ORIGINAL ARTICLE

Geochemical, mineralogical and magnetic characteristicsof vertical dust deposition in urban environment

Peter Sipos • Em}o Marton • Zoltan May •

Tibor Nemeth • Viktoria Kovacs Kis

Received: 5 March 2013 / Accepted: 10 December 2013

� Springer-Verlag Berlin Heidelberg 2013

Abstract Studies on composition and distribution of dust

deposition are necessary for the risk assessment of dust to

atmospheric quality. We studied the vertical distribution

pattern of dust and metal (Cu, Fe, Pb, Zn) deposition up to

33 m height in urban environment. Integrated geochemical,

mineralogical and magnetic study of the seasonally sam-

pled dust helped to specify our knowledge on the use of

magnetic susceptibility for tracking its deposition. Harmful

dust and metal deposition may occur even at great heights

and at the low-traffic side of buildings. Re-suspension of

local surface materials dominates the dust deposition pri-

marily in summer and spring due to weather conditions,

and it may overwrite the influence of recent anthropogenic

activities on dust composition. The accepted air-flow

models should be modified by taking the local conditions

(weather, morphology, etc.) into account. All studied

metals showed strong enrichment in the dust and could be

characterized by similar vertical deposition pattern to dust.

The total susceptibility was found to be much more useful

proxy for tracking dust and metal deposition than mass-

specific susceptibility. Using the former, potential errors

arising from sampling practice of settled dust could be

eliminated. The most important heavy-metal-bearing pha-

ses were iron oxides and clay minerals. Their different

behavior during the dust deposition is reflected by the

vertical metal distribution patterns. Clay minerals originate

primarily from re-suspension and may be one of the most

important sources of potentially mobile heavy metals in

such materials.

Keywords Total susceptibility � Heavy metals � Vertical

metal deposition � Dust re-suspension � Urban dust

Introduction

Airborne particulate matter is widely associated with health

disorders as presented by numerous studies. In the past, their

estimates have tended to be of total suspended particles,

while recent attention has focused on the study of their fine

fractions (\10 lm or even less), which may penetrate to the

innermost regions of the lung easily (Samet et al. 2000).

However, particles with a diameter of up to 100 lm can be

inhaled or ingested, and those below 32 lm may reach the

bronchial tubes causing health damage by diseases or due to

their toxic components (UNEP and WHO 1992).

Airborne particulate matter can be divided into two

groups: the settled dust sediment and the suspended parti-

cles (Remeteiova et al. 2007). The settled dust sediment is

created by particles with great sedimentation power, and

their delay time in the atmosphere is very short. Pollutants

that are carried by settled dust cause generally near-source

pollution. Thus, dust deposition is of important significance

P. Sipos (&) � T. Nemeth

Research Centre for Astronomy and Earth Sciences, Institute

for Geological and Geochemical Research, Hungarian Academy

of Sciences, Budaorsi ut 45, Budapest 1112, Hungary

e-mail: sipos@geochem.hu

E. Marton

Paleomagnetic Laboratory, Geological and Geophysical Institute

of Hungary, Columbus utca 17-23, Budapest 1145, Hungary

Z. May

Research Centre of Natural Sciences, Institute of Materials

and Environmental Chemistry, Hungarian Academy of Sciences,

Pusztaszeri ut 59-67, Budapest 1025, Hungary

V. Kovacs Kis

Institute of Technical Physics and Materials Science, Research

Centre of Natural Sciences, Hungarian Academy of Sciences,

Konkoly-Thege Miklos ut 29-33, Budapest 1121, Hungary

123

Environ Earth Sci

DOI 10.1007/s12665-013-3013-8

as an indicator of local environment quality (Vento and

Dachs 2007).

Magnetic particles have long been recognized to be

associated with atmospheric particulates (Hunt et al. 1984).

The magnetic minerals in the atmosphere are mainly

derived from combustion processes, such as industrial,

domestic and vehicle emissions, or from abrasion products

from asphalt and from vehicle brake systems (Gautam et al.

2005). It is therefore possible to use the magnetic proper-

ties to identify and trace airborne particulate pollutants in

the environment (Muxwothy et al. 2002).

Among the potentially toxic components of urban dust,

heavy metals are the most abundant ones. Generally, lead,

zinc and copper show the highest enrichment in such

materials when compared to their abundances in natural

geological formations (e.g., Manasreh 2010). These metals

can be also found in the magnetic particles. The most

important sources of these metals are the traffic in urban

environment (bearing and brushing, moving engine parts,

brake-lining, tire, lubricating oil and grease, motor oil,

under coating), but they are the common component of

construction materials of the built environment, too

(Sutherland 2000). The dust particles, after their sedimen-

tation, can also contaminate soils, groundwater (and even

the food chain) with their toxic components.

In urban environment, and especially in those areas

where population and traffic density are relatively high,

human exposure to hazardous substances is expected to be

significantly increased. This is often the case near busy

traffic axis in city centers, where urban topography and

microclimate may contribute to the creation of poor air

dispersion conditions giving rise to contamination hot spots

(Vardoulakis et al. 2003). In such areas, airborne particu-

late matter is expected to show unique sedimentation

characteristics.

Studies on sources, compositions and distribution of

dust deposition are necessary for the risk assessment of

dust to atmospheric quality, ecology and human health.

This is especially true for the urban environment, where the

horizontal distribution characteristics of airborne particu-

late matter (and its toxic components) have been widely

studied in the last few decades (e.g., Laidlaw and Filippelli

2008). It has been also shown that people breathing at

different heights are subjected to different concentrations

of airborne particulate matter from the pavement up to

3 meters height (Micallef et al. 1998). However, there are

no data about the deposition characteristics of dust and its

components at greater heights. The aim of this study was to

investigate the deposition characteristics of urban dust and

heavy metals (Cu, Pb, Zn) along a high building up to 33 m

height and on its both sides. Integrated geochemical,

mineralogical and magnetic study of the seasonally sam-

pled dust helped to specify our knowledge on the use of

magnetic susceptibility for tracking dust and metal depo-

sition in urban environment.

