Polymeric ionic liquid gels composed of hydrophilic and hydrophobic units for high adsorption selectivity of perrhenate

Abstract

The removal of TcO₄⁻ from aqueous solutions has attracted more and more attention recently, and ReO₄⁻ has been widely used as its natural analog. In this work, polymeric ionic liquid gel adsorbents, PC₂-C₁₂vimBr, with high adsorption capacity and selectivity towards ReO₄⁻ were synthesized by radiation-induced polymerization and crosslinking. PC₂-C₁₂vimBr was composed of two monomers: a hydrophobic unit, 1-vinyl-3-dodecylimidazolium bromide for high selectivity, and a hydrophilic unit, 1-vinyl-3-ethylimidazolium bromide for improved kinetics. A gel fraction up to 90% could be achieved under 40 kGy with varied monomer ratios. The adsorption of PC₂-C₁₂vimBr gels for ReO₄⁻ was evaluated by batch adsorption. The PC₂-C₁₂vimBr gel containing 20 mol% hydrophilic unit (named PC₂-C₁₂vimBr-A) could significantly improve the adsorption kinetics, which had an equilibrium time of ca. 24 h. The adsorption capacity obtained from the Langmuir model was 559 mg g⁻¹ (Re/gel). The selective factor against NO₃⁻ was 33.4 ± 1.9, which was more than 10 times higher than that of PC₂vimBr, and it could maintain ReO₄⁻ uptake as high as 100 mg g⁻¹ in 0.5 mol kg⁻¹ HNO₃. The ΔH^Θ and ΔS^Θ of the NO₃⁻/ReO₄⁻ ion-exchange reaction of PC₂-C₁₂vimNO₃-A were −16.9 kJ mol⁻¹ and 29 J mol⁻¹ K⁻¹, respectively, indicating physical adsorption. The adsorption mechanism of ReO₄⁻ onto PC₂-C₁₂vimBr-A gel was ion-exchange, and it could be recovered using 5.4 mol kg⁻¹ HNO₃.

---

1. Introduction

Technetium (Tc) is an artificial element, and all its isotopes are radioactive. A large amount of ⁹⁹Tc is produced from nuclear plants nowadays, and exists in high-level liquid waste. Since Tc is hazardous and has high mobility in the environment, and it may affect the geological disposal of other radioactive nuclei, the removal of Tc from aqueous solutions has attracted more and more attention recently.¹ And because of the radioactivity of Tc, it is not allowed to be handled in the normal lab, and rhenium (Re) has been extensively used as its analog owing to their similar chemistry properties.^1–3

Tc mainly exists in the form of TcO₄⁻ in the high-level waste.⁴ Many kinds of method to remove TcO₄⁻ and ReO₄⁻ have been investigated in recent years, including the reductive immobilization of TcO₄⁻ or ReO₄⁻ by Fe(0)^5,6 and the direct adsorption by adsorbents.^7–10 Comparing to the reductive method, the adsorption takes the advantage of high adsorption capacity and the reusability of the adsorbents.

Most adsorbents towards TcO₄⁻ and ReO₄⁻ depended on ion-exchange. MOFs,^11–14 carbon materials,^15–18 and organic adsorbents have been employed as the adsorbents. Organic adsorbents including resins,^19,20 grafted materials^21,22 and gels^23,24 take the advantage of high adsorption capacities and easy design and synthesis. Quaternary ammonium^19,23 including pyridinium^25,26 and imidazolium^3,21,27 have been used as the functional groups in many ion-exchange adsorbents.¹⁹ For the quaternary ammonium, alkyl has been widely used as the group on the quaternary N atom, including ethyl, propyl, butyl, pentyl, hexyl etc.^28,29 And it was found that longer alkyl chain in quaternary ammonium may bring higher selectivity towards ReO₄⁻.²⁸

However, adsorbents depending on ion-exchange would be influenced by the other anions in the solution as competitors, and a large amount of NO₃⁻ exists in the high-level liquid waste. So it is very important to develop adsorbents with high selectivities towards TcO₄⁻ or ReO₄⁻. Wang et al. reported a TcO₄⁻ adsorbent, NDTB-1, with an adsorption capacity of 162.2 mg g⁻¹ and a maximum distribution coefficient K_d of 1.0534 × 10⁴ mL g⁻¹.¹⁰ Shu et al. developed a surface ion-imprinted magnetic microsphere ReO₄⁻ adsorbent containing imidazolium groups with an adsorption capacity of 62.8 mg g⁻¹ and a Langmuir constant of 1.35 L mg⁻¹.³

On the other hand, some of the reported excellent adsorbents contain metal elements, e.g. Ag in SCU-100,¹³ which may result in the increase of radioactive wastes. Adsorbents composed of only C, H, N, O may avoid the secondary radioactive waste since they follow the CHON principle.^30,31 Organic adsorbents such as polymeric ionic liquids may meet this requirement.

