- Email: [email protected]
Using polymer coated nanoparticles for adsorption of micropollutants from water
Accepted Manuscript Title: Using polymer coated nanoparticles for adsorption of micropollutants from water Authors: Mohammad Alizadeh Fard, Ali Vosoogh, Brian Barkdoll, Behnoush Aminzadeh PII: DOI: Reference:
S0927-7757(17)30763-X http://dx.doi.org/doi:10.1016/j.colsurfa.2017.08.008 COLSUA 21870
To appear in:
Colloids and Surfaces A: Physicochem. Eng. Aspects
Received date: Revised date: Accepted date:
4-4-2017 23-7-2017 9-8-2017
Please cite this article as: Mohammad Alizadeh Fard, Ali Vosoogh, Brian Barkdoll, Behnoush Aminzadeh, Using polymer coated nanoparticles for adsorption of micropollutants from water, Colloids and Surfaces A: Physicochemical and Engineering Aspectshttp://dx.doi.org/10.1016/j.colsurfa.2017.08.008 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.
Using polymer coated nanoparticles for adsorption of micropollutants from water
Mohammad Alizadeh Farda,*, Ali Vosooghb, Brian Barkdolla, Behnoush Aminzadehc
a Department
of Civil and Environmental Engineering, Michigan Technological University,
1400 Townsend Dr., Houghton, MI 49931, USA b
Department of Hydraulics and Environment, School of Civil Engineering, Iran University
of Science and Technology, P.O. Box: 16765-163, Tehran, Iran. c Department
of Environmental Engineering, Graduate Faculty of Environment, University
of Tehran, P.O. Box: 1417853111, Tehran, Iran
*Corresponding author. Tel.: +1 906 231 3654, fax: +1 906 487 2943. Email address: [email protected] (M. Alizadeh Fard)
Graphical abstract
1
Highlights
A new hydrothermal method was used to synthesize PVP-coated NPs.
PVP-coated NPs removed >89% of the micropollutants tested.
Thermodynamic analyses revealed endothermic nature of the adsorption process.
Regeneration experiments showed the reliability of methanol regenerating method.
Abstract In this study, polyvinylpyrrolidone (PVP)-coated magnetite nanoparticles were synthesized to adsorb six emerging contaminants (Tonalide, Bisphenol-A, Triclosan, Metolachlor, Ketoprofen and Estriol) from aqueous solutions. The PVP-coated NPs were characterized by dynamic light scattering (DLS) measurements, thermal analysis, and Xray diffraction. Results indicated that PVP-coated NPs were successfully used as a separable adsorbent for removing the micropollutants from water. Adsorption results were modeled using Langmuir and Freundlich isotherms, which showed a better fit of data to the Langmuir model. The adsorbent showed good adsorption performance in which Bisphenol-A and Ketoprofen were the most effectively removed micropollutants, with 98 and 95% removal using only 0.1 mg of the adsorbent within 15 minutes of contact time, respectively. Kinetic studies were performed using the pseudo-second-order model to compare the performance of PVP-coated NPs with granular activated carbon (GAC) revealing the superiority of PVP-coated NPs over GAC. Thermodynamic parameters of the adsorption of the micropollutants onto the adsorbate showed the adsorption process was endothermic and spontaneous. Adsorbate regeneration studies were performed with methanol, ethanol, and the UV/H2O2 process in five regeneration cycles. Methanol treatment ensured the highest level of regeneration.
2
Keywords: Micropollutant removal; Separable adsorbent; adsorption; Adsorbent regeneration.
1. Introduction Nowadays, it is becoming difficult to find unpolluted water resources for drinking purposes. Water resource contamination can emanate from natural sources and/or anthropogenic activities [1, 2]. Industrial development and population growth have imposed thousands of synthetic organic compounds to aquatic fresh ecosystems[3]. Micropollutants, such as pesticides, pharmaceuticals, industrial chemicals, and hormones are present at very low concentrations, and can raise significant toxicological concerns for aquatic ecosystems, and thereby human health [4, 5]. Independent of source, a large degree of the micropolloutants have been found in water resources because they are not bio-degradable [6]. The persistent nature of these compounds means that if released into nature the contaminants will stay in the soil for a long time before reaching the groundwater. Even if they reach the wastewater treatment plant, they are likely to be released with wastewater effluent to surface water bodies [7, 8]. Recently researchers have focused on using adsorption techniques to remove these emerging contaminants from water bodies [9-11]. Activated carbon has been studied widely and showed promising removal efficiency as it has high porosity and internal surface area. However, regeneration of saturated activated carbon is energy demanding (requiring heating to 500–900 oC), and does not fully restore the removal performance [12]. In addition, the probability of releasing carbon dust during the treatment process causes some restrictions over its application in some cases [13, 14]. There are also limitations over application of other adsorbents like zeolite, and carbon nanotubes (CNTs) [15, 16]. Saturated absorbents have to be removed by separation methods such as filtration or centrifugation in order to achieve a free absorbent solution that facilitates regeneration of the absorbent. Applying these costly methods puts constraints on using these absorbents in industrial sites [17].
3
Given the restrictions of current adsorption methods for removing organic micropollutants, novel solutions are being actively sought [18-20]. Polymer-modified iron or iron oxide nanoparticles have been intensively studied for applications in soil and groundwater treatment for various kinds of contaminates including micropollutants [2127]. Phenrat et.al (2016) used a novel electromagnetically enhanced treatment concept for in situ remediation of a source zone of chlorinated dense non-aqueous phase liquid (DNAPL), causing groundwater contamination. Their results showed acceleration in the degradation of the DNAPLs by using polystyrene sulfonate (PSS)-modified zero valent iron NPs in combination with a low frequency AC electromagnetic field. In another study, Bauerlein et.al (2012) [10], studied the removal of some ionic organic compounds by both activated carbon and charged polymeric materials. Their results showed that sorption of all charged compounds to oppositely charged polymers was stronger than to activated carbon, especially for the double-charged cation metformin. Iron oxide nanoparticles (NPs) are attractive for their magnetism and low toxicity characteristics [28]. The former provides easily magnetic separation of NPs from aqueous solutions. Phenrat et.al 2008 [29] showed low toxicity of magnetite in comparison to nanoscale zerovalent iron. In addition, they showed that the addition of a polymer can reduce toxicity even more on mammalian cells (BV2 and N27) due to surface modification. The surface coating can probably change the interaction of the particles with the cell membranes. If the developed adsorbent is able to remove contaminants but cannot be regenerated, it thus will become a contaminant. Regenerating techniques like using organic solvents, inorganic solvents, and redox agents have been developed. In this study two organic solvents and one redox agent were studied to regenerate the saturated PVP-NPs. Composite magnetic materials have been produced by thermal decomposition and used for separation purposes. However, synthesis methods mainly need organic solvents and high temperatures, which make industrial applications difficult [30]. In this study, a simple and inexpensive hydrothermal method was used to produce polyvinylpyrrolidone (PVP)-coated magnetite nanoparticles to adsorb a variety of organic micropollutants from the aqueous phase. This method needs no organic solvents, 4
requires lower energy input, and ambient atmospheric environments. It has been previously shown that PVP-coated NPs are stable in environmental systems [30]. After production of PVP-coated NPs, characteristics of the absorbent, its removal efficiency, and thermodynamic parameters for Tonalide, Bisphenol-A, Triclosan, Metolachlor, Estriol and Ketoprofen were studied.
2. Materials and methods 2.1 Reagents and chemicals Ferrous chloride (FeCl2.4H2O, 99%), ferric chloride (FeCl3.6H2O, 99%), ammonium hydroxide (25%), sodium hydroxide (NaOH), sulfuric acid (98%), hydrogen peroxide (30%), ethanol, and methanol were obtained from Merck (Germany) and used as received. Polyvinylpyrrolidone (PVP) (>95%), Potassium persulfate (KPS), sodium bisulfite (>95%), Suwannee River Fulvic Acid (SRFA), granular activated carbon (GAC), and organic micropollutants (Tonalide, Bisphenol-A, Triclosan, Metolachlor, Ketoprofen and Estriol) were purchased from Sigma-Aldrich (USA). Ultrapure water (Millipore, USA) was used for nanoparticle preparation and kinetic studies. 2.2 Synthesis of nanoparticles Magnetic nanoparticles were synthesized using a hydrothermal method [31]. PVP, ferric chloride, and ferrous chloride were used as precursors, and ammonium hydroxide was used as a precipitation agent. First, 0.50 mmol of PVP was added to 10.0 mL of ultrapure water and was stirred at 90 °C for 2h. Then, 2 mmol of Ferric chloride and 6 mmol of ferrous chloride were added to the solution and stirred at 90 °C for 2h. Then, 10.0 mL of ammonium hydroxide was added into the solution dropwise under continuous stirring. After adding ammonium hydroxide, the color of the mixture turned black and was mixed at 90 °C for 2h. The precipitates were rinsed twice with ultrapure water, separated with a magnet and sonicated to disperse in water. 2.3 Analysis methods
5
Micropollutant
concentrations were
measured by gas chromatography–mass
spectrometry (GC/MS) equipped with a fused capillary column (TRB-5, 60 m, 0.32 mm ID) [5, 32, 33]. The apparatus was operated in the internal ionization mode. A 30 m × 0.25 mm, df= 0.25 μm capillary column SGE (Ringwood, Australia) was used under the following conditions: electron energy=60 eV; transfer-line temperature=200◦C; scan speed= 1000 Hz [4, 8]. Table 1 describes the chemical characteristics and relevance of the tested emerging micropollutants. These compounds were selected from different industry. Total organic carbon (TOC) was measured with a Shimadzu TOC analyzer (5000A, Shimadzu Corp., Japan). The structure of magnetic nanoparticles was analyzed using Cu Ka (k = 0.1542 nm) radiation from an X-ray Diffractometer (Scintag XDS-2000 θ/θ powder diffractometer). The XRD scattering was observed over a range of 10–80o 2θ with step sizes of 0.03 to 0.05° 2θ, and count times between 20 and 55 s. The average crystallite size was calculated according to the full width at the half-maximum height (FWHM) of the (311) plane using Scherrer (Eq.1) [34]: τ=
kγ
(1)
βCOSθ
where τ is the mean size of crystalline domain, k is a dimensionless shape factor (0.89 for magnetite), λ is the X-ray wavelength (1.54059 Å), β is the line broadening at half the maximum intensity (FWHM) after subtracting the instrumental broadening, and θ is Bragg’s angle. By assuming a spherical shape for nanoparticles and using the following equation (Eq.2), particle size (d) can be calculated[35]: 4
d=3 τ
(2)
DLS measurements were performed with a Malvern Zetasizer Nanosizer particle analyzer with a solid-state He–Ne laser at a scattering angle of 1731 at 25 oC (pH 6−8) [36]. For this analysis, all samples were diluted to approximately 0.1 mg/mL. Then, samples were filtered through a Whatman Anotop alumina membrane (pore size of 220 nm). The experiments were performed at NaCl concentrations of 0.001 and 0.16 mol/L at pH 7. All samples were analyzed directly after filtration every three minutes over a one
6
hour period. The average DLS measurements were determined from ten replicates. Zeta potential measurements were also done with a Malvern Zetasizer Nanosizer at 25 oC. Zeta potential measurements were performed by dispersing the sonicated samples (with 10 minutes of sonication) at a concentration of 0.1 mg/mL. The ionic strength was adjusted to 0.001 mol/L NaCl. The pH was initially adjusted to 2.0 and zeta potential was measured by increasing the pH to 12 using sodium hydroxide solution. The average zeta potential and DLS measurements were determined from ten replicates. Thermal analysis was conducted by using a thermogravimetric analyzer (TGA) (Netzsch TG209, Germany) at a heating rate of 10 ºC per minute under constant nitrogen flow [16]. 2.4 Experimental Procedure Adsorption experiments were performed in a 50-mL sealed glass vial. NPs were added to the solution containing 50.0 μg/L of each micropollutant at a pH value of 7.0. Sonication was used for five minutes before conducting the experiments to separate attached nanoparticles (NPs). The pH was adjusted with dilution of NaOH and HCl solutions. Vials were agitated at 150 rpm using a rotary shaker. Containers were wrapped with aluminum foil to minimize photolysis. After conducting the adsorption experiments, nanoparticles were separated with a 1.5 x 0.75 x 0.5 inch neodymium magnet (Grade N 52, K&J Magnetics Inc.) for 30 minutes. To analyze the samples, solutions were collected in 5.0 mL aliquots and sent immediately for analysis. To consider the effect of initial sonication and reduce the chance of analytical uncertainties, a stock sample was maintained with the same experimental conditions except the adsorption process [10]. Results were reported by comparing the final concentration of micropollutants in the stock sample to the experimental samples. All experiments were conducted at least three times and the mean values reported. The isotherms were obtained as follows. PVP-coated NPs (0.1 mg) and the different concentrations of micropollutant were added in a 50 mL sealed glass vial. The micropollutant concentration range was 25–250 µg/L. The vial was shaken at different temperatures (298, 308, and 318 K) for 20 minutes and the precipitation was separated magnetically.
