The chemical speciation of trace metals (Cd, Co, Cr, Cu, Ni, Pb and Zn) in marine sediments from Sulaibikhat Bay, Kuwait was determined using a three-step sequential extraction procedure. To obtain a mass balance, a fourth step, i.e. digestion and analysis of the residue was undertaken using a microwave-assisted acid digestion procedure. The sum of the 4 steps (acid-soluble + reducible + oxidizable + residual) was in good agreement with the total content (71–116%), suggesting that the microwave extraction procedure is efficient. The results showed that all metals except for Pb and Zn were present at higher percentages in the residual fraction. The reducible fraction was the next followed by the oxidizable fraction. The exchangeable fraction was least important as a host for most metals. The mobility order of exchangeable fraction of the elements in surface sediments decreased in the order Cd > Zn > Cu > Co > Pb > Ni > Cr; the reducible fraction in the order Pb > Zn > Cu > Cd > Co > Ni > Cr; the oxidizable fraction in the order Pb > Cu > Cd > Ni > Co > Zn > Cr, and the residual fraction in the order Cr > Ni > Co > Cu > Cd > Zn > Pb. The data showed that the speciation of trace elements in sediments close to a sewage outfall was different from that of sediment from other parts of Sulaibikhat Bay. This suggests that the contribution of the sewage outfall to metal pollution in adjacent marine area is positive and is associated with fine-grained sediments with high level of organic content, which are major controlling factors for the distribution of trace metals in this part of the Bay.
The speciation of metals in sediments is important in understanding the proportion of metals that are potentially bioavailable, mobile and can be transported through the aquatic food chain, and their degree of persistence in the environment (Morillo et al., 2004; Guevara-Riba et al., 2004; Yuan et al., 2004). The total metal concentration in sediments alone is not sufficient to carry out environmental risk assessments of trace metals or for evaluating the environmental impact of trace metal emissions in a defined geographical area (Landner and Reuther, 2004). It is essential to quantify and identify the chemical forms of the element in sediment in order to evaluate its impact on aquatic systems by using element speciation techniques. Sequential extraction techniques have the potential to offer useful information on how trace elements are bound into sediments (El-Hasan and Jiries, 2001; Hlavay, et al., 2004; Yuan, et al., 2004). The concentration of species is significant particularly when setting environmental and ecological standards rather than just the total trace metal concentration. This is because some sediment components such as oxy-hydroxides of manganese and iron, and organic matter are particularly important in term of their scavenging ability (Tessier et al., 1979), and can influence the composition, distribution and bioavailability of trace metals within sediment. Speciation analysis is fundamental for predicting and modelling fate, risk, effects and for decision strategies. Applied speciation principle for risk assessment identifies that marginal data quality and accepted methodologies of total trace metals can only provide a basic representation of estimated contaminant impacts on biota. Accurate risk assessment necessitates qualitative and quantitative identification of the particular contaminants of concern accompanied by characterization of the contaminants environment.
The behaviour of a particular metal in a particular environment can only be fully understood when consideration is given to its distribution between its various host phases. Metals may occur in many different mineralogical phases, each of which will have its own geochemical characteristics. However, these phases may be broadly categorised into four fractions; namely, the exchangeable fraction, the reducible fraction, the oxidizable fraction, and the residual fraction. The exchangeable fraction, held on ion-exchange sites, is freely available to take part in chemical reactions. The reducible fraction occurs in the form of complexes with oxides and hydroxides and might thus be solubilised under reducing conditions. The oxidizable fraction occurs combined with organic matter and sulphides and might become available in environments where these are oxidised. The residual fraction is introduced into the environment in crystalline form by geological processes and is generally unavailable, in the short-term, under earth surface conditions. In this study, marine sediments collected from Sulaibikhat Bay, in Kuwait were analyzed using European Community Bureau of Reference (BCR) 3-step sequential extraction procedure (Rauret et al., 2001). A fourth step, i.e. digestion of the residue from the third step, was included using a microwave-assisted acid digestion procedure.
The aim of this study was to determine and compare the chemical speciation of trace metals in Sulaibikhat Bay of Kuwait to evaluate relative mobility and bioavailability in the context of similar data reported from other areas. This is the first study employing this approach in the state of Kuwait, and to our knowledge, the first in the Arabian Gulf.
