- Email: [email protected]
Extending surfactant-modified 2:1 clay minerals for the uptake and removal of diclofenac from water
Accepted Manuscript Title: Extending surfactant-modified 2:1 clay minerals for the uptake and removal of diclofenac from water Author: Ken Sun Yan Shi Honghan Chen Xiaoyu Wang Zhaohui Li PII: DOI: Reference:
S0304-3894(16)30469-1 http://dx.doi.org/doi:10.1016/j.jhazmat.2016.05.038 HAZMAT 17727
To appear in:
Journal of Hazardous Materials
Received date: Revised date: Accepted date:
3-3-2016 20-4-2016 11-5-2016
Please cite this article as: Ken Sun, Yan Shi, Honghan Chen, Xiaoyu Wang, Zhaohui Li, Extending surfactant-modified 2:1 clay minerals for the uptake and removal of diclofenac from water, Journal of Hazardous Materials http://dx.doi.org/10.1016/j.jhazmat.2016.05.038 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Extending surfactant-modified 2:1 clay minerals for the uptake and removal of diclofenac from water Ken Suna,b, Yan Shia, Honghan Chenb,*, Xiaoyu Wangc, Zhaohui Lic,d,* aSchool
of Environmental and Municipal Engineering, North China University of
Water Resources and Electric Power, Zhengzhou 450045,China bSchool
of Water Resources and Environment, China University of Geosciences,
Beijing 100083, China cSchool
of Materials Science and Technology, China University of Geosciences,
Beijing 100083, China dGeosciences
Department, University of Wisconsin–Parkside, Kenosha, WI
53144, USA
Highlights:
Illite and montmorillonite modified by a cationic surfactant.
Modification resulted in significant uptake of diclofenac (DC) upto 1 mmol/g.
Specific surface area and the anion exchange not limiting factors for DC uptake.
Partitioning into the hemimicelles and admicelles responsible for DC uptake.
Results would extend application of modified clays for removal of anionic drugs.
Abstract The presence and persistency of pharmaceuticals and personal care products
*
Corresponding author: Tel: 1-262-595-2487; Fax: 1-262-595-2056; Email: [email protected] 1
(PPCPs) in the environment attracted great attention recently. Among them, antibiotics and pain-killers accounted for a large quantity. Although many works were devoted to the investigation of their removal in wastewater treatment processes, most of the PPCPs studied were of cationic nature. The net repulsive interactions between anionic PPCPs and negatively charged sorbents make them difficult to be removed in wastewater treatment. In this study, 2:1 clay minerals illite and montmorillonite (MMT) were modified with different amounts of cationic surfactant cetyltrimethylammoium bromide (CTAB). The types and sites of interactions between the surfactant-modified clays and the anionic drug diclofenac (DC) were investigated. The uptake of DC on the modified clays was controlled by the CTAB loading level and its surface configuration on the clays. The adsorption sites of DC were limited to the external surfaces of modified illite due to its non-swelling nature. On the contrary, both the external and interlayer sites were available for the adsorption of DC on modified MMT. A CTAB bilayer formation resulted in significant increase in DC adsorption due to the formation of extensive admicelles. FTIR results showed participation of the benzene ring, N–H, and CH2 CH3 for the interactions between DC and modified MMT, suggesting that partitioning of DC into the admicelles of CTAB played a significant role in DC uptake. The results could extend the application of surfactant-modified clays for the removal of anionic PPCPs in the wastewater treatment processes.
Keywords: adsorption, cationic surfactant, clay minerals, diclofenac, hydrophobic interactions, modification
2
1. Introduction There have been great concerns on the detection of pharmaceuticals and personal care products (PPCPs) in soil, sediments, surface and ground water [1,2]. Among the PPCPs, antibiotics and anti-inflammatory medicines accounted for a large quantity. The non-steroidal inflammatory drug diclofenac (DC) is one of the most commonly detected compounds [3], with a 90-percentile of 1.6 µg/L detected in all 49 wastewater treatment plants (WWTPs) and rivers sampled [4]. Its half-maximal effective concentration (EC50) values were 68 and 72 mg/L in the Daphnia and algal tests in comparison to 166 and 626 mg/L for naproxen [5]. Its lowest observed effective concentrations for fish toxicity were in the ranges of wastewater concentrations, which were reported as 0.3 – 3 µg/L for the influent and 0.3 – 2.5 µg/L for the effluent [6]. In wastewater treatment, biologic degradation pathway of DC was a complex process [7]. Short term investigation of its biodegradation with a pilot sewage plant and biofilm reactors as model systems under oxic and anoxic conditions showed no elimination and its final concentration was approximately 95% of their initial concentration after 48 – 55 h of reaction [8]. Even during conventional treatment, it was only barely removable [9]. A removal efficiency of 22% was calculated in the sewage treatment plant near the Höje River in Sweden [10]. A summary of 20 studies showed its removal efficiencies in WWTPs varied from 0 to 80%, mostly in the scope of 21–40% [11]. In addition, adsorption on the secondary and primary sludge was
3
<5% and 5–15%, respectively [7]. Adsorption process on activated carbon (AC) was most commonly used to remove DC [12]. For its adsorptive removal, materials of positive charges were commonly deployed. DC could be effectively intercalated into the interlayer of layered Mg-Al hydrotalcite with a drug loading up to 55% by weight [13]. However, most of the studies on the uptake of DC by hydrotalcite were focused on extended drug releases [14,15]. As most minerals bear negative charges, they were often modified to reverse the charge in order to increase their uptake of negatively charges organic species. Modification of mesoporous silica with aminopropyl and trimethylsilyl resulted in an increase of the solute distribution coefficient Kd value by 20 and 1000 folds, which was attributed to increased hydrophobicity [16]. Modification of clinoptilolite, phillipsite, and chabazite with a cationic surfactant cetylpyridinium chloride (CP) increased in DC adsorption capacities to 49, 54, and 21 mg/g, respectively [17]. Natural clay minerals have negative charges due to isometric substitution in the tetrahedral and octahedral sites. As clay minerals are inexpensive in material cost, they were studied extensively for the removal of heavy metals and cationic drugs. For the adsorptive removal of gemfibrozil, mefenamic acid, and naproxen, light expanded clay aggregates showed superior removal properties in comparison to vermiculite due to its low density and, thus, low weight of material used [18]. For the uptake of anionic species, clay minerals were often modified with long chain cationic surfactants to reverse the surface charges [19]. Montmorillonite (MMT) has large specific surface
4
area (SSA) and high cation exchange capacity (CEC) and is commonly studied for its removal of anionic contaminants after modifications [19]. MMT modified by poloxamer 188 (F-68) resulted in a DC adsorption of 108 mg/g [20]. A significant increase in binding between DC and clay-admicelle complex was attributed to large surface area and hydrophobic domains created in the complex after addition of surfactant [21]. Column results showed a DC loading capacity up to 19 mg/g for bentonite modified with a cationic surfactant hexadecyltrimethylammonium bromide [22]. The extensive uptake of DC was mainly due to the expandable nature of MMT. In comparison to MMT, illite is also a common soil mineral with a higher charge density, which is compensated by K+ in the interlayer, and as a result, it is non-expandable with much lower SSA and CEC values. As such, its utilization was limited. However, due to the similarity between MMT and illite in the stacking of tetrahedral and octahedral sheets, investigating and contrasting the differences in their uptake of pharmaceuticals after being modified by cationic surfactants would provide ultimate mechanisms of the interactions between the pharmaceuticals and the modified clay minerals. Thus, the goal of this study was to contrast the DC adsorption on surfactant-modified MMT from that on modified illite in order to elucidate the mechanistic similarity or differences on their DC uptake. Ultimately, their potential use to removal anionic phamaceuticals could be optimized in future studies.
2. Materials and methods 2.1. Preparation of surfactant-modified minerals
5
The illite used was IMt-2, while the montmorillonite (MMT) used was SAz-2, a Ca-montmorillonite. Both were obtained from the Source Clay Minerals Repository (Purdue University, West Lafayette, IN). They replaced illite IMt-1 and MMT SAz-1, which were well characterized for their chemical and physical properties but were exhausted and no longer available. The XRD pattern of the illite showed dominant illite with 7.5% quartz, 2% microcline, 0.9% rutile, and 0.3% kaolinite [23]. The reported CEC was 140 mmolc/kg and SSA was 51 m2/g for IMt-1 [24]. For the MMT, the CEC was 1200 mmolc/kg [25]. The external SSA of SAz-1 based on multi-point BET measurements was 65 m2/g [26]. If the MMT was assumed as fundamental particles in aqueous suspension, its calculated SSA would be 717 m2/g [27]. The surfactant used was cetyltrimethylammonium bromide (CTAB) (from Sigma-Aldrich) with a critical micelle concentration (CMC) of 0.9 mmol/L [28]. To ensure admicelle sorption of CTAB, the initial concentrations of CTAB should be higher than this value during MMT and illite modification. The DC used was in a sodium form (CAS#: 15307-79-6), which is water-soluble [29] with a formula mass of 318.14 g/mol. It was purchased from Sigma-Aldrich with a purity > 99%. To investigate the effect of surfactant surface coverage on DC uptake, the MMT and illite were modified to 0.5, 1.0 and 2.0 CEC values of the minerals. To do so, 3 g of MMT was combined with 180 mL of CATB solution or 30 g of illite was combined with 210 mL of CATB solution at the concentrations of 10, 20, and 40 mmol/L. The mixtures were shaken for 24 h at 150 rpm and then centrifuged at 3500 rpm for 20 min. The supernatants were analyzed for their equilibrium CTAB concentrations,
6
which were less than 0.1 mmol/L, confirming achieving the target CTAB loadings on these minerals. The mixtures were then washed with two portion of deionized water to remove the excess counterion bromide, before being air-dried.
