ASSESSMENT OF TRACE ELEMENTS BIOAVAILABILITY – INGESTION OF TOXIC ELEMENTS FROM THE ATTIC DUST COLLECTED FROM THE VICINITY OF THE FERRONICKEL SMELTER PLANT

Research Center for Environment and Materials, Macedonian Academy of Sciences and Arts, 1000 Skopje, Republic of Macedonia Institute of Chemistry, Faculty of Natural Sciences and Mathematics, Ss. Cyril and Methodius University, POB 162, 1000 Skopje, Republic of Macedonia Faculty of Chemistry and Pharmacy, "St. Kliment Ohridski" University of Sofia, Bulgaria Corresponding author, e-mail: kbaceva@manu.edu.mk


INTRODUCTION
The increased industrial activities lead to a rising level of urban environmental pollution.Traffic emissions and waste from municipal activities are major anthropogenic problems and health risks in recent years.Rapid increases of toxic metal concentrations in the atmosphere and environment are commonly coupled to the development of exploitative technologies.The contribution of the human factor to this problem is very broad and complex, which leads to feedback processes.Atmospheric deposition of toxic metals is the main subject of many studies and usually takes place in industrialized areas, where exploitation and processing of natural resources (ore, oil, etc.) are performed [1].Atmospheric deposition poses significant ecological concerns.According to the process of deposition particles can be deposited at the earth's surface in two ways, dry or wet, depending on the phase in which a species strikes the earth's surface and is taken up [2].
The emission of toxic metals into the atmosphere is one of the greatest threats to human health.People are directly exposed to the effects of toxic metals through inhalation of airborne microparticles from atmospheric dust [3].Depth of penetration and deposition of particles depends on their size, the defense capabilities of the respiratory tract and the respiratory pattern [4] which finally also determine the effect to the human health.Chemical elements emitted in the atmosphere by combustion processes usually have relatively high solubility and reactivity; especially under low pH condition [5].They are highly mobile and might be carried far away from the sources by wind, depending upon whether they are in gaseous form or as particulates.The samples of dust from attic beams are very useful for continuous and long term monitoring of the content of toxic metals and other toxic components in areas with high industrial activities [6][7] ensuring reliable assessment of the degree of pollution.The term dust usually comprises street dust and house dust [8][9][10][11][12].However, other types have also been studied in the past.A particular type of house dustthe attic dust is studied in this work.It represents dust deposited in the attics, with reduced (minimized) influence of the householder.The attic dust is derived predominantly from external sources such as aerosol deposits and as a result of soil dusting, and less from household activities [13].The main advantage of attic dust as sampling material is that its composition remains constant, i.e. chemically unchanged with time.Investigations on attic dust composition and chemistry therefore reveal the average of the historical state of the atmosphere [13][14][15][16][17].
Chemical elements from dust can entered in human bodies via ingestion, dermal contact, or breathing, especially dangerous for children due to the "hand to mouth" activity during outdoor activities in playground and recreational areas [18].Low tolerance to toxins and easy ingestion of dust through hand-to mouth pathways are serious health risk hazards that could not be ignored for children.The ingestion and inhalation of dust appears to be the main exposure pathway to As, Pb, Cd, Cr and Cu, which results in high health risk for people living in industrial zones.The most exposed group in the population to air particles are children, the particles penetrate deeper in their lungs as compared to adults because children breathe deeper and faster and they are physically more active.Ingested metals can accumulate in the fatty tissues, affect the nervous system during the period of maximal brain growth [19][20], disrupt the normal functioning of internal organs and act as cofactors for other diseases [19].Senior citizens, especially those with a weakened cardiovascular and respiratory system are a high risk group, too.Another risk group comprises patients with chronic pulmonary emphysema, asthma or cardiovascular diseases [21].Therefore the concentration of metals in house dust is a matter of high concern, together with their bioaccessibility, e.g. the fraction most readily absorbed by humans.
Reliable risk assessment through thorough understanding of the nature and extent of the contamination of house dust with metals [19] is an actual problem from both analytical and environmental point of view.Bioaccessibility (a proxy for bioavailability, which is the proportion of the readily absorbed metal in the sample) is another aspect which is very important in modern health risk assessments [22].
Bioaccessibility has the potential to make a significant impact on current risk assessment and comparing and evaluating the physico-chemical processes within the many models and systems that have been developed over the years to measure bioaccessibility and contaminant exposure [23][24][25].
The significant emission sources that contribute to atmospheric pollution with chemical elements on the territory of the Republic of Macedonia, are predominantly mines and smelter plants [26].The subject of this study is to investigate the degree of air pollution in the town of Kavadarci and its surrounding through analysis of attic dust samples.Total elements content and their bioavailable fractions (obtained after extraction with different media: phosphate buffer, hydrochloric acid solution and carbonate buffer, simulating human buffers) are determined in attic dust samples and used to evaluate the exposure of the population to toxic elements.In our previous studies it was shown that the most important source of deposition of the toxic elements is the ferro-nickel smelter plant situated in this region, responsible for the presence of high content of Ni, as well as of Co, Cr and some other chemical elements in the soil and dust in this area [15,17,[27][28].

