Science of the Total Environment 481 (2014) 209–216
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Co-sorption of ofloxacin and Cu(II) in soils before and after organic matter removal Di Wu a, Hao Li a, Shaohua Liao a, Xiaolong Sun b, Hongbo Peng a, Di Zhang a, Bo Pan a,⁎ a b
Faculty of Environmental Science & Engineering, Kunming University of Science & Technology, Kunming 650500, China College of Environmental Science and Engineering, Southwest Forestry University, Kunming 650224, China
H I G H L I G H T S • • • • •
Sorption of the co-adsorbate increased at low primary adsorbate concentration. Sorption of the co-adsorbate decreased at high primary adsorbate concentration. Complexation of the adsorbate by the adsorbed co-adsorbate increased sorption. Overlapping of the sorption sites decreased the sorption. Apparently increased or decreased sorption may be only a part of the whole picture.
a r t i c l e
i n f o
Article history: Received 17 December 2013 Received in revised form 10 February 2014 Accepted 10 February 2014 Available online 2 March 2014 Keywords: Binary sorption Cation bridge Competitive sorption Complementary sorption Complexation
a b s t r a c t Various mechanisms play roles simultaneously for antibiotic sorption on solid particles. Previous studies simply emphasized mechanisms that match the increased or decreased antibiotic sorption by metal ions, without a general concept including these diverse mechanisms in their co-sorption. We observed both increased and decreased OFL and Cu(II) sorption in their co-sorption system. The comparison of the sorption coefficients of primary adsorco pri bate (Kpri d ) and co-adsorbate (Kd ) suggested that enhanced sorption occurred at high Kd region (low primary was decreased to a certain value deadsorbate concentration). Competitive sorption was observed when Kpri d pending on solid particle properties. We thus summarized that if the adsorbates were introduced with low concentrations, OFL (such as hydrophobic region in solid particles) and Cu(II) (such as inner-sphere complexation sites) occupied their unique high-energy sorption sites. Cu(II) complexed with the adsorbed OFL, and OFL bridged by the adsorbed Cu(II) promoted the sorption for both chemicals. With the increased concentrations, the adsorbates spread to some common sorption sites with low sorption energy, such as cation exchange and electrostatic attraction region. The overlapping of Cu(II) and OFL on these sorption sites resulted in competitive sorption at high concentrations. The previously reported apparently increased or decreased sorption in antibiotic– metal ion co-sorption system may be only a part of the whole picture. Extended study on the turning point of decreased and increased sorption relating to water chemistry conditions and solid particle properties will provide more useful information to predict antibiotic–metal ion co-sorption. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Antibiotics are widely used in medical therapeutic for humans, growth promotion and disease treatment for animals. With decades of massive application, antibiotics are now detected in surface water, ground water, waste water, sewage, soils and sediments (Wu et al., 2012). Previous studies indicated that the continuous input and wide presence of antibiotics in terrestrial system provided an acclimatizing environment for microorganisms, which in turn promoted the development of drug resistance genes through mutation or gene transfer ⁎ Corresponding author. Tel./fax: +86 871 65170906. E-mail address: [email protected] (B. Pan).
http://dx.doi.org/10.1016/j.scitotenv.2014.02.041 0048-9697/© 2014 Elsevier B.V. All rights reserved.
(Seveno et al., 2002; Chee-Sanford et al., 2009). The failure of medical functioning of antibiotics is frequently reported. Antibiotic environmental behavior is the primary process controlling their risks, and thus attracted lots of academic attention (Pan et al., 2009). It was previously pointed out that metal ions could interact with the various functional groups in antibiotics, which altered the sorption behavior and risk of antibiotics. However, various, sometimes controversial results were reported in literature, including both increased and decreased antibiotic adsorption by metal ions. All these impacts were provided well with theoretical explanations. For example, metal ions increased the apparent sorption of antibiotics through electrostatic attraction (Kahle and Stamm, 2007), salting out effects (Zhou, 2006; Li et al., 2007), or cation bridging (Jia et al., 2008). On the other hand, metal
210
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
6.10
Distribution of OFL species
A 8.28 1.0
B
0.8 +
OFL OFL± OFL-
0.6 0.4 0.2 0.0 4.0
5.0
6.0
7.0
8.0
9.0
10.0
pH Fig. 1. Dissociation of ofloxacin (A) and distribution of cationic (OFL+), zwitterionic (OFL±0), and anionic (OFL−) ofloxacin in aqueous solution as a function of pH (pKa1 = 6.10 and pKa2 = 8.28) (B).
ions decreased antibiotic sorption through competition (Ter Laak et al., 2006; Bai et al., 2008) or outer-sphere complexation (Gu and Karthikeyan, 2005). However, the emphasis on one or several mechanisms does not necessarily exclude the other(s). Investigating the sorption variation of antibiotics at different pHs in the presence of metal ions provided useful information to understand the mechanisms that control their co-sorption. Researchers generally reported that metal ions decreased antibiotic sorption at relatively low pH (because of the competition between positively charged antibiotics and metal ions), but increased antibiotic sorption at relatively high pH (because of the cation bridging) (Pei et al., 2009, 2011; Jia et al., 2008). As discussed in previous studies, these above-mentioned interaction mechanisms may simultaneously play roles in antibiotic sorption (Zhang et al., 2010). The overwhelming of certain interaction mechanisms over others determines the apparent increasing or decreasing sorption. We thus hypothesize that both increased and decreased antibiotic sorption occur in a binary sorption system, and the apparent impact is determined by the combination of water chemistry conditions and solid particle properties. In the binary sorption system, investigators generally measured the concentration of primary adsorbate only, as summarized in Table S1 regarding literature results on co-sorption of ionic organic chemicals and metal ions. Thus, some key information may have been missed. For
example, the decreased antibiotic sorption with the addition of metal ions could be explained by competition between positively charged metal ions and antibiotics (Pei et al., 2011), or the blocking of antibiotic sorption sites nearby the complexes formed by the adsorbed metal ions (Chen et al., 2009). Clearly, the former will result in decreased metal sorption, while the latter will not. Another example is that both cation bridging and the sorption of antibiotic–metal ion complexes could result in the increased sorption of antibiotics, which resulted in the similar increased sorption of antibiotics. However, cation bridging will not increase metal ion sorption, while the sorption of antibiotic–metal ion complexes may result in increased sorption of both antibiotics and metal ion. Providing metal sorption information in the binary sorption system will greatly facilitate the discussion on co-sorption mechanisms. According to the above discussion, we separately used antibiotics and metal ions as primary adsorbates. The sorption of the co-adsorbates was also determined and analyzed together with the sorption of the primary adsorbates. Ofloxacin (OFL) was chosen as the model antibiotic chemical because of its massive application and wide presence in the environment (Seifrtova et al., 2008; Karthikeyan and Meyer, 2006). Cu(II) was selected as a representative metal ion because of its strong complexation property (Macias et al., 2001). Soil samples before and after organic matter removal were used as adsorbents to include particles with a wide span of sorption strength. With the knowledge that different mechanisms play roles
2.5
A
logSeOFL (mmol/kg)
logSeCu (mmol/kg)
3.0 2.5 2.0 1.5 1.0
-2.5
-1.5
-0.5
logCeCu (mmol/L)
0.5
B 2.0 1.5 1.0
MA PMA MB PMB
0.5 0.0 -3.0
-2.5
-2.0
-1.5
-1.0
-0.5
logCeOFL (mmol/L)
Fig. 2. The sorption isotherms of Cu(II) (A) and OFL (B) on different adsorbents. The isotherms were fitted using Freundlich model and the obtained parameters are listed in Table 3. The single-point sorption coefficients were also calculated and presented in Table 3. The gray arrows in the figures are used to guide the eye for comparing soils before and after organic matter removal.
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
simultaneously, we intend to summarize a general concept to understand the apparently increased and decreased antibiotic sorption by metal ions. 2. Experimental section
211
All the samples were kept in dark and were shaken in an air-bath shaker at 25 °C for 7 d. During this period, OFL was stable and no apparent degradation was observed. After the equilibration, all of the vials were centrifuged at 2500 g for 10 min and the supernatants were subjected to the quantification of both OFL and Cu(II).
2.1. Materials 2.3. Quantification of Cu(II) and OFL Two soil samples were collected in a mountainous area far from human activities. This area was located in Mengzi, Yunnan province, China, with geographical coordinates of 103° 47′ 24″N, 23° 24′ 36″E and altitude of 1874.40 m. Soil horizons A and B were collected and noted as MA and MB, respectively. The collected soil samples were freeze-dried, ground, and sieved through a 2-mm sieve. Plant residues were picked out manually. The organic matter in these two soils was removed to obtain the inorganic fractions. Briefly, humic and fulvic acids were removed three times using the mixture of 0.1 mol/L NaOH and 0.1 mol/L Na4P2O7 with the aqueous:solid ratio of 50:1 (v:w) (Pan et al., 2007). The residual soil particles were washed using distilled water, freeze-dried and then heated at 680 °C for 24 h with sufficient supply of air. These particles were noted as organic matter-removed soils (Hou et al., 2010). The organic matter-removed MA and MB were noted as PMA and PMB, respectively. The organic elements in these solid particles were analyzed using an elemental analyzer (MicroCube, Elementar, Germany). Their surface areas were quantified through the B.E.T. method using N2 as the adsorbate (Autosorb-1C, Quantachrome). The zeta potentials were analyzed on a Zetasizer (Malvern Instruments). The mineral contents of these two soils were analyzed through X-ray diffraction (XRD). Ofloxacin (OFL) was obtained from Bio Basic Inc. (Japan). It was dissolved in the background electrolyte solution of 0.02 mol/L NaCl (ionic strength adjuster) as 0.55 mmol/L stock solution. The solubility of OFL at pH 7.0 is 9.41 mmol/L according to our measurement. OFL is a zwitterionic compound with two pKas of 6.10 and 8.28 (Fig. 1). Cu(II) was applied as nitrate salt in this study. All the other chemicals were higher than analytical grade. The organic solvents used in this study were purchased from Merck Co. (Germany) with gradient grade for liquid chromatography.