2.2. Adsorption of DC by modified clays To each 50-mL centrifuge tubes, 0.2 g of modified illite was combined with 10-mL DC solutions at concentrations of 0.5 to 2 mmol/L for 24 h. For modified MMT, 0.07 g was combined with 20-mL of DC solutions at concentrations of 0.5 to 5 mmol/L for 24 h. For the kinetic study, the time for mixing varied from 0.1 to 24 h at an initial DC concentration of 5 mmol/L for MMT and 1.7 mmol/L for illite. Although the initial DC concentrations were much higher than its concentrations found in the environment, it was necessary to investigate the DC adsorption capacity on these minerals modified by CTAB to different surface loading levels and their mechanism of interactions with the modified clays.
2.3 Instrumental analyses The SEM observation was made on JSM-IT300 scanning electron microscope under a voltage of 20.0 kV. The contact angle was measured using ZH10692 goniometer with adjustable LED backlight to achieve a better contrast of the liquid drops. The images were scanned at 70 frames per second with a resolution of 2048*1536. The magnification was increased from 0.7 to 4.5 times in a continuous mode. The accuracy was 0.1°. The XRD analyses were conducted on D-8 Advances
7
with a CuK radiation at 40 kV and 100 mA, a scanning speed at 8°/min and 0.02° per step from 3 to 70°. The FTIR analyses were performed on TENSOR 27 FTIR (Bruker) in the range of 400 – 4000 cm–1 with a resolution of 0.4 cm–1 and a speed of 65 scans per second. The equilibrium DC concentration was determined using a UV-Vis method at a wavelength of 276 nm with a linear range for the calibration of 0.01 to 0.07 mM.
3. Results and discussion 3.1. SEM observation The goal of the SEM observation was to determine if there were morphological changes after the minerals being modified to different levels of CTAB. The surfaces of the raw minerals are relatively smooth (Fig. 1a and 1e). After modified to 0.5, 1.0, and 2.0 CEC, the their surfaces are covered with small particles progressively, suggesting better adhesion of the smaller particles onto the large ones (Fig. 1). This could be attributed to a change in surface hydrophobicity and thus, the free energy of adhesion. The increase in surfactant loading would result in enhanced van der Waal’s interactions due to increases in organic molecules on the surface of these hydrophilic minerals and formation of progressive hydrophobic surfaces, which was confirmed in contact angle measurements and FTIR spectra (see sections 3.2 and 3.5), and thus, a reduction for free energy of adhesion. The increased interaction would result in better partitioning of organic molecules onto the surface of the minerals.
8
3.2. Contact angle measurement To confirm the speculation of adhesive interactions of minerals after progressive surfactant modification, the contact angles between water and the mineral surfaces were measured using a contact goniometer. The contact angles are 21.5°, 42.5°, 43°, and 57° for raw illite and illite modified to 0.5, 1.0, and 2.0 CEC (Fig. 2a-d). Similarly, the contact angles increased from 51.5°, 53.5°, 60.5°, to 76.5° for raw MMT and MMT modified to 0.5, 1.0, and 2.0 CEC (Fig. 2e-h). The progressive increase in contact angles indicate that the mineral surfaces are becoming more hydrophobic as the surfactant loading increased. The increased hydrophobicity would result in enhanced uptake of hydrophobic molecules on the surfaces of the minerals.
3.3. DC adsorption isotherm The uptake of DC on surfactant-modified illite and MMT increases as the surfactant surface coverage increased (Fig. 3). The data were fitted to different adsorption models and the Langmuir model fitted the data best. The Langmuir adsorption isotherm has the form:
CS
K L S mCL 1 K LCL
(1)
where CS is the amount of DC adsorbed at equilibrium (mmol/kg), Sm the apparent adsorption capacity (mmol/kg), CL the equilibrium DC concentration (mmol/L), and KL the Langmuir coefficient (L/mmol). The fitted DC adsorption capacities are 50, 28, 18 mmol/kg, while the KL values are 40, 5.6, and 2.8 L/mmol for the illite modified to 2.0, 1.0, and 0.5 CEC, respectively. Similarly, the DC adsorption capacities are 1000, 450, 9
and 330 mmol/kg, while the KL values are 32, 11, and 5 L/mmol for MMT modified to 2.0, 1.0, and 0.5 CEC. As the CEC of the illite is 140 mmolc/kg, at 2.0 CEC, the anion exchange capacity (AEC) would be 140 mmolc/kg as a result of surfactant surface admicelle or bilayer formation. This value is much larger than the DC adsorption capacity of 50 mmol/kg on the illite modified to 2.0 CEC. For MMT its CEC value is 1200 mmolc/kg, which will result in an ACE value of 1200 mmolc/kg under surfactant admicelle or bilayer coverage, in comparison to the DC adsorption capacity of 1000 mmol/kg. These results suggested that the CEC values of the minerals played an important role for CTAB modification, and thus, the total organic carbon (TOC) content of the minerals after modification. In comparison, the DC adsorption capacities on CP-modified zeolite to 1.0, 2.0, and 3.0 of its external CEC values were 70, 135, and 160 mmol/kg [30] and a DC adsorption capacity of 500 mmol/kg was found on activated carbon [31]. Compared with these materials, MMT modified to 2.0 CEC has superior properties on DC uptake and removal, let alone its low material cost.
3.4. DC adsorption kinetics The adsorption of DC on illite and MMT modified by CTAB to 2.0 CEC is almost instantaneous, particularly on illite (Fig. 4). The adsorption data were fitted to several kinetic models and the pseudo-second-order model fitted the experimental data the best:
qt
kqe2t 1 kqet
(2) 10
where qe and qt are amounts of DC adsorbed at equilibrium and at time t (mmol/kg), k is the pseudo-second-order rate constant (kg/mmol-h). The fitted initial rate and rate constant are 10500 mmol/kg-h, and 4.5 kg/mmol-h for DC uptake on illite modified to 2.0 CEC from an initial concentration of 1.7 mmol/L. In contract, the fitted initial rate and rate constant are 17000 mmol/kg-h, and 0.024 kg/mmol-h for DC uptake on MMT modified to 2.0 CEC from an initial concentration of 5 mmol/L. Similarly, DC adsorption on CP-modified zeolite also followed a pseudo-second-order reaction and diffusion through the boundary layer was considered as the rate controlling step of the process [32]. The higher initial rate constant for DC adsorption on modified illite suggested surface uptake of DC. In contrast, the lower rate constant for DC uptake on modified MMT may suggest that most of the adsorption sites were in the interlayer instead of on the surface and could be be due to DC diffusion from the bulk solution into the interlayer space.
3.5. XRD and FTIR characterization To confirm the DC adsorption sites on modified illite and MMT, the XRD patterns of raw minerals, minerals modified to 2.0 CEC, and after DC uptake on modified minerals are shown in Fig. 5. For illite there is essentially no changes for the XRD patterns (Fig. 5a), indicating that the adsorption of CTAB was restricted on the mineral surfaces and the adsorption of DC on modified illite were also on the external surfaces. This confirmed the speculation from the kinetic study due to the non-swelling nature of illite. In contrast, the (001) peak of MMT moves to a lower
11
angle (Fig. 5b), suggesting expansion of the interlayer space after CTAB modification. After DC uptake, the location of the (001) remained the same, suggesting that adsorption of DC did not result in additional interlayer expansion. However, as the adsorbed CTAB molecule would form paraffin-like structure in the interlayer [33], this could create sites for DC to partition or anion exchange with the counterion Br– on the surface or in the interlayer of MMT.
For illite and MMT, the band at 3620 cm–1 (Fig. 6) was attributed to the stretching vibration of structural -OH in the 2:1 layer, while the broader band at 3420 cm–1 was attributed to the adsorbed water on the surface or in the interlayer of MMT [34]. The intensity of the latter band is drastically reduced after CTAB modification, suggesting significant removal of adsorbed or interlayer water. This would drastically increase the hydrophobicity of the modified minerals as confirmed by the contact angle measurements. The bands at 2920 and 2850 cm–1 appear after CTAB modification. These bands were attributed to the asymmetric and symmetric stretching vibrations of C–C in the alkyl chain [35], and their intensities remain more or less the same (Fig. 6), suggesting no changes of CTAB loading before and after DC adsorption.