Polymeric ionic liquid is a kind of polymer which contains ionic liquid units, and they have attracted much attention in many fields, such as polyelectrolytes, polymeric surfactants, anion sensitive smart materials,^32,33 gas separation membranes,³⁴ and also ion-exchange adsorbents.³⁵ Recently we reported a polymeric ionic liquid gel ReO₄⁻ adsorbent with high adsorption capacity of 865 mg g⁻¹, but the selectivity is not high enough, and we proposed that hydrophobic polycations might have higher selectivity towards ReO₄⁻.²⁴ Polymers with both hydrophilic and hydrophobic segments have been reported with interesting self-assembling behaviors³⁶ or as proton conductive membranes.³⁷ Considering the advantages of combination of the hydrophilic and the hydrophobic segments in polymers, i.e. the hydrophobic one is favour for higher adsorption selectivity towards ReO₄⁻ and the hydrophilic one is better for adsorption kinetics, we herein developed a polymeric ionic liquid gel adsorbent, PC₂-C₁₂vimBr, with high adsorption capacity and selectivity, which was synthesized easily by radiation-induced polymerization and crosslinking of 1-vinyl-3-alkylimidazolium. Considering longer alkyl chain on the quaternary N atom can increase the hydrophobicity of gels, which may result in higher adsorption selectivity towards ReO₄⁻, a hydrophobic unit, 1-vinyl-3-dodecylimidazolium bromide (C₁₂vimBr) was chosen to improve the selectivity. On the other hand, a hydrophilic unit, 1-vinyl-3-ethylimidazolium bromide (C₂vimBr), was chosen to improve the adsorption kinetics of ReO₄⁻ in aqueous solution. A series of PC₂-C₁₂vimBr gels with different molar ratio of hydrophobic and hydrophilic units were synthesized and characterized, and their adsorption kinetics towards ReO₄⁻ were investigated. Then the adsorption isotherm, selectivity and mechanism of PC₂-C₁₂vimBr gel were investigated in detail.

2. Experimental

2.1 Materials

C₂vimBr and C₁₂vimBr monomers were purchased from Lanzhou Greenchem ILs, LICP. CAS., dried in vacuum to remove the residual solvent. 3,3′-Divinyl-1,1′(1,6-hexanediyl) diimidazolium dibromide (C₆vim₂Br₂) was easily synthesized as we reported before.²⁴ 1,6-Dibromohexane was purchased from J&K, and 1-vinylimidazole was provided by TCI. KReO₄ was obtained from Alfa Aesar. Re standard solution was provided by NCS Testing Technology Co., Ltd. Some HNO₃ mother solutions with certain concentrations were prepared from GR HNO₃ (Xilong Chemical Co., Ltd.) and titrated. Other chemicals were analytical-grade reagents. Deionized water was used throughout the experiments.

2.2 Preparation and characterization of PC₂-C₁₂vimBr gels

Preparation of PC₂-C₁₂vimBr gels. C₂vimBr, C₁₂vimBr and C₆vim₂Br₂ were mixed into an aqueous solution and added into a glass test tube. It was then γ-irradiated at a dose rate of ca. 40 Gy min⁻¹ to a dose of 40 kGy by the ⁶⁰Co source, which was calibrated by Fricke dosimeter. The solution became into a white opaque gel after radiation. The concentration of C₆vim₂Br₂ crosslinker was set as 0.05 M. The total concentration of C₂vimBr and C₁₂vimBr monomers was 1 M, and their feed ratios and the short names of the resulted gels were shown in Table 1.

Table 1 Feed concentrations of monomers and the resulted gels

Feed concentrations (M)			Names of the resulted gels
C2vimBr	C12vimBr	C6vim2Br2
0	1	0.05	PC12vimBr
0.2	0.8	0.05	PC2-C12vimBr-A
0.4	0.6	0.05	PC2-C12vimBr-B
0.6	0.4	0.05	PC2-C12vimBr-C
0.8	0.2	0.05	PC2-C12vimBr-D

---

The resulted gels were cut into round pieces with a thickness of ca. 3 mm, and then every round piece was evenly cut into 4 sector pieces for further usage. The gels was immersed into excess ethanol for more than 6 times (at least 1 day for every time) to remove the sol part, and was subsequently put into excess water for more than 4 times. Then the gels were dried under 60 °C for more than 2 days to a constant weight. Gel fraction (GF) and equilibrium degree of swelling (EDS) were measured by gravimetry.

GF is defined as:

GF (%) = w_g/w₀ × 100 (1)

where w_g is the weight of dry gel after removing the sol part, and w₀ is the initial weight of dry gel.

EDS is defined as:

EDS = w_e/w_d (2)

where w_d is the dry weight of the gel and w_e is the weight of the swelled gel after swelling equilibrium.

Characterization of PC₂-C₁₂vimBr gels. IR spectra were obtained from Nicolet iS50. XPS was performed on Axis Ultra (Kratos Analytical) and B.E. were calibrated by C 1s hydrocarbon peak as 284.80 eV. Elemental Analysis (EA) was performed on Elementar Vario EL CUBE. SEM images were measured using Nova NanoSEM 430, and the samples were prepared by freeze-drying from gels well swelled in water.

