Hyperbranched multiple polythioamides made from elemental sulfur for mercury adsorption

Akram Yasin a, Yurong Chen ab, Yanxia Liu ab, Letao Zhang a, Xingjie Zan *a and Yagang Zhang *abc
aXinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi 830011, China. E-mail: ygzhang@ms.xjb.ac.cn
bUniversity of Chinese Academy of Sciences, Beijing 100049, China
cDepartment of Chemical & Environmental Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China

Received 12th October 2019 , Accepted 30th November 2019

First published on 3rd December 2019

Different from traditional polyethylenimine (PEI) modified Hg(II) adsorbent materials, a novel hyperbranched polythioamide adsorbent (SPD) was prepared by using sulfur, PEI and 1,4-diethynylbenzene (DEB) as monomers. There are tens and thousands of tons of sulfur produced from oil refinery processes as a major by-product each year, and its utilization becomes a concern due to its limited consumption, environmental problems and safety. Herein, a SPD was fabricated through a one-pot, catalyst-free polymerization which even can be solvent-free via using sulfur to solidify the liquid PEI. A kinetic study showed that the adsorption of Hg(II) ions followed the pseudo-second-order model, indicating chemical adsorption between Hg(II) ions and SPD. The SPD had a maximum adsorption capacity of 1161.99 mg g−1. FT-IR and XPS measurements indicated that the removal of Hg(II) by SPD was mainly controlled by the coordination interaction between Hg(II) ions and nitrogen groups (i.e., amine and imine groups), and also C[double bond, length as m-dash]S bonds of SPD. By using elemental sulfur, PEI becomes the backbone of the polymer system by forming –C([double bond, length as m-dash]S)NH– and –C([double bond, length as m-dash]S)N< groups unlike the grafted part of traditional PEI modified materials. TG, DTG, and DSC measurements showed that the main degradations occurred at around 160 °C and 360 °C.

1. Introduction

Sulfur is a major by-product produced in oil refining and natural gas purification processes. More than 70 million tons of elemental sulfur are produced every year,1,2 and the production has been continuously expanded with industrial development. However, the utilization of sulfur is limited, which is mainly used in the production of sulfuric acid, resulting in rapidly growing mountains of sulfur on the ground.1 Thus, it is of great scientific and industrial significance to directly utilize elemental sulfur as a raw material to synthesize useful functional materials.

High sulfur content polymers have attracted much attention recently because of their outstanding properties in energy3–5 and environmental applications6–8 and other practical areas.9–12 Herein, we report a mercury (Hg) adsorbent material with high adsorption capacity using sulfur as the raw material.

Mercury is one of the neurotoxic metals with non-biodegradable, accumulative and mobile characteristics. It is widely used in industries including coal combustion, oil and natural gas refining, chlorine alkali chemistry, batteries and various metallurgic processes.6 Mercury pollution threatens both the environment and human health. Therefore, developing cost-effective materials for removing Hg(II) ions from aquatic environments with high efficiency is an urgent demand for securing public health.

It is well known that Hg(II) forms a stable complex with the thioamide –C([double bond, length as m-dash]S)NH– ligand, which can be employed to remove Hg(II) ions.13 The formation of thioamide molecules was reported by Nguyen et al. during a selective oxidative reaction between two different aliphatic primary amines using sulfur, in 2012.14 Also Nguyen et al. reported a three-component reaction between alkynes, elemental sulfur, and aliphatic amines to synthesize thioamide molecules in 2014.15 Later, B Z Tang's team reported polymerizations of aromatic diynes, elemental sulfur, and aliphatic diamines toward luminescent linear polythioamides,16 and Sun et al. prepared linear polythioamides with a high refractive index by the direct polymerization of aliphatic primary diamines in the presence of elemental sulfur.17 At the same time, B Z Tang's team reported the one-step synthesis of functional polythioureas, using sulfur, aliphatic diamines, and diisocyanides, utilized to detect and remove mercury pollution.18

However, up to now, few polymers containing thioamide or thiourea groups have been reported for Hg(II) ion removal.18–23 Although a thiourea containing polymer was utilized to detect mercury pollution with high sensitivity (Ksv = 224[thin space (1/6-em)]900 L mol−1),18 the low maximum mercury adsorption capacity (only tens to hundreds mg g−1) and the inability to regenerate and recycle are still the big bottle-neck of these materials.

With accumulative evidence, modifying polymers with amine groups can greatly enhance the ability to remove heavy metals.22,23 Polyethylene imine (PEI) has served as one of the promising amine rich resources to modify various biomass materials such as cellulose, chitosan, and carboxymethyl chitosan for the purpose of removing heavy metal ions.24–26 Zeng et al. reported a PEI functionalized carboxymethyl chitosan composite adsorbent with high maximum adsorption capacity (1594 mg g−1) and high efficiency.27

Inspired by these studies, we utilized elemental sulfur, PEI and 1,4-diethynylbenzene to generate a mercury removal material by combining the removal ability of PEI and thioamide groups in this work. Different from the previously reported linear polythioamides and PEI grafted biomass materials, the hyperbranched polythioamide structure derived from the backbone polymer of hyperbranched PEI was generated and well characterized. In the obtained hyperbranched polythioamide structure, there were two kinds of –C([double bond, length as m-dash]S)NH– groups formed in two different mechanisms during the reaction. In addition, this hyperbranched polythioamide structure exhibited excellent mercury removal ability. Thus, a large amount of sulfur by-product was utilized effectively to prepare a new type of mercury adsorption material with high adsorption capacity.