Materials and Methods

Settled dust samples were collected according to the

Hungarian standard (MSZ 21454/1-83 1983) using glass

pots of 2,000 mL containing 500 mL distilled water and

0.500 ± 0.001 g of algaecide (analytical grade methyl

4-hydroxybenzoate) with continuous supply of the water.

Altogether 8 sampling pots were placed on the front and

the back sides, respectively, of a building at a busy road at

heights of 2, 9, 21 and 33 m (Fig. 1). The number of

bypassing vehicles and trains are over 50,000 and 250 per

day there, respectively. The prevailing wind direction is

toward the road (NW), which is strengthened by the

dolomite hill behind and the plain in the front of the

building. The building towers are above its surroundings

by 15–20 m, so it may influence the wind flow structures

around itself significantly. The continuous seasonal sam-

pling started on the December 1, 2008, and finished on the

November 30, 2010 (64 samples altogether). The sampling

pots were drained at the first day of each season. The dust

and water were separated by vacuum filtering using a

Millipore filter with pores of 2 lm. Our magnetic mea-

surements (data are not shown here) of the dust samples

and the filtered liquid showed that there are no particles

present in the water phase which could give magnetic sign

(even superparamagnetic ones either). Additionally, there

was no detectable particulate matter found in the liquid

phase after centrifuging the liquid at 6,000 rpm for

15 min.

The weight measurements of the air dried dust samples

were followed by magnetic susceptibility measurements

using a KLY-2 Kappabridge instrument operating with one

frequency.

After the non-destructive magnetic experiments, the

dust was separated in ultrasonic bath from the filters.

Concentrations of the most significant urban metal pollu-

tants (Cu, Pb and Zn) and Fe in the samples were analyzed

by a Thermo Niton XL3 type X-ray fluorescence spec-

trometer in alumina sample holders. Relative standard

deviations of the parallel analyses are as follows: 7 % for

Cu, 4 % for Fe, 7 % for Pb and 5 % for Zn. Due to the

separation of liquid phase, water-dissolved metal amounts

are not presented here. Otherwise, water-soluble metal

fraction is not expected to be associated with magnetic

particles. Monthly (30 days) deposition rate (De in mg/m2)

for an element were calculated using the formula

De ¼ Ce �M=A=T � 30, where Ce is the concentration of a

given element (mg/g), M is the weight of the settled dust

(g), A is the area of the sampling pot (m2), and T is the time

Environ Earth Sci

123

of sampling (days). Threshold limit values are also given

for the monthly deposition rates in Hungary.

Bulk mineralogical compositions of the samples were

analyzed by a Philips PW1710 X-ray diffractometer using

the following sample preparation: 40 mg of each of the

samples was suspended in ethanol, and then they were

sedimented onto steel plates with an area of 12 cm2.

Analytical transmission electron microscopy analyses were

carried out to characterize the mineralogy and chemistry of

individual solid particles with special emphasis on those

containing heavy metals. The dust samples were suspended

in ethanol, and then they were dropped onto a holey car-

bon-coated Cu grid for the analyses. The measurements

were taken on a Philips CM 20 TEM with a LaB6 filament,

equipped with a Noran energy-dispersive spectrometer

(EDS). For the chemical analyses, a 20-nm spot size and

counting times of 100 s were used. The chemical compo-

sition was calculated estimating 20 nm sample thickness

and an average density of 3 g/cm3. The relative standard

deviations of the EDS analyses are below 2.5, 10 and 50 %

for element concentrations [10, 1–10 and \1 %, respec-

tively. We pretended to analyze only one discrete particle

in each case, which could be confirmed from the corre-

sponding diffraction pattern. The identification of the

individual mineral phases was performed based on their

diffraction pattern and chemical composition.

Results and discussion

Dust deposition

The average dust deposition rate for a month is 6.9 g/m2

showing large variation between 0.5 and 21 g/m2

(Table 1). These values fall within similar range as

reported for other cities (Krolak 2000; Inomata et al. 2009;

Zhao et al. 2010). The threshold limit value for the dust

deposition is 16 g/m2 for urban areas in Hungary (Bartofi

2000). It is exceeded in 5 samples out of 64, exclusively in

the summer season and at the front side of the building.

However, there is no significant difference between the

front and back side of the building either in averages or in

ranges of the dust deposition values.

Autumn and winter samples are generally characterized

by low dust quantities and nearly uniform vertical distri-

bution pattern (on average 3–4 g/m2 monthly) with the

highest values at 2 m front. In spring samples, the dust

amount sharply decreases upwards at the front side of the

building with about twice as high average (8.0 g/m2) and

maximum values than in winter or autumn. The largest dust

deposition were found in the summer samples (on average

11 g/m2), which was also observed by Krolak (2000) in

similar weather conditions. In the summer, highest dust

deposition was observed at 9 m both at the front and at the

back sides of the building. On the contrary, the very low

deposition at 2 m back (protected site by local morphol-

ogy) results in low average deposition values at 2 m.

