Synthesis and characterization of imidazolium poly(azolyl)borate ionic liquids and their potential application in SO₂ absorption

Abstract

Thirteen new 1,3-substituted imidazolium poly(azolyl)borate salts with the general formula [R₁R₂im][B(H)_4−n(azolyl)_n] (R₁ = methyl, n-butyl, 2-(diethylamino)ethyl; R₂ = methyl; azolyl = pyrazolyl, imidazolyl, 1,2,4-triazolyl, tetrazolyl; n = 2, 3) have been synthesized through salt metathesis of the corresponding imidazolium chloride and potassium poly(azolyl)borates. The newly synthesized dihydrobis(azolyl)borate and hydrotris(azolyl)borate salts are liquids at room temperature except two of the 1,3-dimethylimidazolium derivatives, [mmim][H₂B(pz)₂] and [mmim][HB(pz)₃]. Conductivity, thermal and physicochemical properties of the new borate ionic liquids were systematically investigated. With multiple azolyl groups serving as binding sites, these ionic liquids (ILs) generally exhibit high SO₂ absorption capacity (up to 5.8 mol mol⁻¹ of IL or 1.05 g g⁻¹ of IL for 1-methyl-3-n-butylimidazolium hydrotris(imidazolyl)borate, compound 13).

---

Introduction

Ionic liquids (ILs) have been attracting much attention from researchers because of their advantageous properties of high thermal stability, low vapour pressure, wide liquid temperature range, high electric conductivities, and versatile tunability, etc.^1–3 Functionalized ionic liquids are especially useful and they have found various applications in different fields. For example, they can be employed as reusable solvents or catalysts for reactions in synthetic chemistry;^4–6 they can also be used as electrolytes in batteries,^7–9 solar cells,¹⁰ capacitors,¹¹ and hypergolic fluids.^12,13

Particularly, a number of ILs have also been explored as gas separation materials which can capture acidic gases, such as SO₂, CO₂, NO_x, etc. from post-combustion flue gas.^14–20 There are actually a number of reasons for choosing ILs as the next generation acid gas sorbents. The currently employed scrubbing methods for removing SO₂ use one-time alkaline sorbents like the slurry of limestone or lime, which generate huge amounts of wastewater and waste salts. Reusable sorbents represented by aqueous amines are commonly used for CO₂ removal; however, their high water content accounts for a large part of sorbent regeneration energy, let alone their other drawbacks including highly exothermic reaction enthalpy (even higher for SO₂ (ref. 16)), low thermal stability, high corrosivity, etc.¹⁵ In short, reversibility, high capture capacity (on a per molar/mass basis), and many other above mentioned desirable properties make ILs promising sorbents, which are in theory guaranteed by their virtually unlimited structural tunability.

Poly(azolyl)borates (Fig. 1a) are typically known as the scorpionate ligands in metal complexes.²¹ The first report on this type of ligands appeared in 1966,²² when Trofimenko successfully synthesized hydrotris(pyrazolyl)borate. After that, various poly(azolyl)borates^23–27 have been studied in the past fifty years and until very recently, a full family of these ligands have been completely developed with the final two reports from the Winter's²⁸ and Jenkins'²⁹ groups. With a tetrahedral boron center bearing five-membered N-heterocyclic groups, the poly(azolyl)borate ligands can be structurally and electronically tuned, potentially enabling the constituted metal complexes with desired properties. In addition to the research on poly(azolyl)borates as coordination ligands, Shreeve et al.³⁰ have reported several azolium poly(1,2,4-triazolyl)borate salts which belong to the category of room-temperature ILs, and these ILs could be energetic materials as indicated by the positive heat of formation (109 kJ mol⁻¹) of 1,2,4-triazole.

Fig. 1 Structural illustration of (a) poly(azolyl)borates and (b) imidazolium cations.

Previously, we have theoretically investigated the effects of poly(azolyl)borate ligands on the coordinated metal center, and have found that the electronic structure of the metal center can be accordingly changed by tuning the electron donating capability of the borate ligands.³¹ Inspired by the recent reports on the azole anion based ILs which can absorb SO₂ through multi-site interaction,^32–34 we have carried out a systematic study on a range of azolium poly(azolyl)borate ILs. The DFT calculation results demonstrate that the borate anions can strongly bind with multiple SO₂ molecules with nearly uniform binding energies.³⁵

Herein, we report the synthesis and characterization of a series of imidazolium poly(azolyl)borate salts derived from 1H-pyrazole, 1H-imidazole, 1H-1,2,4-triazole, and 1H-tetrazole. The cations studied in this work are alkyl or aminoalkyl N,N′-substituted imidazolium ions (see Fig. 1b). Although tetraalkyl phosphonium or ammonium ions are also commonly used cations in forming ILs,³⁶ we selected imidazolium because of its relatively good thermal stability and low formula weight, as well as its different symmetric elements than the poly(azolyl)borates, so that the combination of these two asymmetric bulky organic ions can more possibly form room temperature ILs.³⁰ We also present the multiple SO₂ molar uptakes (at 20 °C, 0.1 MPa of SO₂) by the synthesized ILs as verified through the absorption tests.

Results and discussion

Synthetic

The imidazolium poly(azolyl)borate salts were prepared following the synthetic procedure as described in detail in the Experimental section. As illustrated in Scheme 1, the corresponding imidazolium chloride and potassium poly(azolyl)borate reacted in a methanol/dichloromethane mix-solvent system, leading to the formation of product 1–3 and 5–14, while potassium chloride precipitated from the solution as the side product, which could be easily removed through filtration. The product was further purified through extraction with dichloromethane and washing with distilled water. The ILs thus obtained were dried at 60 °C under vacuum to reduce any traces of residual water. The water content of the ILs was determined by Karl Fisher titration and found to be less than 0.1 wt%. For the solid products, recrystallization from ethanol–hexane (1 : 5) gave colourless crystals suitable for X-ray crystallographic determination (detailed structural discussion see the Crystallography part).

Scheme 1 Synthesis of imidazolium poly(azolyl)borate salts.

The structure and purity of the synthesized poly(azolyl)borate salts were confirmed by ¹H, ¹³C, and ¹¹B NMR, infrared spectroscopy, mass spectrometry, and elemental analysis. Compound 9 was also characterized by single-crystal X-ray crystallography. ¹¹B NMR spectra of 1–3 and 5–8 showed the ¹¹B triplet resonances at −6.7 to −11.0 ppm, while 9–14 exhibited the ¹¹B doublet peak at −1.1 to −4.0 ppm, and all of them displayed coupling between the boron and hydrogen atoms. In the infrared spectra, the B–H stretching frequencies appeared around 2400 cm⁻¹ for all the synthesized compounds, which is a very typical characterization for poly(azolyl)borates.^25,29,37 These newly prepared salts are soluble in methanol, dichloromethane, acetone, and water, but insoluble in diethyl ether, hexane, and toluene.