2 Experimental

2.1 Materials

1,4-Diethynylbenzene (DEB) and polyethylenimine (PEI) with Mw = 600, 1800 and 10[thin space (1/6-em)]000 were purchased from Adamas-beta. If not specified, the molecular weight of the PEI mentioned is 600. Pyridine was purchased from Aladdin. Sulfur, HgCl2 and methanol were purchased from Tianjin Baishi Chemical Industry Co. Ltd. All chemical precursors were used as received without any further purification.

2.2 Preparation of hyperbranched polythioamides (SPDs)

The hyperbranched polythioamides were synthesised by “one-pot” polymerization using sulfur (S), PEI (P) (Mw = 600) and DEB (D) as monomers. The mole ratio of sulfur and DEB was fixed at S[thin space (1/6-em)]:[thin space (1/6-em)]PEI[thin space (1/6-em)]:[thin space (1/6-em)]DEB = 4[thin space (1/6-em)]:[thin space (1/6-em)]x[thin space (1/6-em)]:[thin space (1/6-em)]1 in all samples, where x ranged from 0.33 to 0.9, and the samples were named SPD-x. The main processes were similar to those in the studies by B Z Tang and T B Nguyen.15,16 The procedure for preparing SPD-0.33 at 100 °C is given below as an example and the synthesis processes are shown in Fig. S1. 0.3539 g (11.06 mmol) of elemental sulfur and 0.3488 g (2.76 mmol) of DEB were added into a 20 mL glass vial equipped with a magnetic stir bar in air. 1.76 mL of pyridine was added into the tube and stirred at 100 °C, sulfur and DEB dissolved immediately. Then 0.5474 g (0.91 mmol) of PEI dissolved in 1 mL of pyridine was injected with a syringe and kept for 3 h before being cooled to room temperature. The rigid solid product was dug out and immersed in 100 mL of methanol and stirred with a magnetic stir bar overnight at room temperature to remove unreacted PEI and DEB. The powders were filtered and immersed in 20 mL of pyridine and heated to 100 °C. After stirring for 30 min, the mixture was filtered immediately at a high temperature to remove unreacted sulfur and washed with methanol three times (3 × 20 mL) to remove pyridine. Finally, the powder was dried under vacuum at 50 °C overnight and ground by using a grinder into particles and passed through a 200 mesh sieve before use.

2.3 Characterization

The microstructures of all samples were observed with a field emission scanning electron microscope (FE-SEM, ZEISS, Germany). The surface area was tested by the Brunauer-Emmer-Teller (BET) method with use of the nitrogen absorption/desorption measurement (ASAP2460 4MP, micromeritics, USA). Differential scanning calorimetry (DSC) thermograms for the samples were obtained with a PerkinElmer STA 8000 TG/DSC thermal analyzer (USA) in a nitrogen atmosphere from 30 °C to 800 °C at a heating rate of 10 °C min−1. Fourier transform infrared (FT-IR) spectra were recorded using a Bruker VERTEX-70 (Germany) FT-IR spectrometer in the wavelength range from 400 to 4000 cm−1 by making pellets with KBr. The 1H-NMR and 13C-NMR experiments were performed with a VARIAN 400 spectrometer (USA) (1H-NMR at 400 MHz and 13C-NMR at 100 MHz) at room temperature using DMSO-d6 as the solvent and a JNM-ECZ600R 600 spectrometer (Japan) was used for solid state 13C-NMR experiments. Elemental analysis was performed using an Elementar elemental analyzer (vario EL cube, Elementar, Germany). X-ray photoelectron spectroscopy (XPS) was performed by using a X-ray photoelectron spectrometer (K-Alpha+, Thermo Fisher, USA), with monochromatic Al Kα as an excitation source.r The concentration of Hg2+ in the dilute aqueous solution was quantified using an inductively coupled plasma atomic emission spectrometer (ICP-5000, FPI, China).

2.4 Mercury adsorption studies

The Hg2+ adsorption experiments were performed in 10 mL glass bottles. The pH of distilled water was adjusted to 5 with diluted HCl solution, and then solid HgCl2 was dissolved to prepare HgCl2 solution. 10 mg of SPD and 10 mL of HgCl2 solutions were mixed in glass bottles, and shaken in a thermostatic shaker at 30 °C for 3 h at 150 r min−1. In the end, the adsorbent was filtered out and the Hg2+ ion concentration in the residue solution was measured by using an inductively coupled plasma atomic emission spectrometer (ICP-5000, FPI, China).

The removal rate and adsorption capacity were calculated by using the following equations:

image file: c9py01544b-t1.tif(1)
image file: c9py01544b-t2.tif(2)
where C0 and Ce are the initial and equilibrium mercury ion concentrations (mg L−1), Qe is the equilibrium adsorption capacity (mg g−1), m is the mass of SPD (g), and V is the volume of solution (L).

Kinetic studies were performed at 30 °C by adding 10 mg of the adsorbent to 10 mL of HgCl2 solution (2000 mg L−1) at pH 5.0. At each predetermined time point, a suspension sample was filtered and the Hg2+ ion concentration in the filtrate was measured. Experimental data were fitted by using the pseudo-first-order (eqn (3)) and the pseudo-second-order (eqn (4)) equations.22,27,28

ln(QeQt)=ln[thin space (1/6-em)]Qek1t(3)
image file: c9py01544b-t3.tif(4)
where Qt (mg g−1) is the amount of Hg2+ adsorbed by the sorbent at time t (min); Qe (mg g−1) is the amount of Hg2+ adsorbed at equilibrium; k1 (min−1) and k2 (g mg−1 min−1) are the pseudo-first-order and pseudo-second-order adsorption rates, respectively.

