State Agency for Environment and Geology (LfUG), Group 22 Regional Air Quality
Department 2 - Integrative Environmental Protection, Climate, Air, Radiation
Zur Wetterwarte 11, 01109 Dresden, Germany
2
Leibniz Institute for Tropospheric Research (IfT), Department of Physics,
Permoserstr. 15, 04318 Leipzig, Germany
Keywords: Ultrafine particles, number concentration, air quality, monitoring, traffic
The European Union decided a series of directives
to improve the monitoring and evaluation of the air
quality and inform the public. In Saxony, the
concentration of most pollutants could be reduced
successfully in ambient air for the last 15 years. The
authorities tasks of monitoring could be reduced
(SO
2
) or completely stopped (TSP).
Findings of epidemiology studies refer to the
health-relevant meaning of ultrafine aerosol
particles (<100 nm). The State Agency for
Environment and Geology decided to document the
air quality situation for such ultrafine particles at a
single observation site in Saxony. This
measurement is optional without legal base
according to precautionary principle. A traffic-
orientated measuring site was chosen to collect air
quality data for many years, while amongst others
the fraction of low emission vehicles will increase.
The intention from the beginning was to integrate
the measurement into the air quality monitoring
network in Saxony. A project with the Leibniz
Institute for Tropospheric Research in Leipzig
covers not only the setup and installation of the
measuring technique, but also the training of the
personnel and the quality assurance of the data. The
measurements are performed with a TDMPS (Twin
Differential Mobility Particle Sizer). In two
separate ranges, particle number size distributions
are measured from 3 to 800 nm. The measured data
are reduced to 8 particle size classes and ½-h-
average values. The data are validated and
administered like other air pollutants in the data
base and wide-area evaluations can be made.
The measuring container “Dresden-Nord” is
situated on an open square near the railway station
Dresden Neustadt. The container is placed 5 m from
a road and 50 m to a busy crossing (about 50.000
vehicles per day with 5% heavy traffic). The
sampling point height is 3.5 m. Table 1 contains
characteristic values of the pollution levels from
this point for the last 5 years. The average number
concentration of particles in the range from 3 to 800
nm amounted to 22.923 cm
-3
. The concentration of
the ultrafine particles (3 - 100 nm) was determined
to 20.055 cm
-3
. The ultrafine particles represent
87% of all particles. Results for particle classes are
listed in table 2.
The evolution of daily averages of particle number
concentrations from 3 to 800 nm is shown in fig. 1.
A large variation of the daily values is
demonstrated. The data availability is 80%.
Safe statistic data can be made over 5 years. Fig. 2
shows an example. The middle reduction of the
load can be illustrated with the mean value of all
Sundays in relation to the mean value of all
Mondays until Fridays. It is shown, how the
average load can be reduced, if the economic life is
reduced to the Sunday-level.
The presented data set provides the first long term
measurements of number size distributions within a
regular monitoring network. The data are available
for epidemiological investigations now.
A need for a mobile aerosol standard for
examination in the measuring station is seen for the
further improvement of the quality assurance.
Thanks are given to the staffs of the Environmental
Operating Company (UBG) in Radebeul for the
special measurements.
36

image
Number [cm
-3
]
0
25.000
50.000
75.000
Aug 02
Jan 03
Aug 03
Jan 04
Aug 04
Jan 05
Aug 05
Jan 06
Aug 06
Jan 07
Aug 07
Fig. 1:
The variation of the daily averages for the number concentration of particles from 3 to 800 nm at a
traffic-orientated site in Dresden over 5 years.
Gases
number of particles
mass of particles vehicles
Fig. 2:
Mean percentile reduction on Sundays opposite on mean from Mondays to Fridays over 5 years for
gases (left), number of particles for 8 particle classes from 3 to 800 nm (middle), particle mass of
different pollutants (middle-right) at traffic-orientated measuring station “Dresden-Nord” and
number of vehicles (quite right) on a near vehicle counting point.
Table 1:
Pollutant concentration at the measuring point from 1
st
August 2002 to 31
st
July 2007
Pollutant PM
10
[μg/m³]
PM
2.5
[μg/m³]
Soot
[μg/m³]
NO
2
[μg/m³]
Benzene
[μg/m³]
CO
[mg/m³]
5 year average
33
21
4,4
46
2,4
0,6
98-percentil
1
80 56 9,3 72 6,2 1,3
Table 2:
Particles concentration in particles/cm³ at the measuring point from 1
st
August 2002 to 31
st
July
2007
Particles sizes
class
3 – 5 nm
5 – 10
nm
10 – 20
nm
20 – 50
nm
50 – 100
nm
100 –
200 nm
200 –
400 nm
400 –
800 nm
5 year average 603 3.179 6.427 6.209 3.685 2.162 672 68
98-percentil
1
1.421 6.590 13.791 12.587 6.544 3.016 1.382 214
1
related to daily averages
37

image
Ultrafine particles in NRW – case studies in urban background and at an „Autobahn“
T. A. J. Kuhlbusch, A. C. John, U. Quass
“Airborne Particle / Air quality” Unit
Institute of Energy and Environmental Technolgy (IUTA),
Bliersheimer Str. 60, D-47229 Duisburg, Germany
Keywords: SMPS measurements, weekly profiles, traffic emissions
The potential impact of ultrafine particles (UFP) on
human health is increasingly discussed in recent
years. This paper reports about UFP measurements
made at two background sites (urban and suburban,
resp., distance approx. 8 km, Duisburg area, (Quass
et al. 2004)) and upwind/downwind of a highly
trafficked motorway (A61 near Meckenheim, (Quass
et al 2007)). All measurements were made using
SMPS/CNC (TSI). PM10 mass concentrations were
measured by TEOM systems.
Urban/suburban backround measurements
At urban background, the particle number
concentration (PNC) of UFP (particle diameter < 100
nm) show distinct dependence on the day of week for
rush-hour conditions (see fig. 1), while being less
variable at midnight.
Figure 1: Size-resolved number concentration of the
urban background aerosol for different weekdays and
two daytimes (Averages of a 3-month period)
During rush hours, a slight shift of the modal
diameter was observed with increasing
concentrations indicating a higher fraction of freshly
produced combustion particles. Peak concentration
increased during the week with highest values on
Wednesday and Friday. For that daytime (09:00) a.m.
this weekly profile could be found also for PM
10
concentrations and the correlation between PM
10
and
the modal peak concentration was significant
(r² ≈ 0.7). However, at other times, this correlation
was much lower (midnight:r² ≈ 0.1). This clearly
shows that PM
10
concentrations temporarily may
significantly correlate with UFP concentrations, e.g.
during situations with high traffic emission rates.
With respect to the question of spatial
representativeness of UFP measurements a
comparison of UFP concentration data measured at
the suburban und urban site was made. Increasing
correlations were obtained for longer averaging
times, but even the highest r² values did not reach
those usually obtained for PM
10
or PM
2.5
. Clearly,
UFP concentrations are influenced more by local
processes (sources, short atmospheric lifetime) than
PM
10
, which has to be accounted for in e.g.
epidemiologic study designs.
Data averaging time
30 min
6 h
24 h
r² 0.35 0.50 0.54
Table 1: inter-site correlation coefficients for
different data averaging times (total UFP <100nm).
midnight
Highway measurements
Within a measurement program mainly intended to
derive emission factors for non-exhaust emissions a
campaign with upwind-downwind SMPS measure-
ments was carried out. During this period the average
traffic volume was ca. 2.200 vehicles/h with 16%
high duty vehicles.
Figure 2 shows the obtained number size distribution
and the downwind/upwind ratios. The traffic emis-
sions affect the particle concentrations throughout the
entire particle size spectrum. Changes of the size dis-
tribution (new modes), indicated by a change of the
downwind/upwind ratio, could be observed for the
particle size ranges of about 100 nm and for particles
below 30 nm. While the first new mode most pro-
bably consisted of soot particles, the latter may be
interpreted as fresh nucleation particles originating
from unburned or only partially combusted fuel.
9 a. m
This was further underlined by a correlation analysis
of downwind-upwind concentration differences for
particles numbers and nitrogen oxides. R² values
were calculated for the sum of PNC in each size
range given in table 2. In the size range below 50 nm
higher correlation of PNC vs. NO and NOx was
found than PNC vs. NO
2
. On the contrary, NO
2
correlated better with PNC for larger particles
(50-200 nm).
38

1
10
100
1000
10000
100000
10
100
1000
Dp [nm]
dN/dlog(Dp) [#/cm³]
0
4
8
12
16
20
24
28
32
36
40
ratio down/up
upwind
downwind
ratio
Figure 2: Upwind and downwind PNC size distribution (mean and 25/75 percentiles) and downwind/upwind
ratio (N=371 measurements, 0.5 h averages)
<50 nm
UFP
50-200 nm 200-500 nm Total Number
NO2
0.42
0.46
0.69
0.23
0.47
NOx
0.62
0.66
0.67
0.32
0.67
NO
0.63
0.66
0.60
0.32
0.67
PM1
0.51
0.54
0.52
0.13
0.55
PM10
0.25
0.25
0.16
0.02
0.25
PM1-10
0.06
0.06
0.06
0.00
0.06
Table 2: r² correlation coefficients between down-
wind-upwind differences of PNCs (for 3 size ranges)
and of various pollutant concentrations
R²-coefficients with PM concentrations are lower but
still indicate correlation in case of PM
1
, whereas
PM
10
and PM
1-10
do not correlate at all with ultrafine
and fine PNCs.
Comparing the size distributions and PNCs in the
urban background with those obtained at the highway
considerable differences are obvious:
The pronounced size mode around 30-50 nm
observed in the urban aerosol is not found
near the highway. There, smaller particles
<30 nm have a relatively higher
concentration
Above 30 nm, PNCs even downwind the
highway do not reach the concentrations
found in urban environment. This is
probably due to the low background
contribution (1 order of magnitude lower
than midnight concentration in the city) and
good dispersion conditions at the
investigated highway site.
Summary and conclusions
Particle numbers and number size distributions
measured at urban background and near a highway in
rural environment are presented. The urban aerosol
shows a distinct daytime dependence and a weekly
variation partially correlating to that of PM
10
.
Correlation with PM concentrations is significant
only in high emission situations (rush hour). The
spatial representativeness is weaker than for PM due
to higher importance of local sources and lower
particle lifetime. The modal diameter is usually
found between 30 and 60 nm. This mode was also
present in the aerosol downwind the highway,
however a strong mode of finer particles could be
observed additionally. Number concentrations of this
mode better correlate with NO than with NO
2
,
indicating fresh particle production.
Acknowledgements
This work was supported by the North Rhine-
Westphalian Ministry of Environment (MUNLV)
and the Federal Highway Research Institute (BASt).
Citations
Quass, U., Kuhlbusch. T. & Koch, M. (2004).
Identifizierung von Quellgruppen für die
Feinstaubfraktion
. IUTA report LP15/2004
Quass, U., John, A., Beyer, M.,Lindermann, J.,
Hirner, A.V., Sulkowski, M.&M., Hippler, J &.
Kuhlbusch, T.A.J.. (2007)
Ermittlung des
Beitrages von Reifen-, Kupplungs-, Brems- und
Fahrbahnabrieb an den PM10-Emissionen von
Straßen.
IUTA report LP28/2007
39

