Facile synthesis and characterization of thiol-functionalized graphene oxide as effective adsorbent for Hg(II)

Accepted Manuscript Title: Facile synthesis and characterization of thiol-functionalized graphene oxide as effective adsorbent for Hg(II) Author: A. Santhana Krishna Kumar Shiuh-Jen Jiang Wei-Lung Tseng PII: DOI: Reference:

S2213-3437(16)30113-0 http://dx.doi.org/doi:10.1016/j.jece.2016.03.034 JECE 1036

To appear in: Received date: Revised date: Accepted date:

30-12-2015 29-2-2016 21-3-2016

Please cite this article as: A.Santhana Krishna Kumar, Shiuh-Jen Jiang, Wei-Lung Tseng, Facile synthesis and characterization of thiol-functionalized graphene oxide as effective adsorbent for Hg(II), Journal of Environmental Chemical Engineering http://dx.doi.org/10.1016/j.jece.2016.03.034 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.

Facile synthesis and characterization of thiol-functionalized graphene oxide as effective adsorbent for Hg(II)

A.

Santhana

Krishna

Kumara,

Shiuh-Jen

Jianga,b*[email protected],Wei-Lung

Tsenga,c*[email protected]   a

Department of Chemistry, National Sun Yat-sen University, Kaohsiung 80424, Taiwan.

b

Department of Medical Laboratory Science and Biotechnology, Kaohsiung Medical University,

Kaohsiung 80708, Taiwan. c

School of Pharmacy, College of Pharmacy, Kaohsiung Medical University, Kaohsiung 80708,

Taiwan

*Corresponding author. Tel: +886-7-5252000 ext. 3929. +886-7-5252000 ext. 3925  

Highlights ™The method was successfully applied in sea, river and pond water ™ The method was successfully applied to fish tissues sample ™The MBT-GO adsorbent could be reused for 3 adsorption-desorption cycles ™Physico-chemical characterization of the adsorbent was studied in detail ™The maximum adsorption capacity was found to be 107.52 mg g-1

Abstract

1   

In this work, we report the synthesized and characterization of mercaptobenzothiazole (MBT) functionalized graphene oxide (GO) as a novel adsorbent material for the adsorption of Hg(II). This adsorption process could be realized with GO and MBT acting as host-guest in the first instance followed by the subsequent complexation of Hg(II) with MBT, the soft−soft interaction between Hg(II) & sulphur enhances the effective complexation. The prepared graphene-based adsorbent was characterised by x-ray photoelectron spectroscopy (XPS), Raman spectroscopy, Powder-X-ray diffraction (Powder-XRD), Transmission electron microscopes (TEM), Fourier transform infrared spectroscopy (FT-IR), UV-Vis spectroscopy, scanning electron microscopes (SEM) and energy-dispersive x-ray analysis (EDX). The capability of ICP-MS for Hg(II) adsorption was extensively studied under different optimal parameters, the developed method was successfully applied to mercury adsorption from water samples. Keywords: Mercury; Graphene oxide; Mercaptobenzothiazole; Sea water; River water  

1. Introduction Mercury is a highly toxic metal that can accumulate in ecological systems and adversely impact the environment. The main chemical species of mercury are elemental mercury (Hg0), ionic mercury (Hg+, Hg2+), methylmercury (MeHg+), and ethylmercury (CH3CH2Hg+). Methylmercury is a neurotoxin and is commonly found in aquatic environments due to its affinity with fatty tissue in animals; methylmercury tends to bioaccumulate and biomagnify more readily than other species of mercury. In particular, divalent mercury (Hg2+) is a widespread environment pollutant, found mainly in surface water [1]. Hg(II) causes permanent harmful effects of living organisms, even at minimal doses [2]. Mercury could be found in significant amounts in wastewater such as chloralkali manufacturing plants, electrical and electronics manufacturing, sulphide ore roasting operations; manufacture of batteries, lamps, paints, paper industries are common and, therefore, 2   

requires to effective treatment for mercury remediation. Exposure to mercury can have toxic effects on reproduction, central nervous system, liver and kidneys cause sensory and psychological impairments. Mercury is one of the most hazardous heavy metals due to the reinforcement of its intrinsic toxicity by bioaccumulation through metabolic processes of the food chain [3]. The elemental mercury vapour released into the atmosphere is subsequently converted to Hg(II) and transferred to water and sediments. Mercury is well-known for toxicity, even at very low concentrations; the drinking water criterion for mercury [4], established by the US EPA, is 2.0 mg L-1, permitted discharge limit of total mercury in waste water is 10.0 mg L-1. Removal of various toxic heavy metals, including mercury compounds, from the environment, is a big challenge. Among the various available technologies, adsorption is widely used, simple, and easy operation and it could effectively remove various heavy metals from the environment [5, 6]. However, a very similar work which has been recently published by M. Hami Dindar et al [6].

Adsorption techniques are simple and work very effectively for any adsorption process; an adsorbent with large surface areas, pore volume and proper functionalities is the key to success. The adsorption capacities of materials depend on their porous structure and surface properties; nevertheless, given its large specific surface area and strong hydrophilic nature, graphene oxide (GO) shows great potential in the removal of toxic heavy metals in aqueous solutions. Currently, many different porous materials have been developed, such as thiols based on GO adsorbent [3], a glycidyl methacrylate grafted cellulose adsorbent [4], a thiol modified Fe3O4@SiO2 as a robust [7], a cerium functionalized PVA– chitosan composite nanofibers[8], 1-octadecanethiol monolayers modified on graphene [9], mercaptobenzothiazole-modified

cellulose[10], 3 

 

thermally

reduced

GO

decorated

with

functionalized gold nanoparticles[11], and trioctylamine-modified sodium montmorillonite clay[12], polyaniline/reduced GO nanocomposite [13],

showing varying effectiveness in

removing toxic mercury from the environment. GO and graphene are new carbonaceous adsorbent materials and, in the past few years, GO has attracted tremendous interest. Graphene is two-dimensional, with a single layer of sp2-hybridized carbon atoms arranged in six-membered rings [14]. Graphene has strong mechanical, thermal, and electrical properties and a large theoretical specific surface area, GO has many functional groups, indicating their potential for adsorption processes. GO to contain a range of surface oxygen-functionalities, such as carboxyl, carbonyl, hydroxyl and phenol groups these functional groups markedly increase the hydrophilicity of GO, making it easily dispersible in aqueous solutions and stable under common environmental conditions [14]. Carbonaceous-based adsorbents, including activated carbon, carbon nanotubes and GO, usually show high adsorption capacity [15-18]. In the past few years, many investigations focused on the applications of GO in the removal of toxic mercury from aqueous solutions [19-23]. The ability of a material to adsorb toxic heavy metals is controlled, in part, by the number of available functional groups used for binding metals; functional groups with a known affinity for specific metals can be attached to other substances to create an effective adsorbent.

The functionalization of thiol groups (–SHs) on suitable solid supports offers favourable performances by taking advantage of the coordination of sulphur (S) groups with heavy metal species. The –SH forms stable complexes with soft heavy metals of high polarizability, such as Hg, Ag, Au, and, to a lesser extent, Cd and Zn, failing to coordinate well with the more abundant smaller, lighter metals, such as Ca, Na, and Mg [2, 23, 24]. Based on the preference of a cation for complexation with ligands, mercury is classified as a B-type metal cation, characterized by a 4   

‘‘soft sphere’’ of highly polarizable electrons in its outer shell. Soft metals like mercury show a pronounced preference for S ligands. The presence of thiol groups improves adsorption capacities for removal of Hg(II) [2,23,25] since the reaction against the Hg(II) ions and the –SHs are highly thermodynamically favourable [26] and metal adsorption is directly related to the amount of –SHs available in the adsorbent [25]. The adsorption properties of these adsorbents depend on the functional groups of their surfaces, adsorbents containing S-based ligands (such as thiol, thiourea, and thioether groups) effectively form complexes with mercury ions [27-29]; thiol-functionalized adsorbents exhibited a specific binding capability toward highly toxic heavy metal ions, including Hg(II), due to the –SHs [30]. The functionalized MBT onto GO for Hg(II) adsorption was not previously explored, despite how MBT is a well-known complexing agent able to bind soft metal ions in aqueous solution. The MBT bears a –SH that could interact with GO hydroxyl, carboxyl groups through covalent bond interaction, non-covalent pi–pi stacking, hydrophobic interaction, hydrogen bonding, van der Waals forces. Here, we report that thiolfunctionalized GO adsorbent shows a strong affinity for Hg(II) adsorption.