According to Zhao et al. (2010), frequent precipitation

and windy conditions enhance the dust deposition. Addi-

tionally, re-suspension of roadside soil and street dust may

compose the major part (up to 50–70 %) of the dust

deposition (Hunt et al. 1993; Young et al. 2002). This

contribution to the airborne particulate matter is expected

to be much higher in time periods of frequent dry surface

especially when it is coincide with the time of strong wind

conditions. The 30-year-average seasonal meteorological

data for Budapest are shown in Table 2. These data suggest

that weather conditions favor the dust deposition primarily

in summer and in spring. In these periods, the re-suspen-

sion of the surface material is also expected to increase. So

weather conditions result in increased dust deposition in

two ways in our case. Unfortunately, the excess growth of

algae in the sampling pots during summer may have also

influenced the final weight of our samples despite the

hill side (upstream front)downflow of clean air

road side (downstream front)lifting up polluted air

2 m

9 m

21 m

33 m

4-line road railway line

dolomite hill

266 m

~ 100 m~ 600 mNW SE

120 m

hoardingslow-traffic road

Fig. 1 Sketch of the surroundings of the sampling site showing the isolated roughness air-flow model. Sampling locations are shown by the stars

(*), air-flow directions by continuous arrows

Environ Earth Sci

123

algaecide used. However, they did not influence the dust

deposition pattern significantly and did not have any effect

on the deposition of magnetic particles and metals.

According to the isolated roughness air-flow model,

high buildings may affect wind flow structures as follows

(Oke 1988): first, creating down-flow close to the upstream

front elevation (the back side in our case), which transports

clean air to pedestrian levels; and second, developing a

separation bubble over the downstream front elevation (the

front side), which lifts up the polluted air from the

pedestrian level (see Fig. 1). The seasonal dust deposition

patterns found generally correspond to this model except in

summer, when the maximum in dust deposition at 9 m

claims an explanation by a specific air-flow model. This

special case, however, may be related to the co-influence of

local morphology and weather conditions.

Metal enrichment and deposition

The average Cu (290 mg/kg), Fe (21,959 mg/kg), Pb

(1,227 mg/kg) and Zn (1,567 mg/kg) concentrations and

ranges are similar to those found in Central European cities

(Krolak 2000; Popescu and Dumitrescu 2000). There is no

significant linear relationship between the concentrations

of any two studied elements. As metal background values

for settled dust are not available, geoaccumulation indexes

were calculated according to Ji et al. (2008) using the

regional background values after Odor et al. (1997). Iron

could be characterized by no enrichment in the samples as

compared to natural geological formations. Although this

metal may originate from almost any anthropogenic sour-

ces, it is also a major element in natural geological for-

mations. On the contrary, Cu and Zn show moderate, while

Pb heavy enrichment. In several cases, however, Zn also

can be characterized by heavy enrichment, as well as Pb by

extreme enrichment in one single case when a room

was renovated resulting in high concentrations of Pb, Ba

and Ti of paint origin (Fig. 2). As far as Zn shows heavy

Table 1 Averages and ranges of total (TS; 10-6 SI) and mass-specific (MS; 10-6 m3/kg) susceptibility values, as well as monthly dust (g/m2)

and metal deposition rates (mg/m2) with respect to the variation in sampling points

All Front Back 2 m 9 m 21 m 33 m Winter Spring Sum. Aut.

TS

Average 89 111 67 77 137 87 56 76 97 126 48

Range 5.8–480 23–480 5.8–151 5.8–253 18–480 22–193 13–101 47–122 14–253 5.8–480 6.1–132

MS

Average 3.7 4.1 3.3 4.3 4.2 3.2 2.9 3.4 2.9 4.1 5.5

Range 0.7–9.4 0.9–8.0 0.7–9.4 0.7–9.4 1.9–6.4 1.2–7.7 0.9–5.8 1.7–6.2 0.7–5.8 0.9–9.4 2.8–7.7

Dust

Average 6.9 6.9 7.0 5.0 8.9 7.9 6.1 3.8 8.0 10.8 3.2

Range 0.5–21 2.4–21 0.5–20 0.5–15 1.7–21 1.7–20 1.3–16 2.9–4.8 1.8–16 1.2–21 0.5–7.1

Cu

Average 1.7 1.8 1.6 1.9 2.2 1.7 1.2 1.3 1.9 2.7 0.9

Range 0.2–8.3 0.5–5.6 0.2–8.3 0.2–5.6 0.4–8.3 0.4–4.0 0.3–2.6 0.7–1.9 0.5–5.6 0.5–8.3 0.2–2.9

Fe

Average 149.6 168.1 131.2 110.7 202.6 169.0 117.2 100.2 167.6 226.2 71.8

Range 4.2–580 52–580 4.2–392 4.2–328 20–580 35–378 20–254 73–122 17–328 19–580 4.2–180

Pb

Average 10.7 15.7 5.7 0.9 20.3 17.3 4.0 2.4 4.1 19.6 12.0

Range 0.1–147 0.6–147 0.1–32 0.1–4.1 0.1–126 0.4–147 0.4–18 1.3–4.2 0.3–14 0.1–126 0.1–147

Zn

Average 10.8 11.3 10.3 3.0 16.4 14.6 8.6 7.0 9.5 17.6 6.7

Range 0.3–50 2.0–50 0.3–38 0.3–7.7 0.6–50 2.3–35 3.2–22 3.2–9.9 1.0–2.1 0.8–50 0.3–35

Sum. summer, Aut. autumn

Table 2 The 30-year-average seasonal weather conditions for

Budapest, Hungary (data from the Hungarian Meteorological Service

at www.met.hu)

Precipitation

(mm)

Temperature

(�C)

Sunshine

hours

Windy

([10 m/s)

days

Winter 109 1.5 207 34

Spring 134 11.6 548 48

Summer 157 20.8 771 45

Autumn 132 11.4 397 32

Environ Earth Sci

123

enrichment primarily at the front side and at lower heights

as it is expected, Pb can be characterized by heavy

enrichment at the back side, too. Highest geoaccumulation

index values were found primarily in the summer and some

of the spring samples. Ji et al. (2008) found that the total

fraction of the urban dust (\100 lm) shows practically

similar metal concentrations to natural geological forma-

tions, while the fine fractions (\10 lm) were mostly

heavily enriched with Cr, Co, Cu, Pb and Zn during a study

concerning in 15 Chinese cities. Comparison between the

magnetic properties of vehicle exhaust materials, and these

dust samples suggested that the main source of pollution is

the traffic there (Marton et al. 2011). Literature data show

that the studied metals may originate from traffic sources,

but they are also the common components of the built

environment (Sutherland 2000).