Thermal behaviours

The thermal stability is one of the most important properties of ILs, and it is reflected by the decomposition temperature (T_d). A high T_d is desirable for ILs because it leads to a broader utilization temperature range. The decomposition temperature is related to the strength of the ionic bonds, which is determined by the electrostatic interactions between cations and anions. Usually, cleavage of stronger chemical bonds requires more energy, thus higher thermal decomposition temperature.³⁸ The thermal stability of the newly synthesized ILs has been investigated using thermogravimetric analysis (TGA), and the decomposition temperatures (T_{5% onset}) are listed in Table 1. The T_d values of the synthesized ILs vary from 156.9 (salt 3, [Nmim][H₂B(pz)₂]) to 226.8 °C (salt 7, [mmim][H₂B(tetz)₂]). Apparently, the cation structure affects the thermal stability of the ILs, for example, when paired with the same borate anion, the T_d values of the ILs decrease in the cation order of [mmim] > [n-bmim] > [Nmim]. This is not difficult to understand, since a smaller cation is much closer to the anion, and thus experiencing a stronger cation–anion interaction. For the dihydrobis(pyrazolyl)borate salts, the thermal stabilities are enhanced with more nitrogen atoms in the azolyl ring of the anions, i.e. the T_d values increase following the anion order of [H₂B(pz)₂] < [H₂B(tz)₂] < [H₂B(tetz)₂]. The temperature differences might be originated from the different innate thermal stabilities of the borate anions, and the stability sequence is in good agreement with the B–N(azolyl) bond formation energies (zero point energy corrected, calculated at the B3LYP/6-311++G(d,p) level. E_b/kJ mol⁻¹: −98.6 for [H₂B(pz)₂], −121.4 for [H₂B(tz)₂], and −147.7 for [H₂B(tetz)₂]), more negative values imply stronger bonds predicted by our previous theoretical research.³¹ It is worth noting that the theoretical bond formation energies were calculated at 0 K, but they reflect the thermal robustness of the bonds, as they are the lowest (most stable) points on the potential energy surface (PES). Only at sufficiently high temperature (high thermal energy), the system (here the borate anions) can climb up the PES and eventually the bond is broken.

Table 1 Physicochemical properties of the imidazolium poly(azolyl)borate salts

Compd	Tg^a (°C)	Td^b (°C)	Density^c	Viscosity^d (mPa s)	κ^e (μS cm^−1)	State^f
a Glass transition temperature (DSC).b Decomposition temperature (TGA, T5% onset).c In units of g cm^−3 at 25 °C.d Viscosity at 25 °C.e Conductivity at 25 °C; the conductivity for a KCl solution (0.5 mM) is 8.61 μS cm^−1.f At 25 °C.g Legend: im, imidazole; pz, pyrazole; tz, 1,2,4-triazole; tetz, tetrazole; m, methyl; n-b, n-butyl; N, 2-(diethylam-ino)ethyl.h Data taken from ref. 30.
[mmim][H2B(pz)2]^g (1)	80.2	164.6	—	—	8.24	Solid
[n-bmim][H2B(pz)2] (2)	−57.7	157.5	1.24	505	7.76	Liquid
[Nmim][H2B(pz)2] (3)	−60.8	156.9	1.20	2354	7.12	Liquid
[mmim][H2B(tz)2] (4)^h	120.7	214.8	1.29	—	8.51	Solid
[n-bmim][H2B(tz)2] (5)	−56.7	225.6	1.25	398	7.82	Liquid
[Nmim][H2B(tz)2] (6)	−53.0	194.9	1.20	836	6.70	Liquid
[mmim][H2B(tetz)2] (7)	−76.3	226.8	1.34	90	8.43	Liquid
[n-bmim][H2B(tetz)2] (8)	−71.1	223.1	1.26	105	7.70	Liquid
[mmim][HB(pz)3] (9)	110.9	221.0	—	—	8.06	Solid
[n-bmim][HB(pz)3] (10)	−49.2	210.5	1.30	787	6.65	Liquid
[Nmim][HB(pz)3] (11)	−41.7	173.5	1.26	2171	6.25	Liquid
[mmim][HB(im)3] (12)	−45.3	208.5	1.34	645	7.92	Liquid
[n-bmim][HB(im)3]^f (13)	−49.8	205.5	1.30	1568	6.31	Liquid
[Nmim][HB(im)3] (14)	−41.9	176.3	1.28	5116	5.92	Liquid
[mmim][HB(tz)3] (15)^h	115.4	217.0	1.36	—	8.03	Solid
[n-bmim][HB(tz)3] (16)^h	−45.2	202.0	1.27	—	7.61	Liquid

---

The melting points are affected by the size of the cation and anion, as well as the substituent on the cation. Differential scanning calorimetric (DSC) investigation indicates that the hydrotris(azolyl)borate salts exhibit higher melting temperatures compared with the dihydrobis(azolyl)borate salts based on the same azole, when they share a common cation (see Table 1). For example, 9 melts at 110.9 °C, which is higher than the melting point for compound 1 (T_g = 80.2 °C). With the same borate anion, as the imidazolium cation bearing a longer chain, the melting point lowers dramatically (see Fig. 2b). The 1,3-dimethylimidazolium poly(azolyl)borate salts 1, 4, 9, and 12 have melting temperatures higher than 86 °C and are solids at ambient conditions, whereas the 1-methyl-3-n-butylimidazolium analogues are liquids at room temperature with T_g values ranging from −49.2 to −57.7 °C. All of the 1-[2-(diethylamino)ethyl]-3-methylimidazolium based salts studied herein are also room temperature ionic liquids with transition points lower than −41 °C. The fact that most of the 2-(diethylamino)ethyl group containing salts have slightly higher T_g values than the n-butyl containing ones with respect to the same anion might be attributed to the increasing ion–ion and van der Waals interactions in the salts resulted from the introduction of a more polar alkylamino group compared with the alkyl group.³⁹

Fig. 2 (a) Thermal decomposition temperatures (T_d) of the imidazolium poly(azolyl)borate salts. (b) Glass transition temperatures (T_g) of the imidazolium poly(azolyl)borate salts. For data source see note in Table 1.

Density

The densities of the prepared ILs are in the range of 1.20 to 1.34 g cm⁻³ and decrease as the alkyl chain length increases in the cations (see Table 1). The salts containing 1-[2-(diethylamino)ethyl]-3-methylimidazolium cation have lower densities than those with 1-methyl-3-n-butylimidazolium cation possibly because the less efficient packing of the former than the latter. When keeping other aspects all the same, i.e. the same cation and the same number of azole rings on the borate anion, the ILs exhibit higher density when there are more nitrogen atoms in the azole ring.