In the adsorption isotherm experiment, the initial Hg2+ ion concentrations were varied from 200 to 2000 mg L−1 at pH 5.0, and the mixtures were stirred for 180 min at 30 °C. Then the residual Hg2+ ions in each solution were measured. The experimental data were fitted with Langmuir (eqn (5)), Freundlich (eqn (6)) and Langmuir–Freundlich (eqn (7)) equations.22,27,28

image file: c9py01544b-t4.tif(5)
Qe = kFCe1/nF(6)
image file: c9py01544b-t5.tif(7)
where Qm (mg g−1) represents the maximum adsorption capacity of the adsorbent. kL (L mg−1) is the Langmuir constant, which is related to the affinity of the binding sites. kF (mg g−1) and n are both Freundlich constants that represent the adsorption capacity and adsorption favorability of the system, respectively.

In the adsorbent reusability test, the adsorbed Hg2+ ions on SPD were desorbed with 2 mol L−1 HNO3. After 5 h shaking at 30 °C, the suspension was filtered and washed with distilled water several times. Then, the adsorbent was dried in a vacuum oven at 50 °C overnight and reused in the next cycle. To test the reusability of the sorbent, the adsorption–desorption cycles were repeated 5 times.

3 Results and discussion

3.1 Synthesis and characterization

SPD-0.33, SPD-0.5, SPD-0.7, SPD-0.8 and SPD-0.9 were prepared using the method described above. Table 1 shows the monomer feeding ratio of polymerizations. The monomer weight was fixed at 1.25 g in each polymerization. Meanwhile, PEI with Mw = 1800, 10[thin space (1/6-em)]000 were also used as monomers to study the influence of molecular weight of PEI on the Hg2+ adsorption properties and the samples were denoted as SPD-0.8-1800 and SPD-0.8-10[thin space (1/6-em)]000. In detail, 0.8173 g of PEI-600, 1800, 10[thin space (1/6-em)]000 were used to prepare SPD-0.8, SPD-0.8-1800 and SPD-0.8-10[thin space (1/6-em)]000, respectively. Finally, 1.0777 g and 1.0182 g of SPD-0.8-1800 and SPD-0.8-10[thin space (1/6-em)]000 were obtained.
Table 1 Monomer feeding ratio of polymerizations
Samples Sulfur (M) PEI (M) DEB (M) Monomer weight (g) Product weight (g)
SPD-0.33 4 0.33 1 1.25 0.9326
SPD-0.5 4 0.5 1 1.25 0.9498
SPD-0.7 4 0.7 1 1.25 0.7642
SPD-0.8 4 0.8 1 1.25 0.6754
SPD-0.9 4 0.9 1 1.25 0.6188

With the increase of the PEI concentration from 0.33 M to 0.9 M, the final product changed from hard solid to elastic solid. This was because with the increase of the PEI concentration the relative content of sulfur decreased, and the low amount of sulfur induced the low crosslinking density. Meanwhile, the weight of products decreased as the PEI concentration increased.

The structures of SPDs were characterized by FT-IR, 1H-NMR, and solid-state 13C-NMR. In the FT-IR spectra (Fig. 1A), the absorbance at 2102 and 3262 cm−1 is attributed to the stretching vibrations of C[triple bond, length as m-dash]C and [triple bond, length as m-dash]C–H of DEB, respectively. After the reaction, the absorbance at 2102 and 3262 cm−1 disappeared. In the spectra of all SPD samples, a wide peak with the centre around 3277 cm−1 assigned to the NH2 and –NH– stretching vibrations of PEI, the absorbance at 1082 cm−1 belonging to the C[double bond, length as m-dash]S bond, and the peak at 1650 cm−1 attributed to the deformation vibration of –NH– groups beside C[double bond, length as m-dash]S groups were observed. These data strongly suggested the formation of thioamide groups.16,17

image file: c9py01544b-f1.tif
Fig. 1 (A) FT-IR spectra of DEB, PEI monomers and SPDs; (B) 13C-NMR spectra of DEB and PEI in DMSO-d6; (C) solid-state 13C-NMR spectra of SPDs; (D) 1H-NMR spectra of PEI, S + PEI and S + PEI + D2O. The solvent and water peaks are marked with asterisks.

Further proof for the formation of C[double bond, length as m-dash]S can be found in the 13C-NMR spectra (Fig. 1B and C). In the 13C-NMR spectra of DEB and PEI (Fig. 1B), the peaks at δ 132.36, 122.62 and 83.26 represent the benzene ring and acetylene carbons (C[triple bond, length as m-dash]C) of DEB, respectively, and peaks between δ 40–60 represent the CH2 carbons of PEI.

In the solid-state 13C-NMR spectra of SPDs (Fig. 1C), the peaks of CH2 carbons and benzene rings were verified easily at 40–60 ppm and 160–180 ppm, when compared with the spectra of DEB and PEI (Fig. 1B). In addition, the disappearance of the C[triple bond, length as m-dash]C peak at δ 83.26, the persistent presence of the benzene rings, and the newly emerged peaks associated with the C[double bond, length as m-dash]S groups around δ 170[thin space (1/6-em)]16–18 strongly indicated the reaction of DEB and the formation of thioamide groups.

According to the previous reports, the thioamide group can be formed by the direct reaction of primary amines and elemental sulfur.14,17 As a control to confirm the reaction between sulfur and PEI, elemental sulfur and PEI (4[thin space (1/6-em)]:[thin space (1/6-em)]0.9, mol[thin space (1/6-em)]:[thin space (1/6-em)]mol) were mixed directly without a solvent at 100 °C in a 20 mL glass vial. As soon as the sulfur was added into the glass vial, light yellow PEI turned into homogeneous deep red, and the fluid was too viscous to be magnetically stirred in 5 minutes, indicating a quick reaction between sulfur and PEI. After cooling down, the deep red liquid solidified, which was ground into powder (Fig. S3) and is denoted as S + PEI.