Study area
Results obtained from moss biomonitoring study showed that the appearance of some metals (Co, Cr, Fe and Ni) in the air in the south-central part of the Republic of Macedonia (Figure 1) is related to the activities of the ferro-nickel smelter plant situated near the town of Kavadarci [15,[29][30][31].
The population in the city of Kavadarci and its environs (about 60,000) is engaged mainly in agriculture (cultivation of vegetables, grapes and wine production).In 1982, a ferro-nickel production plant started the production of ferro-nickel with an annually processing of about 1.5 million tons of ore.
The oldest formations dominate in the investigated area is and they have direction NW-SE and belong to the inner parts of the Vardar zone.The Lower Paleozoic metamorphic complex is present with two series: amphibole and amphibolechlorite schists with marbles and phylite layers [27].

Sampling and samples preparation
Samples of attic dust were collected in the period of October-December 2008 at 10 settlements in the Kavadarci region.In every settlement, attic dust was collected from two or three houses on different sites.The selection of the locations for the sampling area covered the town of Kavadarci, the surrounding area of ferro-nickel smelter plant and the wider environment around 300 km 2 and taking into account the direction of the winds and the possible influence of meteorological conditions.The collection of attic dust samples was performed according to the adopted protocol [13,[15][16][17].Close to each sample location an old house was chosen with intact attic carpentry.Although some of the selected houses were older than the ferro-nickel smelter plant, most of them were as old as the plant itself.To avoid collecting particles of tiles, wood and other construction materials, the attic dust samples were brushed from parts of wooden constructions that were not in immediate contact with roof tiles or floors [14].
From preliminary studies performed for the same samples of attic dust from Kavadarci and its surroundings it was determined: a) The total content of chemical elements in the attic dust samples collected from the investigated area [15]; b) Total mineralogical composition of the samples of attic dust [32].
The bioavailability of toxic elements from the dust samples was defined by using different model solutions which were applied to determine leachable and mobile metal concentrations.Four methods were applied for the evaluation of trace elements bioavailability and mobility: (1) extraction with 0.1 mol l -1 HCl for 1 h, then suspension was filtered through an acid-resistant filter; (2) extraction with H2O for 1 h then suspension was filtered through membrane (cellulose) filter; (3) extraction of trace elements in a mixed phosphate buffered solution (pH in the range 7.35-7.45)that simulate human blood for 1 h and then filtration trough membrane filter; and (4) extraction of trace elements in a mixed carbonate buffered solution that simulate human lungs and then filtration trough membrane filter.
Optimal instrumental and operating conditions are presented in Table 1.Certified reference materials were used to validate the method for all considered elements.Results obtained showed very good agreement, the difference between measured and certified values was within 15%.Standard soil reference material (JSAC 0401) was also used to ensure quality control.The measured concentrations agree very well with the recommended values.
The theoretical limit for ICP-AES methods is in the ppm (µg ml -1 ) range for the majority of elements.The determination limits achieved for all studied elements are in the typical range for ICP-AES measurements, relative standard deviations varied between 3 and 12% for all measured elements.

Statistical analysis of the data
Principal component analysis (PCA) is a statistical method to classify the relations between the measured variables, i.e. it linearly compresses a set of main data to an essentially smaller set of new, uncorrelated variables which represent nearly all information in the main data set.Meanwhile, understanding and working with a small set of uncorrelated variables is much easier than working with a large set of correlated variables.In the present study, PCA was carried out in PAST 3.04 for Windows.In addition, spatial correlation analysis was applied for each variable to select the most appropriate method of interpolation.In particular the obtained results by the method of PCA showed that in the score plots significant correlation exists between the types of extraction.
In the descriptive statistics it was used the coefficient of variance (CV%) instead of Variance, to characterize the behaviors of the data.Some statistical parameters like skewness (A) and kurtosis (E) are shown in the Table 3. From the values of A and E, it is shown that the data more or less are following the normal distribution.Is it because there is a constant pollution level in all the area?

