1 Introduction

Monitored natural recovery, dredging, and disposal, as well as in situ capping, have been widely adapted for the remediation of contaminated marine sediments. Monitored natural recovery is considered a passive remedial approach in which sediment contaminants are removed via naturally occurring physical, chemical, and biological processes (Hull et al. 1999). Sediment remediation by natural recovery requires a long time, and source control should be accompanied by surveillance to meet cleanup goals. Dredging and disposal are expensive and may cause secondary contamination by disturbed sediments during the dredging process (Nayar et al. 2004; Knox et al. 2008). Capping contaminated sediments is an in situ remediation technique in which a capping material is placed on top of contaminated sediments to prevent the continued contamination of surface water and biota above the contaminated sediments. Although local benthic communities underlying contaminated sediments might be buried by capping, well designed capping can reduce the damage on the benthic communities and provide clean substrate for re-colonization (Hull et al. 1999). Capping is thought to be much cheaper than dredging, and it is expected to be very effective at blocking the diffusion of contaminants in the short term (Föstner and Apitz 2007; Perelo 2010; Sun et al. 2010).

Capping can be categorized as passive or active capping depending on the material used for the capping. In passive capping, an inert material such as sand, gravel, or other non-reactive materials is used to isolate the contaminated sediments, to reduce contaminant migration, and to discourage burrowing and bioturbation for the benthic communities (Hull et al. 1999; Knox et al. 2008). Passive capping does not effectively interrupt the release of toxic contaminants, and it requires a high thickness of passive caps to isolate contaminants in marine sediments (Knox et al. 2008). In contrast with passive capping, active or reactive capping uses the materials that react with the sediment contaminant, block the release of the toxic compounds from the sediments, and minimize the bioavailability of the contaminant.

Research has been conducted on capping materials that can be used to control organic and inorganic contaminants, and on different capping materials used for each contaminant. Inorganic contaminants, especially heavy metals such as Cd, Ni, Pb, and Zn, have been reported to be effectively immobilized by phosphate minerals, including apatite (Knox et al. 2006; Knox et al. 2008), calcium phytate (Knox et al. 2008), and mineral rock phosphate (Basta and McGowen 2004; Knox et al. 2008). Other natural minerals such as zeolite and limestone were also used as a capping material to block the release of heavy metals (Jacobs and Förstner 1999; Eek et al. 2007; Knox et al. 2008). In addition to heavy metals, phosphorus and nitrogen release were also interrupted by zeolite and calcium minerals (Huang et al. 2011; Lin et al. 2011; Yin et al. 2013). Furthermore, the application of zeolite as a capping material could be extended to organic contaminants (Jacobs and Föstner 1999). For the retention of organic contaminants such as polychlorinated biphenyls, polycyclic aromatic hydrocarbons, and nitrobenzene, active capping through the use of organoclay (Knox et al. 2008; Yin et al. 2010) and activated carbon (Zimmerman et al. 2004; Murphy et al. 2006; Wang et al. 2014) were widely adapted. Although various materials have been assessed as potential capping materials, more research is needed on cheap and affordable material for capping.

In this study, we investigated the applicability of crushed concrete (CC) and bentonite (BN), which, to the best of our knowledge, have not been used as an active capping material for remediating contaminated marine sediments by other researchers. CC and BN are easily obtainable due to their low cost and high availability. Sea sand (SS) was also used in the experiment to compare it with those two capping materials. The evaluation was based on the ability of SS to block the release of toxic metals (As, Cd, Cr, Cu, Ni, Pb, and Zn) from the marine sediments and to stabilize these toxic metals in the sediments. While most previous research was performed in batch experimental conditions and few studies were done in a flow condition, all of our experiments were performed under a dynamic leaching condition using a flat flow tank.

2 Materials and Methods

2.1 Sediments and Seawater

The sediment used in this study was sampled from a coastal dock in Inchon, Korea, at a depth of 10 cm below the sediment surface. The sediments were collected in January 2012 from about 10–20 cm below the seafloor using Van Veen grab samplers. Following this, the collected sediments were homogenized by mechanical mixing after the removal of large debris. The seawater for the experiment was collected using a polyvinyl chloride airtight container. After the seawater was filtered with a GF/C filter (1.2 μm pore size, Whatman, UK) and sterilized with an autoclave (HB-506, Hanbaek Scientific Co., Korea), it was placed in a refrigerator before use.

