Study of ciprofloxacin adsorption and regeneration of activated carbon prepared from Enteromorpha prolifera impregnated with H3PO4 and sodium benzenesulfonate

Ecotoxicology and Environmental Safety 139 (2017) 36–42

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv

Study of ciprofloxacin adsorption and regeneration of activated carbon prepared from Enteromorpha prolifera impregnated with H3PO4 and sodium benzenesulfonate

MARK

⁎

Man Wang, Gang Li, Lihui Huang , Jing Xue, Quan Liu, Nan Bao, Ji Huang Shandong Provincial Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, Jinan, Shandong 250100, China

A R T I C L E I N F O

A B S T R A C T

Keywords: Activated carbons Enteromorpha prolifera Sodium benzenesulfonate Regeneration

Activated carbons were derived from Enteromorpha prolifera immersed in H3PO4 solution or the H3PO4 solution mixed with sodium benzenesulfonate (SBS), producing AC and AC-SBS. NaOH solution was employed in regeneration of ciprofloxacin (CIP)-loaded AC and AC-SBS to obtain RAC and RAC-SBS. The properties of the original and regenerated activated carbons were characterized by thermo-gravimetric analysis (TGA), scanning electron microscopy (SEM), N2 adsorption/desorption isotherms and Fourier transform infrared spectroscopy (FTIR). Batched adsorption studies were carried out to compare CIP adsorption behaviors of the four carbons. The results suggested that the four samples exhibited higher proportions of mesopores and similar functional groups. Although AC displayed much higher specific surface area (SBET) (1045.79 m2/g) than AC-SBS (738.03 m2/g), its CIP adsorption capacity was much less than AC-SBS. The maximum adsorption capacity for AC, AC-SBS, RAC and RAC-SBS were found to be 250 mg/g, 286 mg/g, 233 mg/g and 256 mg/g, respectively, with the isotherms adhering to Langmuir isotherm model. The electrostatic attraction and cation exchange between CIP and the four carbons were the dominant adsorption mechanisms. Moreover, the thermodynamic parameters represented that the adsorption process had been confirmed to be a spontaneous and endothermic reaction.

1. Introduction Pharmaceutical antibiotics are produced in enormous quantities and heavily used in the farming industry as veterinary therapeutics and growth promoters for animals (Ji et al., 2011). A small portion of antibiotics are metabolized by human and animals, and most of them are eventually released into the aquatic environment (Hirsch et al., 1999; Kolpin et al., 2002), leading to microorganism antibiotic resistance and potential risks to ecosystem function (Schmitt et al., 2006; Storteboom et al., 2010). Ciprofloxacin (CIP), a kind of quinolone antibiotics, has been detected in sewage concentration typically < 1 μg/L, while higher concentration has been probed in effluent from hospitals (up to 150 μg/L) (Carmosini and Lee, 2009; Carabineiro et al., 2011). Even though CIP removal is of extreme importance, there is limited literature on it compared with other antibiotics (Sun et al., 2012a, 2012b). In this regard, it has been an emerging concern to develop efficient treatment technology for removal of CIP in recent years. Numerous techniques have been implemented to scavenge pollu-

⁎

tants from wastewater, including floatation, coagulation, photochemical degradation, chemical reduction, oxidation and microbial remediation (Chen et al., 2011). It has been demonstrated that these conventional methods may be subjected to low treatment efficiency and high operation expense. Therefore, it warrants development of simple, highefficiency and low-cost treatment methods to remove antibiotics from sewage water (Cho et al., 2011). Activated carbons are carbonaceous material with extremely high surface area, well-developed porous structure, and prominent adsorption capacity (Bautista-Toledo et al., 1994). Thus they provide promising potential for removal of organic and inorganic pollutants from domestic and industrial effluent (Popescu et al., 2003; Aktaş and Çeçen, 2006). Many chemical or physical treatment and surface modification methods have been introduced (Ahn et al., 2009) to enhance the capacity of activated carbon by employing a variety of materials such as surfactants (Moradi, 2014), oxidizing agents (Babel and Kurniawan, 2004) and metal salts (Chen et al., 2007). Among these methods, the utility of surfactant is preferred for the preparation of activated carbon since it involves simple blending with activated carbon. The surfactants

Corresponding author. E-mail address: [email protected] (L. Huang).

http://dx.doi.org/10.1016/j.ecoenv.2017.01.006 Received 14 October 2016; Received in revised form 3 January 2017; Accepted 3 January 2017 0147-6513/ © 2017 Published by Elsevier Inc.