OFL concentration in equilibrium solution was quantified by high performance liquid chromatography (HPLC) (Agilent Technologies 1200) equipped with a reversed-phase C8 column (5 μm, 4.6 × 150 mm) and an UV detector at 286 nm. The mobile phase was 10:90 (v:v) of acetonitrile and deionized water with 0.8% acetic acid. The flow rate was 1 mL/min. The retention time for OFL was 4.6–5.0 min. The concentration of Cu in the aqueous phase was quantified using a flame atomic absorption spectrometer (FAAS, Hitachi Z2000). The detection limits for OFL and Cu were 0.8 μmol/L and 1.0 μmol/L, respectively. All the samples were quantified using external standards. The co-adsorbate did not influence the quantification of the primary adsorbate in the co-sorption experiment. 2.4. Data analysis The solid phase concentrations were calculated based on mass balance: Se = (C0 − Ce) · (V/W), where C0 and Ce (mmol/L) are initially added and equilibrium aqueous phase concentrations, respectively. V/W (L/kg) is the aqueous:solid ratio. Sorption isotherms were fitted using the Freundlich models: logSe ¼ logK F þ n logC e
ð1Þ
where Se (mmol/kg) and Ce (mmol/L) are equilibrium solid-phase and aqueous-phase concentrations, respectively. KF (mmol1 − n Ln kg−1) is Freundlich sorption coefficient and n is the nonlinearity factor. Because the unit of KF is dependent on n value, KF values could not be directly compared. Thus, single-point sorption coefficients, Kd, were calculated at selected aqueous phase concentrations in this study: logK d ¼ logðSe =C e Þ ¼ logSe – logC e ¼ logK F þ ðn−1Þ logC e :
ð2Þ
2.2. Sorption experiments Individual sorption and co-sorption of Cu(II) and OFL were investigated for all the four solid particles using batch sorption experiment as described previously (Pan and Xing, 2010). Briefly, OFL (0.55 mmol/L) and cupric nitrate trihydrate (0.83 mmol/L) were separately prepared in 0.02 mol/L NaCl as stock solutions. All the sorption experiments were conducted in 20 mL glass vials with Teflon-lined screw caps. According to preliminary studies, the solution:solid ratios were 1000:1 (w/w) to ensure 20–80% sorption. The concentration of Cu(II) was in the range of 1.57 × 10−2–1.57 mmol/L and OFL concentration was in the range of 2.77 × 10−3–2.77 × 10−1 mmol/L. In the co-sorption study, both Cu(II) and OFL were separately used as primary adsorbates. The concentrations of the co-adsorbates were carefully selected to ensure significant sorption ratio (around 50%) and comparable sorption coefficients to the primary adsorbates, based on both preliminary experiments and calculation (through their sorption isotherms). When Cu(II) was used as the primary adsorbate, Cu(II) concentration was in the range of 1.57 × 10−2–1.57 mmol/L and OFL concentration was 1.38 × 10−1, 1.38 × 10−1, 1.38 × 10−2 or 2.76 × 10−2 mmol/L for MA, MB, PMA or PMB, respectively. When OFL was used as the primary adsorbate, OFL concentration was in the range of 2.77 × 10−3–2.77 × 10−1 mmol/L and Cu(II) concentrations were 6.29 × 10−1, 1.57 × 10−1, 7.87 × 10−2 or 7.87 × 10−2 mmol/L for MA, MB, PMA or PMB, respectively. Each treatment had two replicates. The pH values of the solutions were measured before and after the sorption equilibration.
Model fitting and statistical analysis were conducted using SigmaPlot 10.0 and SPSS 13.0, respectively. 3. Results and discussion 3.1. Cu(II) sorption on different particles The sorption of Cu(II) on MA was higher than those on MB, and both soils showed decreasing sorption after organic matter removal (Fig. 2). Previous studies have reported the importance of inorganic minerals, such as Mn and Fe oxides for Cu(II) sorption (Wang and Li, 2011). However, these inorganic components may be coated by organic matter in natural soils and thus not accessible for metal sorption. Therefore, for original soils, organic matter played a more important role in Cu(II) sorption than the mineral fractions. The oxygen-containing functional groups, such as hydroxyl and carboxyl groups, are the main sites for Table 1 Selected properties of the solid particles used in this study. Elemental analyzer (%)
MA MB PMA PMB
C
H
O
N
S
11.9 2.9 0.1 0.1
2.4 1.9 0.4 0.4
24.3 14.4 3.7 3.7
0.93 0.32 0.1 0.1
0.09 0.09 0.6 0.2
SSA N2 (m2/g)
CEC cmol(+)/kg
pH
Zeta potential (mv)
22.6 65.8 10.1 22.5
19.9 17.5 11.0 11.5
6.9 6.4 7.5 6.9
−32.1 −30.2 −34.5 −29.8
212
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
7.5
7.5 Cu MA-Cu MB-Cu
6.5
Cu PMA-Cu PMB-Cu
7.0
pH
6.5 6.0
5.5
5.5 4.5
A
B
-2.5
-1.5
-0.5
0.5
5.0 -3.0
-2.0
logCeCu (mmol/L)
-1.0
0.0
logCeCu (mmol/L)
Fig. 3. pH variation of Cu(II) sorption in original (A) and organic matter-removed soils (B). Cu(II) sorption in MA and MB resulted in decreased pH in comparison to Cu(II) solution at the same aqueous Cu(II) concentration. This phenomenon was not observed for the soils after organic matter removal. Error bars stand for standard error of the triplicate measurements.
Cu(II) sorption (Moon and Peacock, 2012; Reddy et al., 2012; Sun et al., 2012; F.J. Zhang et al., 2012). Cation exchange occurs between Cu(II) and proton in hydroxyl or carboxyl groups (Chen et al., 2009), which resulted in decreased pH (Wu et al., 2012). This pH decrease with Cu(II) sorption was observed in this study (Fig. 3). Cu(II) solution (without any particles) showed decreased pH with increasing Cu(II) concentration, possibly resulted from hydrolysis of Cu(II). Cu(II) sorption in MA and MB resulted in decreased pH in comparison to Cu(II) solution (paired t-test at P b 0.05 grouped by Cu(II) initial concentration). MA showed greater decrease of pH values, which might be attributed to its more organic oxygen content (more exchangeable protons) and thus higher cation exchange capacity for Cu(II) (Chen et al., 2009).
This pH decrease was not observed for soil samples after organic matter removal, as suggested by the comparable pH values to Cu(II) solution. The higher sorption of Cu(II) by PMA may be related with its lower zeta potential and thus stronger electrostatic attraction to Cu (II) than PMB (Xu and Zhao, 2013). The lower zeta potential (P b 0.05) of PMA may be resulted from its higher quartz content. The Si hydroxides gen2− erally exist as HSiO− 3 or SiO3 after dissociation. As a result, the colloidal particle of Si hydroxide is negatively charged, and the higher the solution pH, the more negatively charged the Si hydroxide. The pH of PMA (pH 7.45) was higher than PMB (pH = 6.90) (Table 1), suggesting more negatively charged particle surface. Although the SSA for MB which measured by N2 was 3 times of that for MA
logSeCu (mmol/kg)
2.8
2.4
A
B
2.3
2.0
1.8
1.6
1.3
1.2
0.8
-2.5
-1.5
-0.5
0.5
0.8 -3.0
-2.0
logCeCu (mmol/L) MA-Cu
PMA-Cu
MB-Cu
PMB-Cu
MA-Cu OFL
PMA-Cu OFL
MB-Cu OFL
PMB-Cu OFL
0.0
2.3
2.5
logSeOFL (mmol/kg)
-1.0
logCeCu (mmol/L)
C
D
2.0 1.8
1.5 1.0
1.3
0.5 0.0 -3.0
-2.0
-1.0
0.0
0.8
-2.5
-1.5
-0.5
logCeOFL(mmol/L)
logCeOFL(mmol/L) MA-OFL
PMA-OFL
MB-OFL
PMB-OFL
MA-OFL Cu
PMA-OFL Cu
MB-OFL Cu
PMB-OFL Cu
Fig. 4. Cu(II) sorption as affected by OFL (A and B) and OFL sorption as affected by Cu(II) (C and D). Soils before and after organic matter removal were presented in the same panel for a better comparison. The sorption generally increased in the presence of co-adsorbate for original soils, but decreased for organic matter-removed soils.
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
5
5
logKd (L/kg)
KdCu
KdCu
OFL
Kd
KdOFL
4
4
3 3
A
B
-3.0
-2.0
-1.0
0.0
2 -3.0
-2.0
-1.0
0.0
logCeCu (mmol/L)
logCeCu (mmol/L) 4.5
logKd (L/kg)
213
5
4.0
KdOFL
KdOFL
KdCu
KdCu 4
3.5
C 3.0 -3
-2
3 -3
-1
D -2
-1
logCeOFL (mmol/L)
logCeOFL (mmol/L)
Fig. 5. The sorption coefficients, Kd, of the primary (solid symbols) and co-adsorbate (open symbols) in binary sorption systems of original soils. Panels A and B present soils MA (A) and MB (B) when Cu(II) was the primary adsorbate, while panels C and D present soils MA (C) and MB (D) when OFL was the primary adsorbate. The gray bars indicate the range of Kd values for the co-adsorbates without the primary adsorbates.
