You are on page 1of 10

Heavy metal determination in atmospheric particulate matter by

Instrumental Neutron Activation Analysis


P. Avino
a,

, G. Capannesi
b
, A. Rosada
b
a
Dipartimento Insediamenti Produttivi ed Interazione con l'Ambiente, Istituto Superiore per la Prevenzione E la Sicurezza del Lavoro (ISPESL),
via Urbana 167, I-00184, Rome, Italy
b
Dipartimento Tecnologie Fisiche e Nuovi Materiali, Ente Nazionale per le Nuove Tecnologie, l'Energia e l'Ambiente (ENEA),
Centro Ricerche Casaccia, via Anguillarese 301, I-00060, Rome, Italy
Received 1 October 2007; accepted 21 November 2007
Available online 31 December 2007
Abstract
Instrumental Neutron Activation Analysis (INAA) is employed for its important analytical properties. Fundamentally, INAA is a multi-
elemental technique allowing the determination of about 40 elements with a good Limit of Detection. In this paper we applied this nuclear
technique to study the element composition in PM10 determining about 30 elements.
25 filters were collected in downtown Rome from October 1999 to April 2000 and irradiated at the nuclear reactor Triga Mark II (ENEA-
Casaccia Laboratories). The -ray measurements have allowed the quali- and quantitative analysis. The element levels in PM10 with the relative
correlations have been determined: basically, the concentrations are very low.
Furthermore, the enrichment factors of all elements will be reported in order to understand the natural or anthropogenic origins of the
particulate matter: some elements may be attributed to long-range transport phenomena from other natural and/or anthropogenic sources.
2007 Elsevier B.V. All rights reserved.
Keywords: Instrumental Nuclear Activation Analysis; Element; PM10; Anthropogenic sources
1. Introduction
The air quality in Rome area is a notable and complex problem
and it is not easy to find out a good solution. Air pollution in Rome
from gaseous species and particulate matter is basically due to
traffic, local heating and industrial emissions. The major pol-
lutants derived from these sources are sulfur dioxide, suspended
particles, nitrogen oxides, carbon monoxide, hydrocarbons,
ozone and other photochemical pollutants [1]. The ranges of
these pollutants are very wide showing important variations in the
range of seasons, weeks, days, hours and even minutes. Further-
more, the mechanisms of occurrence of the pollutant behavior in
such area are only evidenced by multi-elemental analysis. These
few considerations mean the hard work necessary to understand
the atmospheric pollutants dynamic in a megacity such as Rome.
Among the various species present in the particulate matter, a
great attention has been devoted since many years to the study of
the elements with elevated toxicity and great diffusion in the
environment (As, Cd, Cr, Hg, Ni, Pb, etc.) because of
anthropogenic (autovehicular, domestic heating, industrial emis-
sion) and natural (air mass transport, volcanic eruption, desert,
etc.) pollutant emissions. In particular, the metals differ fromother
toxic substances because they are neither created nor destroyed by
humans. Nevertheless, utilization by human influences the
potential health effects in two major ways: first, by direct emis-
sions, that is, by human or anthropogenic contribution to air,
water, soil and food, and second by altering the speciation or
biochemical form of the element [2]. The geologic and biological
cycles naturally redistribute metals by physical process and
bioconcentration, respectively. Therefore, metals distribution in
the environment reflects both natural sources and contribution
Available online at www.sciencedirect.com
Microchemical Journal 88 (2008) 97106
www.elsevier.com/locate/microc

