Journal of Colloid and Interface Science 424 (2014) 124–131

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science www.elsevier.com/locate/jcis

Extraction of methylmercury and ethylmercury from aqueous solution using surface sulfhydryl-functionalized magnetic mesoporous silica nanoparticles Guangzhu Li a,b, Miao Liu a,⇑, Zhuqing Zhang b, Chao Geng c, Zhongbo Wu b, Xin Zhao b a b c

Key Laboratory of Groundwater Resource and Environment, Ministry of Education, Jilin University, Changchun 130012, PR China Environmental Monitoring Center of Jilin Province, Changchun 130011, PR China Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, PR China

a r t i c l e

i n f o

Article history: Received 31 December 2013 Accepted 10 March 2014 Available online 19 March 2014 Keywords: Fe3O4@SiO2 magnetic nanoparticles Sulfhydryl-functionalized Extraction Methylmercury Ethylmercury

a b s t r a c t Surface sulfhydryl-functionalized magnetic mesoporous silica nanoparticles were prepared, aiming to extract trace alkylmercury from aqueous solution. The prepared nanoparticles were characterized by TEM, ED, EDX, DLS, FTIR, and SERS. Compare with that the non-sulfhydryl-functionalized Fe3O4@SiO2 exhibited almost no affinity for CH3Hg+ and CH3CH2Hg+; the sulfhydryl-functionalized Fe3O4@SiO2 exhibited high adsorption affinity for them, resulting from chelating interaction by surface sulfhydryl group, and the adsorption was not significantly impacted by pH within the range of 3.5–9.0 or coexisting metal ions. The monolayer adsorption on surface of Fe3O4@SiO2–RSH could reach equilibrium in 2 min. Moreover, the CH3Hg+ and CH3CH2Hg+ adsorbed on Fe3O4@SiO2–RSH could be quickly separated from the matrix in a magnetic field and desorbed easily by acetonitrile and L-cysteine aqueous solution or HCl solution, and the recoveries were more than 80%. Findings of the present work highlight the potential for using Fe3O4@SiO2–RSH magnetic nanoparticles as effective and reusable adsorbents for extraction of ultra trace alkylmercury from environmental water samples. Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction More than 2500 tons of mercury is emitted annually from global anthropogenic sources [1]. Mercury is a ubiquitous contaminant that is widely dispersed throughout the environment mostly via atmospheric deposition [2]. For different speciation of mercury, the toxicity is significantly difference. Generally, organic mercury is more toxic than inorganic mercury. The methylmercury (MeHg), CH3Hg+, because of the lipophilic property and biological amplification, is hundreds of times more toxic than inorganic Mercury. However, all of the mercury released in the ecosystem undergoes biogeochemical transformation processes and can be converted into MeHg by microorganisms and microalgae in aquatic environments [3]. Moreover, there have been evidence suggests that ethylation could also play an important role in the biogeochemical cycling of mercury [4]. Therefore, only monitoring the total mercury concentrations in the environment is not enough, and ⇑ Corresponding author. E-mail addresses: [email protected] (G. Li), [email protected] (M. Liu), [email protected] (Z. Zhang), [email protected] (C. Geng), [email protected] (Z. Wu), [email protected] (X. Zhao). http://dx.doi.org/10.1016/j.jcis.2014.03.026 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.

speciation analysis provides more useful information to assess the toxicity and health risks of mercury and further understand biogeochemical cycling of mercury compounds [5]. But the organic Mercury concentrations of the environmental water samples are very difficult to measure precisely because they are so low and suffer from matrix effects when samples are analyzed directly. Therefore, a sample pretreatment step, which can separate the analytes from the matrix components and preconcentrate them, is exceedingly important. Sulfhydryl-cotton fiber Solid-phase extraction (SCF-SPE) and distillation are main approaches for the extraction of trace alkylmercury from water. SCE-SPE technology was developed over 30 years ago and remains the most popular SPE method for determining alkylmercury. For offering a number of advantages such as high enrichment factor and low consumption of organic solvents [6], the method has been adopted by the national standard methods of China (GB/T 14204-1993 and GB/T 17132-1997). But, in practice, operation of the method is complex and time consumed. On the other hand, distillation, as a more effective method in terms of CH3Hg+ recovery [7], has been adopted by the U.S. Environmental Protection Agency as a standard method (EPA 6030). However, the sample is distilled relatively slowly such that the total

