To ease the adsorbent recovery and to increase the adsorption capacity of polyaniline (PANI), aniline was polymerized in the presence of a solvothermally prepared nano-composite of reduced graphene oxide and Fe3O4 (RGO/Fe3O4). The polyaniline was formed along the RGO/Fe3O4 composite in transmission electron microscope (TEM). The thus formed PANI/RGO/Fe3O4 adsorbent was tested and applied in removing Hg(II) in aqueous solution. The initial adsorption rate as well as the adsorption capacity increases with the incorporation of RGO/Fe3O4. The magnetic separation of PANI/RGO/Fe3O4 was easy, and its regeneration can be carried out at the optimal pH of 2. Test results proved the competence of the prepared adsorbent in pollution remediation applications for safer water quality and environmental protection.

INTRODUCTION

Potentially toxic metal pollution is widespread across the world. For potentially toxic metal ions such as mercury (II), which causes Minamata disease (Yorifuji et al. 2009) and bio-transformation forms of methyl mercury makes it even more toxic. Mercury ions can be removed by chemical precipitation (Hutchison et al. 2008), chemical reduction (Wiatrowski et al. 2009) and separated by membrane (Urgun-Demirtas et al. 2012) or ion exchange (Monteagudo & Ortiz 2000). Economical and efficient adsorption (Khaloo et al. 2012; ShamsiJazeyi & Kaghazchi 2013) is also frequently used.

Low-price, highly stable polyaniline-based adsorbents have often been studied for this purpose (Li et al. 2013, 2014), because the nitrogen functional groups on polyaniline (PANI) can complexate with Hg(II) (Wang et al. 2009). Bulk PANI has poor adsorption capacity due to limited surface area and active sites. Compositing PANI with reduced graphene oxide (RGO) formed a favorable morphology, increased specific surface area and adsorption sites, nearly twice the maximum adsorption capacity for Hg(II) than polyaniline was achieved (Li et al. 2013). In spite of the fair adsorption performance, its separation and recovery is not so easy after adsorption.

The recovery and recycling of magnetic composite adsorbent by magnetic separation is easy and consumes less energy. To make adsorbent magnetic, Fe3O4 is often used for magnetic functional materials (Kakavandi et al. 2013) due to its strong magnetism, super-paramagnetism and small coercive force (Girginova et al. 2010). Nano-Fe3O4 itself possesses good adsorption capacity for different pollutants (Paşka et al. 2013). Its composites with powder activated carbon (Kakavandi et al. 2013), silicon dioxide (Girginova et al. 2010) or chitosan (Shalaby et al. 2014) have been studied. The composite of PANI and Fe3O4 can adsorb pollutants such as organic dyes (Mahto et al. 2014), Pb(II) (Shao et al. 2012) and Cr(VI) (Han et al. 2013).

In this study, for high adsorption capacity and easy recovery adsorbent, PANI, RGO and Fe3O4 are composited via a two-step method, first by solvothermal synthesis of RGO/Fe3O4, then modification with PANI by polymerization of aniline. Transmission electron microscope (TEM), Brunauer-Emmett-Teller (BET) the specific surface area analyses, Fourier transform infra-red (FTIR) and vibrating sample magnetometer (VSM) are used to characterize the adsorbent. The effects of RGO/Fe3O4 content, pH of solution, adsorption time and initial concentration of Hg(II) on removal were investigated. The adsorption kinetics, thermodynamics and regeneration/desorption behavior were quantified to evaluate performance. A synthetic Hg(II)-containing wastewater sample was used for the practical competence test.

EXPERIMENTAL

Materials

The Hg(II) stock solution (1,000 mg/L) was prepared using Hg(NO3)2·1/2H2O (97%) supplied by Jiangyan Huanqiu Chemical Works, and further diluted to obtain standard solutions. Aniline (99.5%) was supplied by Tianjin Bodi Chemical Co., Ltd. Graphite powder (99.85%, 30 um) and polyethylene glycol (molecular weight = 2,000) were obtained from Sinopharm Chemical Reagent Co., Ltd. Ammonium peroxydisulfate (APS, (NH4)2S2O8, 98%) was purchased from Tianjin Damao Chemical Reagent Co., Ltd. Iron chloride hexahydrate (FeCl3·6H2O, 99.0%) was purchased from Shantou Xilong Chemical Co., Ltd. Ethylene glycol (99.8%) was supplied by Tianjin Fuyu Fine Chemical Co., Ltd. The reagents were all of analytical grade and used as received.