2. Materials and methods 2.1. Materials The starting material, graphite used in the preparation of GO was procured from Sigma Aldrich. A stock solution of 1000 mg L-1 Hg2+ solution was prepared using A. R. grade mercuric chloride, CH3Hg+ solutions was prepared to use CH3HgCl, and CH3CH2Hg+ solution was prepared to use CH3CH2HgCl, a working solution of 80 mgL-1 of Hg2+, 30 mgL-1 of CH3Hg+ and 30 mgL-1 of CH3CH2Hg+ for batch adsorption was prepared by appropriate dilution with Milli-Q water, 2Mercaptobenzothiazole (MBT) was purchased from Merck chemicals. The other required reagents were procured from Merck chemicals respectively. 5   

2.2. Instruments and characterizations Raman spectra were recorded on a WITec Confocal Raman Microscope Alpha 300R using a 532 nm He-Ne laser with 5 mW. The TEM analysis was carried out using a PHILIPS CM-200 TWIN instrument an operating voltage at 200 kV, samples were prepared by drop-wise addition of an appropriate solution to the carbon-coated 400 mesh Cu grid followed by solvent evaporation in hot air oven. A Perkin Elmer FT-IR spectrum100 spectrometer was used to characterize the material functional groups identifications with the range of 400−4000 cm−1 by mixing 0.01g of the material with 0.1g KBr (spectroscopy grade). The XPS (Kratos Axis Ultra) instruments were used to record samples, X-Ray Sources: 500 mm Rowland Circle Monochromator Al-Mg/Al achromatic Source 450W max power. The spectra were recorded using a monochromatic Al Kα X-ray source (15 mA, 14 KV). All obtained spectra were calibrated to a C 1s peak at 284.6 eV, and fitted with a mixed Gaussian-Lorentzian function by XPSPEAK (a freeware). The pressure in the analyzer chamber was 1 x 10-8 Torr. High-resolution spectra were collected using 40 eV pass energy, the spot size of 300 x 700 µm slots and 0.05 eV step size. UV–Visible spectra were taken by Jasco V-630 (Jasco, Japan), the X-ray diffraction (Powder-XRD) was carried out on a D8 Discover X-ray diffractometer with Cu Ka radiation (k = 0.1541 nm, Bruker, Germany) was utilized to record the characteristic changes in the diffraction pattern of the adsorbent material. A JEOL JSM-6330TF analyzer was used to observe the morphological changes and the energy dispersive X-ray analysis spectra (EDX) were recorded adsorbent and after adsorption of Hg(II). The concentration of mercury in the aqueous solution was measured using ICP-MS PerkinElmer, Sciex- Elan DRC Plus, Software which we used Elan-6100 DRC PLUS respectively. The pH adjustments were done using METTLER TOLEDO pH meter S20. 2.3. Preparation of exfoliated graphene oxide from graphite

6   

An improved method was used for the synthesis of GO [14,15,16,31], which was reported in our earlier study [23]. About 1.5 g of graphite powder was taken and gradually added to a 9:1 mixture of concentrated H2SO4/H3PO4 (180:20 mL). Subsequently, KMnO4 (9.0 g) was gradually added to the mixture with stirring, a maintaining the reaction temperature below 40 oC. The reaction was slightly exothermic to 35-40 oC, after which it was heated up to 60 oC with continuous stirring for 12h. The above reaction mixture was cooled to room temperature and slowly poured onto ice (200 mL) with 2 mL of H2O2 (30%) where the brown color was entirely turned to yellow. Afterwards, centrifugation of the filtrate was done at 4500 rpms for 30 min, the supernatant was decanted. The solid material obtained, centrifugation was thoroughly washed with 200 mL of water followed by 200 mL of 30% HCl and 200 mL of ethanol. After each wash, the filtrate was centrifuged at 4500 rpms for 30 min and the supernatant was decanted off. The leftover solid material was dried at room temperature for 12 h. A 1.0 g of GO was dispersed into mixture of 1:1 ratio, an SOCl2/DMF (25:25mL, v/v) were slowly added, the resulting mixture was refluxed at 70 oC for 24 h, the final product were washed with tetrahydrofuran (THF) and dried under the vacuum [23].

2.4. Preparation of adsorbent A 1.0g of GO was taken in a round-bottom flask followed by drop-wise addition of 20 ml of 0.05 mol MBT was added gradually, the reaction mixture was stirred at room temperature for 10 h, followed by reaction mixture was washed with ethanol an un-reacted MBT were removed by repeated washing with ethanol and dried under the vacuum. The comprehensive characterizations of the adsorbent were done using various physico-chemical techniques to confirm the presence of MBT and GO in the adsorbent [10]. 2.5. Batch adsorption study 7   

Batch adsorption isotherm studies [23] were carried out with varying concentrations of Hg(II) solution, these studies were carried out by equilibrating (20 mL of 80 mg L

−1

of Hg(II), 20 mL

of 30 mg L−1 CH3Hg+, and 20 mL of 30 mg L−1 CH3CH2Hg+ solution) adjusted to pH 5.4 to 6.9 with 0.10 g of the MBT-GO adsorbent material. The equilibration was carried out at 120 rpms for different time intervals (5, 10, 15, 20, and 30 min) using a magnetic orbital stirrer, the amount of Hg(II) adsorbed (qe) was obtained by equilibration at various concentrations using the expression qe:

(1) qe

(C0Ce )V W

Where Co and Ce indicate the initial and equilibrium Hg(II), CH3Hg+ and CH3CH2Hg+ concentrations respectively, V &W specify the volume of sample solution (L) and the weight (g) of the adsorbent. The supernatant was collected for further ICP-MS analysis to determine the Hg(II) concentration after adsorption (Table T1 of supporting information ESI†) [23].  

3. Results and discussion

3.1. X-ray photoelectron spectroscopy analysis The wide-scan XPS spectra (Fig. 1A) of MBT-GO show photoelectron lines at binding energies (BEs) of about 285.1 eV, 399.7 eV and 530.9 eV, attributed to C 1s, N 1s and O 1s, respectively[15,32]. The C 1s core level spectra of GO-MBT (Fig. 1B) could be the curve fitting 8   

into three peak components with BEs of 284.5, 284.9 &285.7, attributable to the sp2-hybridized carbon, C-C, C-O, and C=O species, respectively[33,34]. The N1s spectra (Fig. 1C) of the MBTfunctionalized GO adsorbent exhibits two peaks at 397.7 eV&399.7 eV, respectively which corresponds to the nitrogen peak of the MBT ligand; the successful functionalized GO surface was confirmed by XPS. The O1s spectra (Fig. 1D) were assigned to the corresponding BE’s of 529.4 &530.7 eV attributed to the adsorbent.

Deconvolution of Hg4f spectra(Fig. 1E) shows two peaks at 101.3 eV &105.4 eV indicates to Hg(II) adsorption onto the surface of the adsorbent, which has a spin–orbit splitting of 4.1 eV for the 4f7/2 and 4f5/2 states [15,18,23]. The Hg(II) adsorption is followed by the appearance of a doublet (Hg 4f5/2 and 4f7/2) with symmetric peaks, the BEs of Hg 4f7/2 to the most appropriate reference compounds are 101.3 eV for HgCl2 and 105.4 eV for HgO; it is possible to assume that the adsorbed mercury species are HgCl2&HgO which reported in earlier studies[23].  

3.2. Powder XRD Powder XRD data of graphite [35] at 2θ = 26.2o (d = 0.33 nm) corresponding (Fig.2A) to the plane (002) and graphite oxide (Fig. 2B) which shifted to 2θ = 10.4o (d = 0.84 nm) by chemical oxidation, indicating that graphite completely oxidized to GO which is reported in our earlier studies [23]. The MBT-functionalized GO has new peaks emerge corresponding to 2θ = 13.3o, 16.3 o, 17.0o, 22.2o, 26.3o, 29.1o&32.9o respectively (Fig. 2C), the secondary interaction of Hg(II) which gives a characteristic peak at 31.7o (Figure not shown). These results indicate that the metal ion could penetrate easily onto the adsorbent surface and interact effectively. The

9   

adsorption of Hg(II) onto adsorbent peaks is sharpening further, sharp peaks which indicate that crystalline nature of the adsorbent [10].  

3.3. Raman spectra Raman spectra of graphite which shows a D peak at 1350 cm−1 (D band) of crystalline graphite and a very strong peak at 1575 cm−1 (G band) C=C sp2 stretching vibration of olefinic/conjugated chains, a 2441 cm-1 D’ band and a 2D band at 2701cm-1 (Fig. 3A). GO to show a D band of 1360 cm−1 , G bands at 1601 cm−1 and a D’ band at 2443 cm-1(Fig. 3B). However, D&G bands have equal intensity, confirming that graphite completely oxidized to GO. After oxidation to GO, the intensity of the ID/IG ratio increased from 0.52 to 0.99, which reported in our earlier studies [23]. The functionalized GO to further confirms the formation of new C–S bond the most significant effect of thiolation is the appearance of new peaks (Fig. 3C), approximately 625&981 cm-1 which prove that formation of C–S bond [36, 37].