(Table 1), the sorption of Cu(II) in MA was much higher (t-test for K d values at P b 0.05). This difference may suggest that surface charge of the particles played a more important role than the exposed surface area for Cu(II) sorption.
logKd (L/kg)
3.5
3.2. OFL sorption on different particles OFL sorption in two original soils was comparable, with higher sorption in MB at relatively low OFL concentrations. Previous studies have
3.5 3.0
3.0 2.5 2.5
KdCu
KdCu
2.0
KdOFL 2.0 -2.5
-2.0
A -1.5
-1.0
-0.5
1.5 -3.0
logCeCu (mmol/L)
-2.5
B -2.0
-1.5
-1.0
logCeCu (mmol/L)
3.5
logKd (L/kg)
KdOFL
4.0
3.0 3.0 2.5
2.0 -3.0
KdOFL KdCu
2.0
C -2.5
-2.0
logCeOFL (mmol/L)
-1.5
-2.5
KdOFL KdCu -2.0
D -1.5
-1.0
logCeOFL (mmol/L)
Fig. 6. The sorption coefficients, Kd, of the primary (solid symbols) and the co-adsorbate (open symbols) in binary sorption systems of soils after organic matter removal. Panels A and B present soils PMA (A) and PMB (B) when Cu(II) was the primary adsorbate, while panels C and D present soils PMA (C) and PMB (D) when OFL was the primary adsorbate. The gray bars indicate the range of Kd values for the co-adsorbates without the primary adsorbates.
214
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
Table 2 The mineral contents of MA and MB. %
Quartz
Muscovite
Kaolinite
Montmorillonite
Anatase
Hematite
Graphite
MA MB
32.1 28.4
18.3 14.1
23.4 26.1
12.8 22.5
1.0 1.0
5.1 5.0
7.3 2.9
suggested that the organic matter coating may compete with ionic organic chemicals for sorption sites on mineral particles (Hou et al., 2010). However, it is also reported that OFL could be adsorbed by organic matters by cation-bridging, electrostatic interaction and H-bond (Gu et al., 2007; Agarwal et al., 2008). The diffusion of OFL through organic matter and partitioning of OFL in an organic matter matrix may not be important for the apparent OFL sorption, as suggested by the comparable sorption on MA and MB. The slightly higher sorption of MB at low OFL concentrations may be resulted from the lower pH of MB, which enabled electrostatic attraction between positively charged OFL molecules (35% of OFL+ vs. b1% of OFL− at pH 6.4) and negatively charged soil particles (zeta potential of −30.24 mV). OFL sorption on organic matter-removed soils showed significant difference between two soils (P b 0.05), with PMB sorption up to one order of magnitude higher than PMA. The higher montmorillonite content of PMB may be the reason of its larger surface area and higher OFL sorption (Table 2). The interlayers of montmorillonite provided important sorption sites for IOCs with comparable sizes to OFL, such as oxytetracycline (Aristilde et al., 2013) and ciprofloxacin (Wang et al., 2010). This sorption region was not important for OFL on original soils, because of the organic matter-coating and thus blocking of montmorillonite interlayers. Montmorillonite interlayer sorption was not included in Cu(II) sorption discussion, because of the dominant role of electrostatic interactions. It should also be noted that the pH of PMA was higher than that of PMB (7.45 over 6.90). Thus, electrostatic repulsion between negatively charged OFL and solid particles could not be excluded as an explanation for the lower sorption of OFL in PMA. 3.3. Cosorption of Cu(II) and OFL Fig. 4 presents the sorption isotherms of Cu(II) or OFL in the presence of co-adsorbates. The presence of co-adsorbates increased the sorption of the primary solute in both original soils, but decreased in organic matterremoved soils. As discussed earlier, combining co-adsorbate sorption information will narrow down the possibility of the mechanisms that lead to the change of antibiotic or metal ion sorption. The sorption of co-adsorbates was presented in Figs. 5 and 6 to compare with the sorption of the primary adsorbate. In these figures, the sorption coefficients of the primary adsorbates, noted as Kpri d (L/kg), decreased with their increased aqueous phase concentration, Cpri e (mmol/L), which is consistent with the significant nonlinear sorption as suggested by n values much lower than 1 (Table 3). For the original soils, the Kd values of co-adsorbates (Kco d ) generally increased and then decreased (Fig. 5). For example, the Kd values of OFL
(co-adsorbate) in Cu(II) (primary adsorbate) sorption system increased up to one order of magnitude with increased Cu(II) concentration compared with the Kd values of OFL alone (noted as gray bars in Fig. 5A). When OFL Kd values increased to be comparable with Cu(II) Kd values, OFL sorption started to decrease. However, at Cu(II) concentration where OFL Kd values started to decrease (logCCu e = −2.0), Cu(II) sorption was still increased (Fig. 4A). Similar Kd variation for co-adsorbates was noted in the other three binary sorption systems (Fig. 5B, C, and D). However, these increasing followed by decreasing Kco d values were not observed for soils after organic matter removal. Although the variation of Kco d values was large, a decreasing trend could be observed for these values in PMA and PMB (Fig. 6). However, it should be noted that at low primary adsorbate concentrations, co-adsorbate sorption was higher than that of the co-adsorbate alone (as noted by the gray bars in Fig. 6 when the concentration of the primary adsorbate was pri zero). We thus predict that an increasing zone of Kco d vs. Ce could be ob. This experiment unfortunately could not be repeated served at low Cpri e because of the detection limits. Previous studies have suggested that adsorption of Cu(II) may neutralize the negative charge of organic matter, which will result in the coagulation of organic matter and thus shrink its physical conformation (Pei et al., 2014). This process may partly cause the decreased OFL sorption as presented in Fig. 5A and B at higher Cu(II) concentrations, but could not support the decreased Cu(II) adsorption at high OFL concentrations (Fig. 5C and D). Since both promoted and inhibited sorptions were observed in soils before and after organic matter removal, we believe that this observation was independent of organic matter content, and thus a generally applicable concept of the co-sorption system should be summarized. For the same soil (before and after organic co matter removal), the Kpri d values (x axis) were plotted against Kd values (y axis). The increased aqueous concentration of the primary adsorbate resulted in the decreased Kpri d , as guided by the gray arrows in Fig. 7. The pri pri pri Kco d values increased with decreased Kd (increased Ce ). When Kd and co co Kd were comparable, the Kd values started to decrease. It was reported previously that the adsorbate preferentially occupied sorption sites with high sorption energies, and then spread to sites with low sorption energies (D. Zhang et al., 2012). The primary adsorbates on high-energy sorption sites could not be replaced by the co-adsorbates, and thus the apparent sorption showed no impact (if they occupy sorption sites with different properties) or promoted sorption (if the adsorbed molecules provide additional sorption sites for the newly introduced ones, as for the case of OFL–Cu(II) co-sorption). When the primary adsorbate spread to sorption sites with low sorption energies, the competition with co-adsorbates may occur. The competitive adsorption was generally
Table 3 The fitted parameters of the sorption isotherms and the calculated single-point sorption coefficients. logKF (mmol1 − n Ln kg−1)
MA-Cu MB-Cu PMA-Cu PMB-Cu MA-OFL MB-OFL PMA-OFL PMB-OFL
2.59 2.16 1.73 1.56 2.35 2.21 1.79 1.56
± ± ± ± ± ± ± ±
0.02 0.04 0.05 0.06 0.01 0.02 0.15 0.06
r2adj
n
0.14 0.25 0.17 0.19 0.34 0.20 0.58 0.17
± ± ± ± ± ± ± ±
0.02 0.02 0.04 0.04 0.01 0.01 0.06 0.03
0.771 0.895 0.729 0.681 0.993 0.957 0.839 0.642
logKd (L/kg) at logCe = (mmol/L) −2.5
−2.0
−1.5
−1.0
3.87 3.28 2.98 2.77 3.35 3.41 2.42 2.80
3.45 2.91 2.57 2.37
4.02 4.21 2.84 3.62
4.30 3.66 3.40 3.18 3.68 3.81 2.63 3.21
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
4
logKd OFL (L/kg)
logKd OFL (L/kg)
5
MA-Cu OFL
4
PMA-Cu OFL
3
3 MB-Cu OFL
A 2 2
3
4
4
5
1
2
logKd Cu (L/kg)
logKd Cu (L/kg)
4
5
4
PMA-OFL Cu
3
3
2 MB-OFL Cu
C 3
3
logKd Cu (L/kg)
MA-OFL Cu
2
B
PMB-Cu OFL
2
logKd Cu (L/kg)
2
215
4
logKd OFL (L/kg)
5
1
D
PMB-OFL Cu
2
3
4
5
logKd OFL (L/kg)
Fig. 7. The comparison of sorption coefficients between primary and co-adsorbates. Panels A and B present Kd values when Cu(II) was the primary adsorbate, while panels C and D present Kd values when OFL was the primary adsorbate. The soils before and after organic matter removal were included in the same panel for comparison. The gray arrows indicate the increase of the primary adsorbate concentration.
observed for soils after organic matter removal, which may be understood from the fact that both OFL and Cu(II) occupy relatively low and similar sorption sites. It is now important to identify the possible sorption sites for OFL and Cu(II). The initial increased sorption for both primary and co-adsorbates should be related with their unique and strong sorption sites. Based on literature summary, it is clear that antibiotics could be adsorbed through hydrophobic effect (Zhang et al., 2010), while Cu(II) could form inner sphere complex with the adsorbents (Schlegel and Manceau, 2013). The successive sorption of Cu(II) on the adsorbed OFL through complexation (Pan et al., 2012), or the sorption of OFL on the adsorbed Cu(II) through cation bridging (Jia et al., 2008), resulted in their promoted sorption. This process occurs in the whole concentration span of the adsorbates. With the increase of adsorbate concentration, OFL and Cu(II) spread to other sorption sites that may overlap. For example, both OFL and Cu(II) could be adsorbed through cation exchange and electrostatic attraction (Wang et al., 2010; Kahle and Stamm, 2007). When the overlapping of OFL and Cu(II) on these sorption sites overwhelmed the impact of the promoted sorption, the apparent competitive sorption was observed. We speculate that the competition may occur between OFL and Cu(II), as well as OFL–Cu(II) complex and OFL or Cu(II). As we can compare in Figs. 4A and 5A, at logCCu e higher than − 2.0, OFL sorption was decreased (Fig. 5A), while Cu(II) sorption was still increased (Fig. 4A). This may suggest the competition between Cu and OFL–Cu(II) complex. The excessive Cu(II) or OFL in the aqueous phase may strip OFL or Cu(II) off from the solid particles through their complexation (Macias et al., 2001), which resulted in the observed competitive sorption. If this is the case, the decrease of Cu(II) and OFL should be concurrent. Clearly, this was not the case in this study. The sorption of co-adsorbates decreased at high primary adsorbate concentrations (Fig. 5), while the sorption of the primary adsorbates was always increased (Fig. 4). This study emphasizes that the promoted sorption at relatively low concentration (on the sites with high sorption energies) and inhibited sorption at relative high concentration could be a general process in OFL–Cu(II) co-sorption system. The single-concentration experimental
design and the experimental design without considering the sorption of co-adsorbate may have missed the key information of their cosorption. For example, Jia et al. (2008a) observed increased TC sorption on WS, while decreased TC sorption on RS. The higher sorption of Cu(II) on WS than RS suggested that WS may have more abundant high energy sorption sites for Cu(II). Thus, the competition between TC and Cu(II) on WS was expected at higher Cu(II) concentrations than RS. Without investigation in a large concentration range, conclusion on explicit increased or decreased antibiotic sorption by metal ions may not be accurate or complete.