Viscosity

Viscosity is influenced by the ion–ion interactions including van der Waals interactions and hydrogen bonding.³⁸ The viscosity of the synthesized ILs were determined by shear rate controlled mode with an Anton Paar MCR302 rheometer at 25 °C. The measurements were repeated three times for each sample and the results show good reproducibility. The shear stress and shear rate are directly proportional to each other indicating that the ILs synthesized herein are Newtonian fluids (Fig. 3a). The viscosity increases following the sequence of [mmim] < [n-bmim] < [Nmim] while the anion remains the same (see Fig. 3b). This is due to the possible more hydrogen bonds formed between the [Nmim] cation with the paired anions, and the stronger van der Waals interactions provided by the increment of alkyl chain length.

Fig. 3 (a) Shear stress–shear rate curves. (b) Viscosities of the imidazolium poly(azolyl)borate salts.

Conductivity

The conductivity of ILs depends on their density, molecular weight, viscosity, and size of the ions. The conductivities of the ILs were measured in 0.5 mM acetonitrile solutions and carried out at ambient conditions. As illustrated in Fig. 4, the ILs formed with the same anion show the highest conductivities when paired with the [mmim] cation, followed by [n-bmim], and those with the [Nmim] cation display the lowest conductivities. This is quite expected as the conductivity is primarily affected by the viscosity of ILs, and the two factors are inversely correlated, i.e. higher viscosity leads to lower conductivity. Ion sizes also play an important role in the determination of conductivity and usually smaller and more agile ions result in higher conductivity, which is also true to the data obtained by this research. Compounds 1, 7, and 9 synthesized in this work have the highest conductivities at 8.24, 8.43, and 8.06 μS cm⁻¹, respectively, which are very similar to that for KCl (8.61 μS cm⁻¹ for KCl solution of the same concentration in 95 : 5 acetonitrile–water mixed solvent).

Fig. 4 Conductivities of the imidazolium poly(azolyl)borate salts. For data source see note in Table 1.

Crystallography

Single crystals of compound 9 were obtained by layering a concentrated ethanol solution with five equivalent volume of hexane and were investigated by X-ray diffraction. The structure is shown in Fig. 5a. The hydrotris(pyrazolyl)borate anion is tetrahedral with the N–B–N angles ranging from 108.2 to 108.8° and the H–B–N angles of 108.4 to 112.0°. The B–N distances fall between 1.543(2) and 1.550(2) Å. The bond lengths and angles of compound 9 are very similar to the bis- and tris(triazolyl)borates in the literature.³⁰ For example, the literature geometric values for [mmin][H₂B(tz)₂] and [mmmin][HB(tz)₃] include the N–B–N angles ca. 109–110°, the H–B–N angles ca. 106–111°, and the B–N distances ca. 1.542–1.562 Å, correspondingly.

Fig. 5 (a) Perspective view of 9 with probability ellipsoids at the 50% level. Hydrogen atoms are included but are unlabeled for clarity. Inset: extended crystal structure. (b) Packing diagram of 9 viewed down the a axis. Dashed lines indicate hydrogen bonding.

The extended structure of 9 consists of nearly parallel sheets and each sheet is composed of two layers of the ion pairs connected through C–H⋯N hydrogen bonds (Fig. 5a inset and 5b). In contrast, the hydrogen bonding network is slightly weaker in the crystal structure of 9 than that in [mmin][H₂B(tz)₂] and [mmmin][HB(tz)₃]. In 9, the C–H⋯N hydrogen bonds with C⋯N are 3.45 Å, a little longer compared with the values of 3.2–3.4 Å and 3.0–3.4 Å for [mmin][H₂B(tz)₂] and [mmmin][HB(tz)₃], respectively.³⁰ This is a result of only one N atom on each pyrazolyl ring available for H bonding, which further leads to the absence of hydrogen bond between the two layers, unlike the literature reported structures.³⁰

SO₂ absorption and desorption test

The synthesized ILs were tested for their capability of SO₂ absorption. It turned out that all of them can capture SO₂ at atmospheric pressure and reach equilibrium with molar absorption ratio higher than one. They possess almost no ability for capturing CO₂, which in fact renders them good selectivity for using as selective SO₂ absorbents. The absorption processes of representative ILs synthesized herein are illustrated in Fig. 6a. 1-Methyl-3-n-butylimidazolium hydrotris(imidazolyl)borate (13) can absorb 5.8 moles of SO₂ per mole IL at 20 °C, 0.1 MPa of SO₂, which is comparable to the IL absorbent with the highest absorption molar ratio in the literature ([P₆₆₆₁₄][BenIm]).⁴⁰ More preferably, 13 has a higher mass absorption ratio of 1.05 g SO₂/g of IL as it has a lower molecular weight of 352.23 compared with that of 552.52 for [P₆₆₆₁₄][BenIm]. In principle, the high molar uptake can be categorized into two kinds of interactions. Each of the three imidazolyl ring in 13 can chemically react with one SO₂ (binding energies, calculated at the B3LYP/6-311++G(d,p) level, are −63.1, −56.2, and −52.8 kJ mol⁻¹ (ref. 35)). The binding sites are the idling N atoms, linking to the S(SO₂) atoms at distances around 2.4 Å. The existing bindings make the incoming binding slightly weaker, which can be also seen in the longer N–S distances, from 2.40 Å after the first binding to 2.42 Å after the second binding, and to 2.43 Å after the third binding. The subsequent absorptions are even weaker and physical in nature and these SO₂ can form network with the chemically bound ones.⁴¹ The inner shell of the network is the chemically absorbed SO₂ molecules, whose O atoms are connected either to the second shell SO₂ molecules (with the S atoms, due to the electrostatic interactions) or to the H atoms in cations or anions in the form of hydrogen bonds. Correspondingly, the SO₂ molecules in the second shell orient the O/S atoms according to the positions of the S/O atoms in the inner shell, and form O⋯H hydrogen bonds with other cations and anions. Eventually, a SO₂ network is formed.

Fig. 6 (a) Absorption processes by representative ILs as a function of time at 20 °C, 0.1 MPa SO₂ at a flow rate of 60 mL min⁻¹ (b) FT-IR spectra of [n-bmim][HB(im)₃] before and after the reaction with SO₂.

Hydrotris(pyrazolyl)borate based ionic liquids 10 and 11 also exhibit multiple absorption of SO₂ with molar ratios of 4.9 and 4.0 mol SO₂/mol of IL, respectively. It is unexpected that the alkylamino substituted imidazolium based IL 11 showed a weaker absorption ability compared with the alkyl substituted derivative 10, as it is normally believed that the introduction of an amino group should increase the absorption of SO₂. This might be caused by the much higher viscosity of 11 (2171 mPa S) than 10 (787 mPa S) which makes the delay of the absorption equilibrium.

