Environ. Sci. Technol. 2004, 38, 1960-1969
Levels of PCDD/Fs, PCBs, and PCNs in Soils and Vegetation in an Area with Chemical and Petrochemical Industries M A R T A S C H U H M A C H E R , †,‡ MARTI NADAL,† AND J O S E L U I S D O M I N G O * ,† Laboratory of Toxicology and Environmental Health, School of Medicine, Rovira i Virgili University, San Lorenzo 21, 43201 Reus, Spain, and Environmental Engineering Laboratory, Department of Chemical Engineering, Rovira i Virgili University, Sescelades Campus, 43007 Tarragona, Spain
The concentrations of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/PCDFs), polychlorinated biphenyls (PCBs), and polychlorinated naphthalenes (PCNs) have been determined in soil and wild chard samples collected in an area of Tarragona County (Catalonia, Spain) with an important number of chemical and petrochemical industries. Samples were also collected in urban/ residential zones, as well as in presumably unpolluted sites. In soils, the levels of PCDD/Fs ranged from 0.16 ng I-TEQ/kg (unpolluted sampling points) to 2.65 ng I-TEQ/kg (industrial zone), and those of ΣPCBs ranged from 657 to 12038 ng/ kg in these same zones. In turn, ΣPCNs ranged from 32 (unpolluted sites) to 180 ng/kg (residential/urban sites). In contrast to soil concentrations, there were not significant differences among collection zones in the levels of PCDD/ Fs, PCBs, and PCNs found in chard. However, PCB and PCN concentrations in chard samples collected at the unpolluted sampling points were higher than the respective concentrations in soils. In general terms, the current concentrations of the organic pollutants analyzed in this study are similar or lower than data from previous reports in other countries.
Introduction Persistent organic pollutants (POPs) have been subjected to national and international regulations and controls to restrict their usage/release into the environment. However, certain POPs show some difficulties to be eliminated/controlled taking into account that they have extremely diverse patterns of use (e.g., polychlorinated biphenyls, PCBs, and polychlorinated naphthalenes, PCNs), or they are formed/released unintentionally (polychlorinated dibenzo-p-diozins and dibenzofurans, PCDD/Fs). As result, there are still uncertainties over the main sources of emissions/release to the environment. A number of small, diffuse sources may contribute to the total emission of POPs rather than just easily identifiable and primary sources (1). In relation to PCDD/Fs, there are still considerable uncertainties about the relative importance of atmospheric sources of these pollutants and their fluxes into the environ* Corresponding author telephone: +34-977-759380; fax: +34977-759322; e-mail:
[email protected]. † Laboratory of Toxicology and Environmental Health. ‡ Environmental Engineering Laboratory. 1960
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 7, 2004
ment. This uncertainty undermines efforts to relate sources to air concentrations and will hamper future efforts to further reduce human exposure to these toxic compounds. There is evidence that in recent years atmospheric emission reductions of PCDD/Fs have been occurring, with PCDD/F concentrations in atmospherically impacted media (e.g., vegetation, cow’s milk), human dietary intake, and body burdens showing significant declines (2-4). With regard to this, it is interesting to note that data from sediment cores and archived samples suggest that peak inputs of PCDD/Fs to the environment occurred probably in the late 1960s to early 1970s (5-7). However, in Spain (as in other countries) the large-scale efforts at primary source reductions were not initiated until later, with efforts to tackle emissions from municipal solid waste incinerators (MSWI) and other important primary sources. This has notable implications because it suggests that the reasons why PCDD/Fs levels have been reduced, as well as which sources/measurements were responsible for the observed declines, are not adequately understood yet. On the other hand, PCBs are highly persistent molecules used as dielectric fluids, flame retardants, and industrial lubricant fluids. Because of their persistence and widespread distribution, a number of studies have been performed to evaluate their environmental fate and to detect their potentially harmful effects on live organisms (8-12). In turn, PCNs are a group of compounds with similar physicochemical properties to those of PCBs (13). They have been used in similar applications, such as capacitor fluids, engine oil additives, cable insulation, and wood preservation. It is estimated that the cumulative global production of PCNs is approximately 10% of the PCBs (14). Other sources of PCNs to the atmosphere include chlor-alkali and thermal processes (15), as well as emissions from MSWI (16, 17). It is well-established that soils and vegetation are the major sink for airborne POPs (6). Therefore, the measurement of concentrations of POPs such as PCDD/Fs, PCBs, and PCNs in these media is useful to establish trends in their abundance and their consequences as result of natural and anthropogenic changes. Since approximately 30 years ago, one of the largest chemical and petrochemical complexes in Southern Europe is placed in Tarragona County (Catalonia, Spain). A big petroleum refinery, together with a number of important chemical and petrochemical industries, are placed in the zone. In recent years, public concern over possible adverse health effects for the population living near this industrial complex has increased. In response to this concern, we recently initiated a wide survey focused on determining the current levels of various inorganic and organic pollutants in the area and on establishing the health risks for the population living in the neighborhood of this important industrial complex. The results concerning PCDD/Fs, PCBs, and PCNs in soils and vegetation are presented here. A comparison with data corresponding to residential and unpolluted areas of Tarragona County has also been performed. Another important goal of this investigation was to establish a background value for PCDD/Fs, PCBs, and PCN in Tarragona soils and vegetation to compare the current results with data obtained in future surveys.
