Environ Sci Pollut Res DOI 10.1007/s11356-014-2750-x

RESEARCH ARTICLE

Responses of kinetics and capacity of phenanthrene sorption on sediments to soil organic matter releasing Xiaoyan Zhang & Yaoguo Wu & Sihai Hu & Cong Lu & Hairui Yao

Received: 9 January 2014 / Accepted: 7 March 2014 # Springer-Verlag Berlin Heidelberg 2014

Abstract Soil organic matter (SOM) releasing with dissolved organic matter (DOM) formed in solution was confirmed in a sediment/water system, and the effects of SOM releasing on the sorption of phenanthrene on sediments were investigated. Inorganic salt (0–0.1 mol L−1 NaCl) was used to adjust SOM releasing, and two sediments were prepared, the raw sediment (S1) from Weihe River, Shann’xi, China, and the eluted sediments with and without DOM supernatant remained, termed as S2a and S2b, respectively. The FTIR and 1H NMR analysis indicate that the low molecular weight hydrophilic SOM fraction released prior to the high molecular weight hydrophobic fraction. As a response, phenanthrene sorption kinetics on S1 showed atypical and expressed as three stages: rapid sorption, pseudo sorption with partial desorption, and slow sorption, thus a defined “sorption valley” occurred in kinetic curve. In all cases, partition dominates the sorption, and sorption capacity (Kd) ranked as S2b > S1 > S2a. Compared with the alterations of sediment characters, DOM solubilization produced by SOM releasing exhibited a greater inhibitory effect on sorption with a relative contribution of 0.67. Distribution coefficients (Kdoc) of PHE into DOM clusters were 2.10×104–4.18×104 L kg−1, however a threshold concentration of 6.83 mg L−1 existed in DOM solubilization. The study results will help to clarify PAHs transport and their biological fate in a sediment/water system.

Responsible editor: Zhihong Xu Electronic supplementary material The online version of this article (doi:10.1007/s11356-014-2750-x) contains supplementary material, which is available to authorized users. X. Zhang : Y. Wu (*) : S. Hu : C. Lu : H. Yao Department of Applied Chemistry, Northwestern Polytechnical University, Xi’an 710072, China e-mail: [email protected]

Keywords SOM releasing . Dissolved organic matter . Polycyclic aromatic hydrocarbons . Sorption kinetics . Solubilization . Bioavailability

Introduction Polycyclic aromatic hydrocarbons (PAHs) are classified as priority pollutants according to the US Environmental Protection Agency, due to their mutagenic, carcinogenic, and endocrine-disrupting properties (Liu et al. 2012; Xia et al. 2012). Being hydrophobic, PAHs tend to be sorbed and accumulated in organic rich sediments, and become less available for biodegradation (Mahanty et al. 2011; Ortega-Calvo et al. 2013), which seriously restricts remediation technologies applied for contaminated sediments. Therefore, sorption/ desorption behaviors of PAHs on sediments were of growing concern in past decades (Weber and Huang 1996; Gao et al. 2007; Mahanty et al. 2011; Sun et al. 2013). Numerous researches have been conducted on sorption/ desorption behaviors of PAHs on sediments, and made several progresses. One of the findings is that sorption of PAHs on sediments is intimately related to the content and the structure of soil organic matter (SOM) in sediments: (1) sorption capacity is positively correlated to the content of SOM, particularly when partition is the dominant sorption mechanism (Gao et al. 2007; Yang et al. 2010); (2) SOM can be treated as heterogeneous combinations of “rubbery carbon” and “glassy carbon”, leading to a high sorption capacity as well as desorption hysteresis (Weber and Huang 1996; Chen and Huang 2011; Sun et al. 2013). These observations exhibit the profound effects of SOM on PAHs sorption. However, it is noteworthy that these researches usually made the simplification of allowing the bulk sediment to remain stable during PAHs sorption, which is actually a limitation in these researches.

Environ Sci Pollut Res

Indeed, SOM releasing occurs ubiquitously in a sediment/water system (Wang et al. 2011a, b; Laurén et al. 2012; SanClements et al. 2012), and simultaneously produces dissolved organic matter (DOM) in aqueous solution. It not only changes the physiochemical properties of the sediment (such as SOM content (Gao et al. 2007), specific surface area (Wang et al. 2008), structural heterogeneity (Kang and Xing 2005; Chen and Huang 2011), etc.), but also the nature of aqueous solution with the formation of inherent DOM. Numerous studies have indicated that DOM in aqueous solution generally impeded PAHs sorption, ascribing to the complex formation of DOM with PAHs or competition for sorption sites (Polubesova et al. 2007; Zhang et al. 2011), which enhanced apparent solubilities of PAHs. In addition, similar to synthetic surfactants, DOM can also aggregate into micelles and cause distinct solubilization for PAHs (Wu et al. 2010). Consequently, sediment stability is likely to be destroyed during PAHs sorption, and its physiochemical properties would be changed. Further, with the formation of DOM in solution, it is reasonable to tentatively put forward the assumption that SOM releasing with inherent DOM formation would impact PAHs sorption on sediments. Fortunately, with respect to PAHs desorption, Xu et al. (2008) recently found SOM releasing impacted phenanthrene (PHE) desorption kinetics obviously, making desorption concentration initially increase, then decrease, and finally achieve equilibrium. This strongly prompts us to test the reliability of the assumption, and discuss the mechanisms involved. To our knowledge, most reported studies focused on the influences of the content and structure of SOM, or the concentration and origin of exotic DOM on PAHs sorption capacity (Weber and Huang 1996; Gao et al. 2007; Yang et al. 2010; Sun et al. 2013), few studies of SOM releasing with inherent DOM formation on PAHs sorption, especially on sorption kinetics were reported. Therefore, in the present study, an inorganic salt (0– 0.1 mol L−1 NaCl), which was widely used as an extractant of DOM (Mavi et al. 2012), was applied to adjust SOM releasing, and batch tests were conducted to explore the effects and their mechanisms of SOM releasing with inherent DOM formation on PHE (a model PAH compound) sorption kinetics and capacity. Furthermore, since bioavailability of PAHs combined with DOM differs from that of freely dissolved ones, we will also try to quantify the relative contributions of DOM solubilization and the alterations of sorbent characters to the effects of SOM releasing on sorption. This research will help to reveal the effects of SOM releasing on PAHs’ transport, and the results will provide valuable information for PAHs’ bioavailability and risk assessment.