image
image
Ultrafine and fine particle measurements in Switzerland at various stations and on
different roads
A.S.H. Prévôt
1
, E. Weingartner
1
, S. Weimer
1
, J. Sandradewi
1
, M.R. Alfarra
1
, V. Lanz
2
, C. Hueglin
2
, S. Szidat
3
,
U. Baltensperger
1
1
Laboratory of Atmospheric Chemistry, Paul Scherrer Institute, 5232 Villigen, Switzerland
2
Laboratory for Air Pollution and Environmental Technology, Empa, Swiss Federal Laboratories for Materials
Testing and Research, 8600 Dübendorf, Switzerland
2
Department of Chemistry and Biochemistry, University of Bern, 3000 Bern, Switzerland
Keywords: mobile measurements, aerosol mass spectrometry,
14
C-analysis, FMPS (fast mobility particle sizer)
In the last 2-3 years several field campaigns
have been performed in Switzerland to study the
aerosol composition their sources and the aerosol
number size distributions (Figure 1). Some
campaigns were performed at fixed locations in
urban, rural, and Alpine areas. In addition, mobile
measurements allowed the characterisation of the
spatial distribution of the chemical composition and
the aerosol size distribution.
Reiden (
February
)
Roveredo (
March,
December
)
Härkingen (
May
)
Payerne (
June
,
January
)
Massongex
(
December
)
JFJ (
Winter
,
Summer
)
Riviera (
June
)
Milano (
June
)
Mobile measurements
Zürich (
May
)
Figure 1. Locations of the Aerodyne aerosol mass
spectrometer measurements from 2005-2007 in
Switzerland.
The main instrumentation discussed in this
presentation were the Aerodyne aerosol mass
spectrometer to detect the volatile PM1 components,
the
14
C analysis for the quantification of the fossil
and non-fossil EC and OC fractions and the TSI fast
mobility particle sizer to characterize the aerosol
number size distribution from 5.6 to 560 nanometers.
Measurements in Roveredo (Alpine valley along the
Transalpine San Bernardino Route) showed that the
particulate matter, especially the organic mass is
strongly dominated by wood burning particles. This
could be shown both by the aerosol mass
spectrometer measurements and the
14
C analyses
(Alfarra
et al.
, 2007, Szidat
et al.
, 2007).
The results of the studies in the Alpine valley are
however not representative for the rest of
Switzerland. On the Swiss Plateau, secondary
inorganics ammonium nitrate (especially in winter)
and ammonium sulfate are the main fraction of PM1.
The elemental carbon is mostly due to fossil sources
(especially traffic) (Szidat et al., 2006). The organics
in Zürich summer were mostly secondary (2/3) and
wood burning, charbroiling and wood burning were
similarly important (Lanz et al., 2007a). In winter,
surprisingly also half of the organics are secondary,
and wood burning is about 3 times more important
than traffic (Lanz et al., 2007b). The secondary
organics in winter and summer are non-fossil,
probably coming mostly from terpene and isoprene
emissions in summer and volatile wood burning
emissions in winter.
Mobile measurements in various locations have
shown that the aerosol volume (D<560 nm) is often
highest in villages (especially in winter) whereas the
nanoparticles (D<50 nm) are by far highest on the
highway due to the high load and high fractions of
diesel vehicles.
In summary, wood burning is very important in
Switzerland during winter for the particulate mass.
Traffic contributes significantly aerosol number
(especially nanoparticles), to elemental carbon also to
ammonium nitrate (due to the NO
x
emissions).
Alfarra, M.R. et al., (2007) Identification of the mass
signature of organic aerosols from wood burning
emissions,
Environ. Sci. Technol.
, 41, 5770-5777.
Lanz, V.A. et al. (2007a) Source apportionment of
submicron organic aerosols at an urban site by
factor analytical modelling of aerosol mass
spectra, Atmos. Chem. Phys., 7, 1503-1522.
Lanz, V.A. et al., (2007b) Source attribution of
submicron organic aerosols during wintertime
inversions by advanced factor analysis of aerosol
mass spectra, Environ. Sci. Technol., in press.
Szidat, S. et al., (2006) Contributions of fossil fuel,
biomass-burning, and biogenic emissions to
carbonaceous aerosols in Zurich as traces by 14C,
J. Geophys. Res., 111, D07206.
Szidat, S. et al. (2007) Dominant impact of
residential wood burning on particulate matter in
Alpine valleys during winter,
Geophys. Res. Lett.
,
34, L05820.
40

Temporal and spatial variability of sub-μm aerosol concentrations
in the urban atmosphere of Leipzig
Wolfram Birmili
1
, Susan Klose
1
, Maik Merkel
1
, Birgit Wehner
1
, Korinna König
1
, André Sonntag
1
,
Alfred Wiedensohler
1
, Oswald Knoth
1
, Detlef Hinneburg
1
, Thomas Tuch
1,2
, and Ulrich Franck
2
1
Leibniz Institute for Tropospheric Research, Permoserstrasse 15, 04318 Leipzig, Germany
2
UFZ - Centre for Environmental Research, Permoserstrasse 15, 04318 Leipzig, Germany
Session “Measurement of UFPs in ambient air II” on October 24, 2007.
Keywords: particle exposure, spatial variability, traffic aerosols, UFP exposure
INTRODUCTION
Scientific evidence has consolidated associations
between environmental aerosols (also termed “PM”,
particulate matter) and negative health effects, such
as asthma and cardiopulmonary disease (WHO,
2004). While there is little debate about whether
these effects are real, numerous questions have arisen
which particular types (i.e., sub-fractions) of
environmental particles are responsible for the health
risk, and how the exposure risks for the population
may be reduced in a most efficient way (HEI, 2002).
In urban environments, fine (< 1 μm) and
ultrafine (< 0.1 μm) particles, as well as number
concentration of particles, have moved into the focus
of public health interests. The rationale is that
ultrafine particles are inhalable into the deep lung,
contain increased amounts of soot, aromatic organics,
and trace metals, and account for the majority of
particle number in urban atmospheres.
Current estimates of exposure to fine and
ultrafine particles, however, are particularly ham-
pered by the lack of knowledge when characterising
the spatial and temporal variability of fine and
ultrafine aerosols within urban landscapes. Several
studies have indeed suggested that especially
ultrafine particles are highly variable with location
(Hussein et al., 2005, Tuch et al., 2006). Before
particle metrics including fine and ultrafine particles
can be considered for a systematic monitoring, a
broader scientific understanding of the behaviour of
these aerosol properties in an urban atmosphere is
required.
FIELD EXPERIMENT
A series of field experiments (“PURAT” – Par-
ticles in the urban atmosphere: Behaviour of fine and
ultrafine particles, their spatial variation and
relationships with local policy action) was conceived
to investigate the spatial and temporal distribution of
urban aerosols on different urban length scales.
Long-term measurements between 2004 and 2005
in a street canyon and at an urban background site in
Leipzig, Germany (population ca. 500,000) were
used to evaluate the general impact of traffic density
and meteorological processes on urban aerosol
concentrations (PURAT-0).
A second experiment (PURAT-1) involved simul-
taneous particle size distribution measurements at 6
fixed observation sites in Leipzig: Five out of the six
fixed measurement sites were within a distance of 3
km in the downwind plume of Leipzig’s city centre.
The character of the observation sites spanned a
range between two “roadside” sites (i.e., distance to a
major road < 10 m), and four urban background sites.
In addition, a mobile aerosol laboratory was
deployed to measure number size distributions at
additional 7
th
site.
Particle size distributions were determined using
various scanning mobility analysers (SMPS). In order
to avoid systematic bias due to instrumental differ-
ences regular intercomparisons were made between
individual SMPS systems before and during the
experiment. Individual instruments showed system-
atic differences in their readings, and this needed to
be taken into account when evaluating the final size
distribution data set.
RESULTS
An example for the observed variations in particle
size distributions during PURAT-1 is given in Figure
1, showing simultaneously measured size distribtions
at 7 measurement sites. As expected, particle
concentrations were generally increased near sources
of motorised traffic. This concerned especially the
size range of ultrafine particles (< 100 nm).
However, significant differences were also observed
between urban background sites.
We next examined the correlation of particle
number size distributions between different locations
by evaluating linear correlations of particle number
concentrations. Figure 2 shows the measures of
determination (R
2
) obtained for the particle size
interval 40-120 nm (Aitken particles).
41