7
Some studies have shown incomplete adsorption of micropollutants in natural samples. To investigate the effect of DOM on adsorption of micropollutants, the adsorption removal efficiencies for both PVP-coated NPs and GAC were compared with distilled water and SRFA-added water samples (DOM = 8 mg C/L). Regeneration experiments were done in a volume of 100 mL. For UV/H2O2 regenerating experiments a solution containing 20 mg/L of hydrogen peroxide was mixed with 2 mg of saturated NPs in a 500 mL four-neck cylindrical quartz reactor for 30 minutes. The reactor was placed on a rotary shaker with rotary speed being the same as the adsorption experiments (150rpm). UV irradiation was provided with one 30 W low-pressure UV mercury lamp placed in the center of the reactor. The residual H2O2 was neutralized with sodium bisulfite [37]. For methanol and ethanol regenerating experiments, all experimental conditions were the same as those with UV/H2O2, except UV irradiation and existence of hydrogen peroxide in which were replaced either by methanol (20 mg/l) or ethanol (20 mg/l). The regenerated adsorbents were reacted again with micropollutant solution to study the recycling rate. The regeneration tests were repeated 5 times.
3. Results and discussions 3.1 Characterizations of PVP-coated nanoparticles Fig. 1a shows the size distribution of synthesized nanoparticles using DLS.
The
hydrodynamic diameter measured by DLS was 106.2 ± 3.6 nm with an average polydispersity index (Pdi) of 0.25. According to the DLS results, NPs are well-dispersed and monodispersed enough for the DSL technique to be valid, since the Pdi Value was less than 0.7 indicating that the sample did not have a broad size distribution and was suitable for the dynamic light scattering technique [36]. The XRD pattern of the PVP-coated iron oxide NPs is shown in Fig. 1b which shows that the dominant phase of NPs is magnetite (Fe3O4), though maghemite was also detected. The calculated lattice parameter α was 8.3889 Å. Since the lattice of maghemite
8
and magnetite are 8.3515 Å (JCPDS 39-1346) and 8.3960 Å (JCPDS 19-629), respectively, the main crystalline phase could be identified as magnetite [19, 38]. Using Eqs. (1) and (2), the average crystal size of 14.3 nm and particle size of 19.7 nm were calculated. The TGA curve for PVP-coated iron oxide nanoparticles is depicted in Fig. 2. The initial weight loss before 240 ºC could be due to the desorption of adsorbed water and/or surface hydroxyls. It can be seen that the sample weight significantly decreased around 300 ºC. This can be ascribed to the formation of iron oxides and the decomposition of amorphous iron hydroxides. The major weight loss took place around 372 ºC which could be due to the decomposition of the coated PVP layer as such a loss did not occur in the uncoated sample [38]. Subsequently, the micropollutant removal capacity of PVP-coated iron oxide NPs was studied. The average zeta potential was 11.5 ± 0.4 mV indicating the sorption of PVP shifts the slip plane out of the surface of the particle and to the effect of adsorbed non-charged polymeric layer as suggested by Ohshima's soft particle analysis which decreases the apparent electrophoretic mobility [29]. 3.2 The effect of dosage and contact time Adsorbent dosage and contact time play significant roles in the adsorption removal of the micropollunts. By Increasing contact time and dosage more chance will be given to the absorbents to adsorb pollutants. On the other hand, either over-dosage of adsorbent or excessive contact time increases the operational costs. Therefore, it is indispensable to optimize the adsorbent dose and contact time. In order to evaluate the adsorptive capacity of the polymer coated nanoparticles, the removal of each micropollutant was studied with different dosages of NPs. Fig. 3(a) shows the effect of adsorbent concentration on the micropollulant removal. The experiments were performed at pH 7.5 and the detention time was 20 min. The adsorption efficiency was improved when adsorbent dose increased from 0.75 to 2.5 mg/L. Higher adsorbent dosage can lead to more adsorbable area. Almost complete removal of micropollutants was observed at the 2 mg/L dose. At the 2.5 mg/L dose, little 9
or no improvement was observed in adsorption efficiencies. On the other hand, the removal efficiencies for Bisphenol-A, Triclosan, Ketoprofen, and Estriol were almost the same for 1.5 and 2 mg/L concentrations. However, the average removal efficiencies for Metolachor and Tonalide at an adsorbent dose of 1.5 mg/L was considerably less than those for 2 mg/L dose. Therefore, 2 mg/L was selected as the optimum dose of NPs. Fig. 3(b) shows the removal of micropollutants due to contact time. The optimal contact time was found to be 20 minutes, at this point most of the micropollutants had reached their equilibrium concentration. At this condition little or no change was observed in their concentration. It should be noted that no sign of oxidation was observed during the experiments and all removal efficiencies can be ascribed to adsorption. Adsorption equilibrium was determined when each micropollutant had been contacted with the PVPcoated NPs for enough time, and adsorbate concentration in the aqueous phase was in a dynamic balance with the interface concentration. At this point almost all of the adsorbate concentration was removed. The concentrations of all micropollutants were sharply reduced in the first 10 minutes of contact time. Concentration reduction continued gradually after 10 minutes. Bisphenol-A and Ketoprofen were the most effectively removed micropollutants, with 98 and 95% removal, respectively, using only 0.1 mg of the adsorbent within 15 min. This could be due to their chemical structures, physiochemical properties, as well as the specific surface properties of the adsorbent. Micropollutants with high polar bonds tend to be adsorbed more easily. As Bisphenol-A, Triclosan, Ketoprofen, and Estriol have O-H bond(s) in their chemical structure that can easily bind to PVP [39]. The next significant factor that can affect adsorption capacity could be molecular weight because Bisphenol-A and Ketoprofen had less molecular weight than Triclosan and Estriol. The change of adsorption capacity with increasing reaction time can be attributed to the following stages [16, 28] conceptualized in Fig. 4. First, the instantaneous adsorption of micropollutants occurred within a short period of time. This step can be attributed to interactive forces between the micrpollutant molecules and the surface of the adsorbent. Second, as the contact time increased, a gradual achievement of the equilibrium was observed, since most of the active sites on the surface of the adsorbent were occupied 10
by the micropollutants. In addition, physisorption was the main adsorptive action for which physical bonding of the micropollutant molecules occurs on the surface of the adsorbent. This takes place due to Van der Waals forces. Finally, the adsorption equilibrium between micropollutant molecules and PVP-coated NPs was attained. To study the amount of each micropollutant adsorbed by PVP-coated NPs and GAC at equilibrium condition, the pseudo-second-order model was used to explain the adsorption kinetics, which is expressed in Eq. 3 [40]: 𝑡 𝑞𝑡
=
1 𝑘𝑞𝑒2
+
𝑡
(3)
𝑞𝑒
where k (µg/mg.min) is the second-order rate constant, qt (µg/mg) is the adsorption capacity at any time t (min), and qe (µg/mg) is the adsorption capacity at equilibrium. The results of adsorption kinetics are tabulated in Table 2. In addition, the kinetic removal curves of the tested micropollutants are presented in Fig. S1. According to the results, all correlation of coefficients were greater than 0.99. The corresponding linear regressions were obtained for the values of qe and k from Eq. (3). For experiments with PVP-coated NPs, the value of qe (18.2 µg/mg) for Tonalide was lower than the other micropollutants. This could be attributed to the difference in molecular structure of the target micropollutants. The k value for Bisphenol-A adsorbed by PVP-coated NPs was 15 times greater than that of the GAC experiments. This indicates a high adsorption capacity of PVP-coated NPs. This is probably due to the availability of more readily-accessible binding sites (located at the outer surface of the adsorbent) on the PVP-NPs compared to GAC. Alsbaiee et al. (2016) [9], also showed rapid adsorption of a variety of organic micropollutants with higher adsorption rate constants than those of activated carbon.
3.3 Adsorption isotherms Adsorption isotherms are valuable curves describing the phenomenon leading the retention or release of a substance from the aqueous to the solid phase at a constant temperature and pH. Langmuir and Freundlich adsorption isotherms were used for modeling adsorption data. The Langmuir isotherm assumes homogeneous adsorption, in 11
which all sites have equal affinity for the adsorbate molecule (each molecule has constant sorption activation energy). The Langmuir adsorption isotherm can be expressed by Eq. (4): 𝐶𝑒 𝑞𝑒
=𝑘
1
𝐶
𝑎 𝑞𝑚
+ 𝑞𝑒
(4)
𝑚
where Ce is the equilibrium concentration in the liquid phase (µg/L), qe is the amount of adsorbate adsorbed on the adsorbent at equilibrium conditions (µg/mg), qm is the maximum theoretical adsorption capacity of the adsorbent (µg/mg), and Ka is the adsorption equilibrium constant (L/µg). An equilibrium parameter or a dimensionless constant of separation factor (RL) can also be expressed as a critical characteristic of the Langmuir isotherm, which is defined as Eq. (5) [41]: 1
𝑅𝐿 = 1+𝐾
(5)
𝑎 𝐶0
where C0 is the initial concentration of micropollutants and RL indicates the shape of the isotherm; values between 0 and 1 indicate favorable adsorption, whereas RL = 0, RL = 1, and RL > 1, indicate irreversible, linear, and unfavorable, adsorption isotherms, respectively [42]. On the other hand, the Freundlich isotherm assumes that there are many types of binding sites which have different free energy of sorption values, and there are a large amount of sorbent sites. This isotherm expresses the diversity of free energies related to the adsorbate by the heterogeneous adsorption [43]. The Freundlich adsorption isotherm can be expressed by Eq. (5)[43]: 1
q𝑒 = K C𝑛
(5)
where K ((µg/mg) (L/µg) 1/n) is the adsorption coefficient of the Freundlich isotherm and 1/n is the Freundlich intensity parameter. If the value of 1/n is lower than 1, it indicates a normal Langmuir isotherm. The amount of micropollutant adsorbed per unit adsorbent
12
(qe) (µg/mg) was evaluated by dividing the amount adsorbed by the concentration of PVPcoated NPs. The linearized form of the Freudlich model is shown below [44, 45] in Eq. (6): 1
Log (q 𝑒 ) = Log(k) + 𝑛 Log(C)
(6)
The adsorption isotherms of Tonalide and Bisphenol-A for both Langmuir and Freundlich isotherms are depicted in Fig. 5 (a) and (b), respectively. Adsorption isotherm plots for other tested micropollutants can be found in Fig. S2 and Fig. S3. Results showed that the adsorption data for all micropollutants were well fitted to the Langmuir model (R2 >0.99). Therefore, a monolayer of the micropollutants was formed on the surface of PVPcoated NPs. Corresponding adsorption parameters for all tested micropollutants are listed in Table 3. According to the results, RL values for Langmuir isotherm were between 0 and 1 and the value of 1/n for Freundlich isotherm was smaller than 1, which indicate favorable adsorption. The maximum adsorption capacities (qm) by the Langmuir model were achieved for Bisphenol-A and Ketoprofen as 90.91 and 83.33 µg/mg, respectively. These two compounds have smaller molecular size in comparison to the other studied compounds. The smaller molecular size can help more molecules to be adsorbed on the surface of the adsorbent. The large adsorption capacity of PVP-coated NPs can be ascribed to the strong adsorption affinity of the adsorbent towards the target micropollutants, which emanated from unique surface characteristics. Table 4 lists the maximum theoretical adsorption capacity (qm) and the correlation coefficients (R2) at three different temperatures (298, 308 and 318 K). All the R2 values were above 0.94, indicating that Langmuir isotherm had acceptable agreement with the experimental data. Results indicated that when the temperature increased, qm values increased. This could be probably due to the interactive forces between adsorbate and adsorbent that control the process. Therefore, it can be concluded that the adsorption of the micropollutants onto PVP-coated NPs is an endothermic process.