The characteristic bands of DC are 3430 cm–1 (N–H stretching); 1570 cm–1 (N– H bending); 1505 cm–1 (N–H bending); 1676 cm–1 (C=O stretching); 1452 cm–1 and 1400 cm–1 (CH2 CH3 deformation, CH2 bending of ortho-substituted benzene); 1305 cm–1 and 1150 cm–1 (C–O-bending); 1585 cm–1 (benzene ring carbon–carbon
12
vibration); and 748 cm–1 (C–Cl stretching) [36]. In this study, the N–H bending at 1570 cm–1 is located at 1577 cm–1 and it blue shifts to 1590 cm–1 after its adsorption on modified MMT. Meanwhile, the N–H bending at 1505 cm–1 blue shifts to 1513 cm–1. The benzene ring carbon–carbon vibration is located at 1589 cm–1 and it blue shifts to 1611 cm–1 after its adsorption on modified MMT. The CH2 CH3 deformation at 1455 cm–1 blue shifts to 1457 cm–1, while that at 1400 cm–1 due to the CH2 bending of ortho-substituted benzene blue shifts to 1420 cm–1 after DC adsorption. These shifts suggest participation of the benzene ring, N–H, and CH2 CH3 for the interactions between DC and modified MMT. In comparison to MMT, the FTIR spectra of modified illite do not show much difference before and after DC adsorption (Fig. 6). This is due to the lower amount of DC uptake on the modified illite. Again, the intensities of the 2850 and 2920 cm–1 bands remain the same, indicating no loss of adsorbed CTAB after DC adsorption.
3.6. Removal mechanisms of DC by CTAB-modified minerals The dimension of DC is about 1.0 nm long, 0.5 nm wide, and 0.4 nm high based on the energy-minimized molecular configuration (Fig. 7). With a flat orientation, the surface area is about 0.4 to 0.5 nm2, while the polar surface area of DC was reported as 0.5 nm2 [37]. Using the SSA of 51 m2/g and the Sm of 50 mmol/kg for the illite modified to 2.0 CEC, the DC adsorption density on the illite would be 1.7 nm 2 per DC molecule. This value is much larger than the surface area occupied by DC molecule based on close packing of a monolayer adsorption. For illite modified to 1.0 and 0.5
13
CEC, the DC adsorption densities would be 3.0 and 4.6 nm2 per DC molecule. Thus, the SSA is not a limiting factor for DC adsorption on the illite. In contrast, if the external SSA of 65 m2/g is used for the calculation, the DC sorption density on MMT modified to 2.0 CEC would be 0.11 nm2 per DC molecule. Thus, just the external SSA of MMT is not enough to accommodate monolayer adsorption of DC. If the total SSA of 717 m2/g is used, the DC adsorption density would be 1.2 nm 2 per DC molecule, larger than the surface area of DC. Thus, the interlayer space of MMT must have participated in the DC adsorption and SSA is not a limiting factor for DC adsorption on MMT either. Similarly, for MMT modified to 1.0 and 0.5 CEC, the DC adsorption density would be 2.7 and 3.6 nm2 per DC molecule. As such, for both CTAB-modified illite and MMT, SSA is not a limiting factor at all.
At the 2.0 CEC coverage, the TOC content would be 6.4% and 54% for illite and MMT. The 8 times difference in TOC values resulted in 20 times difference in DC adsorption capacity on these minerals at the 2.0 CEC modification level. Plotting the Sm values against the TOC results in a straight-line relationship with slopes of 18 and 8 mmol/kg-%of TOC (Fig. 7d) for DC adsorption on CTAB-modified MMT and illite, respectively. The interaction of DC with CTAB took place when its concentration was 1.1 mmol/L [38], while the CMC of CTAB was 0.9 mmol/L [28]. At 2.0 CEC coverage, the CTAB molecules form admicelles on the surface of illite and MMT. Partitioning of DC into the hydrophobic domains created due to formation of CTAB admicelles is indeed responsible for the uptake of DC on surfactant-modified clay minerals. In
14
addition, the interlayer space of the swelling clay minerals plays a vital role in increasing DC uptake. The pKa value of DC is in the range of 4.0 – 4.2 [39], while the solution pH under the experimental conditions was higher than 5, suggesting that DC was present in its anionic form DC–. Under 0.5, 1.0, and 2.0 CEC coverage, the adsorbed surfactant would form patch monolayer, monolayer, and admicelle or bilayer surface coverage [40], under which the surface would be partially negative, nearly neutral, and positive. However, the linear correlation of DC adsorption capacity against the TOC of the minerals (Fig. 7d) suggests that the DC uptake was irrelevant to the surface charges, indicating that surface anion exchange may not play a major role in DC uptake on the modified minerals. Thus, the uptake of DC on clay minerals modified to 0.5 and 1.0 CEC may indicate partitioning of DC on to the hemimicelles formed on the mineral surfaces, against suggesting the importance of hydrophobic interactions in the uptake of DC on surfactant-modified clay minerals. In addition, surface complexation between the DC molecules and adsorbed CTAB may also contribute to the increased DC uptake. As both illite and MMT are important minerals in soils and sediments, they could be present in wastewater due to their smaller particle sizes and extended stay in suspension. They are also important natural resources as sorbents for the removal of heavy metals and cationic organic contaminants because of their high CEC and large SSA. This study showed that after being modified by cationic surfactants, their uptake of anionic pharmaceuticals from water increased drastically. With proper engineering
15
designs, they could be readily used in wastewater treatment for the removal of anionic pharmaceuticals.
4. Conclusions This study investigated the interactions of an anionic pharmaceutical diclofenac (DC) sodium with 2:1 clay minerals modified by a cationic surfactant CTAB under different loading levels (different surface or interlayer configurations). The DC adsorption capacities are 50, 28, 18 mmol/kg on illite modified to 2.0, 1.0, and 0.5 CEC, respectively, and are 1000, 450, and 330 mmol/kg on MMT modified to 2.0, 1.0, and 0.5 CEC. Uptake of DC was limited onto the external surface of modified illite while interlayer adsorption contributed significantly to the total uptake of DC on modified MMT. The TOC contents of the clay minerals after CTAB modification played a significant role in DC uptake. Hydrophobic interactions were mostly responsible for the adsorption of DC on the CTAB modified clay minerals. The inexpensive material cost and the high DC adsorption capacity on these modified minerals may extend their potential use as sorbents from the removal of anionic pharmaceuticals in wastewater treatment processes.
Novelty statement This manuscript discusses mechanisms of an anionic drug diclofenac (DC) sodium on surfactant-modified 2:1 clay minerals illite and montmorillonite under different surface configurations. Although some studies were conducted on the uptake of DC on surfactant modified zeolite and montmorillonite. The focus of those reports was for the extended drug release. Moreover, there has been no report on the uptake of DC on surfactant-modified illite. The results in this study address the surfactant loading, 16
thus, the surfactant surface configuration on the uptake of DC based on XRD, FTIR, and contact angle measurements. This latter correlation has not been reported in the literature. The findings would add values to the assessment of extending the use of surfactant-modified clays for the removal of anionic drugs from water.
Acknowledgment This research was supported by the following grants: (1) International Science and Technology Cooperation of China (2014DFA91000); (2) Henan Strategic Projects in Science and Technology (Grant #: 162102210077) for mechanistic studies of oxidative removal of antibiotics from animal manures using iron oxides under the microwave induction; and (3) Key Research Projects of Science and Technology of Ministry of Education of Henan Province (Grant # 70578) for mechanistic studies of interactions between clay minerals and antibiotics in animal manures.
References [1] J.L. Zhou, Z.L. Zhang, E. Banks, D. Grover, J.Q. Jiang, Pharmaceutical residues in wastewater treatment works effluents and their impact on receiving river water, J. Hazard. Mater.166 (2009) 655–661. [2] A. Ziylan, N.H. Ince, The occurrence and fate of anti-inflammatory and analgesic pharmaceuticals in sewage and fresh water: treatability by conventional and non-conventional processes, J. Hazard. Mater. 187 (2011) 24–36. [3] J. Schwaiger, H. Ferling, U. Mallow, H. Wintermayr, R.D. Negele, Toxic effects of the non-steroidal anti-inflammatory drug diclofenac: Part I: histopathological alterations and bioaccumulation in rainbow trout, Aqua. Toxicol. 68 (2004) 141– 150. [4] T.A. Ternes, Occurrence of drugs in German sewage treatment plants and rivers, Water Res. 32 (1998) 3245–3260. [5] M. Cleuvers, Mixture toxicity of the anti-inflammatory drugs diclofenac, ibuprofen, naproxen, and acetylsalicylic acid, Ecotoxicol. Environ. Safety 59 (2004) 309– 315.