Materials and methods Sediments and their characterization Two kinds of sediments were prepared. The raw sediment, termed as S1, was collected from the top 0– 15 cm of Weihe riverbed, in Xi’an, China. Samples were air-dried, ground, sieved through a 130 mesh and stored in a closed container. The eluted sediment, termed as S2, was obtained from S1 via eluting by NaCl solutions of various concentrations (0– 0.1 mol L−1) for 24 h, separated from the supernatant and air-dried. Preliminary experiments showed that SOM releasing equilibrium reached in less than 24 h. S2 with upper DOM solution remained for sorption experiments was referred to as S2a, and with DOM solution discarded to eliminate DOM solubilization was referred to as S2b. S1 and S2 were characterized and selected physiochemical properties are listed in Table 1. Specific surface area and microporosity were evaluated by N 2-BET method (Tristar II 3020, Micromeritics, USA). Elemental composition was determined by Energy Dispersive Spectrometer (INCA X-Act, TESCAN, Czech Republic). DOM preparation and characterization Molecular weight distribution of DOM varied with extracting time is one of the important factors to reveal SOM releasing process. DOM derived from S1 were prepared by placing 20 g S1 in 100-mL deionized water and then extracted on a reciprocal shaker (120 rpm) for 1 h and 24 h at (20±1)°C, respectively. The one after shaking for 24 h should be refreshed and extracted for another 24 h. Suspensions were centrifuged at 4,000 rpm for 20 min and filtered through 0.45 μm membrane. The obtained DOM stock solutions were then fractionated by using ultrafiltration technique as described in Dong et al. (2010), and the percentage of each fraction (10 kDa) was calculated by difference from the total amount. In addition, the ratio of ultraviolet absorbance of DOM solution at 465 and 665 nm (E4/E6), which was measured with a Shimadzu 2550 UVspectrophotometer, was also calculated to evaluate changes in molecular weight (Yu et al. 2011). DOM stock solutions were freeze-dried for further characterization by Fourier transform infrared spectroscopy (FTIR) and solution-state 1H nuclear magnetic resonance (NMR). The FTIR of DOM solid-phase samples were obtained for a wavenumber range of 4,000 to 400 cm−1 in a Nicolet iS10 Magna-IR Spectrometer (Nicolet Instruments Corporation, USA). Solution-state 1H NMR analysis was carried out as in Jones et al. (2013). Briefly, DOM samples (3 mg) were dissolved in D2O, transferred to 5 mm NMR tubes, and

Environ Sci Pollut Res Table 1 Basic physicochemical properties of tested sediments Sediments

foca /%

DOMb/mg L−1

pHc

Asurf /m2 g−1

Pore size /nm

C/%

O/%

Ca/%

Al/%

Fe/%

C/O

S1 S2

4.77 4.65–4.68

9.60–12.54 2.64–2.96

8.1 8.4

7.27 4.73

92.75 117.7

2.83 1.86

55.16 57.33

3.99 3.49

5.64 5.26

3.56 3.38

0.05 0.03

a

Soil organic carbon (SOC) content determined by the wet dichromate oxidation method

b

The amount of inherent DOM eluted off from sediments for 24 h, at sediment/water ratio of 1:100 and NaCl concentrations of 0–0.1 mol L−1

c

Sediment:water 1:10 (w/v)

analyzed on a Bruker Advance 400 MHz spectrometer. Chemical shifts are referenced relative to water at 4.7 ppm. PHE sorption experiments Sorption kinetics Batch experiments were performed following Rivas et al. (2008); 0.2 g S1 was weighed in 50 mL capped vials followed by 20 mL 0.1 mg L−1 PHE (purity >97 %, Merck, Germany) solution containing 0.02 % HgCl2 to inhibit biological activity. Supernatants were centrifuged at 4,000 rpm for 20 min, filtered through 0.45 μm membrane and analyzed for PHE concentrations. Particularly for S2a, after S1 was eluted for 24 h, the upper DOM solution was retained and a given amount (ensure not to disrupt SOM releasing balance) of PHE stock solution in HPLC-grade methanol was added to set the initial concentration of 0.1 mg L−1. For S2b, the upper DOM solution was completely refreshed by 20 mL 0.1 mg L−1 PHE solution containing 0.02 % HgCl2. The following procedures were conducted as mentioned above. Meanwhile, DOM concentrations were also determined to obtain SOM releasing kinetics. Sorption isotherms A series of quantities of PHE were added to set initial concentrations of 0.1–0.6 mg L−1, and the vials were shaken for a time enough to reach equilibrium (48 h). The control vials in the absence of sediments showed that PHE losses caused by photochemical decomposition and volatilization were negligible, and the sorbed quantity can be calculated by mass balance. Methanol concentrations in all solutions were less than 0.6 % (v/v) to avoid cosolvent effect. All experiments were performed in triplicate. Since salinity in aquatic environment is generally lower than that in seawater (0.5 mol L−1), 0.002–0.1 mol L−1 NaCl were used as background solutions to study the effects of SOM releasing on sorption. Analysis PHE concentration was analyzed by a HPLC system equipped with a reverse phase C18 column (TIANHE Kromasil,

150 mm×4.6 mm, 5 μm particle size) and fluorescence detector with an excitation wavelength of 250 nm and an emission wavelength of 364 nm. An isocratic 80/20 was used as mobile phase at a flow rate of 1 mL min−1. The dissolved organic carbon, representing the amount of DOM in solution, was determined by a TOC analyzer (TOC-VCPH, Shimadzu, Japan). Data in figures and tables is an average of triplicates, and standard deviations are shown as error bars.

