A. Study area

The city of Islamabad is located near the Margallah Hills at the northern edge of the Potohar Plateau supporting natural terraces and meadows across the whole area. Streams flow down from Margallah Hills and ranges of the Murree Hills, and converge with a profusion of small streams like Soan and Kurang tributaries passing through the city. The growth of the urban area is taking place at a high pace leading to the deterioration of the vegetation of the area. Islamabad is situated at latitude 33.72N and longitude 73.07E, and at altitudes of 457–610 m (asl). The area is in a semi-arid region with mild summers and winters and with rugged topography varying in elevation, mainly comprising steep slopes and gullies. The natural soils of the area come from wind and water laid deposits, and sedimentary rocks. The soils from wind deposits vary from dark brown to yellowish brown in color and the subsoil is usually calcified or calcareous silt loam [13]. Because of recent accelerated urbanization and developmental activities, adverse effects of discharging untreated sewage wastes and dumping of solid waste on the soil have been observed [14].

B. Sampling

For this study, sector I-9, being the industrial areas of Islamabad, was selected to find the extent of contamination of toxic metals in roadside dust, which might affect the health of the inhabitants of these areas. A total of 95 sampling sites were selected from sector I-9. Because of the rapid changes in soil use in the urban area, the identified sites were not further divided into different categories. The dust samples were collected from both sides of the road using a spatula and a small brush, and then mixed thoroughly to obtain a representative sample for each site. All the collected samples were kept in sealed polyethylene (PE) bags to avoid contamination and thereafter the samples were taken to the laboratory for preservation.

C. Sample preparation

The dust samples were dried at room temperature and sieved using a 2-mm sieve to remove unwanted particles like stones and debris. The dried samples were then mixed thoroughly and about 1 g of the homogenized sample was grounded into powder. From each homogenized sample about 180 mg were pelletized, using a press and a 13 mm die, to obtain 1 mm thick pellets.

D. PIXE analysis

The pelletized samples were irradiated with 3.0 MeV proton beam from the 5 MV Pelletron Tandem accelerator installed at Experimental Physics Lab, National Centre for Physics, Islamabad. The calibration and standardization of the newly installed PIXE chamber for both thick (TiV/Fe alloy) and thin (SRM 2783) targets was carried out using NIST standard reference materials. The diameter of the collimated proton beam was 2 mm. A 100 μm thick Mylar, "funny" filter was used during the measurements and this reduced the count-rate to ensure a dead time of less than 10% at beam currents of 2–5 nA. Observation of the samples after irradiation showed no apparent damage. The emitted X-rays were detected using a 30 mm² Si(Li) detector with an energy resolution of 138 eV (FWHM) at 5.9 keV of Mn. The PIXE data was analyzed using the computer code GUPIXWIN.

A. Total contents of metals

The concentration levels of the elements found in roadside dust in Islamabad are as shown in Fig. 1. The concentrations with respect to sampling locations for Zn and Cu are shown in Fig. 2. It is evident that the concentration of metals varies with distance from source.

Fig. 1.Full-screen View

(Color online) Mean concentration of elements in the roadside dust of Islamabad.

Fig. 2.Full-screen View

Concentration of Zn and Cu in roadside dust samples collected at different locations in sector I-9 in Islamabad.

The elements determined include Si, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, As, Se and Cd, and elemental levels in the dust samples were ranked as Fe >Ca >Co >Sc >K >Ti >Mn >Zn >Pb>Cu >Ni >Cr >V >Si >S >As >Ga >Cl >Se >Cd. The soil background values of Islamabad were used as references [15]. The mean concentration of measured heavy metals, except Cd, greatly exceeded the reference values. The levels of Cu, Ni, Pb and Zn were over 3 times greater than the reference values. Worldwide concentrations of these elements in road dust as found in literatures are given in Table 1. Comparatively, metal concentrations for different studies show different metal diversity and variable magnitude of concentration. In this study, the extent of pollution is significant for several metals at different sites based on the absolute metal content. Compared with metal contents in other cities such as Ottawa (Canada) (Table 1), the mean value of Cd, Cu, Ni, Pb and Zn in the analyzed roadside dust is much higher.

B. Assessment of pollution

Atmospheric particles mainly come from low-temperature crustal weathering, soil remobilization (crustal source), and a variety of anthropogenic sources at high temperatures. Enrichment factors (EF) are used to prove the chemical nature and principal sources of trace elements in aerosols by incorporating reference elements, and evaluate the strength of the crustal and non-crustal sources using Eq. (1), based on the Earth’s crustal mean abundance of the metals given by Wedepohl in 1995 [29], which are indicators of a specific source. Ti is used as a crustal indicator element.

where EFx is the enrichment factor for element x, C_xp and C_Tip are the concentrations of x and Ti in aerosol, respectively, and C_xc and C_Tic are their concentrations in an average crustal material.

When EFx approaches unity, crustal soils become the predominant source for element x. The EF values of trace elements in sources are expected to be greater than 1.0, save for crustal materials. However, since crustal material types and soils originate from different areas, and little is known about uncertainties of fractionation during weathering, the EF values are better resolvable within 10 or a little larger for crustal materials (Fig. 3). The EF of elements, such as Cl, V, Mn, Fe, Ca, and K are close to unity or less than 10, suggesting that they are attributable mainly to crustal sources. The high enrichment values for Cr, Cu, Pb, Zn, As, Se, Cd, Ni, Co and S suggest that their dominant sources are non-crustal and various pollutant emissions may contribute to their loading. Higher EF values of elements show heavy metal contamination level in the roadside dust. This enrichment analysis confirms that the trace elements come from a mixture of crustal particles and anthropogenic sources in the roadside dust of Islamabad. The EF values of Cu, Pb and Ni were 60.22, 84.2 and 22.4, respectively.