Corresponding author. DIPIA-ISPESL, via Urbana 167, I-00184, Rome,


Italy. Tel.: +39 06 4714242; fax: +39 06 4744017.
E-mail address: pasquale.avino@ispesl.it (P. Avino).
0026-265X/$ - see front matter 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.microc.2007.11.005
fromindustrial activities [2]. The aimof the atmospheric pollution
evaluation is thus to understand levels and behavior of tox-
icologically and geochemically important elements (e.g., Au and/
or Sc, used for evaluation of the enrichment factors of an-
thropogenic elements).
The group of heavy metals and other potentially toxic
elements is rendered particularly important for both its relative
abundance and high amount production: for many species the
anthropogenic contribution alters deeply the natural fluxes
(Table 1) [3]. The elements can be distinguished in three various
categories in relation to their potential effects on the biological
and human systems: essential elements, critical elements and
toxic elements.
Because of local and global contamination by these substances
and their relative effects on human health [4] establish the study of
the trace element composition in fine particulate matter (PM10)
between the more important scientific tasks in environmental
chemistry. This paper shows the research performed by Instru-
mental Neutron Activation Analysis (INAA) for determining
element spectra in PM10 sampled in a single site located in a
residential area in downtown Rome. Taking advantage of the
potentiality and the versatility of the INAA, the main purposes of
this study have been aimed to know the pollution levels due to
metals and/or toxic elements investigating also the correlations
between sources, such as autovehicular traffic, domestic heating
and marine aerosol, and pollutant concentrations.
2. Experimental
First of all, in the frame of the whole project, PM10 mass was
determined by both Tapered Element Oscillating Microbalance
(TEOM, Rupprecht & Patashnick Co Inc., Albany, NY, USA)
and SM200 sampler (Opsis, Sweden) both of them operating at
16.7 mL/min and equipped with a 10-m sampling head. The
comparison between the two different instruments is reported.
The two PM10 measurements are based on two different
methodological approaches: TEOM analyzer is an automatic
instrument and the particulate matter amount sampled on filter
(Fig. 2) is calculated using the Stokes' law whereas the SM200
is based on PM deposition onto the filter (2.0 m pore size and
37 m diameter, Millipore Co, Bedford, MA) and relative
weighting (mod. Umx-2, Mettler Toledo Inc., Columbus, OH).
The period investigated was seven months, from October 1999
to April 2000: the instruments operated simultaneously for 24-
hour periods, generally every four days, except during February
were 15 continuous daily samples were taken for understanding
the element trends in relationship with the atmospheric dynamic
[5]. All the sampling campaign was performed in a site located
in downtown Rome which can be considered representative of
the residential urban area because it is surrounded by buildings
and no heavy local traffic is present.
All the samples were undergone to INAA to take more in-
formation as possible about the elemental composition. The
choice of using INAA, a nuclear not-destructive technique, is due
to its well-known characteristic of reference analytical technique
because all the experimental steps are totally traceable and no
physicalchemical sample manipulation reducing the positive
and/or negative artifacts formation [611]. Furthermore, because
of its high sensitivity and multi-elemental character allowing the
determination of many elements with a good Limit of Detection
(LOD) (Table 2) and accuracy, the INAA is a more suitable
technique than other instrumental methods (e.g., Inductively
Coupled Plasma-Atomic Emission Spectrometry, Inductively
Coupled Plasma Mass Spectrometry, X-ray fluorescence) for
trace/ultra-trace metal and rare earth analysis as deeply reported in
refs. [13,14].
Table 1
Natural and anthropogenic element fluxes (value 10
8
g y
1
)
Element Fluxes from crustal
origin
Fluxes from
volcanic origin
Fluxes in gas of
volcanic origin
PM emission of
industrial origin
Element flux from
fossil fuel
Total
Emission
Atmospheric
Alteration Factor
Ag 0.5 0.1 0.0006 40 10 50 8333
Al 356,500 132,750 8.4 40,000 32,000 72,000 15
As 25 3 0.1 620 160 780 2786
Cd 2.5 0.4 0.001 40 15 55 1897
Co 40 30 0.04 24 20 44 63
Cr 500 84 0.005 650 290 940 161
Cu 100 93 0.012 2200 430 2630 1363
Fe 190,000 87,750 3.7 75,000 32,000 107,000 39
Hg 0.3 0.1 0.001 50 60 110 27,500
Mn 4250 1800 2.1 3000 160 3160 52
Mo 10 1.4 0.02 100 410 510 4474
Ni 200 83 0.0009 600 380 980 348
Pb 30 8.7 0.012 16,000 4300 20,300 34,583
Sb 9.5 0.3 0.013 200 180 380 3878
Se 3 1 0.13 50 90 140 3390
Sm 32 9 7 5 12 29
Sn 50 2.4 0.005 400 30 430 821
Ti 23,000 12,000 3600 1600 5200 15
V 500 150 0.05 1000 1100 2100 323
Zn 250 108 0.14 7000 1400 8400 2346
Atmospheric Alteration Factor (%) =Total Emission/ (Fluxes from crustal origin+Fluxes from volcanic origin) 100.
98 P. Avino et al. / Microchemical Journal 88 (2008) 97106
After drying, the samples, blank and standards were irradiated
at a neutron flux of 2.610
12
ncm
2
s
1
for 32.55 h in the
rotary rack Lazy Susan and at 2.6810
13
ncm
2
s
1
for
1 min in the central channel of the nuclear reactor Triga Mark II of
the ENEA-Casaccia Laboratories. These conditions allow high
