Efficient absorption of ammonia with hydroxyl-functionalized ionic liquids

Abstract

Ammonia (NH₃) emitted from the ammonia synthesis process is a kind of waste chemical resource and a major environmental pollutant. The traditional water scrubbing method suffers from high energy consumption due to the concentrated NH₃ from aqueous ammonia. Therefore, it is desirable to develop novel absorbents for the efficient, reversible and environmentally-friendly recovery of NH₃. In this paper, a series of hydroxyl-functionalized imidazolium ILs ([EtOHmim]X, X = [NTf₂], [PF₆], [BF₄], [DCA], [SCN] and [NO₃]) were designed and prepared. Their physical properties and NH₃ absorption capacities under different temperatures and pressures were systematically investigated. The effects of hydroxyl cation, anionic structures, pressure and temperature on absorption performance were sufficiently studied. In addition, the absorption mechanism was investigated in detail by spectral analysis and quantum chemistry calculations. Compared with conventional IL [Emim]X, a higher absorption capacity was achieved by introducing the hydroxyl group on the imidazolium cation. The mechanism results showed the fascinating absorption performance of the task-specific ILs was attributed to the stronger hydrogen bonding interaction between NH₃ and the H atom of the hydroxyl group. Considering the excellent absorption performance, high thermal stability, and super reversibility, this type of IL provides great improvement over conventional IL and shows their enormous potential in NH₃ recovery.

---

1. Introduction

As a typical environmental pollutant, the increase of ammonia (NH₃) release is leading to different environmental problems including eutrophication of ecosystems and the formation of fine particulate matter¹ and seriously threatens human health. Meanwhile, NH₃ is one of the important chemical raw materials and widely applied in producing nitric acid, nitrogenous fertilizer and so on. Therefore, removal and recovery of NH₃ is of great significance. Up to now, water scrubbing has been the most common method due to its high NH₃ absorption capacity. However, this method suffers from several inherent drawbacks such as high energy consumption, large amounts of waste water and difficult reclamation of NH₃. Hence, it is desirable to develop environmentally-friendly, recyclable absorbents for efficient NH₃ absorption.

It is increasing recognized that ionic liquids (ILs) are promising absorbents because of their special properties such as negligible vapor pressure, wide liquid temperature range, high thermal stability² and adjustability³ and have been paid great attentions on separation of CO₂,^4–11 SO₂ ^12–19 and H₂S.^6,20–23 On the contrast, the reports on NH₃ absorption by ionic liquids are scare. In fact, the results reported have shown that ionic liquids have great potential for NH₃ absorption. Yokozeki et al.^24,25 demonstrated the conventional ionic liquids such as [Bmim][PF₆], [Hmim][Cl] had high NH₃ absorption performance and the anion had little effect on NH₃ solubility. Li et al.²⁶ reported NH₃ solubility increased when the length of cation's alkyl increased. Shi et al.²⁷ testified the cation played the leading role in determining NH₃ solubility and they found a strong hydrogen bonding could be formed between NH₃ molecule and the ring H atom of the [Emim] cation by molecular simulation.

Although great efforts had been made to improve the NH₃ absorption, the solubility of NH₃ in these conventional ILs was relatively low. What is more, the research on the mechanism of IL–NH₃ system was very scanty and not comprehensive. Shi et al.²⁷ investigated the NH₃ absorption mechanism just by molecular simulation. As reported, hydrogen bonding interaction between IL and NH₃ was beneficial to enhance solubility,²⁸ thus in this work, hydroxyl, as the hydrogen bonding donor group, was introduced into the imidazolium cation for further improving the absorption. NH₃ absorption capacities of these hydroxyl-functionalized ILs at different temperatures and pressures were systematically investigated. Furthermore, the absorption mechanism was deeply explored by experimental spectra analysis and quantum chemistry calculations.

