Removal of mercury ions from aqueous solution by thiourea-functionalized magnetic biosorbent: Preparation and mechanism study

Removal of mercury ions from aqueous solution by thiourea-functionalized magnetic biosorbent: Preparation and mechanism study

Accepted Manuscript Removal of mercury ions from aqueous solution by thiourea-functionalized Magnetic Biosorbent: Preparation and mechanism study Jian...

1MB Sizes 0 Downloads 27 Views

Accepted Manuscript Removal of mercury ions from aqueous solution by thiourea-functionalized Magnetic Biosorbent: Preparation and mechanism study Jianjun Zhou, Yaochi Liu, Xiaohui Zhou, Jialin Ren, Chubin Zhong PII: DOI: Reference:

S0021-9797(17)30877-9 http://dx.doi.org/10.1016/j.jcis.2017.07.110 YJCIS 22639

To appear in:

Journal of Colloid and Interface Science

Received Date: Revised Date: Accepted Date:

12 April 2017 25 July 2017 28 July 2017

Please cite this article as: J. Zhou, Y. Liu, X. Zhou, J. Ren, C. Zhong, Removal of mercury ions from aqueous solution by thiourea-functionalized Magnetic Biosorbent: Preparation and mechanism study, Journal of Colloid and Interface Science (2017), doi: http://dx.doi.org/10.1016/j.jcis.2017.07.110

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.

Removal of mercury ions from aqueous solution by thiourea-functionalized Magnetic Biosorbent: Preparation and mechanism study Jianjun Zhou, Yaochi Liu*, Xiaohui Zhou, Jialin Ren and Chubin Zhong

College of Chemistry and Chemical Engineering, Central South University, 410083, Changsha, China *Corresponding author: E-mail address: [email protected] Tel: +86-136-1748-6029 Abstract: A novel magnetic bio-material (MCIT) was synthesized via coupling reaction and functional modification after load of Fe3O4 nano-particle on the puckered surface of cyclosorus interruptus (CI). The synthesized material was characterized by fourier transform infrared (FTIR), field emission scanning electron microscope (FE-SEM), X-ray photoelectron spectroscopy (XPS) and X-ray powder diffractometer (XRD). The influence factors like pH, temperatures, contact time, initial concentration and cycle times on the adsorption of Hg (II) in aqueous solution were studied. Adsorption isotherm, kinetics, selectivity and mechanism were investigated. The results indicated that the isotherm model well agreed with monolayer adsorption model. The adsorption process could be divided into three steps, which included a fast step controlled by chemical adsorption, a slow step limited by intraparticle diffusion and an equilibrium stage. The maximum adsorption capacity of Hg (II) was 385.3mg/g at 318K. MCIT possessed high reusability (retained 93% after five successive cycles) and sharply magnetic nature (9.5emu/g), which endowed it easy and efficient separation from aqueous solution. Keywords: Bio-based adsorbent; Magnetic adsorption; Separation; mercury ions.

1. Introduction The presence of contaminants like heavy metals and organic compounds in effluents has become a worldwide problem due to their terrible impacts on human health and environment [1-2]. Hg (II) discharged from industrial-related sources is one of the most hazardous contaminants, which acquires high toxicity even at low concentrations [3]. In order to curb the impacts of heavy metals, wastewater needs to be treated before being discharged to the environment. These traditional treatments like reduction, extraction, precipitation and membrane separation are utilized to remove the mercury compounds [4-5]. However, these treatments expose many problems including high operation cost and low efficiency. Adsorption is one of the most effective and low-cost treatments for removing pollutants from aqueous solution [6]. Adsorption process could achieve via an adsorbent combining pollutants with physical and chemical attractive forces [7]. With increasing emphasis on low-cost and sustainable chemistry, the development of green adsorbents has become a new research trend direction in the past few decades [8]. The bio-based adsorbent is a promising candidate for the removal of pollutant from wastewater, because it exhibits excellent adsorption property and derives from renewable resources [9-10]. In our previous study, a novel bio-based material _ cyclosorus interruptus (CI) with large lignin, cellulose and hemicellulose, a fern plant growing in 1

all around the world, was considered as an adsorbent for the adsorption of Pb2+ and U6+ from the aqueous solution [11]. The experimental results indicated that the bio-material with abundant activity groups could be feasibly modified by coupling reaction to form potential adsorption sites for heavy metal ion. Additionally, this fern plant has abundant porous structure and puckered surface, suggesting that the material can be considered as an ideal carrier to nano-particles like metals and metals oxide. However, the bio-material presents drawback of low separation efficiency, which has restricted its further practical application in removal of contaminants [12-13]. It is well known that Fe3O4 nano-particles possesses rapid responsibility during separation process, which has been obtained considerable attention in the treatment of contaminated solution due to its high surface area, large adsorption capacity and magnetic property [14]. However, bare Fe3O4 nano-particles are easily oxidized in air and prone to aggregation in the aqueous solution. As a result, the magnetic property and adsorption ability would rapidly reduce [15]. Thus, surface encapsulation and modification of Fe3O4 nano-particles is required to prevent loss of mass. For instance, amino-silane was used to stabilize the Fe3O4 nano-particles and inhibit their aggregation, achieving rapid and efficient removal of heavy metal [16]. Reduced graphite oxide was considered as an ideal carrier to stabilize gold and Fe3O4 nano-particles, exhibiting efficient adsorption of mercury ions from aqueous solution. Besides, cellulose serving as a scaffold to support nano-particles has been reported [17]. Compared with filtration, flocculation and centrifugation, this bio-based magnetic adsorbent does not require any additional process for solid-liquid separation, which provides a new potential candidate for low-cost and high efficient removal of heavy metals from aqueous solution. Moreover, the adsorption capacity and selectivity will be greatly improved via the introduction of functional groups. The thiol and amine groups, which can strongly bond with mercury ions, were grafted on the silica to increase adsorption sites [18]. It has been reported that the thoiurea grafted on adsorbent could form a strong affinity with mercury ions based on the theory of hard and soft acid and bases of Pearson [4]. The purpose of this work was to prepare a novel bio-based magnetic cyclosorus interruptus material via coupling reaction and other functional modification. CI was considered as a support to load Fe3O4 nano-particles on the surface, because it offers abundant multi-porous structure and puckered surface as well as active groups. Then the composites were coated a layer of 3-Triethoxysilylpropylamine on the surface, which could improve the stability of Fe3 O4 nano-particle on the surface of CI and increase adsorption sites. To further increase adsorption sites and selectivity for mercury ions, the precursor will be modified with thiourea (TU). The magnetic biosorbent (MCIT) was characterized by FT-IR, SEM, XPS, VSM and XRD. The equilibrium isotherm and kinetic were studied to evaluate removal performance of mercury ions from aqueous solution.