13
Thermodynamic parameters such as the change in free energy (ΔG°), enthalpy (ΔH°), and entropy (ΔS°) were calculated for the adsorption of micropollutants using the following equations [12]: C
K c = Ce
(6)
ΔG° = −RT ln(K c )
(7)
S
ln (K c ) =
ΔS° R
−
ΔH°
(8)
RT
where Kc is equilibrium constant, Ce and Cs are the concentrations of micropollutants adsorbed by the adsorbent (µg/L) and remained in the aqueous phase at equilibrium (µg/L) respectively. R is the universal gas constant (8.314 J/mol K), and T is temperature (K). ΔH° and ΔS° can be calculated from the slope and intercept of ln Kc vs. 1/T (Van’t Hoff plot (Fig. 6)). Van’t Hoff plots for adsorption of other tested micropollutants are shown in Fig. S4 and S5. The thermodynamic parameters are summarized in Table 5. Positive values of ΔH° showed endothermic adsorption processes for all micropollutants. In addition, positive values of ΔS° indicated the affinity of the tested micropollutants and increasing randomness at the liquid-solid interface during the adsorption process. Negative values of ΔG° indicated spontaneous adsorption and the degree of spontaneity increased with increasing temperature.
3.4 The effect of dissolved organic matter (DOM) Micropollutants usually exist with DOM in water bodies [46]. Therefore, the possible competition between micropollutants and DOM should be considered when performing an adsorption experiment. Further experiments showed that 5.5 mg/L of PVP-coated NPs concentration was the optimal concentration when SRFA was added to the sample. This increase in the adsorbate concentration was indispensable due to the fact that the SRFA acts as a competitor for PVP-coated NPs meaning that a higher adsorbate concentration is required to remove the same amount of target micropollutants. Fig. 7 showed the average adsorption removals for micropollutants in the distilled water and surface water for both PVP-coated NPs and GAC. According to the results, GAC was more affected by the addition of DOM to the water sample. However, PVP-coated NPs 14
showed a great performance for the same conditions. It took 25 minutes for PVP-coated NPs to reach equilibrium while the equilibrium time for GAC was 55 minutes. The significant decrease in the removal of micropollutants might be caused by the competitive inhibition of DOM with micropollutants on GAC. This result implies that the inhibition of adsorption in the SRFA water samples is due to competition of hydrophobic adsorption interaction of SRFA with micropollutants on the surface of GAC to occupy adsorption sites.
3.5 Adsorbent regeneration Adsorbent regeneration is critical to ensure its sustainability [47]. In this study three regeneration methods were studied. Fig. 8 compares all regeneration methods to regenerate the removal capacity of the adsorbent for five consecutive cycles. This figure reports the average removal efficiency of micropollutants that are based on the micropollutants concentrations in the solution. Tables S1-S3 in the supplementary materials show the removal efficiency of all micropollutants for each regeneration method. According to the results the average removal efficiency for the methanol method was almost the same for all five cycles, while ethanol’s desorption efficiency decreased slightly. This showed that methanol could effectively detach the micropollutants from the surface of the adsorbent due to lower molecular weight, since ethanol (MW = 46) >methanol (MW = 32). This could be due to the fact that smaller molecules can penetrate more easily into the micropores of the adsorbent and displaces the adsorbate more effectively. Because alcohols have a hydroxyl group in their structures, organic micropollutants are more soluble in them. For example, Filippa et al. [48] found the solubility of Ketrophen in water, ethanol, and methanol to be 6.27×10-4, 1.85 and 1.75 mol/L, respectively. Therefore, methanol and ethanol are able to easily desorb the studied micropollutants. On the other hand, the regeneration performance of the UV/H2O2 process was tested as a system that does not have any waste at the end of the regeneration process [49]. This system is recognized as one of the more effective advanced oxidation processes 15
(AOPs) that is able to remove a wide range of contaminants [4, 50]. Organic pollutants adsorbed onto the adsorbent surface can be mineralized into harmless compounds such as carbon dioxide and water. It can be seen from Fig. 8 that the adsorption efficiency decreased when the UV/H2O2 treatment was applied. The main reason could be the nonselective nature of UV/H2O2 process to degrade oxidizable substances. In fact, this system can degrade adsorbed micropollutants and the surface of the coated polymer at the same time. Therefore, the regenerated adsorbent has probably less effective adsorption sites for the next cycles.
Conclusions In this study, PVP-coated NP was successfully produced using a new method and its characterizations were evaluated by different techniques. The adsorbent was successfully used for removing micropollutants from aqueous solutions. The adsorbent dosage, contact time, temperature, and dissolved organic matter competitor had a significant effect on the adsorption performance of PVP-coated NPs. The maximum adsorption capacities of Bisphenol-A and Ketoprofen on the adsorbent at a pH of 7.5 was 90.91 and 83.33 µg/mg, respectively. Results of isotherm modeling indicated that the Langmuir model was well fitted to the experimental data. In addition, the pseudo second order model was used to describe the adsorption kinetics. The thermodynamic parameters of all tested micropollutant adsorption onto the PVP-coated NPs showed that the adsorption process was endothermic and spontaneous. Methanol regeneration studies showed that PVP-coated NPs could be recycled and used for at least five cycles.
Acknowledgements The authors want to thank the department of Civil and Environmental Engineering at Michigan Technological University for financial support.
16
References: [1] M.T. Amin, A.A. Alazba, U. Manzoor, A Review of Removal of Pollutants from Water/Wastewater Using Different Types of Nanomaterials, Advances in Materials Science and Engineering, 2014 (2014) 24. [2] M. Carballa, F. Omil, J.M. Lema, Removal of cosmetic ingredients and pharmaceuticals in sewage primary treatment, Water Research, 39 (2005) 4790-4796. [3] L. Hernández-Leal, H. Temmink, G. Zeeman, C. Buisman, Removal of micropollutants from aerobically treated grey water via ozone and activated carbon, Water Research, 45 (2011) 2887-2896. [4] L. Guzzella, D. Feretti, S. Monarca, Advanced oxidation and adsorption technologies for organic micropollutant removal from lake water used as drinking-water supply, Water Res, 36 (2002) 4307-4318. [5] M. Hijosa-Valsero, V. Matamoros, R. Sidrach-Cardona, J. Martín-Villacorta, E. Becares, J.M. Bayona, Comprehensive assessment of the design configuration of constructed wetlands for the removal of pharmaceuticals and personal care products from urban wastewaters, Water research, 44 (2010) 3669-3678. [6] W. Qi, H. Singer, M. Berg, B. Müller, B. Pernet-Coudrier, H. Liu, J. Qu, Elimination of polar micropollutants and anthropogenic markers by wastewater treatment in Beijing, China, Chemosphere, 119 (2015) 1054-1061. [7] P.A. Quinlivan, L. Li, D.R. Knappe, Effects of activated carbon characteristics on the simultaneous adsorption of aqueous organic micropollutants and natural organic matter, Water research, 39 (2005) 1663-1673. [8] Z.M. Nagy, M. Molnar, I. Fekete-Kertesz, I. Molnar-Perl, E. Fenyvesi, K. Gruiz, Removal of emerging micropollutants from water using cyclodextrin, The Science of the total environment, 485-486 (2014) 711-719. [9] A. Alsbaiee, B.J. Smith, L. Xiao, Y. Ling, D.E. Helbling, W.R. Dichtel, Rapid removal of organic micropollutants from water by a porous β-cyclodextrin polymer, Nature, 529 (2016) 190-194. [10] P.S. Bäuerlein, T.L. ter Laak, R.C.H.M. Hofman-Caris, P. de Voogt, S.T.J. Droge, Removal of charged micropollutants from water by ion-exchange polymers – Effects of competing electrolytes, Water Research, 46 (2012) 5009-5018. 17
[11] X. Cai, Q. Liu, C. Xia, D. Shan, J. Du, J. Chen, Recyclable Capture and Destruction of Aqueous Micropollutants Using the Molecule-Specific Cavity of Cyclodextrin Polymer Coupled with KMnO4 Oxidation, Environmental Science & Technology, 49 (2015) 92649272. [12] A. Jafari Kang, M. Baghdadi, A. Pardakhti, Removal of cadmium and lead from aqueous solutions by magnetic acid-treated activated carbon nanocomposite, Desalination and Water Treatment, 57 (2016) 18782-18798. [13] I. Oller, S. Malato, J.A. Sánchez-Pérez, Combination of Advanced Oxidation Processes and biological treatments for wastewater decontamination—A review, Science of The Total Environment, 409 (2011) 4141-4166. [14] K. Ebie, F. Li, Y. Azuma, A. Yuasa, T. Hagishita, Pore distribution effect of activated carbon in adsorbing organic micropollutants from natural water, Water Research, 35 (2001) 167-179. [15] S.O. Ganiyu, E.D. van Hullebusch, M. Cretin, G. Esposito, M.A. Oturan, Coupling of membrane filtration and advanced oxidation processes for removal of pharmaceutical residues: A critical review, Separation and Purification Technology, 156, Part 3 (2015) 891-914. [16] H. Gao, S. Zhao, X. Cheng, X. Wang, L. Zheng, Removal of anionic azo dyes from aqueous solution using magnetic polymer multi-wall carbon nanotube nanocomposite as adsorbent, Chemical Engineering Journal, 223 (2013) 84-90. [17] S. Giannakis, F.A. Gamarra Vives, D. Grandjean, A. Magnet, L.F. De Alencastro, C. Pulgarin, Effect of advanced oxidation processes on the micropollutants and the effluent organic matter contained in municipal wastewater previously treated by three different secondary methods, Water Research, 84 (2015) 295-306. [18] J. Fresnais, M. Yan, J. Courtois, T. Bostelmann, A. Bée, J.-F. Berret, Poly (acrylic acid)-coated iron oxide nanoparticles: Quantitative evaluation of the coating properties and applications for the removal of a pollutant dye, Journal of colloid and interface science, 395 (2013) 24-30. [19] J.-F. Liu, Z.-s. Zhao, G.-b. Jiang, Coating Fe3O4 magnetic nanoparticles with humic acid for high efficient removal of heavy metals in water, Environmental science & technology, 42 (2008) 6949-6954. 18
[20] D. Morillo, A. Uheida, G. Pérez, M. Muhammed, M. Valiente, Arsenate removal with 3-mercaptopropanoic acid-coated superparamagnetic iron oxide nanoparticles, Journal of colloid and interface science, 438 (2015) 227-234. [21] T. Phenrat, N. Saleh, K. Sirk, H.-J. Kim, R.D. Tilton, G.V. Lowry, Stabilization of aqueous nanoscale zerovalent iron dispersions by anionic polyelectrolytes: adsorbed anionic polyelectrolyte layer properties and their effect on aggregation and sedimentation, Journal of Nanoparticle Research, 10 (2008) 795-814. [22] T. Phenrat, Y. Liu, R.D. Tilton, G.V. Lowry, Adsorbed polyelectrolyte coatings decrease Fe0 nanoparticle reactivity with TCE in water: conceptual model and mechanisms, Environmental Science & Technology, 43 (2009) 1507-1514. [23] T. Phenrat, F. Fagerlund, T. Illangasekare, G.V. Lowry, R.D. Tilton, Polymer-modified Fe0 nanoparticles target entrapped NAPL in two dimensional porous media: effect of particle concentration, NAPL saturation, and injection strategy, Environmental science & technology, 45 (2011) 6102-6109. [24] T. Phenrat, I. Kumloet, Electromagnetic induction of nanoscale zerovalent iron particles accelerates the degradation of chlorinated dense non-aqueous phase liquid: Proof of concept, Water Research, 107 (2016) 19-28. [25] S. Srirattana, K. Piaowan, G.V. Lowry, T. Phenrat, Electromagnetic induction of foambased nanoscale zerovalent iron (NZVI) particles to thermally enhance non-aqueous phase liquid (NAPL) volatilization in unsaturated porous media: Proof of concept, Chemosphere, (2017). [26] B. Jarošová, J. Filip, K. Hilscherová, J. Tuček, Z. Šimek, J.P. Giesy, R. Zbořil, L. Bláha, Can zero-valent iron nanoparticles remove waterborne estrogens?, Journal of environmental management, 150 (2015) 387-392. [27] I. Arslan-Alaton, T. Olmez-Hanci, The Use of Al and Fe Nanoparticles for the Treatment of Micropollutants, in:
Nanotechnologies for Environmental Remediation,
Springer, 2017, pp. 61-114. [28] D.-P. Li, Y.-R. Zhang, X.-X. Zhao, B.-X. Zhao, Magnetic nanoparticles coated by aminoguanidine for selective adsorption of acid dyes from aqueous solution, Chemical engineering journal, 232 (2013) 425-433.