3
10
100
800
0
2x10
4
4x10
4
6x10
4
8x10
4
1x10
5
D
p
, nm
d
N
/ dlog
D
p
, cm
-3
roadside I
roadside II
roadside III
urban background I
urban background II
urban background III
urban background IV
Figure 1: Particle size distributions measured
simultaneously at seven urban sampling locations.
Exemplary data from a morning rush hour episode in
Leipzig, April 13, 2005, 0530-0700 h.
RS I
RS III
UB I
UB II
UB III
UB IV
RS I
1
0.42
0.61
0.53
0.53
0.43
RS III
0.42
1
0.67
0.52
0.66
0.57
UB I
0.61
0.67
1
0.87
0.90
0.72
UB II
0.53
0.52
0.87
1
0.87
0.66
UB III
0.53
0.66
0.90
0.87
1
0.72
UB IV
0.43
0.57
0.72
0.66
0.72
1
Figure 2: Measure of determination (R
2
) of linear
correlations between particle number concentrations
(size interval 40-120 nm) at 6 different urban
observation sites. “RS” denotes the roadside, “UB”
the urban background sites referred to in Figure 1.
Data coverage is April 7, to May 8, 2005.
The general conclusions from the correlation
matrixes were:
Aitken particle concentrations (40-120 nm)
correlate well between urban background sites.
(The best correlation was found between two
neighbouring urban background sites.)
Particle concentrations at roadside sites correlate
less with urban background concentrations, and
eventually even less between different roadside
sites.
Correlations increase for accumulation mode
particles (> 120 nm), but decrease for nucleation
mode particles (10-40 nm).
Concentrations at all particle sizes correlate
positively with the density of traffic in the
proximity of the measurement sites.
Besides the results obtained from the experiment
PURAT-1, this presentation will also give a summary
of the long-term dual point measurements (PURAT-
0) as well as of a subsequent experiment PURAT-3,
which concentrated on an even smaller transport
length scale 1-100 m.
For the region around the Eisenbahnstrasse street
canyon in Leipzig, microscale transport simulations
of traffic aerosols will be presented.
CONCLUSIONS
During multiple site experiments in Leipzig, the
spatial and temporal variations of aerosol particle
number distributions were characterised across a
range of several observation sites. The results
demonstrate that UPFs tend to be much more
variable within the complex terrain of a city than, for
instance particle mass.
The impact of localised and time-dependent
traffic sources in an urban landscape causes poor
correlations between roadside and urban background
observations. The best correlations in Aitken mode
particles were found between neighbouring
background sites (R
2
=0.90). These results indicate
that the effects of particle emission and transport are
complex in an urban atmosphere, and that a single
point measurement of ultrafine aerosol concentration
in a city may be representative within a limited
spatial radius only. This experimental data set will be
used to validate numerical emission and transport
models in the future.
REFERENCES
HEI (2002), Understanding the health effects of
components of the particulate matter mix: progress and
next steps. Health Effects Institute, 4, Boston, MA.
Hussein, T., Hämeri, K., Aalto, P. P., Kulmala, M. (2005)
Modal structure and spatial–temporal variations of
urban and suburban aerosols in Helsinki–Finland.
Atmos. Env. 39, 1655-1668.
Tuch, Th. M., Herbarth, O., Franck, U., Peters, A. Wehner,
B., Wiedensohler, A., Heintzenberg, J. (2006a) Weak
correlation of ultrafine aerosol particle concentrations <
800 nm between two sites within one city. J. Exposure
Sci Env. Epid., 16, 486–491.
WHO (2004), Health Effects of Air Pollution Results from
the WHO Project “Systematic Review of Health
Aspects of Air Pollution in Europe”. World Health
Organisation, No. E83080, Geneva
We acknowledge broad support by Dr. Dieter Bake
(German Federal Environment Agency – UBA,
Berlin). This work was supported by UBA contract
UFOPLAN No. 20442204/03, and the EU Marie
Curie Reintegration Grant FP6-2002-Mobility-11
contract No. 510583.
42

Chemical composition of aerosol particles including UFPs in Saxony
T. Gnauk
1)
, E. Brüggemann
1)
, H. Gerwig
2)
, H. Herrmann
1)
, K. Müller
1)
, G. Spindler
1)
1) Leibniz-Institut für Troposphärenforschung, Chemistry Dept., Permoserstr. 15, 04318 Leipzig
2) Sächsisches Landesamt für Umwelt und Geologie, Referat Luftqualität, Zur Wetterwarte 11, 01109
Dresden
Keywords: particulate mass, ions, OC/EC, organics, mass closure, source apportionment
Size-segregated particle sampling using
impactors in different seasons and at differently
polluted sites in Saxony in combination with
trajectory analysis for the air mass origin estimation
can be used to find out particle sources, fractions of
different sources and long-range transport effects.
During the last decade a set of projects was
accomplished by the IfT in Saxony including MINT
97 (Melpitz Intensive Measurement Campaign),
Feinstaub 1999/2000 (a source study of PM in
Saxony (I)), Schwebstaub 2003/04 (a study of size-
segregated chemical composition in Saxony (II)),
Ferneintrag 2006/07 (a study of PM long-range
transport in Saxony (III)) or the FAT project 2004/05
(examination of PM in a street canyon). Detailed
descriptions of the measurement campaigns were
published elsewhere (Plewka et al., 2004; Herrmann
et al., 2006; Gerwig et al., 2006; Brüggemann et al.,
2007). Several interesting results will be presented
here:
Significant differences in chemical composition of
particles in different size ranges depending of air
mass origin (east-west contrast)
Fig.1: TC long range transport from East in Winter
Seasonal differences of particle composition by
size-segregated measurements at a roadside in
Dresden (traffic and domestic heating influence)
Source apportionment by size-segregated particle
component concentration measurements in summer
and winter at a traffic, urban background, and
regional background station
Correlation of particulate OC/EC in different size
ranges with traffic density in a busy street canyon
Some concluding remarks are given about the
limited possibilities of local authorities to avoid
exceedances of the PM
10
limit value of 50 μgm
-3
in
the streets. Mass distribution shows a maximum
Fig.2: Contribution of local traffic to TC in a street
canyon depending of particle size range
(about 50%) in the size range of Dp
aer
= 0.42-1.2 μm
originating mostly from long-range transport
processes. Small particles from local emissions
account for a high fraction of the number
concentration, but only for a small fraction of the
mass concentration.
References for more information:
Plewka, A., Gnauk, T., Brüggemann, E., Neusüss,
C., Herrmann, H., 2004. Size-resolved aerosol
characterization for a polluted episode at the IfT
research station Melpitz in Autumn 1997. Journal of
Atmospheric Chemistry 48, 131-156.
Herrmann, H., Brüggemann, E., Franck, U., Gnauk,
T., Löschau, G., Müller, K., Plewka, A., Spindler, G.,
2006. A source study of PM in Saxony by size-
segregated characterisation. Journal of Atmospheric
Chemistry 55, 103-130.
Gerwig, H., Bittner, H., Brüggemann, E., Gnauk, T.,
Herrmann, H., Löschau, G., Müller, K., 2006.
Quellgruppenquantifizierung von PM
10
an einer
Verkehrsmessstation in Dresden. Gefahrstoffe –
Reinhaltung der Luft 66, 175-180.
Brüggemann, E., Gnauk, T., Müller, K., van
Pinxteren, D., Plewka, A., Herrmann, H., Birmili,
W., Tuch, T., Wehner, B., Wiedensohler, A., 2007.
Größenaufgelöste physikalische und chemische
Bestimmung von elementarem und organischem
Kohlenstoff in Nanopartikeln. FAT
(Forschungsvereinigung Automobiltechnik e.V.) –
Schriftenreihe 206, 138 Seiten.
TC MOUDI Winter 2005 Eisenbahnstr.
0
1
2
3
0.010-0.018
0.018-0.032
0.032-0.056
0.056-0.10
0.10-0.18
0.18-0.32
0.32-0.56
0.56-1.0
1.0-1.8
1.8-3.2
3.2-5.6
5.6-10
10-18
>18
Dp(aer) [μm]
conc. [μg/m³]
Run 1
Run 2
Run 3
Run 4
East
North
0.0
0.5
1.0
1.5
2.0
2.5
0.05-0.14
0.14-0.42
0.42-1.2
1.2-3.5
3.5-10
Dp(aer) [μm]
TC [μg/m³]
local traffic
urban background
rural background
80%
37%
46%
32%
61%
28%
21%
57%
54%
17%
43

image
PM abatement from a European perspective –
current legislation and the CAFE thematic strategy
P. Bruckmann
1
1
Landesamt für Natur, Umwelt und Verbraucherschutz NRW, D-45610 Recklinghausen, Germany
Keywords: particle metrics, upcoming Air Quality Directive, European legislation.
Figure 1.
Estimated losses in life expectancy (in months) attributable to exposure to fine particulates 2000
and 2020; Source: IIASA, 2005
Attainment of the current PM10 limit values laid
down in the EC legislation poses big problems in the
majority of the Member States. European PM2.5 data
are still incomplete. Until the end of 2006, about 140
plans and programmes to abate air pollutions have
been set up.
With a view to attain an European air quality which
has no significant negative impacts on health until
2020, the Commission has launched the thematic
strategy “Clean Air For Europe” (CAFE) in 2005. As
a political paper, this strategy lists important
legislative projects such as the revision of the air
quality legislation or the NEC directive. The CAFE
strategy was underpinned by a comprehensive set of
scientific reports, inter alia the 2nd PM Position
Paper, an updated assessment of health effects by
WHO, or an integrated assessment modelling of
abatement options. PM turns out to be the pollutant
with the most severe impact on public health, leading
to a loss of life expectancy of more than 12 months in
parts of Europe (compare Fig. 1). According to the
scientific judgement, PM2.5 is strongly associated
with mortality and cardiopulmonary disease and was
therefore chosen as principal PM metric for the
upcoming regulation. It was felt too early to regulate
UFP or particle numbers, but more research such as
determining and linking exposure to UFP with health
effects was recommended.
The key elements of the Common position of
Council and EC Commission for a new Air Quality
directive as well as the state of proceedings are
outlined in the paper. In respect to PM, the existing
PM10 limit values (annual and daily means) shall
remain unchanged, but the attainment period may be
extended by 3 years under certain conditions. The
PM2.5 burden shall be regulated by a target value of
25 μg/m³ which applies everywhere, which will
become legally binding in 2015. In addition, the
average national exposure of the general public in big
cities shall be reduced by 20 % going from 2010 to
2020. A revision of the directive is foreseen in 2013
with a view of making this reduction target
mandatory. Negotiations between Council,
Commission and European Parliament before the 2nd
reading in Parliament, which will probably take place
in December 2007, are still running. Main remaining
points of discussion are the regulations for PM2.5,
the applicability of the limit values and the
prolongation of attainment dates.
44