Ecotoxicology and Environmental Safety 139 (2017) 36–42

M. Wang et al.

were transferred into a muffle furnace and heated to a desired temperature 450 °C for 1 h. After cooling to room temperature, the obtained carbonaceous materials were washed with hot water followed by distilled water for removal of residual H3PO4 until the pH of the supernatant remained steady at about 7. Then the produced carbons were dried at 105 °C for 8 h, crushed and sieved to particle sizes between 0.075 and 0.019 mm (140–200 mesh) for subsequent experimental use. The EP activated carbons by H3PO4 activation with or without SBS (AC and AC-SBS) were acquired.

are amphipathic substances with hydrophobic and hydrophilic groups, so it is easy to be absorbed at the interface between aqueous and solid. Surface layer of surfactant molecules can be organized on the solid surface via forming the self-associated clusters (Nadeem et al., 2009). However, modification of activated carbon was obtained by traditionally immersing in surfactant-containing solution. To our best knowledge, surfactant serving as dopant to modify activated carbon during the preparation of activated carbon has not been performed thus far. Sodium benzenesulfonate (SBS), a kind of surfactant, was chosen as modifier in this study. Moreover, Enteromorpha prolifera (EP), a widely available marine biomaterial, was utilized as the raw material for the preparation of activated carbon since it is rich in functional groups such as hydroxyl and carboxyl. In addition, it is beneficial to use EP as the precursor for producing AC in terms of waste recovery and waterpollution alleviation. In general, the spent activated carbon was usually incinerated or dumped in a landfill after the saturation or exhaustion. It is necessary to include a regeneration step in respect to environmentally friendly and recycling economy (Ania et al., 2005). This will minimize the operational cost for the adsorbents and consequently reduce waste production which will contribute to the material recycling process (Sabio et al., 2004). The common techniques for activated carbon regeneration are mainly constituted of microwave and thermal treatment, chemical, and electrochemical process (Cazetta et al., 2013). In our work, NaOH solution, one of chemical regeneration methods, was selected as the eluent due to the acidic nature of CIP. In this paper, activated carbons were developed from EP by H3PO4 activation with or without SBS, producing AC-SBS and AC. The regenerated activated carbons (RAC and RAC-SBS) were obtained by soaking spent adsorbents in NaOH solution. The physicochemical properties of original and regenerated carbons, such as morphology, porosity and surface characteristics, were systematically assessed. This study was to discuss and compare the adsorption isotherms and thermodynamic properties of CIP onto the four carbons. The effect of solution pH and ionic strength were also studied. Moreover, our work was also expected to confirm the feasibility of regenerating the spent activated carbons using NaOH solution.

2.3. Regeneration procedure and experimental design The feasibility of regenerating the exhausted activated carbon saturated with CIP was determined using NaOH desorption method. Before the commencement of regeneration, adsorption of CIP by the produced activated carbons (AC and AC-SBS) at an initial CIP concentration of 440 mg/L, agitating for 36 h until the active sites were fully occupied. The obtained samples were added into 1000 mL beakers containing 1000 mL of 0.25 mol/L NaOH for desorption of CIP-loaded adsorbents. The beakers were kept on magnetic stirrers at shaking speed of 125 rpm at room temperature. The regenerated activated carbons (RAC and RAC-SBS) were then separated from NaOH solution, washed with sufficient distilled water to eliminate residual CIP and dried in an oven. After desorption, the concentrations of CIP desorbed, Cde (mg/L), was similarly measured using the UV–vis spectrophotometer. The percentage of desorption was calculated as follows:

Desorption (%) =

Cde × 100% Cad

(1)

where Cde is the amount of CIP desorbed (mg/L); Cad is the amount of CIP adsorbed (mg/L). 2.4. Characterization methods Thermo gravimetric analysis (TGA) and differential scanning calorimetry (DSC) analyzes were obtained from a thermo-gravimetric analyzer (SDT Q600) under N2 gas flow (10 mL/min) from room temperature to 800 °C at a heating rate of 10 °C/min. The surface characteristics of the three adsorbents obtained were analyzed using scanning electron microscopy (SEM Hitachi S4800, Japan). The Brunauer-Emmett-Teller surface area (SBET) and porous properties of the resultant carbons were deduced from the nitrogen adsorption/ desorption isotherms at 77 K using a surface analyzer (Quantachorme Corporation, USA). The pore size distribution was measured following the Density Functional Theory (DFT) method, and the total pore volume (Vt) was determined from the amount of nitrogen absorbed by the BJH theory at the relative pressure of P/P0 around 0.95. The micropore surface area (Smic) and external surface areas (Sext), as well as the micropore volume (Vmic) were calculated by t-plot method. The surface functional groups of the prepared activated carbons were detected using a Fourier transform infrared (FTIR) spectrometer (Fourier-380 FT-IR, USA), where the spectra were recorded in the range of 400– 4000 cm−1. The KBr powder was blended with the samples, and then the mixture was pressed into tablet for analysis.

