University of Zilina
Faculty of Civil Engineering
Department of Highway Engineering
Road transport as the
primary source of
particulate matter in the
ambient air in urban and
non-urban areas of
northern Slovakia
ECTRI – FEHRL – FERSI
Young Researchers Seminar 2015
Dusan Jandacka
1
2
 Particulate
Content of presentation
matter
 Monitoring and chemical analysis of particulate
matter
 Identification process of particulate matter sources


Urban area
Non-urban area
 Conclusions
3
Particulate matter


Course fraction – aerodynamic diameter 2,5 - 10 μm (PM2,5-10) –
mechanical abrasion and removal of road dust.
Fine fraction – aerodynamic diameter < 2,5 μm (PM2,5) – chemical
reactions, condensation of exhaust emissions, coagulation of ultrafine
particles.
Exhaust particle
zoom 30 tis.
Particle from mechanical abrasion
zoom 70 tis.
Vehicular
traffic
Tyres
abrasion
Exhaust
emission
Non-exhaust
emission
Brakes
abrasion
Road surface
abrasion
4
Car-body
components
corrosion
Resuspension
of road dust
Road traffic as the source of
particulate matter
5
o
o
o
o
o
in the urban area in the vicinity of an
urban road in the City of Žilina,
during the years 2010 – 2012,
8 measuring cycles were realized,
total of 56 measurements were
performed,
Urban area is typically city canyon.
o
o
o
o
o
Monitoring of PM
in the non-urban area in the vicinity of a
highway D1
during the years 2013 – 2014,
3 measuring cycles were realized,
total of 36 measurements were performed,
surroundings of Non-urban area - open
area with agro land and water areas.
Concentration of PM
and traffic volume
6
Urban area
Non-urban area
PM1-2.5
23%
PM1
68%
Traffic volume [veh/24h]
PM2.5-10
9%
18,522
20000
15000
PM2.5-10
23%
11,845
10000
5000
PM1-2.5
15%
0
Urban area
Non-urban area
Measuring station
PM1
62%
7
Chemical analysis of PM
In order to determine the amount of chemical elements in the sample
of particulate matter (PM10) the spectroscopic methods (inductively
coupled plasma mass spectrometer ICP MS) were utilized.
Pb
Ba
Cd
Sb
Mo
As
Zn
Cu
Ni
Mn
Fe
Cr
V
Ca
Al
Mg
Na
The average concentration
of chemical elements in the
non-urban area
328
393
9
91
22
Chemical element
Chemical element
The average concentration
of chemical elements in the
urban area
52
1,847
709
20
340
16,127
51
27
28,930
6,557
8,774
6,249
1
10
100
1,000
10,000
100,000
Concentration of chemical element [μg/g]
Pb
Ba
Cd
Sb
Mo
As
Zn
Cu
Ni
Mn
Fe
Cr
V
Ca
Al
Mg
Na
381
177
12
83
13
67
1,551
359
55
235
9,264
65
15
6,861
3,587
2,158
22,131
1
10
100
1,000
10,000
100,000
Concentration of chemical element [μg/g]
8


Chemical analysis of PM
Each of these metals may come from a specific source
(Table).
Based on a sufficiently comprehensive database of data it
deems possible by the utilization of multi-layer statistical
methods (for instance factorial analysis), to more closely
specify the possible source of this particulate matter.
Source
Transportation
Associated elements
road surface
Al, Si, Ca, Mg, C, Na, K, V, Ni
car-body components
Cu, Sn, Cr, Pb, Cd, As, Sb, Fe, Al
brake callipers, pads and
rotors
Cu, Sb, Ba, Cr, Fe, Ni, Pb, Zn
tyres
Zn, Cd, Pb, Cu, Ni, Fe, Mn, Cr, Co
fuel and lubricating oil
diesel
Al, Ca, Mg, Mn, Cu, Fe, Mo, V, Zn
gasoline
Sr, Cu, Mn
oil
Fe, Ca, P, Zn, Mg
catalytic converter
Pt, Pa, Rh (Platinum metals)
road dust
Zn, Al, K, Fe, Na, Mn
Burning coal and wastes
Zn, Sb, Cu, Cd, Hg, Se, As, Cr,
Co, Al
Industry
Sb, Ag, V, Ni, As, In, Cu, Mn, Ce,
Co, Cr, Pb
Biomass burning
K
Incinerators
Cd, Pb, Sb, Zn
9
Used statistical methods – Principal
Component Analysis - PCA
𝑚
𝑦𝑘 =
𝑣𝑘𝑗 𝑥𝑗
𝑗=1
where: yk
xj
vkj
principal components, k = 1, ..., p,
former character, input variable, j = 1, ..., m,
coefficients of own vectors.
Matrix notation:
𝑦1
𝑦2
⋮
𝑦𝑝