2. Experimental section

2.1. Materials

NH₃ (99.999%) was supplied by Beijing Beiwen Gas Factory. 1-Methylimidazole (99.0%), NaBF₄ (98.0%), LiNTf₂ (98.5%) and NaSCN (98.5%) were purchased from Sinopharm Chemical Reagent Co., Ltd. 2-Chloroethanol (99.0%) was purchased from Xiya Chemical Co., Ltd. KPF₆ (98.0%) and NaNO₃ (99.0%) were purchased from Beijing Chemical Works Co., Ltd. All the chemicals above were used without further purification. 1-Ethyl-3-methylimidazolium chloride was obtained from Linzhou Keneng Materials Technology Co., Ltd. 1-2(-Hydroxyethyl)-3-methylimidazolium salt [EtOHmim]X and 1-ethyl-3-methylimidazolium salt [Emim]X were synthesized in our laboratory.

2.2. Synthesis of ionic liquids

[EtOHmim][NTf₂] was synthesized by the method shown in Scheme S1 (seeing ESI†) according to the literature.²⁹ Preparation of [EtOHmim][Cl]: 1-methylimidazole (41.00 g, 0.5 mol) was put in a round-bottom flask (250 mL) and 2-chloroethanol (48.31 g, 0.6 mol) was added slowly. The mixture was stirred for 24 hours at 353.15 K. Then the product was washed several times with ethyl acetate. The [EtOHmim][Cl] white solid was obtained after filtration and vacuum drying for 48 hours at 343.15 K. Preparation of [EtOHmim][NTf₂]: [EtOHmim][Cl] (16.252 g, 0.1 mol) and LiNTf₂ (28.69 g, 0.1 mol) were mixed and some water was added as solvent. Then the mixture was stirred for 24 hours at ambient temperature. By separating the mixture, the supernate was removed and ionic liquid was washed by deionized water for several times until there was no sediment formed when AgNO₃ was added in water. At last, high-purity [EtOHmim][NTf₂] was obtained after rotary evaporation and vacuum drying at 373.15 K for 64 hours. The synthesis methods of other ILs [EtOHmim]X (X = [PF₆]⁻, [BF₄]⁻, [DCA]⁻, [SCN]⁻, [NO₃]⁻) and [Emim]X (X = [NTf₂]⁻, [BF₄]⁻, [NO₃]⁻) were similar to that of [EtOHmim][NTf₂] except the solvent was replaced by acetone. The different structures of [EtOHmim] cation and [Emim] cation were presented in Fig. 1.

Fig. 1 Schematics of [EtOHmim]⁺ (a) and [Emim]⁺ (b), with atom labels.

2.3. Characterization and physical properties

¹H NMR and ¹³C NMR spectra were recorded on a Bruker spectrometer (600 Hz) in deuterated dimethyl sulfoxide (DMSO-d₆) with tetramethylsilane as the internal standard. The NMR data of ILs were presented in ESI.† FT-IR spectra were obtained in the range of 400–4000 cm⁻¹ on a Thermo Nicolet 380 spectrometer. The water contents of the ILs were measured by Karl Fisher coulometers C20 and had been reduced less than 400 ppm. The chlorine contents were tested by ioniza chlorine analyzer (Leici PXSJ-226). The densities of ionic liquids were measured by a density meter (Anton Paar DMA 5000) with an accuracy of ±0.000005 g cm⁻³. The viscosities were measured by an automated micro viscometer (Anton Paar AMVn). The thermal decomposition temperatures were tested by TGA (Q5000 V3.15 Build 263) from room temperature to 773.15 K with a heating rate of 10 K min⁻¹ under N₂ atmosphere. The glass transition temperatures were measured on a Mettler Toledo DSC1 between 123.15 and 298.15 K at a heating rate of 10 K min⁻¹ under N₂ atmosphere.