2. Materials and methods 2.1. Materials All chemicals used in this study were of analytical grade. All solutions were prepared with distilled water. Sodium benzenesulfonate (SBS), (Molecule formula: C6H5SO3Na, Molecule weight: 180 g/mol) was used to modify the activated carbon. Ciprofloxacin hydrochloride (CIP) (molecular weight: 368 g/mol, molecular formula: C17H18FN3O3· HCl, purity > 99.6%), which was selected as adsorbate, was supplied from Sangon Biotech (Shanghai, China). The chemical structure and pH-dependent speciation of this antibiotic are illustrated in Fig. S1, as described in the literature by Carabineiro (Carabineiro et al., 2012). Its pKa1 value is 6.1 for the carboxylic acid group, while pKa2 value is 8.7 for the basic-N-moiety, so it can exist in cation, zwitterions and anion under different pH condition (Drakopoulos and Ioannou, 1997). All chemical reagents used were of analytical grade and the experimental water was distilled water.

2.5. Adsorption experiment

2.2. Synthesis of activated carbons

A set of adsorption experiments were carried out in a water bath temperature-controlled oscillator using 150 mL conical flasks at desired condition for 36 h (shaking speed 150 rpm and 0.6g carbon/50 mL solution). The effects of pH (2–12), temperature (30–50 °C) and ionic strength (0–100 mmol/L) on adsorption by the adsorbents were investigated. Solution pH was adjusted with standard solutions of 0.1 M HCl and 0.1 M NaOH using a pH meter (Model pHS-3C, Shanghai, China). Ionic strength was controlled with NaCl solution. The samples were coated with aluminum foil to avoid possible photodegradation of

Enteromorpha prolifera (EP) was obtained from Qingdao, China. The EP was thoroughly washed to remove impurity, dried in an oven, and shattered to pieces ( < 0.25 mm) using a grinder. About 5 g of EP was soaked in 40 wt% H3PO4 solution at a ratio of 1:2 (g EP/g H3PO4) with or without mixing 8 mmol sodium benzenesulfonate (SBS). The varied amounts of SBS were studied from 2 mmol to 32 mmol to determine the optimized preparation conditions. After impregnation, the mixtures 37

Ecotoxicology and Environmental Safety 139 (2017) 36–42

M. Wang et al.

ce 1 1 = + ce qe qm kL qm

CIP. All the adsorption experiments were run in duplicate, and blank samples were prepared simultaneously under the same condition. After equilibration, the samples were filtered using a 0.45 µm membrane filter, and the residual CIP concentration of the filtrate was determined by a UV–vis spectrophotometer (UV-754, Shanghai) at the maximum wavelength of 275 nm. The uptake amount of CIP adsorbed at equilibrium, Qe (mg/g), was calculated as follows:

Qe = (C0 − Ce ) V / W

ln qe = ln kF +

1 ln ce n

(3)

(4)

where ce (mg/L) is the equilibrium concentration of CIP, qe (mg/g) is the amount of CIP adsorbed at equilibrium, qm (mg/g) is CIP maximum adsorption amount, kL (L/mg) and kF (mg/g(L/mg)1/n) are the Langmuir and Freundlich isotherm constant, respectively.

(2)

where C0 and Ce is the initial and equilibrium concentration of CIP (mg/L), V represents the solution volume (L), and W is the weight of adsorbent used (g). 2.6. Adsorption isotherm models

2.7. Adsorption thermodynamics The adsorption isotherm is of importance to describe how the adsorbent interacted with the adsorbate. Furthermore, the constant of values can help to evaluate the maximum adsorption capacity and understand the mechanisms of adsorption and the homogeneity or heterogeneity on the adsorbent surface. Two most commonly used models, Langmuir isotherm model and Freundlich isotherm model, are applied to fit equilibrium data for the adsorption of CIP onto AC, ACSBS, RAC and RAC-SBS. The Langmuir model presumes monolayer coverage of adsorbent over a homogenous adsorbent surface with a finite number of adsorption sites and there is no interaction between the adsorbed molecules (Allen et al., 2003). Freundlich model is the first known equation representing the adsorption process (Freundlich, 1906). It can be derived assuming multilayer adsorption takes place on a heterogeneous surface with non-uniform distribution of adsorption heat. The linearized forms of the Langmuir and Freundlich equations can be expressed as:

The concept of thermodynamics hypothesizes that energy cannot be gained or lost and the entropy change is the driving force in an isolated system. The thermodynamic parameters deemed to provide a deeper insight into adsorption mechanism were enthalpy change (ΔH, kJ/mol), entropy change (ΔS, kJ/(mol. K)) and Gibbs free energy change (ΔG, kJ/mol). The values of ΔH, ΔS and ΔG were calculated by the following equation (McKay et al., 1985):

ΔG = −RT ln K

(5)

ΔG = ΔH −TΔS

(6)

where R (8.314 J/(mol. K)) is the ideal gas constant, T (K) is absolute temperature, K (L/mol) represents the Langmuir constant, ΔH and ΔS can be obtained by intercept and slope of the plot of ΔG vs. T.