3.4. UV-Visible spectroscopy GO which gives distinct characteristic peaks were observed at 234 nm [33,34] corresponding to π-π* transitions of C=C bond and 305 nm due to n-π* transitions of COOH groups which reported in our earlier studies [23] (Fig. 4A), after reaction with MBT the absorption peaks at 234& 305 nm had disappeared. However, a new absorption peak appeared at 285 & 322nm (Fig. 4B), this could be explained that the absorbance peak was gradually red-shifted, this red-shift indicating that incorporation of MBT ligands [26, 38].

3.5. TEM and FT-IR spectral analysis

10   

The TEM images to provide evidence of monolayer graphene-like structure (Fig. 5A, B), GO has flatly deposited, without coagulation, are visible from the bright-field TEM images (Fig. 5C, D) stretched out holes in the lacy carbon support [18]. The FT-IR spectra of GO, has several functional groups that were introduced after oxidation from graphite to GO, such as hydroxyl, carboxylic acid, and epoxy groups, peak at 3715 cm−1 are attributed to free hydroxyl groups and 3408 cm−1 could be assigned to the O-H stretching from carboxyl groups (O=C−OH and C−OH), the peak at 2303 cm−1 could be associated with O−H stretching from strongly hydrogen-bonded COOH group (Figure not shown). The peak at 1564 cm−1 is related to the carboxylate anion, the peak at 1638 cm−1 could be associated with stretching from the GO backbone [15,16,23]. Moreover, the carbonyl group (C=O) stretching vibrations has emerged as a new peak at 1705 cm-1 which reported on our earlier studies [15, 23].

MBT-GO adsorbent FT-IR spectra (Fig. 6A) show clear bending and stretching vibrations of hydroxyl (OH), carbonyl/carboxyl, epoxy and aromatic groups, OH peaks to indicate a relatively free hydroxyl group and carboxyl group, these functional groups could interact with MBT ligands (Fig. 6A&E), which involves to inter and intra-molecular hydrogen bonding. A strong absorption peak at 1569 cm-1 was assigned to (C=N) stretching vibration, and new strong peak appeared at 1217 cm-1, which was assigned to (C-S) [40-42]. The XPS&FT-IR characteristics studies have fully demonstrated the functionalization of (MBT)-graphene oxide, are well documented. After adsorption of Hg(II) which indicates that FT-IR peaks were broader and sharp, the strong intensity increases in 1199 & 1005 cm−1 which main characteristic changes were observed (Fig. 6)[41-43].

3.6. SEM and EDX analysis 11   

The surface morphology revealed that pure or unmodified graphite which shows (Fig. 7A) vein and graphite flake shape morphology are made up of graphite intrinsic morphology. However, SEM images of GO have revealed a highly porous, homogeneous and interconnected structure (Fig. 7B). The surface of GO has lots of leaf-like veins that come from the scrolling of GO. The synthesized MBT-GO adsorbent SEM images (Fig. 7C) show that thin and slightly crumpled sheets are available on the surface [44], which demonstrates that the formation of individual GO sheets which provides to additional evidence for surface functionalized using a sulphur nucleophile, with subsequently yielding of surface thiol groups (GO-SH). The dense porous morphology got disturbed after adsorption of Hg(II) could accommodate onto the between layer of graphene sheets which we observed in the SEM images (Fig. 7D), after adsorption of Hg(II) onto the adsorbent surface, protuberant folds can be formed which might be due to the conglomeration of Hg(II) on the crumpled sheets of MBT-GO adsorbent surface. We could also authenticate in EDX spectra before and after adsorption of Hg(II) the characteristic peaks were appeared (1-3 keV) in the EDX spectra were confirmed by the presence of adsorbed Hg(II) along with other elements such as C, N. O, S respectively (Fig. 8B)[16,23].

3.7. pH point of zero charges (pHPZC) The adsorbent surface acquire a neutral charge is referred to point of zero charges;- pHpzc is one of the key factors that controls the adsorption process on carbon materials, pHpzc of the adsorbent was measured at 25 oC by taking 0.30 g of adsorbent in 50 mL of 0.1 mol L-1 sodium chloride electrolyte solution the initial pH was adjusted using 0.1 mol L-1 HCl or NaOH, with constant magnetic shaking in 120 rpms, after a time period of 24 h, the final pH was measured[45]. The surface charge of adsorbent is highly negatively charged and zeta potential becomes more negative with the increase of pH within the pH range was well documented. This phenomenon 12   

may be due to the ionization of the different groups (carboxylic/or hydroxyl group) on the surface of the adsorbent. The point of zero charges (pHpzc) was determined from the plot of ΔpH [pH

initial

– pH

final]

versus pH

initial

(Fig. 9A). The pHpzc was found to be 6.68, it is well known

that a solid surface is positively charged at pH< pHPZC and negatively charged at pH> pHPZC. As shown in Fig. 9A, Hg(II) adsorption depends on pH and increases sharply with the increases of pH, which reaches a maximum value of pH = 6.9. This fact may benefit from the ion exchange of Hg(II) with H+ onto adsorbent surfaces during the adsorption process, and the exchanged H+ ions are released to solution thereby leading to the decrease of pH values. Less removal efficiency is observed at the low pH range, which may be related to the fact that there are more protons in acid environment available to protonate active sites of adsorbent surfaces, and competition between Hg(II) ion and H+ occurred in solution [15,23]. In alkaline pH, the lower concentration of H+ and a greater number of ligands with negatives charges lead to greater Hg(II) adsorption. However, electrostatic attraction between positive Hg(II) ion and the negative surface of adsorbent could offer another driving force to enhance Hg(II) adsorption. Hence, in the pH range of 5.4 – 6.9, removal of Hg(II) increased slowly and reached the maximum efficiency of 98.5%.

3.8. Effect of pH The effect of pH(Fig. 9B), adsorption of Hg(II) by MBT-GO adsorbent was studied by varying pH over a range of 1 to 10 using 0.1 N NaOH/HCl. In acidic medium (pH 1 to 5.0), there was less adsorption of Hg(II) because of the chloro complex and competition between monovalent protons (H+) to the binding sites [23,46]. As it is seen, the predominant species between pH 1 to 5.0 is mercury, as HgCl2(aq); when the pH, decreases Cl concentration increases as a result of very stable Hg(II) chloride complexes, namely HgCl3- (aq) and HgCl42- (aq), which become 13   

predominant in aqueous medium. Therefore, under these conditions, there is a true competition between the formations of the Hg(II)-sulfur bond [23].

In acidic pH, which is not favourable to adsorption of Hg(II) onto the adsorbent, adsorbent surface has hydroxyl and carboxyl groups could get protonated which becomes positively charged it leads to electrostatic repulsion and became a less adsorption. In alkaline pH, the attraction between the negatively charged adsorbent surface owing to the deprotonation of the surface hydroxyl groups and positive Hg(II) ion results an increase in the percentage adsorption. The optimum pH is 5.4 to 6.9, (Fig. 9B) thus mercury would be a neutral, existing as divalent mercury, and it’s ready to accept a lone pair of electrons from sulphur, oxygen, or nitrogen atoms, which generates the surface complexation [23,46,47]. According to the Pearson’s concept, Hg(II), being a typical soft acid, has the ability to coordinate with sulphur, oxygen, and nitrogen to form a stable metal chelate complex. It is well known that sulphur binds mercury very strongly and plays an important role in the speciation of mercury in anoxic environments. The stability constants for these complexes are [48]

Hg2++HS‐.HgS0+H+→log K=26.5  Hg2++2HS‐.Hg(S2H)‐+H+→log K=32.0  Hg2++2HS‐.Hg(SH)20+H+→log K=37.5

14   

Furthermore, highly basic pH, adsorption of Hg(II) decreases because of the competition of hydroxide ions (OH-); formation of HgOH+, Hg(OH)2 species

it leads to a reduction in

percentage adsorption and hydroxyl species could compete for the active adsorption sites to the adsorbent surface [15,23]. 3.9. Adsorption mechanism In MBT-functionalized GO, S atoms play a significant role in Hg(II) adsorption, GO normally contains carboxylic, hydroxyl and epoxy groups, therefore easily dispersed in polar solvents, including water and DMF. Based on the advantage, thionation mechanism reported by Ozturk et al [49], and demonstrated for carbon nanotubes by Cech et al [50]. GO directly reacts with MBT ligand by transforming the (-R-Cl) and (R-COCl) functional groups of GO into thiol (R-SH) & thiocarboxylic (R-COSH) groups [40], respectively (schematically diagram shown in Fig. 10). MBT could be dissociated with a pronounced polar character of the CϨ+ SϨ- bond, which is alternatively binds to CϨ+=OϨ- or OϨ- = HϨ+ groups of GO; nucleophilic SϨ- will attack CϨ+, in contrast, and bind to SϨ-, forming a ring transition state that leads to the replacement of chloride (-Cl) from GO with S atoms to form [R-C-S] and [R-CO-S] groups on GO’s adsorbent surface[40]. The increase of Hg(II) adsorption by the MBT-GO adsorbent material is attributed to the metal-chelating thiol functionality, which provides the MBT-GO adsorbent surface with a remarkable affinity for Hg(II), thiol groups were involved in the binding of the Hg(II) ions, suggesting that the Hg(II) ions are coordinated with the adsorbent [15,23]. A large number of binding sites exist on adsorbent surfaces such as sulphur, hydroxyl groups, the nitrogen atom of the amino group (MBT) and oxygen atoms of the hydroxyl groups could bind to a proton or toxic heavy metal ion by electron pair sharing. The electronegativity of oxygen being more than that of nitrogen, donation of a lone pair of electrons from the nitrogen atom will be more facile than the oxygen atom to make bond formation with 15   

mercury [46], under highly acidic conditions (pH 1 to 2), there was less sorption of mercury, it could be explained by binding sites are protonated (OH, COOH) which causes repulsion of the approaching mercury cation. As pH increases (5.4 to 6.9), sorption of mercury increases due to electron rich binding sites is exposed this could allow metal ion binding to the adsorbent. In addition, the adsorbent surface becomes deprotonated resulting to the negative charge, the strong electrostatic attraction between the negatively charged carboxyl group and a hydroxyl group to the positive or neutral mercury species.