G. Li et al. / Journal of Colloid and Interface Science 424 (2014) 124–131

distillation time is 5–6 h. More importantly, regents are added to the sample prior to distillation, which increases the possibility for contamination or increased blank levels [2]. In recent years, modified Silica-coated core–shell magnetic nanoparticles (Fe3O4@SiO2) have shown great potential for using as adsorbents in various fields, such as removal of heavy metal ions from aqueous solution [8–13], selective adsorption of Hg2+ [14], adsorption of phosphate [15], separation of phosphopeptides [16], chiral separation of recemic compounds [17], and recently magnetic solid phase extraction (MSPE) [18–23]. Moreover, the previous studies demonstrated that c-Mercaptopropyltrimethoxysilane (c-MPTMS) modified Fe3O4@SiO2 nanoparticles can rapidly and quantitatively adsorb Cd, Cu, Pb, and Hg from aqueous solution, attributing to metal complexation by the sulfhydryl group [18]. According to the theory of Hard–Soft-Acid–Base (HSAB), sulfhydryl group (soft alkali) also has high adsorption affinity for CH3Hg+ (soft acid). However, to the best of our knowledge, no attempt has been made to extract trace CH3Hg+ and ethylmercury (EtHg), CH3CH2Hg+, from aqueous solution using surface sulfhydrylfunctionalized Fe3O4@SiO2 magnetic nanoparticles. In this study, Fe3O4@SiO2–RSH nanoparticles were prepared as adsorbents for extraction of trace MeHg and EtHg from aqueous solution. And accompanied with high performance liquid chromatography online coupled with atomic fluorescence spectrometry (HPLC-AFS) technique, the extraction properties of Fe3O4@SiO2–RSH toward CH3Hg+ and CH3CH2Hg+ in aqueous solution were investigated for adsorption capacity, effect of pH and coexisting metal ions, desorption and regeneration.

125

Fe3O4@SiO2 subsequently. 100 mg of Fe3O4 nanoparticles and 2 ml of H2O were homogeneously dispersed in 98 mL of ethanol with ultrasonication. Then 0.2 mL of 3-MPTMS added into the flask with stirring and ultrasonication for 10 min. Following, 0.5 mL of NH3H2O was added into the above mixture to start the reaction with stirring at 25 °C for 12 h under a nitrogen atmosphere. The obtained black precipitates were separated by a magnet and thoroughly washed with deionized water and methanol. The materials obtained are referred to as Fe3O4@SiO2–RSH nanoparticles. The transmission electron microscope (TEM), electron diffraction (ED), and energy dispersive X-ray (EDX) analysis of the nanoparticles were used a field emission TEM (TECNAI-F20, FEI, Netherlands) equipped with EDX (EDAX, USA) at an acceleration voltage of 200 kV. The dynamic light scattering (DLS) analysis of the nanoparticles was carried out using an eighteen angle laser light scatterometer (DAWN HELEOS II, WYATT, USA). Briefly, 0.5 mg of nanoparticles was dispersed in 5 mL of H2O and ultrasonication for 10 min. Then the dispersion was used for DLS measurement. The functional groups of the nanoparticles were examined through a Fourier-transform infrared spectroscopy (FTIR, VERTEX-70, BRUKER, Germany) use KBr pressed disk, and a Renishaw 2000 model confocal microscopy Raman spectrometer equipped with a holographic notch filter (Renishaw Ltd., UK). Surface-enhanced Raman scattering (SERS) spectra were collected using a microscope objective (N.A. = 0.4) with 50% amplitude and laser intensity 15 mW at the sample. The acquisition time was typically 10 s. 2.3. Determination methods of MeHg and EtHg

2. Experiment section 2.1. Reagents Methyl mercury (GBW08675), ethyl mercury (GBW(E)081524), Hg2+ (GBW08617) were obtained from National Institute of Metrology, China. K, Na, Ca(II), Mg(II), Cu(II), Pb(II), Zn(II) and Cd(II) (1 g/L) were obtained from Accustandard, USA. Tetraethoxysilane (TEOS, 95.0%) was obtained from Sigma–Aldrich, USA. 3-MPTMS (95.0%) was obtained from aladdin, China. Ferric chloride hexahydrate (FeCl36H2O, 99.0%), Ferrous chloride tetrahydrate (FeCl24H2O, 98.0%), ammonium hydroxide (NH3H2O, 25%), ethanol (99.7%), methanol (99.9%), acetic ammonium (98.0%), potassium hydroxide (KOH, 85.0%), sodium hydroxide (NaOH, 98.0%), potassium persulfate (K2S2O8, 99.0%), and potassium borohydride (KBH4, 95.0%) were obtained from Sinopharm Chemical Reagent Co., Ltd., China. Ultra pure grade hydrochloric acid (HCl, 36–38%) was obtained from Xi-Long Chemical Co., Ltd., Shantou, China. Acetonitrile (HPLC Grade, 99.9%) was obtained from Fisher, USA. L-cysteine (98.5%) was obtained from Hui-Shi Co., Ltd., Shanghai, China. Highly pure deionized water (18.25 MX cm) was obtained from Milli-Q Element System (Millipore France). 2.2. Adsorbents preparation and characterization The Fe3O4 nanoparticles were prepared using a simple chemical coprecipitation method according to previously reported method [12]. Then, the StÖber method [24] to prepare silica-coated Fe3O4 nanoparticles was used. Briefly, 100 mg of Fe3O4 nanoparticles, 20 mL of H2O, and 0.1 mL of TEOS were homogeneously dispersed in 80 mL of ethanol with ultrasonication. Following, 1.5 mL of NH3H2O was added into the above mixture to start the reaction with stirring at 25 °C for 6 h under a nitrogen atmosphere. The materials obtained are referred to as Fe3O4@SiO2 nanoparticles. Fe3O4@SiO2–RSH nanoparticles was prepared by the hydrolysis of 3-MPTMS and condensed with the surface hydroxyl groups of