Fig. 1. Thermal analysis (DSC/TGA) for the pyrolysis of AC (a), AC-SBS (b), RAC (c) and RAC-SBS (d).

38

Ecotoxicology and Environmental Safety 139 (2017) 36–42

M. Wang et al.

3. Results and discussion

carbons suffered from corrosion to varying degrees and showed an apparent difference in the structure. It was obviously observed that the surface of the AC had distinct regular tubular porosity, while for ACSBS, there were irregular cavities on rough and loose surface. Thus, the pores of AC were more ordered than that of AC-SBS, which could be further proved by the results of pore size distributions. The pore size distributions of the four adsorbents as well as the nitrogen adsorption/desorption isotherms (inset) are presented in Fig. 3. The isotherms of the carbons displayed a mixture of types I and IV with a hysteresis loop at high relative pressures, which suggested the existence of mesopores in the four adsorbents. According to International Union of Pure and Applied Chemistry (IUPAC), the adsorbent pores are classified into three groups: micropores (diameter < 2 nm), mesopores (2–50 nm), and macropores ( > 50 nm). A large fraction of the pores of the carbons occupied over the range of micropore and mesopores ( < 20 nm), and were essentially micro-mesoporous structure. The details about the textural parameters of the carbons, including SBET, pore volume, pore area and Dp, are summarized in Table 1. The SBET of AC was 1050 m2/g, which was much larger than that of AC-SBS (738 m2/g). This result was probably caused by mixing SBS in the activation of activated carbon, resulting in the block of pores. The Vmic/Vtot values of AC and AC-SBS were 25.6% and 17.2%, respectively, indicating that AC and AC-SBS contained mainly mesopores. While for RAC and RAC-SBS, regeneration proce-

3.1. Characterization of adsorbents 3.1.1. Thermal analysis of AC, AC-SBS, RAC and RAC-SBS Thermo gravimetric analysis (TGA) and differential scanning calorimeter (DSC) curves of AC, AC-SBS, RAC and RAC-SBS are plotted in Fig. 1. TGA analysis was performed in order to appraise the thermal stability of the adsorbents. It can be observed from Fig. 1 that the remaining mass for the four carbons were higher than 65% at the temperature of 800 °C, indicating high resistance of the materials at high temperature and a similar tendency with regards to thermal process. Additionally, the profiles of TGA showed that the pyrolysis process of the four samples could be divided into two stages. An initial weight loss (WL) occurred in temperature below 180 °C for the adsorbents which was attributed to the elimination of water or highly volatile matters. The obvious WL in the higher temperature range of 180–800 °C for the four adsorbents was shown on the TGA curve due to the decomposition of the organic materials (Cazetta et al., 2013).

3.1.2. Textural structure The surface morphological structure of AC (a1 and a2) and AC-SBS (b1 and b2) are detected using scanning electron microscopy and depicted in Fig. 2 (700× and 7000× magnification). Surfaces of the

Fig. 2. SEM micrographs (left: 700x; right: 7000x magnification) of AC (a1 and a2) and AC-SBS (b1, b2).

39

Ecotoxicology and Environmental Safety 139 (2017) 36–42

M. Wang et al.

vibration of carboxyl or lactone groups (─C═O) (Min et al., 2004). As a comparison, the peak at about 1600 cm−1 for AC-SBS and RAC-SBS was ascribed to the stretching vibration of aromatic ring or C═C bond (Kim et al., 2005). The significant peaks for the adsorbents in the regions of 1000─1500 cm−1 may assign to the presence of various configurations of C, O, H and N bonds of the activated carbons. In addition, it is essential to point out that less amount of peaks for the RAC and RACSBS in comparison to the AC and AC-SBS. This phenomenon could be explained by the fact that NaOH reacted with the surface acidic functional groups of CIP-loaded carbons, leading to the decrease of amount of functional groups.