The average monthly metal deposition rates and their

ranges (Table 1) are much higher than those found in rural

environments (Lithuania—Kvietkus et al. 2011). Also,

lower values were found for Cu, Pb and Zn in several cities

(Tokyo, Japan—Sakata and Marumoto 2004; Chicago,

USA—Yi et al. 2001), but there were observed similar

(Clermont, USA—Shahin et al. 2000; Poland—Krolak

2000) and much higher values (Izmir, Turkey—Odabasi

et al. 2002) in other locations, as well. Threshold limit

value is only given for Pb in Hungary which is 1.2 mg/m2

(Bartofi 2000). This value is exceeded in the great majority

(78 %) of the studied samples.

Metal deposition showed similar pattern to the dust

deposition for the different seasons. The highest average

deposition values and the widest range were found in the

summer samples followed by those collected in spring.

Contrarily, autumn and winter samples showed the lowest

averages and the narrowest ranges. The high summer

deposition values are the most expressed at 9 m at the both

sides, where 2–5 times higher values were found for each

studied metal compared to all other sampling place and

date. As this phenomenon was observed for each studied

metal, it could not be explained by the changes in metal

sources. The weather conditions favoring dust deposition

and the increased rate of re-suspended material in summer

and spring may affect the metal deposition together and

result in its increase. Other studies, however, did not find

any significant seasonal variation either in metal deposition

(Odabasi et al. 2002) or in the amount of suspended par-

ticulate matter (Tahri et al. 2012) in urban areas. Others

have found slightly higher values in the heating season

(Krolak 2000). In our case, however, the contribution of

soil/dust re-suspension is the dominant source of the dust

(mainly in the summer season), which may even overwrite

the effect of recent anthropogenic activities. Moreover, the

re-suspended material contains the contribution of past

anthropogenic activities, which could result in much higher

metal concentration than recent activities. This can be also

the reason for the following differences in metal deposition

between the building’s sides. As far as Pb and Fe could be

characterized by much higher deposition values at the front

side than at the back one, Cu and Zn showed no significant

difference between the two sides similarly to the dust

deposition.

The vertical distribution pattern of metals deposition is

similar to that of dust. The metal deposition generally

decreases upwards at the front side, except summer when a

maximum was observed at 9 meters. This pattern was the

most conspicuous for Pb. Higher metal deposition at lower

levels at the front side and at the higher levels at the back

side of the building correspond to the isolated roughness

air-flow model. However, highest average metal deposition

values were found at 9 m, followed by at 21 m, while the

lowest values were found at 33 m. The unexpectedly low

average values at 2 m can be due to the very low deposition

values at the back side of the building at this level. This

sampling site is protected by the morphology of the natural

and built environment (it is placed in a ‘‘valley’’ between

the dolomite hill and the building). This is shown by the

fact that if the deposition values at the front and back sides

are compared at the different sampling levels, 4–8 times

higher values can be found at the front than at the back side

at 2 m, while no difference were found at all other sam-

pling levels between the sides.

Hierarchical cluster analysis based on the linear corre-

lation between the metal deposition rates at the different

sampling sites showed that the three heavy metals studied

show slightly different spatiotemporal deposition charac-

teristics. Such differences in deposition patterns may sug-

gest differences in host phases for the studied elements,

which show different behavior during the re-suspension or

the effect of active anthropogenic sources. Highest simi-

larity was found for Fe and Zn (with a dissimilarity value

„uncontaminated“

heavily contaminated

extremely contaminated

moderately contaminated

Cu Fe Pb Zn-2

0

2

4

6

8G

eoac

cum

ulat

ion

inde

x

Fig. 2 Averages (white lines) and ranges (black columns) of the

geoaccumulation indexes suggesting anthropogenic enrichment of

heavy metals in the dust samples

Environ Earth Sci

123

of 0.22). These two metals show more similar deposition

characteristics to Pb (dissimilarity = 0.28) than to Cu

(dissimilarity = 0.44). This suggests that Zn and Pb may

be associated with Fe at higher proportion than Cu.

Magnetic susceptibility versus dust and metal

deposition

The magnetic susceptibility measurements were evaluated

first providing total susceptibilities meaning that neither the

volume nor the mass of the pollutants was taken into

account. In this case, we basically obtain information about

the amount of magnetic pollutants in each sample. These

values, varying between 5.8 and 480 9 10-6 SI, are not

influenced, e.g., by possible error in weight measurements,

remnants of algaecide or algae. Next the mass suscepti-

bilities were calculated, which facilitate comparison with

the intensity of magnetic pollution at other sampling sites

and differently collected materials (like total suspended

particles or PM10). The mass susceptibilities varied

between 0.7 and 9.4 9 10-6 m3/kg. These values are

comparable with those measured in other urban areas (e.g.,

Booth et al. 2006; Lu et al. 2008). As mass-specific mag-

netic susceptibility shows significant linear relationship

either with dust deposition or with metal concentrations

and deposition just sporadically (Table 3), only total sus-

ceptibility values will be evaluated below. The only

exception is the strong linear relationship between the iron

concentration in the dust and its mass-specific suscepti-

bility (r = 0.60; p \ 0.05), which is expected as iron oxide

minerals are the major source of magnetism in such kind of

samples (Elzinga et al. 2011). Otherwise, the average

values and ranges of the mass-specific susceptibility show

very little variation with respect to the spatiotemporal

variation of sampling. We found a bit higher mass-specific

susceptibility values at the front side of the building than at

the back one. Additionally, it shows only slight or no

decrease with height and its highest values were found in

the autumn, while the lowest ones in the spring samples

(Table 1).