17
[6] K. Fent, A.A. Weston, D. Caminada, Ecotoxicology of human pharmaceuticals, Aqua. Toxicol. 76 (2006) 122–159. [7] N. Vieno,M. Sillanpää, Fate of diclofenac in municipal wastewater treatment plant a review, Environ. Int. 69 (2014) 28–39. [8] C. Zwiener, F.H. Frimmel, Short-term tests with a pilot sewage plant and biofilm reactors for the biological degradation of the pharmaceutical compounds clofibric acid, ibuprofen, and diclofenac, Sci. Total Environ. 309 (2003) 201–211. [9] A. Jelić, M. Gros, M. Petrović, A. Ginebreda, D.Barceló, Occurrence and elimination of pharmaceuticals during conventional wastewater treatment, in H. Guasch, A. Ginebreda, A. Geiszinger (eds.), Emerging and Priority Pollutants in Rivers, The Handbook of Environmental Chemistry 19 (2012) 1–24, Springer, Heidelberg. DOI:10.1007/978-3-642-25722-3_1. [10] D. Bendz, N.A. Paxeus, T.R. Ginn, F.J. Loge, Occurrence and fate of pharmaceutically active compounds in the environment, a case study: Höje River in Sweden, J. Hazard. Mater. 122 (2005) 195–204. [11] Y. Zhang, S.-U. Geißen, C. Gal, Carbamazepine and diclofenac: Removal in wastewater treatment plants and occurrence in water bodies, Chemosphere 73 (2008) 115 –1161. [12] D. Cherik, M. Benali, K. Louhab, Occurrence, ecotoxicology, removal of diclofenac by adsorption on activated carbon and biodegradation and its effect on bacterial community: A review, World Scientific News 16 (2015) 116–144. [13] V. Ambrogi, G. Fardella, G. Grandolini, L. Perioli, M.C. Tiralti, Intercalation compounds of hydrotalcite-like anionic clays with anti-inflammatory agents, II: uptake of diclofenac for a controlled release formulation, AAPS Pharmscitech 3 (2002) 77–82. [14] F.P. Bonina, M.L. Giannossi, L. Medici, C. Puglia, V. Summa, F. Tateo, Diclofenac-hydrotalcite: In vitro and in vivo release experiments, Appl. Clay Sci. 41 (2008) 165–171. [15] L. Perioli, T. Posati, M. Nocchetti, F. Bellezza, U. Costantino, A. Cipiciani, Intercalation and release of antiinflammatory drug diclofenac into nanosized ZnAl hydrotalcite-like compound, Appl. Clay Sci. 53 (2011) 374–378. [16] T.X. Bui, S.Y. Kang, S.H. Lee, H. Choi, Organically functionalized mesoporous SBA-15 as sorbents for removal of selected pharmaceuticals from water, J. Hazard. Mater. 193 (2011) 156–163. [17] C. Serri, B. de Gennaro, L. Catalanotti, P. Cappelletti, A. Langella, M. Mercurio, L. Mayol, M. Biondi, Surfactant-modified phillipsite and chabazite as novel excipients for pharmaceutical applications? Micropor. Mesopor. Mater. 224 (2016) 143–148. [18] A.V. Dordio, S. Miranda, J.P. Prates Ramalho, A.J. Palace Carvalho, Mechanisms of removal of three widespread pharmaceuticals by two clay materials, J. Hazard. Mater. (2016) in review. [19] Z. Li, Sorption of oxyanions and surface anion exchange by surfactant-modified clay minerals, J. Environ. Qual., 28 (1999) 1457–1463.
18
[20] M. Datta, M. Kaur, In Vitro release of sodium diclofenac from poloxamer 188 modified montmorillonite as an oral drug delivery vehicle, Int. J. Pharm. Pharm. Sci. 6 (2014) 100–110. [21] R. Karaman, M. Khamis, M. Quried, R. Halabieh, I. Makharzeh, A. Manassra, J. Abbadi, A. Qtait, S.A. Bufo, A. Nasser, S. Nir, Removal of diclofenac potassium from wastewater using clay-micelle complex, Environ. Technol. 33 (2012) 1279– 1287. [22] D. Tiwari, Efficient use of hybrid materials in the remediation of aquatic environment contaminated with micro-pollutant diclofenac sodium, Chem. Engineer. J. 263 (2015) 364–373. [23] H. Gailhanou, J.C. van Miltenburg, J. Rogez, J. Olives, A. Amouric, E.C. Gaucher, P. Blanc, Thermodynamic properties of anhydrous smectite MX-80, illite IMt-2 and mixed-layer illite–smectite ISCz-1 as determined by calorimetric methods. Part I: heat capacities, heat contents and entropies, Geochim. Cosmochim. Acta 71 (2007) 5463–5473. [24] M. Kahle, C. Stamm, Time and pH-dependent sorption of the veterinary antimicrobial sulfathiazole to clay minerals and ferrihydrite, Chemosphere 68 (2007) 1224–1231. [25] D. Borden, R.F. Giese, Baseline studies of the clay minerals society source clays: cation exchange capacity measurements by the ammonia-electrode method, Clay Clay Miner. 49 (2001) 444–445. [26] A.U. Dogan, M. Dogan, M. Onal, Y. Sarikaya, A. Aburub, D.E. Wurster, Baseline studies of the clay minerals society source clays: specific surface area by the Brunauer Emmett Teller (BET) method, Clay Clay Miner. 54 (2006) 62–66. [27] A.E. Blum, D.D. Eberl, Measurement of surface areas by polyvinylpyrrolidone sorption, (2007) US Patent #7,264,777 B1. [28] Z. Li, R.S. Bowman, Counterion effects on the sorption of cationic surfactant and chromate on natural clinoptilolite, Environ. Sci. Technol. 31 (1997) 2407–2412. [29] A. Bharathi, N.K. Srinivas, G.R. Reddy, M. Veeranjaneyulu, A. Sirisha, P. Kamala, Formulation and in vitro evaluation of diclofenac sodium sustained release matrix tablets using melt granulation technique, Int. J. Res. Pharm. Biomed. Sci. 2 (2011) 788–808. [30] D. Krajišnik, A. Daković, M. Milojević, A. Malenović, M. Kragović, D.B. Bogdanović, V. Dondur, J. Milić, Properties of diclofenac sodium sorption onto natural zeolite modified with cetylpyridinium chloride, Colloids Surf. B 83 (2011) 165–172. [31] V. Rakić, V. Rac, M. Krmar, O. Otman, A. Auroux, The adsorption of pharmaceutically active compounds from aqueous solutions onto activated carbons, J. Hazard. Mater. 282 (2015) 141–149. [32] B. de Gennaro, L. Catalanotti, P. Cappelletti, A. Langella, M. Mercurio, C. Serri, M. Biondi, L. Mayol, Surface modified natural zeolite as a carrier for sustained diclofenac release: A preliminary feasibility study, Colloids Surf. B 130 (2015) 101– 109. [33] S.Y. Lee, W.J. Cho, P.S. Hahn, M. Lee, Y.B. Lee, K.J. Kim, Microstructural 19
changes of reference montmorillonites by cationic surfactants, Appl. Clay Sci. 30 (2005) 174–180. [34] J. Madejova, P. Komadel, Baseline studies of the clay minerals society source clays: infrared methods, Clay Clay Miner. 49 (2001) 410–432. [35] K.H.S. Kung, K.F. Hayes, Fourier transform infrared spectroscopic study of the adsorption of cetyltrimethylammonium bromide and cetylpyridinium chloride on silica, Langmuir 9 (1993) 263–267. [36]M.H. Baki, F. Shemirani, R. Khani, M. Bayat, Applicability of diclofenac– montmorillonite as a selective sorbent for adsorption of palladium (II); kinetic and thermodynamic studies, Anal. Methods 6 (2014) 1875–1883. [37] NIH, 2016. https://pubchem.ncbi.nlm.nih.gov/compound/Diclofenac_sodium#section=Top [38] S. Choudhary, P. Talele, N. Kishore, Thermodynamic insights into drug– surfactant interactions: Study of the interactions of naporxen, diclofenac sodium, neomycin, and lincomycin with hexadecytrimethylammonium bromide by using isothermal titration calorimetry, Colloids Surf. B 132 (2015) 313–321. [39] S. Babić, A.J. Horvat, D.M. Pavlović, M. Kaštelan-Macan, Determination of pK a values of active pharmaceutical ingredients, Trends Anal. Chem. 26 (2007) 1043– 1061. [40] Z. Li, R.S. Bowman, Sorption of perchloroethylene by surfactant-modified zeolite as controlled by surfactant loading, Environ. Sci. Technol. 32 (1998) 2278–2282.
20
Fig. 1. The SEM photos of raw illite (a), illite modified to 0.5 (b), 1.0 (c), and 2.0 (d) CEC and those of raw MMT (e), MMT modified to 0.5 (f), 1.0 (g), and 2.0 (h) CEC.
Fig. 2. The contact angle measurements of raw illite (a), illite modified to 0.5 (b), 1.0 (c), and 2.0 (d) CEC and those of raw MMT (e), MMT modified to 0.5 (f), 1.0 (g), and 2.0 CEC.
21
Fig. 3. DC adsorption isotherms on illite (a) and MMT (b) modified by CTAB to 0.5, 1.0, and 2.0 CEC. The lines are the Langmuir fits to the experimental data.
Fig. 4. Kinetics of DC adsorption on illite (a) and MMT (b) modified to 2.0 CEC. The lines are the pseudo-second order fits to the experimental data. The Insets are the linear forms of the pseudo-second-order kinetics.
22
Fig. 5. The XRD patterns of illite (a) and MMT (b) before and after CTAB modification and DC adsorption. Black: raw mineral; red: after CTAB modification to 2.0 CEC; Blue: after DC uptake on 2.0 CEC modified minerals.
Fig. 6. The FTIR spectra of illite (a) and MMT (b), after their CTAB modification, and after DC uptake on the modified minerals. The FTRI spectra of crystalline CTAB and crystalline DC are also included for comparison.
