Article Cite This: Langmuir 2017, 33, 11719-11726
pubs.acs.org/Langmuir
Confinement Phenomenon Effect on the CO2 Absorption Working Capacity in Ionic Liquids Immobilized into Porous Solid Supports Nanhua Wu,†,‡ Xiaoyan Ji,‡ Wenlong Xie,§ Chang Liu,† Xin Feng,*,† and Xiaohua Lu*,† †
State Key Laboratory of Materials-Oriented and Chemical Engineering and Jiangsu National Synergetic Innovation Center for Advanced Materials (SICAM), Nanjing Tech University, Nanjing 210009, China ‡ Energy Engineering, Division of Energy Science, Luleå University of Technology, 97187 Luleå, Sweden § China Petroleum Chemicals Kunshan Company, No. 210, Kuntai Road, Kunshan 215337, China
ABSTRACT: In this work, the CO2 absorption working capacity and solubility in ionic liquids immobilized into porous solid materials (substrates) were studied both experimentally and theoretically. The CO2 absorption working capacity in the immobilized ionic liquids was measured experimentally. It was found that the CO2 absorption working capacity and solubility increased up to 10-fold compared to that in the bulk ionic liquids when the film thickness was nearly 2.5 nm in the [HMIm][NTf2] immobilized in the P25. Meanwhile, a new model was proposed to describe the Gibbs free energy of CO2 in the immobilized ionic liquids, and both macro- and microanalyses of the CO2 solubility in the confined ionic liquids were conducted. The theoretical investigations reveal that the substrate has a crucial effect on the gas solubility in the ionic liquid immobilized into the substrates, and the model performance was approved with a consideration of the substrate effect.
1. INTRODUCTION CO2 separation plays an important role in capturing CO2 from the combustion of fossil fuels or producing transportation fuels via biomass gasification.1 However, the cost of separating CO2 is very high using the available technologies.2,3 Ionic liquids (ILs) have been proposed as promising solvents for CO2 separation because of their unique high thermal stability, low volatility, high chemical stability, and nonflammability.4−8 More importantly, the number of ILs that can be synthesized was estimated to be nearly 1018, providing an enormous scope for scientific innovation.9 A lot of research work have been carried out to develop IL for CO2 separation.10 Brennecke et al. investigated the CO2 solubility in ILs at room temperature (RTILs);4,11,12 however, the solubility at atmospheric pressure is a maximum of 3 mol % but increases with increasing pressure.13 To improve the CO2 absorption capacity of ILs at low pressures, task-specific ILs have been developed that can be classified into novel cation designing, novel anion designing, and cation and/or anion functionalizing.5,14−20 For example, Davis et al. synthesized the functional ILs, and the CO2 solubility was increased to 0.5 mol/ mol of IL.5 Jiang et al. synthesized tetraalkylammonium-based © 2017 American Chemical Society
amino acid ILs successfully, which can improve both the reaction and mass transfer rates of CO2 in the ILs.14 However, several drawbacks still need to be solved before practical applications of IL-based technologies, e.g. the high viscosity leading to an equilibrium time of as long as 180 min,7 the high price of ILs, and so on.21 Recently, it has been found that IL immobilization into porous solid supports could significantly enhance the mass transfer rate of CO2 in ILs22−24 and reduce the quantity of ILs needed for CO2 separation.25 For example, the equilibrium time was shortened to less than 15 min for IL-immobilized sorbents.26 Meanwhile, supported ionic liquid membranes (SILMs) that combine the advantages of both membranes and ILs for the selective separation of gases have received growing attention in recent years.27−30 The unique properties of ILs including low vapor pressures, high thermal stability, and Special Issue: Tribute to Keith Gubbins, Pioneer in the Theory of Liquids Received: June 27, 2017 Revised: August 14, 2017 Published: August 28, 2017 11719
DOI: 10.1021/acs.langmuir.7b02204 Langmuir 2017, 33, 11719−11726
Article
Langmuir
Gases were dried by P2O5 before being used in the TGA. P25 was purchased from Degussa, and PMMA was from Sigma-Aldrich. The supports were dried and degassed under vacuum at 80 °C for 8 h before use. 2.2. Preparation of Immobilized IL Sorbents. Immobilizing ILs into the porous supports can be divided into a chemical or physical method,38,39 i.e., by chemical attachment of the ionic species to a solid support or by physical fixation of IL within a support.40−43 In this work, the physical immobilization method was conducted with impregnation according to previous research.26,43 In immobilization, the IL was dissolved in methanol and then the solid particles (substrates) were added in different weight ratios to form a slurry. The formed slurry was placed in a rotary evaporator (RE-52AA, Rongsheng, China) under vacuum in a water bath at 60 °C to remove methanol and resulted in the physical adsorption (immobilization) of IL into the pores of substrates. The prepared sorbents were stored in a desiccator before use. The exact IL loading in the P25-IL sorbents was further determined using TGA (TG209-F3, ± 0.0001 mg, Netzsch). Specifically, approximately 20 mg of each IL-immobilized sorbent was placed in an Al2O3 sample cell filled with flowing N2 at a flow rate of 40 mL/ min. The samples were then heated from 35 to 950 °C at a heating rate of 10 °C/min to remove IL. The IL loading was determined accurately on the basis of the weight before and after heating. However, the [APMIm]Br loading in the PMMA-[APMIm]Br sorbents cannot be determined using TGA because PMMA will decompose at high temperatures. Therefore, the feed ratios in the preparation process were used as the IL loadings. On the basis of the results in our previous work, the immobilization ratio can be well controlled by the feed ratio in the impregnation process, and the error was acceptable.43 The prepared sorbents and the loading weights for the corresponding immobilized ILs are summarized in Table 1.