𝑝×1
𝑣11
𝑣21
= ⋮
𝑣𝑝1
⋯ ⋯ 𝑣1𝑚
⋯ ⋯ 𝑣2𝑚
⋮
⋮
⋮
⋯ ⋯ 𝑣𝑝𝑚
𝑥1
𝑥2
× ⋮
𝑥𝑚
𝑝×𝑚
𝑚×1
The primary goal of PCA is the transformation of the original
characters of xj, j=1, ..., m, into a smaller amount of latent
variables of yk.
The first principal component y1 describes the greatest part of
variability.
Used statistical methods – Factor
10
Analysis - FA
𝑝
𝑋𝑗 =
𝜆𝑗𝑘 𝐹𝑘 + 𝐸𝑗
𝑘=1
where: λjk
factor load of the j-th object to the k-th common factor
Fk, j = 1, ..., m, k = 1, ..., p,
the k-th common factor,
is a random deviation, j=1, ..., m.
Fk
Ej
Matrix notation:
𝑥1
𝑥2
⋮
𝑥𝑚

𝑚×1
𝜆11
𝜆
= 21
⋮
𝜆𝑚1
𝜆12
𝜆22
⋮
⋯
⋯ 𝜆1𝑝
⋯ 𝜆2𝑝
⋮
⋮
⋯ 𝜆𝑚𝑝
𝑚×𝑝
𝐹1
𝐹
× ⋮2
𝐹𝑝
𝑝×1
𝑒1
𝑒2
+ ⋮
𝑒𝑚
𝑚×1
The fundamental principle of factor analysis lies in the fact that
each and every of monitored values Xj (j = 1, …, m) may be
expressed as a sum of a linear combination of a lesser amount p
non-observed (hypothetical) random values F1, ..., Fp – so called
common factors and the further source of variability Ej (j = 1, …,
m) – so called specific (residual) elements.
11

Identification proces of PM
sources – Urban area
Using a data matrix was compiled from the concentrations of
selected metals in ng/m3 (Na, Mg, Al, Ca, Cu, Sb, Ba, Pb, Cd,
Cr, As, Mo, V, Mn, Fe, Ni, Zn) and PM10 in µg/m3 resulting from 8
measurement cycles between the years of 2010 and 2012. The
data matrix contained 18 variables and 56 objects.
PCA
Pursuant to the rate of
eigenvalues (1 – 12.42, 2 –
1.94, 3 – 1.07) there were 3
main components selected
(selection
criteria
of
eigenvalue > 1.0). The three
main components define
85.70
%
of
the
total
dissipation, spread of the
former characters.
Identification proces of PM
sources – Urban area
12
FA

3 factors were selected for the factor analysis.
the factor loads are quoted in relation to particular characters and
particular factors. They may be explained as the correlation between the
factors and characters. They represent the most important unit of
information the interpretation of factors is based on.
F1: Non-exhaust traffic source (tyres and road dust) and local combustion
Factor load