19
[29] T. Phenrat, T.C. Long, G.V. Lowry, B. Veronesi, Partial oxidation (“aging”) and surface modification decrease the toxicity of nanosized zerovalent iron, Environmental science & technology, 43 (2008) 195-200. [30] S. Palchoudhury, J.R. Lead, A facile and cost-effective method for separation of oil– water mixtures using polymer-coated iron oxide nanoparticles, Environmental science & technology, 48 (2014) 14558-14563. [31] S. Mirshahghassemi, J.R. Lead, Oil Recovery from Water under Environmentally Relevant Conditions Using Magnetic Nanoparticles, Environmental Science & Technology, 49 (2015) 11729-11736. [32] C.P. James, E. Germain, S. Judd, Micropollutant removal by advanced oxidation of microfiltered secondary effluent for water reuse, Separation and Purification Technology, 127 (2014) 77-83. [33] V. Matamoros, J. García, J.M. Bayona, Organic micropollutant removal in a full-scale surface flow constructed wetland fed with secondary effluent, Water research, 42 (2008) 653-660. [34] M. Guillemot, J. Mijoin, S. Mignard, P. Magnoux, Adsorption of Tetrachloroethylene on Cationic X and Y Zeolites: Influence of Cation Nature and of Water Vapor, Industrial & Engineering Chemistry Research, 46 (2007) 4614-4620. [35] R.A. Bini, R.F.C. Marques, F.J. Santos, J.A. Chaker, M. Jafelicci Jr, Synthesis and functionalization of magnetite nanoparticles with different amino-functional alkoxysilanes, Journal of Magnetism and Magnetic Materials, 324 (2012) 534-539. [36] W.Y. Chen, DSL: Simulation Techniques and Standards Development for Digital Subscriber Lines, Alpel Publishing, 1997. [37] M.A. Fard, B. Aminzadeh, H. Vahidi, Degradation of petroleum aromatic hydrocarbons using TiO2 nanopowder film, Environmental technology, 34 (2013) 11831190. [38] Z. Shan, W.-S. Yang, X. Zhang, Q.-M. Huang, H. Ye, Preparation and characterization of carboxyl-group functionalized superparamagnetic nanoparticles and t he potential for bio-applications, Journal of the Brazilian Chemical Society, 18 (2007) 1329-1335.
20
[39] H. Chen, H. Luo, Y. Lan, T. Dong, B. Hu, Y. Wang, Removal of tetracycline from aqueous solutions using polyvinylpyrrolidone (PVP-K30) modified nanoscale zero valent iron, Journal of hazardous materials, 192 (2011) 44-53. [40] Y.S. Ho, G. McKay, Pseudo-second order model for sorption processes, Process Biochemistry, 34 (1999) 451-465. [41] S.-C. Tsai, K.-W. Juang, Comparison of linear and nonlinear forms of isotherm models for strontium sorption on a sodium bentonite, Journal of Radioanalytical and Nuclear Chemistry, 243 (2000) 741-746. [42] A. Dada, A. Olalekan, A. Olatunya, O. Dada, Langmuir, Freundlich, Temkin and Dubinin–Radushkevich isotherms studies of equilibrium sorption of Zn2+ unto phosphoric acid modified rice husk, IOSR Journal of Applied Chemistry, 3 (2012) 38-45. [43] K.Y. Foo, B.H. Hameed, Insights into the modeling of adsorption isotherm systems, Chemical Engineering Journal, 156 (2010) 2-10. [44] M. Baghdadi, E. Ghaffari, B. Aminzadeh, Removal of carbamazepine from municipal wastewater effluent using optimally synthesized magnetic activated carbon: Adsorption and sedimentation kinetic studies, Journal of Environmental Chemical Engineering, 4 (2016) 3309-3321. [45] B.A. E. Aghdam, M. Baghdadi, M.A. Fard, Removal of BTEX from aqueous solutions by paper mill sludge-based activated carbon, Journal of Advances in Chemistry, 11 (2015) 3416-3432. [46] R.P. Schwarzenbach, B.I. Escher, K. Fenner, T.B. Hofstetter, C.A. Johnson, U. Von Gunten, B. Wehrli, The challenge of micropollutants in aquatic systems, Science, 313 (2006) 1072-1077. [47] S. Lee, J.-S. Lee, M.-K. Song, J.-C. Ryu, B. An, C.-G. Lee, C. Park, S.-H. Lee, J.-W. Choi, Effective regeneration of an adsorbent for the removal of organic contaminants developed based on UV radiation and toxicity evaluation, Reactive and Functional Polymers, 95 (2015) 62-70. [48] M.A. Filippa, G.M. Melo, E.I. Gasull, Ketoprofen Solubility in Organic Solvents and Aqueous Co-solvent Systems: Interactions and Thermodynamic Parameters of Solvation.
21
[49] S. Notthakun, J.C. Crittenden, D.W. Hand, D.L. Perram, M.E. Mullins, Regeneration of adsorbents using heterogeneous advanced oxidation, Journal of Environmental Engineering, 119 (1993) 695-714. [50] M. Alizadeh Fard, A. Torabian, G.R.N. Bidhendi, B. Aminzadeh, Fenton and photoFenton oxidation of petroleum aromatic hydrocarbons using nanoscale zero-valent iron, Journal of Environmental Engineering, 139 (2013) 966-974.
22
Figure captions
Fig. distribution plot, (b) XRD pattern for the PVP-coated NPs.
23
1
(a)
DLS
size
Fig. 2 TGA plot for the PVP-coated NPs.
24
Fig. 3 Effect of (a) NP dose (micropollutant concentration of 50 µg/L) and (b) contact time (NP concentration of 2 mg/L) on micropollutants removal efficiencies. The data are reported as the average of at least triplicate experiments. Error bars indicate the minimum and maximum removal efficiencies.
25
Fig. Conceptual model of adsorption of the micropollutants.
26
4
Fig.
5
Langmuir (a) and Freundlich (b) isotherms for the adsorption of Tonalide and BisphenolA onto PVP-coated NPs.
27
Fig. 6 Van’t Hoff plot for adsorption of Tonalide and Bisphenol-A onto PVP-coated NPs (initial concentration of 100 µg/L, adsorbent dosage of 2 mg/L).
Fig. 7 Average adsorption removal efficiencies of PVP-coated NPs and GAC. GAC concentrations were the same as NPs for the same experiment. Error bars indicate the minimum and maximum removal efficiencies for all tested micropollutants.
28
Fig.
8
Average
adsorption removal efficiencies of micropollutants onto PVP-coated NPs by multiple cycles of regeneration. Error bars indicate the minimum and maximum removal percentages.
29
Tables: Table. 1. Molecular description of the tested micropollutants. Compound
Tonalide
Bisphenol-A
Triclosan
Metolachlor
Antibacterial
Ketoprofen
Estriol
Anti-
One of the
Musk
Plastic
and antifungal
Leading
Inflammatory
estrogen
compound
component
agent
pesticide
drug
hormones
C18H26O
C15H16O2
C12H7Cl3O2
C15H22ClNO2
C16H14O3
C18H24O3
258.4
228.29
289.53
283.8
254.28
288.39
(mg/L)
1.8
192
10
530
51
119
Log Kow
5.9–6.3
2.2-3.4
4.76
3.13
2.99
2.45
Log Koc
3.80–4.80
2.06-3.59
3.38-4.20
1.34-3.36
2.12
3.08
Relevance Chemical formula Molecular weight (g/mol) Solubility
30
Table 2 Adsorption kinetic constant, k, adsorption capacity at equilibrium, q e, and correlation coefficients, R2, modeled by pseudo second-order equation (Eq. (3)) for both PVP-coated NPs and GAC (micropollutant concentration of 50 µg/L and PVP-coated NPs dose of 2 mg/L). PVP-NP Compound
GAC
qe
k
(µg/mg)
(µg/mg.min)
Tonalide
18.2
1.08
Bisphenol-A
23.5
Triclosan
R2
R2
qe
k
(µg/mg)
(µg/mg.min)
0.999
14.3
0.07
0.996
1.22
0.996
15.3
0.08
0.999
22.8
1.19
0.991
14.6
0.09
0.995
Metolachlor
19.2
1.18
0.992
15.2
0.07
0.995
Ketoprofen
23.7
1.21
0.996
14.1
0.08
0.996
Estriol
22.8
1.20
0.994
15.4
0.09
0.998
31
Table 3 Langmuir and Freundlich parameters for the adsorption of tested micropollutants onto PVP-coated NPs. Langmuir constants
Freundlich constants
Compound
qm (µg/mg)
RL
R2
1/n
K
R2
Tonalide
65.36
0.72
0.998
0.6
1.77
0.78
Bisphenol-A
90.91
0.74
0.994
0.63
1.9
0.86
Triclosan
52.63
0.56
0.991
0.56
2.18
0.85
Metolachlor
57.14
0.65
0.992
0.61
1.95
0.84
Ketoprofen
83.33
0.7
0.996
0.63
2.4
0.87
Estriol
56.17
0.62
0.996
0.51
2.81
0.88
32
Table 4 Langmuir fitting parameters for PVP-coated NPs at different temperatures. Initial micropollutant concentrations were 100 µg/L. Compound
T (K)
qm (µg/mg)
R2
Tonalide
298
76.47
0.9978
308
92.69
0.9684
318
105.61
0.9742
298
102.82
0.9935
308
129.65
0.9421
318
146.24
0.9536
298
63.43
0.9912
308
84.68
0.9856
318
98.24
0.9536
298
68.54
0.9920
308
79.69
0.9927
318
85.27
0.9874
298
95.36
0.9960
308
107.19
0.9766
318
119.45
0.9758
298
68.28
0.9956
308
82.39
0.9661
318
96.45
0.9724
Bisphenol-A
Triclosan
Metolachlor
Ketoprofen
Estriol
33
Table 5 Thermodynamic parameters for the adsorption of tested micropollutants onto PVP-coated NPs. ΔGo (kJ/mol) Compound
ΔHo (kJ/mol)
ΔSo (J/mol K)
298 K
308 K
318 K
Tonalide
158.22
48.56
-0.69
-2.17
-4.00
Bisphenol-A
296.73
85.98
-2.46
-5.35
-8.40
Triclosan
98.27
27.77
-1.53
-2.42
-3.50
Metolachlor
117.56
34.32
-0.69
-1.93
-3.04
Ketoprofen
234.87
67.51
-2.57
-4.64
-7.27
Estriol
147.57
35.58
-0.90
-2.29
-3.35
34
S0927-7757(17)30763-X http://dx.doi.org/doi:10.1016/j.colsurfa.2017.08.008 COLSUA 21870
To appear in:
Colloids and Surfaces A: Physicochem. Eng. Aspects
Received date: Revised date: Accepted date:
4-4-2017 23-7-2017 9-8-2017
Please cite this article as: Mohammad Alizadeh Fard, Ali Vosoogh, Brian Barkdoll, Behnoush Aminzadeh, Using polymer coated nanoparticles for adsorption of micropollutants from water, Colloids and Surfaces A: Physicochemical and Engineering Aspectshttp://dx.doi.org/10.1016/j.colsurfa.2017.08.008 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.