23
Fig. 7. A DC molecule along front (a), side (b), and top (c) views. Green: Cl; Blue: N; Red: O; Dark gray: C; and light gray: H. And plots of DC adsorption capacity against TOC content for modified MMT with the right y-axis and modified illite with the left y-axis (d).
24
S0304-3894(16)30469-1 http://dx.doi.org/doi:10.1016/j.jhazmat.2016.05.038 HAZMAT 17727
To appear in:
Journal of Hazardous Materials
Received date: Revised date: Accepted date:
3-3-2016 20-4-2016 11-5-2016
Please cite this article as: Ken Sun, Yan Shi, Honghan Chen, Xiaoyu Wang, Zhaohui Li, Extending surfactant-modified 2:1 clay minerals for the uptake and removal of diclofenac from water, Journal of Hazardous Materials http://dx.doi.org/10.1016/j.jhazmat.2016.05.038 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Extending surfactant-modified 2:1 clay minerals for the uptake and removal of diclofenac from water Ken Suna,b, Yan Shia, Honghan Chenb,*, Xiaoyu Wangc, Zhaohui Lic,d,* aSchool
of Environmental and Municipal Engineering, North China University of
Water Resources and Electric Power, Zhengzhou 450045,China bSchool
of Water Resources and Environment, China University of Geosciences,
Beijing 100083, China cSchool
of Materials Science and Technology, China University of Geosciences,
Beijing 100083, China dGeosciences
Department, University of Wisconsin–Parkside, Kenosha, WI
53144, USA
Highlights:
Illite and montmorillonite modified by a cationic surfactant.
Modification resulted in significant uptake of diclofenac (DC) upto 1 mmol/g.
Specific surface area and the anion exchange not limiting factors for DC uptake.
Partitioning into the hemimicelles and admicelles responsible for DC uptake.
Results would extend application of modified clays for removal of anionic drugs.
Abstract The presence and persistency of pharmaceuticals and personal care products
*
Corresponding author: Tel: 1-262-595-2487; Fax: 1-262-595-2056; Email: [email protected] 1
(PPCPs) in the environment attracted great attention recently. Among them, antibiotics and pain-killers accounted for a large quantity. Although many works were devoted to the investigation of their removal in wastewater treatment processes, most of the PPCPs studied were of cationic nature. The net repulsive interactions between anionic PPCPs and negatively charged sorbents make them difficult to be removed in wastewater treatment. In this study, 2:1 clay minerals illite and montmorillonite (MMT) were modified with different amounts of cationic surfactant cetyltrimethylammoium bromide (CTAB). The types and sites of interactions between the surfactant-modified clays and the anionic drug diclofenac (DC) were investigated. The uptake of DC on the modified clays was controlled by the CTAB loading level and its surface configuration on the clays. The adsorption sites of DC were limited to the external surfaces of modified illite due to its non-swelling nature. On the contrary, both the external and interlayer sites were available for the adsorption of DC on modified MMT. A CTAB bilayer formation resulted in significant increase in DC adsorption due to the formation of extensive admicelles. FTIR results showed participation of the benzene ring, N–H, and CH2 CH3 for the interactions between DC and modified MMT, suggesting that partitioning of DC into the admicelles of CTAB played a significant role in DC uptake. The results could extend the application of surfactant-modified clays for the removal of anionic PPCPs in the wastewater treatment processes.
Keywords: adsorption, cationic surfactant, clay minerals, diclofenac, hydrophobic interactions, modification
2
1. Introduction There have been great concerns on the detection of pharmaceuticals and personal care products (PPCPs) in soil, sediments, surface and ground water [1,2]. Among the PPCPs, antibiotics and anti-inflammatory medicines accounted for a large quantity. The non-steroidal inflammatory drug diclofenac (DC) is one of the most commonly detected compounds [3], with a 90-percentile of 1.6 µg/L detected in all 49 wastewater treatment plants (WWTPs) and rivers sampled [4]. Its half-maximal effective concentration (EC50) values were 68 and 72 mg/L in the Daphnia and algal tests in comparison to 166 and 626 mg/L for naproxen [5]. Its lowest observed effective concentrations for fish toxicity were in the ranges of wastewater concentrations, which were reported as 0.3 – 3 µg/L for the influent and 0.3 – 2.5 µg/L for the effluent [6]. In wastewater treatment, biologic degradation pathway of DC was a complex process [7]. Short term investigation of its biodegradation with a pilot sewage plant and biofilm reactors as model systems under oxic and anoxic conditions showed no elimination and its final concentration was approximately 95% of their initial concentration after 48 – 55 h of reaction [8]. Even during conventional treatment, it was only barely removable [9]. A removal efficiency of 22% was calculated in the sewage treatment plant near the Höje River in Sweden [10]. A summary of 20 studies showed its removal efficiencies in WWTPs varied from 0 to 80%, mostly in the scope of 21–40% [11]. In addition, adsorption on the secondary and primary sludge was
3
<5% and 5–15%, respectively [7]. Adsorption process on activated carbon (AC) was most commonly used to remove DC [12]. For its adsorptive removal, materials of positive charges were commonly deployed. DC could be effectively intercalated into the interlayer of layered Mg-Al hydrotalcite with a drug loading up to 55% by weight [13]. However, most of the studies on the uptake of DC by hydrotalcite were focused on extended drug releases [14,15]. As most minerals bear negative charges, they were often modified to reverse the charge in order to increase their uptake of negatively charges organic species. Modification of mesoporous silica with aminopropyl and trimethylsilyl resulted in an increase of the solute distribution coefficient Kd value by 20 and 1000 folds, which was attributed to increased hydrophobicity [16]. Modification of clinoptilolite, phillipsite, and chabazite with a cationic surfactant cetylpyridinium chloride (CP) increased in DC adsorption capacities to 49, 54, and 21 mg/g, respectively [17]. Natural clay minerals have negative charges due to isometric substitution in the tetrahedral and octahedral sites. As clay minerals are inexpensive in material cost, they were studied extensively for the removal of heavy metals and cationic drugs. For the adsorptive removal of gemfibrozil, mefenamic acid, and naproxen, light expanded clay aggregates showed superior removal properties in comparison to vermiculite due to its low density and, thus, low weight of material used [18]. For the uptake of anionic species, clay minerals were often modified with long chain cationic surfactants to reverse the surface charges [19]. Montmorillonite (MMT) has large specific surface
4
area (SSA) and high cation exchange capacity (CEC) and is commonly studied for its removal of anionic contaminants after modifications [19]. MMT modified by poloxamer 188 (F-68) resulted in a DC adsorption of 108 mg/g [20]. A significant increase in binding between DC and clay-admicelle complex was attributed to large surface area and hydrophobic domains created in the complex after addition of surfactant [21]. Column results showed a DC loading capacity up to 19 mg/g for bentonite modified with a cationic surfactant hexadecyltrimethylammonium bromide [22]. The extensive uptake of DC was mainly due to the expandable nature of MMT. In comparison to MMT, illite is also a common soil mineral with a higher charge density, which is compensated by K+ in the interlayer, and as a result, it is non-expandable with much lower SSA and CEC values. As such, its utilization was limited. However, due to the similarity between MMT and illite in the stacking of tetrahedral and octahedral sheets, investigating and contrasting the differences in their uptake of pharmaceuticals after being modified by cationic surfactants would provide ultimate mechanisms of the interactions between the pharmaceuticals and the modified clay minerals. Thus, the goal of this study was to contrast the DC adsorption on surfactant-modified MMT from that on modified illite in order to elucidate the mechanistic similarity or differences on their DC uptake. Ultimately, their potential use to removal anionic phamaceuticals could be optimized in future studies.
2. Materials and methods 2.1. Preparation of surfactant-modified minerals
5
The illite used was IMt-2, while the montmorillonite (MMT) used was SAz-2, a Ca-montmorillonite. Both were obtained from the Source Clay Minerals Repository (Purdue University, West Lafayette, IN). They replaced illite IMt-1 and MMT SAz-1, which were well characterized for their chemical and physical properties but were exhausted and no longer available. The XRD pattern of the illite showed dominant illite with 7.5% quartz, 2% microcline, 0.9% rutile, and 0.3% kaolinite [23]. The reported CEC was 140 mmolc/kg and SSA was 51 m2/g for IMt-1 [24]. For the MMT, the CEC was 1200 mmolc/kg [25]. The external SSA of SAz-1 based on multi-point BET measurements was 65 m2/g [26]. If the MMT was assumed as fundamental particles in aqueous suspension, its calculated SSA would be 717 m2/g [27]. The surfactant used was cetyltrimethylammonium bromide (CTAB) (from Sigma-Aldrich) with a critical micelle concentration (CMC) of 0.9 mmol/L [28]. To ensure admicelle sorption of CTAB, the initial concentrations of CTAB should be higher than this value during MMT and illite modification. The DC used was in a sodium form (CAS#: 15307-79-6), which is water-soluble [29] with a formula mass of 318.14 g/mol. It was purchased from Sigma-Aldrich with a purity > 99%. To investigate the effect of surfactant surface coverage on DC uptake, the MMT and illite were modified to 0.5, 1.0 and 2.0 CEC values of the minerals. To do so, 3 g of MMT was combined with 180 mL of CATB solution or 30 g of illite was combined with 210 mL of CATB solution at the concentrations of 10, 20, and 40 mmol/L. The mixtures were shaken for 24 h at 150 rpm and then centrifuged at 3500 rpm for 20 min. The supernatants were analyzed for their equilibrium CTAB concentrations,
6
which were less than 0.1 mmol/L, confirming achieving the target CTAB loadings on these minerals. The mixtures were then washed with two portion of deionized water to remove the excess counterion bromide, before being air-dried.