Preparation of reduced graphene oxide/Fe3O4

Graphene oxide (GO) was prepared according to a modified Hummers method (Hummers Jr & Offeman 1958). Then solvothermally RGO/Fe3O4 was prepared (Zhou et al. 2010). GO (100 mg) was ultrasonically dispersed in 50 mL of ethylene glycol and FeCl3·6H2O (0.412 g) was added to the mixture and stirred for 3 h. After that, polyethylene glycol (1.0 g) and NaAc (3.6 g) were mixed with the solution. After stirring for 0.5 h, the mixture was transferred to a 200 mL autoclave and maintained at 473 K for 16 h. After natural cooling to room temperature, the product was washed with deionized (DI) water and ethanol several times by magnetic separation.

Synthesis of PANI/RGO/Fe3O4 composite and PANI

For preparing magnetic PANI/RGO/Fe3O4, RGO/Fe3O4 was blended with aniline solution. While the concentration of aniline, [aniline]/[APS] and [GO]/[Fe3+] remained unchanged over the study, the GO/aniline weight ratio varied from 0% to 35% to investigate the effect of the RGO/Fe3O4 content, and the corresponding product samples were described as 0% GO (pure PANI) to 35% GO. For 30% GO, 0.18 g aniline was dissolved in 30 mL HCl solution to form aniline solution. Then RGO/Fe3O4 synthesized using 54 mg GO was dispersed in the resulting solution. While maintaining stirring, 0.44 g of ammonium peroxydisulfate (APS) in HCl solution was slowly added into the aniline/RGO/Fe3O4 mixture. For the reaction system, the initial pH value was 1.9 and the temperature was fixed at 0–5 °C. After stirring for 10 h, the product formed was de-doped with 1 M ammonia solution and washed with plenty of water. After vacuum drying at 333 K and sieving, PANI/RGO/Fe3O4 or PANI powder (100–120 mesh) was obtained.

For the sake of accuracy, all the adsorption data for one material were obtained using the material synthesized from the same synthesis experiment.

Characterization

The adsorbents were characterized using transmission electron microscope (TEM, JEM-100CX-II), FTIR (Shimadzu 8400 s spectrometer) and vibrating sample magnetometer (VSM, JDM-13). The sample BET specific surface area was measured using the Quantachrome NOWA 4,000 apparatus. The zeta potential was measured using ZETASIZER nano-series (Nano ZS-90 Malvern, UK).

Adsorption experiments

For the batch adsorption test, 10 mg adsorbent was added into 50 mL of Hg(II) ion solution; the solution pH was adjusted with 0.1 M NaOH/HNO3. The effect of different parameters including GO content (0–35%) and solution pH value (3.0–7.0) was investigated.

For kinetic studies, 0.2 g/L PANI or PANI/RGO/Fe3O4 was added into 125.45 mg/L mercury (II) solution (pH = 6.5 buffered by 2 mM NaAc/HAc solution). In the adsorption process, the suspension samples were withdrawn at certain intervals (0–420 min) for measurement of the Hg(II) concentration.

For the isotherm studies, Hg(II) solutions (pH = 6.5 buffered by 2 mM NaAc/HAc solution) with various initial concentrations (67.69–364.44 mg/L) were prepared and the solution pH was buffered at 6.5 by 2 mM NaAc/HAc solution. A concentration of 0.2 g/L of PANI or PANI/RGO/Fe3O4 was added to the above solutions and the suspensions were kept in a rotating incubator at 298 K for 24 h to reach complete equilibrium. After shaking, the solid-liquid separation was carried out by filtration or magnetic separation for PANI or PANI/RGO/Fe3O4, respectively.

Colorimeter with dithizone method through a UV/Vis-721 spectrophotometer was applied to determine the concentration of Hg(II) in the liquid (GB 7496–87, China). The adsorption capacity qt (mg/g) at time t was calculated using Equation (1): 
formula
1
where Co and Ct are the Hg(II) concentrations before and after adsorption, respectively (mg/L), V is the volume of solution, and W is the mass of the adsorbent used (g).
The amount of adsorption at equilibrium qe (mg/g) was determined by Equation (2): 
formula
2
where Ce is the Hg(II) concentration at equilibrium (mg/L).