RESULTS AND DISCUSSION
The descriptive statistics of the analyzed element concentrations in the attic dust samples collected in 10 houses from the investigated area (Figure 1) are summarized.The average values for the content of all analyzed elements (Ag, Al, As, Ba, Ca, Cd, Co, Cr, Cu, Fe, K, Mg, Mn, Ni, Pb, Sr and Zn) and their minima and maxima in the attic dust samples as well as the basic statistics are presented in the Table 2 and 3.The spatial distribution of the content of some of the analyzed elements (Co, Cr, Ni and Zn) is presented in Figure 2.
It is known that in the studied area, the main source of toxic metals in the air was the dust emitted from the ferro-nickel smelter plant.Namely, the dust from this facility represents the microparticles of ore that are processed as well as the dust from the waste gases from the ferro-nickel smelter plant.The contents of most of the toxic elements are substantially higher than their natural representation in the surrounding rocks and soils [27][28].The ferro-nickel smelter plant uses ore that contains between 1% and 2.5% Ni, about 0.05% Co, 1-3% Cr etc.Therefore, we expect these elements to have significantly higher contents in samples of attic dust compared to their content in the soils from the area [27][28].
The previous studies, moss biomonitoring and attic dust investigations [15,[29][30] conducted in this area defined the anthropogenic group of the following elements: Cd, Co, Cr, Fe, I, Ni, Pb and Zn, present in the air.The distribution of these elements, especially Ni, Co and Cr, shows their increased content in moss and attic dust samples taken from polluted zones in the investigated region.From the mineralogical investigation of the same attic dust samples collected in the Kavadarci region [32] was confirmed that the mineralogical dust composition differs greatly from the typical mineralogical composition of urban dust.It was shown that the serpentinite (chrysotile, lizardite) and amphibole group of minerals (ribecite, tremolite and actinolite) were presented in the attic dust samples from this region.The fact that these minerals are present in the ore processed in the ferro-nickel smelter plant situated in this area and the high content of Ni in the analyzed attic dust samples confirms that the source of the pollution is the ferro-nickel smelter plant.
The results obtained in the current survey show that the samples which were collected around the ferro-nickel smelter plant are characterized with the highest contents of Cr, Ni and Zn.Thus, the highest content of Ni (825 mg kg -1 ) was found in the attic dust samples that were collected from the houses in the village of Šivec the nearest villages to the smelter plant (Table 2).Also, very high content of Ni was found in the houses form the village of Resava (562 mg kg -1 ), and the villages of Drenovo and Trstenik with the contents of 389 and 308 mg kg -1 , respectively.Similar, high contents were found for Cr and Mn.The contents of Cr in the houses from Šivec, Resava, Drenovo and Trstenik were 370, 321, 146 and 170 mg kg -1 , respectively.The values for the content of Pb in attic dust samples varied in the range of 136 to 321 mg kg -1 (Tables 2 and 3).For the comparison, the average contents of these elements in the surrounding soils are much lower: 74 mg kg -1 for Ni, 55 mg kg -1 for Cr, 15 mg kg -1 for Co and 21 mg kg -1 for Pb.From these results it could be concluded that the population in this region is exposed to the dust with a significantly higher contents of toxic elements as a direct impact of the activities performed in the ferronickel smelter plant.Direct exposure to fine particles of attic dust, which are easily absorbed into the human body through the respiratory system as well as uptake of food that is grown in this region, is a disturbing factor in terms of their health.To assess more realistically the effect on human health, one of the goals of this research was to establish the extractability or bioavailability of these elements from the dust.For that reason, four extraction methods were performed to investigated and see the extractability of toxic and trace elements: (1) Extraction into hydrochloric acid solution with the concentration of 0.1 mol l -1 HCl, which simulates gastric juice in the human body; (2) Extraction with water (pH = 7); (3) Extraction in a phosphate buffer solution (pH value around 7.35-7.45)that simulate human blood; and (4) Extraction of toxic elements in carbonate buffer that simulates the human lungs (pulmonary mucus).
The obtained values for the contents of the elements were statistically processed using basic de-scriptive statistics.The summarized values of median, minimal and maximal contents of all 10 samples are presented in Table 4. Data for the contents of Cr, Cu, Mn, Ni, Pb and Zn for each sample are presented in Figure 3.As can be seen the highest extractability of the toxic elements in samples of attic dust was found in a model solution 0.1 mol l -1 HCl, which was used to simulate stomach acid juice.