2.3 Adsorption of ReO₄⁻ onto PC₂-C₁₂vimBr gels

The mother solution of Re was prepared from deionized water and KReO₄. Inductively coupled plasma atomic emission spectroscopy (ICP-AES, Prodigy) was used to measure the Re concentration, and the relative error of Re concentration was estimated as 2%.

Typically, an adsorption experiment was performed as following: a piece of gel was immersed into a Re solution, the mass ratio of the gel to the solution was 1 mg gel per 1 g solution. The system was shaked in a shaker at 25 °C for ca. 2 days to reach equilibrium, and the concentration of Re in the resulted solution was measured by ICP-AES.

The Re uptake, q, was calculated as following:

where c₀ is the concentration of Re before adsorption, and c is the concentration of Re after adsorption, while m_s is the mass of the solution, and m_g is the mass of the dry gel.

Adsorption kinetics. Since the gels were composed of imidazolium units with different molecular weights, sector pieces of gels with the similar sizes before washing were used to compare their adsorption kinetics. The gels were ca. 20–40 mg, and the corresponding m_s was constant at 50 g. 100 μL of solution was taken out from the system at a certain time and diluted to measure the Re concentration.

Adsorption isotherms and selectivity. For adsorption isotherms, a series of q and c were measured with varied c₀. For adsorption selectivity, NO₃⁻ and Br⁻, which were from HNO₃ and NaBr respectively, were chosen as anion competitors.

In the competitive adsorption against NO₃⁻, the anion of the gel were exchanged into NO₃⁻ before the adsorption, forming PC₂-C₁₂vimNO₃-A. In detail, PC₂-C₁₂vimBr-A was immersed into excess ca. 1.5 M HNO₃ to equilibrium for 2 times, then into water for 3 times and dried. XPS and IR were performed to confirm the formation of PC₂-C₁₂vimNO₃-A (Fig. S1†).

The selective factor (SF) was used to evaluate the selectivities toward ReO₄⁻ against NO₃⁻ and Br⁻. SF was defined as:³⁶

where q is the uptake of the corresponding ion, and should be converted into the same unit (mmol g⁻¹). And c is the remained concentration (converted into mol kg⁻¹) of the corresponding ion at equilibrium. Thus, SF is a dimensionless number and suitable for the comparison between different adsorbents.

The SF of PC₂vimBr, a Re adsorbent reported in our previous work,²⁴ towards ReO₄⁻ against NO₃⁻ was measured by the same procedure.

For the selective adsorption in ethanol–water mix solvent, the concentration of HNO₃ was 0.05 mol kg⁻¹.

Adsorption thermodynamics. In the solution, the c₀ was 700 ppm, and the concentration of HNO₃ was 0.05 mol kg⁻¹. The phase ratio was 1 mg gel per 1 g solution, and the solution was more than 60 g in each system. The system was shaken in a shaker for more than 2 days at each temperature, then 100 μL of solution was taken to measure the c. Less than 1% of the solution was taken out for measurement during the whole experiment to avoid the influence by the change of the solution volume.

Adsorption cycles. In the first cycle, a piece of PC₂-C₁₂vimBr-A was used to adsorb ReO₄⁻ (c₀ = 1176 ppm) and became PC₂-C₁₂vimReO₄-A after the adsorption. Then PC₂-C₁₂vimReO₄-A was desorbed by 5.40 mol kg⁻¹ HNO₃ (1 g per mg PC₂-C₁₂vimBr-A), and the solution was collected to measure the amount of desorbed Re. After that it was further immersed into 5.40 mol kg⁻¹ HNO₃ for another time and then into water for 3 times to get rid of the residual ReO₄⁻ and HNO₃. Thus PC₂-C₁₂vimNO₃-A was formed and used for the next cycle. After 4 cycles, IR and XPS were performed to the resulted PC₂-C₁₂vimNO₃-A to confirm the desorption, as well as to the PC₂-C₁₂vimReO₄-A in the first cycle to confirm the adsorption.

3. Result and discussion

3.1 Preparation and characterization of PC₂-C₁₂vimBr gels

Radiation-induced polymerization and crosslinking is a green and clean method widely used to prepare gels,³⁷ and it has been proved to be effective for the preparation of poly ionic liquid gels in our previous works.²⁴ The synthesis of PC₂-C₁₂vimBr gels was illustrated in Scheme 1, and the GF and the EDS of PC₂-C₁₂vimBr gels were shown in Fig. 1.

Scheme 1

Fig. 1 GF and EDS of the PC₁₂vimBr gel and PC₂-C₁₂vimBr gels.

The GFs of all PC₂-C₁₂vimBr gels with different compositions synthesized under 40 kGy were above 90%, which indicated that γ-radiation-induced polymerization and crosslinking was suitable and effective for this system. IR, TGA, EA, and XPS were performed to confirm the composition and structure of the resultant gels, and the results were shown in the ESI (Fig. S2–S4 and Table S1†).