Regeneration tests and material stability

For desorption studies, 0.2 g/L of PANI/RGO/Fe3O4 was first adsorbed in 50 mL Hg(II) (156.15 mg/L) solution for 8 h. Then the Hg(II) loaded adsorbent was placed in 40 mL HCl solution with different pH (1.0–3.5) for desorption. The above steps were carried out at 303 K and repeated 4 times to investigate the reusability of PANI/RGO/Fe3O4. To study the relationship between the pH value of the desorbing agent, desorption efficiency and recovery ratio, the total Fe content was determined (GB 3049-86) before and after the exposure of RGO/Fe3O4 and PANI/RGO/Fe3O4 to the desorbing agent. In addition, a certain amount of PANI/RGO/Fe3O4 and RGO/Fe3O4 with the same total Fe content was exposed to solution of pH 2.5 for 24 h to investigate the effect of PANI on preventing Fe leaching.

Synthetic waste water tests

To investigate the practical application abilities and selective adsorption of PANI/RGO/Fe3O4, a simulated coal-fired utility scrubber wastewater containing 1.37 mg/L Hg(II) was prepared according to previous literature (Nam et al. 2003), and the composition listed in Table 1. The solution pH was adjusted to 6.5. PANI/RGO/Fe3O4 with various dosages (0.0125–0.2 g/L) was added to 100 mL of wastewater sample and shaken continuously for 8 h at 298 K.

Table 1

Characteristics of synthetic wastewater samples

Ion Initial concentration (mg/L) Ion Initial concentration (mg/L) 
Ca2+ 349  206 
Mg2+ 49 Cl 1,101 
Zn2+ 30  549 
Na+ 678  1,338 
Ion Initial concentration (mg/L) Ion Initial concentration (mg/L) 
Ca2+ 349  206 
Mg2+ 49 Cl 1,101 
Zn2+ 30  549 
Na+ 678  1,338 

All the adsorption experiments were repeated three times and the standard deviation was ±3%.

RESULTS AND DISCUSSION

Material characterization

The TEM images of PANI (Figure 1(a)), RGO/Fe3O4 (Figure 1(b)) and PANI/RGO/Fe3O4 (Figure 1(c)) are compared in Figure 1. While PANI prepared here, appears as piles of irregular nanofibers that aggregate loosely, RGO/Fe3O4 has many stacked layers laden with granular Fe3O4, the TEM image of PANI/RGO/Fe3O4 revealed a lamellar structure which is much different from polyaniline. Tiny sized PANI species were seen growing along RGO/Fe3O4, instead of big granular or long fiber forms of PANI.
Figure 1

TEM images of PANI (a), RGO/Fe3O4 (b) and PANI/RGO/Fe3O4 (c).

Figure 1

TEM images of PANI (a), RGO/Fe3O4 (b) and PANI/RGO/Fe3O4 (c).

The effect of RGO/Fe3O4 on the morphology and structure of PANI is also reflected in the BET surface area measurement. As shown in Figure 2(a), the BET specific surface area of PANI/RGO/Fe3O4 (13.50 m2/g) is nearly 3.6 times larger than that of PANI (3.79 m2/g) due to the incorporation of RGO/Fe3O4 with a high specific surface area (41.18 m2/g).
Figure 2

Nitrogen adsorption-desorption isotherm, the inset graph is the pore size distributions (a); zeta potential curves (b); FTIR spectra (c); and magnetization curves, the inset graph is the separation of PANI (left) and PANI/RGO/Fe3O4 (right) by a magnet (d).

Figure 2

Nitrogen adsorption-desorption isotherm, the inset graph is the pore size distributions (a); zeta potential curves (b); FTIR spectra (c); and magnetization curves, the inset graph is the separation of PANI (left) and PANI/RGO/Fe3O4 (right) by a magnet (d).

The result of zeta potential measurement for PANI/RGO/Fe3O4 (30% GO) is shown in Figure 2(b). The electrostatic point is around 5.75, indicating a negatively charged surface when pH > 5.75 and the PANI/RO/Fe3O4 become positively charged when pH < 5.75.