4. Conclusions We observed both increased and decreased apparent sorptions of OFL and Cu(II) in their co-sorption system, and thus hypothesized that both increased and decreased antibiotic sorptions occur in a binary sorption system, and the apparent impact is determined by the combination of water chemistry conditions and solid particle properties. The spreading of adsorbates on heterogeneous sorption sites in solid particles enabled their promoted sorption on their unique and high-energy sorption sites, while the overlapping on relatively low-energy sorption sites resulted in competition. The balance between these two processes resulted in the apparently observed increased (at low concentration) and decreased (at high concentration) sorptions. We call the readers' attention that the turning point of decreased and increased sorptions relating to water chemistry conditions and solid particle properties is very important in predicting antibiotic–metal ion co-sorption. Singleconcentration experiments or the co-sorption experiments without presenting the sorption behavior of co-adsorbate provide only limited information.
Conflict of interest The authors declare that there are no conflicts of interest.
216
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
Acknowledgments This research was supported by the National Scientific Foundation of China (41222025, 41273138), the Program for New Century Excellent Talents in University, the Chinese Ministry of Education, and the Recruitment Program of Highly-Qualified Scholars in Yunnan (2010CI109). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.02.041. References Agarwal SP, Anwer MK, Aqil M. Complexation of furosemide with fulvic acid extracted from shilajit: a novel approach. Drug Dev Ind Pharm 2008;34:506–11. Aristilde L, Lanson B, Charlet L. Interstratification patterns from the pH-dependent intercalation of a tetracycline antibiotic within montmorillonite layers. Langmuir 2013;29: 4492–501. Bai Y, Wu F, Liu C, Guo J, Fu P, Li W, et al. Interaction between carbamazepine and humic substances: a fluorescence spectroscopy study. Environ Toxicol Chem 2008;27:95–102. Chee-Sanford JC, Mackie RI, Koike S, Krapac IJ, Lin YF, Yannarell AC, et al. Fate and transport of antibiotic residues and antibiotic resistance genes following land application of manure waste. J Environ Qual 2009;38:1086–108. Chen G-C, Shan X-Q, Wang Y-S, Wen B, Pei Z-G, Xie Y-N, et al. Adsorption of 2, 4, 6-trichlorophenol by multi-walled carbon nanotubes as affected by Cu (II). Water Res 2009;43:2409–18. Gu C, Karthikeyan KG. Interaction of tetracycline with aluminum and iron hydrous oxides. Environ Sci Technol 2005;39:2660–7. Gu C, Karthikeyan KG, Sibley SD, Pedersen JA. Complexation of the antibiotic tetracycline with humic acid. Chemosphere 2007;66:1494–501. Hou J, Pan B, Niu X, Chen J, Xing B. Sulfamethoxazole sorption by sediment fractions in comparison to pyrene and bisphenol A. Environ Pollut 2010;158:2826–32. Jia D-A, Zhou D-M, Wang Y-J, Zhu H-W, Chen J-L. Adsorption and cosorption of Cu (II) and tetracycline on two soils with different characteristics. Geoderma 2008;146:224–30. Kahle M, Stamm C. Sorption of the veterinary antimicrobial sulfathiazole to organic materials of different origin. Environ Sci Technol 2007;41:132–8. Karthikeyan KG, Meyer MT. Occurrence of antibiotics in wastewater treatment facilities in Wisconsin, USA. Sci Total Environ 2006;361:196–207. Li JH, Zhou BX, Shao JH, Yang QF, Liu YQ, Cai WM. Influence of the presence of heavy metals and surface-active compounds on the sorption of bisphenol A to sediment. Chemosphere 2007;68:1298–303. Macias B, Villa MV, Rubio I, Castineiras A, Borras J. Complexes of Ni(II) and Cu(II) with ofloxacin crystal structure of a new Cu(II) ofloxacin complex. J Inorg Biochem 2001;84:163–70. Moon EM, Peacock CL. Adsorption of Cu(II) to ferrihydrite and ferrihydrite-bacteria composites: importance of the carboxyl group for Cu mobility in natural environments. Geochim Cosmochim Acta 2012;92:203–19. Pan B, Xing BS. Competitive and complementary adsorption of bisphenol A and 17 alpha-ethinyl estradiol on carbon nanomaterials. J Agric Food Chem 2010;58: 8338–43.
Pan B, Xing B, Liu W, Xing G, Tao S. Investigating interactions of phenanthrene with dissolved organic matter: limitations of Stern–Volmer plot. Chemosphere 2007;69: 1555–62. Pan B, Ning P, Xing B. Part V—sorption of pharmaceuticals and personal care products. Environ Sci Pollut Res 2009;16:106–16. Pan B, Han XJ, Wu M, Peng HB, Zhang D, Li H, et al. Temperature dependence of ofloxacin fluorescence quenching and complexation by Cu(II). Environ Pollut 2012;171: 168–73. Pei Z, Shan XQ, Kong J, Wen B, Owens G. Coadsorption of ciprofloxacin and Cu (II) on montmorillonite and kaolinite as affected by solution pH. Environ Sci Technol 2009;44:915–20. Pei ZG, Shan XQ, Zhang SZ, Kong JJ, Wen B, Zhang J, et al. Insight to ternary complexes of co-adsorption of norfloxacin and Cu (II) onto montmorillonite at different pH using EXAFS. J Hazard Mater 2011;186:842–8. Pei Z, Yang S, Li L, Li C, Zhang SZ, Shan XQ, et al. Effects of copper and aluminum on the adsorption of sulfathiazole and tylosin on peat and soil. Environ Pollut 2014;184: 579–85. Reddy DHK, Lee SM, Seshaiah K. Biosorption of toxic heavy metal ions from water environment using honeycomb biomass—an industrial waste material. Water Air Soil Pollut 2012;223:5967–82. Schlegel ML, Manceau A. Binding mechanism of Cu(II) at the clay–water interface by powder and polarized EXAFS spectroscopy. Geochim Cosmochim Acta 2013;113: 113–24. Seifrtova M, Pena A, Lino CM, Solich P. Determination of fluoroquinolone antibiotics in hospital and municipal wastewaters in Coimbra by liquid chromatography with a monolithic column and fluorescence detection. Anal Bioanal Chem 2008;391: 799–805. Seveno NA, Kallifidas D, Smalla K, van Elsas JD, Collard JM, Karagouni AD, et al. Occurrence and reservoirs of antibiotic resistance genes in the environment. Rev Med Microbiol 2002;13:15–27. Sun WL, Xia J, Li S, Sun F. Effect of natural organic matter (NOM) on Cu(II) adsorption by multi-walled carbon nanotubes: relationship with NOM properties. Chem Eng J 2012;200:627–36. Ter Laak TL, Gebbink WA, Tolls J. The effect of pH and ionic strength on the sorption of sulfachloropyridazine, tylosin, and oxytetracycline to soil. Environ Toxicol Chem 2006;25:904–11. Wang XL, Li Y. Measurement of Cu and Zn adsorption onto surficial sediment components: new evidence for less importance of clay minerals. J Hazard Mater 2011;189:719–23. Wang CJ, Li ZH, Jiang WT, Jean JS, Liu CC. Cation exchange interaction between antibiotic ciprofloxacin and montmorillonite. J Hazard Mater 2010;183:309–14. Wu D, Pan B, Wu M, Peng HB, Zhang D, Xing BS. Coadsorption of Cu and sulfamethoxazole on hydroxylized and graphitized carbon nanotubes. Sci Total Environ 2012;427: 247–52. Xu RK, Zhao AZ. Effect of biochars on adsorption of Cu(II), Pb(II) and Cd(II) by three variable charge soils from southern China. Environ Sci Pollut Res 2013;20:8491–501. Zhang D, Pan B, Zhang H, Ning P, Xing B. Contribution of different sulfamethoxazole species to their overall adsorption on functionalized carbon nanotubes. Environ Sci Technol 2010;44:3806–11. Zhang D, Pan B, Wu M, Zhang H, Peng H, Ning P, et al. Cosorption of organic chemicals with different properties: their shared and different sorption sites. Environ Pollut 2012a;160:178–84. Zhang FJ, Ou XX, Chen S, Ran CQ, Quan X. Competitive adsorption and desorption of copper and lead in some soil of North China. Front Environ Sci Eng 2012b;6:484–92. Zhou JL. Sorption and remobilization behavior of 4-tert-octylphenol in aquatic systems. Environ Sci Technol 2006;40:2225–34.