GO-COOH+OH-↔Go‐COO‐+H2O(A)  GO-COO-+Hg2+↔Go‐COO‐Hg+→(B)Electrostatic interaction GO-OH+OH-↔GO‐O‐+H2O→(C)  GO-O-+Hg2+↔GO‐O‐Hg+→(D)Electrostatic interaction

Hg2+ found to be adsorbed at a faster rate for higher percentage removal than organic mercury ions. Since the Hg(II) ion has a smaller ionic radius, it is possible that Hg(II) ions diffuse faster through the adsorbent pores than the bulkier organic mercury[46].The rate of adsorption is fast, in the case of Hg2+ > CH3Hg2+ > C6H5Hg2+. The abundance of carboxylic acids and other oxygen-containing functional groups on GO, Hg(II) is expected to preferentially bind with thiol and other SH-containing groups; in fact, the name ‘‘mercaptan’’ for sulfhydryl compounds stems from its affinity with Hg(II). The adsorbent (-SH) groups would be involved in covalent interaction with the surface hydroxyl, carboxyl groups on the surface of GO; furthermore, MBT electron donor groups, like the nitrogen, oxygen, and sulphur groups, donate electrons to Hg(II), which generates surface complexation [23]. The overall mechanism that could be conceptualized in this adsorption process is given in Fig. 10. Furthermore, equilibrium constant corresponding to 16   

the system has a superior value, (log K = 37.3) this is strong evidence for the S-mercury bond strength [51].

Hg(II)+2HS.Hg(HS)2 log k=37.3

3.10. Amount of adsorbent In the batch experiments, the amount of Hgthe adsorbent material has varied from the range of 0.01 to 0.10g, amount of Hg(II) adsorbed onto the adsorbent was found to be maximum (99.5 %) in a 20 mL sample volume. The available adsorption sites and surface areas increase by varying the adsorbent dose; therefore, the results reveal an increase in the percentage adsorption of Hg(II) [10,12].The percentage adsorption increases in adsorbent doses, amount of Hg(II) adsorbed per unit mass decreases, this is supported from the trend in % adsorption of Hg(II), which shows a sharp increase initially and later attains to maximum (Fig. S1A of supporting information ESI†). The decrease in % adsorption of Hg(II) with an increase in adsorbent doses is attributed to desaturation of adsorption sites in the process of adsorption [ 23]. 3.11. Adsorption isotherm 3.11.1. Langmuir isotherm Langmuir isotherm model [52] adopts uniform adsorption and describes monolayer adsorption on a surface containing a limited number of identical sites. This isotherm portrays the relation between the equilibrium concentration of the Hg(II), Ce and the amount adsorbed at equilibrium (qe) on the adsorbent surface. The Langmuir isotherm model was used to acquire the maximum adsorption capacity, which is calculated as the amount of Hg(II) adsorbed per unit weight of the

17   

MBT-GO adsorbent. By fitting the experimental data to Langmuir isotherm model, we could obtain the various Langmuir isotherm parameters and the linearized Langmuir equation can be expressed as

(2)

Ce qe

1 q0b

Ce q0

‘qo’ and ‘b’ are the Langmuir constants related to the adsorption capacity and intensity, respectively. A plot of Ce/qevsCe gives the qo & b (Fig. S2A-C of supporting information ESI†), and the feasibility of the adsorption process is determined by RL (which is dimensionless), known as the separation factor [23], which is given as

(3) RL

1 1 bC0

The RL value has considerable importance when it is between 0 and 1, where it implies an effective interaction between the adsorbent and the adsorbate. Values greater than 1 are an indication of an unfavorable isotherm, and RL equal to zero is accounted for a totally irreversible isotherm; these features are summarized in TableT2 of the supporting information (ESI†). The Langmuir isotherm parameters, and a relatively good regression coefficient (TableT3 of supporting information ESI†), indicate the effectiveness of interaction between the Hg(II) and the adsorbent [23].

18   

3.11.2. Freundlich isotherm The Freundlich isotherm based on sorption on a heterogeneous surface also serves as an alternative useful model to assess the adsorption from dilute solutions. The linearized expression correlating equilibrium adsorption capacity can be expressed as[53]

(4) log qe

log K F

1 log Ce n

These constants are easily obtained from the log-log plots of qe against Ce (Fig. S2D-F of supporting information ESI†). KF and n are the Freundlich constants that indicate the adsorption capacity and the adsorption intensity, respectively; a favorable adsorption would have a Freundlich constant n between 1 and 10. A higher value of n (smaller value of 1/n) implies an effective interaction between the adsorbent and adsorbate (TableT3 of supporting information ESI†). When 1/n < 1, it corresponds to a normal L-type isotherm, while 1/n >1 reflects a cooperative sorption [23]. 3.12. Adsorption kinetics 3.12.1. Pseudo-first-order kinetics The first-order rate expression of Lagergren [54] can be expressed mathematically as

(5) log(qe

qt )

log qe

K1t 2.303

19   

These equations relate the amount of Hg(II) adsorbed onto the adsorbent and the rate constants at varying time intervals, where qe and qt (mg g-1) are the amounts of Hg(II) adsorbed per unit mass of adsorbent at equilibrium and time in min, respectively, and k1 is the rate constant. The values of the adsorption rate constant (k1) for the sorbents at different initial Hg(II) concentrations were obtained from slopes of the plots of log (qe-qt) vs time (Fig. S3A-C of supporting information ESI†). 3.12.2. Pseudo-second-order kinetics A pseudo-second-order reaction model [55] utilized in the study of adsorption can be expressed mathematically as:

(6)

t qt

1 2 k2 qe

t qe

The t/qt versus t plot (Fig. S3D-F of supporting information ESI†) in the second-order model gave a nice fit to the experimental data in terms of the higher regression coefficient (Table T4 of supporting information ESI†). This proximity in the values (Table T4 of supporting information ESI†) further confirms the suitability of the second-order kinetics to the adsorption data. 3.12.3. Intraparticle diffusion The Weber-Morris model is used to study the intra-particle diffusion and this relates the amount of Hg(II) adsorbed against the square root of time[56]:

qt = k int√t + C

(7)

Where kint is the intra-particle diffusion constant and qt is the amount of Hg(II) adsorbed at times t. The overall rate of adsorption Hg(II) onto the adsorbent could be influenced by (a) film or 20   

surface diffusion, where Hg(II) is transported from the bulk solution to the external adsorbent surface (b) intraparticle or pore diffusion, where the adsorbate Hg(II) molecules move into the interior of the adsorbent particles and (c) Hg(II) adsorption on the interior sites of the adsorbent[10,12,23]. The adsorption step is fairly rapid; it is assumed that it does not bear profound influence on the adsorption kinetics. Therefore, the overall rate of Hg(II) adsorption onto the adsorbent could be controlled by the surface or intraparticle diffusion. The WeberMorris intraparticle diffusion model [57] is appropriate to conclude by intraparticle diffusion is the rate-determining step. According to the W-M model, a plot of qt versus √t (Fig. S3G-I of supporting information ESI†) would be linear [57], intraparticle diffusion is implicated in the adsorption process furthermore, the plot cuts through the origin then intraparticle diffusion is the only rate-limiting step. The plot of qt against √ t yields a finite intercept (Fig. S3G-I of supporting information ESI†) this indicates that more than diffusion, boundary layer mechanism could also be involved in following the adsorption kinetics of Hg(II) onto the adsorbent interface [23]. 3.13. Adsorption thermodynamics The thermodynamics study is an important parameter in order to ascertain the feasibility and nature of the adsorption process. The feasibility of an adsorption process can be best understood from the thermodynamic parameters namely standard free energy (ΔG0), standard enthalpy (ΔH0), and standard entropy (ΔS0), were determined at various temperature ranges. From thermodynamic studies, the values of the equilibrium constants, ΔH0 & ΔS0, were obtained from the slope and intercept of the van’t Hoff plot of lnK against 1/T (Fig. S3J of supporting information ESI†). For an exothermic reaction, the slope is positive and equilibrium constant decreases from increases in temperature [10,12, 23, 58] 21   