PC₁₂vimBr, which had no C₂vimBr units, nearly did not swell in water, but it swelled well in ethanol. The EDS of PC₂-C₁₂vimBr gels in water increased with the increasing C₂vimBr content, while the EDS in ethanol decreased. Since all the five gels had the same crosslinker content, the differences of the EDS of the gels in water and in ethanol were due to the different contents of the short-chain C₂vimBr units and the long-chain C₁₂vimBr units, and the results reflected the hydrophilicities and hydrophobicities of the gels. The more C₂vimBr units the gel contains, the more hydrophilic it was, while the C₁₂vimBr units contributed to the hydrophobicity of the gel.

SEM images were taken from the xerogels prepared by freeze-drying of the corresponding gels swelled in water. SEM images of PC₂-C₁₂vimBr-A, PC₂-C₁₂vimBr-D and PC₁₂vimBr gel were shown in Fig. 2. Among the three samples, PC₂-C₁₂vimBr-D had the largest holes, and it was agreed with its highest EDS in water. PC₂-C₁₂vimBr-A, which had least C₂vimBr units among PC₂-C₁₂vimBr gels, also had small holes, while it could swell a little in water. In the image of PC₁₂vimBr, the section was rough and no hole could be observed. PC₁₂vimBr was relatively hydrophobic, and did not swell in water, while even a small amount of C₂vimBr units could give the gels hydrophilicity and make them swell in water, which would improve their adsorption kinetics.

Fig. 2 SEM images of (a) PC₂-C₁₂vimBr-D, (b) PC₂-C₁₂vimBr-A, and (c) PC₁₂vimBr.

3.2 The adsorption kinetics of ReO₄⁻ onto PC₂-C₁₂vimBr gels

PC₂-C₁₂vimBr gel had a cationic polymer network with Br⁻ as exchangeable counteranion, so anions such as ReO₄⁻ can be attracted easily by PC₂-C₁₂vimBr gel and then be adsorbed by anion-exchange. The adsorption kinetics of ReO₄⁻ onto four PC₂-C₁₂vimBr gels were shown in Fig. 3a, and that of PC₁₂vimBr was shown in Fig. 3b.

Fig. 3 Adsorption kinetics of ReO₄⁻ onto (a) PC₂-C₁₂vimBr gels and (b) PC₁₂vimBr. The c₀ was 557 ppm.

For all the four PC₂-C₁₂vimBr gels, the kinetics had no significant difference. The adsorption equilibriums were achieved at ca. 24 h, which was similar to the equilibrium time of poly(4-vinylpyridine) resins.²⁹ However, PC₁₂vimBr shown a quite slow adsorption kinetics, and could not achieve equilibrium in even 3 months. This was due to the high hydrophobicity of PC₁₂vimBr, and as shown in Fig. 2c, it had no porous structure in water. On the other hand, as little as 20 mol% of the short-chain C₂vimBr units could improve the hydrophilicity of the gel, and further efficiently improve the adsorption kinetics. Upon 20 mol%, more C₂vimBr units did not show significant improvement to the adsorption kinetics. Since we introduced the long-chain C₁₂vimBr units aiming at improving the adsorption selectivity towards ReO₄⁻ (the gels with lower C₁₂vimBr contents had lower adsorption selectivity towards ReO₄⁻, see Fig. S5†), PC₂-C₁₂vimBr-A, which had the highest content of C₁₂vimBr units among the gels with suitable kinetics, was chosen for the further adsorption experiments.

3.3 The adsorption isotherms and selectivity of PC₂-C₁₂vimBr-A towards ReO₄⁻

The adsorption isotherms. The adsorption isotherms of ReO₄⁻ onto PC₂-C₁₂vimBr-A was shown in Fig. 4.

Fig. 4 Adsorption isotherms of ReO₄⁻ onto PC₂-C₁₂vimBr-A.

The adsorption isotherms of ReO₄⁻ onto PC₂-C₁₂vimBr-A agreed with the Langmuir model well:

where q_m is the adsorption capacity, and k is the Langmuir constant.

And it was fitted in the linear form:

The parameters of fitting and the calculated q_m and k were shown in Table 2:

Table 2 Parameters of the adsorption of ReO₄⁻ onto PC₂-C₁₂vimBr-A from the Langmuir model

Intercept, 1/qmk	Slope, 1/qm	R^2	qm, (mg g^−1)	K, (ppm^−1)
0.02036	0.00179	0.9997	559	0.0879

---

The q_m of Re onto PC₂-C₁₂vimBr-A was 559 mg Re per g gel, which was equal to the theoretical value calculated from the N wt% in the EA of PC₂-C₁₂vimBr-A, assuming every imidazolium group could adsorb one ReO₄⁻ ion:

This result also supported that all imidazolium groups in PC₂-C₁₂vimBr-A were utilized with a ratio to ReO₄⁻ of 1 : 1.