As shown in Fig. 1D (1H-NMR spectra of PEI and S + PEI in DMSO-d6), the protons of primary and secondary amines of PEI resonated at δ 1.61, disappeared after the reaction, and new two peaks appeared at δ 8.01 and 5.31, respectively. With one drop of D2O added into DMSO-d6, the peaks at δ 8.01 and 5.31 disappeared, indicating that the peak at δ 8.01 was the proton of –NH– next to the C[double bond, length as m-dash]S group.16–18

For further investigation, the surface chemical properties of SPD-0.8 were characterized by XPS (vide infra, section 3.2). The high-resolution spectrum of S 2p can be fitted by two peaks at 161.32 and 163.13 eV (Fig. 7C), assigned to C[double bond, length as m-dash]S and –S–S–, respectively, and the high-resolution spectrum of N 1s can be fitted by two peaks at 398.85 and 400.71 eV (Fig. 7D), which were attributed to N in the NH/NH2 and tertiary amine (>N–), respectively.

According to the above experimental results, it can be confirmed that we have successfully synthesised hyperbranched polythioamides in which thioamide groups were formed in two different ways. One was formed by the reaction of sulfur and PEI primary amine,14,17 while the second kind was formed by the reaction of both primary and secondary amines of PEI with S and diyne.15,16 The possible chemical structure of SPD was shown in Scheme 1, and there are three types of C[double bond, length as m-dash]S groups (α, β and γ). α groups were formed by the direct reaction of PEI and sulfur, and β and γ were formed by the reaction of primary and secondary amines of PEI with sulfur and DEB, respectively.

image file: c9py01544b-s1.tif
Scheme 1 The possible chemical structure of SPD.

Thermal characterization of SPDs, S + PEI, and monomers was performed by TG, DTG, and DSC. As shown in Fig. 2A–C, PEI degraded at around 355 °C, elemental sulfur melted at 123 °C and the decomposition of polysulfide was observed at 346 °C, and DEB melted at 99 °C and decomposed at 142 °C. The melting endotherm of sulfur and DEB was not visible in SPDs and S + PEI (Fig. 2C and F), indicating the complete removal of unreacted sulfur and DEB by DMSO and methanol. In addition, S + PEI solids also can be removed by DMSO and methanol if they were produced during the reaction of sulfur, PEI and DEB.

image file: c9py01544b-f2.tif
Fig. 2 TG (A and D), DTG (B and E), and DSC (C and F) curves of SPDs, monomers and S + PEI.

As shown in Fig. 2D–F, S + PEI had two weight loss stages. Apparently, the weight loss around 300 °C corresponds to the degradation of the PEI main chain. Hence we can be certain that the weight loss around 190 °C was related to the new products produced by PEI and sulfur (could be –C([double bond, length as m-dash]S)NH– groups).

Interestingly, it was found that there were two obvious DTG peaks with the corresponding endothermic peaks of SPDs around 160 °C and 184 °C (Fig. 2E and F). The peak at 184 °C was caused by the decomposition of –C([double bond, length as m-dash]S)NH– formed by sulfur and PEI, so the other peak at 160 °C was supposed to be caused by the decomposition of the –C([double bond, length as m-dash]S)NH–/–C([double bond, length as m-dash]S)N< groups formed by sulfur, PEI and DEB monomers.

With the doubt that if the peak at 160 °C in the DTG curve was caused by the product produced from the reaction between elemental sulfur and DEB, they were reacted and washed under the same conditions with SPD-0.8 except the addition of PEI. As a result, there was a precipitate produced in pyridine after two hours, and denoted as S + DEB. In contrast, DEB was completely dissolved in pyridine and could not self-polymerize under the same conditions.

As shown in the FT-IR spectra (Fig. S4A), compared to the DEB monomer and sulfur, the new peaks of S + DEB at 921 cm−1 and 754 cm−1 were supposed to be due to the C–S stretching, demonstrating the formation of the C–S chemical bond.3,29 DTG and DSC curves of S + DEB (Fig. S4C and D) showed that no peak appeared between 160–195 °C, and, at the same time, there was a broad DTG peak around 284 °C and a relative endothermic peak around 301 °C. The same peak also exists in the DTG curve of SPD-0.8 around 284 °C, however it is hard to figure out the relative peaks in TG and DSC curves, indicating that a small amount of S + DEB was contained in SPD-0.8. What is more, the two DTG and DSC peaks around 160 °C and 190 °C of SPDs were not contributed by S + DEB. The elemental analysis results of S + DEB are shown in Table S1; the elemental sulfur copolymerized with DEB or initiated the self-polymerization of DEB in pyridine at 100 °C to form S + DEB, and the structure of S + DEB was similar to that in the work reported by Sun et al.3 and will not be discussed further in this work.

The adsorption capacity of SPDs was measured (Table 2); in general, more holes could provide more adsorption sites on the surface and inside the adsorbent to enhance the adsorption capacity. Here, the adsorption–desorption curve and pore size distribution of SPDs were also measured using N2 adsorption–desorption isotherms.