3.2. Adsorption isotherms The fitted results of CIP adsorption isotherm at 30, 40, 50 °C, i.e. the maximum adsorption capacity, isotherm parameters and the correlation coefficients, R2, are summarized in Table 2. Langmuir isotherm model was more suitable to represent the adsorption behavior than Freundlich with greater R2 values (R2 > 0.97), suggesting that the adsorption mechanisms tended to be homogeneous and monolayer coverage. Furthermore, Langmuir model was fairly successful in evaluating the experimental saturation capacities, implying monolayer coverage with the strong interactions between CIP and the carbon surface. The maximum monolayer adsorption capacities of CIP were 250 mg/g, 286 mg/g, 233 mg/g and 256 mg/g for AC, AC-SBS, RAC and RACSBS, respectively. It was reported that when saturation was first reached at the exterior surface, the adsorbate molecules then entered the pores of adsorbent and were adsorbed by the interior surface of the particles (Tan et al., 2009). The adsorbate molecules at the interior surface were difficult to be desorbed by NaOH solution and there was a small amount of CIP residues in the interior surface. Thus, a portion of surface sites were occupied by CIP molecules, resulting in the decrease of uptake capacity for RAC and RAC-SBS. In addition, it is generally considered that the process denotes favorable, moderately difficult and poor adsorption if the values of n calculated from Freundlich model are in the range of 2–10, 1–2, < 1, respectively (Sun et al., 2012a, 2012b). The 1/n values of the results were much lower than 1, indicating that the CIP could be favorably adsorbed by the native activated carbons (AC and AC-SBS) and regeneration carbons (RAC and RAC-SBS) in all cases studied. Moreover, the regeneration rates of NaOH reached to 85.8% and 89.7% for AC and AC-SBS, respectively.

Fig. 3. Pore size distributions and nitrogen adsorption/desorption isotherms (inset) for AC, AC-SBS, RAC and RAC-SBS.

Table 1 Surface areas and porosity characteristics of the AC, AC-SBS, RAC and RAC-SBS. Activated carbon

SBETa (m2/g)

Sextb (m2/ g)

Smicc (m2/g)

Vmicd (cm3/ g)

Vtote (cm3/ g)

Vmic/ Vtot (%)

Dpf (nm)

AC AC-SBS RAC RAC-SBS

1050 738 429 424

547 377 339 294

498 360 90.3 130

0.268 0.159 0.033 0.060

1.05 0.924 0.433 0.697

25.6 17.2 7.62 8.61

4.01 5.01 4.03 6.57

a: BET surface area, b: external surface area, c: micropore surface area, d: micropore volume, e: total pore volume, f: mean pore diameter.

dure showed a significant decrease in the surface area (RAC: 429 m2/g, RAC-SBS: 424 m2/g), this could be due to a part of CIP depositing on the surface of carbonaceous materials. 3.1.3. Surface chemistry FTIR analysis was used to qualitatively display the surface functional groups on the surface of activated carbons, as represented in Fig. 4. The four carbons showed similar peaks in their FTIR spectra, confirming that they had similar surface functional groups. All spectra for the carbons had a peak located at wave number about 3419 cm−1, attributing to the stretching vibration of ─OH groups (carboxyls, alcohols or phenols) (Juan and Ke-qiang, 2009). The bands appeared at around 1558 cm−1 for AC and RAC corresponded to asymmetric

3.3. Effect of ionic strength Effects of ionic strength on CIP adsorption were carried out at natural pH by adding NaCl at different concentrations from 0 to 100 mmol/L. The results are represented in Fig. S2. As reported by Lützenkirchen, there are two different surface complexes formation during adsorption: inner-sphere and outer-sphere (Lützenkirchen, 1997). Covalent bonds form not in outer-sphere but in inner-sphere surface complexes. The adsorption capacity of CIP onto AC-SBS and RAC-SBS were almost constant at pH 6.0 (AC-SBS: 231–235 mg/g, RACSBS: 185–188 mg/g), irrespective of the increase in ionic strength, indicating that the adsorption of CIP primarily proceeded through formation of inner-sphere surface complexation between the CIP species and the surface chemical functional groups. Consequently, CIP was covalently bonded onto AC-SBS and RAC-SBS. While CIP adsorption onto AC and RAC decreased slightly with the increase of ionic strength (AC: 187–195 mg/g, RAC: 149–157 mg/g). Such phenomenon may be taken as an indication of outer-sphere surface complexation, suggesting that electrostatic interaction was partly responsible for the adsorption onto AC and RAC and ion-exchange was a potential adsorption mechanism.

1035 1176

1246 3419

1210 1558

AC AC-SBS RAC RAC-SBS

1600 1263 500

1000

1500

2000

2500

3000

3500

4000

Wavenumber (cm ) Fig. 4. FTIR spectra of AC, AC-SBS, RAC and RAC-SBS.