Concentration of CH3Hg+ and CH3CH2Hg+ was determined by an AFS-930 model atomic fluorescence spectrometer equipped with a SA-20 model speciation analysis pretreatment apparatus (Beijing Ji Tian Apparatus Co., Ltd., China). The process of on-line pretreatment include: Firstly, different speciation of mercury in aqueous solution was separated by Venusil MP-C18 model HPLC column (150.0 mm  4.6 mm  5 lm, agela, China). The mobile phase was 10 mmol/L L-cysteine 60 mmol/L acetic ammonium and 5% (v/v) acetonitrile aqueous solution. Secondly, the outflow from HPLC column, mixed with oxidants (10 g/L K2S2O8 and 3.5 g/L KOH aqueous solution) and air, irradiated by ultraviolet lamp subsequently. Then the organic mercury was oxidized to inorganic mercury. Finally, the above mixture mixed with reducing agent (10 g/L KBH4 and 3.5 g/L KOH aqueous solution) and 7% (v/v) HCl solution. Through hydride reaction, the inorganic mercury was reduced to Hg atom. The Hg atoms, coupled with H2, were purged by the carrier gas (Ar 99.99%), introduced directly into the AFS detector. Instruments operating conditions are given in Table 1. Before adsorption experiments, the methods were evaluated by analyzing 2 lg/L of standard solutions. The relative standard deviations (RSD, n = 7) of CH3Hg+ and CH3CH2Hg+ were 3.2% and 4.8%, and the detection limits (3r) were 0.2 lg/L and 0.3 lg/L, respectively. 2.4. Adsorption experiments In a typical process, the adsorbents were dispersed into the high purity water to form a standard solution (2 g/L). Then 150 lL (0.3 mg) of the aqueous dispersion was added into 10 mL solution. The pH value of the mixture was adjusted by 0.1 M HCl and 0.1 M NaOH solutions. The adsorption experiments were carried out in a temperature incubator at 20 ± 1 °C and 100 rpm. After sorption reached the equilibrium, the adsorbents were separated via an external magnetic field. The equilibrium adsorbed concentration, qe (lg/mg), was calculated according to the following equation:

126

G. Li et al. / Journal of Colloid and Interface Science 424 (2014) 124–131

Table 1 High performance liquid chromatography on-line coupled with atomic fluorescence spectrometry (HPLC-AFS) operating conditions.

a

Pretreatment apparatus Mobile phase flow rate (mL/min) Sample amount (lL) Peristaltic pumpa speed (RMP/min) UV lamp power (W)

1.0 300 60 15

AFS Lamp current (mA) Photomultiplier negative high voltage (V) Carrier gas flow rate (mL/min)

30 280 300

Introduce oxidants, air, reductants, and HCl into pretreatment apparatus.

qe ¼

ðC 0  C e ÞV M

ð1Þ

where C0 (lg/L) is the initial concentration of CH3Hg+ or CH3CH2Hg+ in solution, Ce (lg/L) is the equilibrium concentrations in solution, V (L) is the volume of the solution, and M (mg) is adsorbent mass. Adsorption kinetics of CH3Hg+ or CH3CH2Hg+ on Fe3O4@SiO2–RSH nanoparticles was obtained by placing Fe3O4@SiO2–RSH in a series of aqueous solutions with initial CH3Hg+ or CH3CH2Hg+ concentrations 400 lg/L and pH 7.0. In different adsorption times, the Fe3O4@SiO2–RSH was separated and the concentration of solution was measured. For comparison, the non-sulfhydryl-functionalized Fe3O4@SiO2 as adsorbents for adsorbing CH3Hg+ or CH3CH2Hg+ were performed. Adsorption isotherms were obtained by placing Fe3O4@SiO2–RSH in a series of CH3Hg+ or CH3CH2Hg+ aqueous solutions with initial concentrations ranging from 100 to 1000 lg/L and pH 7.0. A separate set of experiments were conducted to test the PH effect on adsorption of CH3Hg+ onto Fe3O4@SiO2–RSH with pH ranging from 2.0 to 9.0 and initial CH3Hg+ concentrations in solutions 400 lg/L. The effect of coexisting metal ions on CH3Hg+ adsorption was studied at pH 7.0 and initial CH3Hg+ concentrations 400 lg/L. The metal ions which often present together with alkylmercury in environmental water were divided into three groups and added into the solutions, respectively: alkali metal ions (0.5–6 mg/L K+, Na+, Ca2+ and Mg2+, respectively), heavy metal ions (0.5–6 mg/L Cu2+, Pb2+, Zn2+ and Cd2+, respectively), and Hg2+ (50–500 lg/L, added 0.3 mg and 0.6 mg of Fe3O4@SiO2–RSH into the solutions, respectively). 2.5. Desorption and regeneration studies 0.3 mg of Fe3O4@SiO2–RSH was added into a series of 10 mL solutions containing 1 lg/L CH3Hg+ and CH3CH2Hg+, respectively. The adsorption process carried out in a temperature incubator at 20 ± 1 °C and 100 rpm for 2 min. Then the adsorbents were separated by an external magnet, and the supernatant water was decanted directly. Then 1 mL of HCl solutions (5, 10 and 50 mmol/L), or 10 mmol/L L-cysteine and 5% (v/v) acetonitrile aqueous solution were added as eluents. The desorption process carried out in a temperature incubator at 20 ± 1 °C and 100 rpm for 2 min. Finally, the magnet was used again to settle the nanoparticles, and the eluent was transferred into a test tube for subsequent HPLC-AFS analysis. For using low concentration (1 lg/L) of CH3Hg+ and CH3CH2Hg+ in adsorption solution, the eluent can be analyzed by HPLC-AFS without dilution. To evaluate the reusability of the nanoadsorbent, adsorption of alkylmercury and regeneration of alkylmercury-loaded Fe3O4@SiO2-RSH by L-cysteine and acetonitrile aqueous solution were performed in three consecutive cycles.