Table 2 Surface areas, pore volumes, pore sizes and adsorption capacities of 10 mg of SPDs after being immersed in 10 mL 1000 mg L−1 of Hg2+ solution for 3 h at 30 °C
Entry Samples S BET (m2 g−1) Pore volume (cm3 g−1) Pore size (nm) Adsorption efficiency (%) Adsorption capacity (mg g−1)
1 SPD-0.33 4.54 0.0034 3.02 45.30 414.58
2 SPD-0.5 2.21 0.0015 2.72 83.34 762.77
3 SPD-0.7 3.78 0.0042 4.43 86.55 792.20
4 SPD-0.8 2.99 0.0029 3.93 91.51 837.57
5 SPD-0.9 3.50 0.0025 2.89 84.07 769.44
8 SPD-0.8–1800 3.85 0.0047 4.89 82.17 752.07
9 SPD-0.8-10[thin space (1/6-em)]000 2.72 0.0035 5.11 75.72 693.05

As shown, there is no obvious relationship between the specific surface areas, pore volumes, pore sizes and adsorption capacities. With the increase of the PEI content from SPD-0.33 to SPD-0.9, the adsorption capacity first increased to 837.57 mg g−1 and then decreased; at the same time, the specific surface area of SPD-0.8 is only 2.99 m2 g−1, and it is not the maximum specific surface area among the SPDs. This suggests that the specific surface area, pore volume and pore size are not the most critical factors affecting the adsorption capacity of SPD.

The SEM images of the SPDs were used to reveal the surface morphology of the samples (Fig. 3, Fig. S5). As can be seen from the surface of a SPD-0.5 particle (Fig. 3A), the pore size distribution is uneven, and some areas even have no holes. Moreover, as shown in Fig. S5, after observing multiple particles, it was found that there were few particles with a porous structure.

image file: c9py01544b-f3.tif
Fig. 3 SEM images of SPD-0.5 (A), SPD-0.33 (B and b), SPD-0.5 (C and c), SPD-0.7 (D and d), and SPD-0.8 (E and e) before (B–D) and after (b–d) being immersed in 2000 ppm Hg2+ solution.

Fig. 3B–E and b–e show the surface of SPDs before and after being immersed in 2000 ppm Hg2+ solution, respectively. It can be clearly seen that the surface of SPD particles changed greatly after being soaked in HgCl2 solution. Hg2+ ions adsorbed on the surface and formed tiny aggregative particles, making the smooth surface rough.

Furthermore, in order to investigate the elemental distribution of C, N, S and Hg, energy-dispersive X-ray spectroscopy (EDX) mapping was performed and the results are shown in Fig. 4. All elements displayed a homogeneous distribution on the surface of the sample, especially the Hg2+ ions adsorbed on the surface homogeneously (Fig. 4c).

image file: c9py01544b-f4.tif
Fig. 4 SEM image and the corresponding EDX mapping of SPD-0.7 before (A–D) and after (a–d) being immersed in 2000 ppm Hg2+ solution.

Elemental analyses of SPDs were conducted to further study whether the content of the element was related to the adsorption capacity (Table 3). As shown, the contents of H and N were both increased with the increase of PEI from SPD-0.33 to SPD-0.8, and then decreased. The variation of these two elements was the same with the changing trend of the adsorption capacity of the samples, indicating that the contents of N and H have an important effect on the adsorption capacity for Hg2+. On the other hand, the S content was decreased firstly and then increased, and it was the minimum in SPD-0.8. The lower S content means that fewer amines reacted with sulfur and there are more primary or secondary amines existing in the prepared SPD, which promote the adsorption of mercury.

Table 3 The elemental compositions of SPDs
Samples C (%) H (%) N (%) S (%)
SPD-0.33 46.88 6.128 13.60 20.218
SPD-0.5 46.36 6.339 14.52 18.326
SPD-0.7 44.90 6.502 15.38 17.325
SPD-0.8 46.47 6.716 16.38 14.796
SPD-0.9 44.49 6.670 15.62 15.522

For the maximum adsorption capacity, SPD-0.8 was selected for the next tests. Fig. 8 shows the isotherm and the pore size distribution of SPD-0.8. As can be seen in Fig. 5A, SPD-0.8 showed a typical type IV isotherm according to IUPAC classication,27 and the low adsorbed volume (2 cm3 g−1) indicated that there were very few pores in the system; moreover, Fig. 5B shows that the pore diameters of SPD-0.8 were mainly in the range of 3 to 5 nm.

image file: c9py01544b-f5.tif
Fig. 5 (A) N2 adsorption–desorption isotherm, and (B) pore width distribution of SPD-0.8 obtained by the DFT method.

3.2 Adsorption mechanism

In order to investigate the adsorption process, the pseudo-first-order, and the pseudo-second-order kinetics were used to analyse the experimental data. As displayed in Fig. 6A and Table 4, the pseudo-second-order-model provided better correlation with the R2 = 0.9986 and the calculated Qe from the pseudo-second-order model was 1259.84 mg g−1, close to the experimental value (1219.24 mg g−1). The results demonstrated that the adsorption of Hg(II) ions on SPDs can be well described by the pseudo-second-order kinetic model, indicating that the adsorption was controlled by the chemisorption process between SPD and mercury ions.22,25,27
image file: c9py01544b-f6.tif
Fig. 6 (A) Parameters of the kinetics model; (B) equilibrium adsorption isotherm of Hg2+ on SPD-0.8 (experimental conditions: 10 mg of SPD-0.8 was added into a 10 mL Hg2+ solution at a designated concentration after stirring for 3 h. Hg2+ initial concentrations (C0) = 200, 400, 800, 1000, 1400, 1600, 1800, 2000 mg L−1, respectively), (C–E) FT-IR spectra of SPDs before and after being immersed in 2000 ppm Hg2+ solution (pH = 5.0, adsorption time = 180 min).
Table 4 Fitting results of adsorption kinetics of Hg(II) on SPD-0.8
Pseudo-first-order Pseudo-second-order
Q e (mg g−1) k 1 (min−1) R 2 Q e (mg g−1) k 2 (g mg−1 min−1) R 2
1206.67 0.27 0.9832 1259.84 0.0003 0.9986

The effect of initial concentration on the adsorption capacity of SPD-0.8 was further investigated. As shown in Fig. 6B, with the increase of the initial Hg(II) concentration, the adsorption capacity increased, followed by a plateau. Langmuir, Freundlich and Langmuir–Freundlich isotherm models were used to fit the experimental equilibrium data.