Materials and Methods Sampling. In Winter 2002, 24 soil and 12 wild chard (Beta vulgaris) samples were collected in various zones of Tarragona County. This is a residential area (up to 300 000 inhabitants) with an important location of chemical and 10.1021/es034787f CCC: $27.50
2004 American Chemical Society Published on Web 03/05/2004
FIGURE 1. Sampling points in the area of study. petrochemical industries. A municipal solid waste incinerator (MSWI), a hazardous waste incinerator (HWI), a PVC production facility, and a chlor-alkali plant are also located in the area. Moreover, the presence of a highway and several roads with an important traffic density influences the environment of the zone (Figure 1). Soil sampling points were chosen as follows: 15 in the industrial complex divided into industrial-1 (eight samples collected in the vicinity of chemical industries) and industrial-2 (seven samples collected near a large oil refinery and petrochemical industries), five in urban (Tarragona downtown) and residential (several suburbs of Tarragona city) zones, and finally, four outside the study area in zones presumably unpolluted. The last four samples were collected at four different places located more than 30 km from the industrial area (not shown in Figure 1). Every sampling point was chosen depending on the location of relevant focuses of contamination and also according to the possibility of obtaining meteorological data (to have a better understanding of the process of contaminants dispersion). Soil samples were taken from the upper 3 cm of soil and kept in polyethylene bags. They were dried at room temperature until showing a constant weight and sieved through a 2 mm mesh screen. Chard samples were collected in 12 of the 24 sampling points in which soil samples were taken. Six samples were collected in the industrial (industrial-1 and industrial-2) zone, three samples were collected in the residential/urban zones, and three samples were collected in unpolluted sampling points. Chard samples were obtained by cutting the aerial part of the plant, and subsequently, they were packed in aluminum foils. Samples were dried at room temperature and shredded with a domestic shredder. They were kept in double aluminum foil, packed in labeled plastic bags, and stored at room temperature until analysis. Analytical Procedure. Prior to extraction, dried samples were homogenized and spiked with a mixture of internal standards (13C12-labeled PCDD for PCDD/Fs and PCNs and
13 C12-PCB #52, 101, 153, 180 for PCBs). Samples were extracted (Soxhlet extraction), and a multistep-cleanup was performed with adsorption chromatography: a multilayer silica column (from top to bottom: sodium sulfate, silica, silica-sulfuric acid, silica, silica-potassium hydroxide, silica) and alumina columns. For chard samples, additionally, a 70 g BioBead-SX3 gel permeation chromatography column and gel permeation columns were used. The final step involved the reduction to the volume necessary for the analysis. Prior to analysis, a 13C-labeled standard was added for calculation of recovery ratios. The cleaned extract was analyzed by highresolution gas chromatography/high-resolution mass spectrometry (a Fisons CE 8000 gas chromatograph coupled with a VG Autospec Ultima system with electronic impact and a multiple ion detection mode, with a resolution g10 000). PCNs were analyzed using a 60 m DB5ms-column. PCB and PCDD/F congeners were analyzed by a DB-XLB column. For analysis of PCDD/Fs, additional polar columns of a CP-Sil88-type were used for confirmation of the lowest chlorinated congeners. The recovery rates for the POPs analyzed in this study were in the range of 42-117%. The organic matter content and pH of the soil samples were also determined. Fifteen grams of sample was dissolved in 15 mL of deionized water, and after mixing for 24 h, the measurement of pH was done with a pH meter. The organic matter content was evaluated according to the Loss on Ignition (LOI) method. Samples were dried to eliminate water content. Subsequently, they were heated for 2 h at 600 °C, and the weight loss was assessed. Data Analysis. Statistical analysis of the data was performed by one-way analysis of variance (ANOVA) for variables with normal distribution (PCDDs and PCNs). Kruskal-Wallis and Mann-Whitney tests were used for variables without normal distribution (PCDFs and PCBs). A probability of 0.05 or lower was considered significant. A multivariate analysis of the results was also done. Data matrixes were evaluated through the principal component analysis (PCA). The objective of PCA is to derive a few new components (principal components) as a linear combination of the original variables, which will provide a description of the data structure with a minimum loss of information. Each sample was assigned a score in each component, thus allowing the summarized data to be further analyzed and plotted. The hierarchical cluster analysis (HCA), which identifies homogeneous groups of samples, was performed according to the average linkage between groups method on the squared Euclidian distances matrix derived from the PCA scores. In the present study, PCA was used to investigate the relationship of POPs content between the different zones of study. All statistical analyses were carried out with the statistical package SPSS 11.0.
Results and Discussion The mean pH value of soils was 7.66 (7.03-8.22), and the mean content of organic matter in the 24 soil samples was 5.84% (2.61-9.47%). While in soils no significant correlation (Pearson) with the organic matter was found for PCDD/Fs, significant correlations were noted for PCBs and PCNs (p < 0.01 and p < 0.05, respectively). Tables 1-3 summarize PCDD/F, PCB, and PCN concentrations in soils classified according to the respective zone of collection. In turn, Tables 4-6 show the concentrations of these organic pollutants in chard samples. With respect to the concentrations of PCDD/Fs in soil samples, for most congeners significant differences were observed between industrial-1 (chemical industries) and industrial-2 (oil refinery and petrochemical industries) zones. No significant differences were found in TEQ levels between industrial-1 and residential/urban zones (2.65 ng I-TEQ/kg and 1.31 ng I-TEQ/ kg, respectively). It can be also observed that although VOL. 38, NO. 7, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1961
TABLE 1. PCDD/F Concentrations in Soil Samplesa industrial-1 (n ) 8)
industrial-2 (n ) 7)
residential/urban (n ) 4)
unpolluted (n ) 4)
p
0.07 ( 0.0a 0.21 ( 0.13a 0.36 ( 0.18a 0.84 ( 0.56a 0.74 ( 0.49a 18.26 ( 13.60a
0.03 ( 0.00b 0.03 ( 0.02b 0.05 ( 0.00b 0.09 ( 0.05b 0.05 ( 0.00b 0.94 ( 0.60b