ΔG0 = -RT lnK

H0 (9) ln K RT

(8)

S0 R

Where is R is the gas constant (J K-1 mol-1), T is the temperature in Kelvin, the value of K was calculated from the ratio of concentration of Hg(II) adsorbed onto the adsorbent, the concentration in the aqueous phase (K = C(Hg(II) in solid) /C(Hg(II) in solution)). There is a cooperative interaction reflected on the adsorption process ascertained through the equilibrium constant K and Gibb’s free energy values. The overall free energy of adsorption could be expressed as [10,12] ΔG adsorption = ΔG (MBT-GO) + ΔGadsorbent-Hg(II)

The negative free energy (ΔGo) values to confirm the effectiveness of the interaction between Hg(II) and the MBT-GO adsorbent. The free energy values decreased from rises in temperature; negative free energy is a good indication of spontaneous adsorption. Furthermore, the negative value of ΔH shows that the adsorption process is exothermic.

ΔH adsorption = ΔH (MBT-GO) + ΔH adsorbent-Hg(II)  ΔS adsorption = ΔS (MBT-GO)

+ ΔS adsorbent-Hg(II) 

22   

With ΔG, ΔS, ΔH being negative and K > 1 the increase in temperature does not favour the adsorption process in accordance with Le Chatlier principle. Hence,

ΔG adsorption = ΔH adsorption - T (ΔS(GO) + ΔS(MBT) + ΔS(Hg(II))

The adsorption of Hg(II) was more favourable to room temperature and adsorption gradually decrease at a higher temperature; this result reveals that adsorbate–adsorbent interaction weakening at higher temperatures, which is indicating that higher temperature are not favourable for the adsorption process. (Table T5 of supporting information ESI†). The entropy of adsorption (ΔS0) was also negative; this is indicative of decreased randomness at the adsorbent-solution interface. These facts demonstrate the efficacy of the adsorbent material as a useful material for Hg(II) adsorption. The energy of activation (Ea) at different temperatures was obtained using the relation Ea = ΔHads + RT, the negative Ea= -59.44 kJ mol-1 also confirms that exothermic nature of adsorption process [12]. 3.14. Column studies 3.14.1. Effect of sample volume A glass column 3.5 cm in a diameter and 40 cm in length was used for the column adsorption study and 1.5 g of the MBT-GO adsorbent material was packed in the glass column, up to a height of 2.0 cm. A 600 mL volume of 50 mg L-1 Hg(II) was delivered into the column at a flow rate of 5 mL min-1, Hg (II) was effectively adsorbed on the column at pH 5.4 to 6.9, and this was determined using ICP-MS from the concentration of Hg(II) in the solution phase, the retention of Hg(II) was quantitative up to 600 mL sample volume (Fig. S4A of supporting information ESI†). Above 600 mL, there is a decrease in the amount of Hg(II) adsorbed this could be attributed to the column expansion [10, 23], thereby disturbing the close packing of the adsorbent to the 23   

column studies. The performance of an adsorbent could also be assessed based on the adsorbent exhaustion rate. This parameter is expressed as the ratio of the mass of the adsorbent to the maximum sample volume in litres [23], with a short laboratory column containing 1.5 g of the adsorbent, the exhaustion rate of the adsorbent was found to be 2.5g L-1. A lower value denotes the efficacy of the adsorbent towards tolerating a particular sample volume. With an increase in the amount of the adsorbent, the sample volume would also enhance correspondingly. 3.15. Adsorbent regeneration studies The regeneration of the adsorbent subsequent desorption of mercury from the adsorbent surface is an activated process since adsorbed Hg(II) needs to effectively lift from the base of the potential energy well. The regeneration of adsorbent is an important aspect to be examined in an adsorption process; choice of a reagent for desorption depends on its ability to interact with Hg(II) in relieving the adsorbent–adsorbate interaction. In this regard, reagents such as potassium bromide, Ethylenediamine tetraacetic acid (EDTA), thiourea, potassium iodide,  potassium thiocyanate, diethylene triamine Penta acetic acid (DTPA) and potassium chloride were explored and we observed that KI was the most efficient in desorbing mercury as its tetraiodomercurate(II) anion into the aqueous phase. On the overall stability constant [59] (log K4) of mercury complexes with respect to I2, SCN2, Br2 and Cl2, as (Hg (SCN)42-, HgCl42- , HgBr42-, HgI42- stability constant in order wise: 20.9, 15.1, 20.0, and 29.8). In the above order for the stability constant as described Hg(II) has good selectivity towards to potassium iodide. This is due to the strong interaction of iodide with Hg(II). Furthermore, in the aqueous phase, Eo (standard reduction potential) of I2/I−, Br2/Br−, and Cl2/Cl− are +0.54 V, +1.09 V, and +1.36 V, respectively. Hence, in aqueous acidic medium, it is also highly probable that iodide could be oxidized easily and it could be eluted as a covalent HgI2 in aqueous phase [10,60].

24   

Hg2++I‐.HgI+ logK1=12.83(10) 

Hg2++2I‐.HgI2 logK2=24.07(11) 

‐

Hg2++3I‐.HgI 3 logK3=27.77(12) 

2‐

Hg2++4I‐.HgI4  logK4=29.79(13)

It is quite obvious that equation 13 from the above, characterized by a high formation constant value, regeneration of the adsorbent using KI was the most efficient in desorbing mercury as its tetraiodomercurate(II) anion into the aqueous phase.

Hg+I2→HgI2 Hg+I2+KI→KHgI3 Hg+I2→K2HgI4  

However, the proposed methodology we found that 20 mL of 4.0 mol L-1 KI was effective to quantitative desorption of Hg(II). In order to regenerate the adsorbent material, an elution step was carried out after each adsorption cycle, when the adsorbent was saturated. The effective operation of the next sorption process is clearly related to the efficiency of the preceded desorption step. After each elution operation, the adsorbent was washed with Milli-Q water in order to eliminate the remaining iodide in the adsorbent; which used by 0.1 N NaOH/HCl solution to bring pH above 5.5 for subsequent adsorption studies. We were carried out to the 5 adsorption-desorption cycles and found that the adsorbent is stable and retained its performance efficiency for 3 cycles (98.5%). The adsorbent could be reused for 3 adsorption-desorption 25   

cycles without any noticeable decrease in performance efficiency [12,23]. Beyond 3 cycles, the available active sites in the adsorbent get saturated which are not readily available for subsequent adsorption studies due to gradually decreasing adsorption efficiency. However, after 3 cycles the percentage adsorption of Hg(II) decreases, which could be ascribed to KI weakening the interaction between Hg(II) and the adsorbent [23], the soft–soft interaction (Hg–I) favours the facile desorption studies. 3.16. Interference study Industrial effluents systematically contain a series of other elements (cations, anions) that can be used to evaluate the adsorbent; and also important to check how the presence of competitor ions may impact sorption performance. Various cations (Zn2+, Pb2+, Cu2+, Cd2+, Co2+, Ni2+, Se2+, Au+, Cr3+, Fe2+) and anions (such as SO42-, NO3−, PO43-, Cl−) have been examined for their impact on Hg(II) adsorption. A 10 mg L

-1

Hg(II) solution was mixed with the various concentrations of

cations&anions, adsorption studies were carried out by using MBT-GO adsorbent and adsorption studies were optimized in pH=5.4-6.9. The presence of N & S atoms from the MBT ligand could prove to be quite conducive to the adsorption of Hg(II). In order to investigate the adsorption selectivity, studies were conducted to the presence of cations such as Ni2+ (100 mg L-1) Co2+ (100 mg L-1) pb2+ (100 mg L-1), Zn2+ (100 mg L-1) Cu2+ (100 mg L-1), Au+ (200 mg L-1), Al3+ (100 mg L-1), Ba2+ (100 mg L-1), Cs+ (200 mg L-1) Bi3+ (200 mg L-1), Mn2+ (200 mg L-1), Se2+ (200 mg L-1) Ag+ (100 mg L-1), Cd2+ (100 mg L-1), Ca2+ (200 mg L-1), Ga3+ (200 mg L-1), Cr3+ (200 mg L-1) and Fe2+ (200 mg L-1), independently to 100 mL samples volume containing 30 mg L-1 of Hg(II). According to Pearson’s theory (HSAB – hard and soft acid and base theory – soft acid attract soft base and hard acid attract hard base ), S groups are classified as a soft base it can be reacted with soft acids such as Hg(II) and Cd(II) [10, 47,51,61]. Hg(II) which is considered a soft acid that has a strong affinity with S ligands. On the contrary, Zn(II), Pb(II), Cu(II), and 26   