The Freundlich model assumes nonideal adsorption on the heterogeneous surface and multilayer sorption, The Langmuir isotherm is used for the monolayer adsorption on a homogeneous surface, and the Langmuir–Freundlich isotherm suggests that the adsorption of an adsorbent toward a target is the synergistic effect of the monolayer adsorption and the multilayer adsorption.24,27,30

The fitting parameters were calculated and are listed in Table 5. As shown, compared to the Langmuir and Freundlich models, the adsorption behavior was better described by the Langmuir–Freundlich adsorption model due to the higher correlation coefficient (R2 = 0.9926), indicating that the monolayer and multilayer synergistic adsorption isotherms were more suitable to explain the adsorption of Hg2+ ions on SPD-0.8 and the maximum adsorption capacity was 1161.99 mg g−1, which was much higher than most of the traditional absorbents (Table 6).

Table 5 Fitting results of Langmuir, Freundlich and Langmuir–Freundlich isotherm models for the adsorption of Hg(II) on SPD-0.8
Langmuir model Freundlich model Langmuir–Freundlich model
Q m (mg g−1) k L (L mg−1) R 2 k F (mg g−1) (L mg−1)1/n n F R 2 Q m (mg g−1) k LF (L mg−1) n LF R 2
1300.72 0.03 0.9478 225.92 3.54 0.8109 1161.99 0.0016 0.4902 0.9926

Fig. 6C–E are the FT-IR spectra of SPDs before and after being immersed in 2000 ppm Hg2+ solution. As shown, the –NH– peaks at 3277 cm−1 disappeared after being immersed in Hg2+ solution, and the C[double bond, length as m-dash]S peaks at 1082 cm−1 became broader than those of SPDs before coming in contact with Hg2+, indicating the formation of strong binding interactions between Hg2+ and –NH–, C[double bond, length as m-dash]S groups.

Table 6 Comparison of the maximum adsorption capacities of different kinds of adsorbents for Hg(II) removal
Adsorbents t e (h) Adsorption capacity (mg g−1) Ref.
Poly (sulfur-Meldrum's acid-styrene) compound 3 52 8
Ethylenediamie-thiourea crosslinked chitosan 48 217.1 22
PEI (Mw = 70[thin space (1/6-em)]000) functionalized carboxymethyl chitosan 6 1584 27
Polyaniline/attapulgite composite 24 800 31
Porous organic polymer-based mercury ‘nano-trap’ 12 1014 32
SPD-0.8 3 1161.99 This work

As shown in Fig. 7, XPS spectra of both survey and high resolution scans for the key elements on the SPD-0.8 surface before and after Hg2+ ion contact were studied. The characteristic peaks were observed at 161, 284, 400, and 101 eV which correspond to the S 2p, C 1s, N 1s, and Hg 4f, respectively.21,27,28,33

image file: c9py01544b-f7.tif
Fig. 7 (A) XPS survey spectra of SPD-0.8; (B) high-resolution spectra of Hg 4f and (C) S 2p, (D) N 1s before and after being immersed in 2000 ppm Hg2+ solution.

As can be seen in the survey spectra (Fig. 7A), a new peak for Hg 4f at a binding energy of 101 eV was observed after mercury contact, as shown in the high-resolution spectra of Hg 4f in Fig. 7B; two typical strong peaks (100.64 eV and 104.58 eV) were assigned to Hg 4f7/2 and Hg 4f5/2, respectively, confirming the adsorption of mercury by the adsorbent. In addition, it was also confirmed that mercury was in ionic form with no precipitation.22

Fig. 7C shows the high-resolution spectrum of S 2p, before the adsorption of mercury, which can be fitted by two peaks at 161.32 and 163.13 eV, assigned to C[double bond, length as m-dash]S and –S–S–, respectively, and the –S–S– is the bond of polysufide formed by the reaction of sulfur and DEB.6,11,34–36

After the adsorption, there was a new peak that appeared at the bonding energy of 164.64 eV, and at the same time, the two peaks at 161.32 and 163.13 eV were shifted to a higher bonding energy of 161.63 and 163.15 eV, respectively. According to the above results and the changing of the peak areas, it can be concluded that the peak of the C[double bond, length as m-dash]S group at 161.32 eV was divided into two peaks at bonding energies of 161.63 and 164.64 eV after the adsorption of Hg2+, which were assigned to unreacted and reacted C[double bond, length as m-dash]S groups with Hg2+ ions to form S-metal bonds, respectively. The bonding energy of S 2p shifted to a higher value due to the donation of the electrons from S atoms of C[double bond, length as m-dash]S bonds to Hg2+.23,37

Fig. 7D shows the high-resolution spectrum of N 1s; similar to the spectrum of S 2p, there were two peaks fitted at 398.85 and 400.71 eV before SPD-0.8 was immersed in mercury solution, which were attributed to N in the NH/NH2 and tertiary amine (>N–), respectively.23,33,37,38 After the adsorption of Hg2+, a new peak was observed at the binding energy of 401.93 eV, indicating that the other adsorption mechanism was the chelating interaction between NH/NH2 and Hg2+, in which a lone pair of electrons in the N atom was donated to the shared bond between NH/NH2 and Hg2+ and, hence, increasing the N 1s binding energy.23,37Fig. 8 shows schematically the proposed chelation of Hg(II) with SPD-0.8. A stable chelating structure forms between Hg2+ ions, NH/NH2 and C[double bond, length as m-dash]S groups in SPD-0.8.

image file: c9py01544b-f8.tif
Fig. 8 Proposed mode for the complexation of Hg2+ with C[double bond, length as m-dash]S and NH/NH2 groups of SPD-0.8.