3. Results and discussion 3.1. Characterization of adsorbents From the TEM images of the three nanoparticles (Fig. 1a–c), the average particles sizes of Fe3O4, Fe3O4@SiO2 and Fe3O4@SiO2–RSH were observed to be 12.5, 19.5, and 23.1 nm, respectively. The TEM images, Fig. 1b and c, illustrated the core–shell structure of Fe3O4@SiO2 and Fe3O4@SiO2–RSH. Compared with Fe3O4@SiO2, the Silica-shell of Fe3O4@SiO2–RSH was more uniform and similar to sphere, because sol–gel reactions were performed two times during the preparation process. High-resolution TEM image, Fig. 1d, shows that the structure of Fe3O4@SiO2–RSH was a 12.4 nm Fe3O4 core coated of 5.5 nm Silica-shell. The DLS results illustrated that the average particles sizes of Fe3O4, Fe3O4@SiO2 and Fe3O4@SiO2–RSH were 20.0, 22.4, and 23.8 nm, respectively. The RSD were 38%, 29%, and 25%, respectively. Although the saturation magnetization decreased after silica coating of the surface of the Fe3O4 core [25], complete magnetic separation of Fe3O4@SiO2–RSH samples was achieved in 1 min by placing a permanent magnet near the vessels containing 40 mL of aqueous dispersion of the 2 mg nanoparticles (see Fig. 2). Due to the superparamagnetism of the nanoparticles, the adsorbent could be separated rapidly from the sample solution using an external magnetic field instead of filtration or centrifugation. Analysis of the electron diffraction (ED) pattern of Fe3O4 (Fig. 3a and Table 2) confirmed the crystalline structure and interplanar spacing. The different crystal planes were consistent with the standard d-values of magnetite (Fe3O4) listed in the X-ray powder JCPDS diffraction data file [26]. The same diffraction planes (2 2 0), (3 3 1), (4 0 0), (5 1 1), (4 4 0) and (7 3 1), were also observed for Fe3O4@SiO2–RSH (Fig. 3b), indicating the stability of the crystalline phase of Fe3O4 during the preparation process. The EDX spectra of Fe3O4@SiO2–RSH are depicted in Fig. 3c. For being supported by carbon-coated copper grid during the analyses process, C and Cu cannot be identified. The strong peaks of iron, oxygen and silicon were observed, illustrated that Fe, O and Si are the major constituents in Fe3O4@SiO2–RSH. Moreover, the week peak of sulfur was identified, illustrated that low-level S is constituent in Fe3O4@SiO2–RSH. The FTIR spectra of Fe3O4, Fe3O4@SiO2 and Fe3O4@SiO2–RSH are compared in Fig. 4A. For all three nanoparticles, peaks at approximately 1630, 570 cm1 were observed, corresponding to the vibration band of Fe–O from the magnetite phase [27]. The peak at 3392 cm1 observed on Fe3O4@SiO2 is associated with the stretching vibration band of O–H [28], and the peaks at 1205, 1074, 951, 803, and 440 cm1 are associated with the vibration of Si–OH and Si–O–Si groups, reflecting the coating of silica on the magnetite surface. Compared with Fe3O4@SiO2, the O–H and Si–OH bond peaks of Fe3O4@SiO2–RSH at 3328 and 951 cm1 become weaker, and the week peak at 2923 cm1 observed, indicating the low-level CH2 groups are present on Fe3O4@SiO2. But the S–H bond peak cannot be observed from the FTIR spectra, because of low-level S contained in Fe3O4@SiO2–RSH and week infrared adsorption peaks of sulfhydryl. Successful sulfhydryl functionalization of the Fe3O4@SiO2 was evidenced by SERS spectra of Fe3O4@SiO2–RSH (Fig. 4B). The strong SERS bands at 2160 cm1 corresponds to the –SH group. Moreover, the bands located at 496, 665, and 702 cm1 are attributed to the bending of bridging oxygen and the vibration of C–S [29,30], and the bands located at 1376 and 1433 cm1 are attributed to the vibration of CH2. Therefore, from the above-mentioned FTIR and SERS spectra results we can make a conclusion that Si–OH and R–SH are the major groups on surface of the as-obtained Fe3O4@SiO2–RSH.