2. Material and Methods 2.1. Material The cyclosorus interruptus (CI) was collected from YueLu maintain in Changsha, China. FeSO4·7H2O and FeCl3·6H2O were purchased from Sinopharm Chemical reagent Company (Shanghai, China). Thiourea (TU), Epichlorohydrin (EC), 3-Triethoxysilylpropylamine (SA) and other chemical reagents were provided by Huahong reagent Company (Hunan, China). All 2

reagents were of analytical-reagent grade. Distilled water was used throughout all the experiments.

2.2. Synthesis of Fe3O4 nanoparticles Fe3O4 nanoparticles were prepared by coprecipitation method [19]. 4.17g of FeSO4 ·7H2 O and 8.1g of FeCl3·6H2O were dissolved into 150mL of distilled water in a flask. Under the protection of nitrogen, the mixture was stirred for 30 min, then NH3· H2O solution (28%, v/v) was gradually added into flask at 353K, until the pH value was around 10.0. This suspension was vigorously stirred for 5h. The black precipitate was collected by magnet and then washed with distilled water three times, dried at 323K for 5h and stored in the polyethylene bag.

2.3. Preparation of MCIT The collected materials were washed with distilled water several times to remove all dirt particles and dried in oven at 323K for 24h. Next, the dried cyclosorus interruptus were passed a 40-mesh sieve (0.38mm) after grind. Finally, they were stored in polyethylene bag. As shown in Scheme 1, the magnetic bio-adsorbent was synthesized as follows: CI (1g) and Fe3 O4 nano-particles (0.5g) were mixed in 100 ml distilled water at 323K for 2h, then gradually adding SA (5mL) into flask. The magnetic bio-material, named as MCI, was washed with alcohol and distilled water several times after separation using magnet. A medium of 50/50 v/v ethanol/water was prepared, and the prepared particles of MCI and EC (5mL) were added into the medium. Then, the mixture was stirred for 12h at 303K under nitrogen protection. Next, TU (3g) and 1 mol/L NaOH (50mL) were added into the flask at 333K for 12h [18]. After all reactions finished, the particles were washed with ethanol and distilled water after separation by magnet. The materials were dried at 323K for 5 h in the oven, and defined as MCIT. (1) Magnetization and modification of CI particles O

C2H5O

+

C2H5O

Si

CH 2CH2CH2NH 2

Fe3O4



323K 8h

C2H5O

Si

CH2CH2CH2NH2

Si

CH2CH2CH 2NH2

Si

CH 2CH2CH2NH2

O

O

O

CI

SA

MCI

(2) Surface functional modification of MCI particles NH2

+

H2 C

303K

CHCH2Cl

EC S

+

CH CH2CH2Cl

PMCI

CH2CH2Cl

PMCI

OH NH

CH

12h

O

MCI

OH H N

NH2

C

333K NH2

H N

OH

S

CH2 CH2NH

C

CH

12h

NH2

MCIT

TU

Scheme1. Synthetic produce of the bio-based magnetic adsorbent of MCIT.

2.4. Characterization methods and instruments Fourier-transform infrared (FTIR) spectra were recorded on an FTIR spectrophotometer (Thermo Nicolet Nexus Spectrometer), using the KBr disk method. The XRD characterization of CI and MCIT was carried out by an X-ray diffraction diffractometer (Rigaku-TIR III, Japan) with 3

the Co Kα radiation (y=0.15406nm) at 40kV and 30mA and the region of 2θ from 10 to 80o. The surface morphology of CI and MCIT was investigated by field emission scanning electron microscope (SEM, JSM-6700F, Japan). The magnetic property was recorded with vibrating sample magnetometry (VSM, 7410, U.S.A.). The elements of adsorbents were analyzed using X-ray photoelectron spectroscopy (XPS) with a monochromatic Al Kα X-ray source (1486.71eV). The zeta potential of MCIT was measured at different pH with a JS94H (Shanghai, China).

2.5. Batch adsorption experiments The adsorption experiments were carried out in 100ml conical flasks, each of which contained 30ml mercury ions solution. Magnetic bio-adsorbent of 30mg was added into each conical flask and shaken at 160 rpm in thermostatic shaker. 0.1mol/L HNO3 and 0.1 mol/L NaOH solutions were utilized to adjust the pH of solution. The concentration of Hg (II) in the solution was analyzed by ultraviolet spectrophotometer (UV759s) based on colorimetric method of National Standard of China (GB/T 5750.6-2006). The adsorption capacity of heavy metals onto adsorbents was calculated using the following equation:

Q e = (Co - C e ) ×

V m

(1)

where Qe (mg/g) is the adsorption capacity of Hg(II), Co and Ce (mg/L) are the initial and equilibrium concentrations of Hg(II), respectively, m (g) is the weight of the adsorbents, and V (L) is the volume of adsorption solution. The effect of pH on adsorption ability of MCIT was studied at pH in the range of 2.5-7.0 (Co=100mg/L, dosage=1g/L, volume=30mL, reaction time=12h and temperature=303K). In the kinetic experiments, the CI, MCI and MCIT were also studied with contact time. The effect of temperature in range of 303-318K on the kinetic of MCIT was carried out (Co=100mg/L, dosage=1g/L, volume=30mL and pH=5.0). Additionally, the adsorption isotherm studies were investigated with initial Hg(II) concentrations in range of 50-1000mg/L at different temperature (dosage=1g/L, volume=30mL, reaction time=12h and pH =5.0).

2.6. Reusability studies of MCIT To evaluate the reusability of the MCIT, the desorption experiment of Hg (II) was carried out using mixture solutions containing 0.5 mol/L thiourea and saturated ethylene diamine tetraacetic acid. The metal ions loaded MCIT adsorbents were added into the 100mL mixture solution and shaken at 303K for 12h. Then the adsorbents were separated by external magnet, washed with distilled water several times and investigated to further cycle of adsorption process (Co=100mg/L, temperature=303K, and reaction time=12h). Meantime, the adsorption capacity and times were recorded.

3. Results and Discussion 3.1. Characterizations of MCIT Fig.1a shows the FTIR spectra of CI, MCI and MCIT. There was a strong peak at 580 cm-1 in the MCI and MCIT owing to the Fe-O vibrations of Fe3O4, which exhibited that the Fe3 O4 nano-particles were successfully grafted on the puckered surface of CI particles [7,14]. The characteristic band of C=S appeared in the MCIT at 1140 and 1319cm-1 [19], and the peaks at 1634, 1038 and 1509cm-1 responding to the phenyl were enhanced, indicating that the thiourea was successfully grafted on the surface of MCI [20]. The bands which appeared at 2916 cm-1 4

corresponded to -CH2 vibrations. Meantime, the broad peak at 3349 cm-1 of MCIT corresponding to O-H bond stretching of the CI shifted to 3492 cm-1, which suggested that there were a strong hydrogen band interaction between the –OH and –NH2 [18].