3.3. Adsorbent reusability

The adsorbed Hg2+ ions on SPD-0.8 were desorbed with 2 mol L−1 HNO3. After shaking at 30 °C for 5 h, the suspension was filtered and washed with distilled water several times. Then, the adsorbent was dried in a vacuum oven at 50 °C overnight and reused in the next cycle. After repeating the procedure 5 times, an adsorption efficiency of 80% could still be obtained (Fig. 9). This result indicates that SPD-0.8 is stable, reusable and a potential material for Hg2+ ion removal.
image file: c9py01544b-f9.tif
Fig. 9 The adsorption efficiency of SPD-0.8 for Hg2+ ions during reusability cycles (C0 = 2000 mg L−1 in each cycle).

4 Conclusions

For the purpose of consuming a large amount of sulfur byproduct, a new kind of mercury adsorbent was prepared using elemental sulfur, DEB and PEI as raw materials in this work. Different from traditional PEI modified biomass or other materials, PEI acted as the mainchain and was crosslinked by sulfur and DEB by forming three kinds of thioamides. In our research, the maximum adsorption capacity of SPD-0.8 was as high as 1161.99 mg g−1, exhibiting greater potential for Hg2+ removal than most of the PEI modified biomass materials and polythioamides. The adsorption process was well fitted in the pseudo-second order kinetic process and the Langmuir–Freundlich isotherm model. XPS measurements demonstrated that mercury ions were adsorbed by the NH/NH2 and C[double bond, length as m-dash]S groups through chelation. The successful application of elemental sulfur as a raw material to prepare highly efficient mercury removal adsorbents could both utilize the sulfur byproduct and provide a new strategy to design PEI based mercury removal materials.

Conflicts of interest

There are no conflicts to declare.


This work was financially supported by the West Light Foundation of The Chinese Academy of Sciences (2017-XBQNXZ-B-001), Young Elite Scientist Sponsorship Program by CAST (2017QNRC001), the National Natural Science Foundation of China (21464015, 21472235), the Xinjiang Tianshan Talents Program (2018xgytsyc2-3) and the “One Thousand Talents”-Xinjiang Program.