a strong capillary force existing between ILs and support pores make SILMs more competitive than conventional supported liquid membranes (SLMs). In both of these two options, the scale can be down to nanometers, and the quantity of ILs needed is small, which can avoid the high cost from the synthesis of the huge quantity of ILs. However, previous research was mainly focused on experimental measurements, and the effect of the immobilization or membranes down to a small scale on the properties of ILs including gas solubility has not been well examined. However, research from other experiments and molecular simulations has revealed that the thermodynamic properties of the fluids (e.g., the gas solubility at equilibrium) depend on both the size (or film thickness) of the fluids and the substrate contacted when the scale goes down to nanometers (i.e., nanoeffect).31,32 For example, the H2 solubility in the ethanol confined in the 4 nm pores was increased 4−5-fold over the bulk solubility.31 Zhang et al. studied the CO2 solubility in the [BMIm][BF4] confined in mesoporous silica gels, and they observed that the CO2 solubility increased 1.5-fold over the bulk solubility.33 The CO2 permeability in the SILMs was 4fold greater than in the bulk as reported by Close et al.34 Baltus et al. found that when the membrane thickness was 250 μm the CO2 solubility in [C2C1IM][NTf2] or [C6C1IM][NTf2] increased 1.6−2-fold.35 On the basis of the molecular dynamic simulation, Hung et al. investigated the structural and dynamical properties of the confined IL and found that the properties are different from the bulk properties as a function of pore size and pore loading.36 Shi et al. studied the properties of the IL confined in carbon nanotubes. It was found that the molar volume was 12−31% larger than that in the bulk phase, and the solubility of CO2, H2, or N2 was 1.1−3-fold higher.37 However, the research is still limited. For example, only one study has been conducted on the CO2 solubility in confined ILs. Furthermore, no theoretical model has been developed to represent this abnormal phenomenon. The goal of this work was to study the confinement phenomenon effect on the CO2 solubility in the ILs immobilized into porous solid supports experimentally and theoretically. To achieve this, three ILs {1-hexyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([HMIm][NTf2]), 1-butyl-3-methylimidazolium acetate ([BMIm][Ac]), and 1-aminopropyl-3-methylimidazolium bromide ([APMIm]Br)} were loaded into two porous materials (substrates) by means of simple impregnation under ambient conditions. The CO2 absorption capacities of the prepared sorbents with different IL loadings were detected precisely by thermogravimetric analysis (TGA), and the CO2 solubilities in these ILs confined in the porous sorbents were obtained. A new model was developed to describe the Gibbs free energy of CO2 in the immobilized ILs with the consideration of both surface and substrate effects, and the model was further used to analyze the contribution of each part to the CO2 solubility in the confined ILs.
Table 1. Summary of the IL Loadings of the Samples substrate
ionic liquid
loadings, wt %
P25 P25 P25 PMMA
[HMIm][NTf2] [BMIm]Ac [APMIm]Br [APMIm]Br
11.4, 20.7, 39.6, 49.0 32.6, 38.2, 45.8 25.8, 37.2, 40.4, 44.2 15, 25, 35
2.3. Characterization and Testing. The physical properties of substrates were characterized as described in the following text. The substrates were degassed under vacuum at 80 °C for 8 h. The adsorption isotherms of nitrogen (N2) were then measured at −195 °C with a Micromeritics Tristar II 3020 analyzer (Micromeritics, USA). On the basis of the measured adsorption isotherms, the Brunauer−Emmett−Teller (BET) method was used to estimate the specific surface area (SBET). The amount of CO2 absorbed in the prepared immobilized IL sorbents was detected using TGA.44−49 In a typical test, approximately 20 mg of the immobilized IL sorbent was placed in the Al2O3 sample cell. The temperature was increased at a rate of 10 °C/min from 35 to 105 °C and kept at 105 °C until no weight loss was observed under dry helium gas at a flow rate of 100 mL/min. This process was used to remove the solvent, moisture, or other adsorbents from the immobilized IL sorbents. The temperature was then cooled at a rate of 5 °C/min to the desired absorption temperature, and the gas flow was switched from helium gas to a mixture of dry CO2 and helium gases with a flow rate of 100 mL/min and maintained at the desired temperature for CO2 absorption detection. The weight of the sorbent was recorded continuously. On the basis of the recorded weight during the absorption processes, the amount of absorbed CO2 in g-CO2/gsorbent (support + ILs) was then obtained. It is worth noting that the CO2 absorption time was set as 90 min, which is generally enough for the immobilized IL sorbents to approach absorption equilibrium.50 The estimated uncertainty in the determined absorption capacity is 5% as a result of the measured errors on determining the absorbed
2. EXPERIMENT 2.1. Materials. The ILs of [HMIm][NTf2] and [APMIm]Br (>99 wt %) were purchased from Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, China. [BMIm]Ac, with the same purity as used by others (purity ≥96.0 wt %, water ≤0.5%), was purchased from Sigma-Aldrich. Methanol (analytical reagent) was purchased from Lingfeng Chemical Reagent Co., China. Gaseous N2, He, and CO2 were supplied by Nanjing Sanle Gas (99.99% purity). 11720
DOI: 10.1021/acs.langmuir.7b02204 Langmuir 2017, 33, 11719−11726
Article
Langmuir CO2 mass (±0.0001 mg,