flux homogeneity (N99.8%) during the irradiation because each
sample undergoes the same neutron flux (neutron fluency was
3.0510
17
ncm
2
) and the max irradiation temperature in the
rotating rack reaches 40 C. The neutron flux was verified using
an Au standard.
After irradiation, -ray spectrometry measurements of dif-
ferent duration were carried out using a Ge(HP) Ortec detector
(FWHM 1.68 keV at 1332 keV). According to the scheme
reported in Fig. 1 more than 30 elements such as Ag, As, Au, Br,
Ca, Cd, Ce, Co, Cr, Cs, Eu, Fe, Hf, Hg, La, Mg, Mn, Mo, Na,
Nd, Ni, Rb, Sb, Sc, Se, Sm, Ta, Th, V, W, Yb, Zn were
determined [6]. Pb was determined by Inductively Coupled
Plasma-Atomic Emission Spectrometry (ICP-AES, mod. Type-
III Plus, Jobin-Yvon, Munich, Germany) after sample dissolu-
tion in a microwave digestor with 10 mL HNO
3
and few drops
of H
2
O
2
. The analytical spectral line used for Pb analysis was
224.69 nm with a LOD of 0.05 mg/L.
For the analysis primary and secondary standards were
used. Primary standards (Carlo Erba, Milan, Italy) were As,
Cd, Co, Cr, Cs, Fe, Hg, La, Ni, Sb, Se, Sm and Zn whereas
as secondary standards United States Geochemical Survey (U.
S.G.S.) nn. 1, 4, 6 and the Coal Fly Ash (NIST) n. 1633a
were used. The same analytical procedures described above
were applied during an intercomparison organized by the
International Agency for Atomic Energy (IAEA) on air filter
samples: 16 elements were measured by INAA and
Inductively Coupled Plasma-Atomic Emission Spectrometry
(ICP-AES) and the relative values were compared with the
certified values.
Simultaneously, in the same location the meteorological
conditions were followed by natural -radioactivity (SM200):
this parameter used as a tracer of the dynamic of the
boundary layers gives information about the stability and
instability of atmospheric conditions [5]. For this purpose an
SM200 (Opsis, Sweden) analyzer was used giving informa-
tion every 2 h about the remixing atmospheric situation.
3. Results and discussion
3.1. PM mass data
Before applying the nuclear technique for the PM10
composition we investigated the PM10 levels in the
investigated residential urban area. The PM10 levels range
between 14 and 140 g/m
3
in the entire period with an
average value of 58 g/m
3
and a Relative Standard
Deviation (RSD) of 29%. Comparing these PM10 values
with the others that were found in different Roman urban
areas characterized by high density of traffic (e.g., Fermi
Place, Tiburtina Avenue, etc.), the average ratio between
them is about 0.70 with a variability between 0.14 and 2.35
and an elevated data distribution below 1. These ratio values
confirm the thesis that high PM10 levels in Rome are due to
incomplete combustion processes whereas the hypothesis of
resuspended particulate matter in the investigated area is
validated (ratio below 1).
Table 2
Nuclear data and Limit of Detection (LOD) of the elements determined by
INAA
Element Product
nuclide
Cross-section
(barn)
Half life -Ray used
(keV)
LOD
(ng/m
3
)
Ag
110m
Ag 4.5 249.9 d 657.7 0.06
As
76
As 4.3 1.096 d 559.2 1
Au
198
Au 98.8 2.697 d 411.8 0.001
a
Ba
131
Ba 13.5 11.8 d 496.3 5
Br
82
Br 2.69 1.47 d 776.5 0.5
Ca
47
Sc 0.70 3.42 d 159.8 300
Cd
115m
In 300 53.38 h 527.8 0.3
Ce
141
Ce 0.57 32.38 d 145.4 0.05
Co
60
Co 37.2 5.272 y 1332.5 0.01
Cr
51
Cr 15.9 27.7 d 320.0 0.2
Cs
134
Cs 29.0 2.062 y 795.7 0.02
Eu
152
Eu 5900 12.7 y 1408.0 0.01
Fe
59
Fe 1.15 45.1 d 1099.2 50
Hf
181
Hf 12.6 42.5 d 482.2 0.01
Hg
203
Hg 3.8 46.59 d 279.0 0.02
K
42
K 1.46 12.36 h 1524.7 50
La
140
La 9.0 40.27 h 1596.2 0.1
Mg
27
Mg 0.0382 9.46 m 1014.1 50
Mn
56
Mn 13.3 2.576 h 1810.7 0.5
Mo
99
Mo 0.45 2.76 d 141.0 0.2
Na
24
Na 0.53 15.0 h 1368.4 100
Nd
147
Nd 1.3 11.06 d 531.0 1
a
Ni
58
Co 0.113 70.78 d 810.7 1
Rb
86
Rb 72.1 18.66 d 1076.7 1
Sb
122
Sb 6.25 2.70 d 564.0 0.2
Sc
46
Sc 26.5 83.85 d 889.2 0.003
Se
75
Se 51.8 120.4 d 264.6 0.1
Sm
153
Sm 206 1.948 d 103.1 0.41
b
Th
233
Pa 7.40 27.4 d 311.8 0.1
V
52
V 4.88 3.75 m 1434.4 1
W
187
W 37.8 23.9 h 685.7 0.01
a
Yb
175
Yb 65 4.19 d 396.1 0.01
a
Zn
65
Zn 0.78 243.8 d 1115.5 0.5
m: Minutes; h: hours; d: days; y: years.
a
Calculated according to [12], expressed as ppm (g/g).
b
Calculated according to [12], expressed as ppb (ng/g)
Fig. 1. Scheme for INAA of samples and standards. t
i
: irradiation time; t
c
:
cooling time; t
m
: measurement time.
99 P. Avino et al. / Microchemical Journal 88 (2008) 97106
A deep analysis of PM10 trend and its relation to
meteorological conditions was performed for the February
period. During this period (Fig. 2) two different conditions
were present: from the 2nd to the 6th of February atmospheric
stability determined the highest levels of PM10 (almost 100 g/
m
3
) and TEOMvalues always higher than SM200 ones. After the
6th day the situation changed: atmospheric conditions were
unstable, except for two days (7th and 9th due to local
contribution, i.e. road works) when relative higher values of
the pollutant were reached. In particular, in these two days it can
be noted that quite similar PM10 values were determined by the
two methods. Finally, after the 11th day the atmospheric stability
conditions began to establish and, consequently, the TEOM
measurements were higher and the relative concentration levels
rose. About the comparison of two methodologies involved for
determining PM10 and shown in Fig. 2, basically it should be
noted that it is quite good also considering that the TEOM
measures 10% more than the SM200 (except for days 10th and