image
image
LIFE-ENVIRONMENT Project KAPA GS
Klagenfurts Anti PM10 Action Programme with Graz and the South Tyrol
Wolfgang Hafner
1
, Christian Kurz
2
, Gerhard Bachler
2
, Peter Sturm
2
1) City of Klagenfurt, Department Environment Protection, Bahnhofstrasse 35, 9020 Klagenfurt, Austria.
2) Graz University of Technology, Institute for Internal Combustion Engines and Thermodynamics, Inffeldgasse
21A, 8010Graz, Austria.
Keywords: PM10 mitigation options, dispersion model, traffic bans, resupension, domestic fuels, CMA
industry, power
station, trade
3%
backg round
41%
domestic fuel
9%
traffic - non ex
38%
traffic ex
9%
In many European cities the high pollution of
particulate matter is an enormous environmental
and health problem. Within the project KAPA GS
several measures have been simulated and
demonstrated from 2004 – 2007 in order to tackle
the rising particle pollution during the winter
period.
A dispersion model considering all known
emissions has been developed by the Technical
University of Graz and validated by monitoring
stations for PM10 / PM2,5 / PM1.0 and NOx.
Additionally a statistical forecast model enables to
forecast the PM10 concentration (daily mean) for
the next day in Klagenfurt, Graz and Bozen. A
specific innovation has been done in Klagenfurt by
linking the dispersion model with current wind and
PM10-data: the result is an hourly updated
nowcasting model.
Figure 1. Dispersion model: annual Mean of PM10
in Klagenfurt, base case 2005
The dispersion model figures out the sources of
pollution within a grid of 10 X 10 m.
Figure 2. Sources apportionment of a hot spot in
Klagenfurt, Völkermarkter Strasse, 23.000 ADT
winter mean
The model is the main tool for the simulation of
measures to reduce PM10 and calculate szenarios
for mitigation strategies.
Figure 3. The traffic ban in the inner city of
Klagenfurt (Neuer Platz/ Burggasse) reduces the
PM10-level up to 3 μg/m3 (annual mean). .
Due to the remarkable contribution of re-suspension
(38% of PM10 winter mean value) new road
sweeping machines and CMA as dust-binder were
tested.
45

image
Figure 4. Delta PM10/ delta NOx ratio in winter
2004/2005, 2005/2006, 2006/2007(monthly mean,
Völkermarkter Strasse). During periods with CMA-
application a significant reduction of the PM10-
level (up to -17% in January) has been monitored.
Moreover, to reduce traffic exhaust emissions the
following solutions were tested and implemented:
Park and Ride System with special anti-PM10-
shuttle-bus, car free days, traffic bans and
environmental zones, retrofitting of the city bus
fleet in Graz und Klagenfurt by PM-catalytic
converters and PM-filter systems.
PM-pollution from domestic fuel was tackled by
intensive promotion campaign aiming at
substitution of individual heating systems by
natural gas or district heating connections.
8 electronic indicator boards (video walls) inform
the population about the current PM 10-levels,
park&ride possibilities and traffic bans, when
paasing by on the main roads in Klagenfurt.
Many actions were initiated in the field of public
awareness and dissemination to motivate people
participating actively in setting activities against the
PM10 pollution. Information campaigns in schools,
folder, advertising campaigns, placements in
newspapers, press conferences etc. have informed
the public about the problem of the PM10 pollution
and the EU-Project KAPA GS.
The two international project conferences
(November 2005 in Graz, March 2007 in
Klagenfurt) have been very important for the
exchange of experiences to find a common solution.
The project KAPA GS demonstrates best practice
for other cities in Europe.
More information and downloads:
www.kapags.at
www.feinstaubfrei.at
Many measures tested and implemented within the
project KAPA GS were very effective.
Klagenfurt has to save 50 days exceeding the limit
value for PM10. Without the prognosticated traffic
increase this target could be met in 2020 even on a
hot spot in Klagenfurt by combining technical and
planning measures.
Figure 5. Reduction potential of measures in
Klagenfurt, calculated in number of days exceeding
the daily limit value of PM10.
Figure 6. For some measures cost-benefit-analyses
have been carried out.
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
Okt
Nov
Dez
Jän
Feb
Mär
Apr
Mai
delta PM10/delta NOx
2004/05
2005/06
2006/07
mit CMA
ohne CMA
15.0
3
11.8
12.2
5.2
1.4
3.8
-14.8
-20.0
-10.0
0.0
10.0
20.0
30.0
40.0
50.0
60.0
number of days > 50 μg/m3
increase of traffic to
2020 (+51 %)
secondary particles,
backgr ound
particle filter for
HDV (90%)
particle filter for
busses (90%)
particle filter für PC
(90%)
domestic fuels -
45.5%
no split, washing
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
costs per year
Mio €
CMA
Particle filter PC
domestic fuel
Park&Ride W u O
Figure 7. Best case Scenario for Klagenfurt 2020:
Without traffic increase the PM10-levels comply
with limit values in living areas.
46

The PM and NO
x
air pollution in Copenhagen
and assessment of possible measures to reduce the air pollution
F. Palmgren
National Environmental Research Institute, University of ˉrhus, DK-4000, Roskilde, Denmark
Keywords: Traffic, scenarios, environmental zones, ultra fine particles
Particulate matter (PM) and NO
x
/NO
2
are the
most serious air pollution problem, in Copenhagen
and in many other large cities in Europe. The main
sources in Copenhagen to these pollutants are the
local road traffic and the long range transported
pollutants from Europe. Regulations of PM and
NO
x
/NO
2
are closely related, because the road traffic
is an important source to both.
NO
x
, NO
2
and PM
10
/TSP have been measured
in the Danish Urban Air Pollution Monitoring
program for many years. The monitoring strategy is
based on simultaneous measurements at rural
background, urban background and at “hot spots”
(streets) in combination with air quality modelling
(Kemp et al., 2007).
In addition, we made measurements of
ultrafine particles in size fractions 10-700 nm by
DMPS (Differential Mobility Particle Sizer).
Receptor modelling has been used for source
apportionment of the different pollutants (W?hlin,
2003)., and source-receptor modelling for analysis of
the measurement data and for scenario calculations
Both types of modelling are used for assessment of
the air pollution in relation to the EU air quality
directives now and in the future (Brandt and
Palmgren, 2005).
NO
x
shows a clear decreasing trend in busy
streets due to the stricter emissions standards for
motor vehicles. NO
2
has been nearly constant in
streets, because ozone is the limiting factor for
formation of NO
2
, until recently where we observe a
weak increase. This was probably due to more diesel
cars, which emit NO
2
directly. This means that the
limit value for annual average of NO
2
is exceeded
now at busy streets and scenario calculations show
that the limit value will be exceeded many years from
now, if no other measures will be taken.
PM
10
shows generally a decreasing trend, but
the limit value of annual averaged will be exceeded
at busy streets. However, the local contribution is
rather low compared to the regional contribution
(figure 1), which makes it difficult to reduce the PM
pollution locally.
The ultrafine particles in a street can roughly
be divided in 4 main parts, long range transport, soot,
and 2 condensation parts from local traffic, figure 2
(W?hlin, Berkowicz and Palmgren, 2006)
0
10000
20000
30000
40000
50000
60000
70000
80000
Number concentration (cm
-3
)
-3
Long range
Soot
Condensates
Condensates
with sulphur
Monday
Tuesday
Wednesday Thursday
Friday
Saturday
Sunday
0
5
10
15
20
25
30
PM
”0.7”
concentration (μg/m
3
)
Monday
Tuesday
Wednesday Thursday
Friday
Saturday
Sunday
Figure 2. Ultrafine particles in a busy street in
Copenhagen. Upper graph is the number
concentration, and the lower graph is the volume
concentration, both of the average week.
Investigations have been made of the effect of
different types of environmental zones in central
Copenhagen, focussing on retrofit of particles filters
on heavy duty vehicles. Scenario calculations were
performed for 3 different scenarios.
The results of one scenario, where all heavy
duty vehicles should be equipped with particles
filters, if they have EURO III engines or older.
Results are shown in figure 3. The reduction of PM
2.5
in urban background will be rather small, i.e. below
0.4 ?g/m
3
, which should be compared with the
annual average, approx. 20 ?g/m
3
.
Similar scenario calculations were performed
for ultrafine particles. The reductions were up to
2,000 (N/cm
3
), which should be compared to the
annual average 8,000 (N/cm
3
).
All scenario calculations were performed for
urban background, because it was assumed to be
most representative for the exposure of the
population. The relative reductions at busy streets
will be much larger.
PM2.5 in street and urban background
0
5
10
15
20
25
30
Street
Urban background
PM
2.5
(μg/m
3
)
Exhaust
Brakes
Road, tires etc.
Road salt
Secondary
organic+unknown
Primary non-traffic
Secondary
inorganic
Natural (soil, sea
etc)
Figure 1. The different PM
2.5
contributions in streets
and urban background.
47

714000
716000
718000
720000
722000
724000
726000
728000
6168000
6170000
6172000
6174000
6176000
6178000
6180000
6182000
6184000
6186000
Reduction PM2.5 (
µ
g/m
3
) Scenario 1.1
0.00 to 0.05
0.05 to 0.10
0.10 to 0.15
0.15 to 0.20
0.20 to 0.25
0.25 to 0.30
0.30 to 0.36
714000
716000
718000
720000
722000
724000
726000
728000
6168000
6170000
6172000
6174000
6176000
6178000
6180000
6182000
6184000
6186000
Reduction ToN (N/cm
3
) Scenario 1.1
0 to 310
310 to 620
620 to 930
930 to 1240
1240 to 1550
1550 to 1860
1860 to 2170
Figure 3. Expected reduction of PM
2.5
(upper graph)
and ultrafine particles (lower graph) in urban
background in Copenhagen, if all heavy duty EURO
III vehicles or older were equipped with particles
filters.
In addition to traffic wood combustion in wood
stoves is an important source to PM in Denmark,
especially in the residential areas around the
bigger cities (Glasius et al. 2006).
Similar methods have been used to assess the
effect in Denmark of the EU Thematic Strategy
for air pollution.
This work was supported by the Danish
Environmental Protection Agency.
References
Brandt, J. & Palmgren, F. (2005):
Integrated
modelling and monitoring for use in forecasting.
Workshop on real time air pollution data
exchange and forecast in Europe, EEA,
Copenhagen, Denmark, 7-8 April 2005.
Glasius, M., Ketzel, M., W?hlin, P., Jensen, B.,
Młnster, J.G., Berkowicz, R. & Palmgren, F.
(2006):
Impact of wood combustion on particle
levels in a residential area in Denmark.
-
Atmospheric Environment 40(37): 7115-7124.
Kemp, K., Ellermann, T., Brandt, J., Christensen, J.
& Ketzel, M. (2007):
The Danish Air Quality
Monitoring Programme. Annual Summary for
2006.
National Environmental Research Institute,
University of Aarhus. - NERI Technical Report
623: 41 pp. (electronic).
W?hlin, P., Berkowicz, R. & Palmgren, F. (2006):
Characterisation of traffic-generated particulate
matter in Copenhagen.
- Atmospheric
Environment 40(12): 2151-2159.
W?hlin, P. (2003): COPREM–A multivariate
receptor model with a physical approach. -
Atmospheric Environment 37(35): 4861-4867.
48