2.4. Apparatus and procedures

NH₃ solubility was measured by the gas–liquid equilibrium apparatus (shown in Fig. 2) which was similar to that in our previous work.^17,30 In a typical experiment, the temperature was fixed by a thermostatic water bath with an uncertainty of ±0.1 K and the pressure was measured by pressure transmitters with an accuracy of 0.0001 kPa. About 5.0 g IL was placed into the absorption vessel (30 mL), then the gas in absorption vessel was removed by vacuum pump. The NH₃ in storage tank (500 mL) was charged into the absorption vessel slowly through the valve (b₂). Then the magnetic stirrer was opened to enhance the dissolution. It is supposed the equilibrium is reached after the pressure keeps unchanged for 30 min. The solubility of NH₃ was calculated by Peng–Robinson (P–R) eqn (1) through the pressure variation in storage tank and absorption vessel:

a = a_cα (1.1)

α^0.5 = 1 + (0.37464 + 1.54226ω − 0.26992ω²)(1 − T_r^0.5) (1.4)

where p, V_m and T are denoted to pressure, molar volume and temperature, respectively. R is the gas constant and a, b, a_c, α are the parameters of the cubic equation. T_c is the critical temperature, p_c is the critical pressure and T_r is relative temperature. n_NH3 is the amount of NH₃ absorbed and n_IL is the amount of ionic liquid. V_S, V_A and V_IL represent the volumes of storage tank, absorption vessel and ionic liquid, respectively. V_m,S and V_m,A are the molar volume of storage tank and absorption vessel. x_NH3 represents the mole fraction of NH₃ absorbed.

Fig. 2 Schematic diagram of NH₃ absorption apparatus. (a) NH₃ cylinder; (b₁–b₄) valve; (c) storage tank; (d₁ and d₂) pressure sensor; (e) absorption vessel; (f) thermostatic bath; (g) vacuum pump; (h) magnetic stirrer.

2.5. Thermodynamic analysis

Henry's constant (k_H) is an important parameter to present the absorption behavior of NH₃ in ILs. The Henry's constants of NH₃ absorbed in ionic liquids at different temperatures could be obtained by the following eqn (4) and the fugacity coefficient (φ) was calculated according to the P–R equation. Subsequently, the changes of standard Gibbs free energy (Δ_solG^θ), standard enthalpy (Δ_solH^θ) and standard entropy (Δ_solS^θ) could be estimated by the following eqn (6)–(8).

Δ_solG^θ = RT ln(k_H/p^θ) (6)

where f₁(T, p) represents the fugacity of NH₃ and φ₁(T, p) is the fugacity coefficient of NH₃ solute in the gas phase. p^eq is the equilibrium partial pressure of NH₃ and p^θ represents the standard pressure.

3. Results and discussion

3.1. Physical properties of ILs

Density and viscosity of IL were fundamental properties for gas separation process, hence these properties were measured at different temperatures from 298.15 K to 343.15 K. Fig. 3 shows the densities of the functionalized ILs decrease linearly with the increasing of temperature and they fall in the range from 1.15 to 1.58 g cm⁻¹. It can be found the densities are greatly affected by the different anions and [EtOHmim][DCA] shows the lowest density.

Fig. 3 Densities of task-specific ILs at different temperatures.

Fig. 4 shows the viscosities of the functionalized ILs with different anions range from 19.10 to 609.98 mPa s and decrease in an exponential manner with the increasing of temperature, which is in agreement with the other imidazolium ILs.³¹ Viscosity activation energy (E_a) of IL is estimated by a fitting of the measured viscosity to Arrhenius expression³² using the following equation:

η = η_∞ exp(−E_a/RT) (9)

Fig. 4 Viscosities of task-specific ILs at different temperatures.

In this equation, η is the viscosity at any temperature and η_∞ is the apparent viscosity at infinite temperature. The values of E_a for [EtOHmim]X shown in Table S1† and they range from 31 to 45 kJ mol⁻¹. Generally, the larger value will indicate the higher viscosity. The calculated viscosity activation energies of the task-specific ILs are larger than those of the typical solutions (E_{a water} = 17.0 kJ mol⁻¹, E_{a benzene} = 10.4 kJ mol⁻¹, and E_{a acetone} = 7.1 kJ mol⁻¹).³³ Although [EtOHmim][DCA] shows the lowest viscosity, the introducing of hydroxyl group results in the larger viscosity than that of [Emim][DCA] (18.4 kJ mol⁻¹), which indicates that cations also influence the viscosity.