Since the long-chain unit C₁₂vimBr has larger M_w than that of C₂vimBr, the q_m of PC₂-C₁₂vimBr-A was lower than that of the Re adsorbent PC₂vimBr we reported before, which was 865 mg g⁻¹.²⁴ However, as a gel adsorbent, PC₂-C₁₂vimBr-A was fully composed of functional groups without matrix, its adsorption capacity was higher than most of the Re adsorbents reported.^3,10,16

For the practical usage of Re/Tc adsorbents, the selectivity is more important than the adsorption capacities. Because in practical systems, TcO₄⁻ usually presences with competitors, e.g. HNO₃ in the PUREX.⁴ The results of the competitive adsorption against Br⁻ and NO₃⁻ were shown in Fig. 5.

Fig. 5 (a) The competitive adsorption towards ReO₄⁻ against NO₃⁻ on PC₂-C₁₂vimNO₃-A as well as on PC₂vimNO₃. (b) The competitive adsorption towards ReO₄⁻ against Br⁻ on PC₂-C₁₂vimBr-A. The c₀ was 700 ppm.

As shown in Fig. 5, the q of Re decreased with the increasing concentration of the competitor in all the three curves, and it was due to the competition of NO₃⁻ or Br⁻. SF should be a constant according to the Langmuir model, and was fluctuating around the average value among the investigated conditions. The average values of SF were listed in Table 3.

Table 3 The average values of the SF of PC₂-C₁₂vimNO₃-A, PC₂vimNO₃ and PC₂-C₁₂vimBr-A towards ReO₄⁻ against the corresponding competitors

Adsorbent	Competitor	SF
PC2-C12vimNO3-A	NO3^−	33.4 ± 1.9
PC2vimNO3	NO3^−	2.8 ± 0.5
PC2-C12vimBr-A	Br^−	46.9 ± 3.1

---

Comparing to PC₂vimNO₃, the adsorbent we previously reported,²⁴ PC₂-C₁₂vimNO₃-A had an SF against NO₃⁻ more than 10 times higher than that of the former one, which illustrated its excellent selectivity towards ReO₄⁻. Since few literatures about Re adsorbents have reported the SF between NO₃⁻ and ReO₄⁻, it is a little difficult to compare the selectivities widely among these adsorbents, considering some reported quantities, e.g. distribution coefficient K_d or removal rate, are not only decided by the nature of the materials, but also by the experiment parameters such as the phase ratio of adsorbent/solution and the c₀ of Re. We estimated the SF towards ReO₄⁻ against NO₃⁻ of SCU-100, which is an excellent Re/Tc adsorbent reported recently with the highest K_d among inorganic Re adsorbents, from its adsorption capacity as 541 mg ReO₄⁻ per g SCU-100 and removal rate as 73% at 0.15 mM ReO₄⁻ and 15 mM NO₃⁻ with a phase ratio of 1 mL solution per 1 mg SCU-100:¹³

It was illustrated that the measured SF of PC₂-C₁₂vimNO₃-A was higher than the estimated SF of SCU-100, i.e. PC₂-C₁₂vimNO₃-A had better selectivity towards ReO₄⁻. PC₂-C₁₂vimNO₃-A could remain ca. 100 mg g⁻¹ Re uptake under 0.5 mol kg⁻¹ HNO₃. It was even higher than the maximum adsorption capacities of some other reported adsorbents.^3,38 The potential practical usage of PC₂-C₁₂vimBr-A for the treatment of radioactive liquid wastes was revealed by such high Re uptake under this condition. The SF of PC₂vimBr against Br⁻ was higher than that against NO₃⁻, and it is agreed with the sequence of increasing hydration energies (ReO₄⁻ < NO₃⁻ < Br⁻).³⁹

Varied contents of polar organic solvents in the mixed solvent could be used to investigate the effect of solvent on the adsorption.⁴⁰ And the lower hydration energy of ReO₄⁻ than those of the competitors was widely used for the selective adsorption or extraction of ReO₄⁻.¹ Herein, ethanol was added in the solvent for illustrating the importance of water in the selective adsorption of ReO₄⁻. As shown in Fig. 6, the q as well as the SF were decreased with the increasing ethanol content. With increasing ethanol content, the activity of water decreased, thus the contribution of the differences of hydration energies to the selective adsorption was decreased for the system. So it could be concluded that water as a solvent played an important role in the selective adsorption of ReO₄⁻onto PC₂-C₁₂vimNO₃-A gel.

Fig. 6 The competitive adsorption towards ReO₄⁻ against NO₃⁻ on PC₂-C₁₂vimNO₃-A in ethanol–water mix solvent. The c₀ was 700 ppm.

3.4 The adsorption thermodynamics of PC₂-C₁₂vimBr-A towards ReO₄⁻

The SF of PC₂-C₁₂vimNO₃-A towards ReO₄⁻ against NO₃⁻ can be considered as the K of the following equation:

PC₂-C₁₂vimNO₃-A + ReO₄⁻ = PC₂-C₁₂vimReO₄-A + NO₃⁻ (9)

The K under different temperatures were measured, and the relation between ln K and 1000/T (Fig. 7) were fitted by the van't Hoff equation:

Fig. 7 Fitted curve of ln K–1000/T of eqn (9).