Results and discussion SOM releasing process SOM releasing kinetics SOM releasing kinetics in S1, S2a, and S2b at different salinities are shown in Fig. 1. As shown in Fig. 1, SOM releasing occurred significantly in S1 and S2b, while SOM releasing hardly occurred in S2a. DOM concentration in each sediment/water system ranked as S2a>S1>S2b. Slightly lower than that in the system of S2a, DOM equilibrium concentrations in S1 were 9.60–13.97 mg L−1, similar to previous studies (Gao et al. 2007; Rivas et al. 2008), and in which remarkable solubilization produced by DOM was observed. DOM concentration in the system of S2b (2.64–2.96 mg L−1) were much lower than that in S1, since the pretreatment that S2b has been extracted once and the extracted DOM solution was completely refreshed, so the DOM solubilization was considered negligible. Wang et al. (2011a, b) also reported that SOM release amount decreased sharply at first during successive extractions. Hence, significant SOM releasing with evident DOM solubilization only existed in the system of S1, which was further confirmed by the obvious decrease in the content of C element (2.83 % for S1 versus 1.86 % for S2) (Table 1). In the system of S1, SOM releasing process includes rapid and slow reactions (Fig. 1). DOM concentration increased rapidly in the first 1 h to reach 72.6–78.4 % of the total amount, and subsequently increased slowly to reach equilibrium at about 24 h. The rapid reaction of SOM releasing

Environ Sci Pollut Res Fig. 1 SOM releasing kinetics in S1 (solid line), S2a (dashed line) and S2b(dotted line)

occurred within 1 h, consistent with a published report (Wang et al. 2011a, b). Pseudo-first-order equation (1) and pseudosecond-order equation (2) are employed to interpret the kinetic data and their results are shown in Table 2. logðRe −Rt Þ ¼ logRe −

k1 t 2:303

ð1Þ

t 1 1 ¼ þ t Rt k 2 R2e Re

ð2Þ

Where k1 (h−1) and k2 (L mg−1 h−1) are the release rate constants of pseudo-first-order and pseudo-second-order, respectively. Rt and Re are the amount released at time t (h) and at equilibrium, both in milligrams per liter. Based on the correlation coefficients (R2 >0.995) and relative errors (RE 0.01 mol L−1). Addition of cations weakens electrostatic repulsions between DOM molecules and sediment particles, which facilitates DOM sorption on sediment (Setia et al. 2013). In contrast, a relatively high concentration of ions would compete with DOM for sorption sites by themselves, as well as by the released inorganic cations such as Ca2+ and Mg2+ via carbonates dissolution,

Table 2 Parameters in the modeling of SOM releasing kinetics in S1 C(NaCl) mol L−1

Re,exp /mg L−1

Pseudo-first-order

△TDSb /mg L−1

Pseudo-second-order

k1 /h−1

R2

Re,cal /mg L−1

RE a /%

k2 /L mg−1 h−1

R2

Re,cal /mg L−1

RE a /%

0 0.002 0.01

11.41 9.63 9.60

0.166 0.221 0.163

0.700 0.848 0.781

5.26 4.91 5.18

53.9 49.0 46.0

0.294 0.277 0.261

0.997 0.996 0.995

10.64 9.35 8.86

6.8 2.9 7.7

42.6 40.1 39.0

0.05 0.1

12.54 12.50

0.175 0.147

0.680 0.628

5.38 5.67

57.1 54.6

0.321 0.316

0.998 0.998

11.81 11.39

5.8 8.9

60.0 120.0

a

Relative error=|Re,exp −Re,cal|/Re,exp

b

Total dissolved solids (TDS) in solution after subtracting the concentration of NaCl added

Environ Sci Pollut Res

leading to an increase in DOM concentration (Rashad et al. 2010). As predicted, △TDS in solution increased substantially when NaCl concentration was higher than 0.01 mol L−1 (Table 2). Therefore, it is feasible to adjust SOM releasing by NaCl application, and this releasing is closely related to NaCl concentration. Although the phenomenon of SOM releasing has been broadly reported (Wang et al. 2011a, b; Laurén et al. 2012; SanClements et al. 2012), detailed characters of the released compositions still remain obscure. SOM releasing in sequence DOM is a composite consisting of organic matters with different molecular weights and disparate functional groups. In terms of the molecular weight, it has been reported that the low molecular weight (LMW) fraction was more liable to release than the refractory high molecular weight (HMW) fraction (Shao et al. 2009; Laurén et al. 2012). Molecular weight distribution of DOM derived from S1 with varied extracting time is showed in Table 3. Both the release amount and release rate of LMW-fraction (5 kDa) within the first 1 h eluting process (the rapid reaction in SOM releasing). Afterward, the release amount and release rate of HMWfraction increased significantly and turned to be higher than that of LMW fraction. It indicates that SOM releasing was sequential in terms of molecular weight, and the LMWfraction released preferentially compared with the HMWfraction. This would lead to an increase in average molecular weight of DOM solution, and it was verified by the gradual decrease in E4/E6 ratio of DOM solution as extracting time extended (Table 3). Moreover, owing to the diverse solubilities of SOM components (e.g., hydrophilic moieties, hydrophobic moieties such as aliphatic carbons and aromatic carbons) with each containing their own distinct functional groups, SOM releasing is also supposed to be sequential with respect to functional groups. Compared with the components of LMW and high polarity (LMW-hydrophilic fraction), the counterpart of HMW and low polarity (HMW-hydrophobic fraction) would