12th) and the two measurements are based on different
methodological approaches [5].
3.2. INAA quality control
Table 3 shows the analytical quality control of the INAA
data. This control was performed through an intercompar-
ison campaign for 16 elements promoted by IAEA on air
filter samples among different worldwide laboratories (130)
using both spectrochemical (colorimetry, fluorescence, X-
Fig. 2. Comparison between the daily trends of TEOM and SM200 (above) and the -natural radioactivity during the 2nd14th of February 2000.
Table 3
Results of INAA and ICP-AES quality control on IAEA air filter samples (g/g)
Element Measured value Certified value Average value
INAA
meanS.D.
ICP-AES
meanS.D.
Mean Mean 2 (%)
As 4.90.5 5.50.5 5.6 4.59 43
Au 1.260.10 n.d. 1.15 1.06 21
Ba 43.40.5 n.d. 53.8 39.05 40
Cd 10.61.0 9.01.0 9.96 9.8 18
Co 1.30.1 n.d. 1.12 1.18 38
Cr 4.70.8 4.00.4 5.6 4.8 13
Cu 51.60.5 53.60.5 48.8 44.8 16
Fe 19317 n.d. 207.9 200.1 8
Hg n.d. 0.720.18 1.00 0.78 70
Mn 31.21.0 30.22.0 31.9 30.1 14
Mo 1.260.2 n.d. 1.14 1.56 70
Ni n.d. 7.40.7 8.0 7.4 15
Se 1.010.10 n.d. 1.06 1.01 11
U 0.780.10 n.d. 1.02 0.99 14
V 8.040.35 n.d. 8.00 7.2 16
Zn 13218 16816 152 141 12
The measured value is the average of seven determinations on seven different
replicates. S.D.: Standard deviation; n.d.: not detected.
100 P. Avino et al. / Microchemical Journal 88 (2008) 97106
ray fluorescence, infrared spectrometry, atomic absorption
and emission spectrometry, ICP-AES, ICP-MS), electro-
chemical (polarography, voltammetry,) and nuclear (INAA,
isotopic dilution, beta counting) analytical techniques. For
each element in the table the reported values (measured
value) were determined by INAA and ICP-AES and the
certified value: the last column (average value) represents
the value averaged among all the determinations performed
by the different laboratories interested in the measurements.
As it can be noted, the agreement between the two
techniques and the real value is quite good except for
some elements such as Ba, U and Zn. For Ba and U this
discrepancies can be due to the difficulty to analyze such
kind of elements even if for Ba our measured value and
the average value are quite similar. For Zn the situation is
a little bit different: in fact, the certified value (152 g/L)
fall into the INAA and ICP-AES value ranges (13218 and
14117 g/L, respectively).
3.3. Air filter results
Mean concentration values, minimum and maximum
levels and standard deviation between the elements
determined in the PM10 fractions are shown in Table 4
whereas the correlation coefficients of the analyzed
elements are reported in Table 5. Taking in account this
table, we can note the good correlation (0.7bcorrelationb1)
of some elements (in bold in Table 5). Among the good
correlations evidenced in Table 5 (especially for La and
Sm), in particular, for their environmental implications
because they have well-known geochemical mobility and
anthropogenic origin, the very good correlation between Br
and Sb should be underlined (equation curve: y=0.388x
+0.609; R
2
=0.948) and, in general, the good correlation
among Br and other elements. At the same time the good
correlation for Se with other elements should be noted, the
not so good correlation among Zn and other elements
(ranging between 0.023 and 0.652) and the total discre-
pancies among Hg and the other elements meaning there
are no correlation between the Hg (and W as well) and
other element levels in Rome urban area. Finally, it is very
interesting that the Rare Earth Element (REE) levels and
relative good correlations are important for both evaluating
the data quality, understanding the mass transport events
(influencing the sample homogeneity) and the particulate
matter composition and aging [2,10]. In fact, REEs are
considered coming from natural sources and thus their good
correlation is an index of a homogeneous and raw
particulate matter composition.
It should be evidenced that very few papers show results
on a deep investigation on the particulate matter composi-
tion in urban residential area of Rome. In fact, other
studies [17,18] show concentration levels related to few
elements, whereas the complete investigations using INAA
[19,20] were performed by one of the authors and report
data of 10 and 15 years ago, respectively. According to the
considerations on the sampling site reported in the
Experimental section, the data presented in this paper can
be considered as reference values for some elements in
downtown Rome. Basically, the concentration levels of the
all elements are very low: the whole situation is reassuring
regarding the exposition of potential toxic elements. In the
last column of Table 4 are reported the guide values for
very few elements, i.e. Cd, Mn, Ni, Pb and V as
recommended by World Health Organization (WHO) [15]
and European Directive [16]: as it can be seen, for these
toxic elements there are no alarms from a sanitary point of
view for human health.
In Table 6 the comparison among the levels determined in
this study and those analogous found in Rome in previous
decades (years '70 and '80) or in other Italian, European and
international locations are reported. Even if the measure-
ments performed during the 80s campaign are referred to
Table 4
Synoptic table (mean value, minmax values and standard deviation) of
elements concentration (ng/m
3
) determined in the PM10 fractions in downtown
Rome (