2.2. Adsorption of DC by modified clays To each 50-mL centrifuge tubes, 0.2 g of modified illite was combined with 10-mL DC solutions at concentrations of 0.5 to 2 mmol/L for 24 h. For modified MMT, 0.07 g was combined with 20-mL of DC solutions at concentrations of 0.5 to 5 mmol/L for 24 h. For the kinetic study, the time for mixing varied from 0.1 to 24 h at an initial DC concentration of 5 mmol/L for MMT and 1.7 mmol/L for illite. Although the initial DC concentrations were much higher than its concentrations found in the environment, it was necessary to investigate the DC adsorption capacity on these minerals modified by CTAB to different surface loading levels and their mechanism of interactions with the modified clays.
2.3 Instrumental analyses The SEM observation was made on JSM-IT300 scanning electron microscope under a voltage of 20.0 kV. The contact angle was measured using ZH10692 goniometer with adjustable LED backlight to achieve a better contrast of the liquid drops. The images were scanned at 70 frames per second with a resolution of 2048*1536. The magnification was increased from 0.7 to 4.5 times in a continuous mode. The accuracy was 0.1°. The XRD analyses were conducted on D-8 Advances
7
with a CuK radiation at 40 kV and 100 mA, a scanning speed at 8°/min and 0.02° per step from 3 to 70°. The FTIR analyses were performed on TENSOR 27 FTIR (Bruker) in the range of 400 – 4000 cm–1 with a resolution of 0.4 cm–1 and a speed of 65 scans per second. The equilibrium DC concentration was determined using a UV-Vis method at a wavelength of 276 nm with a linear range for the calibration of 0.01 to 0.07 mM.
3. Results and discussion 3.1. SEM observation The goal of the SEM observation was to determine if there were morphological changes after the minerals being modified to different levels of CTAB. The surfaces of the raw minerals are relatively smooth (Fig. 1a and 1e). After modified to 0.5, 1.0, and 2.0 CEC, the their surfaces are covered with small particles progressively, suggesting better adhesion of the smaller particles onto the large ones (Fig. 1). This could be attributed to a change in surface hydrophobicity and thus, the free energy of adhesion. The increase in surfactant loading would result in enhanced van der Waal’s interactions due to increases in organic molecules on the surface of these hydrophilic minerals and formation of progressive hydrophobic surfaces, which was confirmed in contact angle measurements and FTIR spectra (see sections 3.2 and 3.5), and thus, a reduction for free energy of adhesion. The increased interaction would result in better partitioning of organic molecules onto the surface of the minerals.
8
3.2. Contact angle measurement To confirm the speculation of adhesive interactions of minerals after progressive surfactant modification, the contact angles between water and the mineral surfaces were measured using a contact goniometer. The contact angles are 21.5°, 42.5°, 43°, and 57° for raw illite and illite modified to 0.5, 1.0, and 2.0 CEC (Fig. 2a-d). Similarly, the contact angles increased from 51.5°, 53.5°, 60.5°, to 76.5° for raw MMT and MMT modified to 0.5, 1.0, and 2.0 CEC (Fig. 2e-h). The progressive increase in contact angles indicate that the mineral surfaces are becoming more hydrophobic as the surfactant loading increased. The increased hydrophobicity would result in enhanced uptake of hydrophobic molecules on the surfaces of the minerals.
3.3. DC adsorption isotherm The uptake of DC on surfactant-modified illite and MMT increases as the surfactant surface coverage increased (Fig. 3). The data were fitted to different adsorption models and the Langmuir model fitted the data best. The Langmuir adsorption isotherm has the form:
CS
K L S mCL 1 K LCL
(1)
where CS is the amount of DC adsorbed at equilibrium (mmol/kg), Sm the apparent adsorption capacity (mmol/kg), CL the equilibrium DC concentration (mmol/L), and KL the Langmuir coefficient (L/mmol). The fitted DC adsorption capacities are 50, 28, 18 mmol/kg, while the KL values are 40, 5.6, and 2.8 L/mmol for the illite modified to 2.0, 1.0, and 0.5 CEC, respectively. Similarly, the DC adsorption capacities are 1000, 450, 9
and 330 mmol/kg, while the KL values are 32, 11, and 5 L/mmol for MMT modified to 2.0, 1.0, and 0.5 CEC. As the CEC of the illite is 140 mmolc/kg, at 2.0 CEC, the anion exchange capacity (AEC) would be 140 mmolc/kg as a result of surfactant surface admicelle or bilayer formation. This value is much larger than the DC adsorption capacity of 50 mmol/kg on the illite modified to 2.0 CEC. For MMT its CEC value is 1200 mmolc/kg, which will result in an ACE value of 1200 mmolc/kg under surfactant admicelle or bilayer coverage, in comparison to the DC adsorption capacity of 1000 mmol/kg. These results suggested that the CEC values of the minerals played an important role for CTAB modification, and thus, the total organic carbon (TOC) content of the minerals after modification. In comparison, the DC adsorption capacities on CP-modified zeolite to 1.0, 2.0, and 3.0 of its external CEC values were 70, 135, and 160 mmol/kg [30] and a DC adsorption capacity of 500 mmol/kg was found on activated carbon [31]. Compared with these materials, MMT modified to 2.0 CEC has superior properties on DC uptake and removal, let alone its low material cost.
3.4. DC adsorption kinetics The adsorption of DC on illite and MMT modified by CTAB to 2.0 CEC is almost instantaneous, particularly on illite (Fig. 4). The adsorption data were fitted to several kinetic models and the pseudo-second-order model fitted the experimental data the best:
qt
kqe2t 1 kqet
(2) 10
where qe and qt are amounts of DC adsorbed at equilibrium and at time t (mmol/kg), k is the pseudo-second-order rate constant (kg/mmol-h). The fitted initial rate and rate constant are 10500 mmol/kg-h, and 4.5 kg/mmol-h for DC uptake on illite modified to 2.0 CEC from an initial concentration of 1.7 mmol/L. In contract, the fitted initial rate and rate constant are 17000 mmol/kg-h, and 0.024 kg/mmol-h for DC uptake on MMT modified to 2.0 CEC from an initial concentration of 5 mmol/L. Similarly, DC adsorption on CP-modified zeolite also followed a pseudo-second-order reaction and diffusion through the boundary layer was considered as the rate controlling step of the process [32]. The higher initial rate constant for DC adsorption on modified illite suggested surface uptake of DC. In contrast, the lower rate constant for DC uptake on modified MMT may suggest that most of the adsorption sites were in the interlayer instead of on the surface and could be be due to DC diffusion from the bulk solution into the interlayer space.
3.5. XRD and FTIR characterization To confirm the DC adsorption sites on modified illite and MMT, the XRD patterns of raw minerals, minerals modified to 2.0 CEC, and after DC uptake on modified minerals are shown in Fig. 5. For illite there is essentially no changes for the XRD patterns (Fig. 5a), indicating that the adsorption of CTAB was restricted on the mineral surfaces and the adsorption of DC on modified illite were also on the external surfaces. This confirmed the speculation from the kinetic study due to the non-swelling nature of illite. In contrast, the (001) peak of MMT moves to a lower
11
angle (Fig. 5b), suggesting expansion of the interlayer space after CTAB modification. After DC uptake, the location of the (001) remained the same, suggesting that adsorption of DC did not result in additional interlayer expansion. However, as the adsorbed CTAB molecule would form paraffin-like structure in the interlayer [33], this could create sites for DC to partition or anion exchange with the counterion Br– on the surface or in the interlayer of MMT.
For illite and MMT, the band at 3620 cm–1 (Fig. 6) was attributed to the stretching vibration of structural -OH in the 2:1 layer, while the broader band at 3420 cm–1 was attributed to the adsorbed water on the surface or in the interlayer of MMT [34]. The intensity of the latter band is drastically reduced after CTAB modification, suggesting significant removal of adsorbed or interlayer water. This would drastically increase the hydrophobicity of the modified minerals as confirmed by the contact angle measurements. The bands at 2920 and 2850 cm–1 appear after CTAB modification. These bands were attributed to the asymmetric and symmetric stretching vibrations of C–C in the alkyl chain [35], and their intensities remain more or less the same (Fig. 6), suggesting no changes of CTAB loading before and after DC adsorption.