be difficult to release, due to their strong binding affinity to sediment particles and the large steric hindrance against diffusion (Simpson et al. 2001). The FTIR spectra of DOM varied with extracting time are presented in Fig. 2. Apart from a broad band at 3,200–3,600 cm−1 (H-bonds, OH groups), three sharp and strong peaks appeared at 1,627, 1,384, and 1,147 cm−1. The 1,627 cm−1 band is characteristic of the C═C stretching vibration in aromatic groups, the 1,384 and 1,147 cm−1 bands can be assigned to aliphatic CH2 bending vibration, and a small peak at ∼830 cm−1 is ascribed to C–H out-of-plane bending in aromatic compounds (Polubesova et al. 2007). The latter extracted DOM showed relatively high intensities and better resolution for these peaks, suggesting a gradual increase of aromatic carbons and a particularly obvious increase of aliphatic carbons, and the DOM would exhibit pronounced paraffinic and nonpolar characteristics. According to the studies conducted by Kang and Xing (2005) and Tao et al. (2012), modified ratio of the sum of aliphatic carbon peak heights (1,384 and 1,147 cm−1) to that of aromatic carbons (1,627 cm−1), expressed as CAli/CAro, is used to reflect aliphaticity of the extracted DOM. The ratio increased from 1.35 to 2.12 as extracting time extended (Table 3), corroborating the increasing aliphaticity. In addition, the nonpolarity, defined as atomic ratio of C/O (Chen and Huang 2011; Zhang et al. 2013), decreased from 0.05 for S1 to 0.03 for S2 (Table 1), indicative of enhanced nonpolarity of the extracted DOM. Consequently, the LMW-hydrophilic fraction probably released prior to the HMW-hydrophobic fraction. This finding agrees well with 1H NMR integration data. As shown in Fig. 3, the 1H NMR spectrum of the 24-hextracted DOM exhibited sharp and well-defined peaks (1.80, 3.54, and 3.69 ppm), whereas the spectrum of the 1-hextracted DOM showed unresolved peaks. According to data in literatures (Lu et al. 2004; Guéguen et al. 2012; Xing et al. 2012), the proton distribution of DOM in the regions of 0.6– 3.1, 3.1–6.5, and 6.5–9.0 ppm can be assigned to aliphatic, carbohydrate (amino acids, sugars, CH2 adjacent to ester and ether/hydroxyl) and aromatic portions, respectively, and their relative abundances are listed in Table 3. During the first 1 h

Table 3 Distributions of molecular weight and hydrogen moieties in DOM varied with extracting time E4/E6 H moieties /%

Extracting time Molecular weight /% 0.45 μm-10 kDa 10–5 kDa 5–3 kDa 0.942) and n values (≈1), PHE sorption on S1, S2a, and S2b can be better described by the linear equation in the concentration range examined. This indicates that the SOM dominates PHE sorption on sediments as the reported studies showed (Gao et al. 2007; Yang et al. 2010). Considering the present test situations, partition coefficient (Kd) has the potential to represent sorption capacity on the studied sediments. As shown in Table 5, PHE sorption capacity on S2b was the greatest compared with results on S1 and S2a, following the order: S2b>S1>S2a, which was contrary to the range in DOM concentration of each system, indicating that sorption capacity would be reduced by DOM solubilization proportional to DOM concentration. For S2b, SOM has been extracted once and the extracted DOM was stopped, thus its Kd variation was completely attributed to salinity change. Numerous studies have reported the promoting effect of salinity on PAHs sorption, and the well-known “salting-out

Environ Sci Pollut Res Table 5 PHE sorption coefficients and Freundlich parameters on S1, S2a and S2b Sediments C(NaCl) /mol L−1 Linear isotherma (Qe =KdCe) Freundlich isotherma (Qe =KfCen) DOM concentration /mg L−1 f′oc b /% Koc c /L kg−1

S1

S2a

S2b

Kd/L kg−1

R2

n

lgKf

R2

0 0.002 0.01 0.05 0.1 0 0.002 0.01 0.05 0.1 0 0.002 0.01 0.05

202.2 240.0 272.2 227.9 222.8 165.9 196.3 219.4 167.3 188.3 229.7 257.7 278.4 264.1

0.996 0.942 0.992 0.993 0.989 0.973 0.986 0.979 0.949 0.983 0.983 0.996 0.988 0.982

0.908 1.137 0.962 1.048 0.958 0.802 0.865 0.764 0.719 0.751 0.952 1.101 0.969 0.919

2.36 2.37 2.47 2.39 2.36 2.36 2.53 2.60 2.50 2.52 2.35 2.48 2.48 2.49

0.999 0.974 0.982 0.997 0.994 0.991 0.996 0.988 0.991 0.996 0.999 0.997 0.994 0.982

11.41 9.63 9.60 12.54 12.50 13.12 12.84 12.74 13.97 13.94 2.82 2.64 2.87 2.96

4.658 4.676 4.676 4.647 4.647 4.641 4.644 4.645 4.632 4.633 4.630 4.649 4.647 4.617