expressed as pg/m
3
;

expressed as g/m
3
; by ICP-AES)
Element Mean MinMax St. Dev. Guide values
WHO European
a
Ag 0.176 0.1010.234 0.045
As 1.35 0.1383.578 0.890
Au 0.008 bLOD0.050 0.011
Ba 12.8 2.9062.9 16.0
Br 22.2 4.5063.8 17.5
Ca 1500 4002800 360
Cd 0.526 bLOD1.24 0.253 5
Ce 0.843 0.1301.58 0.455
Co 0.379 0.1800.921 0.179
Cr 7.28 2.9912.9 2.40
Cs 0.151 0.0170.390 0.115
Eu

12.3 1.1020.1 7.11


Fe

0.566 0.0341.22 0.353


Hf 0.020 0.0060.032 0.010
Hg 1.07 0.0503.33 0.842
K 1100 3501800 420
La

188 34.9384 106


Mg 240 1001300 169
Mn 40.0 1560 24.8 0.15

Mo 2.10 0.3804.83 1.36


Na 420 1102100 315
Nd 0.245 0.1800.310 0.091
Ni 4.54 2.2411.6 2.34 20
Pb 92 32195 47.8 0.5

0.5

Rb 2.19 0.2003.70 1.18


Sb 9.22 2.2826.3 6.97
Sc

46.1 5.60101 23.7


Se 0.687 0.1281.66 0.450
Sm

53.4 6.80111 31.1


Th 0.204 0.0380.362 0.107
V 4.02 bLOD9.30 2.0 1
b

W 1.25 0.4503.19 0.740


Yb 0.015 0.0070.025 0.007
Zn 80.0 1.20295 71.6
The guide values (ng/m
3
) are provided by the (WHO) [15] and limit values
are suggested by the European Council Directive 1999/30/EC [16]. a:
Annual mean 24-hour values revealed; b: guide value is referred to the mean
of 24 h.
101 P. Avino et al. / Microchemical Journal 88 (2008) 97106
elements determined in lichens used as biomonitoring species
for heavy metals pollution (the lichen system acts like a
sponge accumulating the pollutants), the comparison could
be useful for understanding the decreasing trend during these
30 years. In comparison with the other European cities it
should be noted that the pollutant levels determined in
Table 5
Correlation coefficients for the concentration levels of analyzed elements present in the PM10 samples
As Au Ba Br Ce Co Cr Cs Fe Hg La Mo Sb Sc Se Sm Th W Zn
0.28 0.76 0.83 0.63 0.87 0.55 0.64 0.83 0.15 0.73 0.67 0.81 0.78 0.72 0.75 0.70 0.02 0.45 As
0.28 0.60 0.47 0.50 0.25 0.51 0.53 0.23 0.55 0.47 0.60 0.39 0.57 0.56 0.51 0.00 0.25 Au
0.77 0.55 0.80 0.53 0.75 0.67 0.35 0.75 0.62 0.81 0.68 0.68 0.73 0.60 0.03 0.31 Ba
0.78 0.91 0.57 0.82 0.84 0.09 0.92 0.67 0.97 0.85 0.87 0.90 0.83 0.07 0.64 Br
0.70 0.73 0.56 0.78 0.10 0.88 0.50 0.81 0.80 0.65 0.82 0.90 0.05 0.63 Ce
0.48 0.78 0.86 0.03 0.83 0.58 0.89 0.80 0.76 0.85 0.81 0.11 0.57 Co
0.24 0.59 0.15 0.64 0.67 0.63 0.59 0.71 0.62 0.62 0.08 0.55 Cr
0.70 0.08 0.84 0.52 0.83 0.77 0.65 0.85 0.73 0.14 0.38 Cs
0.02 0.85 0.56 0.85 0.90 0.69 0.88 0.84 0.04 0.52 Fe
0.08 0.23 0.01 0.08 0.03 0.11 0.03 0.21 0.45 Hg
0.64 0.94 0.91 0.79 0.98 0.94 0.01 0.60 La
0.73 0.55 0.70 0.64 0.59 0.09 0.40 Mo
0.86 0.86 0.94 0.86 0.03 0.65 Sb
0.72 0.92 0.85 0.11 0.64 Sc
0.78 0.65 0.12 0.61 Se
0.93 0.05 0.60 Sm
0.03 0.59 Th
0.02 W
Zn
Table 6
Levels (ng/m
3
) of elements determined by INAA or other analytical techniques in different Italian and European region in various sampling years
Element Rome Rome
[20]
(ppm)
Rome
[19]
Rome
[18]
Rome
[19]
(rural)
Lecce
[21]
(rural)
Milan
[22]
Zurich
[23]
Stuttgart
[24]
Bucharest
[24]
Greece
[25]
Israel
[26]
Turkey
[26]
India
[14]
Artic
[27]
Antartic
[28]
Year 2000 8992 6578 9901 1979 2004 1987 9899 9394 9394 8990 9901 9901
Ag 0.176 0.069 0.25 0.25
As 1.35 4.0 4 3.03 0.44 1.5 12.3 4.6 14
Au 0.008
Ba 12.8 36.1 30 30 27.4 5.2 16.2 330
Br 22.2 19 70 70 24.3 4.9 16.2 120.4 6700 0.63
Ca 1500 1200 1200 b4000 242 510 1050 55 0.5
Cd 0.526 0.75 0.616 0.75 0.8 10 0.23 0.29 0.23 0.37
Ce 0.843 1.33 2 2 3900
Co 0.379 0.21 0.5 0.5 0.3 0.17 0.53 1.1 0.001
Cr 7.28 16 4.05 16 27.4 7.2 1.6 8.5 5.1 3.7 7.9 0.005
Cs 0.151 0.22 0.6 0.6 0.14 2.3
Eu 0.012 0.021 0.03 0.03 0.02
Fe 566 582 1100 1293 1100 582 695 474 240 920 1002 1351 725 0.84
Hf 0.020 0.41 0.13 0.13 0.08
Hg 1.07 0.092 0.13 0.13 0.09 5
K 1100 10700 320 320 367 219 204 726 0.3
La 0.188 2 2 0.42 0.10 1800 0.0005
Mg 240 2310 560 560 1496 49 120 160 1
Mn 40.0 72.9 35 17.7 35 11.1 138 7.3 8.3 49.3 27.2 13.9 1.1 0.1
Mo 2.10 2 2 1.28 2800
Na 420 260.6 730 730 326 192 770 7.2
Nd 0.245 0.07
Ni 4.54 8 6.66 8 1.8 2.1 12.2
Pb 92 63.4 270 67.9 270 11.1 20 25 460 29.1 41.7 1600 0.63
Rb 2.19 25.4 7 7 3.3 0.50
Sb 9.22 0.78 2.0 2 2.7 1.4 2.1 10.7 3.7 14 0.002
Sc 0.046 0.1 0.1 0.08 0.0001
Se 0.697 0.5 0.5 0.3 0.43 0.6 1.6
Sm 0.053 0.02
Th 0.204 0.34 0.9 0.9 0.07
V 4.02 2.37 14 3.76 14 9 25.8 1.0 1.8 23.2 30 0.65 0.0001
W 1.25
Yb 0.015 0.05
Zn 80.0 16.2 190 59.2 190 28 35.5 403 203 28.8 19.6 15 0.03
n.d.: Not detected; : determined by ICP-AES.
102 P. Avino et al. / Microchemical Journal 88 (2008) 97106
German city, Stuttgart, are below those found in Rome
whereas the others are above them especially in east regions
(i.e., Greece, Romania and Turkey) due probably to both
different meteo and horographic conditions and local
sources. Further, it should be underlined the very high
values found in India in an industrial site just close to a
refinery and, for opposite considerations, the low element
values investigated in Artic and Antarctic areas which can
be considered as background levels of the two hemispheres,
respectively. Finally, it is interesting the comparison with