Materials and methods
Study area and sample collection
Sulaibikhat Bay (Location, 29o21'N 47o51'E; Area, 45 km2), is a small and sheltered embayment located on the south side of Kuwait Bay (Figure 1). This portion of Kuwait Bay has undergone considerable development over the years, with a major cargo port in the Shuwaikh area of its eastern part. These developments have significantly contributed to Kuwait's coastal area pollution, with Sulaibikhat Bay being described as one of the most vulnerable coastal areas to be affected (Al-Bakri, 1996; Al-Ghadban et al., 2002). The Bay has also undergone substantial filling and dredging activity receives different types of effluents in different locations in the Bay including thermal effluents from Power and Desalination Plants in its western part and sewage effluents, particularly the Ghazali outfall, southern part of Sulaibikhat Bay, in addition to major storm water drains along the coast of the Bay. Although much of Sulaibikhat Bay is shallow, it is relatively deep in the central channel area, where a maximum depth of 5.5 m is reached. The central channel is 3 km long and 0.5 km wide. Most of the rest of Sulaibikhat Bay is intertidal flats that are up to 3 km in width (Khalaf et al., 1988; Alshemmari1, 2010).
Surface sediment samples (0–5 cm) were collected from 35 stations of Sulaibikhat Bay, during December 2003 (Figure 1) Surface sediment samples from the lower tidal flat and central channel of the Bay were collected using a Van Veen grab sampler (stainless steel; 0.5 litres) while samples from the upper tidal flat were collected by hand, using an acid-rinsed plastic scoop and for both type of samples they were transferred to cleaned polythene box.
Sample preparation and analysis
Samples were analysed in batches of twelve, with each batch also containing blanks, certified reference materials and one triplicate sample. One sub-sample of each wet sediment sample was freeze-dried (Virtis, USA), ground and homogenized in preparation for total and sequential trace element analysis and a second sub-sample of wet sediment (1 to 10 g) was taken for determination of water content (allowing a salt correction to be made). Salt content correction is required for samples because high salt content introduces significant errors in trace element determination. An average salinity of 35‰ and a density of 1.025 g ml−1 may be assumed for pore water (MOOPAM, 1999). The weight of dry salt-corrected sediment was then calculated as follows:
As dried surface sediment samples were used for trace element analyses, the data required correction for salt precipitated out during drying. This was applied as follows:
Salt corrected concentration (mg kg dry wt−1) = Uncorrected concentration (mg kg dry wt−1) × (weight of dry sediment [weight Y]/weight of dry salt corrected sediment [Z])
The three-stage BCR-701 sequential extraction procedure (Rauret et al., 2001) was used to determine element speciation in the sediments with slight modifications. The extraction was performed in polythene centrifuge tubes (PTFE is recommended by BCR-701), with shaking done using an orbital rotary shaker (KS-501D, IKA-Labortechnik, Germany) (an end-over-end shaker is recommended by the BCR-701 protocol). Centrifugation was performed using a refrigerated centrifuge to control any heat generated during this process. Prior to use, laboratory glassware used was cleaned by immersing in warm 10% nitric acid overnight and then rinsing with deionised water (18.2 MΩ cm−1). The residues from stage 3 of the sequential extraction process and separate sediment samples were digested for analysis of total elements by the modified MOOPAM (1999) method using 1 ml of aqua regia (HNO3:HCl, 1:3 v/v) (Nitric acid; 69%, AnalaR, BDH, England), (Hydrochloric acid; 37%, Riedel-deHaën, Germany) and 6 ml of concentrated hydrofluoric acid (HF; 48%, Panreac, Espana).
Solutions resulting from sequential extraction and total element digestion procedures were analysed for Co, Cd, Cr, Cu, Ni, Pb and Zn by Inductively Coupled Plasma Optical Emission Spectrometry (Vista-MPX CCD Simultaneous ICP-OES; Varian, Australia). The ICP-OES was calibrated prior to analyses using 1000 mg l−1 stock ICP multi-element standard solution (Spectrosol from BDH, Poole, England), and were made up with the relevant extractant solution. The calibration standards for each stage were run at the beginning of the analysis on the ICP-OES and after every ten samples. A range of not less than four standard concentrations was selected and linear calibration curves were calculated using the least-squares method and used to determine metal contents of the analyzed samples.