Using polymer coated nanoparticles for adsorption of micropollutants from water
Mohammad Alizadeh Farda,*, Ali Vosooghb, Brian Barkdolla, Behnoush Aminzadehc
a Department
of Civil and Environmental Engineering, Michigan Technological University,
1400 Townsend Dr., Houghton, MI 49931, USA b
Department of Hydraulics and Environment, School of Civil Engineering, Iran University
of Science and Technology, P.O. Box: 16765-163, Tehran, Iran. c Department
of Environmental Engineering, Graduate Faculty of Environment, University
of Tehran, P.O. Box: 1417853111, Tehran, Iran
*Corresponding author. Tel.: +1 906 231 3654, fax: +1 906 487 2943. Email address: [email protected] (M. Alizadeh Fard)
Graphical abstract
1
Highlights
A new hydrothermal method was used to synthesize PVP-coated NPs.
PVP-coated NPs removed >89% of the micropollutants tested.
Thermodynamic analyses revealed endothermic nature of the adsorption process.
Regeneration experiments showed the reliability of methanol regenerating method.
Abstract In this study, polyvinylpyrrolidone (PVP)-coated magnetite nanoparticles were synthesized to adsorb six emerging contaminants (Tonalide, Bisphenol-A, Triclosan, Metolachlor, Ketoprofen and Estriol) from aqueous solutions. The PVP-coated NPs were characterized by dynamic light scattering (DLS) measurements, thermal analysis, and Xray diffraction. Results indicated that PVP-coated NPs were successfully used as a separable adsorbent for removing the micropollutants from water. Adsorption results were modeled using Langmuir and Freundlich isotherms, which showed a better fit of data to the Langmuir model. The adsorbent showed good adsorption performance in which Bisphenol-A and Ketoprofen were the most effectively removed micropollutants, with 98 and 95% removal using only 0.1 mg of the adsorbent within 15 minutes of contact time, respectively. Kinetic studies were performed using the pseudo-second-order model to compare the performance of PVP-coated NPs with granular activated carbon (GAC) revealing the superiority of PVP-coated NPs over GAC. Thermodynamic parameters of the adsorption of the micropollutants onto the adsorbate showed the adsorption process was endothermic and spontaneous. Adsorbate regeneration studies were performed with methanol, ethanol, and the UV/H2O2 process in five regeneration cycles. Methanol treatment ensured the highest level of regeneration.
2
Keywords: Micropollutant removal; Separable adsorbent; adsorption; Adsorbent regeneration.
1. Introduction Nowadays, it is becoming difficult to find unpolluted water resources for drinking purposes. Water resource contamination can emanate from natural sources and/or anthropogenic activities [1, 2]. Industrial development and population growth have imposed thousands of synthetic organic compounds to aquatic fresh ecosystems[3]. Micropollutants, such as pesticides, pharmaceuticals, industrial chemicals, and hormones are present at very low concentrations, and can raise significant toxicological concerns for aquatic ecosystems, and thereby human health [4, 5]. Independent of source, a large degree of the micropolloutants have been found in water resources because they are not bio-degradable [6]. The persistent nature of these compounds means that if released into nature the contaminants will stay in the soil for a long time before reaching the groundwater. Even if they reach the wastewater treatment plant, they are likely to be released with wastewater effluent to surface water bodies [7, 8]. Recently researchers have focused on using adsorption techniques to remove these emerging contaminants from water bodies [9-11]. Activated carbon has been studied widely and showed promising removal efficiency as it has high porosity and internal surface area. However, regeneration of saturated activated carbon is energy demanding (requiring heating to 500–900 oC), and does not fully restore the removal performance [12]. In addition, the probability of releasing carbon dust during the treatment process causes some restrictions over its application in some cases [13, 14]. There are also limitations over application of other adsorbents like zeolite, and carbon nanotubes (CNTs) [15, 16]. Saturated absorbents have to be removed by separation methods such as filtration or centrifugation in order to achieve a free absorbent solution that facilitates regeneration of the absorbent. Applying these costly methods puts constraints on using these absorbents in industrial sites [17].
3
Given the restrictions of current adsorption methods for removing organic micropollutants, novel solutions are being actively sought [18-20]. Polymer-modified iron or iron oxide nanoparticles have been intensively studied for applications in soil and groundwater treatment for various kinds of contaminates including micropollutants [2127]. Phenrat et.al (2016) used a novel electromagnetically enhanced treatment concept for in situ remediation of a source zone of chlorinated dense non-aqueous phase liquid (DNAPL), causing groundwater contamination. Their results showed acceleration in the degradation of the DNAPLs by using polystyrene sulfonate (PSS)-modified zero valent iron NPs in combination with a low frequency AC electromagnetic field. In another study, Bauerlein et.al (2012) [10], studied the removal of some ionic organic compounds by both activated carbon and charged polymeric materials. Their results showed that sorption of all charged compounds to oppositely charged polymers was stronger than to activated carbon, especially for the double-charged cation metformin. Iron oxide nanoparticles (NPs) are attractive for their magnetism and low toxicity characteristics [28]. The former provides easily magnetic separation of NPs from aqueous solutions. Phenrat et.al 2008 [29] showed low toxicity of magnetite in comparison to nanoscale zerovalent iron. In addition, they showed that the addition of a polymer can reduce toxicity even more on mammalian cells (BV2 and N27) due to surface modification. The surface coating can probably change the interaction of the particles with the cell membranes. If the developed adsorbent is able to remove contaminants but cannot be regenerated, it thus will become a contaminant. Regenerating techniques like using organic solvents, inorganic solvents, and redox agents have been developed. In this study two organic solvents and one redox agent were studied to regenerate the saturated PVP-NPs. Composite magnetic materials have been produced by thermal decomposition and used for separation purposes. However, synthesis methods mainly need organic solvents and high temperatures, which make industrial applications difficult [30]. In this study, a simple and inexpensive hydrothermal method was used to produce polyvinylpyrrolidone (PVP)-coated magnetite nanoparticles to adsorb a variety of organic micropollutants from the aqueous phase. This method needs no organic solvents, 4
requires lower energy input, and ambient atmospheric environments. It has been previously shown that PVP-coated NPs are stable in environmental systems [30]. After production of PVP-coated NPs, characteristics of the absorbent, its removal efficiency, and thermodynamic parameters for Tonalide, Bisphenol-A, Triclosan, Metolachlor, Estriol and Ketoprofen were studied.
2. Materials and methods 2.1 Reagents and chemicals Ferrous chloride (FeCl2.4H2O, 99%), ferric chloride (FeCl3.6H2O, 99%), ammonium hydroxide (25%), sodium hydroxide (NaOH), sulfuric acid (98%), hydrogen peroxide (30%), ethanol, and methanol were obtained from Merck (Germany) and used as received. Polyvinylpyrrolidone (PVP) (>95%), Potassium persulfate (KPS), sodium bisulfite (>95%), Suwannee River Fulvic Acid (SRFA), granular activated carbon (GAC), and organic micropollutants (Tonalide, Bisphenol-A, Triclosan, Metolachlor, Ketoprofen and Estriol) were purchased from Sigma-Aldrich (USA). Ultrapure water (Millipore, USA) was used for nanoparticle preparation and kinetic studies. 2.2 Synthesis of nanoparticles Magnetic nanoparticles were synthesized using a hydrothermal method [31]. PVP, ferric chloride, and ferrous chloride were used as precursors, and ammonium hydroxide was used as a precipitation agent. First, 0.50 mmol of PVP was added to 10.0 mL of ultrapure water and was stirred at 90 °C for 2h. Then, 2 mmol of Ferric chloride and 6 mmol of ferrous chloride were added to the solution and stirred at 90 °C for 2h. Then, 10.0 mL of ammonium hydroxide was added into the solution dropwise under continuous stirring. After adding ammonium hydroxide, the color of the mixture turned black and was mixed at 90 °C for 2h. The precipitates were rinsed twice with ultrapure water, separated with a magnet and sonicated to disperse in water. 2.3 Analysis methods
5
Micropollutant
concentrations were
measured by gas chromatography–mass
spectrometry (GC/MS) equipped with a fused capillary column (TRB-5, 60 m, 0.32 mm ID) [5, 32, 33]. The apparatus was operated in the internal ionization mode. A 30 m × 0.25 mm, df= 0.25 μm capillary column SGE (Ringwood, Australia) was used under the following conditions: electron energy=60 eV; transfer-line temperature=200◦C; scan speed= 1000 Hz [4, 8]. Table 1 describes the chemical characteristics and relevance of the tested emerging micropollutants. These compounds were selected from different industry. Total organic carbon (TOC) was measured with a Shimadzu TOC analyzer (5000A, Shimadzu Corp., Japan). The structure of magnetic nanoparticles was analyzed using Cu Ka (k = 0.1542 nm) radiation from an X-ray Diffractometer (Scintag XDS-2000 θ/θ powder diffractometer). The XRD scattering was observed over a range of 10–80o 2θ with step sizes of 0.03 to 0.05° 2θ, and count times between 20 and 55 s. The average crystallite size was calculated according to the full width at the half-maximum height (FWHM) of the (311) plane using Scherrer (Eq.1) [34]: τ=
kγ
(1)
βCOSθ
where τ is the mean size of crystalline domain, k is a dimensionless shape factor (0.89 for magnetite), λ is the X-ray wavelength (1.54059 Å), β is the line broadening at half the maximum intensity (FWHM) after subtracting the instrumental broadening, and θ is Bragg’s angle. By assuming a spherical shape for nanoparticles and using the following equation (Eq.2), particle size (d) can be calculated[35]: 4
d=3 τ
(2)
DLS measurements were performed with a Malvern Zetasizer Nanosizer particle analyzer with a solid-state He–Ne laser at a scattering angle of 1731 at 25 oC (pH 6−8) [36]. For this analysis, all samples were diluted to approximately 0.1 mg/mL. Then, samples were filtered through a Whatman Anotop alumina membrane (pore size of 220 nm). The experiments were performed at NaCl concentrations of 0.001 and 0.16 mol/L at pH 7. All samples were analyzed directly after filtration every three minutes over a one
6
hour period. The average DLS measurements were determined from ten replicates. Zeta potential measurements were also done with a Malvern Zetasizer Nanosizer at 25 oC. Zeta potential measurements were performed by dispersing the sonicated samples (with 10 minutes of sonication) at a concentration of 0.1 mg/mL. The ionic strength was adjusted to 0.001 mol/L NaCl. The pH was initially adjusted to 2.0 and zeta potential was measured by increasing the pH to 12 using sodium hydroxide solution. The average zeta potential and DLS measurements were determined from ten replicates. Thermal analysis was conducted by using a thermogravimetric analyzer (TGA) (Netzsch TG209, Germany) at a heating rate of 10 ºC per minute under constant nitrogen flow [16]. 2.4 Experimental Procedure Adsorption experiments were performed in a 50-mL sealed glass vial. NPs were added to the solution containing 50.0 μg/L of each micropollutant at a pH value of 7.0. Sonication was used for five minutes before conducting the experiments to separate attached nanoparticles (NPs). The pH was adjusted with dilution of NaOH and HCl solutions. Vials were agitated at 150 rpm using a rotary shaker. Containers were wrapped with aluminum foil to minimize photolysis. After conducting the adsorption experiments, nanoparticles were separated with a 1.5 x 0.75 x 0.5 inch neodymium magnet (Grade N 52, K&J Magnetics Inc.) for 30 minutes. To analyze the samples, solutions were collected in 5.0 mL aliquots and sent immediately for analysis. To consider the effect of initial sonication and reduce the chance of analytical uncertainties, a stock sample was maintained with the same experimental conditions except the adsorption process [10]. Results were reported by comparing the final concentration of micropollutants in the stock sample to the experimental samples. All experiments were conducted at least three times and the mean values reported. The isotherms were obtained as follows. PVP-coated NPs (0.1 mg) and the different concentrations of micropollutant were added in a 50 mL sealed glass vial. The micropollutant concentration range was 25–250 µg/L. The vial was shaken at different temperatures (298, 308, and 318 K) for 20 minutes and the precipitation was separated magnetically.