Moreover, for the three ILs, clearly their isotherms fall into two classical sorption isotherm types:⁴² the “L” isotherm (10 and 11) and the “S” isotherm (13, red dots in Fig. 6a). The “L” isotherm features a single absorption mechanism and a progressive saturation of the sorbent. The “S” isotherm implies more than one absorption mechanisms, which usually consist of a slow/hard absorption and then a following fast/easy absorption. However, the exact mechanisms leading to the “S” isotherm of 13 is still unclear and we hope to clarify this through controlled step-wise absorption study in the future.

Fig. 6b shows the IR spectra of the IL before and after SO₂ absorption. In comparison with the fresh IL, new IR bands at 954 and 1530 cm⁻¹ appeared as a result of the absorption of SO₂, which are attributable to sulphate S–O and S=O stretches, respectively.

We have also tested the reversibility of SO₂ absorption of a representative IL ([n-bmim][HB(pz)₃], 10) by carrying out adsorption–desorption cycles. The absorption conditions were the same as before and desorption was performed at 80 °C under N₂ for 60 minutes. As shown in Fig. 7, at least the IL 10 can be recycled for five times without losing the initial SO₂ absorption capacity. However, about 1.5 mole absorbed SO₂ per mole IL is not released at the desorption conditions, reducing the active SO₂ absorption uptake from 5.0 mol SO₂/mol IL to 3.5 mol SO₂/mol IL per cycle. The incomplete SO₂ desorption has been observed in other ILs as well,^32,40,41 which indicates the complexity of IL–SO₂ interactions at elevated temperature. Still a lack of clear explanation of this phenomenon could be provided and possible reasons include water contamination during measurement,⁴¹ strong IL–SO₂ interactions under high temperature,^32,40 etc. Reversibility research of other ILs is still undergoing in the group.

Fig. 7 SO₂ absorption–desorption cycles of 10 as a function of time. Five consecutive cycles are carried out. Absorption (black dots) is measured at 20 °C, 0.1 MPa SO₂ at a flow rate of 60 mL min⁻¹ for 1 hour. Desorption (blue dots) is measured at 80 °C, 0.1 MPa N₂ at a flow rate of 60 mL min⁻¹ for 1 hour.

Conclusions

Thirteen new imidazolium poly(azolyl)borate salts were conveniently synthesized by simple salts metathesis reactions. Many of the prepared salts are liquids at ambient conditions. The trends and correlations between the physicochemical properties including the thermal stability, glass transition points, viscosity, and conductivity of the newly prepared ILs and their structures were fully discussed. Most ILs showed multi-molar SO₂ uptake at ambient conditions, especially with 1-methyl-3-n-butylimidazolium hydrotris(imidazolyl)borate (13) exhibiting one of the highest SO₂ capacity of 5.8 mol mol⁻¹ of IL and 1.05 g g⁻¹ of IL. However, regarding industrial application, there is still a long road for these ILs and the major obstacles would be the cost and the slow reaction kinetics with SO₂. Nevertheless, our research enriches the research of poly(azolyl)borate-based ILs.

Experimental

Materials and methods

1H-pyrazole (99%), 1H-imidazole (99%), 1,2,4-triazole (99%), 1H-tetrazole (98%), 1-methylimidazole (99%), 1-butyl-3-methylimidazolium chloride (98%), 1,3-dimethylimidazolium chloride (99%), and 2-diethylamidoethylchloride hydrochloride (99%) were obtained from Energy Chemical. Potassium borohydride (97%) was purchased from Aladdin. Potassium hydrotris(pyrazolyl)borate (99%) was purchased from TCI. All the above chemicals were used as received without purification. SO₂ gas with a purity of 99.9% was obtained from Beijing Yanglilai Technology Co. Ltd., China. Potassium dihydrobis(pyrazolyl)borate,⁴³ potassium hydrotris(imidazolyl) borate,⁴⁴ potassium dihydrobis(1,2,4-tiazolyl)borate,⁴⁵ potassium dihydrobis(tetrazolyl)borate,²⁵ 1-[2-(diethylamino)ethyl]-3-methylimidazolium chloride⁴⁶ were prepared and purified according to literature reported procedures. ¹H NMR, ¹¹B NMR, and ¹³C NMR spectra were measured on a Bruker Avance-400 spectrometer at 400, 128, and 100 MHz in DMSO-d6. Chemical shifts (δ) are reported in parts per million (ppm) and coupling constants are reported in Hz with multiplicities denoted as s (singlet), d (doublet), t (triplet), m (multiplet). Infrared spectra were collected on a Nicolet Avatar 330 FT-IR spectrometer and absorptions (ν) are reported in wavenumbers (cm⁻¹). High resolution mass spectra (HRMS) were obtained on a Bruker Apex IV FTMS spectrometer with an ESI source. Elemental analyses were performed on an Elementar Vario EL cube elemental analyzer. Nitrogen contents tend to be lower than expected values, due to the formation of refractory boron nitrides.^27,47 Decomposition temperatures were determined by thermogravimetric analysis (TGA) conducted on a NETZSCH STA 409C instrument under nitrogen with a heating rate of 10 °C min⁻¹. Differential scanning calorimetry (DSC) was performed with a NETZSCH DSC 200 PC unit under a nitrogen flow at a heating rate of 10 °C min⁻¹. Conductivities were determined on an INESA DDBJ-350 conductivity meter. Densities were measured at 25 °C with a 1.00 mL pipette and an OHAUSE AR124CN analytical balance. Each sample was measured at least five times and the average values were reported. The relative standard deviation of the density measurements was 0.5%. Viscosities were measured by an Anton Paar MCR302 rheometer, using shear rate controlled mode (shear rate changing from 1 to 1000 s⁻¹). Single-crystal X-ray diffraction data were collected on a Bruker SMART Apex II DUO diffractometer at 296 K.

General procedure for the synthesis of poly(azolyl)borate organic salts 1–3 and 5–14

The potassium poly(azolyl)borate salt (10.0 mmol) was dissolved in 15 mL of methanol in a 100 mL round bottom flask under ambient conditions, then a 15 mL solution of equivalent amount of the proper imidazolium chloride in dichloromethane was added while stirring. White precipitate started to form during the addition. After stirring at room temperature for 4 h, the reaction mixture was filtered to remove the precipitate and the filtrate was dried under reduced pressure to remove the solvents. The residual substance was extracted with 10 mL of dichloromethane and the solution was washed three times with distilled water to remove traces of sodium chloride. The dichloromethane was then evaporated under reduced pressure. The product was vacuum dried at 60 °C for 24 h to reduce any traces of water and then subjected to further characterization and application.