image
Implementation of air quality directives in a candidate state
A Turkish - German Twinning Project under the project leadership of the German
Federal Ministry for the Environment, Nature Conservation and Nuclear Safety
D. Gömer
1
1
FREIE UND HANSESTADT HAMBURG, BEHÖRDE FÜR SOZIALES, FAMILIE, GESUNDHEIT UND
VERBRAUCHERSCHUTZ, Institut für Hygiene und Umwelt, Bereich Umweltuntersuchungen,
Marckmannstr.129a/b, 20539 Hamburg
Keywords: Twinning Project, Air Quality Directives, Transposition and Implementation..
The Twinning Project “Air Quality” started in
October 2004 and ended in December 2006. The
Turkish Ministry of Environment and Forestry
(MoEF) and the German Federal Ministry for
Environment, Nature Conversation and Nuclear
Safety (BMU) were partners in this project. It had to
fulfill four main tasks:
-
Transposition of the Air Quality Framework
Directive 96/62/EC and the Large
Combustion Plants Directive 2001/80/EC
into Turkish (Draft) Regulation
-
Draft Agreed Framework Regulation on Air
Quality which defines the Roles and the
Responsibilities of the involved ministries
(considering both directives)
-
Strengthening of the qualification of the
administration (Know-How-Transfer) –
Strengthening of the Quality management
and preparation of the accreditation of the
two laboratories – Refik Saydam Hygienic
Center (RSHC) and Gölbasi
-
Agreed strategic Action Plans on further
implementation steps of the two directives
During the first year legal gap analysis, a structure
analysis concerning the administration and
assessments were carried out. Two drafts have been
developed, one for a Turkish Air Quality Framework
by-law and one for a Large Combustion Plant by-
law. Both drafts consider for the transition time the
old regulation. These by-laws are supposed to fulfill
the requirements of the European Air Quality
Framework and Large Combustion Directives. In
addition, trainings were arranged for strengthening
the Quality Management in the two above mentioned
laboratories. Additional trainings and working groups
informed about topics like zones and agglomerations,
preliminary assessment and assessment of air quality,
emission inventories, air pollution modeling, data
validation, reporting, permission and inspection
procedures etc. One result of these activities was a
proposal concerning zones and agglomerations in
Turkey. Another important result was the
implementation of the EURAD model (Rhenish
Institute in Cologne, University of Cologne,
Germany) at the Turkish Meteorological Service
which is part of the MoEF.
In connection with the Twinning project the
equipment for an ambient air pollution network for
Ankara was ordered and was delivered shortly before
the end of the project. Trainings about maintenance,
repairs,
quality
assurance
and
accreditation
procedures were given.
The second year of the Twinning project was
generally concerned with the preparation for the
implementation of the above mentioned Directives.
Cities like Istanbul, Izmir, Bursa, Erzurum were
visited in a campaign to check the automatic
instruments which were already in place in different
local networks. The station in Erzurum is part of the
national network. The MoEF built up this network
during the time of the project from it’s own
resources. Therefore some of the experts who were
employed in the project were also involved in these
activities. Each one of the 81 provinces received one
station which is equipped with one SO
2
monitor and
one PM10 monitor. The MoEF plans to equip these
stations with additional components.
Calibration and maintenance tasks are normally
carried out by companies or in the case of Izmir by
the local university. In general, gas cylinders with
certificates are used for calibration.
In addition to practical trainings two strategy papers
were prepared for the future implementation of the
prepared
by-laws.
These
strategy
papers
recommended new administration structures for the
implementation of the directives and gave estimates
for the costs concerning the implementation of the
directives.
The following proposal concerning future ambient air
pollution network stations and regional network
centres for Turkey was made: see Figure 1.
49

image
Figure 1: Proposal for future ambient air pollution
networks and regional network centres
Main future tasks are:
The two by – law drafts need to become
effective in order to provide legal certainty
for future investors
Further trainings on running an air pollution
network are needed
Build up of a nationwide quality
management system to ensure the equal
standard of the measured emission and air
quality data in all parts of Turkey
(introducing a primary standard, round
robin tests, cooperation with international
institutions etc.)
Build up of emission data bases (national
and regional); ensuring the quality of the
data
Further implementation of models (national
and regional); validation of the model
results
Build up of action and clean air plans for
regional areas and agglomerations
where
the limit values have been exceeded
This work was supported under Project Title “Air
Quality, Chemicals, Waste; Component 1: Air
Quality”, TR 03-IB-EN-01 by the European
Commission
The project leadership was carried out by the German
Federal Ministry for the Environment, Nature
Conservation and Nuclear Safety
Reports and strategy papers: unpublished information
Distribution of station types in the
regional networks
re residential station
tr traffic station
#
ru
rural station
12
#
To total number of stations per network
(inclusive additional ozone monitoring stations in agglomerations)
#
#
Regional network center
re 12 tr 5 in 4 ru 4 To 25
re18 tr 4 ru 8
re 19 tr 9 in 7 ru 4 To 39
re11 tr 4 in2 ru 5 To 22
re 10 tr 5 in 5 ru 4 To 24
re15 tr 6 in5 ru 7 To 33
re7 tr 3 in 1 ru 5 To 16
re10 tr 3 in 0 ru 6 To 19
re15 tr 8 in1 ru 7 To 31
#
in industry station
50

image
image
Street-Detailed Calculation Of The Air Quality In Saxony
Uwe Wolf
Sächsisches Landesamt für Umwelt und Geologie
Keywords: air quality, immission register, spread calculation
The Saxony immission register is an
outstanding tool to calculate and display the air
quality from NO
2
and PM
10
in Saxony
Through the combination from different
disperation models is an area-wide calculation of the
immission situation in rural areas as well as street-
detailed calculation in urban areas possible.
Picture 1, modelled yearly average values of PM
10
-
pollution (average 2001-2005)
Following methods were used:
an interpolation with inverse distance
weighting to calculate the background
pollution
the Lagrange spread model LASAT to
calculate the additional pollution from the
area
the Gauss model PROKAS to calculate the
net entry through transport in urban areas
the diagnostic spread model PROKAS B to
calculate induced transport additional
pollution in the streets
For Practical applications (e.g. clean air plans for
Saxony communes) the variation between measured
and calculated values are mostly under 20%.
Table 1, comparison / modelling
With this procedure it is possible to calculate
not only current but also future situations. With these
results it is possible to review the success of the
actions taken. Therefore it is used in five cities of
Saxony to create clean air plans.
Measured Calculated
PM
10
NO
2
PM
10
NO
2
Dresden -
Bergstraße
33
58
31 (31)
47 (49)
Dresden -
Mitte
31 31 30 31
Dresden -
Nord
33 47 35 49
51

image
image
UFIPOLNET LIFE04 ENV/DE/000054
52

image
image
UFIPOLNET LIFE04 ENV/DE/000054
POSTER
53

image
Particle number concentration in the urban area of Rome
P. Avino
1
, S. Casciardi
2
, C. Fanizza
1
, M. Manigrasso
1
1
DIPIA-ISPESL, via Urbana 187, 00184 Rome, Italy
2
DIL-ISPESL, via Fontana Candida 1, 00040 Monteporzio Catone (Rome), Italy
Keywords: Ultrafine particles, Polycyclic Aromatic Hydrocarbons, Nitrogen oxides
Atmospheric particulate matter (PM) pollution
is presently regulated on mass basis through PM
10
convention that measures the mass of particles
collected with a 50% efficiency for particles with an
aerodynamic diameter of 10 μm, in that way
including, coarse particles down to ultrafine particles
(UFPs), <0.1 μm. UFPs deposit very efficiently in
the lungs. Their small size, high number
concentration and surface area allow them increased
ability of absorption of organic molecules and
penetration into cellular targets in the lung and
systemic circulation (Li et al., 2003). Such
characteristics account for their importance in
explaining the health effects of PM.
This work describes the first results of a study started
in April 2007 to investigate the UFP pollution in the
urban area of Rome. Particle number concentration
and size distribution have been measured by means
of TSI 3936 Scanning Mobility Particle Sizer,
configured with 3080 Electrostatic Classifier, 3081
Differential Mobility Analyzer and 3786 water-based
ultrafine condensation particle counter. Number
concentration data have been compared with nitrogen
oxides (NOx), measured by chemiluminescent
analysis and particle-phase Polycyclic Aromatic
Hydrocarbons (PAHs), measured by Ecochem PAS
2000 PAH Monitor. Positively charged UFPs are
being sampled using the TSI 3089 electrostatic
nanoparticle precipitator, after passing the aerosol
through a TSI 3080 Electrostatic Classifier.
Transmission Electron Microscopy (TEM) is a
technique widely used to study the size, morphology,
composition, microstructure and crystallinity of
individual particles (Chen et al., 2005). Such
technique is being used in the ongoing UFP
characterization study, by means of an Energy-
Filtered Transmission Electron Microscopy
(EFTEM) FEI TECNAI 12.
Figures 1 and 2 show the daily trend of number
particle concentration in the range 20-880 nm
compared respectively with NOx and particle-phase
PAH. The pattern of variation of such pollutants are
very similar, suggesting a common autovehicular
origin. Number particle concentrations show a typical
daily modulation with minimum values measured
during nocturnal hours when the effect due to the
reduction of the autovehicular traffic emission
overcomes the decrease of the atmospheric mixing
height.
Hourly averages
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
4/6
4/7
4/8
4/9
4/10
4/11
4/12
4/13
4/14
4/15
Day
#/cm3 (20-880 nm
0
50
100
150
200
250
300
350
400
450
500
550
600
NOx (ug/m3
Particles 20-880 nm
NOx
Figure 1. Daily trends of particle number
concentration (20-880 nm) and NOx.
Hourly averages
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
4/6
4/7
4/8
4/9
4/10
4/11
4/12
4/13
4/14
4/15
Day
#/cm3 (20-880 nm
0
50
100
150
200
250
300
350
400
450
500
PAH (ng/m3
Particles 20-880 nm
PAH
Figure 2. Daily trends of particle number
concentration (20-880 nm) and particle-phase PAH.
The first results of TEM study indicates the presence
of single (figure 3), and aggregates carbonaceous
UFPs.
Figure 3. EFTEM micrograph of a carbonaceous
UFP.
Li N. et al., (2003).
Environmental Health
Perspectives,
11 (4), 455-460.
Chen N. et al., (2005).
Aerosol Science and
Technology,
39, 509-518.
54