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Co-sorption of ofloxacin and Cu(II) in soils before and after organic matter removal Di Wu a, Hao Li a, Shaohua Liao a, Xiaolong Sun b, Hongbo Peng a, Di Zhang a, Bo Pan a,⁎ a b
Faculty of Environmental Science & Engineering, Kunming University of Science & Technology, Kunming 650500, China College of Environmental Science and Engineering, Southwest Forestry University, Kunming 650224, China
H I G H L I G H T S • • • • •
Sorption of the co-adsorbate increased at low primary adsorbate concentration. Sorption of the co-adsorbate decreased at high primary adsorbate concentration. Complexation of the adsorbate by the adsorbed co-adsorbate increased sorption. Overlapping of the sorption sites decreased the sorption. Apparently increased or decreased sorption may be only a part of the whole picture.
a r t i c l e
i n f o
Article history: Received 17 December 2013 Received in revised form 10 February 2014 Accepted 10 February 2014 Available online 2 March 2014 Keywords: Binary sorption Cation bridge Competitive sorption Complementary sorption Complexation
a b s t r a c t Various mechanisms play roles simultaneously for antibiotic sorption on solid particles. Previous studies simply emphasized mechanisms that match the increased or decreased antibiotic sorption by metal ions, without a general concept including these diverse mechanisms in their co-sorption. We observed both increased and decreased OFL and Cu(II) sorption in their co-sorption system. The comparison of the sorption coefficients of primary adsorco pri bate (Kpri d ) and co-adsorbate (Kd ) suggested that enhanced sorption occurred at high Kd region (low primary was decreased to a certain value deadsorbate concentration). Competitive sorption was observed when Kpri d pending on solid particle properties. We thus summarized that if the adsorbates were introduced with low concentrations, OFL (such as hydrophobic region in solid particles) and Cu(II) (such as inner-sphere complexation sites) occupied their unique high-energy sorption sites. Cu(II) complexed with the adsorbed OFL, and OFL bridged by the adsorbed Cu(II) promoted the sorption for both chemicals. With the increased concentrations, the adsorbates spread to some common sorption sites with low sorption energy, such as cation exchange and electrostatic attraction region. The overlapping of Cu(II) and OFL on these sorption sites resulted in competitive sorption at high concentrations. The previously reported apparently increased or decreased sorption in antibiotic– metal ion co-sorption system may be only a part of the whole picture. Extended study on the turning point of decreased and increased sorption relating to water chemistry conditions and solid particle properties will provide more useful information to predict antibiotic–metal ion co-sorption. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Antibiotics are widely used in medical therapeutic for humans, growth promotion and disease treatment for animals. With decades of massive application, antibiotics are now detected in surface water, ground water, waste water, sewage, soils and sediments (Wu et al., 2012). Previous studies indicated that the continuous input and wide presence of antibiotics in terrestrial system provided an acclimatizing environment for microorganisms, which in turn promoted the development of drug resistance genes through mutation or gene transfer ⁎ Corresponding author. Tel./fax: +86 871 65170906. E-mail address: [email protected] (B. Pan).
http://dx.doi.org/10.1016/j.scitotenv.2014.02.041 0048-9697/© 2014 Elsevier B.V. All rights reserved.
(Seveno et al., 2002; Chee-Sanford et al., 2009). The failure of medical functioning of antibiotics is frequently reported. Antibiotic environmental behavior is the primary process controlling their risks, and thus attracted lots of academic attention (Pan et al., 2009). It was previously pointed out that metal ions could interact with the various functional groups in antibiotics, which altered the sorption behavior and risk of antibiotics. However, various, sometimes controversial results were reported in literature, including both increased and decreased antibiotic adsorption by metal ions. All these impacts were provided well with theoretical explanations. For example, metal ions increased the apparent sorption of antibiotics through electrostatic attraction (Kahle and Stamm, 2007), salting out effects (Zhou, 2006; Li et al., 2007), or cation bridging (Jia et al., 2008). On the other hand, metal
210
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
6.10
Distribution of OFL species
A 8.28 1.0
B
0.8 +
OFL OFL± OFL-
0.6 0.4 0.2 0.0 4.0
5.0
6.0
7.0
8.0
9.0
10.0
pH Fig. 1. Dissociation of ofloxacin (A) and distribution of cationic (OFL+), zwitterionic (OFL±0), and anionic (OFL−) ofloxacin in aqueous solution as a function of pH (pKa1 = 6.10 and pKa2 = 8.28) (B).
ions decreased antibiotic sorption through competition (Ter Laak et al., 2006; Bai et al., 2008) or outer-sphere complexation (Gu and Karthikeyan, 2005). However, the emphasis on one or several mechanisms does not necessarily exclude the other(s). Investigating the sorption variation of antibiotics at different pHs in the presence of metal ions provided useful information to understand the mechanisms that control their co-sorption. Researchers generally reported that metal ions decreased antibiotic sorption at relatively low pH (because of the competition between positively charged antibiotics and metal ions), but increased antibiotic sorption at relatively high pH (because of the cation bridging) (Pei et al., 2009, 2011; Jia et al., 2008). As discussed in previous studies, these above-mentioned interaction mechanisms may simultaneously play roles in antibiotic sorption (Zhang et al., 2010). The overwhelming of certain interaction mechanisms over others determines the apparent increasing or decreasing sorption. We thus hypothesize that both increased and decreased antibiotic sorption occur in a binary sorption system, and the apparent impact is determined by the combination of water chemistry conditions and solid particle properties. In the binary sorption system, investigators generally measured the concentration of primary adsorbate only, as summarized in Table S1 regarding literature results on co-sorption of ionic organic chemicals and metal ions. Thus, some key information may have been missed. For
example, the decreased antibiotic sorption with the addition of metal ions could be explained by competition between positively charged metal ions and antibiotics (Pei et al., 2011), or the blocking of antibiotic sorption sites nearby the complexes formed by the adsorbed metal ions (Chen et al., 2009). Clearly, the former will result in decreased metal sorption, while the latter will not. Another example is that both cation bridging and the sorption of antibiotic–metal ion complexes could result in the increased sorption of antibiotics, which resulted in the similar increased sorption of antibiotics. However, cation bridging will not increase metal ion sorption, while the sorption of antibiotic–metal ion complexes may result in increased sorption of both antibiotics and metal ion. Providing metal sorption information in the binary sorption system will greatly facilitate the discussion on co-sorption mechanisms. According to the above discussion, we separately used antibiotics and metal ions as primary adsorbates. The sorption of the co-adsorbates was also determined and analyzed together with the sorption of the primary adsorbates. Ofloxacin (OFL) was chosen as the model antibiotic chemical because of its massive application and wide presence in the environment (Seifrtova et al., 2008; Karthikeyan and Meyer, 2006). Cu(II) was selected as a representative metal ion because of its strong complexation property (Macias et al., 2001). Soil samples before and after organic matter removal were used as adsorbents to include particles with a wide span of sorption strength. With the knowledge that different mechanisms play roles
2.5
A
logSeOFL (mmol/kg)
logSeCu (mmol/kg)
3.0 2.5 2.0 1.5 1.0
-2.5
-1.5
-0.5
logCeCu (mmol/L)
0.5
B 2.0 1.5 1.0
MA PMA MB PMB
0.5 0.0 -3.0
-2.5
-2.0
-1.5
-1.0
-0.5
logCeOFL (mmol/L)
Fig. 2. The sorption isotherms of Cu(II) (A) and OFL (B) on different adsorbents. The isotherms were fitted using Freundlich model and the obtained parameters are listed in Table 3. The single-point sorption coefficients were also calculated and presented in Table 3. The gray arrows in the figures are used to guide the eye for comparing soils before and after organic matter removal.
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
simultaneously, we intend to summarize a general concept to understand the apparently increased and decreased antibiotic sorption by metal ions. 2. Experimental section
211
All the samples were kept in dark and were shaken in an air-bath shaker at 25 °C for 7 d. During this period, OFL was stable and no apparent degradation was observed. After the equilibration, all of the vials were centrifuged at 2500 g for 10 min and the supernatants were subjected to the quantification of both OFL and Cu(II).
2.1. Materials 2.3. Quantification of Cu(II) and OFL Two soil samples were collected in a mountainous area far from human activities. This area was located in Mengzi, Yunnan province, China, with geographical coordinates of 103° 47′ 24″N, 23° 24′ 36″E and altitude of 1874.40 m. Soil horizons A and B were collected and noted as MA and MB, respectively. The collected soil samples were freeze-dried, ground, and sieved through a 2-mm sieve. Plant residues were picked out manually. The organic matter in these two soils was removed to obtain the inorganic fractions. Briefly, humic and fulvic acids were removed three times using the mixture of 0.1 mol/L NaOH and 0.1 mol/L Na4P2O7 with the aqueous:solid ratio of 50:1 (v:w) (Pan et al., 2007). The residual soil particles were washed using distilled water, freeze-dried and then heated at 680 °C for 24 h with sufficient supply of air. These particles were noted as organic matter-removed soils (Hou et al., 2010). The organic matter-removed MA and MB were noted as PMA and PMB, respectively. The organic elements in these solid particles were analyzed using an elemental analyzer (MicroCube, Elementar, Germany). Their surface areas were quantified through the B.E.T. method using N2 as the adsorbate (Autosorb-1C, Quantachrome). The zeta potentials were analyzed on a Zetasizer (Malvern Instruments). The mineral contents of these two soils were analyzed through X-ray diffraction (XRD). Ofloxacin (OFL) was obtained from Bio Basic Inc. (Japan). It was dissolved in the background electrolyte solution of 0.02 mol/L NaCl (ionic strength adjuster) as 0.55 mmol/L stock solution. The solubility of OFL at pH 7.0 is 9.41 mmol/L according to our measurement. OFL is a zwitterionic compound with two pKas of 6.10 and 8.28 (Fig. 1). Cu(II) was applied as nitrate salt in this study. All the other chemicals were higher than analytical grade. The organic solvents used in this study were purchased from Merck Co. (Germany) with gradient grade for liquid chromatography.