The FTIR spectra of RGO/Fe3O4, PANI and PANI/RGO/Fe3O4 composite are shown in Figure 2(c). For RGO/Fe3O4, the peaks around 1,539 cm−1 and 1,188 cm−1 are related to the aromatic C=C stretching and C-O stretching, respectively (Chandra & Kim 2011). Accordingly, for PANI, peaks at 1,561 cm−1 and 1,126 cm−1 represent quinoid (Q) and C=N stretching (−N=Q=N−) (Palaniappan 2001), respectively, demonstrating the de-doped PANI in its oxidized form. Other PANI peaks include the C = C in benzenoid (B) ring (1,484 cm−1) (Ruckenstein & Yin 2001), C-N stretching (−N–B–N−, 1,226 cm−1) (Boyer et al. 1998), C-N stretching (1,299 cm−1) (Athawale et al. 2002) and C-H stretching (802 cm−1) (Campos et al. 1999). All these peaks for PANI are observed in the FTIR spectrum of PANI/RGO/Fe3O4 (Figure 1(c)), indicating the compositing of PANI and RGO/Fe3O4 has been successful. The nitrogen-containing functional groups are the adsorption sites and they play a major role in Hg(II) adsorption.

The magnetic property of PANI/RGO/Fe3O4 is critical for recovering adsorbent from the aqueous phase by magnetic force. It was investigated by VSM at room temperature with an applied field −4394 Oe < H < 4394 Oe. As shown in Figure 2(d), the sample is super-paramagnetic and the specific saturation magnetization (Ms) of PANI/RGO/Fe3O4 reaches 21.37 emu/g, being lower than that of RGO/Fe3O4 and Fe3O4 reported in previous literature (Zhou et al. 2010; Han et al. 2013) because of the bigger proportion of PANI. The separation of PANI/RGO/Fe3O4 and PANI with applied magnetic field is shown in the inset of Figure 2(d). When a magnet is put beside the beaker, PANI/RGO/Fe3O4 particles are attracted quickly toward the magnet, and the solution becomes clear in 10 s, indicating that a good adsorbent separation can be realized.

Effect of RGO/Fe3O4 content

The ratio of aniline to RGO/Fe3O4 was investigated, and it is expected to influence Hg(II) removal in three ways: (1) the morphology, functional group and the adsorption capacity changes due to the presence of RGO (Li et al. 2013); (2) adsorption due to the surface complexation between Hg(II) and hydroxyl groups in Fe3O4 (Wiatrowski et al. 2009); and (3) chemical reducing Hg(II) by magnetite (Wiatrowski et al. 2009). The effect of RGO/Fe3O4 on the Hg(II) adsorption is shown in Figure 3. The adsorption capacity for Hg(II) dropped with a decrease of PANI content, such as 10% GO. But the adsorption capacity increases with the further addition of RGO/Fe3O4 until the maximum (389 mg/g) is reached at 30% GO, higher than that of nano fiber aggregated PANI (339 mg/g). The adsorption capacity then decreases when the RGO/Fe3O4 content continues increasing, because of the low removal capacity of RGO/Fe3O4 (53 mg/g, much lower than PANI and PANI/RGO/Fe3O4). This indicates the major role of PANI in PANI/RGO/Fe3O4 in Hg(II) adsorption. PANI/RGO/Fe3O4 prepared at 30% GO possess the optimal structure and maximum accessible adsorption sites, resulting in maximal adsorption for Hg(II).
Figure 3

Optimization of RGO/Fe3O4 content in PANI (C0 = 135.50 mg/L; dosage = 0.2 g/L; pH = 6.3; adsorption time = 8 h; temperature = 298 K).

Figure 3

Optimization of RGO/Fe3O4 content in PANI (C0 = 135.50 mg/L; dosage = 0.2 g/L; pH = 6.3; adsorption time = 8 h; temperature = 298 K).