The ΔH^Θ and ΔS^Θ of eqn (9) were calculated from the slope and the intercept, which were −16.9 kJ mol⁻¹ and 29 J mol⁻¹ K⁻¹, respectively.

The negative ΔH^Θ indicated that the adsorption process was exothermic, and lower temperature was beneficial to the adsorption process. The type of adsorption can be indicated by the magnitude of the enthalpy change value. Lower enthalpy change corresponds to physical adsorption, and higher one corresponds to chemical adsorption. The range of enthalpy change of physical adsorption is considered as 0–20 kJ mol⁻¹ (ref. 41) or 0.5–5 kcal mol⁻¹ (2.1–20.9 kJ mol⁻¹).⁴² The absolute value of ΔH^Θ of eqn (9) was in the range of enthalpy change of physical adsorption, so the adsorption process of PC₂-C₁₂vimNO₃-A towards ReO₄⁻ was physical adsorption. This was agreed with the ion-exchange mechanism, where no covalent bond forms or breaks.

3.5 The adsorption cycles and the adsorption mechanism of PC₂-C₁₂vimBr-A towards ReO₄⁻

Though PC₂-C₁₂vimBr-A had high adsorption selectivity towards ReO₄⁻, ReO₄⁻ can be exchanged into solution by high concentration NO₃⁻. So the used adsorbent can be renewed by HNO₃ with high concentration resulting in the formation of PC₂-C₁₂vimNO₃-A. After washing off HNO₃ by deionized water, PC₂-C₁₂vimNO₃-A can adsorb ReO₄⁻ again by anion-exchanging. The adsorption cycles of PC₂-C₁₂vimBr-A towards ReO₄⁻ were shown in Fig. 8, where the Re uptake were marked by red and the Re desorbed in the first wash by 5.40 mol kg⁻¹ HNO₃ were marked by blue.

Fig. 8 The adsorption cycles of ReO₄⁻ on PC₂-C₁₂vimBr-A.

The q in the second cycle was reduced to ca. 96% of that of the first cycle and kept well in the next two cycles. The decrease of q after the first cycle could be explained by the anion exchange from Br⁻ to NO₃⁻ after the first cycle. PC₂-C₁₂vimReO₄-A could be renewed into PC₂-C₁₂vimNO₃-A by HNO₃ and the adsorbent had excellent reusability. Most ReO₄⁻ could be desorbed in the first wash by 5.4 mol kg⁻¹ NO₃⁻, and the second wash could renew almost all the ion-exchange positions. The eluted Re in HNO₃ solution exists without any other salts, and can be made into Re products for other usages easily, or disposed separately from other wastes.

IR and XPS were performed to characterize PC₂-C₁₂vimBr-A, PC₂-C₁₂vimReO₄-A and the PC₂-C₁₂vimNO₃-A after desorption (Fig. 9).

Fig. 9 The (a) IR and (b) XPS pattern of PC₂-C₁₂vimBr-A, PC₂-C₁₂vimReO₄-A and the PC₂-C₁₂vimNO₃-A after desorption.

As Br⁻ has no IR absorbance, the peaks of the polymeric imidazolium cation were revealed clearly in the IR spectrum of PC₂-C₁₂vimBr-A. After the adsorption of ReO₄⁻, in the IR spectrum of PC₂-C₁₂vimReO₄-A, no change of the peaks of the polymeric imidazolium cation could be observed, and the peak of ReO₄⁻ at 891 cm⁻¹ occurred. After the desorption by 5.4 mol kg⁻¹ HNO₃, the peak of ReO₄⁻ disappeared, and the strong peak of NO₃⁻ at 1325 cm⁻¹ occurred, while the other peaks showed no change. It could be concluded from the IR spectra that the polymeric imidazolium cation did not change during the adsorption and the desorption process, and ReO₄⁻ and NO₃⁻ were exchanged onto the polycation without valence change. The XPS also confirmed the adsorption and the desorption was successful. The peak of Re occurred in PC₂-C₁₂vimReO₄-A instead of that of Br in PC₂-C₁₂vimBr-A, and it disappeared in the XPS of PC₂-C₁₂vimNO₃-A after desorption. The peak of N in PC₂-C₁₂vimNO₃-A was a double peak, instead of the single peaks in PC₂-C₁₂vimBr-A and PC₂-C₁₂vimReO₄-A, for the N in PC₂-C₁₂vimNO₃-A could be divided into two kinds with different B.E., which belonged to NO₃⁻ and imidazolium respectively (see also Fig. S1†). So an ion-exchange mechanism could be concluded from the IR and the XPS results.

Furthermore, as an anion-exchange gel, PC₂-C₁₂vimBr may be used for the adsorption of other anions, such as Cr(VI) anions, anionic surfactants and anionic dyes in the future.