G. Li et al. / Journal of Colloid and Interface Science 424 (2014) 124–131

127

Fig. 1. Transmission electron microscopy (TEM) images of nanoparticles: (a) Fe3O4, (b) Fe3O4@SiO2, (c) and (d) Fe3O4@SiO2–RSH.

observation also suggested that CH3Hg+ and CH3CH2Hg+ could not be adsorbed by Si-OH group. On the contrary, after modification, the Fe3O4@SiO2–RSH exhibited high adsorption capacities for CH3Hg+ and CH3CH2Hg+, the adsorption equilibrium could be reached in 2 min, equilibrium concentrations were 18.5 and 13.3 lg/L, and adsorption percentage were more than 95% (Fig. 5). To further investigate the adsorption process. The pseudo-firstorder model, pseudo-second-order adsorption model were utilized. The two models can be expressed as:

log



qe qe  qt



t 1 t ¼ þ qt k2 q2e qe

Fig. 2. The separation of Fe3O4@SiO2–RSH nanoparticles from suspension under an external magnetic field and the same bottle of sample without the magnet.

3.2. The adsorption kinetics It was found that after 180 min of contact time, the non-sulfhydryl-functionalized Fe3O4@SiO2 exhibited almost no affinity for CH3Hg+ and CH3CH2Hg+; compared with before adsorption, the concentration of CH3Hg+ and CH3CH2Hg+ in aqueous solution were no significant differences (t-test, n = 7, confidence range 95%), the

¼

k1 t 2:303

ð2Þ

ð3Þ

where k1 (min1) is the equilibrium rate constant of pseudo-firstorder, k2 (mg/lg min) is the equilibrium rate constant of pseudosecond-order, qe and qt (lg/mg) are the adsorbed concentration at time t (min) and equilibrium adsorbed concentration, respectively. The fitting parameters are summarized in Table 3. The values of the correlation coefficient (R2) for the pseudo-second-order model (both 1.0000) and the pseudo-first-order model (0.7240, 0.7094) indicated that the adsorption kinetics date fit well with the pseudo-second-order model. Moreover, the values of qe calculated from the pseudo-second-order curves were in good agreement with the experimental values (12.74 and 12.93 lg/mg). The results suggested that the adsorption processes appear to be controlled by the chemical processes. The good linearity of the pseudo-second-order model suggested the homogeneity of adsorbent surface. In addition, as mention

128

G. Li et al. / Journal of Colloid and Interface Science 424 (2014) 124–131

Fig. 3. Electron diffraction (ED) patterns of (a) Fe3O4, (b) Fe3O4@SiO2–RSH, and (c) energy dispersive X-ray (EDX) spectra of Fe3O4@SiO2–RSH. Table 2 A comparison of experimental and standard interplanar spacing (d) values with their respective diffracting plane index (h k l) in Fe3O4 nanoparticles. Experimental d (nm)

Standard d/(nm)

hkl

Fe3O4

Fe3O4@ SiO2–RSH

Fe3O4 (cubic)

Diffraction plane

4.84 2.96 2.53 2.10 1.72 1.63 1.49 1.33 1.10

– 2.96 2.53 2.10 – 1.63 1.49 – 1.09

4.86 2.97 2.53 2.10 1.71 1.62 1.48 1.33 1.09

111 220 311 400 422 511 440 620 731

above, CH3Hg+ and CH3CH2Hg+ could not be adsorbed by Si-OH group. Then, the conclusion could be made that adsorption of CH3Hg+ and CH3CH2Hg+ on Fe3O4@SiO2–RSH occurred at surface sulfhydryl group. The fact that adsorption equilibrium could be quickly reached, also suggested that sulfhydryl group on the surface of nanoparticles could easier accessing to the alkylmercury in solution.

Ce Ce 1 ¼ þ qe qm bqm

ð4Þ

where qe (lg/mg) and Ce (lg/mg) are the adsorbed concentration and the aqueous concentration at adsorption equilibrium, qm (lg/mg) is the adsorption capacity, and b (L/lg) is the affinity coefficient. The fitting parameters are summarized in Table 4. The isotherm date of CH3Hg+ and CH3CH2Hg+ fitted well to the Langmuir model with the values of the R2 0.9999 and 0.9998, respectively, and the values of qm calculated from the Langmuir curves were in good agreement with the experimental values (14.4 and 15.0 lg/mg), suggesting that the adsorption was monolayer which occurred at sites with indifferent energy of adsorption. Hence the isotherm data also illustrated that CH3Hg+ and CH3CH2Hg+ adsorption onto Fe3O4@SiO2–RSH through chelating adsorption of the surface sulfhydryl groups. According to previous papers [31,32], one sulfhydryl group will react with one CH3Hg+, as shown below.