Average total susceptibility values and their ranges

show the same vertical distribution pattern as dust or metal

deposition do. It shows generally higher values at the front

than at the back side. Its highest values were found at 9 m,

followed by at 21 m, 2 m and the least at 33 m. Again,

much higher values were found in the summer and spring

samples than in the autumn and winter samples. Total

susceptibility generally shows close linear relationship

both with dust and metal deposition (Table 4). Its linear

correlation is the strongest with Fe deposition as it is

expected, and the weakest with Pb and Zn, suggesting that

these metals can be associated not solely with magnetic

particles. Sequential chemical extraction of metals from

urban dust samples showed that besides the residual phases

(which magnetite also belongs to), oxidizable (for Cu, Pb),

reducible (for Zn, Pb) and easily extractable (for Zn)

fractions are the dominating ones generally (Banerjee

2003; Wang et al. 2011).

In the winter samples, however, there are mostly poor

linear relationships between the total susceptibility and

dust or metal deposition. This is probably due to the

weather conditions resulting in different deposition condi-

tions for the particulate matter (Parameswaran and Vija-

yakumar 1994). In contrast, closest relations were found in

the summer and spring samples, namely the higher the

potential of re-suspension, the closer the linear relationship

between total susceptibility and metal deposition. This

suggests that magnetic particles are important components

of the re-suspendable material, while other potential metal-

bearing phases such as clay minerals are not due to their

different sedimentation abilities as compared to magnetic

particles. This behavior of particles is shown by the fact

that the linear relationship between total susceptibility and

metal deposition is weaker at greater heights than at lower

ones as clay minerals with relatively large surface are able

to reach greater heights than the isomorphic magnetic

particles.

Total susceptibility was found to be a useful indicator

not only for the anthropogenic contribution to dust depo-

sition but also to that of dust weight (Fig. 3). In case of

both dust weight and deposition, around 50 % of their

variation could be explained by the corresponding variation

in total susceptibility. The potential of environmental

magnetism as a proxy for atmospheric pollution levels has

Table 3 Pearson’s correlation coefficients (r; p = 0.05) between mass-specific susceptibility and monthly dust/metal deposition rates with

respect to the variation in sampling points

All Front Back 2 m 9 m 21 m 33 m Spring Sum. Aut. Winter

Dust -0.26 0.10 -0.47 0.10 -0.36 -0.48 -0.58 -0.16 -0.03 -0.21 -0.35

Cu -0.15 0.09 -0.30 0.17 -0.70 -0.23 -0.58 0.55 -0.22 0.08 -0.38

Fe -0.07 0.18 -0.36 0.17 -0.12 -0.25 -0.58 0.11 0.29 -0.03 0.07

Pb -0.03 -0.03 -0.35 0.27 0.22 -0.25 -0.33 -0.33 0.47 -0.21 0.58

Zn -0.20 -0.08 -0.33 0.19 -0.27 -0.40 -0.36 -0.44 0.10 -0.29 0.62

Sum. summer, Aut. autumn

Environ Earth Sci

123

been reported in several studies. They concluded that

magnetic susceptibility provides a first indication of the

concentration of ferrimagnetic minerals in the dust. Hence,

such measurements on the dust can serve as a comple-

mentary tool for the routinely used geochemical methods

(Lu et al. 2008).

The distribution of the total susceptibilities among col-

lecting sites, however, does not always follow strictly that

of the amounts of dust. The differences are most striking

for the back side, summer of 2009 and spring of 2010,

where the total susceptibilities show a decreasing trend

from 9 m upwards. The susceptibility values are lower in

the back than in the front side of the building, whereas the

dust amount is unexpectedly high at the back side. A

possible explanation is that the mass of the dust is

increased by algae, while the susceptibility truly reflects

the degree of pollution. That is why total susceptibility is

much more useful proxy for anthropogenic dust and metal

deposition than mass-specific susceptibility as the former is

not influenced by potential errors resulted from sampling

and/or weight measurements. It is important to note that the

sharp peak at 9 m in summer samples can be observed in

the magnetic susceptibilities in each case.

When compared to the relation between total suscepti-

bility and dust weight or deposition, much better fit was

found for the relation between total susceptibility and iron

deposition (r2 = 0.77), similar for Cu (r2 = 0.48) and Zn

(r2 = 0.41) deposition, and slightly lower for Pb deposition

(r2 = 0.30) (Fig. 4). This suggests that significant part of

Cu and Zn can be associated with magnetic particles, while

slightly lower ratio of Pb could be found in such phases.

Although these metals are often associated with magnetic

particles in urban environment (Filipelli et al. 2005), sev-

eral studies showed that significant proportion (up to 60 %)

of total metal concentration in the dust could be dissolved

by weak acids which do not mobilize them from magnetite

(Duong and Lee 2009). It also corresponds to the obser-

vations of Barrett et al. (2010) who found that Pb occurs

primarily in the form of Pb-sorbed goethite, as well as lead

chromate, chloride, carbonate, oxide and phosphate in the

urban dust of Manchester, UK. Most of these phases are

non-magnetic so their detection by magnetic measurements

could fail, and it explains the absence of very strong cor-

relation between susceptibility and metal deposition in

urban dust.