The characteristic bands of DC are 3430 cm–1 (N–H stretching); 1570 cm–1 (N– H bending); 1505 cm–1 (N–H bending); 1676 cm–1 (C=O stretching); 1452 cm–1 and 1400 cm–1 (CH2 CH3 deformation, CH2 bending of ortho-substituted benzene); 1305 cm–1 and 1150 cm–1 (C–O-bending); 1585 cm–1 (benzene ring carbon–carbon
12
vibration); and 748 cm–1 (C–Cl stretching) [36]. In this study, the N–H bending at 1570 cm–1 is located at 1577 cm–1 and it blue shifts to 1590 cm–1 after its adsorption on modified MMT. Meanwhile, the N–H bending at 1505 cm–1 blue shifts to 1513 cm–1. The benzene ring carbon–carbon vibration is located at 1589 cm–1 and it blue shifts to 1611 cm–1 after its adsorption on modified MMT. The CH2 CH3 deformation at 1455 cm–1 blue shifts to 1457 cm–1, while that at 1400 cm–1 due to the CH2 bending of ortho-substituted benzene blue shifts to 1420 cm–1 after DC adsorption. These shifts suggest participation of the benzene ring, N–H, and CH2 CH3 for the interactions between DC and modified MMT. In comparison to MMT, the FTIR spectra of modified illite do not show much difference before and after DC adsorption (Fig. 6). This is due to the lower amount of DC uptake on the modified illite. Again, the intensities of the 2850 and 2920 cm–1 bands remain the same, indicating no loss of adsorbed CTAB after DC adsorption.
3.6. Removal mechanisms of DC by CTAB-modified minerals The dimension of DC is about 1.0 nm long, 0.5 nm wide, and 0.4 nm high based on the energy-minimized molecular configuration (Fig. 7). With a flat orientation, the surface area is about 0.4 to 0.5 nm2, while the polar surface area of DC was reported as 0.5 nm2 [37]. Using the SSA of 51 m2/g and the Sm of 50 mmol/kg for the illite modified to 2.0 CEC, the DC adsorption density on the illite would be 1.7 nm 2 per DC molecule. This value is much larger than the surface area occupied by DC molecule based on close packing of a monolayer adsorption. For illite modified to 1.0 and 0.5
13
CEC, the DC adsorption densities would be 3.0 and 4.6 nm2 per DC molecule. Thus, the SSA is not a limiting factor for DC adsorption on the illite. In contrast, if the external SSA of 65 m2/g is used for the calculation, the DC sorption density on MMT modified to 2.0 CEC would be 0.11 nm2 per DC molecule. Thus, just the external SSA of MMT is not enough to accommodate monolayer adsorption of DC. If the total SSA of 717 m2/g is used, the DC adsorption density would be 1.2 nm 2 per DC molecule, larger than the surface area of DC. Thus, the interlayer space of MMT must have participated in the DC adsorption and SSA is not a limiting factor for DC adsorption on MMT either. Similarly, for MMT modified to 1.0 and 0.5 CEC, the DC adsorption density would be 2.7 and 3.6 nm2 per DC molecule. As such, for both CTAB-modified illite and MMT, SSA is not a limiting factor at all.
At the 2.0 CEC coverage, the TOC content would be 6.4% and 54% for illite and MMT. The 8 times difference in TOC values resulted in 20 times difference in DC adsorption capacity on these minerals at the 2.0 CEC modification level. Plotting the Sm values against the TOC results in a straight-line relationship with slopes of 18 and 8 mmol/kg-%of TOC (Fig. 7d) for DC adsorption on CTAB-modified MMT and illite, respectively. The interaction of DC with CTAB took place when its concentration was 1.1 mmol/L [38], while the CMC of CTAB was 0.9 mmol/L [28]. At 2.0 CEC coverage, the CTAB molecules form admicelles on the surface of illite and MMT. Partitioning of DC into the hydrophobic domains created due to formation of CTAB admicelles is indeed responsible for the uptake of DC on surfactant-modified clay minerals. In
14
addition, the interlayer space of the swelling clay minerals plays a vital role in increasing DC uptake. The pKa value of DC is in the range of 4.0 – 4.2 [39], while the solution pH under the experimental conditions was higher than 5, suggesting that DC was present in its anionic form DC–. Under 0.5, 1.0, and 2.0 CEC coverage, the adsorbed surfactant would form patch monolayer, monolayer, and admicelle or bilayer surface coverage [40], under which the surface would be partially negative, nearly neutral, and positive. However, the linear correlation of DC adsorption capacity against the TOC of the minerals (Fig. 7d) suggests that the DC uptake was irrelevant to the surface charges, indicating that surface anion exchange may not play a major role in DC uptake on the modified minerals. Thus, the uptake of DC on clay minerals modified to 0.5 and 1.0 CEC may indicate partitioning of DC on to the hemimicelles formed on the mineral surfaces, against suggesting the importance of hydrophobic interactions in the uptake of DC on surfactant-modified clay minerals. In addition, surface complexation between the DC molecules and adsorbed CTAB may also contribute to the increased DC uptake. As both illite and MMT are important minerals in soils and sediments, they could be present in wastewater due to their smaller particle sizes and extended stay in suspension. They are also important natural resources as sorbents for the removal of heavy metals and cationic organic contaminants because of their high CEC and large SSA. This study showed that after being modified by cationic surfactants, their uptake of anionic pharmaceuticals from water increased drastically. With proper engineering
15
designs, they could be readily used in wastewater treatment for the removal of anionic pharmaceuticals.
4. Conclusions This study investigated the interactions of an anionic pharmaceutical diclofenac (DC) sodium with 2:1 clay minerals modified by a cationic surfactant CTAB under different loading levels (different surface or interlayer configurations). The DC adsorption capacities are 50, 28, 18 mmol/kg on illite modified to 2.0, 1.0, and 0.5 CEC, respectively, and are 1000, 450, and 330 mmol/kg on MMT modified to 2.0, 1.0, and 0.5 CEC. Uptake of DC was limited onto the external surface of modified illite while interlayer adsorption contributed significantly to the total uptake of DC on modified MMT. The TOC contents of the clay minerals after CTAB modification played a significant role in DC uptake. Hydrophobic interactions were mostly responsible for the adsorption of DC on the CTAB modified clay minerals. The inexpensive material cost and the high DC adsorption capacity on these modified minerals may extend their potential use as sorbents from the removal of anionic pharmaceuticals in wastewater treatment processes.
Novelty statement This manuscript discusses mechanisms of an anionic drug diclofenac (DC) sodium on surfactant-modified 2:1 clay minerals illite and montmorillonite under different surface configurations. Although some studies were conducted on the uptake of DC on surfactant modified zeolite and montmorillonite. The focus of those reports was for the extended drug release. Moreover, there has been no report on the uptake of DC on surfactant-modified illite. The results in this study address the surfactant loading, 16
thus, the surfactant surface configuration on the uptake of DC based on XRD, FTIR, and contact angle measurements. This latter correlation has not been reported in the literature. The findings would add values to the assessment of extending the use of surfactant-modified clays for the removal of anionic drugs from water.
Acknowledgment This research was supported by the following grants: (1) International Science and Technology Cooperation of China (2014DFA91000); (2) Henan Strategic Projects in Science and Technology (Grant #: 162102210077) for mechanistic studies of oxidative removal of antibiotics from animal manures using iron oxides under the microwave induction; and (3) Key Research Projects of Science and Technology of Ministry of Education of Henan Province (Grant # 70578) for mechanistic studies of interactions between clay minerals and antibiotics in animal manures.
References [1] J.L. Zhou, Z.L. Zhang, E. Banks, D. Grover, J.Q. Jiang, Pharmaceutical residues in wastewater treatment works effluents and their impact on receiving river water, J. Hazard. Mater.166 (2009) 655–661. [2] A. Ziylan, N.H. Ince, The occurrence and fate of anti-inflammatory and analgesic pharmaceuticals in sewage and fresh water: treatability by conventional and non-conventional processes, J. Hazard. Mater. 187 (2011) 24–36. [3] J. Schwaiger, H. Ferling, U. Mallow, H. Wintermayr, R.D. Negele, Toxic effects of the non-steroidal anti-inflammatory drug diclofenac: Part I: histopathological alterations and bioaccumulation in rainbow trout, Aqua. Toxicol. 68 (2004) 141– 150. [4] T.A. Ternes, Occurrence of drugs in German sewage treatment plants and rivers, Water Res. 32 (1998) 3245–3260. [5] M. Cleuvers, Mixture toxicity of the anti-inflammatory drugs diclofenac, ibuprofen, naproxen, and acetylsalicylic acid, Ecotoxicol. Environ. Safety 59 (2004) 309– 315.