Ni(II) are classified as borderline metals that have less affinity with S ligands (compared to soft acids like Hg(II)). This could be explained; the bi-component solutions to Hg(II) adsorption which are not affected by the presence of other bivalent metal ions; Cd(II) did not interfere significantly to Hg(II) adsorption under the selected experimental conditions, which indicates that the adsorbent had a preference for Hg(II) adsorption over Cd(II). The effects of anions (i.e., SO42-, NO3−, PO43-, Cl−) interference were studied by addition of the potassium and sodium salts of these anions which presence of the Hg(II) ion solutions (10 mg L-1); and also anions such as Cl- (250 mg L-1), NO3-(250 mg L-1), SO42- (250 mg L-1), PO43- (250 mg L-1), did not cause the adsorption of Hg(II). 3.17. Application to study the removal of mercury from a fish tissue sample Mercury occurs in the environment in various chemicals forms and among them methylmercury (CH3Hg+), which are formed by methylation of inorganic mercury (Hg[II]) in aquatic environments. It’s most toxic and this could be biomagnified by a factor of 105-107 in predatory fish, causing adverse effects to humans via fish consumption [62]. Inconsequently, “Hg compounds” are recently added an entry on the list of priority pollutants recognized by the Environmental Protection Agency (EPA). In their issue of the “Guidance of Implementing January 2001 Methylmercury Water Quality Criterion,” a maximum level of 0.3 μg g-1 of fish tissue (methylmercury, wet weight) is recommended for freshwater fish. Therefore, the development of reliable and fast, cost-effective methodologies for the determination of Hg(II) and CH3Hg+ in food stuff is required, particularly in the field of food safety [62]. A 0.5-g sample of fish tissue was pre-treated using HNO3−H2SO4 mixture to destroy organic components, after that fish tissue dissolved completely; samples were allowed to cool in room temperature, resulting in a mixture, were filtered and brought to a known volume. The Hg(II) was adsorbed onto the adsorbent material, the adsorption of Hg(II) was quantitative, 27   

and this was ascertained by using ICP-MS technique shown in Fig. 11. The MBT-functionalized adsorbent was effective against the removal of Hg(II) from the fish tissue sample, and furthermore, the adsorbed Hg(II) could also be desorbed quantitatively using potassium iodide. The concentration of Hg(II) in the fish tissue sample was found to be 19.1 μg kg−1 obtained as the average of three replicating determinations (Table 1). From these results, it’s evident that MBT-GO adsorbent is highly efficient to the adsorption of Hg(II) from fish tissues. Finally, the developed method has proved to be successful in the adsorption of Hg(II) from fish tissue, which is a vital source of mercury pollution. 3.18. Analysis of real samples The proposed method was applied to river water, pond water and seawater samples, water samples were collected from Kaohsiung in Taiwan. The reliability of the method was investigated by spiking the samples, results from data obtained by ICP-MS analysis, the concentration of Hg(II) was found to be quantitative in the real water samples, results are shown in Table 2. 3.19. Comparison of adsorption behaviour based on literature data The efficacy of MBT-functionalized GO adsorbent was evaluated with regard to its adsorption capacity against other adsorbents [63]. The comparison given in Table 3 exemplifies the fact that MBT-GO adsorbent shows very good prospect in adsorbing Hg(II). 4. Conclusion

A thiol-functionalized graphene oxide was synthesized for the first time the prospective application of Hg(II) adsorption has been well documented in this study, and this method could be highly selective for the adsorption of Hg(II) resulting from the soft acid of the mercury by specifically binding to the soft base of thiol. The adsorption of Hg(II) correlated with the Langmuir isotherm, maximum adsorption capacity was found to be 107.52 mg g-1 and kinetics 28   

processes reached their equilibrium state within 30 min, which is faster than most other graphene-based adsorbents, thermodynamically favourable adsorption process is driven by negative enthalpy and entropy changes, respectively. The adsorbent material could be effectively regenerated by using KI and method could be scaled up to a large sample volume of 600 mL. Regeneration and stability of the adsorbent for 3 repetitive cycles are yet another benefit to this interesting adsorbent material; there are considerable changes to the adsorption of Hg(II) from the aqueous solutions, which presence of various cations and anions. Finally, the developed method has proved to be successful to the adsorption of Hg(II) from fish tissue, which is a vital source of mercury pollution. Acknowledgments

Authors greatly acknowledge the financial support from Ministry of Science and Technology (MOST), Taiwan (Republic of China), under grant NSC: 100-2113-M-110-002-MY3. We also thank MOST for the post-doctoral grant under contract No: MOST-102-2811-M 110-039. References [1] S. Mokhtari, H. Faghihian,;1; Modification of activated carbon by 2, 6‐diaminopyridine for separation  of Hg2+ from aqueous solutions, J. Environ. Chem.  Eng.  3 (2015) 1662–1668.    [2] S.Thakur, G. Das, P. K. Raul, N. Karak,;1; Green one‐step approach to prepare sulfur/reduced  graphene oxide nanohybrid for effective mercury ions removal, J. Phys. Chem. C 117 (2013) 7636−7642.    [3]  L.Wang, T. Yao, S. Shi, Y. Cao, W. Sun,;1; A label‐free fluorescent probe for Hg2+ and biothiols based  on graphene oxide and Ru‐complex, Scientific reports, 4 (2014) 5320‐5325.    [4] ;1;  A. Santhana Krishna Kumar, M. Barathi, S. Puvvada, N. Rajesh, Microwave assisted preparation of  glycidyl methacrylate grafted cellulose adsorbent for the effective adsorption of mercury from a coal fly  ash sample, J. Environ. Chem.  Eng.  1 (2013) 1359‐1367.  29   

  [5]  S. P. Bayen, P. Chowdhury,;1; Synthesis of chromatographic material by immobilization of  thioacetamide onto silica gel for easy detection and removal of mercury, J. Environ. Chem.  Eng. 3 (2015)  70–78.     [6]  M. Hami Dindar, M. Reza Yaftian, S. Rostamnia,;1; Potential of functionalized SBA‐15 mesoporous  materials for decontamination of water solutions from Cr(VI), As(V) and Hg(II) ions, J. Environ. Chem.  Eng. 3 (2015) 986–995.    [7]  S. Zhang, Y. Zhang, J. Liu, Q. Xu, H. Xiao, X. Wang, H. Xu, J. Zhou,;1; Thiol modified Fe3O4@SiO2 as a  robust, high effective, and recycling magnetic sorbent for mercury removal, Chem. Eng. J. 226 (2013)  30–38.    [8]  R. Sharma, N. Singh, S. Tiwari, S. K. Tiwari, S. R. Dhakate,;1; Cerium functionalized PVA–chitosan  composite nanofibers for effective remediation of ultra‐low concentrations of Hg(II) in water, RSC Adv. 5  (2015)16622–16630.    [9] T. Zhang, Z. G. Cheng, Y. Wang, Z. Li, C. Wang, Y. Li, Y. Fang,;1; Self‐assembled 1‐octadecanethiol  monolayers on graphene for mercury detection, Nano Lett. 10 (2010) 4738‐4741.    [10] A.;1; Santhana Krishna Kumar, S. Kalidhasan, V. Rajesh, N. Rajesh, Adsorptive demercuration by  virtue of an appealing interaction involving biopolymer cellulose and mercaptobenzothiazole,  Ind. Eng.  Chem. Res. 52 (2013) 11838–11849.    [11] K. Chen, G. Lu, J. Chang, S. Mao, K. Yu, S. Cui, J. Chen,;1; Hg(II) ion detection using thermally reduced  graphene oxide decorated with functionalized gold nanoparticles, Anal. Chem. 84 (2012) 4057−4062.    [12] A.;1; Santhana Krishna Kumar, S. Kalidhasan, V. Rajesh, N. Rajesh, A meticulous study on the  adsorption of mercury as a tetrachloromercurate(II) anion with trioctylamine modified  sodium  montmorillonite and its application to a coal fly ash sample, Ind. Eng. Chem. Res. 51(2012)  11312−11327.    30   