Table 4 Pearson’s correlation coefficients (r; p = 0.05) between apparent susceptibility and monthly dust/metal deposition rates with respect to

the variation in sampling points

All Front Back 2 m 9 m 21 m 33 m Spring Sum. Aut. Winter

Dust 0.72 0.91 0.76 0.89 0.79 0.63 0.74 0.70 0.67 0.64 0.03

Cu 0.70 0.74 0.79 0.94 0.47 0.78 0.54 0.92 0.52 0.78 -0.10

Fe 0.88 0.94 0.86 0.92 0.92 0.84 0.80 0.86 0.89 0.76 0.40

Pb 0.54 0.51 0.77 0.84 0.95 0.03 0.73 0.35 0.95 0.12 0.48

Zn 0.64 0.63 0.83 0.92 0.77 0.53 0.47 0.33 0.74 0.10 0.62

Sum. summer, Aut. autumn

y = 0.0511x + 2.3467R² = 0.5138

y = 0.002x + 0.0946R² = 0.5082

0.01

0.1

1

10

100

0 100 200 300 400 500 600

Dus

t wei

gth

(g)

and

mon

thly

dep

ositi

on (

mg/

m2 )

Total susceptibility (10-6 SI)

Deposition Weigth

Fig. 3 Relationship between apparent susceptibility and dust weight

and deposition

y = 0.0201x + 0.0906R² = 0.4834

y = 1.3376x + 28.037R² = 0.7715

y = 0.1845x -5.9045R² = 0.295

y = 0.092x + 2.4077R² = 0.4081

0.01

0.1

1

10

100

1000

0 100 200 300 400 500 600 700

Mon

thly

met

al d

epos

ition

(m

g/m

2 )

Total susceptibility (10-6 SI)

Cu Fe Pb Zn

Fig. 4 Relationship between apparent susceptibility and metal

deposition

Environ Earth Sci

123

Dust mineralogy

The bulk mineralogical composition of the samples reflects

primarily the geological characteristics of the sampling

area. The main components are in the order of frequency:

quartz (60–90 wt%), dolomite (2–20 wt%), calcite

(1–15 wt%), feldspar (3–6 wt%), mica (1–5 wt%), chlorite

(1–5 wt %). These phases are characteristic natural com-

ponents of settled dust (Farkas and Weiszburg 2006).

However, Zhao et al. (2010) found that some portion of

these phases (quartz, feldspar, carbonates) could appear

also in amorphous forms in the urban settled dust which

suggest their anthropogenic origin. The most characteristic

temporal change in the bulk mineralogical composition of

the dust samples is the higher quartz content of the autumn

and winter samples (80–90 %) than that of the other ones

(60–80 %), as well as the increase in the contribution of

dolomite and calcite to the spring and summer samples.

This suggests the higher importance of local material (cp.

dolomitic hills behind the building) to dust mineralogy in

spring and summer, when weather conditions favor the re-

suspension and deposition of high amount of local material.

This is partly supported by the outstandingly high calcite

content at 9 meters in summer (up to 15 %). Interestingly,

the amount of dolomite is twice as high at the front as at the

back side of the building despite the fact that the dolomite

hill is at the back side of the building. Significant amount

of amorphous (organic?) material was found in the summer

samples, probably due to increased contribution of plant

materials (debris, pollen, etc.) and the unfortunate presence

of algae in the sampling pots. Trace amounts of gypsum

also appears in the autumn samples. Gypsum is general

component of the construction materials, but it may form

also due to the reaction between sulfuric acid and calcic

material in several anthropogenic processes (Panigrahy

et al. 2003).

Transmission electron microscopic analyses showed that

the most significant components of the settled dust are

different mineral phases. The most frequent particles are

gypsum, quartz, feldspar, layer silicates, calcite, dolomite

as well as Fe and Ti oxides. The most characteristic size

range is the one of 10–20 lm for most of the particles, but

its dominance may be partly due to the samples preparation

technique. The most important potentially toxic metal-

bearing phases are magnetite and clay minerals in the

samples (Fig. 5). Additionally, Zn was found to be asso-

ciated with a calcite particle in one case. Zinc could be

associated with both clay and Fe-oxide particles, while lead

primarily to the latter ones. The silicate and oxide particles

are often form aggregates with each other. The Zn content

Fig. 5 TEM microphotographs

and diffraction patterns, as well

as EDS spectra for dominant

heavy metal-bearing mineral

particles in the dust (from the

samples collected at 2 m in

the spring of 2010).

A = Zn-bearing smectite,

B = Zn-bearing magnetite,

C = Pb-nearing magnetite

Environ Earth Sci

123

of clay minerals can be as high as 5 wt%, while Fe-oxides

are characterized by a slightly lower Zn content (up to

2.5 wt%). The Pb content of the latter one phases is gen-

erally between 2 and 3 wt%, and they also contain a small

amount of Mn (around 0.5 wt%). Among Fe-oxide parti-

cles, both magnetite and hematite were identified. Addi-

tionally, ilmenite and titanite were also found in the

samples, but they do not contain significant amount of

heavy metals (except one ilmenite particle which contained

0.5 wt% Mn). Urban anthropogenic particles are often

enriched in toxic trace metals (Maher 2009). Magnetite

particles in the dust may be partly originated from the

anthropogenic emissions, while clay particles derived

rather from the re-suspension of urban soils. Magnetite

particles are resistant to weathering releasing its toxic

components slowly to the environment. However, its close

association with hematite suggests its oxidation, which

may proceed already in the anthropogenic combustion

process as showed by the results of Zajzon et al. (2013)

who found close association between magnetite and

hematite in vehicle exhaust materials. This latter phase is

much less resistant than magnetite (Silva et al. 2007), and

together with layer silicates, they may be the potential

sources of mobile toxic metals in the studied dust samples.

Conclusions

The settled dust samples were moderately contaminated by

Cu and Zn, while heavily by Pb. The enrichment of these

metals is generally the highest at the roadside of the

building and at lower sampling heights, but Pb and Zn may

show strong enrichment even at 33 m and the former metal

at the low-traffic side of the building, too. The seasonal

variation in metal deposition is generally related to weather

conditions, which can be also related to the increase in the

re-suspension of urban soil and road dust.

This process dominates the seasonal deposition pattern

not only for these metals but also for the dust. Their similar

vertical deposition pattern also supports this phenomenon.