The effect of solution pH value

Solution pH value affects adsorption by changing both the surface charge density of PANI/RGO/Fe3O4 and the speciation of Hg(II). There will be an optimal pH for the adsorption of Hg(II). As shown in Figure 4(a), a rising trend in adsorption is observed when the pH value increases from 3.0 to 7.0. It should be stated that no precipitation occurred up to pH 7.0. Although the Hg(II) cations decreased with the increase of solution (Figure 4(b)), the Hg(II) cations still existed even at pH 7.0 (0.006% Hg(II)). The improved adsorption capacity of PANI/RGO/Fe3O4 can be explained by two reasons: (1) the competition of hydrogen ions with Hg(II) for bonding sites becomes weaker as the solution pH increases; (2) the deprotonation of functional groups (imine, amine and hydroxyl groups) on PANI/RGO/Fe3O4 makes its surface negatively charged, which attracts Hg(II) cations through electrostatic interaction. (Wiatrowski et al. 2009).
Figure 4

(a) Effect of pH on Hg(II) removal by PANI/RGO/Fe3O4 (C0 = 125.45 mg/L; dosage = 0.2 g/L; adsorption time = 8 h; temperature = 298 K) and (b) distribution of Hg(II) species as a function of pH calculated by MINEQL+ at same conditions.

Figure 4

(a) Effect of pH on Hg(II) removal by PANI/RGO/Fe3O4 (C0 = 125.45 mg/L; dosage = 0.2 g/L; adsorption time = 8 h; temperature = 298 K) and (b) distribution of Hg(II) species as a function of pH calculated by MINEQL+ at same conditions.

Adsorption kinetics for Hg(II) and the effect of adsorption time

The Hg(II) adsorbed on the PANI/RGO/Fe3O4 and PANI is shown in Figure 5 as a function of adsorption time. For both PANI and PANI/RGO/Fe3O4, 70% of the adsorption capacity is reached in 1 h. The adsorption becomes stabilized after 5 h. For PANI and PANI/RGO/Fe3O4, the equilibrium adsorption capacities are 313 mg/g and 375 mg/g, respectively.
Figure 5

Adsorption kinetics, pseudo-first-order and pseudo-second-order plots for the adsorption of Hg(II) by PANI and PANI/RGO/Fe3O4 (C0 = 125.45 mg/L; dosage = 0.2 g/L; pH = 6.5 buffered by 2 mM NaAc/HAc solution; temperature = 298 K).

Figure 5

Adsorption kinetics, pseudo-first-order and pseudo-second-order plots for the adsorption of Hg(II) by PANI and PANI/RGO/Fe3O4 (C0 = 125.45 mg/L; dosage = 0.2 g/L; pH = 6.5 buffered by 2 mM NaAc/HAc solution; temperature = 298 K).

All adsorption kinetic and initial adsorption rates were analyzed similarly as previously described (Khaloo et al. 2012), the results are shown in Figure 5.

The kinetic parameters obtained from fitting results are presented in Table 2. According to the correlation coefficient, the adsorption could be better explained by the pseudo-second-order equation for both PANI and PANI/RGO/Fe3O4. And the equilibrium adsorption capacities calculated from the pseudo-second-order equation are close to the experimental value. In addition, the initial adsorption rate is 45 mg/(g·min) for PANI/RGO/Fe3O4, which is twice as fast as 22 mg/(g·min) for PANI.

Table 2

Kinetic parameters for the adsorption of Hg(II) by PANI and PANI/RGO/Fe3O4 composite

Kinetics equation Kinetic parameters PANI PANI/RGO/Fe3O4 
Pseudo-first-order qe (mg/g) 290.65 346.44 
K1 (1/min) 0.05213 0.08174 
R2 0.881 0.879 
qe (mg/g) 314.16 367.45 
Pseudo-second-order K2 (g/(mg min)) 2.29 × 10−4 3.34 × 10−4 
R2 0.950 0.955 
Kinetics equation Kinetic parameters PANI PANI/RGO/Fe3O4 
Pseudo-first-order qe (mg/g) 290.65 346.44 
K1 (1/min) 0.05213 0.08174 
R2 0.881 0.879 
qe (mg/g) 314.16 367.45 
Pseudo-second-order K2 (g/(mg min)) 2.29 × 10−4 3.34 × 10−4 
R2 0.950 0.955 

Adsorption isotherm

In isotherm studies, 0.2 g/L of adsorbent was contacted with Hg(II) solution with a wide range of initial Hg(II) concentration for 24 h. The equilibrium adsorption capacities of two materials and the corresponding equilibrium Hg(II) concentration (Ce) are shown in Figure 6. The isotherm data were analyzed using the Langmuir and Freundlich models as previously described (Ling et al. 2014).
Figure 6

Adsorption isotherms of Hg(II) on PANI and PANI/RGO/Fe3O4 modeling with Langmuir and Freundlich equation (dosage = 0.2 g/L; pH = 6.5 buffered by 2 mM NaAc/HAc solution; temperature = 298 K; contact time = 24 h).