2.2 Capping Materials

SS, CC, and BN were used as capping material, and their physical and chemical properties were analyzed. SS was collected from the harbor in Dangjin City, Korea. CC and BN were supplied by Dawon Environment and Korea Bentonite Company, respectively. All capping materials were used without any purification. A specific surface area was determined using the nitrogen adsorption isotherm technique at 77 K using an Autosorb-iQ surface area analyzer (Quantachrome Corp., USA). The data obtained from the surface area analyzer were interpreted using the Brunauer-Emmett-Teller (BET) method. X-ray fluorescence (XRF) analysis was performed using an XRF spectrometer (XRF-1700, Shimadzu Corporation, Japan) to quantify the elemental composition of SS, CC, and BN.

2.3 Bench-Scale Capping Experiments

Figure 1 depicts a schematic diagram of a flat flow tank. This system can be used to evaluate the effects of capping materials on the release and transformation of toxic metals. The device is a flat flow tank that consists of two parts. The bottom part includes a 4-cm-thick layer of contaminated sediment and a 1-cm-thick layer of capping material, with a length of 80 cm and a width of 15 cm. The top part holds the seawater, which is 86 cm in length, 15 cm in width, and 1 cm in depth. The total volume of seawater used for each experiment was 6 L. To investigate the blocking and stabilizing efficiency of SS, CC, and BN, four different experimental systems were designed: (1) no capping; (2) 1 cm thickness of SS; (3) 1 cm thickness of CC; and (4) 1 cm thickness of BN capping layer above the contaminated sediments. To mimic the sediments contaminated by trace metals, artificially contaminated sediments were prepared by adding a known amount of trace metal solution to the sampled sediments. Preparation of artificially contaminated sediments was described in supplementary material. The 4-cm-thick contaminated sediments and the 1-cm-thick capping material were filled in the water tank, and the thickness of contaminated sediments for no capping was set to 5 cm. The water enters evenly into the tank by overflowing through the inflow weir, then it flows across the tank, and then it exits the tank through the outlet opposite to the inlet. The exited seawater is stored in the reservoir and recirculated into the tank using the peristaltic pump (Model 7527-15, Cole-Parmer, USA) to maintain a flow rate of 10 mL min−1. Sampling and analysis of seawater are described in supplementary material.

Fig. 1
figure 1

Experimental setup of a bench-scale flat flow experimental system to evaluate capping materials

2.4 Analysis of Sediments

The particle size fraction was analyzed using a particle size analyzer (Bluewave, Microtrac, USA). The sediment solution pH was determined in a 1:1 mixture of sediment and deionized water using a pH meter. The total carbon and nitrogen contents in the sediment were analyzed using an elemental analyzer (LECO CHN-2000, Leco Corporation, USA). The trace metal content in the sediment used in this study was determined using ICP-OES after digesting 0.5 g of sediment in a mixture of 1.5 mL of aqua regia and 5 mL of 40 v v-1 % HF (Knox et al. 2003).

The concentration of trace metals in the sediment below the capping material was determined upon completion of the 32-day flow tank experiments. The sediment samples from all treatments were taken at two different sediment depths: upper-layer sediment (adjacent to the capping layer) and middle-layer sediment (1 cm below the capping layer). The sediment samples were dried at room temperature and homogenized to obtain a 1.0-g aliquot for conducting sequential extractions (Tessier et al. 1979).

3 Results

3.1 Characteristics of Sediments and Capping Materials

The physicochemical characterizations of the sediments sampled from the coastal dock in Inchon are shown in Table 1. The grain size distribution of the sediments was sorted into clay, silt, and sand fractions. The sediment in this study was composed mainly of the silt fraction, followed by the sand fraction. According to the biological effect-based sediment quality guidelines (SQGs) developed by the National Oceanic and Atmospheric Administration (NOAA) in the USA (Long et al. 1995; McCready et al. 2006), the contaminant concentrations corresponding to the 10th and 50th percentage of adverse biological effects were designated as the effects-range-low (ERL) and effects-range-median (ERM) guidelines, respectively. The As, Cu, and Pb concentrations recorded in this study ranged between ERL and ERM, indicating that adverse biological effects due to As, Cu, and Pb are expected occasionally. The Cd, Cr, Ni, and Zn concentrations are in the ranges where adverse biological effects are expected only rarely (<ERL).