Milan area: the total PM10 composition was performed in
this study [22] and for this reason it has been involved in
this comparison. As it can be seen, there is good agreement
between Rome and Milan data except for As and V whose
levels are comparable with those found during the '70 in
Rome.
3.4. Element sources in PM10
For quantifying the element sources in the PM10 samples,
the possible local earth crust contribution was investigated
taking more information as possible from soil samples taken
in the suburban areas of Rome, exactly at the Casaccia
Research Center and Ladispoli (a rural site and a sea town
located far few kilometers North from Rome, respectively).
These samples are representative of two geological typolo-
gies predominating in the Rome area [22]: volcanic
formation developed over pyroclastic formations and qua-
ternary formations characterized by beaches, respectively. In
Table 7 are reported the element levels determined by INAA
and ICP-AES for the two soil typologies (Casaccia and
Ladispoli). Using the Factor Analysis on all data (Fig. 3), it
can be possible to extrapolate some final associations and
considerations such as: between Na and Cl, elements of
marine origin; among Al, Ti, Sc, Ta, Fe, Zr, Eu, Mg, Ba, Rb,
Cs, La, Th, Ce, Tb and U, elements of earth crust origin,
and between MgCa and AlTi; among V, Zr, Zn and Ni,
elements of anthropogenic origin to which it necessary to
include Se, Br, Pb, Pb, Mo, As, Ag and Sb.
In addition, a specific study for the evaluation of the
origin of the trace elements through enrichment factors
(EFs) was carried out. The EFs have been calculated in
accordance to the equation reported in ref. [29]: in our
case, the ambient consists of airborne PM10 samples while
the background consists of soil samples collected outside
Rome and, in particular, in the close suburban areas of
Rome, Casaccia and Ladispoli (the soil collection was
performed in two rural areas and regarded both the upper
strata and the 20-cm down it). Elements with EF values
significantly higher than 1 can be considered not originat-
ing from local soil background and may be attributed to
long-range transport phenomena from other natural sources
and/or anthropogenic sources, or, in some cases, to a
possible preferential uptake [29]. For this purpose either
scandium or lanthanum [29] can be independently used as
a normalizing trace element [30]. The results obtained on
the soil samples collected at the two locations using both
the normalizing agents (La and Sc) are reported in Fig. 4.
The EFs calculated by the two ways allow interesting
considerations. The soil samples collected in Ladispoli are
constituted by only crustal natural origin elements and their
respective EF
La
and EF
Sc
comparisons are quite similar.
For the Casaccia soil samples the situation is different
because the comparison between EF
La
and EF
Sc
is almost
constantly dissimilar; however, also in this case all the
elements can be considered of natural origin (few doubts
for Sm, Th, Br, Ba, Cs, Pb, Sb, La, As even if their EFs
are very low, max 3.7). The discrepancy between the two
EFs may be attributed to two different causes: no-
homogeneity of the soil samples collected at Casaccia
Center and/or difference of Sc and La levels between the
reference values [30] and local situation (volcanic forma-
tion developed over pyroclastic formations). After, we have
investigated the airborne PM10 samples collected down-
town Rome and we have compared these EFs with those
found by one of the authors in previous decade: all the EF
data are given on the basis of the lanthanum concentra-
tions. In Fig. 5 are reported the EFs of elements analyzed
on airborne PM10 filters. As can be noted, all the elements
may be attributed to long-range transport phenomena from
other natural and/or anthropogenic sources: this behavior is
common to the three decades even if a very light
decreasing trend can be evidenced from 1970 to 2000.
Table 7
Levels (g/g) of elements in soil sampled in two suburban areas of Rome,
Casaccia Research Center and Ladispoli
Element Casaccia Ladispoli
As 63 38
Ba 1540 600
Br 9.2 5.5
Ca 6700 660
Cd 0.22 0.15
Ce 50 24
Co 24 22
Cr 82 110
Cs 45 11
Eu 3 2
Fe 62,000 45,000
Hf 13 16
K 23,000 20,000
La 180 77
Mg 6300 3500
Mn 1300 1600
Mo 0.70 0.65
Na 4040 8900
Ni 45 38
Pb 110 33
Rb 240 206
Sb 7.5 2.8
Sc 16.9 17.9
Sm 23 13
Th 69 24
V 180 150
Yb 4 4
Zn 210 170
: Determined by ICP-AES.
103 P. Avino et al. / Microchemical Journal 88 (2008) 97106
Looking at Fig. 5, two element groups can be identified
according to their EFs: La, Hf, Sm, Sc, Th, Cs, Eu, Ba, Fe,
Rb, Co, Ce, As, V, Mn, Mg, K, Cr, Ni and Na with EF
ranging between 5 and 100 and Hg, Ca, Zn, Se, Pb, Sb,
Cd, Br and Mo with EFs ranging between 200 and 3000.
These EF values determined using Casaccia concentrations
are quite similar to those found using Ladispoli concentra-
tions. Finally, some particular considerations can be