Quality control and quality assurance
The accuracy of the sequential extraction data was assessed by including one certified reference material (CRM) of BCR-701 in every batch of samples analysed. The accuracy of the total digestion of trace-elements was assessed by including two certified reference materials (CRM) LGC (6137) and GBW (07311) in every batch of samples analysed. A further check on data quality was made by comparing the sum of an element's concentration in each of the four fractions determined by the sequential technique with the directly measured total trace element concentration. This was expressed as% recovery and was calculated as follows:
The analytical precision for the trace-elements sequential extractions and the total extraction were monitored through repeated analyses of CRM BCR-701 and harbour sediment reference material LGC (6137) (UK), respectively. Table 1 shows the recovery of trace elements of CRM BCR-701 within a batch of surface sediment samples. The results indicate good recovery for most elements for steps one and two, but poor recovery for step three. When the sum of the four fractions determined is compared with the total elemental concentrations, the recoveries ranged between 71.3 and 116% (Table 1). These results are in agreement with other studies from China (Yuan et al., 2004) and Singapore (Cuong and Obbard, 2006) reporting recoveries between 78–117% and 84–125%, respectively. This implies that the method used is reliable and reproducible.
Results and Discussion
The extractable contents of Cd, Co, Cr, Cu, Ni, Pb and Zn and the extracted percentages of these metals with respect to the sums of 4 extraction fractions of sediments were calculated (Figure 2). All metals except for Pb and Zn were present at higher percentages in the residual fraction. The reducible fraction was the second most significant followed by the oxidizable fraction. The exchangeable fraction was least important as a host for most metals. As known, among the various fractions, exchangeable elements are commonly considered to be most dangerous for the environment because they are weakly bound and equilibrate easily and rapidly with the water column (Tessier et al., 1980; Morillo et al., 2004). In the sediments from Sulaibikhat Bay, the proportion of trace elements in the exchangeable fraction decreased in the order Cd > Zn > Cu> Co > Pb > Ni > Cr. This pattern was also observed in studies carried out in Spain, Belgium, and Singapore (Singh et al., 2000; Guevara-Riba, et al. 2004; Cuong and Obbard, 2006). High concentrations of trace elements in the exchangeable fraction can be regarded as a pollution indicator (Förstner and Wittmann, 1979). Exchangeable Cr and Ni were found in negligible amounts in the sediments from Sulaibikhat Bay, which means that these metals are unlikely to be either mobile or bioavailable. Moderate amounts of exchangeable Co, Cu, Pb, and Zn than Cd suggest that although they were less mobile and bioavailable than Cd, a significant proportion of these metals is only very weakly bound (Figure 2).
The scavenging efficiency of the reducible fraction (iron oxides and hydroxides) for trace elements has been widely reported (Tessier et al. 1979; Jones and Turki, 1997) and is an important influence on the behaviour (through co-dissolution and/or co-precipitation) of trace elements. In the sediments from Sulaibikhat Bay, the proportion of trace elements in the reducible fraction decreased in the order Pb > Zn > Cu > Cd > Co > Ni > Cr. There were significant trends in the concentration levels of Pb and Zn bound in the reducible fraction observed in the sediments (Figure 2). The reducible fraction was dominant for Pb and Zn at all stations, with the highest concentrations at station 2, close to the Ghazali sewage outfall. Here, the reducible fraction accounted for 80% of total Pb, and 60% of total Zn. These findings agree with earlier studies (Salomons and Förstner, 1980; Tessier et al., 1980; Jones and Turki, 1997; Morillo et al., 2004; Cuong and Obbard, 2006; Larner et al., 2007).
In unpolluted marine sediments in the Bay, most Zn is originally in the residual fraction, but, as the intensity of pollution increases, an increasingly large proportion of total Zn is found in the reducible fraction. Thus, pollutant inputs sometimes appear as host phases that are different from those containing contributions from natural inputs. Similarly to Zn, the association of Cu with the reducible fraction in sediment of the Bay increased in polluted sediments. It is possible that, in polluted sediments, Cu is found binding to the Fe–Mn oxide fraction more than the residual fraction. The high proportions of Cu bound to reducible fractions found in this study differs from the results reported by other workers, who found Cu bound to the oxidizable fraction (Boughriet et al., 1994; Jones and Turki, 1997; Morillo et al., 2004; Yuan et al., 2004; Cuong and Obbard, 2006). This might be because pollutant inputs tend to redistribute in host phases under changing environmental conditions that are different from those from their input sources.