4. Conclusions

Polymeric ionic liquid gels, PC₂-C₁₂vimBr, was synthesized by radiation-induced polymerization and crosslinking, and its adsorption behavior towards ReO₄⁻ was investigated. The PC₂-C₁₂vimBr gel contained two kinds of imidazolium units, namely the hydrophilic units composed of ethylimidazolium and the hydrophobic units composed of dodecylimidazolium. Addition of small amount of hydrophilic units could effectively improve the adsorption kinetics of PC₁₂vimBr gel, and the addition of hydrophobic units could improve the adsorption selectivity towards ReO₄⁻ of PC₂vimBr gel. PC₂-C₁₂vimBr-A gel, with 20 mol% hydrophilic units and 80 mol% hydrophobic units, had an adsorption equilibrium time of ca. 24 h and adsorption capacity as high as 559 mg g⁻¹ (mg Re per g gel). Its SF towards ReO₄⁻ against NO₃⁻ was 33.4 ± 1.9, which was more than 10 times of that of PC₂vimBr, and it could keep a Re uptake as high as 100 mg g⁻¹ in 0.5 mol kg⁻¹ HNO₃, which revealed its potential application in the treatment of Tc in radioactive liquid wastes. The adsorption mechanism was ion-exchange, and the selective adsorption was driven by the low hydration energy of ReO₄⁻ and the hydrophobicity of the polycation. The ΔH^Θ and ΔS^Θ of the NO₃⁻/ReO₄⁻ ion-exchange reaction of PC₂-C₁₂vimNO₃-A were −16.9 KJ mol⁻¹ and 29 J mol⁻¹ K⁻¹, respectively. The adsorbent could be renewed by 5.4 mol kg⁻¹ HNO₃ for several times with high adsorption performance maintenance. The work provided a new approach to prepare novel ionic gel adsorbent by controlling the hydrophilicity and hydrophobicity of adsorbents which are related to the kinetics and selectivity towards adsorbates, respectively.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The National Natural Science Foundation of China (NNSFC, Project No. 21471161 and 11375019) and the Scientific and Technological Plan of Shenzhen City (JCYJ 20150616163111759) are acknowledged for supporting this research.