4,341.0 5,132.9 5,821.2 4,904.7 4,794.5 3,574.8 4,227.3 4,723.8 3,611.6 4,064.7 4,961.4 5,542.8 5,990.6 5,720.2

0.1

258.9

0.983

0.990

2.47

0.995

2.82

4.619

5,605.4

Qe (μg g−1 ) and Ce (mg L−1 ) are the equilibrium solid-phase and aqueous-phase solute concentrations, respectively. Kd is the partition coefficient (L kg−1 ) and Kf is the Freundlich capacity parameter [(μg/g)(mg/L)−n ]. n (dimensionless) is a measure of isotherm nonlinearity

a

b

SOC content after subtracting the amount of SOM releasing

c

Organic normalized distribution coefficient Koc =Kd/f′oc

effect” on both PAHs solubilities and SOM structures are the main mechanisms (Luo et al. 2008). With NaCl concentration increased to 0.01 mol L−1, Kd value on S2b increased as expected, but decreased as concentration continuously increased, since the extensive salinity enabled SOM to aggregate into a more rigid and packed conformation, and resulted in reduced sorption sites available for PHE (Tremblay et al. 2005). For S1 and S2a, Kd values also first increased and then decreased, while their extents of Kd variation were higher than that of S2b. For example, at 0.01 mol L−1 NaCl, Kd values on S1, S2a, and S2b reached their maximums, and the enhanced extents were 34.6, 32.2, and 21.2 %, respectively. It suggests that the “salting-out effect” was not solely responsible for Kd variations on S1 and S2a. Thus, the obvious DOM solubilization in the systems of S1 and S2a would be another factor to impact sorption capacity. Gao et al. (2007) also found that PHE sorption on deionized water-eluted soils were greater than that on control soils, due to the evident DOM solubilization produced by SOM releasing, which impeded PHE sorption. Yu et al. (2011) also had the similar findings. Rivas et al. (2008) even observed some anomalous isotherms of PAHs sorption on soils with negative slopes, ascribing to the trapping of aqueous PAHs by DOM (19 mg L−1 at solid/liquid ratio of 1:50). As presented in Fig. 6, negative correlation was found between Kd values and DOM concentrations examined in the range of 9.60– 13.97 mg L−1 (R2 =0.632), implying the inhibitory effect of

DOM solubilization on sorption. Meanwhile, SOM releasing also made changes in sediment characters, including the reduced SOC content, Asurf, aliphaticity and nonpolarity, which also inhibited PHE sorption on sediment, and it was the reason that correlation coefficient between Kd values and DOM concentrations was not extremely high. Therefore, alterations of the sediment characters and the solution properties (i.e., DOM solubilization), both produced by SOM releasing, play important roles in PHE sorption. This is noteworthy and should be

Fig. 6 Relationships between Kd values and DOM concentrations

Environ Sci Pollut Res

K d ðS2b−S2aÞ ¼ K dðS2bÞ −K d ðS2aÞ

ð3Þ

K d ðS1−S2aÞ ¼ K dðS1Þ −K d ðS2aÞ

ð4Þ


0 K d ðS1−S2aÞ ¼ C ðS2aÞ −C ðS1Þ  K dðS2b−S2aÞ =C ðS2aÞ

ð5Þ

Where C(S1), C(S2a) and C(S2b) (mg L−1) are DOM concentrations in the systems of S1, S2a, and S2b, and their corresponding solubilization are termed as DOM (S1) , DOM(S2a) and DOM(S2b), respectively. α1 and α2 are the relative contributions of DOM solubilization and the alterations of sediment characters to the effects of SOM releasing on sorption, respectively. Kd(S2b − S2a) is totally attributed to DOM(S2a). Kd(S1−S2a) is attributed to the difference between DOM(S1) and DOM(S2a) (termed as DOM(S2a−S1)), and the alterations of sediment characters between S1 and S2. Kd(S1−S2a) ′ denotes the Kd variation individually caused by DOM(S2a−S1), and K′′d(S1−S2a) denotes the Kd variation individually caused by the alterations of sediment characters between S1 and S2. As shown in Table 6, the mean values of α1 and α2 were 0.67 and 0.33, respectively, indicating that DOM solubilization exhibited a greater inhibitory effect on PHE sorption than the alterations of sediment characters, and the former is two times that of the latter. It also suggested that only ∼33 % of the reduced sorption amount can be directly biodegraded, while biodegradation of the rest would be complex due to their potential combination with DOM. Obviously, this finding further confirmed more attention should be put on the study about the components and solubilization of DOM produced by SOM releasing, and their influences on both sorption kinetics and capacity. It has been reported that DOM has a character as surfactants (Wu et al. 2010), and its solubilization was actually the partition of PAHs into DOM clusters and can be expressed as distribution coefficient Kdoc (Kalmykova et al. 2013). As shown in Table 5, due to the significant DOM solubilization in the system of S2a, organic normalized distribution coefficient on S2a (Koc(S2a)) was lower than that on S2b (Koc(S2b)), so the Kdoc can be determined by the following equation:

K d ðS1−S2aÞ ¼ K dðS1−S2aÞ −K d ðS1−S2aÞ

ð6Þ

K ocðS2aÞ ¼ K ocðS2bÞ = 1 þ C ðS2aÞ ⋅K doc

  00 α1 ¼ K dðS2b−S2aÞ = K d ðS2b−S2aÞ þ K d ðS1−S2aÞ

ð7Þ

  00 00 α2 ¼ K dðS1−S2aÞ = K dðS2b−S2aÞ þ K dðS1−S2aÞ

ð8Þ

As presented in Table 6, Kdoc was calculated to be in the range of 2.10×104–4.18×104 L kg−1 (logKdoc of 4.3–4.6), highly consistent with the results (logKdoc of 4.2–4.6) in Barret et al. (2010). Kdoc values were one order of magnitude

addressed in future studies, although sediment may have a very small proportion of soluble SOM. As discussed above, effects of SOM releasing on sorption kinetics and capacity are disparate. At equilibration, sorption capacity is dependent on the total amount of SOM releasing, and it can be predicted via their negative correlation. However for the sorption kinetics, including the release amount, the features of each released fraction are also important for evaluating sorption quantity at each moment, since sediment characters and solution properties always change with contact time. Hence, records in the equilibrium sorption are insufficient, and a detailed study about the sorption kinetics is crucial for clarifying PAHs transport. Furthermore, bioavailability of PAHs combined with DOM clusters differs from that of freely dissolved PAHs (Neale et al. 2011; Akkanen et al. 2012), thus the relative contributions of DOM solubilization and the alterations of sediment characters to the inhibitory effects of SOM releasing on sorption should be quantified. Quantification of DOM solubilization Sorption capacities on S1, S2a, and S2b (termed as Kd(S1), iKd(S2a) and Kd(S2b), respectively) at each NaCl concentration are compared as follows, and results obtained are listed in Table 6.

00

0




ð9Þ

Table 6 Relative contributions of DOM solubilization and the alterations of sediment characters to Kd variation C(NaCl) /mol L−1 Kd(S2b−S2a)/L kg−1 Kd(S1−S2a)/L kg−1 C(S2a−S1) a /mg L−1 Kd(S1–S2a)′ /L kg−1 Kd(S1–S2a)″ /L kg−1 α1

α2

Kdoc /×104 L kg−1

0 0.002 0.01 0.05 0.1

0.31 0.32 0.39 0.34 0.28

2.96 2.42 2.10 4.18 2.72

a

63.8 61.4 59.0 96.8 70.6

36.3 43.7 52.8 60.6 34.5

1.71 3.21 3.14 1.43 1.44

Difference of DOM concentrations between the systems of S1 and S2a

8.3 15.4 14.5 9.9 7.3

28.0 28.3 38.3 50.7 27.2

0.69 0.68 0.61 0.66 0.72

Environ Sci Pollut Res

greater than Koc values on S1, S2a, and S2b, implying that DOM is more effective in binding with PHE than the native SOM in sediment. This is supported by many previous studies (Yu et al. 2011; López-Vizcaíno et al. 2012), in which DOM and surfactants were found efficient in improving contaminants’ mobility and satisfactory remediation effects for contaminated soils and sediments were achieved. Furthermore, Kdoc in high NaCl concentrations (≥0.05 mol L−1) were generally greater than that in low concentrations. It was ascribed to the reduction in electrostatic repulsion between DOM molecules, which induced DOM to aggregate into macromolecules, thus DOM with enhanced hydrophobicity would be an efficient phase for PHE partition. Similarly, a recent study from Wang et al. (2013) suggested the minimum concentration of NaCl required to cause humic coagulation was 61.3– 84.4 mmol L−1. However, threshold concentration seems to exist in DOM solubilization. As shown in Table 5, a strong positive correlation was found between C(S2a) and Koc(S2b−S2a) (denotes the difference of Koc values between S2a and S2b): Koc(S2b−S2a) = −1484.2+217.4C(S2a) (R2 =0.972). When Koc(S2b−S2a) equaled zero, DOM concentration of 6.83 mg L−1 was obtained. It theoretically indicates that solubilization was considered negligible when DOM concentration was not higher than 6.83 mg L−1. The result is consistent with the report of Ojwang and Cook (2013), at a concentration close to or higher than 5 mg L−1, humic acid started to form aggregates. As a result, since DOM concentrations in the system of S2b were merely 2.64–2.96 mg L−1, no sorption valley occurred in sorption kinetics on S2b. The threshold concentration of DOM solubilization is intimately related to the components and their features. For example, the afterward-released HMW-hydrophobic fraction in SOM releasing process resulted in DOM solution of elevated molecular weight, hydrophobicity, and aliphaticity, which would lower the threshold concentration, thus sorption valley occurred in sorption kinetic curve. Threshold concentration is also dependent on PHE concentration, and would increase as PHE concentration enhanced. It was because that a smaller part of the dissolved PHE would eventually complex all the released SOM, and the ratio of combined PHE to “free” PHE would be sufficiently low to be neglected, as reported in the study of Rivas et al. (2008). Accordingly, sorption valley occurred at initial PHE concentration of 0.1 mg L−1, but disappeared at 0.52 mg L−1 as shown in Fig. 4d. Hence, the threshold concentration of 6.83 mg L−1 was obtained under certain conditions that (1) PHE concentration was about 0.1 mg L−1; (2) the sediments were collected from Weihe riverbed, in Xi’an, China; and (3) salinity was 0– 0.1 mol L−1. Actually, the threshold concentration of DOM solubilization widely exists in a sediment/water system. This obtained value could serve as reference and provide valuable information for the systems of similar conditions, especially

for the northwest area in China, and also help to clarify PAHs’ transport and bioavailability in a sediment/water system.