The oxidizable fraction is especially significant in polluted sediments (Förstner and Wittmann, 1979; Hlavay et al., 2004) and has a large influence on trace element distributions (Ridgway and Price, 1987). In this study, the proportion of trace elements in the oxidizable fraction decreased in the order Pb > Cu > Cd > Ni > Co > Zn > Cr. This study shows very high amounts of total Pb and total Cu. Oxidizable Pb was in the range 5–40% of total Pb, and oxidizable Cu was in the range 5–20% of total Cu. This probably indicates an association of both Pb and Cu with organic matter. The importance of the oxidisable fraction as a Cu host observed in this study is similar to the conclusions reached by other workers (Tessier et al.,1980, Salomons and Förstner, 1980, and Rauret et al., 1988). The oxidizable fraction was also observed to contain moderate amounts of Cd and Ni, but was found to contain low amounts of Co, Zn, and Cr.
The residual fraction is characteristically composed of detrital silicates and resistant mineral phases that are important carriers of trace elements (Tessier et al., 1979; Jones and Turki, 1997). The trace elements in this fraction are generally structurally bound in detrital silicates such that under natural conditions their release into the overlying water is unlikely (Samant et al., 1990; Boughriet et al., 1994; Morillo et al., 2004). The residual fraction is, thus, considered environmentally non-toxic. In this study, the proportion of trace elements in the residual fraction decreased in the order Cr > Ni > Co > Cu > Cd > Zn > Pb. The concentration levels of Cr and Ni bound in the residual fraction was especially significant (Figure 2). The residual fraction was predominant for both Cr (up to 90% of total Cr) and Ni (up to 75% of total Ni). These findings agree with the results of previous studies from the East China Sea (Hlavay et al., 2004; Yuan et al., 2004). The residual fraction was also predominant for both Co (>70% of total Co) and Cu (>50% of total Cu) (Figure 2). In areas close to sewage discharge, Cd, Pb, and Zn were predominantly in non-residual fractions: Cd > 60% total Cd, Pb > 10% total Pb, and Zn > 30% total Zn. At sampling stations distant from the source of sewage discharge, Cd, Pb, and Zn were predominantly in the residual fraction: Cd > 65% of total Cd, Pb > 50% of total Pb, and Zn > 80% of total Zn. These findings agree with the results of a previous study from East China Sea (Yuan et al., 2004) and indicate that there are significant differences in the associations of Pb and Zn with the residual fraction as one moves from the more polluted to less polluted parts of Sulaibikhat Bay. In general, this study showed that the contribution of the sewage outfall to metal pollution in adjacent marine area is positive and is associated with fine-grained sediments with high level of organic content, which are major controlling factors for the distribution of trace metals in this part of the Bay.
The partitioning of trace elements of a three-stage sequential leaching procedure of sediments from Sulibikat Bay, Kuwait was determined, and provided the first information about metal speciation in the marine environment of Kuwait. Amongst the various fractions, the residual fraction was found to be the dominant fraction for most of the trace elements considered in this study. In terms of elements content, the residual fractions in sediments decreased in the order Cr > Ni > Co > Cu > Cd > Zn > Pb, the reducible fractions decreased in the order Pb > Zn > Cu > Cd > Co > Ni > Cr, the oxidizable fractions decreased in the order Pb > Cu > Cd > Ni > Co > Zn > Cr, and the exchangeable fraction of trace elements decreased in the order Cd > Zn > Cu > Co > Pb > Ni > Cr. For most trace elements, concentrations were highest in the south-east of Sulaibikhat Bay, principally influenced by the sewage waste discharge, with concentrations decreasing to the west.
H. Alshemmari would like to thank Dr. A. N. Al-Ghadban (Manager of Environmental Science Department at Kuwait Institute for Scientific Research [KISR]) for his encouragement and KISR for financial support. Thanks are due to Dr. Bryn Jones from the School of Civil Engineering and Geosciences, the University of Newcastle upon Tyne, UK, for his guidance during this study.