Table 1 Characteristics of sediments sampled from a coastal dock in Inchon, Korea

Table 2 presents the physical and chemical properties of SS, CC, and BN. The grain sizes of the capping materials used in this study are in the following decreasing order: SS (1.68–2.00 mm) > CC (0.10–4.75 mm) > BN (0.10–1.18 mm). The size of the grains is highly correlated with the surface areas, which have a significant influence on the adsorption capacity for contaminants. The surface areas for SS, CC, and BN are 0.76, 5.98, and 16.25 m2 g−1, respectively. SS is mainly composed of more than 80 % SiO2, and Al2O3 and CaO follow the order of abundance. SiO2 (38.4 %) and CaO (36.2 %) are the main chemical components of CC, and CC has a higher amount of CaO than other capping material. BN is composed mainly of SiO2 (72.8 %), and Al2O3 makes up a larger portion of BN than SS and CC.

Table 2 Chemical composition and physical properties of sea sand (SS) crushed concrete (CC), and bentonite (BN) used as capping materials

3.2 Effect of Capping Materials on Trace Metals Released from Sediments

The effect of capping materials on the chemical condition of seawater can be observed by comparing the pH and EC with and without a capping layer. It can be seen in Fig. 2 that, when compared with an uncapped condition, capping material can increase the pH of seawater drastically or only slightly, depending on the capping material. The addition of CC as a capping material led to a distinct increase of pH. The average pH of CC was 8.88 ± 0.18, which was much higher than that (7.50 ± 0.11) in the uncapped condition. The average pH of SS and BN were 7.62 ± 0.12 and 7.75 ± 0.11, respectively, which were slightly higher than that in the uncapped condition. A ranking of capping materials by their EC values is as follows: BN (40.83 ± 3.45 mS cm−1) > CC (40.39 ± 3.79 mS cm−1) > Uncapped (40.10 ± 3.21 mS cm−1) > SS (38.86 ± 3.29 mS cm−1), but their difference was not noticeable.

Fig. 2
figure 2

pH and EC of the seawater above the uncapped sediment and the sediments capped with SS, CC, and BN, respectively

Figure 3 shows the release of trace metals from the contaminated sediments to overlying water over the course of 32 days. Cu and Ni can be classified as trace metals of which a small amount was released from the uncapped sediments to the overlying water and was detected within a short time. The release of Cr was lasted for longer time than Cu and Ni. Zn and As concentrations have a tendency to vary sharply, and their variation is larger than that of other trace metals. Zn concentrations of four different experiments fluctuated, and the highest peak concentration of Zn for different experimental conditions was observed at different times. Cd is considered the most mobile metal among the seven trace metals investigated in this study because it was released in the largest quantity, and its release was prolonged until the completion of the experiments.

Fig. 3
figure 3

a As; b Cd; c Cr; d Cu; e Ni; f Pb; and g Zn elution curves in water tank experiments for untreated sediments and sediments treated with SS, CC, and BN, respectively

In Fig. 3 and Table 3, Cd concentration curves show that significant differences were observed depending on the capping materials. The Cd concentration for the uncapped sediment increased continuously up to 14.51 mg L−1. The Cd concentration for both SS and BN also increased continuously, but the concentration curve for BN was below that for SS. The average Cd concentration (0.75 mg L−1) of CC capping was 1 order of magnitude less than that of SS and BN capping. In Table 3, the average and highest peak concentration indicated that the blocking efficiency for Cu and Ni followed the order CC > BN > SS. SS was effective at blocking Cr release, but the effectiveness of CC and BN at blocking Cr release was not significant. The peak concentrations of Pb for uncapped, SS, CC, and BN were 1.95, 2.07, 1.67, and 0.96 mg L−1, respectively, indicating that BN was considered the most efficient capping material for blocking Pb release. The Zn concentration curve for SS was below that for the uncapped condition during the 32-day experiment, indicating that SS effectively blocked the release of Zn from the sediments. As shown in Table 3, the average concentration of As was in the following order: uncapped (1.61 mg L−1) > BN (1.42 mg L−1) > SS (1.25 mg L−1) > CC (1.05 mg L−1). This indicates that capping the sediments with SS, CC, and BN influenced the interruption of As release. The concentration of trace metals in overlying water showed that SS capping is appropriate for blocking the release of Cr and Zn; CC capping for As, Cd, Cu, and Ni; and BN capping for Pb.