OFL concentration in equilibrium solution was quantified by high performance liquid chromatography (HPLC) (Agilent Technologies 1200) equipped with a reversed-phase C8 column (5 μm, 4.6 × 150 mm) and an UV detector at 286 nm. The mobile phase was 10:90 (v:v) of acetonitrile and deionized water with 0.8% acetic acid. The flow rate was 1 mL/min. The retention time for OFL was 4.6–5.0 min. The concentration of Cu in the aqueous phase was quantified using a flame atomic absorption spectrometer (FAAS, Hitachi Z2000). The detection limits for OFL and Cu were 0.8 μmol/L and 1.0 μmol/L, respectively. All the samples were quantified using external standards. The co-adsorbate did not influence the quantification of the primary adsorbate in the co-sorption experiment. 2.4. Data analysis The solid phase concentrations were calculated based on mass balance: Se = (C0 − Ce) · (V/W), where C0 and Ce (mmol/L) are initially added and equilibrium aqueous phase concentrations, respectively. V/W (L/kg) is the aqueous:solid ratio. Sorption isotherms were fitted using the Freundlich models: logSe ¼ logK F þ n logC e
ð1Þ
where Se (mmol/kg) and Ce (mmol/L) are equilibrium solid-phase and aqueous-phase concentrations, respectively. KF (mmol1 − n Ln kg−1) is Freundlich sorption coefficient and n is the nonlinearity factor. Because the unit of KF is dependent on n value, KF values could not be directly compared. Thus, single-point sorption coefficients, Kd, were calculated at selected aqueous phase concentrations in this study: logK d ¼ logðSe =C e Þ ¼ logSe – logC e ¼ logK F þ ðn−1Þ logC e :
ð2Þ
2.2. Sorption experiments Individual sorption and co-sorption of Cu(II) and OFL were investigated for all the four solid particles using batch sorption experiment as described previously (Pan and Xing, 2010). Briefly, OFL (0.55 mmol/L) and cupric nitrate trihydrate (0.83 mmol/L) were separately prepared in 0.02 mol/L NaCl as stock solutions. All the sorption experiments were conducted in 20 mL glass vials with Teflon-lined screw caps. According to preliminary studies, the solution:solid ratios were 1000:1 (w/w) to ensure 20–80% sorption. The concentration of Cu(II) was in the range of 1.57 × 10−2–1.57 mmol/L and OFL concentration was in the range of 2.77 × 10−3–2.77 × 10−1 mmol/L. In the co-sorption study, both Cu(II) and OFL were separately used as primary adsorbates. The concentrations of the co-adsorbates were carefully selected to ensure significant sorption ratio (around 50%) and comparable sorption coefficients to the primary adsorbates, based on both preliminary experiments and calculation (through their sorption isotherms). When Cu(II) was used as the primary adsorbate, Cu(II) concentration was in the range of 1.57 × 10−2–1.57 mmol/L and OFL concentration was 1.38 × 10−1, 1.38 × 10−1, 1.38 × 10−2 or 2.76 × 10−2 mmol/L for MA, MB, PMA or PMB, respectively. When OFL was used as the primary adsorbate, OFL concentration was in the range of 2.77 × 10−3–2.77 × 10−1 mmol/L and Cu(II) concentrations were 6.29 × 10−1, 1.57 × 10−1, 7.87 × 10−2 or 7.87 × 10−2 mmol/L for MA, MB, PMA or PMB, respectively. Each treatment had two replicates. The pH values of the solutions were measured before and after the sorption equilibration.
Model fitting and statistical analysis were conducted using SigmaPlot 10.0 and SPSS 13.0, respectively. 3. Results and discussion 3.1. Cu(II) sorption on different particles The sorption of Cu(II) on MA was higher than those on MB, and both soils showed decreasing sorption after organic matter removal (Fig. 2). Previous studies have reported the importance of inorganic minerals, such as Mn and Fe oxides for Cu(II) sorption (Wang and Li, 2011). However, these inorganic components may be coated by organic matter in natural soils and thus not accessible for metal sorption. Therefore, for original soils, organic matter played a more important role in Cu(II) sorption than the mineral fractions. The oxygen-containing functional groups, such as hydroxyl and carboxyl groups, are the main sites for Table 1 Selected properties of the solid particles used in this study. Elemental analyzer (%)
MA MB PMA PMB
C
H
O
N
S
11.9 2.9 0.1 0.1
2.4 1.9 0.4 0.4
24.3 14.4 3.7 3.7
0.93 0.32 0.1 0.1
0.09 0.09 0.6 0.2
SSA N2 (m2/g)
CEC cmol(+)/kg
pH
Zeta potential (mv)
22.6 65.8 10.1 22.5
19.9 17.5 11.0 11.5
6.9 6.4 7.5 6.9
−32.1 −30.2 −34.5 −29.8
212
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
7.5
7.5 Cu MA-Cu MB-Cu
6.5
Cu PMA-Cu PMB-Cu
7.0
pH
6.5 6.0
5.5
5.5 4.5
A
B
-2.5
-1.5
-0.5
0.5
5.0 -3.0
-2.0
logCeCu (mmol/L)
-1.0
0.0
logCeCu (mmol/L)
Fig. 3. pH variation of Cu(II) sorption in original (A) and organic matter-removed soils (B). Cu(II) sorption in MA and MB resulted in decreased pH in comparison to Cu(II) solution at the same aqueous Cu(II) concentration. This phenomenon was not observed for the soils after organic matter removal. Error bars stand for standard error of the triplicate measurements.
Cu(II) sorption (Moon and Peacock, 2012; Reddy et al., 2012; Sun et al., 2012; F.J. Zhang et al., 2012). Cation exchange occurs between Cu(II) and proton in hydroxyl or carboxyl groups (Chen et al., 2009), which resulted in decreased pH (Wu et al., 2012). This pH decrease with Cu(II) sorption was observed in this study (Fig. 3). Cu(II) solution (without any particles) showed decreased pH with increasing Cu(II) concentration, possibly resulted from hydrolysis of Cu(II). Cu(II) sorption in MA and MB resulted in decreased pH in comparison to Cu(II) solution (paired t-test at P b 0.05 grouped by Cu(II) initial concentration). MA showed greater decrease of pH values, which might be attributed to its more organic oxygen content (more exchangeable protons) and thus higher cation exchange capacity for Cu(II) (Chen et al., 2009).
This pH decrease was not observed for soil samples after organic matter removal, as suggested by the comparable pH values to Cu(II) solution. The higher sorption of Cu(II) by PMA may be related with its lower zeta potential and thus stronger electrostatic attraction to Cu (II) than PMB (Xu and Zhao, 2013). The lower zeta potential (P b 0.05) of PMA may be resulted from its higher quartz content. The Si hydroxides gen2− erally exist as HSiO− 3 or SiO3 after dissociation. As a result, the colloidal particle of Si hydroxide is negatively charged, and the higher the solution pH, the more negatively charged the Si hydroxide. The pH of PMA (pH 7.45) was higher than PMB (pH = 6.90) (Table 1), suggesting more negatively charged particle surface. Although the SSA for MB which measured by N2 was 3 times of that for MA
logSeCu (mmol/kg)
2.8
2.4
A
B
2.3
2.0
1.8
1.6
1.3
1.2
0.8
-2.5
-1.5
-0.5
0.5
0.8 -3.0
-2.0
logCeCu (mmol/L) MA-Cu
PMA-Cu
MB-Cu
PMB-Cu
MA-Cu OFL
PMA-Cu OFL
MB-Cu OFL
PMB-Cu OFL
0.0
2.3
2.5
logSeOFL (mmol/kg)
-1.0
logCeCu (mmol/L)
C
D
2.0 1.8
1.5 1.0
1.3
0.5 0.0 -3.0
-2.0
-1.0
0.0
0.8
-2.5
-1.5
-0.5
logCeOFL(mmol/L)
logCeOFL(mmol/L) MA-OFL
PMA-OFL
MB-OFL
PMB-OFL
MA-OFL Cu
PMA-OFL Cu
MB-OFL Cu
PMB-OFL Cu
Fig. 4. Cu(II) sorption as affected by OFL (A and B) and OFL sorption as affected by Cu(II) (C and D). Soils before and after organic matter removal were presented in the same panel for a better comparison. The sorption generally increased in the presence of co-adsorbate for original soils, but decreased for organic matter-removed soils.
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
5
5
logKd (L/kg)
KdCu
KdCu
OFL
Kd
KdOFL
4
4
3 3
A
B
-3.0
-2.0
-1.0
0.0
2 -3.0
-2.0
-1.0
0.0
logCeCu (mmol/L)
logCeCu (mmol/L) 4.5
logKd (L/kg)
213
5
4.0
KdOFL
KdOFL
KdCu
KdCu 4
3.5
C 3.0 -3
-2
3 -3
-1
D -2
-1
logCeOFL (mmol/L)
logCeOFL (mmol/L)
Fig. 5. The sorption coefficients, Kd, of the primary (solid symbols) and co-adsorbate (open symbols) in binary sorption systems of original soils. Panels A and B present soils MA (A) and MB (B) when Cu(II) was the primary adsorbate, while panels C and D present soils MA (C) and MB (D) when OFL was the primary adsorbate. The gray bars indicate the range of Kd values for the co-adsorbates without the primary adsorbates.
