We investigated the influences of the changes in the electronic structure and size of the anion of an imidazolium ionic liquid (IL) on gas adsorption and separation performance of the IL/ZIF-8 (zeolitic imidazolate framework) composites
Trang 1Available online 2 July 2020
1387-1811/© 2020 The Authors Published by Elsevier Inc This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Influence of anion size and electronic structure on the gas separation
performance of ionic liquid/ZIF-8 composites
Muhammad Zeeshana,b, Harun Kulaka,b, Safiyye Kavakb,c, H Mert Polatb,c, Ozce Duraka,b,
Seda Keskina,b,*, Alper Uzuna,b,d,**
aDepartment of Chemical and Biological Engineering, Koç University, Rumelifeneri Yolu, 34450, Sariyer, Istanbul, Turkey
bKoç University TÜPRAŞ Energy Center (KUTEM), Koç University, Rumelifeneri Yolu, 34450, Sariyer, Istanbul, Turkey
cDepartment of Materials Science and Engineering, Koç University, Rumelifeneri Yolu, 34450, Sariyer, Istanbul, Turkey
dKoç University Surface Science and Technology Center (KUYTAM), Koç University, Rumelifeneri Yolu, 34450, Sariyer, Istanbul, Turkey
A R T I C L E I N F O
Keywords:
Ionic liquids (ILs)
Metal organic frameworks (MOFs)
Gas adsorption
Porous material
Hybrid composites
A B S T R A C T
We investigated the influences of the changes in the electronic structure and size of the anion of an imidazolium ionic liquid (IL) on gas adsorption and separation performance of the IL/ZIF-8 (zeolitic imidazolate framework)
composites We studied four different imidazolium ILs having the same cation, 1-n-butyl-3-methylimidazolium,
[BMIM]þ, with anions having structures allowing a systematic comparison of the changes in the electronic structure and size To examine the influence of changes in the electronic structure, we considered anions rep-resenting the fluorination on the anion, methanesulfonate, [MeSO3] , and trifluoromethanesulfonate, [CF3SO3] To investigate the change in the anion size, methyl sulfate, [MeSO4] , and octyl sulfate, [OcSO4] , were studied Characterization of IL/ZIF-8 composites demonstrated successful incorporation of each IL in ZIF-8 without causing any detectable changes in the crystal structure and morphology of ZIF-8 Thermogravimetric analysis and infrared (IR) spectroscopy indicated the presence of direct interactions between ILs and ZIF-8, which directly control gas separation performance of the composite Gas adsorption measurements illustrated that incorporation of ILs significantly improves the gas separation performance of the pristine ZIF-8 [BMIM] [MeSO4]/ZIF-8 composite had 3.3- and 1.8-times higher CO2/N2 and CH4/N2 selectivities compared to ZIF-8, respectively, at 1 bar When the IL has a fluorinated anion, CO2/CH4 selectivity improved 3-times compared
to its non-fluorinated counterpart Upon the incorporation of IL with a small anion, IL/ZIF-8 composite showed higher CO2/N2 and CH4/N2 selectivities compared to the composite having an IL with a bulky anion These results will contribute in guiding rational design of IL/MOF composites for different gas separations
1 Introduction
Excessive combustion of fossil fuels led to a significant increase in
CO2 concentration in the atmosphere This increase is the main reason
for the climate change and global warming [1–4] Moreover,
purifica-tion of natural gas is a crucial process because the presence of impurities,
such as CO2, reduces the total calorific value of natural gas and promotes
the corrosion in pipelines and equipment [5] Among the existing CO2
capture and separation technologies, adsorption-based gas separation
process by nanoporous materials has emerged as an energy- and
cost-effective technology [6,7] Thus, it is critical to design and
syn-thesize novel microporous materials that have a potential to selectively
capture CO2 from a mixture of gas streams, such as CH4 and N2 Metal organic frameworks (MOFs), a novel class of porous crystalline mate-rials, have been recently considered for the capture and separation of