RSH þ CH3 Hgþ ! RSHgCH3 þ Hþ

ð5Þ +

3.3. Adsorption isotherms Adsorption isotherms of CH3Hg+ and CH3CH2Hg+ on Fe3O4@SiO2–RSH at 20 ± 1 °C are shown in Fig. 6. The adsorption data were fitted to the Langmuir model according to the equation:

And according to the experimental value of qm for CH3Hg , the CH3Hg+-bound sulfhydryl group on the surface of Fe3O4@SiO2–RSH was calculated to be 0.067 lmol/mg. If each the sulfhydryl group reacts with one CH3CH2Hg+, the adsorbed concentration was calculated to be 15.3 lg/mg. It was in good agreement with the experimental value of qm for CH3CH2Hg+. According to this and

129

G. Li et al. / Journal of Colloid and Interface Science 424 (2014) 124–131

Fig. 5. Adsorption kinetics of 400 lg/L CH3Hg+ (MeHg) and CH3CH2Hg+ (EtHg) onto 30 mg/L Fe3O4@SiO2–RSH, at shaking rate 100 rpm, PH = 7.0, temperature 20 ± 1 °C (mean ± s.d., n = 5).

Table 3 Parameters of the kinetics models of 400 lg/L CH3Hg+ (MeHg) and CH3CH2Hg+ (EtHg) onto 30 mg/L Fe3O4@SiO2–RSH, at shaking rate 100 rpm, PH = 7.0, temperature 20 ± 1 °C. Pseudo-first order

MeHg EtHg

Pseudo-second order

k1 (min1)

R2

k2 (mg/lg min)

qe (lg/mg)

R2

0.0288 0.0257

0.7240 0.7094

11.9100 6.2923

12.7372 12.9282

1.0000 1.0000

Fig. 6. Adsorption isotherms of 100–1000 lg/L CH3Hg+ (MeHg) and CH3CH2Hg+ (EtHg) adsorption onto 30 mg/L Fe3O4@SiO2–RSH, at contact time 2 min, shaking rate 100 rpm, PH = 7.0, temperature 20 ± 1 °C. Fig. 4. (A) Fourier transform infrared (FTIR) spectra of (a) Fe3O4, (b) Fe3O4@SiO2, (c) Fe3O4@SiO2–RSH. (B) Surface-enhanced Raman scattering (SERS) spectrum of Fe3O4@SiO2–RSH.

Table 4 Langmuir model parameters, qm, b, and R2, for 100–1000 lg/L CH3Hg+ (MeHg) and CH3CH2Hg+ (EtHg) adsorption onto 30 mg/L Fe3O4@SiO2–RSH, at contact time 2 min, shaking rate 100 rpm, PH = 7.0, temperature 20 ± 1 °C.

the similar affinity coefficient (values of b) for CH3Hg+ and CH3CH2 Hg+, we conjectured that the mechanism of CH3CH2Hg+ adsorption onto sulfhydryl group is same to CH3Hg+.

Langmuir model parameters

MeHg EtHg

qm (lg/mg)

b (L/lg)

R2

14.4051 14.9813

0.3226 0.3431

0.9999 0.9998

3.4. Effect of pH The effect of pH on the adsorption of CH3Hg+ onto Fe3O4@SiO2– RSH is illustrated in Fig. 7. Within the pH range of 3.5–9.0, Fe3O4@ SiO2–RSH exhibited high adsorption capacity for CH3Hg+ and was not significantly impacted by the change of pH. The maximum uptake of CH3Hg+ occurs at pH 8.0 and decreased slightly at pH 9.0

and 3.5, whereas the adsorption capacity decreased dramatically with decreasing pH from 3.5 to 2.0. This observation is consistent with the sulfhydryl-MeHg chelating and ion-exchange adsorption mechanism, Eq. (5). Slightly alkaline pH conditions are favorable

130

G. Li et al. / Journal of Colloid and Interface Science 424 (2014) 124–131

to the adsorption reaction of sulfhydryl-MeHg. At low pH, the sulfhydryl groups are protonized, passivating adsorption sites and hence suppressing CH3Hg+ adsorption. Noteworthily, even pH decreased to as low as 3.5, adsorption capacity of Fe3O4@SiO2–RSH did not decreased significantly, also suggesting the high adsorption affinity of Fe3O4@SiO2–RSH for CH3Hg+. 3.5. Effect of coexisting metal ions The impact of coexisting alkali metal ions (K+, Na+, Ca2+, Mg2+) and heavy metal ions (Cu2+, Pb2+, Zn2+, Cd2+) on the uptake of CH3Hg+ is presented in Fig. 8A. With the concentration of alkali metal ions increased from 0 to 2 mg/L, or the concentration of heavy metal ions increased from 0 to 0.5 mg/L, the adsorbed concentration shows no significant differences (t-test, n = 7, confidence range 95%) and then slightly decreased with the concentration of alkali metal ions increased from 2 to 6 mg/L, or heavy metal ions increased from 0.5 to 6 mg/L, which may be attributed to the competitive binding of the cations for the surface sulfhydryl groups of the nanoparticles. Compared with the alkali metal ions, K+, Na+, Ca2+, and Mg2+, heavy metal ions, Cu2+, Pb2+, Zn2+, and Cd2+, exerted a more significant suppressive effect on CH3Hg+ adsorption, likely due to the stronger complexing ability of sulfhydryl groups with them. Hence the adsorption ability of sulfhydryl groups for the above tested ions is ordered as follows: CH3Hg+ > Cu2+, Pb2+, Zn2+, Cd2+ > K+, Na+, Ca2+, Mg2+, which is in accordance with the HSAB theory. The impact of coexisting Hg2+ on the uptake of CH3Hg+ is presented in Fig. 8B. When CH3Hg+ present together with Hg2+, the surface sulfhydryl groups of the nanoparticles prefer to adsorb Hg2+(see Fig. 8B(a)), suggesting the complexing ability of sulfhydryl groups with divalent Hg2+ are stronger than monovalent CH3Hg+. But the adsorbed concentration of CH3Hg+ was not obvious affected by the presence of Hg2+, on the condition that the dosage of adsorbent was sufficient (see Fig. 8B(b)). Generally, the concentrations of all mercury species in environmental water are far lower than the experimental concentration. Hence, Fe3O4@SiO2–RSH could extract trace CH3Hg+, CH3CH2Hg+, and Hg2+ from environmental water at same time. 3.6. Desorption and regeneration studies Desorption is a crucial factor in application of the nanoparticles for extraction and enrichment of CH3Hg+ and CH3CH2Hg+ from aqueous solution. The results of effect of pH on CH3Hg+ adsorption and the mobile phase of HPLC used in this study imply that both