Conclusions SOM releasing with DOM formed in solution were confirmed in a sediment/water system. The study results underlined the fact that SOM releasing impact PHE sorption on sediments in a significant way, due to the alterations of both the sediment characters and solution properties (i.e., DOM solubilization). As a response to SOM releasing, PHE sorption kinetics were atypical and contained three stages: rapid sorption, pseudo sorption with partial desorption, and slow sorption. Meanwhile, the sorption capacity was inhibited by SOM releasing. Relative contributions of DOM solubilization and alterations of sediment characters to the inhibitory effects of SOM releasing on sorption were quantified, since bioavailability of PAHs combined with DOM differs from that of freely dissolved PAHs. The results were 0.67 and 0.33, respectively, suggesting that only ∼33 % of the reduced sorption amount can be directly biodegraded, while the rest would be complex due to their potential combination with DOM. Furthermore, PHE affinity for DOM was stronger than for particle, while a threshold concentration of 6.83 mg L−1 existed in DOM solubilization. Further investigation is necessary to include the effects of both sediment and PAHs characteristics in the link between SOM releasing and PAHs sorption. This research would help to clarify the transport and biological fate of PAHs in a sediment/water system. Acknowledgments The project was financially supported by the National Natural Science Foundation of China (Grant No. 40872164), NPU Foundation for Fundamental Research (Grant No. JCY20130145), and the project titled “survey and assessment of groundwater pollution in main cities of Northwestern China (1212011220982)”.

References Akkanen J, Tuikka A, Kukkonen JVK (2012) On the borderline of dissolved and particulate organic matter: partitioning and bioavailability of polycyclic aromatic hydrocarbons. Ecotoxicol Environ Saf 78:91–98 Barret M, Carrère H, Delgadillo L, Patureau D (2010) PAH fate during the anaerobic digestion of contaminated sludge: do bioavailability and/or cometabolism limit their biodegradation. Water Res 44: 3797–3806 Chen B, Huang W (2011) Effects of compositional heterogeneity and nanoporosity of raw and treated biomass-generated soot on adsorption and absorption of organic contaminants. Environ Pollut 159: 550–556 Cheng KY, Wong JWC (2006) Combined effect of nonionic surfactant Tween 80 and DOM on the behaviors of PAHs in soil-water system. Chemosphere 62:1907–1916

Environ Sci Pollut Res Cheng X, Forsythe J, Peterkin E (2013) Some factors affecting SPME analysis and PAHs in Philadelphia’s urban waterways. Water Res 47:2331–2340 Dong MM, Mezyk SP, Rosario-ortiz FL (2010) Reactivity of effluent organic matter (EfOM) with hydroxyl radical as a function of molecular weight. Environ Sci Technol 44:5714–5720 Gao YZ, Xiong W, Ling W, Wang X, Li Q (2007) Impact of exotic and inherent dissolved organic matter on sorption of phenanthrene by soils. J Hazard Mater 140:138–144 Guéguen C, Burns DC, McDonald A, Ring B (2012) Structural and optical characterization of dissolved organic matter from the lower Athabasca River, Canada. Chemosphere 87:932–937 Ho YS, Ng JCY, McKay G (2000) Kinetics of pollutant sorption by biosorbents: review. Sep Purif Rev 29:189–232 Hur J, Lee BM, Shin HS (2011) Microbial degradation of dissolved organic matter (DOM) and its influence on phenenthrene-DOM interaction. Chemosphere 85:1360–1367 Jones V, Meador TB, Gogou A et al (2013) Characterisation and dynamics of dissolved organic matter in the Northwestern Mediterranean Sea. Prog Oceanogr 119:78–89 Kalmykova Y, Björklund K, Strömvall AM, Blom L (2013) Partitioning of polycyclic aromatic hydrocarbons, alkylphenols, bisphenol A and phthalates in landfill leachates and stormwater. Water Res 47:1317– 1328 Kang S, Xing BS (2005) Phenanthrene sorption to sequentially extracted soil humic acids and humins. Environ Sci Technol 39:134–140 Laurén A, Lappalainen M, Saari P et al (2012) Nitrogen and carbon dynamics and the role of enchytraeid worms in decomposition of L, F and H layers of boreal mor. Water Air Soil Pollut 223:3701– 3719 Liu LY, Wang JZ, Wei GL et al (2012) Sediment records of polycyclic aromatic hydrocarbons (PAHs) in the continental shelf of china: implications for evolving anthropogenic impacts. Environ Sci Technol 46:6497–6504 Long C, Lu JD, Li A, Hu D (2008) Adsorption of naphthalene onto the carbon adsorbent from waste ion exchange resin: equilibrium and kinetic characteristics. J Hazard Mater 150:656–661 López-Vizcaíno R, Sáez C, Cañizares P, Rodrigo MA (2012) The use of a combined process of surfactant-aided soil washing and coagulation for PAH-contaminated soils treatment. Sep Purif Technol 88:46–51 Lu J, Chang AC, Wu L (2004) Distinguishing sources of groundwater nitrate by 1H NMR of dissolved organic matter. Environ Pollut 132: 365–374 Luo L, Zhang S, Ma Y, Christie P, Huang H (2008) Facilitating effects of metal cations on phenanthrene sorption in soils. Environ Sci Technol 42:2414–2419 Mahanty B, Pakshirajan K, Dasu VV (2011) Understanding the complexity and strategic evolution in PAH remediation research. Crit Rev Environ Sci Technol 41:1697–1746 Mavi MS, Marschner P, Chittleborough DJ, Cox JW, Sanderman J (2012) Salinity and sodicity affect soil respiration and dissolved organic matter dynamics differentially in soil varying in texture. Soil Biol Biochem 45:8–13 Neale PA, Antony A, Gernjak W, Leslie G, Escher BI (2011) Natural versus wastewater derived dissolved organic carbon: implications for the environmental fate of organic micropollutants. Water Res 45: 4227–4237 Ojwang LM, Cook RL (2013) Environmental conditions that influence the ability of humic acids to induce permeability in model biomembranes. Environ Sci Technol 47:8280–8287 Ortega-Calvo JJ, Tejeda-Agredano MC, Jimenez-Sanchez C et al (2013) Is it possible to increase bioavailability but not environmental risk of PAHs in bioremediation? J Hazard Mater 261:733–745 Polubesova T, Sherman-Nakache M, Chefetz B (2007) Binding of pyrene to hydrophobic fractions of dissolved organic matter: effect of polyvalent metal complexation. Environ Sci Technol 41:5389–5394