(Table 1), the sorption of Cu(II) in MA was much higher (t-test for K d values at P b 0.05). This difference may suggest that surface charge of the particles played a more important role than the exposed surface area for Cu(II) sorption.
logKd (L/kg)
3.5
3.2. OFL sorption on different particles OFL sorption in two original soils was comparable, with higher sorption in MB at relatively low OFL concentrations. Previous studies have
3.5 3.0
3.0 2.5 2.5
KdCu
KdCu
2.0
KdOFL 2.0 -2.5
-2.0
A -1.5
-1.0
-0.5
1.5 -3.0
logCeCu (mmol/L)
-2.5
B -2.0
-1.5
-1.0
logCeCu (mmol/L)
3.5
logKd (L/kg)
KdOFL
4.0
3.0 3.0 2.5
2.0 -3.0
KdOFL KdCu
2.0
C -2.5
-2.0
logCeOFL (mmol/L)
-1.5
-2.5
KdOFL KdCu -2.0
D -1.5
-1.0
logCeOFL (mmol/L)
Fig. 6. The sorption coefficients, Kd, of the primary (solid symbols) and the co-adsorbate (open symbols) in binary sorption systems of soils after organic matter removal. Panels A and B present soils PMA (A) and PMB (B) when Cu(II) was the primary adsorbate, while panels C and D present soils PMA (C) and PMB (D) when OFL was the primary adsorbate. The gray bars indicate the range of Kd values for the co-adsorbates without the primary adsorbates.
214
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
Table 2 The mineral contents of MA and MB. %
Quartz
Muscovite
Kaolinite
Montmorillonite
Anatase
Hematite
Graphite
MA MB
32.1 28.4
18.3 14.1
23.4 26.1
12.8 22.5
1.0 1.0
5.1 5.0
7.3 2.9
suggested that the organic matter coating may compete with ionic organic chemicals for sorption sites on mineral particles (Hou et al., 2010). However, it is also reported that OFL could be adsorbed by organic matters by cation-bridging, electrostatic interaction and H-bond (Gu et al., 2007; Agarwal et al., 2008). The diffusion of OFL through organic matter and partitioning of OFL in an organic matter matrix may not be important for the apparent OFL sorption, as suggested by the comparable sorption on MA and MB. The slightly higher sorption of MB at low OFL concentrations may be resulted from the lower pH of MB, which enabled electrostatic attraction between positively charged OFL molecules (35% of OFL+ vs. b1% of OFL− at pH 6.4) and negatively charged soil particles (zeta potential of −30.24 mV). OFL sorption on organic matter-removed soils showed significant difference between two soils (P b 0.05), with PMB sorption up to one order of magnitude higher than PMA. The higher montmorillonite content of PMB may be the reason of its larger surface area and higher OFL sorption (Table 2). The interlayers of montmorillonite provided important sorption sites for IOCs with comparable sizes to OFL, such as oxytetracycline (Aristilde et al., 2013) and ciprofloxacin (Wang et al., 2010). This sorption region was not important for OFL on original soils, because of the organic matter-coating and thus blocking of montmorillonite interlayers. Montmorillonite interlayer sorption was not included in Cu(II) sorption discussion, because of the dominant role of electrostatic interactions. It should also be noted that the pH of PMA was higher than that of PMB (7.45 over 6.90). Thus, electrostatic repulsion between negatively charged OFL and solid particles could not be excluded as an explanation for the lower sorption of OFL in PMA. 3.3. Cosorption of Cu(II) and OFL Fig. 4 presents the sorption isotherms of Cu(II) or OFL in the presence of co-adsorbates. The presence of co-adsorbates increased the sorption of the primary solute in both original soils, but decreased in organic matterremoved soils. As discussed earlier, combining co-adsorbate sorption information will narrow down the possibility of the mechanisms that lead to the change of antibiotic or metal ion sorption. The sorption of co-adsorbates was presented in Figs. 5 and 6 to compare with the sorption of the primary adsorbate. In these figures, the sorption coefficients of the primary adsorbates, noted as Kpri d (L/kg), decreased with their increased aqueous phase concentration, Cpri e (mmol/L), which is consistent with the significant nonlinear sorption as suggested by n values much lower than 1 (Table 3). For the original soils, the Kd values of co-adsorbates (Kco d ) generally increased and then decreased (Fig. 5). For example, the Kd values of OFL
(co-adsorbate) in Cu(II) (primary adsorbate) sorption system increased up to one order of magnitude with increased Cu(II) concentration compared with the Kd values of OFL alone (noted as gray bars in Fig. 5A). When OFL Kd values increased to be comparable with Cu(II) Kd values, OFL sorption started to decrease. However, at Cu(II) concentration where OFL Kd values started to decrease (logCCu e = −2.0), Cu(II) sorption was still increased (Fig. 4A). Similar Kd variation for co-adsorbates was noted in the other three binary sorption systems (Fig. 5B, C, and D). However, these increasing followed by decreasing Kco d values were not observed for soils after organic matter removal. Although the variation of Kco d values was large, a decreasing trend could be observed for these values in PMA and PMB (Fig. 6). However, it should be noted that at low primary adsorbate concentrations, co-adsorbate sorption was higher than that of the co-adsorbate alone (as noted by the gray bars in Fig. 6 when the concentration of the primary adsorbate was pri zero). We thus predict that an increasing zone of Kco d vs. Ce could be ob. This experiment unfortunately could not be repeated served at low Cpri e because of the detection limits. Previous studies have suggested that adsorption of Cu(II) may neutralize the negative charge of organic matter, which will result in the coagulation of organic matter and thus shrink its physical conformation (Pei et al., 2014). This process may partly cause the decreased OFL sorption as presented in Fig. 5A and B at higher Cu(II) concentrations, but could not support the decreased Cu(II) adsorption at high OFL concentrations (Fig. 5C and D). Since both promoted and inhibited sorptions were observed in soils before and after organic matter removal, we believe that this observation was independent of organic matter content, and thus a generally applicable concept of the co-sorption system should be summarized. For the same soil (before and after organic co matter removal), the Kpri d values (x axis) were plotted against Kd values (y axis). The increased aqueous concentration of the primary adsorbate resulted in the decreased Kpri d , as guided by the gray arrows in Fig. 7. The pri pri pri Kco d values increased with decreased Kd (increased Ce ). When Kd and co co Kd were comparable, the Kd values started to decrease. It was reported previously that the adsorbate preferentially occupied sorption sites with high sorption energies, and then spread to sites with low sorption energies (D. Zhang et al., 2012). The primary adsorbates on high-energy sorption sites could not be replaced by the co-adsorbates, and thus the apparent sorption showed no impact (if they occupy sorption sites with different properties) or promoted sorption (if the adsorbed molecules provide additional sorption sites for the newly introduced ones, as for the case of OFL–Cu(II) co-sorption). When the primary adsorbate spread to sorption sites with low sorption energies, the competition with co-adsorbates may occur. The competitive adsorption was generally
Table 3 The fitted parameters of the sorption isotherms and the calculated single-point sorption coefficients. logKF (mmol1 − n Ln kg−1)
MA-Cu MB-Cu PMA-Cu PMB-Cu MA-OFL MB-OFL PMA-OFL PMB-OFL
2.59 2.16 1.73 1.56 2.35 2.21 1.79 1.56
± ± ± ± ± ± ± ±
0.02 0.04 0.05 0.06 0.01 0.02 0.15 0.06
r2adj
n
0.14 0.25 0.17 0.19 0.34 0.20 0.58 0.17
± ± ± ± ± ± ± ±
0.02 0.02 0.04 0.04 0.01 0.01 0.06 0.03
0.771 0.895 0.729 0.681 0.993 0.957 0.839 0.642
logKd (L/kg) at logCe = (mmol/L) −2.5
−2.0
−1.5
−1.0
3.87 3.28 2.98 2.77 3.35 3.41 2.42 2.80
3.45 2.91 2.57 2.37
4.02 4.21 2.84 3.62
4.30 3.66 3.40 3.18 3.68 3.81 2.63 3.21
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
4
logKd OFL (L/kg)
logKd OFL (L/kg)
5
MA-Cu OFL
4
PMA-Cu OFL
3
3 MB-Cu OFL
A 2 2
3
4
4
5
1
2
logKd Cu (L/kg)
logKd Cu (L/kg)
4
5
4
PMA-OFL Cu
3
3
2 MB-OFL Cu
C 3
3
logKd Cu (L/kg)
MA-OFL Cu
2
B
PMB-Cu OFL
2
logKd Cu (L/kg)
2
215
4
logKd OFL (L/kg)
5
1
D
PMB-OFL Cu
2
3
4
5
logKd OFL (L/kg)
Fig. 7. The comparison of sorption coefficients between primary and co-adsorbates. Panels A and B present Kd values when Cu(II) was the primary adsorbate, while panels C and D present Kd values when OFL was the primary adsorbate. The soils before and after organic matter removal were included in the same panel for comparison. The gray arrows indicate the increase of the primary adsorbate concentration.
observed for soils after organic matter removal, which may be understood from the fact that both OFL and Cu(II) occupy relatively low and similar sorption sites. It is now important to identify the possible sorption sites for OFL and Cu(II). The initial increased sorption for both primary and co-adsorbates should be related with their unique and strong sorption sites. Based on literature summary, it is clear that antibiotics could be adsorbed through hydrophobic effect (Zhang et al., 2010), while Cu(II) could form inner sphere complex with the adsorbents (Schlegel and Manceau, 2013). The successive sorption of Cu(II) on the adsorbed OFL through complexation (Pan et al., 2012), or the sorption of OFL on the adsorbed Cu(II) through cation bridging (Jia et al., 2008), resulted in their promoted sorption. This process occurs in the whole concentration span of the adsorbates. With the increase of adsorbate concentration, OFL and Cu(II) spread to other sorption sites that may overlap. For example, both OFL and Cu(II) could be adsorbed through cation exchange and electrostatic attraction (Wang et al., 2010; Kahle and Stamm, 2007). When the overlapping of OFL and Cu(II) on these sorption sites overwhelmed the impact of the promoted sorption, the apparent competitive sorption was observed. We speculate that the competition may occur between OFL and Cu(II), as well as OFL–Cu(II) complex and OFL or Cu(II). As we can compare in Figs. 4A and 5A, at logCCu e higher than − 2.0, OFL sorption was decreased (Fig. 5A), while Cu(II) sorption was still increased (Fig. 4A). This may suggest the competition between Cu and OFL–Cu(II) complex. The excessive Cu(II) or OFL in the aqueous phase may strip OFL or Cu(II) off from the solid particles through their complexation (Macias et al., 2001), which resulted in the observed competitive sorption. If this is the case, the decrease of Cu(II) and OFL should be concurrent. Clearly, this was not the case in this study. The sorption of co-adsorbates decreased at high primary adsorbate concentrations (Fig. 5), while the sorption of the primary adsorbates was always increased (Fig. 4). This study emphasizes that the promoted sorption at relatively low concentration (on the sites with high sorption energies) and inhibited sorption at relative high concentration could be a general process in OFL–Cu(II) co-sorption system. The single-concentration experimental
design and the experimental design without considering the sorption of co-adsorbate may have missed the key information of their cosorption. For example, Jia et al. (2008a) observed increased TC sorption on WS, while decreased TC sorption on RS. The higher sorption of Cu(II) on WS than RS suggested that WS may have more abundant high energy sorption sites for Cu(II). Thus, the competition between TC and Cu(II) on WS was expected at higher Cu(II) concentrations than RS. Without investigation in a large concentration range, conclusion on explicit increased or decreased antibiotic sorption by metal ions may not be accurate or complete.