Fig. 8. Effect of (A) alkali metal ions (K+, Na+, Ca2+, Mg2+) and heavy metal ions (Cu2+, Pb2+, Zn2+, Cd2+) on 400 lg/L CH3Hg+ (MeHg) adsorption onto 30 mg/L Fe3O4@SiO2–RSH (mean ± s.d., n = 5), (B) Hg2+ on 400 lg/L CH3Hg+ (MeHg) adsorption onto (a) 30 mg/L and (b) 60 mg/L Fe3O4@SiO2–RSH, at contact time 2 min, shaking rate 100 rpm, temperature 20 ± 1 °C.

acid and L-cysteine acetonitrile aqueous solution could be chosen as eluents. The recoveries of CH3Hg+ and CH3CH2Hg+ for using different eluents are presented in Table 5. For using HCl solution, concentration of 10 mmol was sufficient, and the similar recoveries could be reached by using L-cysteine and acetonitrile aqueous solution. Compared with the low pH HCl solution, the L-cysteine and acetonitrile aqueous solution is more suitable to be separated by HPLC directly. Therefore, the solution was used in regeneration study. The recoveries of CH3Hg+ and CH3CH2Hg+ maintained relatively stable in three adsorption–regeneration recycles (see Fig. 9), indicating the Fe3O4@SiO2–RSH nanoparticles can be reused.

Table 5 The recoveries of CH3Hg+ (MeHg) and CH3CH2Hg+ (EtHg) for using different aqueous solution as eluents, adsorption: add 0.3 mg adsorbent into 10 mL solutions containing 1 lg/L MeHg and EtHg, contact time 2 min, shaking rate 100 rpm, temperature 20 ± 1 °C; desorption: add 1 mL eluents, contact time 2 min, shaking rate 100 rpm, temperature 20 ± 1 °C. Eluents

+

Fig. 7. Effect of pH on 400 lg/L CH3Hg (MeHg) adsorption onto 30 mg/L Fe3O4@SiO2–RSH, at contact time 2 min, shaking rate 100 rpm, temperature 20 ± 1 °C (mean ± s.d., n = 5).

5 mmol/L HCl 10 mmol/L HCl 50 mmol/L HCl 10 mmol/L L-cysteine and 5% (v/v) acetonitrile

Recovery (%) MeHg

EtHg

82.6 85.2 85.6 83.5

76.8 81.5 81.6 85.9

G. Li et al. / Journal of Colloid and Interface Science 424 (2014) 124–131

131

Compared with established methods, the MSPE method based on the Fe3O4@SiO2–RSH is rapid, simple, and suitable for batch operations. These unique attributes give the Fe3O4@SiO2–RSH nanoparticles a potential for effective extraction of ultra trace alkylmercury from environmental water samples. Acknowledgments This work was financially supported by the Science and Technology Project of Environmental Protection Agency of Jilin Province, China (Grants 2011-13) and also supported in part by Environmental Monitoring Center of Jilin province, China. References

Fig. 9. The recoveries of CH3Hg+ (MeHg) and CH3CH2Hg+ (EtHg) for using virgin and regenerated Fe3O4@SiO2–RSH adsorbent in three adsorptions and regenerations recycles. adsorption: add 0.3 mg adsorbent into 10 mL solutions containing 1 lg/L MeHg and EtHg, contact time 2 min, shaking rate 100 rpm, temperature 20 ± 1 °C; desorption: add 1 mL eluents, contact time 2 min, shaking rate 100 rpm, temperature 20 ± 1 °C.

It should be noted that alkylmercury could be extracted from aqueous solution in less than 8 min for using the proposed method. Analysis time is shortened greatly compared to the traditional methods. Additionally, compared to extract alkylmercury from the matrix by distillation, no reagents were needed for species conversion by the developed method, benefiting to reduced contamination and lowered blank; and compared to the traditional columnpassing SPE such as SCF-SPE, target analytes adsorbed on the magnetic nanoparticles can be quickly separated from the matrix in a magnetic field, which facilities the SPE processes.