Table 3 Effect of SS, CC, and BN on trace metal concentration released from contaminated sediments during 32 days

3.3 Effect of Capping Materials on Trace Metal Stabilization

The sediments uncapped and capped with SS, CC, and BN were sequentially extracted so that As, Cd, Cr, Cu, Ni, Pb, and Zn could be partitioned into five operationally defined geochemical fractions: exchangeable, bound to carbonates, bound to Fe-Mn oxides, bound to organic matter, and residual, as shown in Fig. 4, Fig. S1, and Table S1. For the purpose of comparison, upper-layer and middle-layer sediments were used for sequential extraction. On the assumption that the mobility and biological availability of metals are highly associated with the solubility of the geochemical form, and that successive extractions can be used to quantify the strength of metals bound to sediments. The apparent mobility and potential metal bioavailability in uncapped upper-layer sediment is as follows: Cd > Zn > Ni > Pb > Cr ≃ As > Cu. In the uncapped middle-layer sediments, the mobility of trace metals follows the same order obtained from uncapped upper-layer sediments (Fig. S1).

Fig. 4
figure 4

Fraction of As, Cd, Cr, Cu, Ni, Pb, and Zn in upper-layer sediments obtained from a sequential extraction procedure

Figure 4 shows the effect of capping treatment on the metal fraction by comparing the capping material used, and Table S1 describes the fraction of trace metals in the upper-layer sediments. In the SS capped sediments, the residual fraction of As (68.71 %), Cu (41.22 %), and Pb (49.77 %) was higher than that of As (48.28 %), Cu (34.85 %), and Pb (39.51 %) in uncapped sediments. When SS was added on top of the contaminated sediments, the residual fraction of Cd (1.65 %), Cr (48.50 %), Ni (31.46 %), and Zn (27.27 %) was slightly higher than or similar to that of Cd (1.15 %), Cr (48.74 %), Ni (29.05 %), and Zn (24.24 %) in untreated sediments. Lower exchangeable and carbonate fractions of all trace metals in SS capped sediments than uncapped sediments were also found. The residual fraction of Zn (61.65 %) in CC capped sediments was shown to be slightly less than that in SS capped sediments, but the lowest exchangeable and carbonate fraction of Zn in CC capped sediments was observed. It is apparent that the carbonate fraction of Cd, Cr, Cu, and Pb in CC capped sediments was higher than that in the other condition. BN capped sediments showed a larger residual fraction of Cu, Pb, and Zn than the uncapped and the other capped condition. Sequential extractions demonstrated that SS capping is appropriate for stabilization of As, Cu, and Ni; CC capping for Zn; and BN capping for Pb.

Figure S1 illustrates the influence of the application of SS, CC, and BN capping on the stabilization of trace metals in middle-layer sediments. When compared with the trace metals in the upper-layer sediments, the trace metals placed in the middle-layer sediments are less influenced by the capping layer due to the limitation of mass transfer by low hydraulic conductivity of the sediments. The addition of SS and BN did not significantly influence the fraction of Cr, Ni, Pb, and Zn in the middle-layer sediments. Capping treatments with CC may decrease As and Cu solubility in the middle-layer sediments by forming a more stabilized form of As and Cu, i.e., a residual fraction and an organic bound fraction. The effect of capping on the immobilization of trace metals is discussed in the following section in more detail.

4 Discussion

As shown in Fig. 2, pH in the presence of capping materials was higher than that in the uncapped condition. A significant increase of pH in CC compared with SS and BN was observed. The concomitant increase in pH with CC addition is probably the result of the alkaline nature of CC, as indicated by its relatively high CaO content (Table 2). It is well known that pH has a significant influence on the behavior of trace metals in soils, and heavy metals are less mobile under higher pH conditions (Alloway 1995; Welp and Brümmer 1999). Therefore, it can be inferred that the pH increase due to CC addition can influence the mobility of trace metals in the sediments. The silanol groups become negatively charged at a pH range of 2–3 and the aluminol groups at around pH 6, where the uptake of cationic metal ions on the surface occurs (Shi et al. 2009). At low pH, the silanol and aluminol groups are protonated, and the metal adsorption to mineral surfaces is hindered (Batchelor 1998).