CO2 from gas mixtures containing CH4 and N2 as alternatives to tradi-tional adsorbents, such as zeolites, alumina, silica gels, carbon molec-ular sieve, and carbon nanotubes [8–11] Furthermore, owing to the ability of changing the metal nodes and linkers, MOFs offer large surface areas, high pore volumes, variety of pore sizes and shapes, and reason-able chemical and thermal stabilities [12,13] Several studies demon-strated tuning of the physicochemical properties of a parent MOF by various post-synthesis modification techniques, such as amine func-tionalization, metal, and ligand exchange, and surface functionalization
* Corresponding author Department of Chemical and Biological Engineering, Koç University, Rumelifeneri Yolu, 34450, Sariyer, Istanbul, Turkey
** Corresponding author Koç University TÜPRAŞ Energy Center (KUTEM), Koç University, Rumelifeneri Yolu, 34450, Sariyer, Istanbul, Turkey
E-mail addresses: skeskin@ku.edu.tr (S Keskin), auzun@ku.edu.tr (A Uzun)
Contents lists available at ScienceDirect Microporous and Mesoporous Materials journal homepage: http://www.elsevier.com/locate/micromeso
https://doi.org/10.1016/j.micromeso.2020.110446
Received 25 May 2020; Received in revised form 25 June 2020; Accepted 28 June 2020
Trang 2[14–17] Among these approaches, post-synthesis modification of MOFs
by combining them with ionic liquids (ILs) has offered a broad prospect
in tuning gas adsorption and separation performance of a parent MOF
[18–20] ILs are novel solvents that are composed of cations and anions,
and generally have lower melting points than 100 �C [21] The unique
properties of ILs, such as low vapor pressure, high thermal stability, and
tunable physicochemical properties, offer a broad potential for various
applications, such as catalysis [22], lubricants [23], electrolytes [24],
sensors [25], and gas adsorption and separation processes [26,27]
Among these applications, post-synthesis modifications of MOFs by
combining them with ILs offer opportunities especially for designing
novel materials having a high performance in CO2 adsorption and
sep-aration because of the high solubility of CO2 in most ILs
Several studies reported that upon the incorporation of ILs into the
pores of a MOF, gas adsorption and separation performance of the
parent MOF improved significantly [28–37] For instance,
1-n-butyl-3-methylimidazolium tetrafluoroborate ([BMIM][BF4]) was
incorporated into copper benzene-1,3,5-tricarboxylate (CuBTC) and
zeolite imidazolate framework, ZIF-8 [28,29] Results showed
im-provements in gas separation performance of both MOFs Mohamedali
et al [30–32] reported impregnation of 1-n-butyl-3-methylimidazolium
acetate ([BMIM][Ac]) and 1-ethyl-3-methylimidazolium acetate
([EMIM][Ac]) into CuBTC, ZIF-8, and MOF-77 Results demonstrated an
improved CO2 adsorption capacity and CO2/N2 separation performance
for each IL/MOF composite Our group reported that incorporation of
1-n-butyl-3-methylimidazolium hexafluorophosphate ([BMIM][PF6])
and 1-n-butyl-3-methylimidazolium thiocyanate ([BMIM][SCN])
resul-ted in enhanced CO2/CH4, CO2/N2, and CH4/N2 separation
perfor-mances of ZIF-8 [33,34] Ma et al [35] studied incorporation of a
task-specific IL 1-(3-aminopropyl)-2-butylimidazolium
tris(tri-fluoromethanesulfonyl)methide ([C3NH2BIM][Tf2N]) into chromium 1,
4-benzenedicarboxylate (NH2-MIL-101(Cr)) and reported an
improve-ment in CO2/N2 selectivity In a similar study, Ding et al [36] explored
the incorporation of imidazolium-based poly (ionic liquid)s (polyILs)
into MIL-101 and reported an improvement in both CO2 uptake capacity
and CO2/N2 separation performance Besides, our group recently
examined six different imidazolium-based ILs by incorporating them
into aluminum 1,4-benzenedicarboxylate (MIL-53(Al)) and reported
increased CO2/CH4 and CO2/N2 selectivities compared to those of
parent MOF [37,38]