However, the lack of the closest linear relationship between

most of the metals and dust deposition suggests the con-

tribution of recent anthropogenic activities to metal

enrichment in the dust. The vertical dust (and metal)

deposition pattern mostly corresponds to the supposed air-

flow model, but just in case of highest chance of re-sus-

pension, a special model is needed to clearly describe them.

This is probably due to the simultaneous effect of weather

conditions and the morphology of natural and built

environment.

Although the mass-specific susceptibility of the settled

dust show very close linear relationship with its iron con-

centration, neither the dust nor the metal deposition (even

that of iron) could be traced by this proxy in this case. In

contrast, total susceptibility could be a very useful tool to

monitor both dust and metal (Cu, Fe, Pb and Zn) deposition

in such samples, suggesting that these metals travel toge-

ther at least party with magnetic particles even in case of

re-suspension. The generally possible errors concerning to

the sampling of settled dust (such as appearance of algae,

remnants of algaecide, errors in weight measurements due

to changes in relative humidity) could be also avoided if

total susceptibility values are used to monitor settled dust.

The most important heavy metal-bearing phases in the

settled dust are iron oxides and clay minerals. Iron oxides

was found to be the primary sink of Pb in the samples,

while Zn was mostly associated with clay minerals, but

both mineral groups contained the other metals, too. The

strongly different sedimentation characteristics of these

phases seemed to be reflected in the vertical metal depo-

sition patterns. Iron oxide particles are relatively resistant

to weathering releasing their components slowly to the

environment, while layer silicates may be one of the

potential sources of mobile metals in the settled dust.

Acknowledgments This study was financially supported by the

Hungarian Scientific Research Fund (OTKA K 76317 and K 75395).

Peter Sipos also thanks for the support of the Janos Bolyai Research

Scholarship of the Hungarian Academy of Sciences.

References

Banerjee ADK (2003) Heavy metal levels and solid phase speciation

in street dusts of Delhi, India. Environ Pollut 123:95–105.

doi:10.1016/S0269-7491(02)00337-8

Barrett JS, Taylor KG, Hudson-Edwards K, Charnock JM (2010)

Solid-phase speciation of Pb in urban road dust sediment: a

XANES and EXAFS study. Environ Sci Technol 44:2940–2946.

doi:10.1021/es903737k

Bartofi I (2000) Environmental technology (in Hungarian). Mez}ogaz-

da Kiado, Budapest

Booth CA, Shilton V, Fullen MA, Walden J, Worsley AT, Power AL

(2006) Environmental magnetism: measuring, monitoring and

modeling urban street dust pollution. In: Longhurst JWS and

Brebbia CA (eds.) WIT transactions on ecology and the

environment, vol 86, Air Pollution XIV. WIT Press, pp 333–342

Duong TTT, Lee BK (2009) Partitioning and mobility behavior of

metals in road dusts from national-scale industrial areas in

Korea. Atmos Environ 43:3502–3509. doi:10.1016/j.atmosenv.

2009.04.036

Elzinga EJ, Gao Y, Fitts JP, Tappero R (2011) Iron speciation in

urban dust. Atmos Environ 45:4528–4532

Farkas I, Weiszburg T (2006) Mineralogical investigation of settled

and airborne dust collected from Cluj County, Romania. Foldtani

Kozlony 136:547–572

Filipelli GM, Laidlaw MAS, Latimer JC, Raftis R (2005) Urban lead

poisoning and medical geology: an unfinished story. GSA Today

15:4–11

Gautam P, Blaha U, Appel E (2005) Magnetic susceptibility of dust-

loaded leaves as a proxy of traffic-related heavy metal pollution

in Kathmandu city, Nepal. Atmos Environ 39:2201–2211.

doi:10.1016/j.atmosenv.2005.01.006

Environ Earth Sci

123

Hunt A, Jones J, Oldfield F (1984) Magnetic measurements and heavy

metals in atmospheric particulates of anthropogenic origin. Sci

Total Environ 33:129–139. doi:10.1016/0048-9697(84)90387-5

Hunt A, Johnson DL, Thornton I (1993) Apportioning the sources of

lead in house dusts in the London borough of Richmond,

England. Sci Total Environ 138:183–206. doi:10.1016/0048-

9697(93)90414-2

Inomata Y, Igarashi Y, Chiba M, Shinoda Y, Takahashi H (2009) Dry

and wet deposition of water-insoluble dust and water-soluble

chemical species during spring 2007 in Tsukuba, Japan. Atmos

Environ 43:4503–4512. doi:10.1016/j.atmosenv.2009.06.048

Ji Y, Feng Y, Wu J, Zhu T, Bai Z, Duan C (2008) Using

geoaccumulation index to study source profiles of soil dust in

China. J Environ Sci 20:571–578. doi:10.1016/S1001-0742(08)

62096-3

Krolak E (2000) Heavy metals in falling dust in Eastern Mazowieckie

province. Pol J Environ Stud 9:517–522

Kvietkus K, Sakalys J, Valiulis D (2011) Trends of atmospheric

heavy metal deposition in Lithuania. Lith J Phys 51:359–369.

doi:10.3952/physics.v51i4.2258

Laidlaw MAS, Filippelli GM (2008) Resuspension of urban soils as a

persistent source of lead poisoning in children: a review and new

directions. Appl Geochem 23:2021–2039. doi:10.1016/j.appgeo

chem.2008.05.009

Lu SG, Zheng YW, Bai SQ (2008) A HRTEM/EDX approach to

identification of the source of dust particles on urban tree leaves.

Atmos Environ 42:6431–6441. doi:10.1016/j.atmosenv.2008.04.