In addition, thermal stability of IL is also very significant for gas separation, because high thermal stability is beneficial for reversible absorption and desorption. The TGA curves are shown in Fig. S2† and the results of DSC and TGA are summarized in Table 1. The thermal decomposition temperatures (T_d) of the six functionalized ILs are all above 500 K and [EtOHmim][NTf₂] shows the highest decomposition temperature. Pablo's group reported the T_d of [Emim][SCN] was 538.6 K,³⁴ which was similar to that of [EtOHmim][SCN]. It shows the introducing of hydroxyl is almost no influence on the thermal stability.

Table 1 The melting points (T_m), glass transition temperatures (T_g) and the thermal decomposition temperatures (T_d) of the functionalized ILs

[EtOHmim]X	Tm (K)	Tg (K)	Td (K)
[NTf2]	—	190.02	643.75
[PF6]	308.32	187.67	564.11
[BF4]	—	187.98	542.81
[DCA]	—	185.28	507.18
[SCN]	286.32	185.64	543.91
[NO3]	281.32	183.35	546.42

---

3.2. NH₃ absorption performance of [EtOHmim]X

First, NH₃ absorption capacity of [Bmim][BF₄] was investigated at 298.15 K and compared with the values reported from the available literatures in order to validate the reliability of data in this work. Fig. 5 shows the solubility is similar to that in literatures,^25,26 which indicates that the experiment apparatus is reliable and the data is credible. Then, a series of vapor liquid equilibrium measurements were performed at pressures from 0 to 0.6 MPa (A (absolute)) and temperatures from 298.15 to 343.15 K. The solubility data of pressure–temperature–composition (p–T–x) of the task-specific ILs are presented in ESI Table S2.†

Fig. 5 NH₃ absorption capacity of [Bmim][BF₄] in this work and the literatures^25,26 at 298.15 K.

3.2.1 Effect of hydroxyl group cation. To assess the effect of hydroxyl group cation on NH₃ solubility, the absorption capacity of functionalized ILs [EtOHmim]X were compared with that of conventional ILs [Emim]X at 333.15 K. The results show that the former captures higher amount of NH₃. Fig. 6 shows that when the anion is same, an obvious increase of solubility is found in [EtOHmim]X. For example, at 102.7 kPa, the solubility of [EtOHmim][BF₄] is 0.27 in mole fraction, and at 547.0 kPa, the mole fraction is 0.56. For [Emim][BF₄], the solubility at 103.1 kPa is 0.10 and at 548.2 kPa the mole fraction is 0.40. It suggests that the absorption capacities of imidazolium ILs are distinctly enhanced due to the introducing of hydroxyl group. A possible explanation is that hydrogen bonding interaction brings about the improved absorption performance. The acid hydroxyl, regarded as the hydrogen bonding donor group, could form stronger hydrogen bonding with the N atom of NH₃ than the conventional [Emim]X. In addition, because the viscosities of [EtOHmim]X are slightly larger than [Emim]X due to the introducing of hydroxyl group, the equilibrium time for NH₃ absorption in [EtOHmim][NTf₂] and [Emim][NTf₂] have been also measured respectively using the same apparatus. The variations of pressure with time are shown in Fig. S3.† We can find the equilibrium time of [EtOHmim][NTf₂] (about 6 min) is slightly longer than [Emim][NTf₂] (about 4 min). In conclusion, hydroxyl-functionalized ILs may be more suitable to be the solvents for NH₃ recovery than conventional ILs.

Fig. 6 Comparison of NH₃ solubilities in [EtOHmim]X and [Emim]X (X = [NTf₂]⁻, [BF₄]⁻, [NO₃]⁻) at 333.15 K.