7
Some studies have shown incomplete adsorption of micropollutants in natural samples. To investigate the effect of DOM on adsorption of micropollutants, the adsorption removal efficiencies for both PVP-coated NPs and GAC were compared with distilled water and SRFA-added water samples (DOM = 8 mg C/L). Regeneration experiments were done in a volume of 100 mL. For UV/H2O2 regenerating experiments a solution containing 20 mg/L of hydrogen peroxide was mixed with 2 mg of saturated NPs in a 500 mL four-neck cylindrical quartz reactor for 30 minutes. The reactor was placed on a rotary shaker with rotary speed being the same as the adsorption experiments (150rpm). UV irradiation was provided with one 30 W low-pressure UV mercury lamp placed in the center of the reactor. The residual H2O2 was neutralized with sodium bisulfite [37]. For methanol and ethanol regenerating experiments, all experimental conditions were the same as those with UV/H2O2, except UV irradiation and existence of hydrogen peroxide in which were replaced either by methanol (20 mg/l) or ethanol (20 mg/l). The regenerated adsorbents were reacted again with micropollutant solution to study the recycling rate. The regeneration tests were repeated 5 times.
3. Results and discussions 3.1 Characterizations of PVP-coated nanoparticles Fig. 1a shows the size distribution of synthesized nanoparticles using DLS.
The
hydrodynamic diameter measured by DLS was 106.2 ± 3.6 nm with an average polydispersity index (Pdi) of 0.25. According to the DLS results, NPs are well-dispersed and monodispersed enough for the DSL technique to be valid, since the Pdi Value was less than 0.7 indicating that the sample did not have a broad size distribution and was suitable for the dynamic light scattering technique [36]. The XRD pattern of the PVP-coated iron oxide NPs is shown in Fig. 1b which shows that the dominant phase of NPs is magnetite (Fe3O4), though maghemite was also detected. The calculated lattice parameter α was 8.3889 Å. Since the lattice of maghemite
8
and magnetite are 8.3515 Å (JCPDS 39-1346) and 8.3960 Å (JCPDS 19-629), respectively, the main crystalline phase could be identified as magnetite [19, 38]. Using Eqs. (1) and (2), the average crystal size of 14.3 nm and particle size of 19.7 nm were calculated. The TGA curve for PVP-coated iron oxide nanoparticles is depicted in Fig. 2. The initial weight loss before 240 ºC could be due to the desorption of adsorbed water and/or surface hydroxyls. It can be seen that the sample weight significantly decreased around 300 ºC. This can be ascribed to the formation of iron oxides and the decomposition of amorphous iron hydroxides. The major weight loss took place around 372 ºC which could be due to the decomposition of the coated PVP layer as such a loss did not occur in the uncoated sample [38]. Subsequently, the micropollutant removal capacity of PVP-coated iron oxide NPs was studied. The average zeta potential was 11.5 ± 0.4 mV indicating the sorption of PVP shifts the slip plane out of the surface of the particle and to the effect of adsorbed non-charged polymeric layer as suggested by Ohshima's soft particle analysis which decreases the apparent electrophoretic mobility [29]. 3.2 The effect of dosage and contact time Adsorbent dosage and contact time play significant roles in the adsorption removal of the micropollunts. By Increasing contact time and dosage more chance will be given to the absorbents to adsorb pollutants. On the other hand, either over-dosage of adsorbent or excessive contact time increases the operational costs. Therefore, it is indispensable to optimize the adsorbent dose and contact time. In order to evaluate the adsorptive capacity of the polymer coated nanoparticles, the removal of each micropollutant was studied with different dosages of NPs. Fig. 3(a) shows the effect of adsorbent concentration on the micropollulant removal. The experiments were performed at pH 7.5 and the detention time was 20 min. The adsorption efficiency was improved when adsorbent dose increased from 0.75 to 2.5 mg/L. Higher adsorbent dosage can lead to more adsorbable area. Almost complete removal of micropollutants was observed at the 2 mg/L dose. At the 2.5 mg/L dose, little 9
or no improvement was observed in adsorption efficiencies. On the other hand, the removal efficiencies for Bisphenol-A, Triclosan, Ketoprofen, and Estriol were almost the same for 1.5 and 2 mg/L concentrations. However, the average removal efficiencies for Metolachor and Tonalide at an adsorbent dose of 1.5 mg/L was considerably less than those for 2 mg/L dose. Therefore, 2 mg/L was selected as the optimum dose of NPs. Fig. 3(b) shows the removal of micropollutants due to contact time. The optimal contact time was found to be 20 minutes, at this point most of the micropollutants had reached their equilibrium concentration. At this condition little or no change was observed in their concentration. It should be noted that no sign of oxidation was observed during the experiments and all removal efficiencies can be ascribed to adsorption. Adsorption equilibrium was determined when each micropollutant had been contacted with the PVPcoated NPs for enough time, and adsorbate concentration in the aqueous phase was in a dynamic balance with the interface concentration. At this point almost all of the adsorbate concentration was removed. The concentrations of all micropollutants were sharply reduced in the first 10 minutes of contact time. Concentration reduction continued gradually after 10 minutes. Bisphenol-A and Ketoprofen were the most effectively removed micropollutants, with 98 and 95% removal, respectively, using only 0.1 mg of the adsorbent within 15 min. This could be due to their chemical structures, physiochemical properties, as well as the specific surface properties of the adsorbent. Micropollutants with high polar bonds tend to be adsorbed more easily. As Bisphenol-A, Triclosan, Ketoprofen, and Estriol have O-H bond(s) in their chemical structure that can easily bind to PVP [39]. The next significant factor that can affect adsorption capacity could be molecular weight because Bisphenol-A and Ketoprofen had less molecular weight than Triclosan and Estriol. The change of adsorption capacity with increasing reaction time can be attributed to the following stages [16, 28] conceptualized in Fig. 4. First, the instantaneous adsorption of micropollutants occurred within a short period of time. This step can be attributed to interactive forces between the micrpollutant molecules and the surface of the adsorbent. Second, as the contact time increased, a gradual achievement of the equilibrium was observed, since most of the active sites on the surface of the adsorbent were occupied 10
by the micropollutants. In addition, physisorption was the main adsorptive action for which physical bonding of the micropollutant molecules occurs on the surface of the adsorbent. This takes place due to Van der Waals forces. Finally, the adsorption equilibrium between micropollutant molecules and PVP-coated NPs was attained. To study the amount of each micropollutant adsorbed by PVP-coated NPs and GAC at equilibrium condition, the pseudo-second-order model was used to explain the adsorption kinetics, which is expressed in Eq. 3 [40]: 𝑡 𝑞𝑡
=
1 𝑘𝑞𝑒2
+
𝑡
(3)
𝑞𝑒
where k (µg/mg.min) is the second-order rate constant, qt (µg/mg) is the adsorption capacity at any time t (min), and qe (µg/mg) is the adsorption capacity at equilibrium. The results of adsorption kinetics are tabulated in Table 2. In addition, the kinetic removal curves of the tested micropollutants are presented in Fig. S1. According to the results, all correlation of coefficients were greater than 0.99. The corresponding linear regressions were obtained for the values of qe and k from Eq. (3). For experiments with PVP-coated NPs, the value of qe (18.2 µg/mg) for Tonalide was lower than the other micropollutants. This could be attributed to the difference in molecular structure of the target micropollutants. The k value for Bisphenol-A adsorbed by PVP-coated NPs was 15 times greater than that of the GAC experiments. This indicates a high adsorption capacity of PVP-coated NPs. This is probably due to the availability of more readily-accessible binding sites (located at the outer surface of the adsorbent) on the PVP-NPs compared to GAC. Alsbaiee et al. (2016) [9], also showed rapid adsorption of a variety of organic micropollutants with higher adsorption rate constants than those of activated carbon.
3.3 Adsorption isotherms Adsorption isotherms are valuable curves describing the phenomenon leading the retention or release of a substance from the aqueous to the solid phase at a constant temperature and pH. Langmuir and Freundlich adsorption isotherms were used for modeling adsorption data. The Langmuir isotherm assumes homogeneous adsorption, in 11
which all sites have equal affinity for the adsorbate molecule (each molecule has constant sorption activation energy). The Langmuir adsorption isotherm can be expressed by Eq. (4): 𝐶𝑒 𝑞𝑒
=𝑘
1
𝐶
𝑎 𝑞𝑚
+ 𝑞𝑒
(4)
𝑚
where Ce is the equilibrium concentration in the liquid phase (µg/L), qe is the amount of adsorbate adsorbed on the adsorbent at equilibrium conditions (µg/mg), qm is the maximum theoretical adsorption capacity of the adsorbent (µg/mg), and Ka is the adsorption equilibrium constant (L/µg). An equilibrium parameter or a dimensionless constant of separation factor (RL) can also be expressed as a critical characteristic of the Langmuir isotherm, which is defined as Eq. (5) [41]: 1
𝑅𝐿 = 1+𝐾
(5)
𝑎 𝐶0
where C0 is the initial concentration of micropollutants and RL indicates the shape of the isotherm; values between 0 and 1 indicate favorable adsorption, whereas RL = 0, RL = 1, and RL > 1, indicate irreversible, linear, and unfavorable, adsorption isotherms, respectively [42]. On the other hand, the Freundlich isotherm assumes that there are many types of binding sites which have different free energy of sorption values, and there are a large amount of sorbent sites. This isotherm expresses the diversity of free energies related to the adsorbate by the heterogeneous adsorption [43]. The Freundlich adsorption isotherm can be expressed by Eq. (5)[43]: 1
q𝑒 = K C𝑛
(5)
where K ((µg/mg) (L/µg) 1/n) is the adsorption coefficient of the Freundlich isotherm and 1/n is the Freundlich intensity parameter. If the value of 1/n is lower than 1, it indicates a normal Langmuir isotherm. The amount of micropollutant adsorbed per unit adsorbent
12
(qe) (µg/mg) was evaluated by dividing the amount adsorbed by the concentration of PVPcoated NPs. The linearized form of the Freudlich model is shown below [44, 45] in Eq. (6): 1
Log (q 𝑒 ) = Log(k) + 𝑛 Log(C)
(6)
The adsorption isotherms of Tonalide and Bisphenol-A for both Langmuir and Freundlich isotherms are depicted in Fig. 5 (a) and (b), respectively. Adsorption isotherm plots for other tested micropollutants can be found in Fig. S2 and Fig. S3. Results showed that the adsorption data for all micropollutants were well fitted to the Langmuir model (R2 >0.99). Therefore, a monolayer of the micropollutants was formed on the surface of PVPcoated NPs. Corresponding adsorption parameters for all tested micropollutants are listed in Table 3. According to the results, RL values for Langmuir isotherm were between 0 and 1 and the value of 1/n for Freundlich isotherm was smaller than 1, which indicate favorable adsorption. The maximum adsorption capacities (qm) by the Langmuir model were achieved for Bisphenol-A and Ketoprofen as 90.91 and 83.33 µg/mg, respectively. These two compounds have smaller molecular size in comparison to the other studied compounds. The smaller molecular size can help more molecules to be adsorbed on the surface of the adsorbent. The large adsorption capacity of PVP-coated NPs can be ascribed to the strong adsorption affinity of the adsorbent towards the target micropollutants, which emanated from unique surface characteristics. Table 4 lists the maximum theoretical adsorption capacity (qm) and the correlation coefficients (R2) at three different temperatures (298, 308 and 318 K). All the R2 values were above 0.94, indicating that Langmuir isotherm had acceptable agreement with the experimental data. Results indicated that when the temperature increased, qm values increased. This could be probably due to the interactive forces between adsorbate and adsorbent that control the process. Therefore, it can be concluded that the adsorption of the micropollutants onto PVP-coated NPs is an endothermic process.