40

Ecotoxicology and Environmental Safety 139 (2017) 36–42

M. Wang et al.

Table 2 Langmuir and Freundlich isotherm parameters for the adsorption of CIP by AC, AC-SBS, RAC and RAC-SBS at different temperatures. Activated carbon

AC

AC-SBS

RAC

RAC-SBS

Langmuir

Freundlich

T °C

qm mg·g−1

kL L·mg−1

R2

kF (mg·g−1) (l/ mg)1/n

1/ n

R2

30 40 50 30 40 50 30 40 50 30 40 50

244 250 238 286 286 286 233 208 204 256 244 256

0.0903 0.0860 0.0915 0.168 0.193 0.145 0.0288 0.0575 0.0748 0.0561 0.0643 0.0780

0.997 0.998 0.998 0.998 0.999 0.998 0.972 0.997 0.998 0.987 0.989 0.993

142 133 140 193 183 170 78.2 107 87.4 127 137 143

0.0913 0.109 0.0916 0.0665 0.0796 0.0947 0.170 0.109 0.159 0.115 0.0927 0.0961

0.947 0.955 0.873 0.938 0.953 0.930 0.865 0.922 0.686 0.874 0.788 0.890

Table 3 Thermodynamic parameters for the adsorption of CIP onto AC, AC-SBS, RAC and RAC-SBS at different temperatures. Activated carbon

T (K)

ΔG (kJ/mol)

ΔH (kJ/mol)

ΔS (J/(mol·K))

AC

303 313 323 303 313 323 303 313 323 303 313 323

−26.2 −27.0 −28.0 −27.8 −29.1 −29.2 −23.4 −25.9 −27.5 −25.0 −26.2 −27.6

0.486

88.0

5.59

73.2

36.5

202

13.4

127

AC-SBS

RAC

RAC-SBS

3.4. Effect of pH

3.5. Adsorption thermodynamics

Solution pH played a significant function in affecting the extent of adsorption and it can alter the surface charge of activated carbon and species of CIP. The adsorption of CIP by the adsorbents at different initial pH values is illustrated in Fig. S3, indicating that the adsorption was pH-dependent. The similar trends were found during the adsorption of CIP to original carbons (AC and AC-SBS) and regeneration carbons (RAC and RAC-SBS). In detail, the adsorption capacity was higher under acid condition (pH 2–7) than the alkaline condition (pH 8–12) and the maximum adsorption capacity occurred at about pKa2 of CIP for all the adsorbents. In addition, a majority of acidic groups were in protonated forms under acid condition. The low adsorption capacity at lower pH (2–4) could be attributed to the less electrostatic attraction between CIP+ and the protonated surfaces of the carbons as well as competition between CIP+ and H+ ions. The decrease of protonated groups markedly hindered the bonding interaction with the increase in pH values from 5 to 8, because the electron acceptor abilities of the cationic groups were weakened by deprotonation. When initial pH was between 5 and 8, electrostatic attraction between carbons and the zwitterionic form (CIP°) was dominant. At high pH values (8–12), which was higher than pKa2 of the CIP, the surface of the carbons gradually became negatively charged and the anion form (CIP-) gradually increased, which could also lead to the decrease of CIP adsorption. Such phenomenon has also been observed in CIP adsorption on montmorillonite (Wu et al., 2010) and kaolinite (Li et al., 2011).

Thermodynamic parameters related to the adsorption process, i.e., ΔG (kJ/mol), ΔH (kJ/mol) and ΔS (J/(mol. K)) values are listed in Table 3. The free energy change ΔG values for the four adsorbents are negative, illustrating that the adsorption process were feasible and spontaneous in the range of temperature studied. It was reported that −20 kJ/mol < ΔG < 0 kJ/mol was for physical adsorption and −400 kJ/mol < ΔG < −80 kJ/mol was for chemical adsorption (Milmile et al., 2011). In this study, the ΔG values were in the range between −29.2 kJ/mol and −23.4 kJ/mol, revealing that the adsorption process for all the adsorbents were mainly physical in nature enhanced by chemisorptions. The positive ΔH values further confirmed that the adsorption procedures were endothermic reaction. It could be demonstrated that the diffusion rate of CIP molecular was intense as temperature rose and entered into the aperture easily. The positive values of ΔS indicated the increasing randomness and the good binding affinity of CIP toward the adsorbents. For comparison, Table S1 summarized the maximum adsorption capacity of various types of adsorbents. A comparison with these adsorbents also suggested high CIP adsorption capacity of the four carbons prepared in this study and further confirmed the suitability for CIP adsorption. 4. Conclusion Enteromorpha prolifera was used to produce activated carbon with phosphoric acid activation mixed with sodium benzenesulfonate (SBS),