Transmittance(%)

a

b

CI MCI MCIT

O1S

C1 S

N 1S

1 31 9 1 50 9 11 40 16 34 10 38

2 9 16 334 9

58 0

4000 3500 3000 2500 2000 1500 1000

500

-1 Wavenumber(cm )

0

200

d

c

O1

400

Si 2 P

s

S2

F e2 P S

s Fe2 p 1 /2

Fe3 +sa t. 2 p 3 /2 7 15

72 0

C-N 285.39

7 25

Bin d ing En ergy ( E) ( ev)

40 0

6 00

80 0

100 0

1 200

282

B inding E nergy (E ) (ev)

e

C=O 288.11

C-O 286.51

C=C/C-H 284.77

Fe2 p 3 /2

P

2 00

1200

S

710

0

1000

C-O 286.57

C-N 285.65 N1

Si 2

800

C=C/C-H 284.84

C1 S

S2

600

B inding E nergy (E ) (ev)

284

286

C=O 288.09 C=S 289.17 288

290

Binding Energy (E) (ev)

N=C 400.08

f

O-H 532.84

O=C 531.47 NH/NH2 399.53

394

396

398

400

402

O-H 532.34

NH3+ 401.62

404

O=C 530.51

406

408 528

Binding Energy (E) (ev)

530

O-Si

533.31

532

534

536

Binding Energy (E) (ev)

Fig. 1. (a) FTIR spectrums of CI, MCI and MCIT; XPS spectra of CI (b) and MCIT (c); C 1s (d), N 1s (e) and O 1s (f) narrow XPS scan for CI and MCIT.

The surface chemical properties of CI and MCIT were analyzed using the X-ray photoelectron spectroscopy. As shown in Fig. 1b and 1c, there were not distinct differences between the peaks of C1S and O1S in all samples. Compared to the raw CI, the strong peaks at 105.1eV and 166.2eV appeared due to the binding energies of O-Si-O and C=S, respectively [4]. Meantime, the new peak at 710.9eV was attributed to the formation of Fe-O, which illustrated the magnetic nano-particles were successfully loaded on the surface of CI. It is well know that the XPS analysis method describes the oxidation state of the composites. The Fe2P peaks at range of 711.4 and 724.8eV correspond to the Fe2P3/2 and Fe2P1/2 spin-orbit peak of magnetic compounds. The weak peak at 719.2eV associated with Fe satellite peak was an evidence to further indicate that the Fe element existed in the MCIT [21]. Additionally, Fig. 1d-f showed the high resolution C1s, N1s and O1s spectra for CI and MCIT. Compared to the CI, the intensity of C-N peak in MCIT significantly increased, and the new peaks of C=S, NH/NH2 and O-Si appeared at 289.17, 399.53 5

and 533.31eV,respectively, which indicated that the TU and SA reagents were successfully grafted on the surface of CI. Moreover, the XPS analysis results of the bio-adsorbents on selected elements were listed in Table 1. The results exhibited that the S and N atoms highly increased after functional modification, which also suggested the objective product of magnetic bio-material was successfully prepared. The SEM images of the CI and MCIT were presented in Fig. 2a and 2b. Abundant three-dimensional porous structure and zigzag morphology appeared in the interior and exterior of raw CI and MCIT particles, indicating that this bio-material possessed high surface area and provides an ideal carrier to Fe3O4 nano-particles. Considering the like-gully and multi-porous structure of MCIT, the dissolved mercury ions can lightly permeate into the interior, which could enhance the adsorption ability. Compared to CI, there was no obvious change after modification, which illustrated that Fe3O4 nano-particles and other functional modification could hardly influence the morphology. Additionally, the average pore diameter of CI and MCIT reached to around 40µm. The X-ray diffraction patterns of CI and MCIT were showed in Fig. 2c. The strong peaks of the CI at 2θ=15.7 and 22.6 corresponding to the (110) and (200) planes were the characteristic peaks of cellulose or lignin crystal, which indicated that the CI consisted of cellulose and lignin [4]. Compared to the CI crystal peak, the new diffraction peaks of Fe3O4 (2θ=30.3o, 35.7 o, 43.4 o, 57.5 o and 63.2 o) were regarded as the (111), (220), (400), (511) and (440) planes of Fe3O4, and the diffraction peaks at 2θ =15.7 and 22.6o became weaker, which suggested that the Fe3 O4 nano-particles had a strong interaction with the functional groups in MCIT [20]. As shown in Fig. 2d, the curves increased sharply with the increase of magnetic field, and the saturation magnetization (Ms) value reached to 9.5emu/g for the MCIT. Obviously, the characteristic of superparamagnetic behavior was caused by the modification with Fe3 O4 nano-particles, which could dramatically improve the separation efficiency from aqueous solution.

10

c

M CIT CI

d

Fe(220) Fe(311) Fe(511)

5

Fe(440)

M(emu/g)

Intensity(a.u.)

Fe(400) CI(200)

CI(110)

0

-5

10

20

30

40

50

2Theta(degree)

60

70

-10 -10000

-5000

0

5000

10000

H(Oe)

Fig. 2. The SEM images of the CI (a) and MCIT (b). (c) The XRD spectrums of CI and MCIT. (d) The magnetization saturation curves for MCIT.

6

Table 1 XPS Analysis data of the bio-adsorbents on selected elements (wt%). Adsorbents

Elements (wt%) C

O

N

Fe

Si

S

CI

64.61

32.56

2.12

0

0.71

0

MCIT

52.48

24.37

4.84

5.81

6.36

6.14

3.2 Effect of pH on the adsorption The pH is an important factor for adsorption of mental ions in aqueous solution, because the chemical properties of metal ions and adsorption sites could vary with pH values in aqueous solution. Thus, the effect of pH on the adsorption ability of MCIT for Hg(II) was studied in range of 2.5-7.0, and the experimental results were displayed in Fig. 3b. It can be obviously found that the adsorption ability of MCIT increased with the increase of initial pH values and then reached to a plateau. The zeta-potential of MCIT was obtained at different pH. As shown in Fig. 3a, the pH value of pHpzc (point of zero charge) of the MCIT was about 4.2. The surface adsorption sites of MCIT showed positive charge at pH range of 2.5-4.2. At low pH, the most of functional groups like amine and thiol were protonated and the competition between hydrogen ions and mercury ions occurred in aqueous solution, which caused the electrostatic repulsion between the adsorbates and the positively charged functional groups. This condition may prevent the adsorption of mercury ions onto the surface of MCIT [22]. With pH increasing, the surface charge of adsorbent become negative, more adsorption sites were available, which brought electrostatic attraction between surface sites and adsorbates. Thus, the adsorption ability of the MCIT for Hg(II) increased. The variation of solution pH after the adsorption of Hg(II) was also obtained. At low pH, a slight increase of the final pH was observed. While above pH >4.5, the final pH tended to slightly reduced. This result also illustrated that the adsorption sites were prone to protonation at lower pH. With pH increasing, the surface ion exchange occurred during adsorption process. Moreover, according to acid-base theory, the thiourea and amine groups acting as the Lewis soft alkali had a strongly force with mercury ions acting as Lewis soft acid [23]. When the solution pH >7, the Hg(II) could combine with OH- to form precipitation in the aqueous solution [24]. Thus, the experiments of Hg(II) adsorption in alkaline solution were not investigated. a