3.2.2 Effect of anions. With the comparison of NH₃ solubilities in ILs with different anions, it is observed that anions also have an obviously effect on NH₃ absorption. Fig. 7 presents the ILs of anions containing fluoride ([NTf₂]⁻, [PF₆]⁻, [BF₄]⁻) have higher solubility than others ([DCA]⁻, [SCN]⁻, [NO₃]⁻). The absorption capacities of these ILs decrease in the following order: [NTf₂]⁻ > [PF₆]⁻ > [BF₄]⁻ > [SCN]⁻ > [NO₃]⁻. Especially, for [DCA] anion, when the pressure is less than 150 kPa, its absorption capacity is lower than that of [NO₃]⁻; but when the pressure is between 150 to 400 kPa, its absorption capacity falls in that between [SCN]⁻ and [NO₃]⁻. The similar conclusion can be obtained at other temperatures. The reason for the higher absorption capacities of ILs containing fluoride may be that the F atom of the anions can also form hydrogen bonding interaction with the H atom of NH₃.²⁷ From a molecular perspective, there are three hydrogen bonding donors on NH₃, four acceptors on [BF₄], six acceptors on [PF₆] and 10 acceptors on [NTf₂]. Therefore, the relatively stronger hydrogen bonding interaction may be formed between the H atom of NH₃ and [NTf₂] anion, which is consistent with the highest solubility of [EtOHmim][NTf₂]. Because of the highest absorption capacity of [EtOHmim][NTf₂], this ionic liquid is investigated as a typical example.

Fig. 7 The solubilities of NH₃ in task-specific ILs with different anions at 313.15 K.

3.2.3 Effects of temperature and pressure. The effects of temperature and pressure on NH₃ absorption in [EtOHmim][NTf₂] are displayed in Fig. 8. It can be seen that with the increasing of temperature, the absorption capacity of hydroxyl-functionalized ILs dramatically decreases. Besides, the molar fraction of NH₃ increases continuously with the increasing of pressure. It suggests that low temperature is better for NH₃ absorption and the desorption can be carried out at high temperature, which is in good agreement with the literature.²⁴

Fig. 8 The solubilities of NH₃ in [EtOHmim][NTf₂] at different temperatures.

3.3. Desorption and recycling

Recyclability is another important factor to evaluate an ionic liquid solvent and a key issue for potential industrial application, thus the absorption cycles of [EtOHmim][NTf₂] have been measured at 298.15 K and 100 kPa. First, the solubility was investigated in the fresh IL, then NH₃-saturated IL was desorbed under a vacuum of 0.15 kPa for 1 hour at 353.15 K and reused for NH₃ absorption. The absorption–desorption experiments were performed for five cycles. Fig. 9 shows that the solubility keeps almost unchanged after five cycles. It is revealed that [EtOHmim][NTf₂] can be reused with almost same capacity, which suggests the functionalized ILs have good recyclability.

Fig. 9 Recycling of [EtOHmim][NTf₂] for NH₃ absorption at 298.15 K and 100 kPa.

3.4. Thermodynamic properties of NH₃ in ILs

Henry's constants were calculated according to the solubility of NH₃. In Fig. 10, with the increasing of temperature, Henry's constants increase for each ionic liquid. The Henry's constants in the ILs containing fluoride are smaller than those of others, which accords with the higher NH₃ absorption capacity of the ILs containing fluoride. Thermodynamic parameters including standard enthalpy, standard Gibbs free energy, and standard entropy were also calculated and the results were presented in Table 2. Solution entropy represents the direction of disorder in the absorption process. Table 2 shows standard Gibbs free energy changes are positive and entropy changes are negative. Namely, at standard pressure, ammonia solution in [EtOHmim]X is a process of entropy decreasing. The value of ΔG increases with the increasing of temperature, which means higher temperature is unfavorable for NH₃ absorption. This conclusion is consistent with the experimental results. Solution enthalpy is an important parameter to evaluate the gas–liquid interaction. From Table 2, the enthalpy changes are negative, which indicates the solution at standard pressure is exothermal. And this conclusion corresponds with the ordinary solution rules for most gas. The enthalpy changes increases with the increasing of temperature, it demonstrates the heat release in this solution process decreases with the increasing of temperature. Usually, a higher value of enthalpy signifies a stronger gas–liquid interaction. But for [EtOHmim]X, the order of solubility is not as same as that of enthalpy. For example, the solubility of [EtOHmim][NTf₂] is significantly higher than [EtOHmim][NO₃], but the absolute value of enthalpy in [EtOHmim][NO₃]–NH₃ system is higher than that in [EtOHmim][NTf₂]–NH₃ system. Therefore, for the ionic liquids with different anions, not only the interaction between NH₃ and ILs but also the free volume of anions plays important influence on solubility.