17
[6] K. Fent, A.A. Weston, D. Caminada, Ecotoxicology of human pharmaceuticals, Aqua. Toxicol. 76 (2006) 122–159. [7] N. Vieno,M. Sillanpää, Fate of diclofenac in municipal wastewater treatment plant a review, Environ. Int. 69 (2014) 28–39. [8] C. Zwiener, F.H. Frimmel, Short-term tests with a pilot sewage plant and biofilm reactors for the biological degradation of the pharmaceutical compounds clofibric acid, ibuprofen, and diclofenac, Sci. Total Environ. 309 (2003) 201–211. [9] A. Jelić, M. Gros, M. Petrović, A. Ginebreda, D.Barceló, Occurrence and elimination of pharmaceuticals during conventional wastewater treatment, in H. Guasch, A. Ginebreda, A. Geiszinger (eds.), Emerging and Priority Pollutants in Rivers, The Handbook of Environmental Chemistry 19 (2012) 1–24, Springer, Heidelberg. DOI:10.1007/978-3-642-25722-3_1. [10] D. Bendz, N.A. Paxeus, T.R. Ginn, F.J. Loge, Occurrence and fate of pharmaceutically active compounds in the environment, a case study: Höje River in Sweden, J. Hazard. Mater. 122 (2005) 195–204. [11] Y. Zhang, S.-U. Geißen, C. Gal, Carbamazepine and diclofenac: Removal in wastewater treatment plants and occurrence in water bodies, Chemosphere 73 (2008) 115 –1161. [12] D. Cherik, M. Benali, K. Louhab, Occurrence, ecotoxicology, removal of diclofenac by adsorption on activated carbon and biodegradation and its effect on bacterial community: A review, World Scientific News 16 (2015) 116–144. [13] V. Ambrogi, G. Fardella, G. Grandolini, L. Perioli, M.C. Tiralti, Intercalation compounds of hydrotalcite-like anionic clays with anti-inflammatory agents, II: uptake of diclofenac for a controlled release formulation, AAPS Pharmscitech 3 (2002) 77–82. [14] F.P. Bonina, M.L. Giannossi, L. Medici, C. Puglia, V. Summa, F. Tateo, Diclofenac-hydrotalcite: In vitro and in vivo release experiments, Appl. Clay Sci. 41 (2008) 165–171. [15] L. Perioli, T. Posati, M. Nocchetti, F. Bellezza, U. Costantino, A. Cipiciani, Intercalation and release of antiinflammatory drug diclofenac into nanosized ZnAl hydrotalcite-like compound, Appl. Clay Sci. 53 (2011) 374–378. [16] T.X. Bui, S.Y. Kang, S.H. Lee, H. Choi, Organically functionalized mesoporous SBA-15 as sorbents for removal of selected pharmaceuticals from water, J. Hazard. Mater. 193 (2011) 156–163. [17] C. Serri, B. de Gennaro, L. Catalanotti, P. Cappelletti, A. Langella, M. Mercurio, L. Mayol, M. Biondi, Surfactant-modified phillipsite and chabazite as novel excipients for pharmaceutical applications? Micropor. Mesopor. Mater. 224 (2016) 143–148. [18] A.V. Dordio, S. Miranda, J.P. Prates Ramalho, A.J. Palace Carvalho, Mechanisms of removal of three widespread pharmaceuticals by two clay materials, J. Hazard. Mater. (2016) in review. [19] Z. Li, Sorption of oxyanions and surface anion exchange by surfactant-modified clay minerals, J. Environ. Qual., 28 (1999) 1457–1463.
18
[20] M. Datta, M. Kaur, In Vitro release of sodium diclofenac from poloxamer 188 modified montmorillonite as an oral drug delivery vehicle, Int. J. Pharm. Pharm. Sci. 6 (2014) 100–110. [21] R. Karaman, M. Khamis, M. Quried, R. Halabieh, I. Makharzeh, A. Manassra, J. Abbadi, A. Qtait, S.A. Bufo, A. Nasser, S. Nir, Removal of diclofenac potassium from wastewater using clay-micelle complex, Environ. Technol. 33 (2012) 1279– 1287. [22] D. Tiwari, Efficient use of hybrid materials in the remediation of aquatic environment contaminated with micro-pollutant diclofenac sodium, Chem. Engineer. J. 263 (2015) 364–373. [23] H. Gailhanou, J.C. van Miltenburg, J. Rogez, J. Olives, A. Amouric, E.C. Gaucher, P. Blanc, Thermodynamic properties of anhydrous smectite MX-80, illite IMt-2 and mixed-layer illite–smectite ISCz-1 as determined by calorimetric methods. Part I: heat capacities, heat contents and entropies, Geochim. Cosmochim. Acta 71 (2007) 5463–5473. [24] M. Kahle, C. Stamm, Time and pH-dependent sorption of the veterinary antimicrobial sulfathiazole to clay minerals and ferrihydrite, Chemosphere 68 (2007) 1224–1231. [25] D. Borden, R.F. Giese, Baseline studies of the clay minerals society source clays: cation exchange capacity measurements by the ammonia-electrode method, Clay Clay Miner. 49 (2001) 444–445. [26] A.U. Dogan, M. Dogan, M. Onal, Y. Sarikaya, A. Aburub, D.E. Wurster, Baseline studies of the clay minerals society source clays: specific surface area by the Brunauer Emmett Teller (BET) method, Clay Clay Miner. 54 (2006) 62–66. [27] A.E. Blum, D.D. Eberl, Measurement of surface areas by polyvinylpyrrolidone sorption, (2007) US Patent #7,264,777 B1. [28] Z. Li, R.S. Bowman, Counterion effects on the sorption of cationic surfactant and chromate on natural clinoptilolite, Environ. Sci. Technol. 31 (1997) 2407–2412. [29] A. Bharathi, N.K. Srinivas, G.R. Reddy, M. Veeranjaneyulu, A. Sirisha, P. Kamala, Formulation and in vitro evaluation of diclofenac sodium sustained release matrix tablets using melt granulation technique, Int. J. Res. Pharm. Biomed. Sci. 2 (2011) 788–808. [30] D. Krajišnik, A. Daković, M. Milojević, A. Malenović, M. Kragović, D.B. Bogdanović, V. Dondur, J. Milić, Properties of diclofenac sodium sorption onto natural zeolite modified with cetylpyridinium chloride, Colloids Surf. B 83 (2011) 165–172. [31] V. Rakić, V. Rac, M. Krmar, O. Otman, A. Auroux, The adsorption of pharmaceutically active compounds from aqueous solutions onto activated carbons, J. Hazard. Mater. 282 (2015) 141–149. [32] B. de Gennaro, L. Catalanotti, P. Cappelletti, A. Langella, M. Mercurio, C. Serri, M. Biondi, L. Mayol, Surface modified natural zeolite as a carrier for sustained diclofenac release: A preliminary feasibility study, Colloids Surf. B 130 (2015) 101– 109. [33] S.Y. Lee, W.J. Cho, P.S. Hahn, M. Lee, Y.B. Lee, K.J. Kim, Microstructural 19
changes of reference montmorillonites by cationic surfactants, Appl. Clay Sci. 30 (2005) 174–180. [34] J. Madejova, P. Komadel, Baseline studies of the clay minerals society source clays: infrared methods, Clay Clay Miner. 49 (2001) 410–432. [35] K.H.S. Kung, K.F. Hayes, Fourier transform infrared spectroscopic study of the adsorption of cetyltrimethylammonium bromide and cetylpyridinium chloride on silica, Langmuir 9 (1993) 263–267. [36]M.H. Baki, F. Shemirani, R. Khani, M. Bayat, Applicability of diclofenac– montmorillonite as a selective sorbent for adsorption of palladium (II); kinetic and thermodynamic studies, Anal. Methods 6 (2014) 1875–1883. [37] NIH, 2016. https://pubchem.ncbi.nlm.nih.gov/compound/Diclofenac_sodium#section=Top [38] S. Choudhary, P. Talele, N. Kishore, Thermodynamic insights into drug– surfactant interactions: Study of the interactions of naporxen, diclofenac sodium, neomycin, and lincomycin with hexadecytrimethylammonium bromide by using isothermal titration calorimetry, Colloids Surf. B 132 (2015) 313–321. [39] S. Babić, A.J. Horvat, D.M. Pavlović, M. Kaštelan-Macan, Determination of pK a values of active pharmaceutical ingredients, Trends Anal. Chem. 26 (2007) 1043– 1061. [40] Z. Li, R.S. Bowman, Sorption of perchloroethylene by surfactant-modified zeolite as controlled by surfactant loading, Environ. Sci. Technol. 32 (1998) 2278–2282.
20
Fig. 1. The SEM photos of raw illite (a), illite modified to 0.5 (b), 1.0 (c), and 2.0 (d) CEC and those of raw MMT (e), MMT modified to 0.5 (f), 1.0 (g), and 2.0 (h) CEC.
Fig. 2. The contact angle measurements of raw illite (a), illite modified to 0.5 (b), 1.0 (c), and 2.0 (d) CEC and those of raw MMT (e), MMT modified to 0.5 (f), 1.0 (g), and 2.0 CEC.
21
Fig. 3. DC adsorption isotherms on illite (a) and MMT (b) modified by CTAB to 0.5, 1.0, and 2.0 CEC. The lines are the Langmuir fits to the experimental data.
Fig. 4. Kinetics of DC adsorption on illite (a) and MMT (b) modified to 2.0 CEC. The lines are the pseudo-second order fits to the experimental data. The Insets are the linear forms of the pseudo-second-order kinetics.
22
Fig. 5. The XRD patterns of illite (a) and MMT (b) before and after CTAB modification and DC adsorption. Black: raw mineral; red: after CTAB modification to 2.0 CEC; Blue: after DC uptake on 2.0 CEC modified minerals.
Fig. 6. The FTIR spectra of illite (a) and MMT (b), after their CTAB modification, and after DC uptake on the modified minerals. The FTRI spectra of crystalline CTAB and crystalline DC are also included for comparison.
23
Fig. 7. A DC molecule along front (a), side (b), and top (c) views. Green: Cl; Blue: N; Red: O; Dark gray: C; and light gray: H. And plots of DC adsorption capacity against TOC content for modified MMT with the right y-axis and modified illite with the left y-axis (d).
24