Figure 3 and Table 3 show that the trace metals were released to the overlying water from uncapped sediments in the following decreasing order: Cd > As > Zn > Ni > Pb > Cr > Cu. As a consequence of the results obtained from the sequential extraction experiments with uncapped sediments, the mobility of trace metals placed in the following decreasing order: Cd > Zn > Ni > Pb > As > Cr > Cu (Fig. 4). With the exception of As, the results obtained from the metal concentration eluted from the sediment were fairly consistent with the sequential extraction experiments. High amounts of Cd in exchangeable and carbonate fractions can contribute to the highest release of Cd. The high mobility of Cd is in accordance with the findings of several other researchers. Stephens et al. (2001) reported that Cd and Zn were more mobile than Pb, Cu, Ni, Zn, and Fe. Hickey and Kittrick (1984) found that metals were extracted in the order Cd > Zn > Cu ≃ Ni. Singh et al. (1998) reported the order of metal mobility as Cd = Zn > Ni > Cu > Pb for dredged sediment. Both our data and those of previous studies have shown that Cd and Zn were more mobile and less adsorbed to the sediment surfaces than other trace metals. Unlike As and Cr, which are mainly present as oxyanion form (McBride 1994), heavy metals including Cd, Cu, Ni, Pb, and Zn have different mechanisms associated with the adsorption to solid surfaces. The adsorption of cationic metal ions on the surface of solids is strongly influenced by the charge-to-radius ratio, metal electronegativity, and the hydrolysis constant.

Electronegativity is an important factor in determining the chemisorption of trace metals, and the metal with higher electronegativity can form stronger covalent bonds with O atoms on the edges of clay minerals (McBride 1994). The trace metals can be placed in the order of electronegativity: Pb (2.33) > Ni (1.91) > Cu (1.90) > Cd (1.69) > Zn (1.65) (McBride 1994; Shi et al. 2009). A lower metal electronegativity of Cd and Zn leads to a lower tendency of the ion for specific adsorption, resulting in a higher amount of metal released from sediments, as shown in Fig. 3. However, Pb with the highest metal electronegativity does not exhibit a higher tendency for specific adsorption than Cu, and Ni is more released from uncapped sediments than Cu even with the similar electronegativity.

The metal adsorption to clay mineral surfaces via surface complexation reactions is also influenced by the electrostatic force between metal ions and the mineral surfaces (Shi et al. 2009). Even if Cd, Cu, Ni, Pb, and Zn have the same valence, they have a different charge density because of their different ionic radius. Cd and Pb have a larger ionic radius (0.97 and 1.20 Å, respectively) than that of Ni, Cu, and Zn (0.69, 0.73, and 0.74 Å), indicating that Cd and Pb have a lower charge density and lower Coulombic attraction toward the surface of the clay mineral (Shi et al. 2009). The charge density of a metal ion has a significant influence on its adsorption to clay mineral, but some results are inconsistent with the order of the metal charge density. Pb with a lower charge density was less mobile in uncapped sediment than Zn and Ni with a higher charge density.

The metals that can form hydroxyl complexes more easily are specifically adsorbed to a greater extent. In more detail, the adsorption of metals to the mineral surfaces increases upon decreasing the pK (equilibrium constant) values for the hydrolysis reaction M2+ + H2O = MOH+ + H+ (Alloway 1995). The order for increasing specific adsorption with decreasing pK value was reported as follows: Cd (pK = 10.1) < Ni (pK = 9.9) < Co (pK = 9.7) < Zn (pK = 9.0) < < Cu (pK = 7.7) < Pb (pK = 7.7) < Hg (pK = 3.4) (Brummer 1986). The results of the metal concentration of overlying water showed that Cd was the most mobile and least bound to the uncapped sediments, and Pb and Cu were shown to have the lowest concentration of effluent from uncapped sediments. These results coincided with the order of pK values for the hydrolysis reaction. However, Zn with a lower pK value was released from uncapped sediments more than Ni with a higher pK value. Electronegativity, electrostatic force, or metal hydrolysis cannot purely describe the surface bonding of metals, but these physicochemical properties had a significant influence on the mobility of trace metals in sediments.