41

Ecotoxicology and Environmental Safety 139 (2017) 36–42

M. Wang et al.

surfactants modified montmorillonite and its application for the removal of methyl orange. Chem. Eng. J. 171, 1150–1158. Chen, W., Parette, R., Zou, J., Cannon, F.S., Dempsey, B.A., 2007. Arsenic removal by iron-modified activated carbon. Water Res. 41, 1851–1858. Cho, D.-W., Chon, C.-M., Kim, Y., Jeon, B.-H., Schwartz, F.W., Lee, E.-S., Song, H., 2011. Adsorption of nitrate and Cr(VI) by cationic polymer-modified granular activated carbon. Chem. Eng. J. 175, 298–305. Drakopoulos, A.I., Ioannou, P.C., 1997. Spectrofluorimetric study of the acid–base equilibria and complexation behavior of the fluoroquinolone antibiotics ofloxacin, norfloxacin, ciprofloxacin and pefloxacin in aqueous solution. Anal. Chim. Acta 354, 197–204. Freundlich, H., 1906. Ueber die Adsorption in Loesungen. Leipzig, Engelmann. Hirsch, R., Ternes, T., Haberer, K., Kratz, K.-L., 1999. Occurrence of antibiotics in the aquatic environment. Sci. Total Environ. 225, 109–118. Ji, L., Wan, Y., Zheng, S., Zhu, D., 2011. Adsorption of tetracycline and sulfamethoxazole on crop residue-derived ashes: implication for the relative importance of black carbon to soil sorption. Environ. Sci. Technol. 45, 5580–5586. Juan, Y., Ke-qiang, Q., 2009. Preparation of activated carbon by chemical activation under vacuum. Environ. Sci. Technol. 43, 3385–3390. Kim, S., Eichhorn, P., Jensen, J.N., Weber, A.S., Aga, D.S., 2005. Removal of antibiotics in wastewater: effect of hydraulic and solid retention times on the fate of tetracycline in the activated sludge process. Environ. Sci. Technol. 39, 5816–5823. Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B., Buxton, H.T., 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999–2000: a national reconnaissance. Environ. Sci. Technol. 36, 1202–1211. Li, Z., Hong, H., Liao, L., Ackley, C.J., Schulz, L.A., MacDonald, R.A., Mihelich, A.L., Emard, S.M., 2011. A mechanistic study of ciprofloxacin removal by kaolinite. Colloids Surf. B-Biointerfaces 88, 339–344. Lützenkirchen, J., 1997. Ionic strength effects on cation sorption to oxides: macroscopic observations and their significance in microscopic interpretation. J. Colloid Interface Sci. 195, 149–155. McKay, G., Bino, M.J., Altamemi, A.R., 1985. The adsorption of various pollutants from aqueous solutions on to activated carbon. Water Res. 19, 491–495. Milmile, S.N., Pande, J.V., Karmakar, S., Bansiwal, A., Chakrabarti, T., Biniwale, R.B., 2011. Equilibrium isotherm and kinetic modeling of the adsorption of nitrates by anion exchange Indion NSSR resin. Desalination 276, 38–44. Min, S.H., Han, J.S., Shin, E.W., Park, J.K., 2004. Improvement of cadmium ion removal by base treatment of juniper fiber. Water Res. 38, 1289–1295. Moradi, S.E., 2014. Microwave assisted preparation of sodium dodecyl sulphate (SDS) modified ordered nanoporous carbon and its adsorption for MB dye. J. Ind. Eng. Chem. 20, 208–215. Nadeem, M., Shabbir, M., Abdullah, M.A., Shah, S.S., McKay, G., 2009. Sorption of cadmium from aqueous solution by surfactant-modified carbon adsorbents. Chem. Eng. J. 148, 365–370. Popescu, M., Joly, J.P., Carré, J., Danatoiu, C., 2003. Dynamical adsorption and temperature-programmed desorption of VOCs (toluene, butyl acetate and butanol) on activated carbons. Carbon 41, 739–748. Sabio, E., González, E., González, J.F., González-Garcı́a, C.M., Ramiro, A., Gañan, J., 2004. Thermal regeneration of activated carbon saturated with p-nitrophenol. Carbon 42, 2285–2293. Schmitt, H., Stoob, K., Hamscher, G., Smit, E., Seinen, W., 2006. Tetracyclines and tetracycline resistance in agricultural soils: microcosm and Field Studies. Microb. Ecol. 51, 267–276. Storteboom, H., Arabi, M., Davis, J.G., Crimi, B., Pruden, A., 2010. Tracking antibiotic resistance genes in the south platte river basin using molecular signatures of urban, agricultural, and pristine sources. Environ. Sci. Technol. 44, 7397–7404. Sun, Y., Yue, Q., Gao, B., Huang, L., Xu, X., Li, Q., 2012a. Comparative study on characterization and adsorption properties of activated carbons with H3PO4 and H4P2O7 activation employing Cyperus alternifolius as precursor. Chem. Eng. J. 181, 790–797. Sun, Y., Yue, Q., Gao, B., Wang, B., Li, Q., Huang, L., Xu, X., 2012b. Comparison of activated carbons from Arundo donax Linn with H4P2O7 activation by conventional and microwave heating methods. Chem. Eng. J. 192, 308–314. Tan, I.A.W., Ahmad, A.L., Hameed, B.H., 2009. Adsorption isotherms, kinetics, thermodynamics and desorption studies of 2,4,6-trichlorophenol on oil palm empty fruit bunch-based activated carbon. J. Hazard. Mater. 164, 473–482. Wu, Q., Li, Z., Hong, H., Yin, K., Tie, L., 2010. Adsorption and intercalation of ciprofloxacin on montmorillonite. Appl. Clay Sci. 50, 204–211.