Figure 6

Adsorption isotherms of Hg(II) on PANI and PANI/RGO/Fe3O4 modeling with Langmuir and Freundlich equation (dosage = 0.2 g/L; pH = 6.5 buffered by 2 mM NaAc/HAc solution; temperature = 298 K; contact time = 24 h).

The isotherm parameters obtained from fitting results are shown in Table 3. According to the correlation coefficients, the obtained isotherm data prove that the Freundlich model is a better fit for PANI/RGO/Fe3O4, and two isotherm models both fit for PANI. It suggests a multilayer adsorption mechanism exists for the Hg(II) adsorption by these two materials, indicating the non-uniformity of adsorption sites on both adsorbents. Especially for layer-by-layer assembly PANI/RGO/Fe3O4, the result can be explained by the higher BET surface area of PANI/RGO/Fe3O4, much richer functional groups and multiple Hg(II) removal mechanisms (adsorption and redox). In addition, the Kf value related to the adsorption capacity indicated that the adsorption capacity of PANI/RGO/Fe3O4 is larger than that of PANI. This means magnetization of PANI with RGO/Fe3O4 is realized without compromising the adsorption capacity of PANI. Considering the great potential of RGO in improving adsorption capacity (Li et al. 2013), it is suggested that further exploration be carried out to replace the solvothermally prepared RGO/Fe3O4 to form PANI/RGO/Fe3O4 composites with a much higher specific surface area.

Table 3

Isotherm parameters for the adsorption of Hg(II) by PANI and PANI/RGO/Fe3O4 composite (pH = 6.5 buffered by 2 mM NaAc/HAc solution; temperature = 298 K; contact time = 24 h)

Adsorption Isotherm Isotherm parameters PANI PANI/RGO/Fe3O4 
Langmuir qm (mg/g) 491.73 460.24 
b (L/mg) 0.017 0.29 
R2 0.997 0.736 
Kf 65.09 227.91 
Freundlich n 3.08 10.11 
R2 0.977 0.907 
Adsorption Isotherm Isotherm parameters PANI PANI/RGO/Fe3O4 
Langmuir qm (mg/g) 491.73 460.24 
b (L/mg) 0.017 0.29 
R2 0.997 0.736 
Kf 65.09 227.91 
Freundlich n 3.08 10.11 
R2 0.977 0.907 

Regeneration test and material stability

To evaluate the regeneration and stability of adsorbent, different concentrations of HCl solution were used as desorption agent. For regeneration, PANI/RGO/Fe3O4 saturated with Hg(II) (after adsorption equilibrium in 50 mL of Hg(II) solution (156.14 mg/L)) was mixed with 40 mL of HCl solution (pH 1–3.5) for 12 h. Then it was tested for the next cycle of adsorption.

In Figure 7, the effect of HCl concentration on regeneration is shown. The adsorptive Hg(II) removal increases with the increase of HCl concentration after first regeneration, indicating that a high concentration of HCl facilitates the desorption of Hg(II) from PANI/RGO/Fe3O4. The HCl-dependent desorption behavior can be explained by the pH effect on adsorption and the favorable formation of stable nonsorbing Hg-Cl complex at high chloride concentrations (Figure 8). However, as the regeneration times increase, the Hg(II) removal ratio of PANI/RGO/Fe3O4 regenerated with solution of pH = 1 became worse, possibly due to adsorbent loss caused by leaching of Fe in strong acid. So, the solution pH at 2 seems to be optimal for regeneration efficiency and adsorption capacity recovery.
Figure 7

Adsorption/desorption cycles of Hg(II) by PANI/RGO/Fe3O4 using HCl solutions with different pH; inset graph is the total Fe content in PANI/RGO/Fe3O4 after re-use (40 mL of desorption medium; dosage = 0.2 g/L; C0 = 156.14 mg/L; temperature = 298 K; adsorption time = 8 h; desorption time = 12 h).