1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.9
0.9
0.7
0.7
0.6
0.5
0.4
0.1
Na
Mg
0.3
Ca
V
0.9
0.7
0.5
0.3
0.9
0.3
0.4
0.4
0.1
Al
Cr
Fe
Mn
Ni
Cu
Zn
Chemical element (PM10)
As
Mo
Sb
Cd
Ba
Pb PM10
Identification proces of PM
sources – Urban area
13
Factor load
F2: Exhaust traffic source - diesel fuel, lubricating oil
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
1.0
0.9
0.6
0.6
0.6
0.3
0.2
0.2
Na
Mg
Al
Ca
V
Cr
0.3
0.2
0.1
0.2
0.3
0.3
0.3
0.2
0.2
0.1
Fe
Mn
Ni
Cu
Zn
As
Mo
Sb
Cd
Ba
Pb PM10
Chemical element (PM10)
Factor load
F3: Non-exhaust traffic source - brakes, car-body, road surface
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.9
0.7
0.8
0.6
0.5
0.5
0.6
0.7
0.7
0.6
0.5
Na
0.3
0.3
0.3
0.1
0.7
0.3
0.1
Mg
Al
Ca
V
Cr
Fe
Mn
Ni
Cu
Zn
Chemical element (PM10)
As
Mo
Sb
Cd
Ba
Pb PM10
14

Identification proces of PM
sources – Non-Urban area
Using a data matrix was compiled from the
concentrations of selected metals in ng/m3 (Na, Mg, Al,
Ca, Cu, Sb, Ba, Pb, Cd, As, Mo, V, Mn, Fe, Zn) and PM10
in µg/m3 resulting from 3 measurement cycles between
the years of 2013 and 2014. The data matrix contained
16 variables and 36 objects
PCA
Pursuant to the rate of
eigenvalues (1 – 8.74, 2 –
2.01, 3 – 1.84) there were 3
main components selected
(selection criteria eigenvalue
> 1.0). The three main
components define 78.71 %
of the total dissipation,
spread
of
the
former
characters.
Identification proces of PM
sources – Non-Urban area
15
FA

3 factors were selected for the factor analysis.
Individual elements are broken down into three factors after performing
the FA. These factors were named on the basis of the elements involved.
F1: Local combustion and non-exhaust traffic source (tyres)
Factor load

1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
0.9
0.7
0.6
0.6
0.6
0.7
0.8
0.8
0.7
0.7
0.5
0.4
0.3
0.1
-0.1
Na
-0.1
Mg
Al
Ca
V
Fe
Mn
Cu
Zn
As
Chemical element (PM10)
Mo
Sb
Cd
Ba
Pb
PM10
Identification proces of PM
sources – Non-Urban area
16
Factor load
F2: Exhaust traffic source - diesel fuel, lubricating oil and earth crust
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
0.9
0.8
0.7
0.9
0.7
0.7
0.7
0.4
0.2
0.3
0.4
0.4
0.2
0.1
-0.1
-0.2
Na
Mg
Al
Ca
V
Fe
Mn
Cu
Zn
As
Mo
Sb
Cd
Ba
Pb
PM10
Chemical element (PM10)
Factor load
F3: Road dust - winter salting
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
0.9
0.6
0.5
0.3
0.1
0.1
0.1
0.0
-0.1
-0.1
Na
Mg
Al
Ca
V
0.1
-0.3
Fe
0.0
-0.1
-0.2
-0.3
Mn
Cu
Zn
As
Chemical element (PM10)
Mo
Sb
Cd
Ba
Pb
PM10
17
Conclusions
 The
average concentrations of PM were twice
higher in Urban area like in Non-urban area,
 Traffic
volume – Urban area 11,845 [veh/24h], Nonurban area 18,522 [veh/24h],
 The
highest concentration of chemical element: Ca
– Urban area, Na – Non-urban area,
 The
lowest concentration of chemical element: Cd
– both areas,
18
Conclusions

Urban area: Factor 1 - Non-exhaust traffic source (tires
and road dust) and local combustion, Factor 2 - Exhaust
traffic source - diesel fuel, lubricating oil, Factor 3 - Nonexhaust traffic source - brakes, car-body, road surface.

Non-urban area: Factor 1 – Local combustion and nonexhaust traffic source (tyres), Factor 2 – Exhaust traffic
source – diesel fuel, lubricating oil and earth crust,
Factor 3 – Road dust – winter salting.

Usability of method - identification of sources of
particulate matter in specific risk areas,

Following this analysis - determining the contribution of
each named sources to the formation of PM.
19
...thanks for your attention
[email protected]
This contribution is the result of the project implementation: "Promotion
& Enhancement of Transportation Research Centre" (ITMS:
26220220160) supported by the Research & Development Operational
Programme funded by the ERDF.