[13] R. Li, L. Liu, F. Yang,;1; Preparation of polyaniline/reduced graphene oxide nanocomposite and its  application in adsorption of aqueous Hg(II), Chem. Eng. J. 229 (2013) 460–468.     [14] D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. Dommett, G. Evmenenko,;1;  S.T. Nguyen,  R. S. Ruoff, Preparation and characterization of graphene oxide, Nature  448 (2007) 457‐460.    [15] V. Chandra, K. S. Kim,;1; Highly selective adsorption of Hg2+ by a polypyrrole–reduced  graphene oxide composite, Chem. Commun. 47 (2011) 3942–3944.    [16] T. S. Sreeprasad, S. M. Maliyekkal, K. P. Lisha, T. Pradeep,;1; Reduced graphene oxide‐metal/metal  oxide composites: Facile synthesis and application in water purification, J. Hazard. Mater. 186 (2011)  921–931.    [17] P. N. Diagboya, B. I. Olu‐Owolabi, K. O. Adebowale,;1; Synthesis of covalently bonded graphene  oxide– iron magnetic nanoparticles and the kinetics of mercury removal, RSC Adv. 5 (2015) 2536–2542.    [18] Y. C. Shih, C. Y. Ke, C. J. Yu, C.Y. Lu, W. L. Tseng,;1; Combined tween 20‐stabilized gold nanoparticles  and reduced graphite oxide−Fe3O4 nanoparticle composites for rapid and efficient removal of mercury  species from a complex matrix, ACS Appl. Mater. Interfaces 6 (2014)17437–17445.    [19] X. Gong, Y. Bi, Y. Zhao, G. Liu, W. Y. Teoh,;1; Graphene oxide‐based electrochemical sensor: A  platform for ultrasensitive detection of heavy metal,  RSC Adv. 4 (2014) 24653–24657.    [20] Y. Liu, C. Tian, B. Yan, Q. Lu, Y. Xie, J. Chen, R. Gupta, Z. Xu, S. M. Kuznicki, Q. Liu, H. Zeng,;1;  Nanocomposites of graphene oxide, Ag nanoparticles, and magnetic ferrite nanoparticles for elemental  mercury (Hg0) removal, RSC Adv. 5 (2015) 15634–15640.    [21] Y. Si, X. Wang, Y. Li, K. Chen, J. Wang, J. Yu, H. Wang, B. Ding,;1; Optimized colorimetric sensor strip  for mercury(II) assay using hierarchical nanostructured conjugated polymers,  J. Mater. Chem. A 2 (2014)  645–652.    31   

[22] Q. Zhai, S. Zhang, H. Jiang, Q. Wei, E. Wang, J. Wang,;1; Biomimetic nanopore for sensitive and  selective detection of Hg(II) in conjunction with single‐walled carbon nanotube, J. Mater. Chem. B 2  (2014) 6371–6377.    [23] A.;1; Santhana Krishna Kumar, S. J. Jiang, Preparation and characterization of exfoliated graphene  oxide–L‐cystine as an effective adsorbent of Hg(II) adsorption, RSC Adv. 5 (2015) 6294–6304.     [24] A. Gupta, S. R. Vidyarthi, N. Sankararamakrishnan,;1; Enhanced sorption of mercury from compact  fluorescent bulbs and contaminated water streams using functionalized multiwalled carbon nanotubes,  J. Hazard. Mater. 274 (2014) 132‐144.    [25] S. Muralikrishna, K. Sureshkumar, T. S. Varley, D. H. Nagaraju, T. Ramakrishnappa,;1; In situ  reduction and functionalization of graphene oxide with L‐cysteine for simultaneous electrochemical  determination of cadmium(II), lead(II), copper(II), and mercury(II) ions, Anal. Methods 6 (2014) 8698– 8705.    [26] H. Zhou, X. Wang, P.Yu, X. Chen, L. Mao,;1; Sensitive and selective voltametric measurement of Hg2+  by rational covalent functionalization of graphene oxide with cysteamine,  Analyst 137 (2012) 305–308.    [27] Avinash Gupta, S. R. Vidyarthi, N. Sankararamakrishnan,;1; Thiol functionalized sugarcane bagasse‐a  low cost adsorbent for mercury remediation from compact fluorescent bulbs and contaminated water  streams, J. Environ. Chem.  Eng.  2 (2014) 1378–1385.    [28] C. Zhang, J. Sui, J. Li, Y. Tang, W. Cai,;1; Efficient removal of heavy metal ions by thiol‐functionalized  super magnetic carbon nanotubes, Chem. Eng. J. 210 (2012) 45‐52.    [29] N. M. Bandaru, N. Reta, H. Dalal, A.V. Ellis, J. Shapter, N. H. Voelcker,;1; Enhanced adsorption of  mercury ions on thiol derivatized single wall carbon nanotubes, J. Hazard. Mater. 261 (2013) 534–541.    [30] J. Bao, Y. Fu, Z. H. Bao,;1; Enhanced adsorption of mercury ions on thiol derivatized single wall  carbon nanotubes, Nanoscale Res. Lett. 8 (2013) 486‐492.  32   

  [31] D. C. Marcano, D.V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Z. Sun, A. Slesarev, L.B. Alemany, W. Lu, J.  M. Tour,;1; Improved synthesis of graphene oxide, ACS Nano 4 (2010) 4806–4814.     [32] H. R. Thomas, A. J. Marsden, M. Walker, N. R. Wilson, J. P. Rourke,;1; Sulfur‐functionalized graphene  oxide by epoxide ring‐opening, Angew. Chem. Int. Ed. 53 (2014) 7613 –7618.    [33] S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S. T. Nguyen, R.  S.;1;  Ruoff, Synthesis of graphene‐based nanosheets via chemical reduction of exfoliated graphite oxide,  Carbon 45 (2007)1558–1565.    [34] S. Stankovich, R. D. Piner, X. Chen, N. Wu, S. T. Nguyen, R. S. Ruoff,;1; Stable aqueous dispersions of  graphitic nanoplatelets via the reduction of exfoliated graphite oxide in the presence of poly(sodium 4‐ styrenesulfonate), J. Mater. Chem. 16 (2006)155–158.    [35]  A. Banazadeh, S. Mozaffari, B. Osoli,;1; Facile synthesis of cysteine functionalized magnetic  graphene oxide nanosheets: Application in solid phase extraction of cadmium from environmental  sample, J. Environ. Chem.  Eng. 3 (2015) 2801–2808.    [36] G. Liu, Q. Fang, W. Xu, H. Chen, C. Wang,;1; Vibration assignment of carbon–sulfur bond in 2‐thione‐ 1,3‐dithiole‐4,5‐dithiolate derivatives, Spectrochim. Acta A 60 (2004) 541–550.    [37] J. Dong, L. Luo, P. H. Liang, D. D. Mariano, P. R. Carey,;1; Raman difference spectroscopic studies of  dithiobenzoyl substrate and product analogs binding to the enzyme dehalogenase: Pi‐electron  polarization is prevented by the C=O to C= S substitution,  J. Raman Spectrosc. 31 (2000) 365‐371.     [38] V. H. Pham, T. V. Cuong, T. D. N.;1;  Phan, H. D. Pham, E. J. Kim, S. H. Hur, S. Shin, E. W. Kim, J. S.  Chung, One‐step synthesis of superior dispersion of chemically converted graphene in organic solvents,   Chem. Commun. 46 (2010)4375–4377.   

33   

[39] A.;1;  Santhana Krishna Kumar, N. Rajesh, Exploring the interesting interaction between graphene  oxide, Aliquat‐336 (a room temperature ionic liquid) and chromium(VI) for wastewater treatment, RSC  Adv. 3 (2013)2697‐2709.    [40] C. V. Pham, M. Eck, M. Krueger,;1; Thiol functionalized reduced graphene oxide as a base material  for novel graphene‐nanoparticle hybrid composites, Chem. Eng. J. 231, (2013)146–154.    [41] A. Hamza, A. S. Bashammakh, A. A. Al‐Sibaai, H. M. Al‐Saidi, M. S. El‐Shahawi,;1; Part 1.  Spectrophotometric determination of trace mercury (II) in dental‐unit wastewater and fertilizer samples  using the novel reagent 6‐hydroxy‐3‐(2‐oxoindolin‐3‐ylideneamino)‐2‐thioxo‐2H‐1,3‐thiazin‐4(3H)‐one  and the dual‐wavelength‐correction spectrophotometry,  J. Hazard. Mater. 178 (2010) 287–292.    [42] K. Nakamoto, Infrared and;1; Raman spectra of Inorganic and Co‐ordination Compounds, Third ed.,  Wiley, New York, 1978.     [43] J. C. Feo, A. J. Aller,;1; Speciation of mercury, methylmercury, ethylmercury and phenylmercury by  Fourier transform infrared spectroscopy of whole bacterial cells, J. Anal. At .Spectrom.16 (2001) 146‐51.    [44] Y. Li, P. Zhang, Q. Du, X. Peng, T. Liu, T. Wang, Y. Xia, W. Zhang, K. Wang, H. Zhu, D. Wu,;1;  Adsorption of fluoride from aqueous solution by graphene, J. Colloid Interface Sci. 363 (2010) 348‐354.    [45] A.;1; Santhana Krishna Kumar, S. J. Jiang, W. L. Tseng, Effective adsorption of chromium(VI)/Cr(III)  from aqueous solution using ionic liquid functionalized multiwalled carbon nanotubes as a super  sorbent, J. Mater. Chem. A 3 (2015) 7044–7057.    [46] S. Kushwaha, B. Sreedhar, P. Padmaja,;1; Sorption of phenyl mercury, methyl mercury, and  inorganic mercury onto chitosan and barbital immobilized chitosan: spectroscopic, potentiometric,  kinetic, equilibrium, and selective desorption studies, J. Chem. Eng. Data 55 (2010) 4691–4698.    [47] M. Puanngam, F. Unob,;1; Preparation and use of chemically modified MCM‐41 and silica gel as  selective adsorbents for Hg(II) ions, J. Hazard. Mater. 154 (2008) 578–587.  34   