Figure 7

Adsorption/desorption cycles of Hg(II) by PANI/RGO/Fe3O4 using HCl solutions with different pH; inset graph is the total Fe content in PANI/RGO/Fe3O4 after re-use (40 mL of desorption medium; dosage = 0.2 g/L; C0 = 156.14 mg/L; temperature = 298 K; adsorption time = 8 h; desorption time = 12 h).

Figure 8

Distribution of Hg(II) species as a function of HCl concentration (pH) calculated by MINEQL+ at 90 mg/L of Hg(II) concentration.

Figure 8

Distribution of Hg(II) species as a function of HCl concentration (pH) calculated by MINEQL+ at 90 mg/L of Hg(II) concentration.

After the desorption experiments, the adsorbents were recovered/collected for the measurement of residual Fe content. As shown in the inset of Figure 8, the solution pH is an important parameter affecting Fe leaching. The amount of Fe dissolved is higher under more acidic conditions, and when pH > 2, Fe leaching is reduced gradually.

In addition, after being exposed to a solution of pH 2.5 for 24 h, for RGO/Fe3O4 and PANI/RGO/Fe3O4 with the same initial Fe dosage (1.91 mg/L), the stability appears to be different. For PANI/RGO/Fe3O4, leaching of Fe is 2.29%, versus 3.29% for RGO/Fe3O4, indicating that PANI coated on RGO/Fe3O4 inhibited Fe leaching to some extent.

Synthetic wastewater tests

The prepared PANI/RGO/Fe3O4 was tested to remove Hg(II) in synthetic wastewater. As shown in Figure 9, the adsorbent has low Hg(II) adsorption capacity in simulated wastewater, which can be attributed to the low initial Hg(II) concentration and the high chloride concentration (31.02 mM). In this high chloride solution, the Hg(II) species (% Hg(II)) exists as HgCl2 (71.37), Hg(OH)2 (22.55), HgCl4 (4.43) and HgClOH (1.65), which have low attraction to the surface functional groups of the adsorbent. Conversely, the Hg(II) removal ratio increases with the increasing dosage of adsorbent, and reaches 75.16% at 0.2 g/L of PANI/RGO/Fe3O4, indicating that the prepared PANI/RGO/Fe3O4 could still be applied in the treatment of practical Hg(II) containing wastewater.
Figure 9

The effect of adsorbent dosage on the adsorption of Hg(II) from a simulated wastewater sample (pH = 6.5; adsorption time = 8 h; temperature = 298 K).

Figure 9

The effect of adsorbent dosage on the adsorption of Hg(II) from a simulated wastewater sample (pH = 6.5; adsorption time = 8 h; temperature = 298 K).

CONCLUSIONS

PANI/RGO/Fe3O4 nano-composite was prepared successfully by solvothermal synthesis of RGO/Fe3O4 and followed by polymerization of aniline. The nano-composite adsorbent with 30% GO/aniline weight ratio has excellent adsorption capacity for Hg(II) and the magnetic property for easy magnetic recovery. TEM and BET analysis indicated a better structure of PANI/RGO/Fe3O4 than PANI that grows along and covers the RGO/Fe3O4. Having a higher specific surface area than PANI, its initial adsorption rate for Hg(II) is much higher than PANI. Its mercury (II) removal was mainly caused by complexation between mercury (II) and the N-containing groups in PANI. The adsorbent could be well regenerated in HCl solution (pH 2), PANI in PANI/RGO/Fe3O4 could inhibit leaching of iron during acidic regeneration.

ACKNOWLEDGEMENTS

The financial support from the National Natural Science Foundation of China (No. 21177018), the Program of Introducing Talents of Discipline to Universities (B13012) and Basic Research Project from Dalian University of Technology (DUT14QY48) is greatly acknowledged.