20

10

0

-10

-20

-30 2

3

4

5

6

7

b

100

7

80

6

60

5

40

4

20

3

0

Final pH

Adsorption capacity (mg/g)

Zeta potential (mV)

30

2 2

3

4

5

6

7

pH Initial pH Fig. 3. (a) Zeta potential of MCIT at different pH; (b) The variations with respect to final pH and effect of pH on the adsorption of Hg (II) onto MCIT (T=303K; adsorbent dosage=1.0g/L; initial Hg(II) concentration = 100 mg/L; contact time = 12 h).

3.3 Adsorption kinetic studies 7

The effect of contact time on the removal ability of different adsorbents was performed (Co=100mg/L, pH=5.0). The experimental results were presented in Fig. 4. It can be easily seen that the adsorption ability of MCIT increased rapidly within the initial 100min, because more adsorption sites were available, which easily interacted with Hg(II). After 200min the adsorption capacity reached to saturation due to the free adsorption sites were occupied. Additionally, the adsorption ability of MCIT increased with the increase of temperature, which illustrated that the adsorption ability can be improved by adjusting the temperature. In order to evaluate the contributions of the 3-Triethoxysilylpropylamine and thiourea in Hg(II) removal, the effect of different biosorbents on adsorption performance was investigated. As shown in Fig. 4b, the experimental data exhibited that the adsorption ability of CI exhibited remarkable adsorption performance. Compared to CI, the adsorption ability of MCI increased from 51.9 to 61.3mg/g, and the MCIT reached to 82.8mg/g, which illustrated that the prepared magnetic biosorbent obviously improved the adsorption ability of Hg(II) after modification. Table 2 Pseudo-second-order kinetic parameters of Hg(II) removal onto CI, MCI and MCIT. Qexp Adsorbents

(mg/g)

Pseudo-second-order Qcal (mg/g)

k2 (min· mg/g)

2

R

CI(303K)

51.9

56.2

0.000402

0.993

MCI(303K)

61.3

64.9

0.000560

0.998

MCIT(303K)

82.8

86.2

0.000704

0.998

MCIT(308K)

68.6

87.4

0.000857

0.999

MCIT(313K)

71.5

89.4

0.000999

0.997

MCIT(318K)

74.4

93.1

0.001160

0.998

Additionally, in order to investigate the adsorption mechanism, the surface complexation model was established to illustrate the adsorption mechanism [25]. It was assumed that the adsorption process is homogeneous adsorption on the active sites, namely, this adsorption is similar to a homogeneous phase chemical equilibrium. The formation of metal complexation was defined as following:

M + S ⇔ MS (2) where M (mg/L) is the free metal ions, S (g/L) represents the free surface adsorption sites, and MS is the formation of complexation. Additionally, it can be deduced that the surface complexation model could be described as the second-order model according to the chemical reaction kinetic model [26]. Thus, the pseudo-second-order model was utilized to analyze the adsorption kinetic. The equation of the pseudo-second-order model was given as following [27-29]: t 1 t = + 2 Qt k 2 Qe Qe

(3)

where Qe and Qt (mg/g) are the adsorption capacity of mercury ions at equilibrium and at time t (min), k2 (g/mg· min) is the rate constant of second-order model. The fitting curve of t/Qt versus t was presented in Fig 5a. The related adsorption kinetic model parameters were listed in Table 2. It can be observed that the value of correlation coefficient (R2>0.993) of the second-order for MCIT was high, which indicated that the 8

second-order model fitted well with the experimental results. In addition, the maximum adsorption capacity (Qe(cal)) obtained from the second-order model was well in agreement with the experimental result (Qe(exp)). In short, the results illustrated the adsorption of Hg(II) on the MCIT was a chemical adsorption. According to the chemical adsorption theory, it was assumed that the adsorption ability is proportional to the content of adsorption sites. The order of adsorption ability and rate was MCIT>MCI>CI, which further confirmed that the adsorption ability of Hg(II) onto these adsorbents was limited by active sites (chemical adsorption). Thus, the adsorption rate and ability have a close relationship with the type and amount of adsorption sites on the surface of adsorbents. Adsorption capacity (mg/g )

a

80 70

303K 308K 313K 318K

60 50 40 30

Adsorption capacity (mg/g )

90 90

b

80 70 60 50

MCIT MCI CI

40 30 20 10

20 0

100

20 0

30 0

4 00

0

10 0

2 00

300

400

500

T ime (min)

Tim e (mi n)

Fig. 4. (a) The adsorption kinetics plot of MCIT toward mercury ions at different temperatures. (b) The adsorption kinetics plot of CI, MCI and MCIT at 303K, pH=5.0 and dosage=1.0g/L.

Table 3 Webber’s pore-diffusion model kinetic parameters of Hg(II) removal onto CI, MCI and MCIT. Fast stage

Slow stage

Adsorbents

k1d (min )

R

k2d (min· mg/g)

R2

CI(303K)

2.78

0.988

1.47

0.991

MCI(303K)

4.38

0.984

1.59

0.998

MCIT(303K)

7.98

0.982

2.60

0.990

MCIT(308K)

7.49

0.989

2.31

0.991

MCIT(313K)

6.57

0.987

2.08

0.991

MCIT(318K)

7.47

0.990

1.77

0.989

-1

2

It is notable that the above kinetic adsorption model derived from chemical reaction are based on the whole reaction process without considering the adsorbates diffusion process. Hence, the Webber’s pore-diffusion model originated from Fick’s second law of diffusion was carried out. The equation was defined by:

Q t = k id t 0.5

(4)

where Kid is the Webber’s pore-diffusion model constant. Based on pore-diffusion theory, the adsorption process can be divided into three steps, namely a fast step involving external diffusion, followed by a slow step limited by intraparticle diffusion, finally, an equilibrium state. It can be observed from the equation of Webber’s pore-diffusion that if diffusion was the controlling step, then intercept of the fitting curves should be zero. As shown in Fig. 5,it can be found that the intercept values calculated from the fitting results of first segment were clearly not zero. The intercept values exhibited that the diffusion step did not limit the whole adsorption rate at early 9

stage. Additionally, the kinetic results of different adsorbents with various amount adsorption sites further indicated that the controlling rate of first stage was a chemical reaction (Fig. 5c). It could be found from Table 3 that the diffusion parameter (K2d) of MCIT increased obviously with increase of adsorption sites. This could be the influence of the active sites (amino and thiol groups), which effectively increased the chemical adsorption rate and capacity, and thus accelerated the diffusion rate during the slow stage. Additionally, the diffusion parameter (K2d ) of MCIT increased with temperature which was ascribed to the influence of adsorption capacity at fast step. Therefore, the second stage was a gradual adsorption step where diffusion was controlling rate. In this stage, the kinetic behavior was similar to diffusion process (Fig. 5d). The third stage was equilibrium process (Fig. 5e).