and NaOH solution was employed to regenerate CIP-loaded carbons. The results of N2 adsorption/desorption, SEM, FTIR analysis indicated that addition of SBS in activation process reduced the activated carbon's surface area (AC: 1050 m2/g, AC-SBS: 738 m2/g), but increased the carbon's surface functional groups. The adsorption study represented that AC-SBS showed much larger adsorption capacity for CIP than AC (AC: 250 mg/g, AC-SBS: 286 mg/g). Moreover, the regenerated carbons (RAC and RAC-SBS) had 429 m2/g and 424 m2/g of surface area, and regeneration rates were 85.8% and 89.7%, respectively. The CIP uptake onto the four adsorbents was found to increase with increasing initial pH but have no significant change with the increase of ionic strength. The equilibrium data at the three temperatures were better fitted by Langmuir model. The negative ΔG and positive ΔH values for the four adsorbents indicated that the adsorption process was spontaneous and endothermic reaction. The main mechanisms for the adsorption of CIP onto the carbons were cation exchange and electrostatic attraction. Acknowledgements The authors would like to acknowledge financial support for this work provided by Shandong Province Postdoctoral (No. 2009sd1034) fund. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ecoenv.2017.01.006. References Ahn, C.K., Park, D., Woo, S.H., Park, J.M., 2009. Removal of cationic heavy metal from aqueous solution by activated carbon impregnated with anionic surfactants. J. Hazard. Mater. 164, 1130–1136. Aktaş, Ö., Çeçen, F., 2006. Effect of type of carbon activation on adsorption and its reversibility. J. Chem. Technol. Biotechnol. 81, 94–101. Allen, S.J., Gan, Q., Matthews, R., Johnson, P.A., 2003. Comparison of optimised isotherm models for basic dye adsorption by kudzu. Bioresour. Technol. 88, 143–152. Ania, C.O., Parra, J.B., Menéndez, J.A., Pis, J.J., 2005. Effect of microwave and conventional regeneration on the microporous and mesoporous network and on the adsorptive capacity of activated carbons. Microporous Mesoporous Mat. 85, 7–15. Babel, S., Kurniawan, T.A., 2004. Cr(VI) removal from synthetic wastewater using coconut shell charcoal and commercial activated carbon modified with oxidizing agents and/or chitosan. Chemosphere 54, 951–967. Bautista-Toledo, I., Rivera-Utrilla, J., Ferro-García, M.A., Moreno-Castilla, C., 1994. Influence of the oxygen surface complexes of activated carbons on the adsorption of chromium ions from aqueous solutions: effect of sodium chloride and humic acid. Carbon 32, 93–100. Carabineiro, S.A.C., Thavorn-Amornsri, T., Pereira, M.F.R., Figueiredo, J.L., 2011. Adsorption of ciprofloxacin on surface-modified carbon materials. Water Res. 45, 4583–4591. Carabineiro, S.A.C., Thavorn-amornsri, T., Pereira, M.F.R.P., Serp, P., Figueiredo, J.L., 2012. Comparison between activated carbon, carbon xerogel and carbon nanotubes for the adsorption of the antibiotic ciprofloxacin. Catal. Today 186, 29–34. Carmosini, N., Lee, L.S., 2009. Ciprofloxacin sorption by dissolved organic carbon from reference and bio-waste materials. Chemosphere 77, 813–820. Cazetta, A.L., Junior, O.P., Vargas, A.M.M., da Silva, A.P., Zou, X., Asefa, T., Almeida, V.C., 2013. Thermal regeneration study of high surface area activated carbon obtained from coconut shell: characterization and application of response surface methodology. J. Anal. Appl. Pyrolysis 101, 53–60. Chen, D., Chen, J., Luan, X., Ji, H., Xia, Z., 2011. Characterization of anion–cationic

42