  1. W. J. Chung, J. J. Griebel, E. T. Kim, H. Yoon, A. G. Simmonds, H. J. Ji, P. T. Dirlam, R. S. Glass, J. J. Wie, N. A. Nguyen, B. W. Guralnick, J. Park, A. Somogyi, P. Theato, M. E. Mackay, Y.-E. Sung, K. Char and J. Pyun, Nat. Chem., 2013, 5, 518–524 CrossRef CAS PubMed.
  2. D. A. Boyd, Angew. Chem., Int. Ed., 2016, 55, 15486–15502 CrossRef CAS.
  3. Z. Sun, M. Xiao, S. Wang, D. Han, S. Song, G. Chen and Y. Meng, J. Mater. Chem. A, 2014, 2, 9280–9286 RSC.
  4. S. Huang, R. Guan, S. Wang, M. Xiao, D. Han, L. Sun and Y. Meng, Prog. Polym. Sci., 2019, 89, 19–60 CrossRef CAS.
  5. A. Hoefling, Y. J. Lee and P. Theato, Macromol. Chem. Phys., 2017, 218, 1600303 CrossRef.
  6. M. J. H. Worthington, R. L. Kucera, I. S. Albuquerque, C. T. Gibson, A. Sibley, A. D. Slattery, J. A. Campbell, S. F. K. Alboaiji, K. A. Muller, J. Young, N. Adamson, J. R. Gascooke, D. Jampaiah, Y. M. Sabri, S. K. Bhargava, S. J. Ippolito, D. A. Lewis, J. S. Quinton, A. V. Ellis, A. Johs, G. J. L. Bernardes and J. M. Chalker, Chemistry, 2017, 23, 16219–16230 CrossRef CAS PubMed.
  7. N. A. Lundquist, M. J. H. Worthington, N. Adamson, C. T. Gibson, M. R. Johnston, A. V. Ellis and J. M. Chalker, RSC Adv., 2018, 8, 1232–1236 RSC.
  8. H.-K. Lin, Y.-S. Lai and Y.-L. Liu, ACS Sustainable Chem. Eng., 2019, 7, 4515–4522 CrossRef CAS.
  9. M. J. H. Worthington, R. L. Kucera and J. M. Chalker, Green Chem., 2017, 19, 2748–2761 RSC.
  10. H.-K. Lin and Y.-L. Liu, Macromol. Rapid Commun., 2017, 38, 1700051 CrossRef PubMed.
  11. J. J. Griebel, S. Namnabat, E. T. Kim, R. Himmelhuber, D. H. Moronta, W. J. Chung, A. G. Simmonds, K.-J. Kim, J. van der Laan, N. A. Nguyen, E. L. Dereniak, M. E. Mackay, K. Char, R. S. Glass, R. A. Norwood and J. Pyun, Adv. Mater., 2014, 26, 3014–3018 CrossRef CAS PubMed.
  12. M. Mann, J. E. Kruger, F. Andari, J. McErlean, J. R. Gascooke, J. A. Smith, M. J. H. Worthington, C. C. C. McKinley, J. A. Campbell, D. A. Lewis, T. Hasell, M. V. Perkins and J. M. Chalker, Org. Biomol. Chem., 2019, 17, 1929–1936 RSC.
  13. M. H. Habibi, S. Tangestaninejad, A. Fallah-Shojaie, I. Mohammadpoor-Baltork, S. F. Tayyari, G. Emtiazi and R. Hamidimotlagh, J. Coord. Chem., 2005, 58, 955–962 CrossRef CAS.
  14. T. B. Nguyen, L. Ermolenko and A. Al-Mourabit, Org. Lett., 2012, 14, 4274–4277 CrossRef CAS PubMed.
  15. T. B. Nguyen, M. Q. Tran, L. Ermolenko and A. Al-Mourabit, Org. Lett., 2014, 16, 310–313 CrossRef CAS PubMed.
  16. W. Li, X. Wu, Z. Zhao, A. Qin, R. Hu and B. Z. Tang, Macromolecules, 2015, 48, 7747–7754 CrossRef CAS.
  17. Z. Sun, H. Huang, L. Li, L. Liu and Y. Chen, Macromolecules, 2017, 50, 8505–8511 CrossRef CAS.
  18. T. Tian, R. Hu and B. Z. Tang, J. Am. Chem. Soc., 2018, 140, 6156–6163 CrossRef CAS PubMed.
  19. K. Kaeriyama, T. Rokusha and K. Naomi, J. Appl. Polym. Sci., 1978, 22, 3075–3080 CrossRef CAS.
  20. S. Kagaya, H. Miyazaki, M. Ito, K. Tohda and T. Kanbara, J. Hazard. Mater., 2010, 175, 1113–1115 CrossRef CAS PubMed.
  21. L. Bai, H. Hu, W. Fu, J. Wan, X. Cheng, L. Zhuge, L. Xiong and Q. Chen, J. Hazard. Mater., 2011, 195, 261–275 CrossRef CAS PubMed.
  22. W. Liang, M. Li, Z. Zhang, Y. Jiang, M. K. Awasthi, S. Jiang and R. Li, Int. J. Biol. Macromol., 2018, 113, 106–115 CrossRef CAS PubMed.
  23. X. Yao, H. Wang, Z. Ma, M. Liu, X. Zhao and D. Jia, Chin. J. Chem. Eng., 2016, 24, 1344–1352 CrossRef CAS.
  24. Y. Shi, T. Zhang, H. Ren, A. Kruse and R. Cui, Bioresour. Technol., 2018, 247, 370–379 CrossRef CAS PubMed.
  25. L. Qin, Y. Y. Ge, B. W. Deng and Z. L. Li, J. Taiwan Inst. Chem. Eng., 2017, 71, 84–90 CrossRef CAS.
  26. H. Ge, H. Huang, M. Xu and Q. Chen, Cellulose, 2016, 23, 2527–2537 CrossRef CAS.
  27. H. Zeng, L. Wang, D. Zhang, P. Yan, J. Nie, V. K. Sharma and C. Wang, Chem. Eng. J., 2019, 358, 253–263 CrossRef CAS.
  28. Y. Jiang, Z. Zhang, Y. Zhang, X. Zhou, L. Wang, A. Yasin and L. Zhang, RSC Adv., 2018, 8, 42405–42414 RSC.
  29. H.-K. Lin and Y.-L. Liu, Macromol. Rapid Commun., 2018, 39, 1700832 CrossRef PubMed.
  30. D. Xu, W. D. Wu, H.-J. Qi, R.-X. Yang and W.-Q. Deng, Chemosphere, 2018, 196, 174–181 CrossRef CAS PubMed.
  31. H. Cui, Y. Qian, Q. Li, Q. Zhang and J. Zhai, Chem. Eng. J., 2012, 211–212, 216–223 CrossRef CAS.
  32. B. Y. Li, Y. M. Zhang, D. X. Ma, Z. Shi and S. Q. Ma, Nat. Commun., 2014, 5, 7 Search PubMed.
  33. S. Deng and Y.-P. Ting, Water Res., 2005, 39, 2167–2177 CrossRef CAS PubMed.
  34. T. Hasell, D. J. Parker, H. A. Jones, T. McAllister and S. M. Howdle, Chem. Commun., 2016, 52, 5383–5386 RSC.
  35. M. P. Crockett, A. M. Evans, M. J. H. Worthington, I. S. Albuquerque, A. D. Slattery, C. T. Gibson, J. A. Campbell, D. A. Lewis, G. J. L. Bernardes and J. M. Chalker, Angew. Chem., Int. Ed., 2016, 55, 1714–1718 CrossRef CAS PubMed.
  36. J. Cao, C. Chen, Q. Zhao, N. Zhang, Q. Lu, X. Wang, Z. Niu and J. Chen, Adv. Mater., 2016, 28, 9629–9636 CrossRef CAS PubMed.
  37. X. Huang, X. Cao, W. Wang, H. Zhong and Z. Cao, J. Dispersion Sci. Technol., 2018, 39, 1316–1323 CrossRef CAS.
  38. S. Deng, R. Bai and J. P. Chen, Langmuir, 2003, 19, 5058–5064 CrossRef CAS.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c9py01544b

This journal is © The Royal Society of Chemistry 2020