Using a sampling and monitoring device, a solution for PM10-PM2,5 assessment ?
Experience with the Swam 5-A
L.Bertrand
1
, G.Gérard and S.Fays
ISSeP, rue du Chéra 200, 4000 Liège, Belgium
1
L.bertrand@issep.be
Keywords: PM10, PM2,5, Equivalence
Present European PM regulation (directive) is
from 1999, measurement standards from 1998
(PM10) and 2005 (PM2,5).
However, reconciliation, through
“Equivalence trials” automatic analysers actually
used by member states the gravimetric reference
method is still an on-going process in 2007.
Presently the Walloon network for PM relies
on the beta attenuation analyser MP101M.C. It
should be run with a 24h cycle to satisfy to
Equivalence criteria. But that time resolution is not
acceptable for the Competent Authority.
Accordingly, it is foreseen to supplement and
eventually replace that analyser by others, which
means more Equivalence trials ahead.
The correction factor (PM10) currently used
for the MP101M.C is 1,37.
Besides, from our experience, reference PM
samplers available up to now do not completely meet
requirements of the gravimetric standard. This they
bring an initial uncertainty, using part of the budget,
when candidate analysers are processed through
gravimetric equivalence trials.
ISSeP ran and tested two dual channel
sampler and beta attenuation analysers of FAI-
Instruments (Fonte-Nuova, Italy) between October
2006 and August 2007. Although more sophisticated
uses are put forward, ISSeP ran exclusively the
devices as a PM10 and PM2,5 sampler and analyser,
operated at the reference flow of 2,3 m³/h. Cycle was
always 24H (hence two filters sampled - one PM10
and one PM2,5) and two beta attenuation results (one
PM10 and one PM2,5) for each device and each day.
In April 2007, initial devices were exchanged
against a more transportable version. One of these
exchange Swams was with an embarked optical
analyser delivering results for two sizes or classes.
Remedies or improvements of the
manufacturer in reaction to ISSeP’s feedback were an
upgrade of the filter cartridge (to avoid filter
adhesions observed with some brands of quartz
filters) and a subsequent mechanical adjustment to
maintain perfect shape of dust spots (as mandatory
for the beta attenuation analysis).
Experienced achievements of the Swam as an
analyser include actual permanent measurement of
flow, temperature conditioning for filters to ensure
respect of the maximum allowable 5°C (ambient vs.
filters) temperature difference and recording of that
difference.
Table 1. Precision of Swam 5β analyser
Table 2. Equivalence :
β
analyser result vs. gravime-
tric result (PM10, quartz Macherey-Nagel
Table 3. MP101.C-pm10-(2h cycle)versus Swam
β
result considered as reference
This confirms MP101M.C is not appropriate
for compliance monitoring when run with a cycle
shorter than 24H. However, as long as it it still used,
the appropriate correction is presumably in the range
1,15-1,25 rather than the currently used 1,37.
Support for this work of DGRNE (Cellule
Air), Walloon Ministerial competent authority, is
gratefully acknowledged.
EN 12341 (1998) Air quality, determination of the
PM10 fraction of suspended particulate matter
EN 14907 (2005) Standard gravimetric measurement
method for the determination of the PM2,5 mass
fraction of suspended particulate matter.
Demonstration of equivalence of ambient air
monitoring methods (report of EC working group)
(http://ec.europa.eu/environment/air/cafe/index.htm)
Filter material
PM10
u.
bs
(μg/m³)
PM2.5
u.
bs
(μg/m³)
fiberglas whatman
0.70
0,67
quartz macherey-nagel
0.99
0.99
quartz what. winter
0.70
0.67
quartz what. summer
0.76
0.72
equation
Y=1.03x-0.66
u% 7,4
factor from ratio the means
1.00
factor from orthogonal
regression through origin o.R.o.
1.01
swam filter material
u%
derived corr. factor
from
ratio
From
o.R.o.
fiberglass whatman
1.15
quartz macherey
7,5
1,20
1,21
quartz what. winter
1,17
quartz what summer
25
1,25
1,26
55

image
image
image
Measurement and analysis in space and time of ultrafine particle number concentration
in ambient air. The case of Parma
F.Costabile
1
and I. Allegrini
1
1
Institute of Air Pollution, National Research Council, Monterotondo, 00016, Rome, Italy
Keywords: ultrafine particles, number concentration, urban ambient air.
Several studies have analysed the influence of
ultrafine particles (UFP) on human health in
atmosphere (e.g., Pope et al., 1995). Even though
preliminarily, such works suggest that motor vehicles
are a significant UFP source (Kittelson at al., 2006).
Additionally, this source could be even more
important than for PM
10
(Harrison et al., 1999).
In this work, the major objective was to
collect a data-set of UFP number concentration in a
middle-size urban area in Italy (Parma) with the aim
to evaluate connections with the local traffic sources.
Measurements were taken every 2 seconds at
two urban sites, 700 meters far each other: a traffic
site (less than 1 meter from the road) and a urban
background site (around 100 meters from the closest
road). UFPs were measured during winter and
summer time of 2007 (both week-days and week-
ends) by using two tandem Water CPCs (TSI, mod
3781).
UFPs concentrations were relatively low
(fig.1-4) when compared to similar works (e.g.,
Harrison et al., 1999).
Figure 1. UFP number concentration during the
winter campaign at an urban background station in
Parma.
As found in previous studies(e.g., Zhang et
al., 2004), UFP concentrations measured at the two
sites showed significantly different tendencies.
Differences related mainly to two factors (fig.1-4): at
the traffic site (i) higher concentrations and (ii) faster
and
more
intensive
dilution
processes
were
measured.
Figure 2. UFP number concentration during the
winter campaign at a traffic station in Parma.
Beyond a more rapid reduction of UFP number, the
faster dilution processes at the traffic site could also
trigger other physical processes (such as nucleation)
in connection to other parameters (such as solar
radiation).
Figure 3. UFP number concentration during the
summer campaign at a urban background station in
Parma.
56

image
Figure 4. UFP number concentration during the
summer campaign at a traffic station in Parma.
UFPs measured at the traffic site clearly
shows (fig.2 and 4) as expected (e.g., Shi et al., 1999)
the two peaks related to the traffic rush hours
indicating traffic emissions to be the prevailing
source.
The influence of the traffic flows can also be
recognised at the urban background site likely in
connection to transport phenomena (e.g., due to wind
speed). At this site, the influence of meteorology can
easier be analysed. This is particularly true for a clear
correlation with the total solar radiation ( Harrison et
al., 1999). Its influence can be seen at both the sites:
in effect, beyond the rush hours peaks, the traffic site
shows clearly a third peak at midday (fig.1-4).
The analysis of the all measurements and the
comparison between winter and summer time, week-
days and week-end showed, in summary, three major
contributions. Firstly, a very low urban background
UFP
concentration
(lower
than
1000
#/cm
3
)
particularly
visible
in
summer.
Secondly,
a
significant contribution due to local traffic sources,
up to 100000 #/cm
3
(traffic site in winter time).
Finally, a significant contribution due to secondary
processes closely linked to meteorology, particularly
solar radiation, and evident at midday.
This work was supported by the Italian Ministry of
Environment, Land and Sea in the framework of the
project “Air Pollution Emission Monitoring in the
city of Shanghai”.
Kittelson, D.B., Watts, W.F., Johnson, J.P., Schauer,
J.J., Lawson, D.R., 2006.
On road and
laboratory evaluation of combustion aerosol-
Part 2: summary of spark-ignition engine
results
. Aerosol science 37, 931-949.
Harrison, R.M., Jones, M., Collins, G., 1999.
Measurements of the physical properties of
particles in the urban atmosphere.
Atmospheric
Environment 33, 309-321.
Pope III, C. A., Thun, M. J., Namboodiri, M. M.,
Dockery, D. W., Evans, J. S., Speizer, F. E. and
Heath, Jr. C. W. (1995)
Particulate air pollution
as a predictor of mortality in a prospective study
of U.S. adults
. Am. J. Respir. Crit. Care Med.
151, 669–674
Shi, J.P., Khan, A.A., Harrison, R.M. (1999)
Measurements of ultrafine particle concentration
and size distribution in the urban atmosphere.
The Science of Total Environment. 235, 51-64.
Zhang, K. M., Wexler, A.S., Zhu, Y.F., Hinds, W.
C., Sioutas, C., 2004.
Evolution of particle
number distribution near roadways. Part II: the
‘Road-to-Ambient’
process.
Atmospheric
Environment 38, 6655–6665
57