Fig. 3.Full-screen View

(Color online) Enrichment factor of elements in roadside dust of Islamabad.

Copper is an essential trace element which is widely distributed in the environment and it occurs naturally in elemental form and as a component of many minerals. It is present in flora and fauna, and in various foods and beverages, including drinking water. Severe inhalation exposure to Cu dust or fumes at concentrations of 0.075–0.12 mg/m³ may cause metal fume fever with symptoms such as cough, chills and muscle ache [30]. The recommended safe and adequate dietary intake for Cu is 1.5–3.0 mg per day for adults, 0.7–2.5 mg per day for children and adolescents, and 0.4–0.7 mg per day for infants [30]. The findings of this study regarding copper were: The average concentration of Cu in the roadside dust samples from Islamabad I-9 sector, as given in Table. 1, is 100.2 μg/g. This value is greater than 25 μg/g, which is the Cu average concentration in the soil. The variation in Cu concentration with sampling location observed in this study is shown in Fig. 2. The global concentration of Cu as reported in Table 1 varies from 25 μg/g to 1075 μg/g. From the 19 places in Table 1, 8 places have higher and 11 places have lower Cu concentrations than that found in Islamabad.

Nickel is found in the earth’s crust in combination with other elements. It is present in all soils, and in the environment where it is in combination with oxygen or sulfur as oxides or sulfides. Nickel is also released into the atmosphere through burning coal and oil in power generator plants. Health hazards from Ni exposure in the occupational environment are due primarily from inhalation. About 10% of women and 2% of men in the world are extremely sensitive to Ni exposure. The most common type of reaction to Ni exposure is skin rash called nickel dermatitis through direct contact with Ni [31]. The following results were obtained in the present study. The concentration of Ni as given in Table. 1 is 58.3 μg/g. In the soil samples, the average concentration of Ni was found to be 32 μg/g. The highest Ni concentration, 347.6 μg/g, was found in sample No.55. From these results it would appear that there is a higher Ni contribution from sediments and soils in roadside dust and also from traffic emissions. The amount of Ni measured at all sampling points along Islamabad I-9 sector was two times higher than the reference value [15]. The concentration of Ni measured at various places around the world is given in Table 1, and it varies from 4.2 μg/g to 177 μg/g. In Table 1, there are 4 places where the level of Ni is greater and 13 places where it is lower than that found in Islamabad.

Zinc is an essential trace element which is present in soil. It can cause deficiency symptoms and can be toxic when intake and exposures exceed physiological needs. The primary source of Zn exposure to people is from food, though oral exposure can become excessive through non-dietary sources. Certain occupational exposures can be hazardous. High Zn levels may disrupt the homeostasis of other essential elements. The observations made from Zn results in this study were as follows: In Table. 1 it can be seen that the average Zn concentration found in roadside dust samples was 162.6 μg/g, and this is higher than the average Zn concentration (91 μg/g) in soil. The global concentration of Zn in Table 1 is in the range 16.6–840 μg/g. In Table 1, there are 13 places where the Zn concentration is greater and 5 places where it is lower than the 162 μg/g found in the roadside dust of Islamabad.

Lead is ubiquitous in industrialized societies, and evidence of the negative effects of Pb on humans has been noted for centuries. Lead poses a unique threat to developing minds and learning capacities of young children, who can be exposed from eating Pb-based paint chips or playing in contaminated soils. Pb can also damage the nervous system, kidneys, and reproductive system. The observations made from Pb results in this study were as follows: From Table 1, it can be seen that the average Pb concentration was 128 μg/g in the roadside dust samples and 62.5 μg/g in the soil samples. The highest Pb concentration in the dust samples was about twice the average Pb concentration found in the soil [15]. The high Pb concentration could have been due to traffic burden, brick kilns, and usage of leaded gasoline, which was phased out in Pakistan in 2005. Furthermore, most of the light vehicles on this highway use CNG (compressed natural gas), therefore the concentration of Pb is somewhat under control and should decrease with the passage of time. From the compiled data given in Table 1, it is evident that there is a big variation in Pb concentration from city to city; the concentration varies from 11.2 μg/g to 1927 μg/g. There are 7 places where Pb concentration is greater, 10 places where concentration is lower and one place (Beijing China) where the concentration is comparable to that in the dust of Islamabad industrial area.

C. Correlation of metal levels in roadside dust samples

In order to establish inter-elemental relationships in the roadside dust samples, principal component analysis was applied to the whole elemental concentration data set. Principal component analysis is an important method for qualitative source distribution in the study of aerosols [32]. Its purpose is to establish the minimum factors which can explain the main variance of the system. Statistical analysis software SSPS10.0 was used for rotated component matrix factor analysis of element mass concentration in the dust samples (Table 2). The first, second and third factors are 80.8, 10.2 and 1.5%, respectively. It is sufficient to retain the three factors to explain 92.5% of the total variance. The first factor has high loading for Ca, Sc, Ti, V, Mn and Fe suggesting that these elements come from geological resources [33]. The second factor is related to Cr, Cu, Ni, Pb, Zn and Ga, indicating an industrial or traffic source, probably comprising different sources from industrial emissions [15, 34], while the third factor shows high loading of S, Cl, K, As and Se, suggesting that they may have originated from oil/coal combustion [35].