Heavy metals emitted by smelter plant processes usually have relatively high solubility and reactivity, because of the small sizes of particles on which they are carried.In addition, from the results the higher leachibility of heavy metals (Ni, Co, Cr, Pb) from attic dust samples was in this extraction model solution 0.1 mol l -1 HCl, compared with the other 3 extraction media.The reasons for the high-est extractability is the low pH value of the model solution 0.1 mol l -1 HCl, the other solutions have pH around 7 or 7.5.
For the attic dust samples that were extracted with the solution of 0.1 mol l -1 HCl, the highest median values were obtained for the elements: Al, Ca, Fe, K, Mg, Mn, Ni, Pb, Sr and Zn (Table 4).From these studies it can be concluded that if the dust in any way was entered into the stomach it will pose a significant risk (often by eating unwashed fruits and vegetables) for the human body.The city of Kavadarci and surrounding villages are characterized as agricultural region where the population is mostly engaged in wine culture and making wine, so there is a serious risk to the population health.If we compare the results for all the extraction procedures from each sampling location of attic dust samples with the total content of these elements, the results will be significant.Thus, the content of Ni in the attic dust samples from the city of Kavadarci was 149 mg kg -1 from which 5.7% were extracted with a solution of 0.1 mol l -1 HCl.The total contents of the other analyzed toxic elements for Cd, Cu, Mn, Pb and Zn in the attic dust from Kavadarci were 14.5, 52.4,460, 264 and 395 mg kg -1 , respectively, showing significant extractability the same simulation solution (44, 49, 55, 53 and 60%, respectively).There were no significant values for the extraction of the toxic elements in the other extraction solutions, except for the extraction with water from which 14% Cu and 17% Mn were extracted from the same attic dust samples (Table 4, Figure 3).
The village Šivec is located 1.4 km from the ferro-nickel smelter plant directly exposed to the emitted dust.The total content of Ni in the dust from this village was 825 mg kg -1 .By using the extraction media of water and carbonate buffer only about 1% Ni were extracted, while with the extractions with HCl solution 5% of Ni were extracted.For the other elements that were analyzed, again the highest content was extracted with the model solutions of HCl, as follows: 40% Cd, 9.2% Co, 26% Cu, 64% Mn, 36% Pb, 62% Sr and 28% Zn (Table 4, Figure 3).Very similar data were obtained by the treatment of the attic dust samples collected from the houses in the village of Vozarci located at a distance of 2 km from the smelter plant.The total content of the Ni in the attic dust was 142 mg kg -1 , and from the simulated extractions the highest values that were obtained were from the model solution that simulated gastric juice: 7.6% Ni, 3.5% Cr, 27% Cu, 52% Mn, 42% Pb, 37% Sr and 31% Zn.Similar values for these elements were established for the samples from the other villages in the vicinity of the smelter plan (Drenovo, Trstenik, Resava, Rosoman, Sirkovo) (Table 4, Figure 3).Because the environment of Kavadarci and the surrounding villages is exposed to high amounts of dust in the air, there is a very high risk of bioavailabilityingestion of toxic elements from the dust by the people, especially those living in the vicinity of the ferro-nickel smelter plant.
From each variables of all samples in the score plots of PCA three groups were distinguished (Figure 4a): the first group consists of the extractions with phosphate and carbonate buffers, a second group includes the extractions with HCl and the third group the samples with the total content of elements that were completely dissolved.The Loading Plot is used as illustration of the loadings, for the interpretation of relationships among variables.From the Figure 4b it can be seen that the correlated variables are for Ni and Cr and second related group is between Cu, Mn, Pb and Zn.Namely, the ferronickel smelter plant uses ore that contains between 1 and 2.5 % Ni, about 0.05 % Co, 1-3 % Cr, etc.Therefore, these elements have significantly higher contents in samples of attic dust compared to other elements [15,32].

Figure 1 .
Figure 1.Location of the investigated region and sampling location

Figure 2 .
Figure 2. Spatial distribution of the content of: (a) Co, (b) Cr, (c) Ni and (d) Pb

Figure 3 .Figure. 4 .
Figure 3. Bioavailability assessment for the Cr, Cu, Mn, Ni, Pb and Zn from the attic dust samples from the town of Kavadarci and surrounding villages

Table 1 .
Instrumentation and operating conditions for the ICP-AES (Varian, 715-ES) system

Table 2 .
The total content of investigated elements in attic dust samples collected from various setlements (in mg kg-1 )

Table 4 .
The calculated contents of the analyzed elements in the extracts of attic dust samples, N = 10; in (mg kg-1 )