13
Thermodynamic parameters such as the change in free energy (ΔG°), enthalpy (ΔH°), and entropy (ΔS°) were calculated for the adsorption of micropollutants using the following equations [12]: C
K c = Ce
(6)
ΔG° = −RT ln(K c )
(7)
S
ln (K c ) =
ΔS° R
−
ΔH°
(8)
RT
where Kc is equilibrium constant, Ce and Cs are the concentrations of micropollutants adsorbed by the adsorbent (µg/L) and remained in the aqueous phase at equilibrium (µg/L) respectively. R is the universal gas constant (8.314 J/mol K), and T is temperature (K). ΔH° and ΔS° can be calculated from the slope and intercept of ln Kc vs. 1/T (Van’t Hoff plot (Fig. 6)). Van’t Hoff plots for adsorption of other tested micropollutants are shown in Fig. S4 and S5. The thermodynamic parameters are summarized in Table 5. Positive values of ΔH° showed endothermic adsorption processes for all micropollutants. In addition, positive values of ΔS° indicated the affinity of the tested micropollutants and increasing randomness at the liquid-solid interface during the adsorption process. Negative values of ΔG° indicated spontaneous adsorption and the degree of spontaneity increased with increasing temperature.
3.4 The effect of dissolved organic matter (DOM) Micropollutants usually exist with DOM in water bodies [46]. Therefore, the possible competition between micropollutants and DOM should be considered when performing an adsorption experiment. Further experiments showed that 5.5 mg/L of PVP-coated NPs concentration was the optimal concentration when SRFA was added to the sample. This increase in the adsorbate concentration was indispensable due to the fact that the SRFA acts as a competitor for PVP-coated NPs meaning that a higher adsorbate concentration is required to remove the same amount of target micropollutants. Fig. 7 showed the average adsorption removals for micropollutants in the distilled water and surface water for both PVP-coated NPs and GAC. According to the results, GAC was more affected by the addition of DOM to the water sample. However, PVP-coated NPs 14
showed a great performance for the same conditions. It took 25 minutes for PVP-coated NPs to reach equilibrium while the equilibrium time for GAC was 55 minutes. The significant decrease in the removal of micropollutants might be caused by the competitive inhibition of DOM with micropollutants on GAC. This result implies that the inhibition of adsorption in the SRFA water samples is due to competition of hydrophobic adsorption interaction of SRFA with micropollutants on the surface of GAC to occupy adsorption sites.
3.5 Adsorbent regeneration Adsorbent regeneration is critical to ensure its sustainability [47]. In this study three regeneration methods were studied. Fig. 8 compares all regeneration methods to regenerate the removal capacity of the adsorbent for five consecutive cycles. This figure reports the average removal efficiency of micropollutants that are based on the micropollutants concentrations in the solution. Tables S1-S3 in the supplementary materials show the removal efficiency of all micropollutants for each regeneration method. According to the results the average removal efficiency for the methanol method was almost the same for all five cycles, while ethanol’s desorption efficiency decreased slightly. This showed that methanol could effectively detach the micropollutants from the surface of the adsorbent due to lower molecular weight, since ethanol (MW = 46) >methanol (MW = 32). This could be due to the fact that smaller molecules can penetrate more easily into the micropores of the adsorbent and displaces the adsorbate more effectively. Because alcohols have a hydroxyl group in their structures, organic micropollutants are more soluble in them. For example, Filippa et al. [48] found the solubility of Ketrophen in water, ethanol, and methanol to be 6.27×10-4, 1.85 and 1.75 mol/L, respectively. Therefore, methanol and ethanol are able to easily desorb the studied micropollutants. On the other hand, the regeneration performance of the UV/H2O2 process was tested as a system that does not have any waste at the end of the regeneration process [49]. This system is recognized as one of the more effective advanced oxidation processes 15
(AOPs) that is able to remove a wide range of contaminants [4, 50]. Organic pollutants adsorbed onto the adsorbent surface can be mineralized into harmless compounds such as carbon dioxide and water. It can be seen from Fig. 8 that the adsorption efficiency decreased when the UV/H2O2 treatment was applied. The main reason could be the nonselective nature of UV/H2O2 process to degrade oxidizable substances. In fact, this system can degrade adsorbed micropollutants and the surface of the coated polymer at the same time. Therefore, the regenerated adsorbent has probably less effective adsorption sites for the next cycles.
Conclusions In this study, PVP-coated NP was successfully produced using a new method and its characterizations were evaluated by different techniques. The adsorbent was successfully used for removing micropollutants from aqueous solutions. The adsorbent dosage, contact time, temperature, and dissolved organic matter competitor had a significant effect on the adsorption performance of PVP-coated NPs. The maximum adsorption capacities of Bisphenol-A and Ketoprofen on the adsorbent at a pH of 7.5 was 90.91 and 83.33 µg/mg, respectively. Results of isotherm modeling indicated that the Langmuir model was well fitted to the experimental data. In addition, the pseudo second order model was used to describe the adsorption kinetics. The thermodynamic parameters of all tested micropollutant adsorption onto the PVP-coated NPs showed that the adsorption process was endothermic and spontaneous. Methanol regeneration studies showed that PVP-coated NPs could be recycled and used for at least five cycles.
Acknowledgements The authors want to thank the department of Civil and Environmental Engineering at Michigan Technological University for financial support.
16
References: [1] M.T. Amin, A.A. Alazba, U. Manzoor, A Review of Removal of Pollutants from Water/Wastewater Using Different Types of Nanomaterials, Advances in Materials Science and Engineering, 2014 (2014) 24. [2] M. Carballa, F. Omil, J.M. Lema, Removal of cosmetic ingredients and pharmaceuticals in sewage primary treatment, Water Research, 39 (2005) 4790-4796. [3] L. Hernández-Leal, H. Temmink, G. Zeeman, C. Buisman, Removal of micropollutants from aerobically treated grey water via ozone and activated carbon, Water Research, 45 (2011) 2887-2896. [4] L. Guzzella, D. Feretti, S. Monarca, Advanced oxidation and adsorption technologies for organic micropollutant removal from lake water used as drinking-water supply, Water Res, 36 (2002) 4307-4318. [5] M. Hijosa-Valsero, V. Matamoros, R. Sidrach-Cardona, J. Martín-Villacorta, E. Becares, J.M. Bayona, Comprehensive assessment of the design configuration of constructed wetlands for the removal of pharmaceuticals and personal care products from urban wastewaters, Water research, 44 (2010) 3669-3678. [6] W. Qi, H. Singer, M. Berg, B. Müller, B. Pernet-Coudrier, H. Liu, J. Qu, Elimination of polar micropollutants and anthropogenic markers by wastewater treatment in Beijing, China, Chemosphere, 119 (2015) 1054-1061. [7] P.A. Quinlivan, L. Li, D.R. Knappe, Effects of activated carbon characteristics on the simultaneous adsorption of aqueous organic micropollutants and natural organic matter, Water research, 39 (2005) 1663-1673. [8] Z.M. Nagy, M. Molnar, I. Fekete-Kertesz, I. Molnar-Perl, E. Fenyvesi, K. Gruiz, Removal of emerging micropollutants from water using cyclodextrin, The Science of the total environment, 485-486 (2014) 711-719. [9] A. Alsbaiee, B.J. Smith, L. Xiao, Y. Ling, D.E. Helbling, W.R. Dichtel, Rapid removal of organic micropollutants from water by a porous β-cyclodextrin polymer, Nature, 529 (2016) 190-194. [10] P.S. Bäuerlein, T.L. ter Laak, R.C.H.M. Hofman-Caris, P. de Voogt, S.T.J. Droge, Removal of charged micropollutants from water by ion-exchange polymers – Effects of competing electrolytes, Water Research, 46 (2012) 5009-5018. 17
[11] X. Cai, Q. Liu, C. Xia, D. Shan, J. Du, J. Chen, Recyclable Capture and Destruction of Aqueous Micropollutants Using the Molecule-Specific Cavity of Cyclodextrin Polymer Coupled with KMnO4 Oxidation, Environmental Science & Technology, 49 (2015) 92649272. [12] A. Jafari Kang, M. Baghdadi, A. Pardakhti, Removal of cadmium and lead from aqueous solutions by magnetic acid-treated activated carbon nanocomposite, Desalination and Water Treatment, 57 (2016) 18782-18798. [13] I. Oller, S. Malato, J.A. Sánchez-Pérez, Combination of Advanced Oxidation Processes and biological treatments for wastewater decontamination—A review, Science of The Total Environment, 409 (2011) 4141-4166. [14] K. Ebie, F. Li, Y. Azuma, A. Yuasa, T. Hagishita, Pore distribution effect of activated carbon in adsorbing organic micropollutants from natural water, Water Research, 35 (2001) 167-179. [15] S.O. Ganiyu, E.D. van Hullebusch, M. Cretin, G. Esposito, M.A. Oturan, Coupling of membrane filtration and advanced oxidation processes for removal of pharmaceutical residues: A critical review, Separation and Purification Technology, 156, Part 3 (2015) 891-914. [16] H. Gao, S. Zhao, X. Cheng, X. Wang, L. Zheng, Removal of anionic azo dyes from aqueous solution using magnetic polymer multi-wall carbon nanotube nanocomposite as adsorbent, Chemical Engineering Journal, 223 (2013) 84-90. [17] S. Giannakis, F.A. Gamarra Vives, D. Grandjean, A. Magnet, L.F. De Alencastro, C. Pulgarin, Effect of advanced oxidation processes on the micropollutants and the effluent organic matter contained in municipal wastewater previously treated by three different secondary methods, Water Research, 84 (2015) 295-306. [18] J. Fresnais, M. Yan, J. Courtois, T. Bostelmann, A. Bée, J.-F. Berret, Poly (acrylic acid)-coated iron oxide nanoparticles: Quantitative evaluation of the coating properties and applications for the removal of a pollutant dye, Journal of colloid and interface science, 395 (2013) 24-30. [19] J.-F. Liu, Z.-s. Zhao, G.-b. Jiang, Coating Fe3O4 magnetic nanoparticles with humic acid for high efficient removal of heavy metals in water, Environmental science & technology, 42 (2008) 6949-6954. 18
[20] D. Morillo, A. Uheida, G. Pérez, M. Muhammed, M. Valiente, Arsenate removal with 3-mercaptopropanoic acid-coated superparamagnetic iron oxide nanoparticles, Journal of colloid and interface science, 438 (2015) 227-234. [21] T. Phenrat, N. Saleh, K. Sirk, H.-J. Kim, R.D. Tilton, G.V. Lowry, Stabilization of aqueous nanoscale zerovalent iron dispersions by anionic polyelectrolytes: adsorbed anionic polyelectrolyte layer properties and their effect on aggregation and sedimentation, Journal of Nanoparticle Research, 10 (2008) 795-814. [22] T. Phenrat, Y. Liu, R.D. Tilton, G.V. Lowry, Adsorbed polyelectrolyte coatings decrease Fe0 nanoparticle reactivity with TCE in water: conceptual model and mechanisms, Environmental Science & Technology, 43 (2009) 1507-1514. [23] T. Phenrat, F. Fagerlund, T. Illangasekare, G.V. Lowry, R.D. Tilton, Polymer-modified Fe0 nanoparticles target entrapped NAPL in two dimensional porous media: effect of particle concentration, NAPL saturation, and injection strategy, Environmental science & technology, 45 (2011) 6102-6109. [24] T. Phenrat, I. Kumloet, Electromagnetic induction of nanoscale zerovalent iron particles accelerates the degradation of chlorinated dense non-aqueous phase liquid: Proof of concept, Water Research, 107 (2016) 19-28. [25] S. Srirattana, K. Piaowan, G.V. Lowry, T. Phenrat, Electromagnetic induction of foambased nanoscale zerovalent iron (NZVI) particles to thermally enhance non-aqueous phase liquid (NAPL) volatilization in unsaturated porous media: Proof of concept, Chemosphere, (2017). [26] B. Jarošová, J. Filip, K. Hilscherová, J. Tuček, Z. Šimek, J.P. Giesy, R. Zbořil, L. Bláha, Can zero-valent iron nanoparticles remove waterborne estrogens?, Journal of environmental management, 150 (2015) 387-392. [27] I. Arslan-Alaton, T. Olmez-Hanci, The Use of Al and Fe Nanoparticles for the Treatment of Micropollutants, in:
Nanotechnologies for Environmental Remediation,
Springer, 2017, pp. 61-114. [28] D.-P. Li, Y.-R. Zhang, X.-X. Zhao, B.-X. Zhao, Magnetic nanoparticles coated by aminoguanidine for selective adsorption of acid dyes from aqueous solution, Chemical engineering journal, 232 (2013) 425-433.