extrapolated: as reported in other paper [25], the high EF
values found for Br by both the elaborations could be
attributed to the use of leaded gasoline (in 2000, there
were two fuels, unleaded and leaded, available in the
Italian market); the sources of As, Pb, Sb and Zn would be
looked for among the various anthropogenic activities in
the Rome area and particularly Sb and Zn could be of
traffic origin being essential components of anti-friction
alloys and car tires [19,20,25].
Finally, we have tried to define the origin of each
element: in Table 8 is reported a master scheme of this
attribution on the basis of both the experimental results
obtained and theoretical consideration. Of course, the lack of
a large data bank do not allow definitive considerations but
the whole methodology involved is valid and is useful to
understand contribution of local anthropogenic sources to the
air pollution.
4. Conclusion
The INAA technique has allowed to reach such elevated
sensibility/accuracy levels to furnish discreet values for
elements present at trace and ultra-trace levels. Particular
attention has been devoted to reach elevated degrees of
precision and accuracy for each element.
The results obtained and the relative implications for human
health can obviously be referred only to the element
investigated in PM10: the concentrations found do not show a
significant level of attention from a toxicological point of view.
The methodological approach is very interesting and may
be used for evaluating the impact of new technological
activities on both a territory and air quality in relation to the
various critical steps (emissions, pollutant fall-out, water
contamination, food chain, etc.). In particular, this approach is
the preliminary study of a project (funded by Italian Ministry
of Health) based on a deep investigation on PM10 and PM2.5
in Rome urban area: in fact, the sampling will be performed in
different sites and after a 3-D matrix (considering data,
Fig. 3. Vertical diagram relative to the application of Factor Analysis to the Casaccia soil data.
104 P. Avino et al. / Microchemical Journal 88 (2008) 97106
Fig. 5. Comparison of trace element enrichment factors in airborne PM10 samples calculated using lanthanum as normalizing elements and Casaccia (above) and
Ladispoli (below) data as reference values.
Fig. 4. Comparison of trace element enrichment factors in soil samples collected at Casaccia (above) and Ladispoli (below) locations calculated using lanthanum and
scandium as normalizing elements.
105 P. Avino et al. / Microchemical Journal 88 (2008) 97106
location and date) or 4-D matrix will be built according to the
literature [31,32].
Acknowledgments
This study was performed under the grant ISPESL/DIPIA/
P14, 20052007. The authors would be very grateful to the
referees for their useful suggestions.
References
[1] X. Querol, A. Alastuey, A. Lpez-Soler, A. Boix, T. Sanfeliu, V.V. Martynov,
P.I. Piven, L.P. Kabina, P.A. Souschov, Trace element contents in atmospheric
suspended particles: inferences frominstrumental neutronactivation analysis,
Fresenius' Journal of Analytical Chemistry 357 (1997) 934940.
[2] K. Beijer, A. Jernelv, Sources, transport and transformation of metals in
the environment, in: L. Frindberg, G.F. Nordberg, V.B. Vouk (Eds.),
Handbook on the Toxicology of Metals, Vol. 1, 2nd ed. Elsevier Scientific
Publ., Amsterdam, ISBN: 0444904131, 1986, pp. 6874.
[3] R.J. Lantzy, F.T. Mackenzie, Atmospheric trace metals: global cycles and
assessment of man impact, Geochimica and Cosmochimica Acta 43 (1974)
511525.
[4] R.M. Harrison, J. Yin, Particulate matter in the atmosphere: which particle
properties are important for its effects on health? The Science of Total
Environment 249 (2000) 85101.
[5] P. Avino, D. Brocco, L. Lepore, S. Pareti, Interpretation of atmospheric
pollution phenomena in relationship with the vertical atmospheric
remixing by means of natural radioactivity measurements (Radon) of
particulate matter, Annali di Chimica 93 (2003) 589594.
[6] R. Djingova, S. Arpadjan, I. Kuleff, INAA and flame AAS of various
vegetable reference materials, Fresenius' Journal of Analytical Chemistry
339 (1991) 181186.
[7] IAEA, Sampling and analytical methodologies for instrumental neutron
activation analysis of airborne particulate matter, Training Course Series
No.4, IAEA, Vienna, 1992.
[8] P. Avino, P.L. Carconi, L. Lepore, A. Moauro, Nutritional and environmental
properties of algal products used in healthy diet by INAA and ICP-AES,
Journal of Radioanalytical and Nuclear Chemistry 244 (2000) 247252.
[9] W. Tian, Reactor neutron activation analysis of airborne particulate matter,
NAHRES-53, IAEA, Vienna, 2000.
[10] G. Guzzi, A. Colombo, F. Girardi, R. Pietra, G. Rossi, N. Toussaint,
Comparison of various analytical techniques for homogeneity test of
candidate standard reference materials, Journal of Radioanalytical and
Nuclear Chemistry 39 (1977) 263276.