REFERENCES

REFERENCES
Athawale
A. A.
Kulkarni
M. V.
Chabukswar
V. V.
2002
Studies on chemically synthesized soluble acrylic acid doped polyaniline
.
Materials Chemistry and Physics
73
(
1
),
106
110
.
Boyer
M. I.
Quillard
S.
Rebourt
E.
Louarn
G.
Buisson
J. P.
Monkman
A.
Lefrant
S.
1998
Vibrational analysis of polyaniline: a model compound approach
.
The Journal of Physical Chemistry B
102
(
38
),
7382
7392
.
Campos
T. L. A.
Kersting
D. F.
Ferreira
C. A.
1999
Chemical synthesis of polyaniline using sulphanilic acid as dopant agent into the reactional medium
.
Surface and Coatings Technology
122
(
1
),
3
5
.
Girginova
P. I.
Daniel-da-Silva
A. L.
Lopes
C. B.
Figueira
P.
Otero
M.
Amaral
V. S.
Pereira
E.
Trindade
T.
2010
Silica coated magnetite particles for magnetic removal of Hg2+ from water
.
Journal of Colloid and Interface Science
345
(
2
),
234
240
.
Han
X.
Gai
L.
Jiang
H.
Zhao
L.
Liu
H.
Zhang
W.
2013
Core-shell structured Fe3O4/PANI microspheres and their Cr(VI) ion removal properties
.
Synthetic Metals
171
,
1
6
.
Hummers
W. S.
Jr
Offeman
R. E.
1958
Preparation of graphitic oxide
.
Journal of the American Chemical Society
80
(
6
),
1339
.
Hutchison
A.
Atwood
D.
Santilliann-Jiminez
Q. E.
2008
The removal of mercury from water by open chain ligands containing multiple sulfurs
.
Journal of Hazardous Materials
156
(
1
),
458
465
.
Kakavandi
B.
Esrafili
A.
Mohseni-Bandpi
A.
Jonidi
J. A.
Rezaei
K. R.
2013
Magnetic Fe3O4@C nanoparticles as adsorbents for removal of amoxicillin from aqueous solution
.
Water Science & Technology
69
(
1
),
591
598
.
Khaloo
S. S.
Matin
A. H.
Sharifi
S.
Fadaeinia
M.
Kazempour
N.
Mirzadeh
S.
2012
Equilibrium, kinetic and thermodynamic studies of mercury adsorption on almond shell
.
Water Science & Technology
65
(
8
),
1341
1349
.
Ling
C.
Liu
F.
Long
C.
Wei
M.
Li
A.
2014
Highly efficient co-removal of copper (II) and phthalic acid with self-synthesized polyamine resin
.
Water Science & Technology
69
(
9
),
1879
1885
.
Mahto
T. K.
Chowdhuri
A. R.
Sahu
S. K.
2014
Polyaniline-functionalized magnetic nanoparticles for the removal of toxic dye from wastewater
.
Journal of Applied Polymer Science
131
(
19
),
40840
40848
.
Monteagudo
J. M.
Ortiz
W. J.
2000
Removal of inorganic mercury from mine waste water by ion exchange
.
Journal of Chemical Technology and Biotechnology
75
(
9
),
767
772
.
Nam
K. H.
Gomez-Salazar
S.
Tavlarides
L. L.
2003
Mercury(II) adsorption from wastewaters using a thiol functional adsorbent
.
Industrial & Engineering Chemistry Research
42
(
9
),
1955
1964
.
Paşka
O.
Ianoş
R.
Păcurariu
C.
Brădeanu
A.
2013
Magnetic nanopowder as effective adsorbent for the removal of Congo Red from aqueous solution
.
Water Science & Technology
69
(
6
),
1234
1240
.
Urgun-Demirtas
M.
Benda
P. L.
Gillenwater
P. S.
Negri
M. C.
Xiong
H.
Snyder
S. W.
2012
Achieving very low mercury levels in refinery wastewater by membrane filtration
.
Journal of Hazardous Materials
215
,
98
107
.
Wang
J.
Deng
B.
Chen
H.
Wang
X.
Zheng
J.
2009
Removal of aqueous Hg(II) by polyaniline: sorption characteristics and mechanisms
.
Environmental Science & Technology
43
(
14
),
5223
5228
.
Wiatrowski
H. A.
Das
S.
Kukkadapu
R.
Ilton
E. S.
Barkay
T.
Yee
N.
2009
Reduction of Hg(II) to Hg(0) by magnetite
.
Environmental Science & Technology
43
(
14
),
5307
5313
.
Yorifuji
T.
Kashima
S.
Tsuda
T.
Harada
M.
2009
What has methylmercury in umbilical cords told us? — Minamata disease
.
Science of The Total Environment
408
(
2
),
272
276
.
Zhou
K.
Zhu
Y.
Yang
X.
Li
C.
2010
One-pot preparation of graphene/Fe3O4 composites by a solvothermal reaction
.
New Journal of Chemistry
34
(
12
),
2950
2955
.