Fig. 10 Henry's constants of NH₃ in task-specific ILs versus temperature.

Table 2 Thermodynamic parameters of NH₃ in [EtOHmim]X

T (K)	[EtOHmim][NTf2]	[EtOHmim][PF6]	[EtOHmim][BF4]	[EtOHmim][SCN]	[EtOHmim][NO3]
ΔsolG^θ (kJ mol^−1)
298.15	4.09	4.24	4.28	4.64	4.73
313.15	5.03	5.16	5.22	5.45	5.62
323.15	5.55	5.60	5.70	5.97	6.09
333.15	5.97	6.02	6.18	6.64	6.82
343.15	6.35	6.41	6.67	7.08	7.34
ΔsolH^θ (kJ mol^−1)
298.15	−9.39	−9.46	−9.83	−10.27	−11.01
313.15	−10.35	−10.44	−10.84	−11.33	−12.15
323.15	−11.03	−11.11	−11.55	−12.07	−12.94
333.15	−11.72	−11.81	−12.27	−12.83	−13.75
343.15	−12.43	−12.53	−13.02	−13.61	−14.59
ΔsolS^θ (J mol^−1K^−1)
298.15	−45.21	−45.94	−47.33	−50.02	−52.80
313.15	−49.14	−49.80	−51.30	−53.58	−56.75
323.15	−51.31	−51.72	−53.39	−55.82	−58.88
333.15	−53.09	−53.52	−55.38	−58.43	−61.74
343.15	−54.75	−55.20	−57.37	−60.29	−63.91

---

3.5. Mechanism of NH₃ absorption

The mechanism of NH₃ absorption was investigated by the comparison of FT-IR and ¹H NMR spectra of [EtOHmim]X before and after NH₃ absorption. Fig. 11 shows that there is no obvious change of FT-IR spectra for [EtOHmim][NTf₂] before and after NH₃ absorption. Furthermore, in situ FT-IR (shown in Fig. S4†) is applied to monitor the spectral variation of NH₃ absorption process for one hour, which also shows no distinct changes. Considering the absorption performance, the easy desorption behavior and no formation of new functional group, a possible explanation is that these novel ionic liquids absorb NH₃ through hydrogen bonding interaction, such as O–H⋯N.

Fig. 11 FT-IR spectra of [EtOHmim][NTf₂] before and after absorption of NH₃.

Meanwhile, ¹H NMR spectra of [EtOHmim]X before and after absorption of NH₃ were further studied. First, by comparing the ¹H NMR spectra of [Emim][NTf₂] and [EtOHmim][NTf₂] (seeing ESI Fig. S5†), the peak at 5.16 ppm is confirmed to be the shift of H₁₁ atom on the hydroxyl. Fig. 12 indicates the peak at 5.16 ppm disappears after NH₃ absorption and reappears after desorption. And the same phenomenon is also happened in the other functionalized ILs. That shows the shift of the H₁₁ on hydroxyl may be dramatically affected due to the strong interaction between the electronegative N atom of NH₃ and H₁₁ of hydroxyl.

Fig. 12 ¹H NMR spectra of [EtOHmim][NTf₂] before and after NH₃ absorption.