4. Conclusions We observed both increased and decreased apparent sorptions of OFL and Cu(II) in their co-sorption system, and thus hypothesized that both increased and decreased antibiotic sorptions occur in a binary sorption system, and the apparent impact is determined by the combination of water chemistry conditions and solid particle properties. The spreading of adsorbates on heterogeneous sorption sites in solid particles enabled their promoted sorption on their unique and high-energy sorption sites, while the overlapping on relatively low-energy sorption sites resulted in competition. The balance between these two processes resulted in the apparently observed increased (at low concentration) and decreased (at high concentration) sorptions. We call the readers' attention that the turning point of decreased and increased sorptions relating to water chemistry conditions and solid particle properties is very important in predicting antibiotic–metal ion co-sorption. Singleconcentration experiments or the co-sorption experiments without presenting the sorption behavior of co-adsorbate provide only limited information.
Conflict of interest The authors declare that there are no conflicts of interest.
216
D. Wu et al. / Science of the Total Environment 481 (2014) 209–216
Acknowledgments This research was supported by the National Scientific Foundation of China (41222025, 41273138), the Program for New Century Excellent Talents in University, the Chinese Ministry of Education, and the Recruitment Program of Highly-Qualified Scholars in Yunnan (2010CI109). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.02.041. References Agarwal SP, Anwer MK, Aqil M. Complexation of furosemide with fulvic acid extracted from shilajit: a novel approach. Drug Dev Ind Pharm 2008;34:506–11. Aristilde L, Lanson B, Charlet L. Interstratification patterns from the pH-dependent intercalation of a tetracycline antibiotic within montmorillonite layers. Langmuir 2013;29: 4492–501. Bai Y, Wu F, Liu C, Guo J, Fu P, Li W, et al. Interaction between carbamazepine and humic substances: a fluorescence spectroscopy study. Environ Toxicol Chem 2008;27:95–102. Chee-Sanford JC, Mackie RI, Koike S, Krapac IJ, Lin YF, Yannarell AC, et al. Fate and transport of antibiotic residues and antibiotic resistance genes following land application of manure waste. J Environ Qual 2009;38:1086–108. Chen G-C, Shan X-Q, Wang Y-S, Wen B, Pei Z-G, Xie Y-N, et al. Adsorption of 2, 4, 6-trichlorophenol by multi-walled carbon nanotubes as affected by Cu (II). Water Res 2009;43:2409–18. Gu C, Karthikeyan KG. Interaction of tetracycline with aluminum and iron hydrous oxides. Environ Sci Technol 2005;39:2660–7. Gu C, Karthikeyan KG, Sibley SD, Pedersen JA. Complexation of the antibiotic tetracycline with humic acid. Chemosphere 2007;66:1494–501. Hou J, Pan B, Niu X, Chen J, Xing B. Sulfamethoxazole sorption by sediment fractions in comparison to pyrene and bisphenol A. Environ Pollut 2010;158:2826–32. Jia D-A, Zhou D-M, Wang Y-J, Zhu H-W, Chen J-L. Adsorption and cosorption of Cu (II) and tetracycline on two soils with different characteristics. Geoderma 2008;146:224–30. Kahle M, Stamm C. Sorption of the veterinary antimicrobial sulfathiazole to organic materials of different origin. Environ Sci Technol 2007;41:132–8. Karthikeyan KG, Meyer MT. Occurrence of antibiotics in wastewater treatment facilities in Wisconsin, USA. Sci Total Environ 2006;361:196–207. Li JH, Zhou BX, Shao JH, Yang QF, Liu YQ, Cai WM. Influence of the presence of heavy metals and surface-active compounds on the sorption of bisphenol A to sediment. Chemosphere 2007;68:1298–303. Macias B, Villa MV, Rubio I, Castineiras A, Borras J. Complexes of Ni(II) and Cu(II) with ofloxacin crystal structure of a new Cu(II) ofloxacin complex. J Inorg Biochem 2001;84:163–70. Moon EM, Peacock CL. Adsorption of Cu(II) to ferrihydrite and ferrihydrite-bacteria composites: importance of the carboxyl group for Cu mobility in natural environments. Geochim Cosmochim Acta 2012;92:203–19. Pan B, Xing BS. Competitive and complementary adsorption of bisphenol A and 17 alpha-ethinyl estradiol on carbon nanomaterials. J Agric Food Chem 2010;58: 8338–43.
Pan B, Xing B, Liu W, Xing G, Tao S. Investigating interactions of phenanthrene with dissolved organic matter: limitations of Stern–Volmer plot. Chemosphere 2007;69: 1555–62. Pan B, Ning P, Xing B. Part V—sorption of pharmaceuticals and personal care products. Environ Sci Pollut Res 2009;16:106–16. Pan B, Han XJ, Wu M, Peng HB, Zhang D, Li H, et al. Temperature dependence of ofloxacin fluorescence quenching and complexation by Cu(II). Environ Pollut 2012;171: 168–73. Pei Z, Shan XQ, Kong J, Wen B, Owens G. Coadsorption of ciprofloxacin and Cu (II) on montmorillonite and kaolinite as affected by solution pH. Environ Sci Technol 2009;44:915–20. Pei ZG, Shan XQ, Zhang SZ, Kong JJ, Wen B, Zhang J, et al. Insight to ternary complexes of co-adsorption of norfloxacin and Cu (II) onto montmorillonite at different pH using EXAFS. J Hazard Mater 2011;186:842–8. Pei Z, Yang S, Li L, Li C, Zhang SZ, Shan XQ, et al. Effects of copper and aluminum on the adsorption of sulfathiazole and tylosin on peat and soil. Environ Pollut 2014;184: 579–85. Reddy DHK, Lee SM, Seshaiah K. Biosorption of toxic heavy metal ions from water environment using honeycomb biomass—an industrial waste material. Water Air Soil Pollut 2012;223:5967–82. Schlegel ML, Manceau A. Binding mechanism of Cu(II) at the clay–water interface by powder and polarized EXAFS spectroscopy. Geochim Cosmochim Acta 2013;113: 113–24. Seifrtova M, Pena A, Lino CM, Solich P. Determination of fluoroquinolone antibiotics in hospital and municipal wastewaters in Coimbra by liquid chromatography with a monolithic column and fluorescence detection. Anal Bioanal Chem 2008;391: 799–805. Seveno NA, Kallifidas D, Smalla K, van Elsas JD, Collard JM, Karagouni AD, et al. Occurrence and reservoirs of antibiotic resistance genes in the environment. Rev Med Microbiol 2002;13:15–27. Sun WL, Xia J, Li S, Sun F. Effect of natural organic matter (NOM) on Cu(II) adsorption by multi-walled carbon nanotubes: relationship with NOM properties. Chem Eng J 2012;200:627–36. Ter Laak TL, Gebbink WA, Tolls J. The effect of pH and ionic strength on the sorption of sulfachloropyridazine, tylosin, and oxytetracycline to soil. Environ Toxicol Chem 2006;25:904–11. Wang XL, Li Y. Measurement of Cu and Zn adsorption onto surficial sediment components: new evidence for less importance of clay minerals. J Hazard Mater 2011;189:719–23. Wang CJ, Li ZH, Jiang WT, Jean JS, Liu CC. Cation exchange interaction between antibiotic ciprofloxacin and montmorillonite. J Hazard Mater 2010;183:309–14. Wu D, Pan B, Wu M, Peng HB, Zhang D, Xing BS. Coadsorption of Cu and sulfamethoxazole on hydroxylized and graphitized carbon nanotubes. Sci Total Environ 2012;427: 247–52. Xu RK, Zhao AZ. Effect of biochars on adsorption of Cu(II), Pb(II) and Cd(II) by three variable charge soils from southern China. Environ Sci Pollut Res 2013;20:8491–501. Zhang D, Pan B, Zhang H, Ning P, Xing B. Contribution of different sulfamethoxazole species to their overall adsorption on functionalized carbon nanotubes. Environ Sci Technol 2010;44:3806–11. Zhang D, Pan B, Wu M, Zhang H, Peng H, Ning P, et al. Cosorption of organic chemicals with different properties: their shared and different sorption sites. Environ Pollut 2012a;160:178–84. Zhang FJ, Ou XX, Chen S, Ran CQ, Quan X. Competitive adsorption and desorption of copper and lead in some soil of North China. Front Environ Sci Eng 2012b;6:484–92. Zhou JL. Sorption and remobilization behavior of 4-tert-octylphenol in aquatic systems. Environ Sci Technol 2006;40:2225–34.