As, Cr, and Se are present in aqueous as oxyanions; thus, they are absorbed on the mineral surfaces via different mechanisms from cationic metals described above (McBride 1994). Anions can be adsorbed on the mineral surfaces by both nonspecific electrostatic force and ligand exchange reaction. Anions can be retained on a positively charged surface by anion exchange through a process involving nonspecific electrostatic forces (McBride 1994). The anion adsorption to the mineral surfaces via electrostatic interaction is influenced by the point of zero charge (PZC) of the mineral component and the pH of the aqua solution. The pH of seawater in these experiments was above 7.8 and higher than PZC (Al2O3, 9.1; Fe2O3, 6.7; and SiO2, 2.0, refereed from Sparks (1995)) of the main mineral component of the capping materials. This indicates that the surface of the capping materials was negatively charged, and the electrostatic interaction between oxyanions such as As and Cr and the surfaces of capping materials was not favorable. Anion adsorption to a mineral via chemisorption, also termed the ligand exchange reaction, occurs by replacing hydroxyl groups present on the mineral surfaces. Therefore, seawater with a higher pH than fresh water is not considered an easy condition for interrupting these trace metals from the marine sediment by the capping materials used in this study.

As shown in Fig. 3, four different experiments with different capping materials were performed to analyze the efficiency of SS, CC, and BN for blocking the release of trace metals from marine sediments. SS, widely used as a passive capping material, was studied to compare it with CC and BN, used as active capping materials in this study. It was reported that sand capping retarded the emergence time of the contaminant by decreasing the concentration gradient between the sediment and the water (Wang et al. 1991). Silanol and aluminol groups on the surface of the mineral play an important role in the adsorption process via forming inner-sphere complexation (Abollino et al. 2003). Therefore, it can be inferred that the silanol and aluminol groups on the surface of SS interrupted the release of trace metals from sediments. As with SS, the BN with silanol and aluminol groups on its surface can adsorb heavy metals. It was reported that montmorillonite, also referred to as bentonite, can adsorb heavy metals via cation exchange in the interlayers and inner-sphere complexation (Abollino et al. 2003). The increased pH due to the addition of CC was effective for interrupting the release of trace metals via surface precipitation. It was reported that Cu2+ and Zn2+ ions were removed by crushed concrete fines via surface precipitation, whereas the mechanism of interaction of Pb2+ with the adsorbent media was via isomorphic substitution of Pb2+ for Ca2+ after diffusion into the cement matrix (Coleman et al. 2005).

CC capping treatment on the contaminated sediments increased the carbonate fraction of Cd, Cu, Ni, and Pb in the contaminated sediments when compared with the uncapped condition. Gray et al. (2006) reported that the soluble and extractable metal concentration in the soil was decreased due to the increase of soil pH with the addition of lime, a calcium-containing inorganic mineral that is the predominant component of CC, as shown in Table 2. As shown in Fig. 3, CC was better at interrupting the release of Cd, Cu, Ni, and Pb than other capping materials used in this study. From the results of eluted trace metal concentration and sequential extraction experiments, it can be inferred that the carbonate fraction, considered a weakly bound fraction, can contribute to reduce the mobility of trace metals in marine sediments. Although a sequential extraction method was developed to estimate the mobility of trace metals in soil (Mulligan et al. 2001), it was not a perfect method for describing the effect of capping material on the immobilization of trace metals in the short term.

5 Conclusions

In this study, an evaluation was carried out on the use of SS, CC, and BN as capping materials to reduce the release of trace metals from marine sediments and to sequestrate those pollutants in marine sediments. The trace metals released to the overlying water from uncapped sediments follow in decreasing order: Cd > As > Zn > Ni > Pb > Cr > Cu. The mobility of trace metals obtained from the sequential extraction experiments are placed in the following decreasing order: Cd > Zn > Ni > Pb > As > Cr > Cu. The order of the mobility of trace metals is not purely but significantly influenced by their electronegativity, electrostatic force, and metal hydrolysis. The release of Cd, which is the most mobile metal, was effectively blocked by the CC capping layer. As, Cu, and Ni as well as Cd release was reduced by the addition of CC. However, except for Zn, the addition of CC did not contribute to the stabilization of trace metals, but it did increase the carbonate fraction of Cd, Cr, Cu, and Pb. The carbonate fraction, considered a weakly bound fraction, can contribute to reducing the trace metal concentration in the aqueous phase. From the sequential experiments, SS capping was proven an appropriate capping material for the stabilization of As, Cu, and Ni.