1,3-Dimethylimidazolium dihydrobis(pyrazolyl)borate (1; [mmim]-[H₂B(pz)₂]). Yield: 84%, colourless solid. ¹H NMR (δ (ppm)): 8.99 (s, 1H), 7.63 (s, 2H), 7.33 (s, 2H), 7.21 (s, 2H), 5.93 (s, 2H), 3.79 (s, 6H). ¹³C NMR (δ (ppm)): 137.60, 137.10, 133.40, 123.43, 102.35, 35.69. ¹¹B NMR (δ (ppm)): −6.71 (t, J_BH = 185.97 Hz). IR (KBr, cm⁻¹): 3454 (b), 3161 (s), 3107 (s), 2390 (s), 2279 (s), 1573 (s), 1496 (s), 1417 (s), 1392 (s), 1287 (s), 1198 (s), 1171 (s), 1081 (s), 1047 (s), 961 (s), 878 (s), 824 (s), 756 (s), 718 (s), 643 (s), 620 (s). MS: 97.0760 for [mmim], 147.0848 for [H₂B(pz)₂]. Anal. calcd for C₁₁H₁₇BN₆ (244.10): C, 54.12; H, 7.02; N, 34.43. Found: C, 53.86; H, 7.11; N 33.01.

1-Methyl-3-n-butylimidazolium dihydrobis(pyrazolyl)borate (2; [n-bmim][H₂B(pz)₂]). Yield: 89%, colourless liquid. ¹H NMR (δ (ppm)): 9.13 (s, 1H), 7.75 (s, 1H), 7.68 (s, 1H), 7.32 (s, 2H), 7.21 (s, 2H), 5.92 (s, 2H), 4.14 (t, J = 14.17 Hz, 2H), 3.82 (s, 3H), 1.78–1.71 (m, 2H), 1.30–1.20 (m, 2H), 0.90 (t, J = 14.92 Hz, 3H). ¹³C NMR (δ (ppm)): 137.47, 136.65, 133.34, 123.65, 122.35, 102.27, 48.52, 35.76, 31.45, 18.84, 13.38. ¹¹B NMR (δ (ppm)): −6.68 (t, J_BH = 194.92 Hz). IR (neat, cm⁻¹): 3147 (b), 2961 (s), 2935 (s), 2874 (s), 2386 (s), 2268 (s), 1571 (s), 1495 (s), 1465 (s), 1416 (s), 1392 (s), 1287 (s), 1200 (s), 1152 (s), 1081 (s), 1047 (s), 964 (s), 871 (s), 755 (s), 718 (s), 645 (s), 622 (s). MS: 139.1230 for [n-bmim] and 147.0848 for [H₂B(pz)₂]. Anal. calcd for C₁₄H₂₃BN₆ (286.18): C, 58.76; H, 8.10; N, 29.37. Found: C, 58.37; H, 7.97; N 29.07.

1-[2-(Diethylamino)ethyl]-3-methylimidazolium dihydrobis(pyrazo-lyl)borate (3; [Nmim][H₂B(pz)₂]). Yield: 83%, colourless liquid. ¹H NMR (δ (ppm)): 9.04 (s, 1H), 7.72 (s, 1H), 7.64 (s, 1H), 7.33 (d, J = 1.64 Hz, 2H), 7.22 (d, J = 1.09 Hz), 5.93 (t, J = 3.31 Hz, 2H), 4.17 (t, J = 11.50 Hz, 2H), 3.83 (s, 3H), 2.70 (t, J = 11.50 Hz, 2H), 2.49–2.44 (m, 4H), 0.86 (t, J = 11.30 Hz, 6H). ¹³C NMR (δ (ppm)): 137.42, 136.77, 133.27, 123.10, 122.62, 102.18, 51.81, 47.36, 46.27, 35.58, 11.67. ¹¹B NMR (δ (ppm)): −6.68 (t, J_BH = 190.11 Hz). IR (KBr, cm⁻¹): 3146 (b), 2969 (s), 2816 (s), 2385 (s), 2267 (s), 1570 (s), 1495 (s), 1457 (s), 1414 (s), 1390 (s), 1286 (s), 1201 (s), 1152 (s), 1080 (s), 1152 (s), 1080 (s), 1045 (s), 962 (s), 877 (s), 754 (s), 647 (s), 622 (s). MS: 97.0760 for [mmim] and 147.0848 for [H₂B(pz)₂]. Anal. calcd for C₁₆H₂₈BN₇ (329.25): C, 58.37; H, 8.57; N, 29.78. Found: C, 57.64; H, 8.72; N 29.23.

1-Methyl-3-n-butylimidazolium dihydrobis(1,2,4-triazolyl)borate (5; [n-bmim][H₂B(tz)₂]). Yield: 87%, colourless liquid. ¹H NMR (δ (ppm)): 9.12 (s, 1H), 7.97 (s, 2H), 7.75 (s, 1H), 7.69 (s, 1H), 7.66 (s, 2H), 4.14 (t, J = 14.40 Hz, 2H), 3.83 (s, 3H), 1.79–1.71 (m, 2H), 1.30–1.20 (m, 2H), 0.90 (t, J = 14.69 Hz, 3H). ¹³C NMR (δ (ppm)): 151.09, 147.44, 136.51, 123.61, 122.29, 48.51, 35.71, 31.31, 18.75, 13.27. ¹¹B NMR (δ (ppm)): −9.35 (t, J_BH = 187.20 Hz). IR (neat, cm⁻¹): 3389 (b), 3152 (s), 3111 (s), 2962 (s), 2875 (s), 2408 (s), 1572 (s), 1500 (s), 1466 (s), 1411 (s), 1315 (s), 1267 (s), 1155 (s), 1130 (s), 1019 (s), 969 (s), 870 (s), 723 (s), 682 (s), 646 (s). MS: 139.1230 for [n-bmim] and 149.0752 for [H₂B(tz)₂]. Anal. calcd for C₁₂H₂₁BN₈ (288.16): C, 50.02; H, 7.35; N, 38.89. Found: C, 49.49; H, 7.96; N 37.56.