19
[29] T. Phenrat, T.C. Long, G.V. Lowry, B. Veronesi, Partial oxidation (“aging”) and surface modification decrease the toxicity of nanosized zerovalent iron, Environmental science & technology, 43 (2008) 195-200. [30] S. Palchoudhury, J.R. Lead, A facile and cost-effective method for separation of oil– water mixtures using polymer-coated iron oxide nanoparticles, Environmental science & technology, 48 (2014) 14558-14563. [31] S. Mirshahghassemi, J.R. Lead, Oil Recovery from Water under Environmentally Relevant Conditions Using Magnetic Nanoparticles, Environmental Science & Technology, 49 (2015) 11729-11736. [32] C.P. James, E. Germain, S. Judd, Micropollutant removal by advanced oxidation of microfiltered secondary effluent for water reuse, Separation and Purification Technology, 127 (2014) 77-83. [33] V. Matamoros, J. García, J.M. Bayona, Organic micropollutant removal in a full-scale surface flow constructed wetland fed with secondary effluent, Water research, 42 (2008) 653-660. [34] M. Guillemot, J. Mijoin, S. Mignard, P. Magnoux, Adsorption of Tetrachloroethylene on Cationic X and Y Zeolites: Influence of Cation Nature and of Water Vapor, Industrial & Engineering Chemistry Research, 46 (2007) 4614-4620. [35] R.A. Bini, R.F.C. Marques, F.J. Santos, J.A. Chaker, M. Jafelicci Jr, Synthesis and functionalization of magnetite nanoparticles with different amino-functional alkoxysilanes, Journal of Magnetism and Magnetic Materials, 324 (2012) 534-539. [36] W.Y. Chen, DSL: Simulation Techniques and Standards Development for Digital Subscriber Lines, Alpel Publishing, 1997. [37] M.A. Fard, B. Aminzadeh, H. Vahidi, Degradation of petroleum aromatic hydrocarbons using TiO2 nanopowder film, Environmental technology, 34 (2013) 11831190. [38] Z. Shan, W.-S. Yang, X. Zhang, Q.-M. Huang, H. Ye, Preparation and characterization of carboxyl-group functionalized superparamagnetic nanoparticles and t he potential for bio-applications, Journal of the Brazilian Chemical Society, 18 (2007) 1329-1335.
20
[39] H. Chen, H. Luo, Y. Lan, T. Dong, B. Hu, Y. Wang, Removal of tetracycline from aqueous solutions using polyvinylpyrrolidone (PVP-K30) modified nanoscale zero valent iron, Journal of hazardous materials, 192 (2011) 44-53. [40] Y.S. Ho, G. McKay, Pseudo-second order model for sorption processes, Process Biochemistry, 34 (1999) 451-465. [41] S.-C. Tsai, K.-W. Juang, Comparison of linear and nonlinear forms of isotherm models for strontium sorption on a sodium bentonite, Journal of Radioanalytical and Nuclear Chemistry, 243 (2000) 741-746. [42] A. Dada, A. Olalekan, A. Olatunya, O. Dada, Langmuir, Freundlich, Temkin and Dubinin–Radushkevich isotherms studies of equilibrium sorption of Zn2+ unto phosphoric acid modified rice husk, IOSR Journal of Applied Chemistry, 3 (2012) 38-45. [43] K.Y. Foo, B.H. Hameed, Insights into the modeling of adsorption isotherm systems, Chemical Engineering Journal, 156 (2010) 2-10. [44] M. Baghdadi, E. Ghaffari, B. Aminzadeh, Removal of carbamazepine from municipal wastewater effluent using optimally synthesized magnetic activated carbon: Adsorption and sedimentation kinetic studies, Journal of Environmental Chemical Engineering, 4 (2016) 3309-3321. [45] B.A. E. Aghdam, M. Baghdadi, M.A. Fard, Removal of BTEX from aqueous solutions by paper mill sludge-based activated carbon, Journal of Advances in Chemistry, 11 (2015) 3416-3432. [46] R.P. Schwarzenbach, B.I. Escher, K. Fenner, T.B. Hofstetter, C.A. Johnson, U. Von Gunten, B. Wehrli, The challenge of micropollutants in aquatic systems, Science, 313 (2006) 1072-1077. [47] S. Lee, J.-S. Lee, M.-K. Song, J.-C. Ryu, B. An, C.-G. Lee, C. Park, S.-H. Lee, J.-W. Choi, Effective regeneration of an adsorbent for the removal of organic contaminants developed based on UV radiation and toxicity evaluation, Reactive and Functional Polymers, 95 (2015) 62-70. [48] M.A. Filippa, G.M. Melo, E.I. Gasull, Ketoprofen Solubility in Organic Solvents and Aqueous Co-solvent Systems: Interactions and Thermodynamic Parameters of Solvation.
21
[49] S. Notthakun, J.C. Crittenden, D.W. Hand, D.L. Perram, M.E. Mullins, Regeneration of adsorbents using heterogeneous advanced oxidation, Journal of Environmental Engineering, 119 (1993) 695-714. [50] M. Alizadeh Fard, A. Torabian, G.R.N. Bidhendi, B. Aminzadeh, Fenton and photoFenton oxidation of petroleum aromatic hydrocarbons using nanoscale zero-valent iron, Journal of Environmental Engineering, 139 (2013) 966-974.
22
Figure captions
Fig. distribution plot, (b) XRD pattern for the PVP-coated NPs.
23
1
(a)
DLS
size
Fig. 2 TGA plot for the PVP-coated NPs.
24
Fig. 3 Effect of (a) NP dose (micropollutant concentration of 50 µg/L) and (b) contact time (NP concentration of 2 mg/L) on micropollutants removal efficiencies. The data are reported as the average of at least triplicate experiments. Error bars indicate the minimum and maximum removal efficiencies.
25
Fig. Conceptual model of adsorption of the micropollutants.
26
4
Fig.
5
Langmuir (a) and Freundlich (b) isotherms for the adsorption of Tonalide and BisphenolA onto PVP-coated NPs.
27
Fig. 6 Van’t Hoff plot for adsorption of Tonalide and Bisphenol-A onto PVP-coated NPs (initial concentration of 100 µg/L, adsorbent dosage of 2 mg/L).
Fig. 7 Average adsorption removal efficiencies of PVP-coated NPs and GAC. GAC concentrations were the same as NPs for the same experiment. Error bars indicate the minimum and maximum removal efficiencies for all tested micropollutants.
28
Fig.
8
Average
adsorption removal efficiencies of micropollutants onto PVP-coated NPs by multiple cycles of regeneration. Error bars indicate the minimum and maximum removal percentages.
29
Tables: Table. 1. Molecular description of the tested micropollutants. Compound
Tonalide
Bisphenol-A
Triclosan
Metolachlor
Antibacterial
Ketoprofen
Estriol
Anti-
One of the
Musk
Plastic
and antifungal
Leading
Inflammatory
estrogen
compound
component
agent
pesticide
drug
hormones
C18H26O
C15H16O2
C12H7Cl3O2
C15H22ClNO2
C16H14O3
C18H24O3
258.4
228.29
289.53
283.8
254.28
288.39
(mg/L)
1.8
192
10
530
51
119
Log Kow
5.9–6.3
2.2-3.4
4.76
3.13
2.99
2.45
Log Koc
3.80–4.80
2.06-3.59
3.38-4.20
1.34-3.36
2.12
3.08
Relevance Chemical formula Molecular weight (g/mol) Solubility
30
Table 2 Adsorption kinetic constant, k, adsorption capacity at equilibrium, q e, and correlation coefficients, R2, modeled by pseudo second-order equation (Eq. (3)) for both PVP-coated NPs and GAC (micropollutant concentration of 50 µg/L and PVP-coated NPs dose of 2 mg/L). PVP-NP Compound
GAC
qe
k
(µg/mg)
(µg/mg.min)
Tonalide
18.2
1.08
Bisphenol-A
23.5
Triclosan
R2
R2
qe
k
(µg/mg)
(µg/mg.min)
0.999
14.3
0.07
0.996
1.22
0.996
15.3
0.08
0.999
22.8
1.19
0.991
14.6
0.09
0.995
Metolachlor
19.2
1.18
0.992
15.2
0.07
0.995
Ketoprofen
23.7
1.21
0.996
14.1
0.08
0.996
Estriol
22.8
1.20
0.994
15.4
0.09
0.998
31
Table 3 Langmuir and Freundlich parameters for the adsorption of tested micropollutants onto PVP-coated NPs. Langmuir constants
Freundlich constants
Compound
qm (µg/mg)
RL
R2
1/n
K
R2
Tonalide
65.36
0.72
0.998
0.6
1.77
0.78
Bisphenol-A
90.91
0.74
0.994
0.63
1.9
0.86
Triclosan
52.63
0.56
0.991
0.56
2.18
0.85
Metolachlor
57.14
0.65
0.992
0.61
1.95
0.84
Ketoprofen
83.33
0.7
0.996
0.63
2.4
0.87
Estriol
56.17
0.62
0.996
0.51
2.81
0.88
32
Table 4 Langmuir fitting parameters for PVP-coated NPs at different temperatures. Initial micropollutant concentrations were 100 µg/L. Compound
T (K)
qm (µg/mg)
R2
Tonalide
298
76.47
0.9978
308
92.69
0.9684
318
105.61
0.9742
298
102.82
0.9935
308
129.65
0.9421
318
146.24
0.9536
298
63.43
0.9912
308
84.68
0.9856
318
98.24
0.9536
298
68.54
0.9920
308
79.69
0.9927
318
85.27
0.9874
298
95.36
0.9960
308
107.19
0.9766
318
119.45
0.9758
298
68.28
0.9956
308
82.39
0.9661
318
96.45
0.9724
Bisphenol-A
Triclosan
Metolachlor
Ketoprofen
Estriol
33
Table 5 Thermodynamic parameters for the adsorption of tested micropollutants onto PVP-coated NPs. ΔGo (kJ/mol) Compound
ΔHo (kJ/mol)
ΔSo (J/mol K)
298 K
308 K
318 K
Tonalide
158.22
48.56
-0.69
-2.17
-4.00
Bisphenol-A
296.73
85.98
-2.46
-5.35
-8.40
Triclosan
98.27
27.77
-1.53
-2.42
-3.50
Metolachlor
117.56
34.32
-0.69
-1.93
-3.04
Ketoprofen
234.87
67.51
-2.57
-4.64
-7.27
Estriol
147.57
35.58
-0.90
-2.29
-3.35
34