a

4

t/qt

b

90

3

303K 308K 313K 318K

2

1

Adsorption capacity (mg/g )

5

80 70 60

303K 308K 313K 318K

50 40 30 20

0 0

100

200

300

400

0

5

c

10

15

20

t0.5(min0.5 )

Time (min)

d

Chemical reaction stage

e

Diffusion stage

Equilibrium stage

Fig. 5. (a) Pseudo-second-order curve of Hg (II) adsorption; (b) Intraparticle diffusion model curve of Hg (II) adsorption; (c-e) Schematics of kinetic process.

3.4. Adsorption Isotherm Adsorption isotherm that describes the interactive behavior between heavy metal and adsorbent could provide important information for optimizing the practical application of the adsorbent. The adsorption isotherms of mercury ions were presented in Fig. 6a. It was clear that the adsorption capacity increased with increase of mercury ions concentration and temperature, then reached to a plateau. Moreover, the equilibrium Hg(II) adsorption of MCIT were 261.8, 386.6, 321.5 and 364.9mg/g at 303, 308,313 and 323K, respectively. The reasons for these phenomena can be regarded as the increase of interaction chance. As shown in Fig. 6a, this trend of adsorption capacity versus equilibrium concentration was fitted with an L-type isotherms model. And the capacity of MCIT had a saturation status at high concentration. According to above discussion, the adsorption capacity could be improved after enhancing adsorption sites on the surface of adsorbent, which consisted with assumption of 10

Langmuir model. Thus, the Langmuir isotherms model can suitably described this experimental result. It was assumed that the adsorption reaction occurred on a homogeneous surface by monolayer adsorption at certain temperature and there was no interaction between the adsorbates. Moreover, according to the complexing formation (Equation (2)), it can be deduced that the conditional formation constant was expressed as following:

Ko =

[ MS] [ M ][S]

(5)

Based on mass-balance theory, the related mass equation was defined as following:

ST = S+

[ MS] [ Nm ]

(6)

where ST(g/L) represents the total mass of adsorbent, and Nm(mol/g) is the number of complexing sites that determines the maximum adsorption capacity at certain temperature. According to the Equations (5) and (6), the equation can be obtained:

Qe =

Ko Nm [M ]

(7)

N m + K o [M]

The equation of Langmuir was given by [30]:

Qe =

k L Qm C e 1+k L C e

(8)

where Qe(mg/g), Ce(mg/L) are the adsorption equilibrium capacity and the equilibrium concentration of Hg(II), respectively. kL(L/mg) is Langmuir adsorption constant and Qm (mg/g) is the theoretical equilibrium capacity. It can be observed that Equation (7) was identical to Langmuir model equation (Equation (8)). The modeling results further indicated that the adsorption process was a homogeneous adsorption. 350

a

b

3.0 2.5

300

2.0

250

303K 308K 313K 318K

200 150 100

Ce /Qe

Adsorption capacity (m g/g)

400

1.5

303K 308K 313K 318K

1.0 0.5

50 0.0 0 0

100

200

300

400

500

600

700

800

0

100

200

300

400

500

600

700

800

C e(mg/L)

C e(m g/L)

Fig. 6. (a) Adsorption isotherm for mercury ions adsorption onto the MCIT; (b) Linear plot of Langmuir isotherm.

The values of the parameter and coefficient of correlation (R2) for Langmuir were listed in Table 4. The linear curve of Langmuir model was presented in Fig.6b. It can be observed that the theoretical adsorption data were well fitted with the Langmuir model (R2=0.997, R2=0.998, R2=0.996 and R2=0.993 at 303,308,313 and 318K, respectively). Additionally, the saturation adsorption capacities of MCIT toward Hg(II) were closed to the maximum adsorption abilities (290.8, 314.6, 351.1, and 385.3mg/g at 303K, 308K, 313K and 318K, respectively) calculated from Langmuir model. It suggested that the adsorption well obeys the Langmuir isotherms model, 11

which indicated that the adsorption process was homogeneous on the uniform surface. Separation factor of RL is a dimensionless constant expressed as [31]:

RL =

1 1+k L Co

(6)

where Co (mg/L) represents the initial metal concentration. The values of R L classified the adsorption to be irreversible (RL=0), favorable (0< RL<1) and unfavorable (RL >1). The values of RL for MCIT were calculated to be 0.072-0.61,0.057-0.55,0.045-0.48 and 0.033-0.41 at 303,308,313 and 318K, respectively with initial concentration ranging from 50 to 1000mg/L, which indicated that the adsorption process was favorable as 0< RL<1.

Table 4 Isotherm parameters for the adsorption of Hg (II) on MCIT at different temperatures. T (K)

Langmuir isotherm model R2

Q m (mg/g)

kL(L/mg)

RL

303

290.8

0.0129

0.072-0.61

0.997

308

314.6

0.0165

0.057-0.55

0.998

313

351.1

0.0215

0.045-0.48

0.996

318

385.3

0.0291

0.033-0.41

0.993

3.5. Thermodynamic Studies According to above discussion, it can be deduced that the equilibrium isotherms was a homogeneous adsorption. Thus, the conditional formation constant (Ko) can be utilized to calculate the thermodynamic parameter. The equation of Van’t Hoff was given as following [32-33]:

ln (K o )=

∆So ∆H o R RT

(11)

The ∆G0 was defined as [32]:

∆G o = ∆H o - T∆So

(12)

where R value is 8.314J/(mol·K). ∆ Go, ∆Ho and ∆So represent the standard free energy,, enthalpy change and entropy change, respectively. -6.7

4.8

4.7

a

4.6

-6.9

4.5

lnk2

ln(K o)

b

-6.8

-7.0

4.4

-7.1 4.3

-7.2 4.2 0.00316

0.00320

0.00324

0.00328

0.00332

-7.3

1/T(1/K)

0.00316

0.00320

0.00324

1/T(1/K)

0.00328

0.00332

Fig. 7. (a) The fitting of thermodynamic parameter; (b) The fitting of Arrhenius parameter at different temperatures.