1-[2-(Diethylamino)ethyl]-3-methylimidazolium dihydrobis(1,2,4-triazolyl)borate (6; [Nmim][H₂B(tz)₂]). Yield: 85%, colourless liquid. ¹H NMR (δ (ppm)): 9.06 (s, 1H), 7.99 (s, 2H), 7.75 (s, 1H), 7.68 (s, 3H), 4.19 (t, J = 11.43 Hz, 2H), 3.86 (s, 3H), 2.70 (t, J = 11.50 Hz, 2H), 2.49–2.44 (m, 4H), 0.86 (t, J = 14.05 Hz, 6H). ¹³C NMR (δ (ppm)): 151.12, 147.47, 136.80, 123.20, 122.76, 51.82, 47.47, 46.31, 35.73, 11.64. ¹¹B NMR (δ (ppm)): −9.40 (d, J_BH = 213.57 Hz). IR (neat, cm⁻¹): 3389 (b), 3152 (s), 3111 (s), 2962 (s), 2875 (s), 2408 (s), 1573 (s), 1500 (s), 1466 (s), 1411 (s), 1315 (s), 1267 (s), 1156 (s), 1130 (s), 1019 (s), 968 (s), 870 (s), 682 (s), 646 (s). MS: 182.1652 for [Nmim] and 149.0752 for [H₂B(tz)₂]. Anal. calcd for C₁₄H₂₆BN₉ (331.23): C, 50.77; H, 7.91; N, 38.06. Found: C, 49.91; H, 8.19; N 36.63.

1,3-Dimethylimidazolium dihydrobis(tetrazolyl)borate (7; [mmim]-[H₂B(tetz)₂]). Yield: 83%, colourless liquid. ¹H NMR (δ (ppm)): 9.04 (s, 1H), 8.86 (s, 2H), 7.66 (s, 2H), 3.84 (s, 6H). ¹³C NMR (δ (ppm)): 147.32, 137.06, 123.43, 35.74. ¹¹B NMR (δ (ppm)): −10.98 (d, J_BH = 183.53 Hz). IR (neat, cm⁻¹): 3445 (b), 3122 (s), 2443 (s), 1633 (s), 1575 (s), 1466 (s), 1419 (s), 1359 (s), 1244 (s), 1174 (s), 1157 (s), 1137 (s), 1103 (s), 1034 (s), 981 (s), 868 (s), 744 (s), 713 (s), 623 (s). MS: 97.0760 for [mmim] and 151.0657 for [H₂B(tetz)₂]. Anal. calcd for C₇H₁₃BN₁₀ (248.07): C, 38.89; H, 5.28; N, 56.47. Found: C, 38.43; H, 5.75; N 53.43.

1-Methyl-3-n-butylimidazolium dihydrobis(tetrazolyl)borate (8; [n-bmim][H₂B(tetz)₂]). Yield: 86%, colourless liquid. ¹H NMR (δ (ppm)): 9.10 (s, 1H), 8.83 (s, 2H), 7.76 (s, 1H), 7.70 (s, 1H), 4.15 (t, J = 13.96 Hz, 2H), 3.85 (s, 3H), 1.79–1.72 (m, 2H), 1.29–1.20 (m, 2H), 0.89 (t, J = 14.77 Hz, 3H). ¹³C NMR (δ (ppm)): 147.22, 136.47, 123.59, 122.22, 48.55, 35.75, 31.42, 18.71, 13.39. ¹¹B NMR (δ (ppm)): −10.89 (t, J_BH = 205.13 Hz). IR (KBr, cm⁻¹): 3434 (b), 3115 (s), 2962 (s), 2936 (s), 2875 (s), 2439 (s), 1572 (s), 1464 (s), 1357 (s), 1240 (s), 1165 (s), 1134 (s), 1101 (s), 1031 (s), 978 (s), 868 (s), 742 (s), 712 (s), 623 (s). MS: 139.1230 for [n-bmim] and 151.0657 for [H₂B(tetz)₂]. Anal. calcd for C₁₀H₁₉BN₁₀ (290.14): C, 41.40; H, 6.60; N, 48.28. Found: C, 41.03; H, 6.91; N, 46.64.

1,3-Dimethylimidazolium hydrotris(pyrazolyl)borate (9; [mmim]-[HB(pz)₃]). Yield: 89%, colorless solid. ¹H NMR (δ (ppm)): 9.00 (s, 1H), 7.65 (s, 2H), 7.32 (s, 3H), 7.28 (s, 3H), 5.99 (s, 3H), 3.82 (s, 6H). ¹³C NMR (δ (ppm)): 138.20, 137.01, 132.58, 123.42, 102.57, 35.63. ¹¹B NMR (δ (ppm)): −1.10 (d, J_BH = 106.67 Hz). IR (KBr, cm⁻¹): 3447 (b), 3115 (s), 2917 (s), 2850 (s), 2434 (s), 1573 (s), 1497 (s), 1456 (s), 1416 (s), 1387 (s), 1286 (s), 1213 (s), 1171 (s), 1108 (s), 1045 (s), 964 (s), 884 (s), 750 (s), 729 (s). MS: 97.0760 for [mmim] and 213.1065 for [HB(Pz)₃]. Anal. calcd for C₁₄H₁₉BN₈ (310.16): C, 54.21; H, 6.17; N, 36.13. Found: C, 54.15; H, 6.06; N 35.03.

1-Methyl-3-n-butylimidazolium hydrotris(pyrazolyl)borate (10; [n-bmim][HB(pz)₃]). Yield: 85%, colorless liquid. ¹H NMR (δ (ppm)): 9.08 (s, 1H), 7.76 (s, 1H), 7.69 (s, 1H), 7.34 (d, J = 0.73 Hz, 3H), 7.30 (d, J = 1.77 Hz, 3H), 6.03 (t, J = 3.38 Hz, 3H), 4.11 (t, J = 14.34 Hz, 2H), 3.79 (s, 3H), 1.77–1.70 (m, 2H), 1.29–1.20 (m, 2H), 0.90 (t, J = 14.66 Hz, 3H). ¹³C NMR (δ (ppm)): 138.24, 136.55, 132.67, 123.54, 122.21, 102.66, 48.47, 35.71, 31.40, 18.82, 13.30. ¹¹B NMR (δ (ppm)): −1.12 (d, J_BH = 109.67 Hz). IR (neat, cm⁻¹): 3090 (b), 2960 (s), 2934 (s), 2874 (s), 2571 (s), 2433 (s), 2159 (s), 1704 (s), 1570 (s), 1497 (s), 1465 (s), 1413 (s), 1384 (s), 1288 (s), 1212 (s), 1184 (s), 1104 (s), 1079 (s), 1040 (s), 960 (s), 920 (s), 872 (s), 838 (s), 753 (s), 671 (s), 624 (s). MS: 139.1230 for [n-bmim] and 213.1065 for [HB(pz)₃]. Anal. calcd for C₁₇H₂₅BN₈ (352.24): C, 57.97; H, 6.17; N, 31.81. Found: C, 57.54; H, 6.95; N 30.69.