Rashad M, Dultz S, Guggenberger G (2010) Dissolved organic matter release and retention in an alkaline soil from the Nile River Delta in relation to surface charge and electrolyte type. Geoderma 158:385– 391 Rivas FJ, García de la Calle R, Álvarez P, Acedo B (2008) Polycyclic aromatic hydrocarbons sorption on soils: some anomalous isotherms. J Hazard Mater 158:375–383 SanClements MD, Oelsner GP, McKnight DM, Stoddard JL, Nelson SJ (2012) New insight into the source of decadal increases of dissolved organic matter in acid-sensitive lakes of the Northeastern United States. Environ Sci Technol 46:3212–3219 Santos PSM, Santos EBH, Duarte AC (2012) First spectroscopic study on the structural features of dissolved organic matter isolated from rainwater in different seasons. Sci Total Environ 426:177–179 Setia R, Rengasamy P, Marschner P (2013) Effect of exchangeable cation concentration on sorption and desorption of dissolved organic carbon in saline soils. Sci Total Environ 465:226–232 Shao ZH, He PJ, Zhang DQ, Shao LM (2009) Characterization of waterextractable organic matter during the biostabilization of municipal solid waste. J Hazard Mater 164:1191–1197 Simpson AJ, Kingery WL, Spraul M, Humpfer E, Dvortsak P, Kerssebaum R (2001) Separation of structural components in soil organic matter by diffusion ordered spectroscopy. Environ Sci Technol 35:4421–4425 Sun K, Ran Y, Yang Y, Xing B, Mao J (2013) Interaction mechanism of benzene and phenanthrene in condensed organic matter: Importance of adsorption (nanopore-filling). Geoderma 204–205:68–74 Tao Y, Xue B, Yao S, Deng J, Gui Z (2012) Triolein embedded cellulose acetate membrane as a tool to evaluate sequestration of PAHs in lake sediment core at large temporal scale. Environ Sci Technol 46: 3851–3858 Tremblay L, Kohl SD, Rice JA, Gagné JP (2005) Effects of temperature, salinity, and dissolved humic substances on the sorption of polycyclic aromatic hydrocarbons to estuarine particles. Mar Chem 96:21– 34 Wang L, Niu J, Yang Z, Shen Z, Wang J (2008) Effects of carbonate and organic matter on sorption and desorption behavior of polycyclic aromatic bydrocarbons in the sediments from Yangtze River. J Hazard Mater 154:811–817 Wang S, Jiao L, Yang S, Jin X, Liang H, Wu F (2011a) Organic matter compositions and DOM release from the sediments of the shallow lakes in the middle and lower reaches of Yangtze River region. China Appl Geochem 26:1458–1463 Wang X, Shu L, Wang Y, Xu B, Bai Y, Tao S, Xing B (2011b) Sorption of peat humic acids to multi-walled carbon nanotubes. Environ Sci Technol 45:9276–9283 Wang LF, Wang LL, Ye XD, Li WW, Ren XM, Sheng GP, Yu HQ, Wang XK (2013) Coagulation kinetics of humic aggregates in mono- and di-valent electrolyte solutions. Environ Sci Technol 47:5042–5049 Weber WJ, Huang W (1996) A distributed reactivity model for sorption by soils and sediments. 4. Intraparticle heterogeneity and phasedistribution relationships under nonequilibrium conditions. Environ Sci Technol 30:881–888 Wu W, Sun H, Wang L, Li K, Wang L (2010) Comparative study on the micelle properties of synthetic and dissolved organic matters. J Hazard Mater 174:635–640 Xia K, Hagood G, Childers C, Atkins J, Rogers B, Ware L, Armbrust K, Jewell J, Diaz D, Gatian N, Folmer H (2012) Polycyclic aromatic hydrocarbons (PAHs) in Mississippi seafood from areas affected by the Deepwater Horizon oil spill. Environ Sci Technol 46:5310–5318 Xing M, Li X, Yang J, Huang Z, Lu Y (2012) Changes in the chemical characteristics of water-extracted organic matter from vermicomposting of sewage sludge and cow dung. J Hazard Mater 205–206:24–31

Environ Sci Pollut Res Xu M, Liu W, Xing B, Pan B, Tao S (2008) Influence of dissolved organic matter on the desorption kinetics of phenanthrene in a soil-water system. Acta Sci Circumst 28:976–981 Yang Y, Tao S, Zhang N, Zhang DY, Li XQ (2010) The effect of soil organic matter on fate of polycyclic aromatic hydrocarbons in soil: a microcosm study. Environ Pollut 158:1768–1774 Yu H, Huang GH, An CJ, Wei J (2011) Combined effects of DOM extracted from site soil/compost and biosurfactant on the sorption

and desorption of PAHs in a soil-water system. J Hazard Mater 190: 883–890 Zhang S, Shao T, Karanfil T (2011) The effects of dissolved natural organic matter on the adsorption of synthetic organic chemicals by activated carbons and carbon nanotubes. Water Res 45:1378–1386 Zhang Y, Ran Y, Mao J (2013) Role of extractable and residual organic matter fractions on sorption of phenanthrene in sediments. Chemosphere 90:1973–1979