12

The fitting curve was presented in Fig. 7a, which provides high correlation coefficients (R >0.993), and thermodynamic parameters were listed in Table 5. In all cases, the negative values of ∆Go indicated that the adsorption of mercury ions on the surface of MCIT was favorable and 2

spontaneous. In addition, the values of ∆ Go increased with increase of temperature, suggesting that the lower temperature was in favor of the adsorption process [33]. The negative values of ∆Ho confirmed that the Hg(II) adsorption on the MCIT was an exothermic process. Meanwhile, ∆ So <0 exhibited that Hg(II) adsorption on MCIT from aqueous solution was a entropy decreased process. The result demonstrated that the adsorption of Hg(II) on MCIT was a spontaneous and exothermic process.

Table 5 Thermodynamic parameters for the adsorption of Hg (II)(100mg/L) on MCIT. Temperature (K)

∆ G0 (kJ/mol)

∆H0 (kJ/mol)

∆S0 (J/mol. K)

303

-12.2

-28.02

-53.2

308

-11.8

313

-11.3

318

-10.7

3.6. Activation Energy The equation of Arrhenius was expressed the activation energy as following [33]:

ln k 2 =

-E a + constant RT

(9)

where k2 (g/(mg · min)) is the pseudo-second-order rate constant, and R is equal to 8.314 J/(mol·K). The k2 values were calculated from the slope of t/Qt versus t at 303, 308, 313K and 318K. Then, the activation energy was obtained by the Arrhenius plot of ln k2 versus 1/T for the adsorption of Hg(II) in the aqueous solution. The fitting curve was presented in Fig. 7b, and the activation energy value was equal to 26.52kJ/mol. This value of activation energy was in range of 5-40kJ/mol, which suggested the adsorption system mainly involved chemical adsorption [32]. Therefore, the adsorption process of mercury ions onto magnetic bio-adsorbent could include chemical adsorption.

Table 6 Comparison of mercury adsorption capacity for different adsorbents. Adsorbent

Metal capacity (mg/g)

Refs.

Ammonium-magnetic meso-porous silia

20.5

[34]

MNPs-polyAEMA-DTC

59.5

[34]

Thiol-functionalied-zeolite

89.3

[36]

HO-SG-GPTs-Ts

148.4

[37]

hybrid ZnCl2 -MCM-41

156.3

[38]

220.5

[39]

385.3

This study

Meso-porous silica-SBA-15 Bio-based magnetic adsorbent (MCIT)

3.7. Reusability of the Bio-based Adsorbent The reusability of the adsorbent was an important parameter, which determined whether it 13

could be feasibly applied in the treatment of wastewater or not. As shown in Fig. 8a, the adsorption abilities exhibited slight fluctuation during five successive adsorption/desorption cycles, and the adsorption capacity maintained about 93%. The result indicated that the bio-based magnetic adsorbent of MCIT exhibited a remarkable reusability, which can be used as a promising candidate for the removal of heavy metal from wastewater.

a

90

80

Adsorption capacity ( mg/g)

Adsorption capacity ( mg/g)

90

70 60 50 40 30 20 10 0

80

b

70 60 50 40 30 20 10 0

0

1

2

3

Numbers of cycles

4

5

Hg

Ni

Pb

Cd

Zn

Heavy metal

Fig. 8. (a) Reusability of the MCIT for the Hg (II) adsorption at 303K, pH=5.0, concentration=100mg/L and dosage=1 g/L; (b) Comparative adsorption of mercury ions and other heavy metals by MCIT.

a

b

399.53

399.90

401.62 401.91

394

396

398

400

402

404

406

408

394

396

Binding Energy (E) (ev)

c

398

400

402

404

406

408

Binding Energy (E) (ev)

d

164.19

164.27

165.47

158

160

162

164

166

168

170

158

160

162

164

166

168

170

Binding Energy (E) (ev)

Binding Energy (E) (ev)

Fig. 9. (a, b) N 1s narrow XPS scan for MCIT before and after adsorption; (c, d) S 2p narrow XPS scan for MCIT before and after adsorption.

3.8. Comparison Adsorption of Heavy Metals by bio-based magnetic adsorbent Solutions separately containing Cd (II), Ni (II), Zn (II), Pb (II) and Hg (II) (each 14

concentration=100mg/L, pH=5) were prepared. 30mg MCIT was added into 30ml of each of these solutions at 303K for 12h. After that, the adsorption capacities of these metals on MCIT were calculated. The experimental results were presented in Fig. 8b. It can be seen that the equilibrium adsorption capacity of Hg (II) was much higher than that of the other metals, and the order was Hg (II)> Pb (II)> Ni (II)> Cd (II)> Zn (II), suggesting that the magnetic bio-adsorbent has an obviously adsorption selectivity to mercury ions. A large number of N and S atoms on the magnetic bio-adsorbent can coordinate with metals in aqueous solution [15]. According to Lewis acid-base theory, mercury ions acting as the Lewis acid have precedence over others metals in interaction with the amine and thiol groups, which can be considered as Lewis base [16]. Moreover, as shown in Fig. 4b, the adsorption capacity of MCIT was much higher than that of other adsorbents at same condition. Thus, this sort of green adsorbent was more feasible and favorable for mercury ions removal from aqueous solution. Additionally, as shown in Table 6, the maximum adsorption capacity of MCIT was much higher than other magnetic adsorbents. Compared to other adsorbent, the MCIT could become a promising candidate for the removal of mercury ions from the aqueous solution, because of its reusability, availability and efficient separation.

3.9. The adsorption mechanism of Hg (II) on MCIT Different adsorption mechanisms could be involved in adsorption system such as electrostatic interaction and complexing reaction [6]. As shown in Fig. 3b, the adsorption capacity was significantly affected by pH when pH <4. This might be attributed to the functional modification of amine and thiol groups on the surface of CI where the amine and thiol groups were protonated [4, 11, 21, 23]. Obviously, the positively charged functional groups of MCIT repulsed the metal ions through electrostatic interaction. Therefore, the electrostatic force was adverse rather than favorable to the adsorption. Additionally, it has been proved that metal ions have strong affinity with electron-rich groups such as amine, thiol and hydroxyl groups via forming coordination with nitrogen, sulfur and oxygen atoms [22-23]. To further provide the evidence of adsorption mechanism, XPS analysis method was applied. The changes in binding energies of N 1s and S 2p on the surface of MCIT were analyzed by XPS before and after Hg (II) adsorption (Fig. 9). The N 1s spectra of MCIT before and after adsorption were presented in Fig. 9a and b. Before adsorption, two major peaks at 399.53 and 401.62 eV could be attributed to the amine (–NH2) and protonated (–NH3+), respectively. After adsorption of Hg (II), the peaks of amine (–NH2 ) and protonated amine (–NH3+) shifted to 399.90 and 401.91 eV, respectively, due to the decrease of electron density of N atoms. The result indicated that Hg (II) was adsorbed in the form of coordination by amine [2-3]. In Fig. 9c and 9d, the binding energies of S 2p appeared not only at 164.27 eV but also at 165.47 eV, which implied that the lone pair of electron in thiol groups was shared with metal ions, leading to the decrease of electron density of sulfur atoms and the increase of binding energy [5-6]. Moreover, the MCIT has large amounts of multi-porous structure and puckered morphology, which increased the interaction area between the adsorbent and the mercury ions [7]. Based to the above discussion, the possible adsorption mechanism of MCIT for Hg (II) was presented by the following complexing formation.