1-[2-(Diethylamino)ethyl]-3-methylimidazolium hydrotris(pyrazol-yl)borate (11; [Nmim][HB(pz)₃]). Yield: 83%, colorless liquid. ¹H NMR (δ (ppm)): 9.03 (s, 1H), 7.73 (s, 1H), 7.65 (s, 1H), 7.34 (s, 3H), 7.30 (s, 3H), 6.01 (s, 3H), 4.17 (t, J = 11.43 Hz, 2H), 3.84 (s, 3H), 2.70 (t, J = 11.70 Hz, 2H), 2.49–2.44 (m, 4H), 0.86 (t, J = 14.20 Hz, 6H). ¹³C NMR (δ (ppm)): 138.23, 136.73, 132.68, 123.11, 122.64, 102.63, 51.79, 47.39, 46.21, 35.62, 11.61. ¹¹B NMR (δ (ppm)): −1.10 (d, J_BH = 113.01 Hz). IR (neat, cm⁻¹): 3092 (b), 2970 (s), 2873 (s), 2817 (s), 2435 (s), 1570 (s), 1498 (s), 1457 (s), 1413 (s), 1385 (s), 1289 (s), 1213 (s), 1185 (s), 1105 (s), 1080 (s), 1042 (s), 961 (s), 920 (s), 754 (s), 654 (s). MS: 182.1652 for [Nmim] and 213.1065 for [HB(Pz)₃]. Anal. calcd for C₁₉H₃₀BN₉ (395.31): C, 57.73; H, 7.65; N, 31.89. Found: C, 57.18; H, 8.03; N 30.97.

1,3-Dimethylimidazolium hydrotris(imidazolyl)borate (12; [mmim][HB(im)₃]). Yield: 85%, colorless liquid. ¹H NMR (δ (ppm)): 9.06 (s, 1H), 7.69 (s, 2H), 7.27 (s, 3H), 6.83 (s, 3H), 6.82 (s, 3H), 3.85 (s, 6H). ¹³C NMR (δ (ppm)): 139.75, 137.08, 128.16, 123.43, 120.61, 35.83. ¹¹B NMR (δ (ppm)): −4.05 (d, J_BH = 108.96 Hz). IR (KBr, cm⁻¹): 3441 (b), 3103 (s), 2404 (s), 1574 (s), 1479 (s), 1419 (s), 1298 (s), 1254 (s), 1203 (s), 1173 (s), 1125 (s), 1075 (s), 1012 (s), 924 (s), 833 (s), 755 (s), 719 (s), 670 (s), 721 (s). MS: 97.0760 for [mmim] and 213.1065 for [HB(im)₃]. Anal. calcd for C₁₄H₁₉BN₈ (310.17): C, 54.21; H, 6.17; N, 36.13. Found: C, 53.13; H, 6.44; N 34.27.

1-Methyl-3-n-butylimidazolium hydrotris(imiazolyl)borate (13; [n-bmim][HB(im)₃]). Yield: 86%, colorless liquid. ¹H NMR (δ (ppm)): 9.13 (s, 1H), 7.76 (s, 1H), 7.69 (s, 1H), 7.25 (s, 3H), 6.82 (s, 3H), 6.80 (s, 3H), 4.14 (t, J = 14.34 Hz, 2H), 3.83 (s, 3H), 1.78–1.71 (m, 2H), 1.29–1.20 (m, 2H), 0.89 (t, J = 14.68 Hz, 3H). ¹³C NMR (δ (ppm)): 139.70, 136.52, 128.12, 123.62, 122.28, 120.60, 48.53, 35.74, 31.36, 18.79, 13.28. ¹¹B NMR (δ (ppm)): −4.02 (d, J_BH = 108.10 Hz). IR (KBr, cm⁻¹): 3104 (b), 2961 (s), 2874 (s), 2621 (s), 2399 (s), 2399 (s), 2207 (s), 1672 (s), 1572 (s), 1479 (s), 1383 (s), 1328 (s), 1298 (s), 1253 (s), 1201 (s), 1169 (s), 1124 (s), 1073 (s), 1021 (s), 923 (s), 832 (s), 718 (s), 670 (s). MS: 139.1230 for [n-bmim] and 213.1065 for [HB(im)₃]. Anal. calcd for C₁₇H₂₅BN₈ (352.24): C, 57.97; H, 6.17; N, 31.81. Found: C, 57.21; H, 7.03; N 30.94.

1-[2-(Diethylamino)ethyl]-3-methylimidazolium hydrotris(imidazo-lyl)borate (14; [Nmim][HB(im)₃]). Yield: 84%, colorless liquid. ¹H NMR (δ (ppm)): 9.06 (s, 1H), 7.73 (s, 1H), 7.65 (s, 1H), 7.25 (s, 3H), 6.81 (s, 6H), 4.17 (t, J = 12.11 Hz, 2H), 3.85 (s, 3H), 2.70 (t, J = 11.97 Hz, 2H), 2.48–2.43 (m, 4H), 0.86 (t, J = 14.01 Hz, 6H). ¹³C NMR (δ (ppm)): 139.73, 136.80, 128.10, 123.23, 122.68, 120.62, 51.81, 47.46, 46.27, 35.67, 11.71. ¹¹B NMR (δ (ppm)): −4.00 (d, J_BH = 111.46 Hz). IR (KBr, cm⁻¹): 3160 (b), 2971 (s), 2823 (s), 2399 (s), 1572 (s), 1479 (s), 1386 (s), 1299 (s), 1254 (s), 1203 (s), 1171 (s), 1124 (s), 1074 (s), 1012 (s), 923 (s), 832 (s), 753 (s), 719 (s), 670 (s). MS: 182.1652 for [Nmim] and 213.1065 for [HB(im)₃]. Anal. calcd for C₁₉H₃₀BN₉ (395.31): C, 57.73; H, 7.65; N, 31.89. Found: C, 56.74; H, 8.33; N 30.83.

Absorption and desorption of SO₂

In a typical process of SO₂ absorption, SO₂ gas at atmospheric pressure was bubbled through about 1 g of IL in a glass container with an inner diameter of 10 mm at a flow rate of 60 mL min⁻¹. The glass container were partly immersed in an oil bath kept at desired temperature. The amount of SO₂ absorbed was determined at regular intervals by an electronic balance with an accuracy of ±0.1 mg.

Acknowledgements

We are grateful for financial support by National Science Foundation of China (Grant No. 21406175), the Scientific Research Foundation for the Returned Overseas Chinese Scholars from State Education Ministry of China (SRF for ROCS, SEM), and the Fundamental Research Funds for the Central Universities.