The interaction between hydroxyl-functionalized cation and NH₃ molecule was further studied by the quantum chemistry calculations at the B3LYP/6-311++G** level with Gaussian 09 software. By comparing [EtOHmim]⁺–NH₃ system with [Emim]⁺–NH₃ system, it can be found both the hydrogen bonding and interaction energy of [EtOHmim]⁺–NH₃ system are stronger than those of [Emim]⁺–NH₃ system. First, two representative structures of [EtOHmim] cation and NH₃ have been calculated. One is NH₃ located around H₁₁ atom, the other is NH₃ located around H₁ atom. The optimized structures are shown in Fig. 13a and b and they are different from those in Shi's work.²⁷ It can be found that the hydrogen bonding is formed between N atom of NH₃ and H₁₁ with distance of 1.810 Å, and the relative interaction energy is −56.32 kJ mol⁻¹. Similarly, the hydrogen bonding also can be formed between the N atom of NH₃ and H₁ with a distance of 2.131 Å, and the interaction energy is −47.42 kJ mol⁻¹. The shorter bonding length and higher interaction energy show that NH₃ prefers to interact with the H₁₁ atom of hydroxyl than H₁ atom for [EtOHmim] cation and NH₃ system. For comparison, the same calculation method is used for [Emim] cation and NH₃ system. The structures are shown in Fig. 13c and d. The interaction energy between NH₃ and [Emim] cation around H₁ atom is −46.16 kJ mol⁻¹, while that around H₁ is only −20.98 kJ mol⁻¹. What's more, the hydrogen bonding length between H₁ and the N atom of NH₃ (2.145 Å) is shorter than that between H₁₁ and the N atom of NH₃ (2.402 Å). The results show that NH₃ is more inclined to interact with H₁ atom of imidazole ring rather than H₁₁ of C₂H₅ group for [Emim] cation and NH₃ system. By the comparison of the two systems, the higher NH₃ absorption capacity of hydroxyl-functionalized ILs than that of conventional imidazolium ILs should be attributed to the strong hydrogen bonding and interaction energy.

Fig. 13 Optimized structures and interaction energies for [EtOHmim]⁺–NH₃ (a and b) and [Emim]⁺–NH₃ (c and d) systems. O, red; N, blue; H, white; N, gray. Hydrogen bonds are indicated by dotted lines, and distances are in angstroms.

4. Conclusion

A series of hydroxyl-functionalized imidazolium ILs were synthesized and their physical properties as well as NH₃ absorption performance were systematically investigated. Compared with the conventional imidazolium ILs, the introducing of hydroxyl group results into the slight increase of viscosity, good thermal stability and improved NH₃ absorption capacity. Among these ILs, [EtOHmim][NTf₂] has the highest NH₃ solubility, the mole fraction is 0.56 at 298.15 K and 159 kPa. The absorption cyclic experiments suggest these ILs can be regenerated. Furthermore, the absorption mechanism was detailedly studied by spectra analysis and quantum chemical calculations. The results demonstrate the excellent NH₃ absorption ability of these novel ILs is due to the strong hydrogen bonding between the N atom of NH₃ and the H atom of the hydroxyl on cation. Owing to the efficient absorption performance, simple preparation, good recyclability and excellent thermal stability, these hydroxyl-functionalized ILs could be promising candidates for NH₃ recovery.

Acknowledgements

The authors would like to acknowledge the National Basic Research Program of China (No. 2014CB744306), the National Natural Science Fund for Distinguished Young Scholars (No. 21425625), the Key Program of National Natural Science Foundation of China (21436010), and the Science and Technology Innovation Team of Cross and Cooperation of Chinese Academy of Sciences.