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]

4. Summary In this work, surface sulfhydryl-functionalized Fe3O4@SiO2 magnetic nanoparticles were prepared, aiming to extract trace alkylmercury from aqueous solution. The synthesized and characterized Fe3O4@SiO2–RSH exhibited high adsorption affinity for CH3Hg+ and CH3CH2Hg+, resulting from chelating interaction by surface sulfhydryl group, and the adsorption was not significantly impacted by pH within the range of 3.5–9.0 or coexisting metal ions. The monolayer adsorption on surface of Fe3O4@SiO2–RSH could reach equilibrium in 2 min. Moreover, the CH3Hg+ and CH3CH2Hg+ adsorbed on Fe3O4@SiO2–RSH could be quickly separated from the matrix in a magnetic field and desorbed easily by acetonitrile and L-cysteine aqueous solution or HCl solution.

[18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32]

N. Lubick, Nature 459 (2009) 620. B. Jackson, V. Taylor, R.A. Baker, E. Miller, Environ. Sci. Technol. 43 (2009) 2463. I. Lehnherr, V.L. St Louis, H. Hintelmann, J.L. Kirk, Nat. Geosci. 4 (2011) 298. Y.X. Mao, Y.G. Yin, Y.B. Li, G.L. Liu, X.B. Feng, G.B. Jiang, Y. Cai, Environ. Pollut. 158 (2010) 3378. Y. Gao, Z.M. Shi, Z. Long, P. Wu, Ch.B. Zheng, X.D. Hou, Microchem. J. 103 (2012) 1. Y. Cai, R. Jaffé, A. Alli, R.D. Jones, Anal. Chim. Acta 334 (1996) 251. M. Horvat, L. Liang, N.S. Bloom, Anal. Chim. Acta 282 (1993) 153. J.H. Wang, S.R. Zheng, Y. Shao, J.L. Liu, Z.Y. Xu, D.Q. Zhu, J. Colloid Interface Sci. 349 (2010) 293. F. Zhang, J. Lan, Z.S. Zhao, Y. Yang, R.Q. Tan, W.J. Song, J. Colloid Interface Sci. 387 (2012) 205. H.B. Hu, Z.H. Wang, L. Pan, J. Alloys Compd. 492 (2010) 656. F.L. Fan, Z. Qin, J. Bai, W.D. Rong, F.Y. Fan, W. Tian, X.L. Wu, Y. Wang, L. Zhao, J. Environ. Radioact. 106 (2012) 40. Q.X. Zheng, H. Liu, X.Q. Xu, M.X. Du, Chem. J. Chin. Univ. 33 (2012) 107. Q.Y. Liu, Z. Xing, Y.L. Bei, J. Colloid Interface Sci. 394 (2013) 646. Y.J. Jiang, X.T. Li, J. Gao, X.C. Guo, J. Guan, X.D. Mu, J. Nanopart. Res. 13 (2011) 939. W.J. Wang, J. Zhou, D. Wei, H.Q. Wan, S.R. Zheng, Z.Y. Xu, D.Q. Zhu, J. Colloid Interface Sci. 407 (2013) 442. Z.C. Wang, G. Cheng, Y.L. Liu, J.L. Zhang, D.H. Sun, J.Z. Ni, J. Colloid Interface Sci. 417 (2014) 217. J.W. Wu, P. Su, J. Huang, S.M. Wang, Y. Yang, J. Colloid Interface Sci. 399 (2013) 107. C.Z. Huang, B. Hu, Spectrochim. Acta, Part B 63 (2008) 437. J.S. Suleiman, B. Hu, H.Y. Peng, C.Z. Huang, Talanta 77 (2009) 1579. N. Zhang, H.Y. Peng, B. Hu, Talanta 94 (2012) 278. G.H. Cheng, M. He, H.Y. Peng, B. Hu, Talanta 88 (2012) 507. H. Bagheri, A. Afkhami, M. Saber-Tehrani, H. Khoshsafar, Talanta 97 (2012) 87. M.D. Farahani, F. Shemirani, J. Colloid Interface Sci. 407 (2013) 442. W. StÖber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62. L. Ling, S.G.C. Eugene, X. Junmin, Chem. Phys. Lett. 461 (2008) 114. S. Sun, H. Zeng, D.B. Robinson, S. Raoux, P.M. Rice, S.X. Wang, G. Li, J. Am. Chem. Soc. 126 (2004) 273. R.Y. Hong, T.T. Pan, H.Z. Li, Magn. Magn. Mater. 303 (2006) 60. Y.T. Zhou, H.L. Nie, C.B. White, Z.Y. He, L.M. Zhu, J. Colloid Interface Sci. 330 (2009) 29. G. Hernándz, R. Rodríguez, J. Non-Cryst. Solids 246 (1999) 209. L.R. Allain, T. Vo-Dinh, Anal. Chim. Acta 469 (2002) 149. V. Celo, R.V. Ananth, S.L. Scott, D.R.S. Lean, Anal. Chim. Acta 516 (2004) 171. D.H. Wang, S.H. Liu, Z.L. Zhao, F.J. Feng, Environ. Sci. (Chin.) 3 (1982) 42.