[11] A. Moauro, P.L. Carconi, A European intercomparison of vegetal standard
reference materials, based on INAAand some non nuclear spectrochemical
techniques, Journal of Radioanalytical and Nuclear Chemistry 151 (1991)
149157.
[12] L.A. Currie, Limits for qualitative detection and quantitative determination.
Application to radiochemistry, Analytical Chemistry 40 (1968) 586593.
[13] J. Tolgyessy, E.H. Klehr, Nuclear Environmental Chemical Analysis, Ellis
Harwood Ltd, West Sussex, UK0470208341, 1987.
[14] N.L. Chutke, M.N. Ambulkar, A.L. Aggarwal, A.N. Garg, Instrumental
Neutron Activation Analysis of ambient air dust particulates from
metropolitan cities in India, Environmental Pollution 85 (1994) 6776.
[15] World Health Organization (WHO), Air quality guidelines for Europe, 2nd
ed. ISBN 92-890-1358-3, 2000 (WHO regional publications, European
series, n. 91).
[16] European Council Directive 1999/30/EC, Limit values for sulphur dioxide,
nitrogen dioxide and oxides of nitrogen, particulate matter and lead in
ambient air, 1999 22 April 1999.
[17] F. D'Innocenzio, P. Di Filippo, L. Lepore, A. Marconi, PM
10
and PM
2.5
concentrations in urban air and size fraction distribution of toxic metals,
Annali di Chimica 88 (1998) 281289.
[18] G. Ziemacki, G. Cattani, M.C. Cusano, G. Stacchini, A. Marconi,
Contenuto di metalli nelle varie frazioni dimensionali di materiale
particellare, Annali dell'Istituto Superiore di Sanit 39 (2003) 371379.
[19] G. Capannesi, G. Mastinu, Trace elements in air particulates from 1965 to
1978, CNEN Report RT/AMB (79)10, 1978, pp. 127.
[20] F. Bruno, G. Capannesi, L. Gratani, F. Manes, Characterization of mineral
content in Quercus ilex leaves by photon and neutron activation analysis
(Castelporziano, Rome), Giornale Botanico Italiano 114 (1980) 175186.
[21] A. Buccolieri, G. Buccolieri, A. Dell'Atti, M.R. Perrone, A. Turione,
Natural sources and heavy metals, Annali di Chimica 96 (2006) 167182.
[22] G. Mastino, L. Testa, I. Michetti, E. Ghiara, Elementi in traccia nel
particolato atmosferico in Italia, Acqua-Aria 1 (1987) 1733.
[23] C. Hueglin, R. Gehrig, U. Baltensperger, M. Gysel, C. Monn, H. Vonmont,
Chemical characterisation of PM2.5, PM10 and coarse particles at urban,
near-city and rural sites in Switzerland, Atmospheric Environment 39 (2005)
637651.
[24] V. Cercasov, A. Pantelic, M. Slgean, H. Schreiber, Comparative
evaluation of some pollutants in the airborne particulate matter in Eastern
and Western Europe: two-city study, BucharestStuttgart, Environmental
Pollution 101 (1998) 331337.
[25] P. Misaelides, C. Samara, F. Noli, Th. Kouimtzis, I. Anou, Toxic element
concentrations in airborne particulate matter in the area of Thessaloniki,
Greece, The Science of Total Environment 130 (1993) 139146.
[26] M. Koak, M. Nimmob, N. Kubilay, B. Herut, Spatio-temporal aerosol
trace metal concentrations and sources in the Levantine Basin of the
Eastern Mediterranean, Atmospheric Environment 38 (2004) 21332144.
[27] K.A. Rahn, R.J. McCaffrery, Compositional differences between Artic
aerosol and snow, Nature 280 (1979) 479480.
[28] C. Boutron, C. Lorious, Trace metals in Antarctic snow since 1914, Nature
277 (1979) 551554.
[29] L. Bergamaschi, E. Rizzio, M.G. Valcuvia, G. Verza, A. Profumo, M.
Gallorini, Determination of trace elements and evaluation of their
enrichment factors in Himalayan lichens, Environmental Pollution 120
(2002) 137144.
[30] B. Mason, C.B. Moore, Principles of Geochemistry, 4th ed., John Wiley,
New York, ISBN: 04-71575224, 1982, pp. 4647.
[31] F. Gong, B.-T. Wang, Y.-S. Fung, F.-T. Chau, Chemometric characteriza-
tion of the quality of the atmospheric environment in Hong Kong,
Atmospheric Environment 39 (2005) 63886397.
[32] I. Stanimirova, V. Simeonov, Modeling of environmental four-way data
from air quality control, Chemometrics and Intelligent Laboratory Systems
77 (2005) 115121.
Table 8
Master scheme of element origin attribution
Element Group
Na, Nd, Ca Normal global crustal contribution
Insufficient local crustal contribution
Sm, V, W Normal global crustal contribution
Very insufficient local crustal contribution
Cr, Mn, Br Insufficient global crustal contribution
Elevated local non-crustal contribution
Zn, As, Cd, Sb, Pb Elevated global anthropogenic contribution
Elevated local anthropogenic contribution
Se, Br Insufficient global anthropogenic contribution
Elevated local anthropogenic contribution
Sc, Cs, Ba, La, Ce, Th Insufficient global crustal contribution
Elevated local crustal contribution
Hf, K, Fe, Co, Ni, Rb, Eu, Mg,
Yb
Normal global crustal contribution
Normal local crustal contribution
Mg, Na Insufficient global crustal contribution
Elevated marine contribution
Mo, Ag Insufficient global crustal contribution
Normal local crustal contribution
Hg Insufficient global crustal-anthropogenic
contribution
Normal local crustal-anthropogenic
contribution
106 P. Avino et al. / Microchemical Journal 88 (2008) 97106

You might also like