Notes and references

1. D. Banerjee, D. Kim, M. J. Schweiger, A. A. Kruger and P. K. Thallapally, Chem. Soc. Rev., 2016, 45, 2724–2739.
2. S. Sarri, P. Misaelides, D. Zamboulis, X. Gaona, M. Altmaier and H. Geckeis, J. Radioanal. Nucl. Chem., 2016, 307, 681–689.
3. X. Shu, L. Shen, Y. Wei and D. Hua, J. Mol. Liq., 2015, 211, 621–627.
4. I. Yamagishi and M. Kubota, J. Nucl. Sci. Technol., 1993, 30, 717–719.
5. B. A. Lenell and Y. Arai, J. Hazard. Mater., 2017, 321, 335–343.
6. T. L. Yu, S. Q. Liu, M. Xu, J. Peng, J. Q. Li and M. L. Zhai, Radiat. Phys. Chem., 2016, 125, 94–101.
7. J. J. Neeway, R. M. Asmussen, A. R. Lawter, M. E. Bowden, W. W. Lukens, D. Sarma, B. J. Riley, M. G. Kanatzidis and N. P. Qafoku, Chem. Mater., 2016, 28, 3976–3983.
8. C. D. Williams and P. Carbone, Environ. Sci. Technol., 2016, 50, 3875–3881.
9. S. A. Wang, E. V. Alekseev, D. W. Juan, W. H. Casey, B. L. Phillips, W. Depmeier and T. E. Albrecht-Schmitt, Angew. Chem., Int. Ed., 2010, 49, 1057–1060.
10. S. A. Wang, P. Yu, B. A. Purse, M. J. Orta, J. Diwu, W. H. Casey, B. L. Phillips, E. V. Alekseev, W. Depmeier, D. T. Hobbs and T. E. Albrecht-Schmitt, Adv. Funct. Mater., 2012, 22, 2241–2250.
11. D. Banerjee, S. K. Elsaidi, B. Aguila, B. Li, D. Kim, M. J. Schweiger, A. A. Kruger, C. J. Doonan, S. Ma and P. K. Thallapally, Chem.–Eur. J., 2016, 22, 17581–17584.
12. D. Banerjee, W. Xu, Z. Nie, L. E. V. Johnson, C. Coghlan, M. L. Sushko, D. Kim, M. J. Schweiger, A. A. Kruger, C. J. Doonan and P. K. Thallapally, Inorg. Chem., 2016, 55, 8241–8243.
13. D. Sheng, L. Zhu, C. Xu, C. Xiao, Y. Wang, Y. Wang, L. Chen, J. Diwu, J. Chen, Z. Chai, T. E. Albrecht-Schmitt and S. Wang, Environ. Sci. Technol., 2017, 51, 3471–3479.
14. A. V. Desai, B. Manna, A. Karmakar, A. Sahu and S. K. Ghosh, Angew. Chem., Int. Ed., 2016, 55, 7811–7815.
15. H. Hu, B. Jiang, H. Wu, J. Zhang and X. Chen, J. Environ. Radioact., 2016, 165, 39–46.
16. H. Hu, B. Jiang, J. Zhang and X. Chen, RSC Adv., 2015, 5, 104769–104778.
17. P. Rajec, O. Rosskopfova, M. Galambos, V. Fristak, G. Soja, A. Dafnomili, F. Noli, A. Dukic and L. Matovic, J. Radioanal. Nucl. Chem., 2016, 310, 253–261.
18. P. Rajec, M. Galambos, M. Dano, O. Rosskopfova, M. Caplovicova, P. Hudec, M. Hornacek, I. Novak, D. Berek and L. Caplovic, J. Radioanal. Nucl. Chem., 2015, 303, 277–286.
19. W. R. Wilmarth, G. J. Lumetta, M. E. Johnson, M. R. Poirier, M. C. Thompson, P. C. Suggs and N. P. Machara, Solvent Extr. Ion Exch., 2011, 29, 1–48.
20. C. Nash, B. Musall, M. Morse and D. McCabe, Sep. Sci. Technol., 2015, 50, 2881–2887.
21. P. Xiao, D. Han, M. L. Zhai, L. Xu and H. B. Li, J. Hazard. Mater., 2017, 324, 711–723.
22. J.-H. Zu, Y.-Z. Wei, M.-S. Ye, F.-D. Tang, L.-F. He and R.-Q. Liu, Nucl. Sci. Tech., 2015, 26, 69–75.
23. B. Gierczyk, M. Ceglowski and M. Zalas, PLoS One, 2015, 10, e0122891.
24. D. Han, X. Li, J. Peng, L. Xu, J. Li, H. Li and M. Zhai, RSC Adv., 2016, 6, 69052–69059.
25. J. Zu, M. Ye, P. Wang, F. Tang and L. He, RSC Adv., 2016, 6, 18868–18873.
26. W. Luo, A. Inoue, T. Hirajima and K. Sasaki, Appl. Surf. Sci., 2017, 394, 431–439.
27. M. Petrova, M. Guigue, L. Venault, P. Moisy and P. Hesemann, Phys. Chem. Chem. Phys., 2015, 17, 10182–10188.
28. B. H. Gu, G. M. Brown, P. V. Bonnesen, L. Y. Liang, B. A. Moyer, R. Ober and S. D. Alexandratos, Environ. Sci. Technol., 2000, 34, 1075–1080.
29. D. Jermakowicz-Bartkowiak and B. N. Kolarz, React. Funct. Polym., 2011, 71, 95–103.
30. D. Pan, G. Ye, F. Wang and J. Chen, Prog Chem., 2012, 24, 2167–2176.
31. S. Tachimori, S. Suzuki and Y. Sasaki, J. At. Energy Soc. Jpn., 2001, 43, 1235–1241.
32. J. Y. Yuan, D. Mecerreyes and M. Antonietti, Prog. Polym. Sci., 2013, 38, 1009–1036.
33. J. Y. Yuan and M. Antonietti, Polymer, 2011, 52, 1469–1482.
34. L. C. Tome and I. M. Marrucho, Chem. Soc. Rev., 2016, 45, 2785–2824.
35. Y. Jiang, F. Li, G. Ding, Y. Chen, Y. Liu, Y. Hong, P. Liu, X. Qi and L. Ni, J. Colloid Interface Sci., 2015, 455, 125–133.
36. Z. Dong, J. Z. Liu, W. J. Yuan, Y. P. Yi and L. Zhao, Chem. Eng. J., 2016, 283, 504–513.
37. J. Chen, Y. Y. Ao, T. R. Lin, X. Yang, J. Peng, W. Huang, J. Q. Li and M. L. Zhai, Polymer, 2016, 87, 73–80.
38. S. Milicevic, L. Matovic, D. Petrovic, A. Aukic, V. Milosevic, D. Dokic and K. Kumric, J. Radioanal. Nucl. Chem., 2016, 310, 805–815.
39. J. Behnsen and B. Riebe, Appl. Geochem., 2008, 23, 2746–2752.
40. R. Konasova, J. J. Dytrtova and V. Kasicka, J. Sep. Sci., 2016, 39, 4429–4438.
41. H. Demey, T. Vincent, M. Ruiz, A. M. Sastre and E. Guibal, Chem. Eng. J., 2014, 244, 576–586.
42. D. Liping, S. Yingying, S. Hua, W. Xinting and Z. Xiaobin, J. Hazard. Mater., 2007, 143, 220–225.

---

Footnotes

† Electronic supplementary information (ESI) available: IR, TGA, EA, and XPS of PC₂-C₁₂vimBr. See DOI: 10.1039/c8ra00838h

‡ These authors contributed equally to this work.

---