image
Experiences with ultra fine particle monitoring in air quality monitoring networks in
Europe
C. Gerhart
1
, T. Petry
1
, T. Rettenmoser
1
, A. Kranapeter
2
, HP.Lötscher
3
1
Grimm Aerosol Technik GmbH & Co. KG, Ainring, 83404, Germany
2
Abteilung Imissionsschutz Stadt Salzburg, Salzburg, 5020, Austria
3
Amt für Natur und Umwelt, Chur, 7001, Switzerland
Keywords: Combustion particles, Health effects of aerosols, Outdoor aerosols, PAH(s), SMPS
In air monitoring networks particles in the
range of a few hundred nm are not measured. For
mass related measurements the ultra fine particles
are negligible. Nevertheless these small particles
(below 500nm) contribute with 80% to the particle
number concentration in ambient aerosols. Special
in urban regions the total particle concentration is
determined by very small particles like diesel soot,
generated by combustion processes.
In
epidemiological discussions on air born particles
the focus of interest has shifted from mass to
number concentration in the recent years.
Therefore
it a very interesting task to measure the ultra fine
particles additionally to the coarse fraction and
compare the data with the meteorological
parameters and the values of important volatiles.
particle diameter [μm]
Figure1. A typical volume and number distribution
in an urban aerosol (Seinfeld and Pandis, 1997)
To compare the measurements of ultra fine
particles with the data, obtained from the well
proofed instrumentation of the air quality
monitoring networks, it is essential to be sure that
the devices for measuring nano particles are
working stable and comparable. Such investigations
were carried out in the last years for the GRIMM
SMPS+C instruments with good results (for
example in December 2006 in Leipzig at the
Leibniz Institute for Tropospheric Research
.
There are also strong efforts at the moment
in Germany to work out the details of measuring
ultra fine particles for a VDI DIN directive. A first
draft defines already the use of a CPC
(Condensation Particle Counter) and a DMA
(Differential Mobility Analyser).
In the following, some examples of
measuring campaigns are mentioned in which
GRIMM instruments were used in air quality
monitoring networks to obtain additional
information about the aerosol particles. Where
“SMPS+C” means a combination of a CPC with a
DMA for the size range between 5 and 1100 nm
and “WRAS” (Wide Range Aerosol Spectrometr)
the combination of a SMPS+C with an OPC
(Optical Particle Counter) for particle diameters up
to 30 μm.
In April 2006 such measurements with a
WRAS were done in Graz (Austria). The measured
particle concentrations related strongly to the traffic
and folkloric events (Easter fires).
In May 2006 during eight days a measure-
ment in Salzburg (Austria) was carried out. Here
additionally to the WRAS a PAH sensor was
installed. The concentrations of the ultra fine
particles and the values of the PAH sensor
correlated strongly with the values for NOx and
CO
2
, which were measured by the network.
In winter 2007 in Graubünden (Switzerland)
a long-term measurement (over six weeks) was
done. Here the particle concentration, measured
with a SMPS+C, in rural and urban regions were
compared.
Special thanks to the involved networks.
Alexander, F. R., & Nathan J. O. (1986).
An
Introduction to Ultrasonic Nebulisation
.Cambridge,
U.K.: Cambridge University Press.
Chapman, D. H. (1975).
J. Aerosol Science
, 36,
3456-3467.
Finn, P., Diver, G. N., & Wake, K. T.
(1998). In
Proc. 13th Int. Conf. on Marine
Aerosols
, Reykjavik (Wiley, New York), 631-633.
58

UFIPOLNET: Concentration of Particle Number Distributions at 4 Stations in Europe
H. Gerwig
1
, G. Löschau
1
, L. Hillemann
2
, B. Wehner
3
, A. Wiedensohler
3
, A. Zschoppe
4
, C. Peters
4
, A. Rudolph
4
,
C. Johansson
5
, J. Cyrys
6
, M. Pitz
6
, R. Rückerl
6
, J. Novak
7
, H.G. Horn
8
, R. Caldow
9
, G.J. Sem
9
1
LfUG - Section Air Quality, Saxon State Agency for Environment and Geology, 01109 Dresden, Germany
2
UBG – Staatliche Umweltbetriebsgesellschaft, 01445 Radebeul, Germany
3
Leibniz-Institute for Tropospheric Research, 04318 Leipzig, Germany
4
Topas GmbH, 01279 Dresden, Germany
5
ITM – Department of Applied Environmental Science, Stockholm University, 106 91 Stockholm, Sweden
6
GSF National Research Centre for Environment and Health, 85764 Neuherberg, Germany
7
CHMI – Czech Hydrometeorological Institute, 14306 Prague, Czech Republic
8
TSI Gmbh, 52068 Aachen, Germany
9
TSI Inc., Shoreview, Minnesota, 55126, USA
Keywords: atmospheric aerosols, instrument development, number concentration, number size distribution
Several studies show a decline of particle
mass concentrations in Central Europe of TSP and
PM10 1990 - 99. In contrast, particle number
concentrations of ultrafine particles (< 100 nm =
UFP) were not changed during winter periods 1991 –
1999 in Erfurt/Germany (Cyrys et al. 2002). There
are however only a limited number of long-term UFP
measurements in Europe. Epidemiological studies
showed a relationship between high number
concentrations of UFP and adverse health effects.
The European Commission needs therefore
more information about UFP concentrations for
evaluation processes within the CAFE process and
the Thematic Strategy on Air Pollution.
The project UFIPOLNET (Ultrafine Particle
Size Distributions in Air Pollution Monitoring
Networks) intends to demonstrate that the newly
developed Ultrafine Particle Monitor UFP 330 is able
to perform adequately in routine network operation.
The instrument produces a number size
distribution (20 – 800 nm). Only 6 size classes >20, >
30, > 50, > 70, > 100, >200 (N1 – N6) are transferred
to the central measurement network stations to
reduce the amount of data collected in the databases.
First comparisons with a DMPS for ambient
aerosols (Wehner et al. this issue) show a good
correlation with a DMPS measuring in parallel at a
street canyon site.
Since December 2006 in Dresden and
February 2007 in Augsburg, Stockholm and Prague,
the UFP 330 will run continuously until October. It is
planned to run the instruments on a permanent basis
for a longer period. All sites are near busy roads;
Augsburg is an urban background site. The number
concentrations will be correlated with nitrogen
oxides, benzene and other continuously measured
parameters in a routine measuring network. In some
places, traffic numbers will be correlated with the
measurements.
At three stations, SMPS/DMPS size
spectrometers have been monitoring for several
years. Figure 1 compares the annual mean
concentrations of total number concentrations per
station (2003 – 2005). Augsburg shows about half,
Stockholm twice as many particles as Dresden.
Prague and Dresden show almost the same
concentration of NOx in 2005, while the street
canyon of Stockholm shows almost twice the
concentration. The correlation with NOx indicates
the traffic influence (Birmili, 2006).
0
3000
6000
9000
12000
15000
Augsburg
Dresden
Prague
Stockholm
Particle number [cm
-
³]
0
30
60
90
120
150
NOx [μg
.
m
-
³]
UFP 20 - 800 nm
NOx
Figure 1: Number concentrations at the 3 stations
with reference instruments
One aim of UFIPOLNET is to harmonise the
sampling conditions (particle pre-impaction and
humidity) as well as the evaluation of identical size
classes. In this way, interpretations of particle
number concentrations and size distributions will be
facilitated. Comparable results will permit analysis of
absolute differences between ultrafine aerosol size
distributions at many polluted sites over long periods.
UFIPOLNET
(www.ufipolnet.eu)
is financed by the LIFE
financial instrument of the European Community under No.
LIFE04 ENV/D/000054. The authors wish to thank Heinz Ott for
providing NOx-data in Augsburg from the routine air measuring
network of the Bavarian State Agency of Environment (LfU).
J. Cyrys, J. Heinrich, A. Peters, W. Kreyling, and
H.E. Wichmann (2002).
Umweltmed Forsch
Prax.
, 7, 67-77.
Birmili, W. (2006). Editor: D. Bake;
Forschungsbericht
203 43 257/05 UBA-FB
000942 ;
UBA Texte
26 – 06, Umweltbundesamt,
Berlin.
59

Determination of particle emission factors of individual vehicles
under real-life conditions
C.S. Hak, E. Ljungström, M. Hallquist, M. Svane and J.B.C. Pettersson
Department of Chemistry, Atmospheric Science, Göteborg University, 412 96 Göteborg, Sweden
Keywords: ultrafine particles, traffic, urban pollution, vehicle emissions.
Road traffic constitutes an important source of
particulate matter and trace gases. Most particles in
vehicle exhaust are in the ultrafine size range. In
contrast to large particles which are a result of wear
of road pavement, tyres and brakes, ultrafine particles
are respirable and penetrate deep into the lungs,
posing a threat to health. Especially in the densely
populated urban areas, road traffic can lead to severe
pollution of the ambient air. Since today half of the
global population lives in urban areas, it is a matter
of public and scientific concern to examine the
emissions under real-life conditions. Particularly the
measurement of the particle emissions from
individual vehicles that form a car fleet delivers
insight into the variability of particle emissions
among cars and might allow the assignment of
emission properties according to vehicle type, age,
fuel technology etc.
Here we present an experimental setup for the
measurement of particle emission factors (EF) from
individual vehicles, which was designed for
continuous on-road sampling. The system has been
applied in two test experiments conducted in the
Göteborg area. The measurement sites were at a
rather busy major street in Göteborg (Western
Sweden) and a sparsely frequented two-lane country
road 25 km from Göteborg, respectively. The
measurements were each performed monitoring the
traffic on one of the two lanes. The inlet of a
sampling line was installed in the centre of the lane
and attached to the street surface to extract air
directly (
in-situ
) from the plumes of passing vehicles.
To measure CO
2
and number density of particles
> 10 nm simultaneously, the collected air was
distributed to a CO
2
monitor and a condensation
particle counter (CPC, TSI 3010), respectively. The
additional registration of licence plate numbers from
the vehicles driving past gives information about the
vehicle type, engine power and further technical data
provided by the National Road Administration.
While during the first experiment everyday
traffic was observed to prove the applicability of the
described approach, the second experiment served to
study the intra-vehicle variability of particle
emissions under predetermined driving conditions
(specified speed and gear). A set of four selected
vehicles, consisting of a diesel car (car I), an older
medium-sized petrol car (car II) and two compact
cars (car III, car IV) was applied.
Figure 1. Measurement setup for simultaneous
sampling of CO
2
and particle concentration.
Although the sampled air volume was diluted
with a known amount of particle-free background air,
the very high particle numbers within the vehicle
exhaust plumes gave concentrations exceeding the
upper limit of the particle counter. A solution to
nevertheless quantify the excess particle numbers in
the plume is presented and the particle EF’s are
derived from the enhancement ratios of particle
number to CO
2
mixing ratio in the exhaust plumes of
passing vehicles. When assuming a CO
2
emission of
164 g km
-1
veh
-1
, which is an average value for
petrol-fuelled vehicles (cp.
www.starterre.fr/voiture-
auto/emission_co2), particle emission factors of the
order 2 x 10
13
part km
-1
veh
-1
were derived from the
observed car passages (cp. Figure 2). This value
agrees well with those published in the literature for
petrol-driven vehicles (e.g. Jones & Harrison, 2006;
see also lower side bar in Figure 2), which constitute
more than 90% of the Swedish fleet (Ahlvik, 2002).
Emission factors for diesel driven passenger cars are
one order of magnitude higher (cp. upper side bar in
Figure 2). Given the obtained results, the used setup
proved appropriate to quantify particle emission
factors from road traffic.
The measurements and observations of this
study demonstrated that the simultaneous
measurement of CO
2
and particle number for the
purpose of characterising individual vehicles’
sampling tube
inlet
valve
(11 l min
-1
)
particle filter
compressor (dilution)
sampling
pump
valve
(30 l min
-1
)
(optional)
CO
2
-scrubber
CO
2
CPC
mixing
volume
sampling tube
inlet
valve
(11 l min
-1
)
particle filter
compressor (dilution)
sampling
pump
valve
(30 l min
-1
)
(optional)
CO
2
-scrubber
CO
2
CPC
mixing
volume
60

emissions is a feasible approach. Furthermore, by
using the described method the particle emission can
be linked to other indicators for traffic exhaust such
as NO
x
, VOC or CO in order to better understand
real-world vehicle to vehicle variation in particle
production.
0.0
2.0x10
13
4.0x10
13
6.0x10
13
8.0x10
13
1.0x10
14
1.2x10
14
Particle EFs [km
-1
veh
-1
]
mixed fleet
average EF
Figure 2. Single vehicle number particle emission
factors derived from measurements in Göteborg.
This work was supported by the Swedish Foundation
for Strategic Environmental Research MISTRA and
the National Swedish Road Administration. Benny
Lönn, Senior Research Engineer, is acknowledged
for skilful technical support.
Ahlvik, P. (2002).
Environmental and Health Impact
from Modern Cars
, Report for the Swedish
National Road Administration, Publication
2002:62.
Jones, A. M., & Harrison, R. M. (2006).
Atmos.
Environ.
, 40, 7125-7137.
61