R-NH 2 + Hg(II) ⇔ R-NH 2 -Hg(II)

(10)

R-NH 3 + + Hg(II) ⇔ R-NH 2 -Hg(II) + H +

(11)

15

-C=S + Hg(II) ⇔ -C=S-Hg(II)

(12)

R-OH + Hg(II) ⇔ R-OH-Hg(II)

(13)

4. Conclusion A novel magnetic bio-adsorbent, MCIT, was synthesized via functional modification for adsorption mercury ions. The results of FTIR, XPS and SEM for the MCIT exhibited that the Fe3O4 nanoparticles and thioureas were successfully grafted on the surface of CI. The experimental results exhibited that the adsorption isotherm model of mercury ions on the MCIT closely obeyed to the complexation model, which was identical to Langmuir model. The maximum adsorption capacities of MCIT calculated from Langmuir model were 290.8, 314.6, 351.1 and 385.3mg/g at 303K, 308K, 313K and 318K, respectively, which were higher than that of other adsorbents in the references. The adsorption kinetic process could be divided into three steps, namely, a fast step controlled by chemical adsorption, followed by a slow step limited by intraparticle diffusion, finally, an equilibrium stage. The adsorption reaction was a spontaneous and exdothermic process, and the MCIT exhibited impressive selectivity for adsorption of Hg(II). The adsorption capacity maintained 94% after five successive adsorption/desorption cycles. Additionally, the magnetic adsorbent (reached to 9.5emu/g) was easily and efficiently separated from the solution by magnetic forces. Thus, the sustainable and low-cost magnetic biosorbent MCIT has great potential to be a promising candidate biomaterial for the removal of Hg(II) from aqueous solution.

Acknowledgements We acknowledge the Fundamental Research Funds for the Central Universities of Central South University (No.502211726) and the Hunan Provincial Science and Technology Plan Project, China (No.2016TP1007) for financial support of this research.

Reference [1] S. Luo, X. Xu, G. Zhou, C. Liu, Y. Tang, and Y. Liu, Amino Siloxane Oligomer-Linked Graphene Oxide as an Efficient Adsorbent for Removal of Pb(II) from Wastewater, J. Hazard. Mater. 274 (2014) 145-155. [2] L. Chu, C. Liu, G. Zhou, R. Xu, Y. Tang, Z. Zeng, and S. Luo, A Double Network Gel as Low Cost and Easy Recycle Adsorbent: Highly Efficient Removal of Cd(II) and Pb(II) Pollutants from Wastewater, J. Hazard. Mater. 300 (2015), 153-160. [3] Lytuong Tran, Pingxiao Wu, Yajie Zhu, Shuai Liu, and Nengwu Zhu, Comparative Study of Hg(II) Adsorption by Thiol- and Hydroxyl-Containing Bifunctional Montmorillonite and Vermiculite, Appl. Surf. Sci. 356 (2015), 91-101. [4] Ke, F.; Qiu, L. G. Thiol-functionalization of metal-organic framework by a facile coordination-based postsynthetic strategy and enhanced removal of Hg2+ from water, J. Hazard. Mater. 196 (2011) 36-43. [5] Katarzyna Tyszczuk-Rotko, Ilona Sadok, and Mariusz Barczak, Thiol-Functionalized Polysiloxanes Modified by Lead Nanoparticles: Synthesis, Characterization and Application for Determination of Trace Concentrations of Mercury(II), Micropor. Mesopor. Mater. 230 (2016) 109-117. [6] J. Y. Bae, H. J. Lee, and W. S. Choi, Cube Sugar-Like Sponge/Polymer Brush Composites for Portable and User-Friendly Heavy Metal Ion Adsorbents, J. Hazard. Mater. 320 (2016) 133-142. [7] Feiyun Chen, Qiuping Wu, Qiufeng Lü, Yali Xu, and Yan Yu, Synthesis and Characterization of Bifunctional

16

Mesoporous Silica Adsorbent for Simultaneous Removal of Lead and Nitrate Ions, Separat. Purific. Technol. 151 (2015) 225-231. [8] Rusen Zhou, Renwu Zhou, Xianhui Zhang, Song Tu, Yingwu Yin, Size Yang, and Liyi Ye, An Efficient Bio-Adsorbent for thce Removal of Dye: Adsorption Studies and Cold Atmospheric Plasma Regeneration, J. Taiwan Inst. Chem. E. 68 (2016) 372-78. [9] J. Anwar, U. Shafique, Zaman Waheed uz, M. Salman, A. Dar, and S. Anwar, Removal of Pb(II) and Cd(II) from Water by Adsorption on Peels of Banana, Bioresour. Technol. 101 (2010) 1752-1755. [10] S.M. Yakout, et al. Uranium sorption onto activated carbon prepared from rice straw: Competition with humic acids, Appl. Surf. Sci. 280(2013) 745-750. [11] Khalid Z. Elwakeel, and Eric Guibal, Selective Removal of Hg(II) from Aqueous Solution by Functionalized Magnetic-Macromolecular Hybrid Material, Chem. Eng. J. 281 (2015) 345-359. [12] Ke Zhou, Zhaoguang Yang, Yaochi Liu, and Xiu Kong, Kinetics and Equilibrium Studies on Biosorption of Pb(Ii) from Aqueous Solution by a Novel Biosorbent: Cyclosorus Interruptus, J. Environ. Chem. Eng. 3 (2015) 2219-2228. [13] Ke Zhou, Yaochi Liu, Zhaoguang Yang, Houzhi Liu, and Ting Xie, High-Capacity Sorption of U(Vi) from Aqueous Solution Using a Bio-Based Oxidized Polymeric Material, J. Taiwan Inst. Chem. E. 63 (2016) 453-462. [14] Hassan Alijani, Zahra Shariatinia, and Abdolreza Aroujalian Mashhadi, Water Assisted Synthesis of Mwcnts over Natural Magnetic Rock: An Effective Magnetic Adsorbent with Enhanced Mercury(II) Adsorption Property, Chem. Eng. J. 281(2015) 468-481. [15] Shengdong Pan, Yun Zhang, Haoyu Shen, and Meiqin Hu, An Intensive Study on the Magnetic Effect of Mercapto-Functionalized Nano-Magnetic Fe3O4 Polymers and Their Adsorption Mechanism for the Removal of Hg(II) from Aqueous Solution, Chem. Eng. J. 210 (2012) 564-574. [16] Zhuoxing Wang, Jiang Xu, Yunjun Hu, Heng Zhao, Junliang Zhou, Yu Liu, Zimo Lou, and Xinhua Xu, Functional Nanomaterials: Study on Aqueous Hg(II) Adsorption by Magnetic [email protected] Nanoparticles, J. Taiwan Inst. Chem. E. 60 (2016) 394-402. [17] H. Tian, and J. He, Cellulose as a Scaffold for Self-Assembly: From Basic Research to Real Applications, Langmuir 32 (2016) 12269-12282. [18]