039

Maher BA (2009) Rain and dust: magnetic record of climate and

pollution. Elements 5:229–234. doi:10.2113/gselements.5.4.229

Manasreh WA (2010) Assessment of trace metals in street dusts of

Mutah city, Karak, Jordan. Carpath J Earth Environ 5(1):5–12

Marton E, Sipos P, Nemeth T, May Z (2011) Transport of pollutants

around a high building: integrated magnetic, mineralogical and

geochemical study. In: Conference Proceedings and Exhibitor’s

Catalogue, 6th Congress of the Balkan Geophysical Society (3–6

Oct 2011, Budapest, Hungary), European Association of Geo-

scientists and Engineers, B9 1–6

Micallef A, Deuchar CN, Colls JJ (1998) Indoor and outdoor

measurements of vertical concentration profiles of airborne

particulate matter. Sci Total Environ 215:209–216. doi:10.1016/

S0048-9697(98)00128-4

Hungarian standard No. MSZ 21454/1-83 (1983) Test of solid

impurities in ambient atmosphere. Determination of settled dust

mass. Hungarian Standards Institution, G 23:1–6

Muxwothy AR, Scmidbauer E, Petersen N (2002) Magnetic proper-

ties and Mossbauer spectra of urban particulate matter: a case

study from Munich, Germany. Geophys J Inter 150:558–570.

doi:10.1046/j.1365-246X.2002.01725.x

Odabasi M, Muezzinoglu A, Bozlaker A (2002) Ambient concentra-

tions and dry deposition fluxes of trace elements in Izmir,

Turkey. Atmos Environ 36:5841–5851. doi:10.1016/S1352-2310

(02)00644-1

Odor L, Horvath I, Fugedi U (1997) Low-density geochemical

mapping in Hungary. J Geochem Explor 60:55–66. doi:10.1016/

S0375-6742(97)00025-3

Oke TR (1988) Street design and urban canopy layer climate. Energy

Build 11:103–113. doi:10.1016/0378-7788(88)90026-6

Panigrahy PK, Goswami G, Panda JD, Panda RK (2003) Differential

comminution of gypsum in cements ground in different mills.

Cem Concr Res 33:945–947. doi:10.1016/S0008-8846(02)

00992-4

Parameswaran K, Vijayakumar G (1994) Effect of atmospheric

relative humidity on aerosol size distribution. Indian J Radio

Space 23:175–188

Popescu GC, Dumitrescu L (2000) Heavy metals distribution in dust

from the central part of Bucharest. Rom J Miner Dep 79:85–87

Remeteiova D, Smincakova E, Florian K (2007) Study of the

chemical properties of gravitation dust sediments. Microchim

Acta 156:109–113. doi:10.1007/s00604-006-0603-0

Sakata M, Marumoto K (2004) Dry deposition fluxes and deposition

velocities of trace metals in the Tokyo metropolitan area

measured with a water surface sampler. Environ Sci Technol

38:2190–2197. doi:10.1021/es030467k

Samet JM, Dominici F, Curriero FC, Zeger SL, Coursac I (2000) Fine

particulate air pollution and mortality in 20 U.S. cities,

1987–1994. N Engl J Med 343:1742–1749. doi:10.1056/

NEJM200012143432401

Shahin U, Yi SM, Paode RD, Holsen TM (2000) Long-term elemental

dry deposition fluxes measured around Lake Michigan with an

automated dry deposition sampler. Environ Sci Technol

34:1887–1892. doi:10.1021/es9907562

Silva M, Kyser K, Beauchemin D (2007) Enhanced flow injection

leaching of rocks by focused microwave heating with in-line

monitoring of released elements by inductively coupled plasma

mass spectrometry. Anal Chim Acta 584:447–454. doi:10.1016/

j.aca.2006.11.043

Sutherland RA (2000) Bed sediment-associated trace metals in an

urban stream, Oahu, Hawaii. Environ Geol 39:611–627. doi:10.

1007/s002540050473

Tahri M, Bounakhla M, Ait Bouh H, Benyaich F, Noack Y, Essaid B

(2012) Application of nuclear analytical techniques (XRF and

NAA) to the evaluation of air quality in Moroccan cities—case

of Meknes city. Carpath J Earth Environ 7(2):231–238

UNEP and WHO (1992) Urban air pollution in megacities of the world.

In: Earthwatch: global environmental system. Blackwell, Oxford

Vardoulakis S, Fisher BEA, Pericleous K, Gonzalez-Fresca N (2003)

Modelling air quality in street canyons: a review. Atmos Environ

37:155–182. doi:10.1016/S1352-2310(02)00857-9

Vento SD, Dachs J (2007) Atmospheric occurrence and deposition of

polycyclic aromatic hydrocarbons in the northeast tropical and

subtropical Atlantic ocean. Environ Sci Technol 41:5608–5613.

doi:10.1021/es0707660

Wang LJ, Lu XW, Lei K (2011) Speciation and transfer of heavy

metals in street dust, soil and river sediment of Baoji city. Urban

Environ Urban Ecol 24:22–26

Yi SM, Shahin U, Sivadechathep J, Sofuoglu SC, Holsen TM (2001)

Overall elemental dry deposition velocities measured around

Lake Michigan. Atmos Environ 35:1133–1140. doi:10.1016/

S1352-2310(00)00242-9

Young TM, Heeraman DA, Sirin G, Ashbaugh LL (2002) Re-

suspension of soil as a source of airborne lead near industrial

facilities and highways. Environ Sci Technol 36:2484–2490.

doi:10.1021/es015609u

Zajzon N, Marton E, Sipos P, Kristaly F, Nemeth T, kovacs Kis V,

Weiszburg T (2013) Integrated mineralogical and magnetic

study of magnetic airborne particles from potential pollution

sources in industrial-urban environment. Carpath J Earth Envi-

ron 8(1):179–186

Zhao J, Peng P, Song J, Ma S, Sheng G, Fu J (2010) Research on flux

of dry atmospheric falling dust and its characterization in a

subtropical city, Guangzhou, South China. Air Qual Atmos

Health 3:139–147. doi:10.1007/s11869-009-0062-y

Environ Earth Sci

123