References

1. T. Welton, Chem. Rev., 1999, 99, 2071–2084.
2. P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed., 2000, 39, 3772–3789.
3. S. Zhang, J. Sun, X. Zhang, J. Xin, Q. Miao and J. Wang, Chem. Soc. Rev., 2014, 43, 7838–7869.
4. J. Dupont, R. F. de Souza and P. A. Z. Suarez, Chem. Rev., 2002, 102, 3667–3692.
5. X. Wang, M. Sternberg, F. T. U. Kohler, B. U. Melcher, P. Wasserscheid and K. Meyer, RSC Adv., 2014, 4, 12476.
6. W. Qian, E. Jin, W. Bao and Y. Zhang, Angew. Chem., Int. Ed., 2005, 44, 952–955.
7. F. Wu, Q. Zhu, R. Chen, N. Chen, Y. Chen and L. Li, Nano Energy, 2015, 13, 546–553.
8. L. H. Finger and J. Sundermeyer, Chem.–Eur. J., 2016, 22, 4218–4230.
9. J. Xiang, F. Wu, R. Chen, L. Li and H. Yu, J. Power Sources, 2013, 233, 115–120.
10. L. Wang, H. Zhang, R. Ge, C. Wang, W. Guo, Y. Shi, Y. Gao and T. Ma, RSC Adv., 2013, 3, 12975.
11. M. A. B. H. Susan, A. Noda, S. Mitsushima and M. Watanabe, Chem. Commun., 2003, 938–939.
12. X. Chen, X. Bao, B. Billet, S. G. Shore and J.-C. Zhao, Chem.–Eur. J., 2012, 18, 11994–11999.
13. Q. Zhang, P. Yin, J. Zhang and J. n. M. Shreeve, Chem.–Eur. J., 2014, 20, 6909–6914.
14. E. D. Bates, R. D. Mayton, I. Ntai and J. H. Davis, J. Am. Chem. Soc., 2002, 124, 926–927.
15. B. E. Gurkan, J. C. de la Fuente, E. M. Mindrup, L. E. Ficke, B. F. Goodrich, E. A. Price, W. F. Schneider and J. F. Brennecke, J. Am. Chem. Soc., 2010, 132, 2116–2117.
16. H. Tang and C. Wu, ChemSusChem, 2013, 6, 1050–1056.
17. S. Sun, Y. Niu, Q. Xu, Z. Sun and X. Wei, RSC Adv., 2015, 5, 46564–46567.
18. G. Cui, Y. Huang, R. Zhang, F. Zhang and J. Wang, RSC Adv., 2015, 5, 60975–60982.
19. C. Wang, H. Luo, D.-e. Jiang, H. Li and S. Dai, Angew. Chem., 2010, 122, 6114–6117.
20. A.-L. Revelli, F. Mutelet and J.-N. Jaubert, J. Phys. Chem. B, 2010, 114, 8199–8206.
21. S. Trofimenko, Chem. Rev., 1993, 93, 943–980.
22. S. Trofimenko, J. Am. Chem. Soc., 1966, 88, 1842–1844.
23. C. Janiak, T. G. Scharmann, P. Albrecht, F. Marlow and R. Macdonald, J. Am. Chem. Soc., 1996, 118, 6307–6308.
24. C. Santini, M. Pellei, G. G. Lobbia and G. Papini, Mini-Rev. Org. Chem., 2010, 7, 84–124.
25. D. Lu and C. H. Winter, Inorg. Chem., 2010, 49, 5795–5797.
26. M. Pellei, G. G. Lobbia, G. Papini and C. Santini, Mini-Rev. Org. Chem., 2010, 7, 173–203.
27. B. H. Hamilton, K. A. Kelly, W. Malasi and C. J. Ziegler, Inorg. Chem., 2003, 42, 3067–3073.
28. C. J. Snyder, P. D. Martin, M. J. Heeg and C. H. Winter, Chem.–Eur. J., 2013, 19, 3306–3310.
29. B. C. Hughes, Z. Lu and D. M. Jenkins, Chem. Commun., 2014, 50, 5273–5275.
30. Z. Zeng, B. Twamley and J. n. M. Shreeve, Organometallics, 2007, 26, 1782–1787.
31. D. Lu and H. Tang, Phys. Chem. Chem. Phys., 2015, 17, 17027–17033.
32. C. Wang, G. Cui, X. Luo, Y. Xu, H. Li and S. Dai, J. Am. Chem. Soc., 2011, 133, 11916–11919.
33. D. Yang, M. Hou, H. Ning, J. Ma, X. Kang, J. Zhang and B. Han, ChemSusChem, 2013, 6, 1191–1195.
34. X. Luo, Y. Guo, F. Ding, H. Zhao, G. Cui, H. Li and C. Wang, Angew. Chem., Int. Ed., 2014, 53, 7053–7057.
35. H. Tang and D. Lu, ChemPhysChem, 2015, 16, 2854–2860.
36. X. Zhang, X. Zhang, H. Dong, Z. Zhao, S. Zhang and Y. Huang, Energy Environ. Sci., 2012, 5, 6668–6681.
37. Effendy, G. Gioia Lobbia, C. Pettinari, C. Santini, B. W. Skelton and A. H. White, Inorg. Chim. Acta, 2000, 308, 65–72.
38. T. A. Siddique, S. Balamurugan, S. M. Said, N. A. Sairi and W. M. D. W. Normazlan, RSC Adv., 2016, 6, 18266–18278.
39. Z.-B. Zhou, H. Matsumoto and K. Tatsumi, Chem.–Eur. J., 2004, 10, 6581–6591.
40. G. Cui, W. Lin, F. Ding, X. Luo, X. He, H. Li and C. Wang, Green Chem., 2014, 16, 1211–1216.
41. S. Y. Hong, J. Im, J. Palgunadi, S. D. Lee, J. S. Lee, H. S. Kim, M. Cheong and K.-D. Jung, Energy Environ. Sci., 2011, 4, 1802.
42. G. Limousin, J.-P. Gaudet, L. Charlet, S. Szenknect, V. Barthes and M. Krimissa, Appl. Geochem., 2007, 22, 249–275.
43. A. F. Hill, J. M. Malget, A. J. P. White and D. J. Williams, Eur. J. Inorg. Chem., 2004, 2004, 818–828.
44. R. Fränkel, U. Kernbach, M. Bakola-Christianopoulou, U. Plaia, M. Suter, W. Ponikwar, H. Nöth, C. Moinet and W. P. Fehlhammer, J. Organomet. Chem., 2001, 617–618, 530–545.
45. C. Janiak, T. G. Scharmann, H. Hemling, D. Lentz and J. Pickardt, Chem. Ber., 1995, 128, 235–244.
46. S. Tian, Y. Hou, W. Wu, S. Ren and J. Qian, J. Hazard. Mater., 2014, 278, 409–416.
47. D. Williams, B. Pleune, J. Kouvetakis, M. D. Williams and R. A. Andersen, J. Am. Chem. Soc., 2000, 122, 7735–7741.

---

Footnote

† Electronic supplementary information (ESI) available: Thermogravimetric curves and X-ray structural data. CCDC 1475597. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra10356a

---