Xin

Lu,

Qiangfeng

Yin,

Zhong

Xin,

Yang

Li,

and

Ting

Han,

Synthesis

of

Poly(Aminopropyl/Methyl)Silsesquioxane Particles as Effective Cu(II) and Pb(II) Adsorbents, J. Hazard. Mater. 196 (2011) 234-241. [19] Yang Wang, Yun Zhang, Chen Hou, Xinghua He, and Mingzhu Liu, Preparation of a Novel Teta Functionalized Magnetic Pgma Nano-Absorbent by Atrp Method and Used for Highly Effective Adsorption of Hg(II), J. Taiwan Inst. Chem. E. 58 (2016) 283-289. [20] G. Huang, D. Wang, S. Ma, J. Chen, L. Jiang, and P. Wang, A New, Low-Cost Adsorbent: Preparation, Characterization, and Adsorption Behavior of Pb(II) and Cu(II), J. Colloid Interface Sci. 445 (2015) 294-302. [21] S. Venkateswarlu, and M. Yoon, Core-Shell Ferromagnetic Nanorod Based on Amine Polymer Composite ([email protected]) for Fast Removal of Pb(II) from Aqueous Solutions, ACS Appl Mater. Interfaces 7 (2015) 25362-25372. [22] Xia Wang, Pengfei Lv, Hua Zou, Ying Li, Xiaoyan Li, and Yaozu Liao, Synthesis of Poly(2-Aminothiazole) for Selective Removal of Hg(II) in Aqueous Solutions, Indus. Eng. Chem. Res. 55 (2016) 4911-4918. [23] Yuzhong Niu, Jinyun Yang, Rongjun Qu, Yanhong Gao, Na Du, Hou Chen, Changmei Sun, and Wenxiang Wang, Synthesis of Silica-Gel-Supported Sulfur-Capped Pamam Dendrimers for Efficient Hg(II) Adsorption: Experimental and Dft Study, Indus. Eng. Chem. Res. 55 (2016) 3679-3688. [24] Xiaohuan Wang, Li Yang, Junping Zhang, Chuanyi Wang, and Qiuye Li, Preparation and Characterization of

17

Chitosan–Poly(Vinyl Alcohol)/Bentonite Nanocomposites for Adsorption of Hg(II) Ions, Chem. Eng. J. 251 (2014) 404-412. [25] Renjie Li, Lifen Liu, and Fenglin Yang, Preparation of Polyaniline/Reduced Graphene Oxide Nanocomposite and Its Application in Adsorption of Aqueous Hg(II), Chem. Eng. J. 229 (2013) 460-468. [26] María-José López-Muñoz, Amaya Arencibia, Luis Cerro, Raquel Pascual, and Álvaro Melgar, Adsorption of Hg(II) from Aqueous Solutions Using TiO2 and Titanate Nanotube Adsorbents, Appl. Surf. Sci. 367 (2016) 91-100. [27] Ho, Y. S.; McKay, G. Kinetic models for the sorption of dye from aqueous solution by wood, Process Saf. Environ. Prot. 76(1998) 183-186. [28] Ho, Y. S. Second-order kinetic model for the sorption of cadmium onto tree fern: A comparison of linear and non-linear methods, Water Res. 40(2006) 119-123. [29] Boyd, G. E.; Adamson, A. W.; Myers, L. S. The exchange adsorption of ions from aqueous solutions by organic zeolites: II Kinetics, J. Am. Chem. Soc. 69(1947) 2836-2841. [30] Ho, Y. S. Removal of copper ions from aqueous solution by tree fern, Water Res. 37(2003): 2323-2328. [31] Ho, Y. S. Selection of optimum sorption isotherm, Carbon. 42(2004): 2115-2119. [32] Yuzhong Niu, Huanhuan Liu, Rongjun Qu, Shujie Liang, Hou Chen, Changmei Sun, and Yuming Cui, Preparation and Characterization of Thiourea-Containing Silica Gel Hybrid Materials for Hg(II) Adsorption, Indus. Eng. Chem. Res. 54 (2015) 1656-1664. [33] Milonjic, S. K., A consideration of the correct calculation of thermodynamic parameters of adsorption, J. Serb. Chem. Soc. 72(2007) 1363-1368. [34] Jinshui; Liuab and Xuezhong, Fast removal of aqueous Hg (II) with quaternary ammonium-functionalized magnetic mesoporous silica and regeneration, Mater. Chem. 21(2011) 6981-6988. [35] Farrukh, A., et al. Design of polymer-brush-grafted magnetic nanoparticles for highly efficient water remediation, ACS Appl. Mater. Interfaces, 5(2013) 3784-3793. [36] Zhang, X. Y., et al. Stabilization/solidification (S/S) of mercury-contaminated hazardous wastes using thiol-functionalized zeolite and Portland cement, J. Hazard. Mater. 168(2009) 1575-1580. [37] Niu, Y., et al. Preparation and Characterization of Thiourea-Containing Silica Gel Hybrid Materials for Hg (II) Adsorption, Industrial & Engineering Chemistry Research, 54(2015) 1656-1664. [38] Foad Raji, Majid Pakizeh, Study of Hg(II) species removal from aqueous solution using hybrid ZnCl2-MCM-41 adsorbent, Appl. Surf. Sci. 282(2013) 415-424. [39] Pérez-Quintanilla, D., et al. 2-Mercaptothiazoline modified mesoporous silica for mercury removal from aqueous media, J. Hazard Mater. 134(2006) 245-256.

18

Graphical abstract

H N

raw CI

adsorption process Hg 2+

Hg2+

Hg

Hg 2+Hg2+ 2+ 2+ Hg Hg Hg2+ Hg2+Hg 2+ Hg2+ 2+ 2+ Hg 2+ 2+ Hg 2+ Hg 2+Hg Hg Hg2+ Hg Hg2+ 2+ 2+ 2+ Hg Hg Hg Hg2+Hg 2+ Hg2+Hg2+ Hg 2+

2+

magnet

b

c

Equilibrium stage

S

CH2CH2NH

C

MCIT

CI/Fe3O4

Hg2+ Hg2+

OH CH

a

Chemical reaction stage

Diffusion stage

19

NH2