References

1. J. W. Erisman, A. Bleeker, J. Galloway and M. S. Sutton, Environ. Pollut., 2007, 150, 140–149.
2. C. Maton, N. D. Vos and C. V. Stevens, Chem. Soc. Rev., 2013, 42, 5963–5977.
3. C. P. Fredlake, J. M. Crosthwaite, D. G. Hert, S. N. Aki and J. F. Brennecke, J. Chem. Eng. Data, 2004, 49, 954–964.
4. C. N. Dai, W. J. Wei, Z. G. Lei, C. X. Li and B. H. Chen, Fluid Phase Equilib., 2015, 391, 9–17.
5. H. S. Gao, S. J. Zeng, X. M. Liu, Y. Nie, X. P. Zhang and S. J. Zhang, RSC Adv., 2015, 5, 30234–30238.
6. A. H. Jalili, M. Shokouhi, G. Maurer and M. Hosseini-Jenab, J. Chem. Thermodyn., 2014, 74, 286.
7. J. L. Anderson, K. D. Janeille and F. B. Joan, Acc. Chem. Res., 2007, 40, 1208–1216.
8. T. Nonthanasin, A. Henni and C. Saiwan, RSC Adv., 2014, 4, 7566.
9. M. Ramdin, S. P. Balaji, J. Manuel Vicent-Luna, J. J. Gutierrez-Sevillano, S. Calero, T. W. de Loos and T. J. H. Vlugt, J. Phys. Chem. C, 2014, 118, 23599–23604.
10. L. C. Tome, M. Isik, C. S. R. Freire, D. Mecerreyes and I. M. Marrucho, J. Membr. Sci., 2015, 483, 155–165.
11. Z. Z. Yang, Y. N. Zhao and L. N. He, RSC Adv., 2011, 1, 545.
12. P. J. Carvalho and J. A. P. Coutinho, Energy Fuels, 2010, 24, 6662–6666.
13. G. K. Cui, W. J. Lin, F. Ding, X. Y. Luo, X. He, H. R. Li and C. M. Wang, Green Chem., 2014, 16, 1211–1216.
14. G. K. Cui, C. M. Wang, J. J. Zheng, Y. Guo, X. Y. Luo and H. R. Li, Chem. Commun., 2012, 48, 2633–2635.
15. S. Y. Sun, Y. X. Niu, Q. Xu, Z. Sun and X. H. Wei, RSC Adv., 2015, 5, 46564–46567.
16. Z. Z. Yang, L. N. He, Q. W. Song, K. H. Chen, A. H. Liu and X. M. Liu, Phys. Chem. Chem. Phys., 2012, 14, 15832–15839.
17. S. J. Zeng, H. Y. He, H. S. Gao, X. P. Zhang, J. Wang, Y. Huang and S. J. Zhang, RSC Adv., 2015, 5, 2470–2478.
18. S. J. Zeng, H. Y. He, H. S. Gao and X. P. Zhang, RSC Adv., 2015, 5, 2470–2478.
19. Y. Zhao and G. X. Hu, RSC Adv., 2013, 3, 2234–2240.
20. M. A. Ahmadi, R. Haghbakhsh, R. Soleimani and M. B. Bajestani, J. Supercrit. Fluids, 2014, 92, 60–69.
21. C. A. Faundez, J. F. Diaz-Valdes and J. O. Valderrama, Fluid Phase Equilib., 2014, 375, 152–160.
22. K. Huang, Y. L. Chen, X. M. Zhang, S. L. Ma, Y. T. Wu and X. B. Hu, Fluid Phase Equilib., 2014, 378, 21–33.
23. Y. Q. Ma and R. Wang, Chem. Res. Chin. Univ., 2014, 35, 1515–1522.
24. A. Yokozeki and B. S. Mark, Ind. Eng. Chem. Res., 2007, 46, 1605–1610.
25. A. Yokozeki and M. B. Shiflett, Appl. Energy, 2007, 84, 1258–1273.
26. G. H. Li, Q. Zhou, X. P. Zhang, L. Wang, S. J. Zhang and J. W. Li, Fluid Phase Equilib., 2010, 297, 34–39.
27. W. Shi and E. J. Maginn, AIChE J., 2009, 55, 2414–2421.
28. J. Palomar, G. Miquel-Maria, J. Bedia, F. Rodriguez and J. J. Rodriguez, Sep. Purif. Technol., 2011, 82, 43–52.
29. S. K. Tang, G. A. Baker and H. Zhao, Chem. Soc. Rev., 2012, 41, 4030–4066.
30. J. Z. Zhang, C. Jia, H. F. Dong, J. Q. Wang, X. P. Zhang and S. J. Zhang, Ind. Eng. Chem. Res., 2013, 52, 5835–5841.
31. M. Zhu, M. X. Zhu, Y. Song, W. Z. Liang, W. Han and Y. N. Chen, Engineering and manufacturing technologies, 2014, 541–542, 78–82.
32. P. Brown, B. E. Gurkan and T. A. Hatton, AIChE J., 2015, 61, 2280–2285.
33. L. J. Wazer JRV, K. Y. Kim and R. E. Colwell, Viscosity and Flow Measurement, New York: Wiley, 1963, 406.
34. P. Navarro, M. Larriba, E. Rojo, J. García and F. Rodríguez, J. Chem. Eng. Data, 2013, 58, 2187–2193.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra13730f

---