  [48] M. Ravichandran,;1; Interactions between mercury and dissolved organic matter‐a review,  Chemosphere 55 (2004) 319–331.    [49] T. Ozturk, E. Ertas, O. Mert,;1; A berzelius reagent, phosphorus decasulfide (P4S10), in organic  syntheses, Chem. Rev. 110 (2010) 3419–3478.    [50] J. Cech, S. A. Curran, D. Zhang, J. L. Dewald, A. Avadhanula, M. Kandadai, S. Roth,;1;  Functionalization of multi‐walled carbon nanotubes: Direct proof of sidewall thiolation, Phys. Stat. Sol. B  243 (2006) 3221–3225.    [51] A. Arencibia, J. Aguado, J. M. Arsuaga,;1; Regeneration of thiol‐functionalized mesostructured silica  adsorbents of mercury, Appl. Surface Sci. 256(2010) 5453–5457.    [52] I. Langmuir,;1; The adsorption of gases on plane surface of glass, mica and platinum, J. Am. Chem.  Soc.  40 (1918) 1361–1403.    [53]  H.M.F. Freundlich,;1; Over the adsorption in solution, Z. Phys.Chem. 57 (1906) 385‐470.    [54] S. Lagergren,;1; Zur theorie der  sogennanten adsorption geloster stoffe, K. Sven Vetenskapsakad.  Handl. 24 (1898) 1‐39.     [55] Y. S. Ho,;1; Review of second‐order models for adsorption systems, J. Hazard. Mater. 136 (2006)  681–689.    [56]  S. Muthu Prabhu, S. Meenakshi,;1; Chemistry of defluoridation by one‐pot synthesized dicarboxylic  acids mediated polyacrylamide–zirconium complex, Chem. Eng. J. 262 (2015) 224–234.    [57] W.J. Weber, J.C. Morris,;1; Kinetics of adsorption on carbon from solution, J. Sanit. Eng. Div. Am.  Soc. Civ. Eng. 89 (1963) 31‐60.  35   

  [58]  D. K. Mondal;1; , B. K. Nandi, M. K. Purkait, Removal of mercury (II) from aqueous solution using  bamboo leaf powder: Equilibrium, thermodynamic and kinetic studies, J. Environ. Chem.  Eng.  1 (2013)  891–898.    [59] M. Ghaedi, A. F. Shojaie, M. Montazerozohori, B. Karami, S. Gharaghan,;1; Selective electrodes  based on bis[N(2‐methyl‐phenyl) 4‐Nitro‐thiobenzamidato]mercury(II) and bis[N‐phenyl 3,5‐Dinitro  thiobenzamidato]mercury(II) carriers, Electroanalysis 17 (2005) 1746 ‐1754.    [60]  C. Cox, M. A. Shoesmith, M. M. Ghosh,;1; Electrokinetic remediation of mercury‐contaminated soils  using iodine/iodide lixiviant, Environ. Sci. Technol. 30 (1996) 1933‐1938.    [61]  M. A. K. C. Gavilan, A.V. Pestov, H. M. Garcia, Y. Yatluk, J.;1;  Roussy, E. Guibal, Mercury sorption on  a thiocarbamoyl derivative of chitosan, J. Hazard. Mater. 165 (2009) 415–426.    [62]  A. Castillo, P. Gonzalez, G. Centineo, A. F. Roig‐Navarro, J. I. G. Alonso,;1; Multiple spiking species‐ specific isotope dilution analysis by molecular mass spectrometry: simultaneous determination of  inorganic mercury and methylmercury in fish tissues. Anal. Chem. 82 (2010) 2773–2783.    [63] O. Hakami, Y. Zhang, C. J. Banks,;1; Thiol‐functionalised mesoporous silica‐coated magnetite  nanoparticles for high efficiency removal and recovery of Hg from water, Water Res. 46(2012)3913‐ 3922. 

Fig. 1 X-ray photoelectron spectroscopy (A) total survey scan (B) carbon spectra

(C) nitrogen spectra (D) oxygen spectra (E) mercury spectra Fig. 2 Powder XRD spectra of (A) graphite (B) graphene oxide (C) MBT functionalized-GO Fig. 3 Raman spectrum of (A) graphite (B) graphene oxide (B) MBT functionalized-GO Fig. 4 UV spectra of (A) graphene oxide (B) MBT functionalized-GO 36   

Fig. 5 TEM images of (A-B) graphite (C-D) graphene oxide Fig. 6 FT-IR spectra of scale between(500-4000 cm-1) MBT functionalized-GO adsorbent(A),

after adsorption of (B)Hg(II) (C)CH3Hg+ , (D)CH3CH2Hg+ Fig. 6 FT-IR spectra of scale between (500-2000 cm-1) MBT functionalized-GO adsorbent (E),

after adsorption of (F)Hg(II), (G)CH3Hg+, (H)CH3CH2Hg+ Fig. 7 SEM image of (A) graphite (B) graphene oxide (C) MBT functionalized-GO (D) after

adsorption of Hg(II) Fig. 8 EDX spectra of (A) MBT functionalized-GO (B) after adsorption of Hg(II) Fig. 9 (A) pH Point of Zero Charge (pHPZC) (B) Effect of pH Fig.10 Illustration of interaction between thiol functionalized-GO, after adsorption of Hg2+,

CH3Hg+, and CH3CH2Hg+ Fig. 11 Schematic representation to the adsorption of inorganic Hg(II) from fish tissue sample

37   

Figure

Fig. 1

Fig. 1

Fig. 1

Fig. 1

Fig. 1

Fig. 2

Fig. 3

Fig. 3

Fig. 4

A

B

C

D

Fig. 5

Fig. 6

Fig. 6

Graphite

Fig. 7

GO

Fig. 8

Element

Weight%

Atomic%

CK

79.88

85.40

OK

16.26

13.05

SK

3.86

1.54

Totals

100.00

Element

Weight%

Atomic%

CK

26.12

33.49

NK

4.99

5.48

OK

62.92

60.57

Hg M

5.98

0.46

Totals

100.00

Fig. 9

Fig. 9

Fig. 10

Fig. 11

Table

Table 1. Application of fish tissue sample

S.

Amount of Hg(II) to fish

No

tissue sample before

fish tissue sample after

adsorption (µg kg−1)

adsorption (µg kg−1)

1

19.1

SD

RSD

0.06 0.32

Amount of Hg(II) to

18.6

SD = Standard deviation RSD = Relative standard deviation

1

SD

RSD

% Removal efficiency

0.04

8.0

97.38

Table 2. Recovery of Hg(II) from real samples

S. No

Water samples

Added Hg(II) in (µg L-1)

1

2

3

a

River watera

Pond waterb

Sea waterc

d

Found in

e

(µg L-1) ± gSD

recovery (%)

0

0.45±0.02

2.0

2.49 ±0.08

102.0

5.0

5.53±0.03

101.6

0

0.32±0.06

-

2.0

2.30±0.04

99.0

5.0

5.41±0.02

101.8

0

0.29±0.09

-

2.0

2.32±0.04

101.5

5.0

5.31±0.06

100.4

River water collected from love river, kaohsiung in Taiwan.

kaohsiung Art museum in taiwan. d

c

Relative

-

b

pond water collected from

sea water collected from sizihwan bay in Taiwan.

for three replicates (n = 3). e determined by ICP-MS. gSD= Standard deviation

2

Table 3. Comparison of adsorption capacity against similar adsorbents

SI. No

Adsorbent material

Adsorption capacity (mg g-1)

1

Modification of activated carbon by 2,6-diaminopyridine

377.68

2

Glycidyl methacrylate grafted cellulose

37.03

3

TAA immobilized silica gel (TAA@FSG)

17.5

4

SBA-15 mesoporous materials modified with aminopropyl

7.59

5

Thiol modified Fe3O4@SiO2 magnetic sorbent

148.8

6

Cerium functionalized PVA–chitosan composite

31.44

7

RGO–MnO2

9.0

8

Graphene oxide–iron magnetic nanoparticles

16.6

9

L -cystine functionalized EGO

79.36

10

Thiol-functionalized mesoporous silica-coated nanoparticles MBT-functionalized graphene oxide

11

3

magnetite 111.93 107.52

References

1 4 5 6 7 8 16 17 23 63 Present work