Particle number size distributions of ambient-state and non-volatile aerosols
in the city of Augsburg, Germany
K. Heinke
1
, W. Birmili
1
, A. Wiedensohler
1
, M. Pitz
2,3
, J. Cyrys
2,3
, and A. Peters
2
1
Leibniz Institute for Tropospheric Research, 04318 Leipzig, Germany
2
GSF National Research Center for Environment and Health,
Institute of Epidemiology, 85758 Neuherberg/Munich, Germany
3
WZU - Environmental Science Center of the University Augsburg, 86159 Augsburg, Germany
Keywords: aerosol size distribution, urban aerosols, soot particles
Fine and ultrafine (< 100 nm) aerosol particles
in the environment have moved into the interest of
public health research due to their presumed adverse
effects upon human health, such as cardiovascular
and respiratory disease. While the adverse effects of
ambient particles are in general widely ack-
nowledged, there has been only little epidemiological
evidence on the role of particular sub-fractions of the
aerosol.
The city of Augsburg in Southern Germany
hosts a centre of environmental medical research for
the quantification of air pollutants and their effects
on sensitive parts of the population (KORA; Holle et
al. 2005). Specialized aerosol particle measurements
were started in November 2004 in order to support
future epidemiological studies within KORA.
As ambient particles are a complex mixture of
a myriad of chemical compounds, there is a growing
need to characterise and isolate those particular sub-
fractions that are relevant to human health. Volatility
analysis (with thermodenuder) is a method that
makes use of different volatilisation temperatures of
chemical compounds, thereby separating volatile
compounds, such as organic matter, sulfates and
nitrates, from non-volatile compounds, such as soot
and mineral dust. A volatilisation temperature of
300°C allows to identify soot particles in the ultrafine
size range emitted from vehicular traffic (Wehner at
al., 2004; Rose et al. 2006).
At the GSF research station Augsburg
ambient state and non-volatile particle size distrib-
utions (3 - 800 nm) have been measured continuously
since 11/2004 using a twin differential mobility
particle sizer (TDMPS).
Figure 1 presents median particle number size
distribution for ambient and non-volatile compounds
during the rush hour traffic; while the ambient
particle number distribution peaks in the Aitken
mode (~40 nm), the curve of non-volatile residues
peaks in the nucleation mode (~10 nm) and in less-
volatile particle mode (~80 nm) which presents the
externally mixed population of soot particles. These
measurements imply that within the measurement
accuracy, every ambient particle contains a non-
volatile core. The chemical composition of the non-
volatile residues < 20 nm is, however, not known yet.
We will present a statistical summary of the 2-
year data set, including an analysis of the relationship
between total and non-volatile particle size
distributions, and the meteorological factors that
cause high concentrations of total and non-volatile
particle fractions, such as wind direction, mixed layer
height, and remote transport.
10
100
1000
10
1
10
2
10
3
10
4
dN/dlog Dp, cm-³
Dp, nm
Figure 1: Median particle number size distribution of
total (grey) and non-volatile (black) aerosol between
7:00 – 9:00 am.
Holle, R., Happich, M., Löwel, H., Wichmann, H. E.
(2005): KORA – A Research platform for
population based health research,
Gesund-
heitswesen 2005
, 67 Sonderheft 1, S19-S25.
Wehner, B., Philippin, S., Wiedensohler, A., Scheer,
V., Vogt, R. (2004): Variability of non-
volatile fractions of atmospheric aerosol
particles with traffic influence,
Atmos. Env
.,
38, 6081-6090.
Rose, D., Wehner, B., Ketzel, M., Engler, C.,
Voigtländer, J., Tuch, T., Wiedensohler, A.
(2006): Atmospheric number size
distributions of soot particles and estimation
of emission factors,
Atmos. Chem. Phys.
, 6,
1021-1031.
62

image
Determination of the charge distribution of highly charged aerosols
Lars Hillemann
1
, Michael Stintz
1
, Christoph Helsper
2
1
Arbeitsgruppe Mechanische Verfahrenstechnik, Institut für Verfahrenstechnik und Umwelttechnik, TU Dresden,
01062 Dresden, Germany
2
Fachbereich Elektrotechnik und Automation, Fachhochschule Aachen, 52428 Jülich, Germany
Keywords: electrical analyzers, electrical charging, charge distribution
INTRODUCTION
Bipolar chargers are widely used in aerosol science
because their charge distribution is well understood
(Wiedensohler, 1988). However, for the determination
of aerosol size distributions unipolar charging may
have some advantages. Using a unipolar charger in this
application requires that the charge distribution is well
known. Biskos et. al. (2005) examined charge
distributions for soot particles with a Hewitt-type
charger.
The work presented here explores the possibility of
using the unipolar charger of the Electrical Aerosol
Detector (Medved, 2000) for the quantification of
ambient aerosols. This requires the measurement of the
charge distribution applied by the charger on several
particle systems varying in particle material and shape.
EXPERIMENTAL SETUP
The charge distribution for several particle systems
(latex, silica, soot) was experimentally determined in a
Tandem DMA setup (Figure 1). It consists of three
parts. In the left part a monodisperse fraction of the test
aerosol is generated. This aerosol is subsequently
diluted and positively charged in the middle part.
Passing the DMA 2 the particles are classified
according to their electrical mobility which depends on
the number of charges carried by the particles. The
combination of an electrometer and CPC 1 in the right
part of the setup returns the mean charge of the
particles. This allows the verification of the charge
distribution.
Figure 1: Experimental setup for measuring the charge
distribution
DATA CONVERSION
The high number of charges impedes the proper
separation of the charge modes by the DMA. Therefore
it is necessary to model the analyzing DMA using its
transfer function. Fitting the output of the second CPC
to the measured data determines the charge distribution
of the aerosol.
RESULTS
For small particles carrying a low number of
charges a direct fit of the charge numbers is possible.
When dealing with larger particles the dimension of the
search space increases with the number of charges.
Here, the charge distribution is described by a modified
normal distribution.
0
2
4
6
8
10
12
0
5
10
15
20
25
30
frequency [%]
Number of charges [-]
355 nm
Figure 2: charge distribution, produced by the charger at
5 l/min (aerosol: ammonium chloride)
UFIPOLNET
(www.ufipolnet.eu)
is financed by the
LIFE financial instrument of the European Community
under No. LIFE04 ENV/D/000054.
Wiedensohler A. (1988). An approximation of the
bipolar charge-distribution for particles in the sub-
micron size range,
J. Aerosol Science
, 19, 387-389.
Medved A., Dorman F., Kaufman, S. L. (2000). A new
corona-based charger for aerosol particles,
J.
Aerosol Science
, 31 SUPP/1, 616-617.
Biskos G., Reavell K., Collings N. (2005). Unipolar
diffusion charging of aerosol particles in the
transition regime,
J. Aerosol Science
, 36, 247-265.
63

image
image
image
Quantification of nanoparticle releases from surfaces
Lars Hillemann
1
, Michael Stintz
1
, Mario Heinemann
2
1
Institute of Process Engineering and Environmental Technology, TU Dresden, D-01062 Dresden, Germany
2
Wacker-Chemie AG, Werk Burghausen, Johannes-Hess-Str. 24, D-84480 Burghausen, Germany
Keywords: ultrafine particles, nanoparticles, release, modelling
Modelling fine and ultrafine particles in urban
air requires quantification of particle releases from
environmental or industrial surfaces under relevant
conditions.
In many fields of application ultrafine
particles or nanoparticles are employed to improve
the properties of surfaces. Easy-to-clean coatings,
corrosion protection and fiber reinforcing are some
examples. Unfortunately, particles in this size range
may be harmful to health if inhaled and deposited in
the respiratory tract.
For the resuspension of particles in gas flows,
the ratio of the drag force to adhesion force is the
determining criterion. Larger particles easily detach
from surfaces, but with decreasing particle size the
decrease in drag force is larger than the decrease in
adhesion force. Furthermore even when using a
turbulent airflow, there is a laminar sublayer at the
surface, reducing the effective drag force on the
particles. Consequently, particles smaller than 10 μm
are usually not removed from a surface by air
currents. For new nanoparticle-doted products this
has to be proven by the manufacturer.
In the presented project, a test device that
quantifies nanoparticle releases from surfaces has
been developed. It focuses on the particle
reentrainment by drag force into an air flow.
Furthermore it can be adapted to assess textile
samples like gas filters or clothes with regard to
particles released from the filter material.
Figure 1. Scheme of the test device, a sample can be
moved in two directions under the nozzle
The device consists of a nozzle and a sample
carrier which can be moved in two directions.
Through the nozzle, a controlled side channel blower
draws a flow rate of up to 20 l/min. The nozzle has a
diameter of 5 mm. This narrow bore hole is
necessary to attain large shear stresses at low flow
rates. Therefore, it becomes possible to avoid
unacceptably high dilution ratios. The examined