Studies discussed above imply that gas adsorption capacity and gas
separation performance of IL/MOF composites are significantly
controlled by the IL‒MOF interactions Furthermore, these studies also
suggest that anion part of the IL is dominant in controlling the IL‒MOF
and IL‒adsorbate interactions None of these studies, however, focused
on systematically investigating the structural changes on the individual
components of IL/MOF composites and their consequences on the gas
separation performance of the materials Such investigations potentially
provide insights on the structure-performance relationships of these
composites and, therefore, they are crucial for the rational design of IL/
MOF composites with a high gas separation performance To contribute
into this field, in this work, we geared at examining the impact of
sys-tematic changes on the electronic structure and size of the anion of an
imidazolium-type IL on the gas separation performance of the
corre-sponding IL/MOF composite
We studied four different imidazolium ILs having the same [BMIM]þ
cation and different anions (methanesulfonate, [MeSO3] ;
tri-fluoromethanesulfonate, [CF3SO3] ; methyl sulfate, [MeSO4] ; octyl
sulfate, [OcSO4] ) and incorporated them into ZIF-8 at comparable IL
loadings We chose ZIF-8, as this MOF offers a versatile platform for the
incorporation of ILs [33,34,39] The composites were prepared by the
post-synthesis modification as before [29] and characterized in detail by
combining different experimental techniques, such as X-ray diffraction
(XRD), Brunauer-Emmett-Teller (BET) analysis, scanning electron
mi-croscopy (SEM), thermogravimetric analysis (TGA), X-ray fluorescence
(XRF) and infrared spectroscopies (IR) Afterward, to examine the gas
adsorption and separation performance of the pristine ZIF-8 and IL/ZIF-8 composites, volumetric adsorption measurements for CO2, CH4, and N2 were performed Results showed that IL/ZIF-8 composite with a fluorinated anion led to an improved CO2/CH4 separation performance, whereas incorporation of the IL with a smaller anion into ZIF-8 resulted
in a superior CO2/N2 and CH4/N2 separation performance Character-ization data indicated that these changes in separation performance are directly related with the changes in the interactions between the IL molecules and the ZIF-8 cage These results illustrate that the changes in both the electronic environment and the size of the IL’s anion play a significant role in determining the interactions and their consequences
on the separation performance Thus, they provide much needed in-sights for the rational design of IL/MOF composites having improved gas separation performances
2 Materials and methods
2.1 Materials
All the ILs, ZIF-8 (Basolite Z1200, 2-methylimidazole zinc salt), and acetone were obtained from Sigma–Aldrich and stored in an Ar-filled glove box (Labconco) CH4 (99.95%), CO2 (99.9%), N2 (99.998%), and He (99.999%) were purchased from Air Liquide
2.2 Sample preparation
Pristine ZIF-8 was first activated at 105 �C overnight under vacuum prior to incorporation of the IL Each IL/ZIF-8 composite was prepared
by wet impregnation, as previously reported [33] The IL/ZIF-8 com-posites with a targeted IL loading of 30 wt% were prepared by dissolving
300 mg of IL in 20 mL of acetone by stirring for 1 h under ambient conditions Then 700 mg of dehydrated ZIF-8 powder was added to the solution and the mixture was stirred at 35 �C in an open atmosphere allowing the solvent to evaporate itself at a slow pace After the solvent was completely evaporated, the resulting IL/ZIF-8 composites were further dried at 105 �C overnight The synthesized IL/ZIF-8 composites were stored in a desiccator
2.3 X-ray fluorescence (XRF) spectroscopy
The elemental analyses of the IL/ZIF-8 composites were conducted
on a Bruker S8 Tiger spectrometer The analyses were performed under
He atmosphere and an X-ray tube with 4 kW Rh anode was used to generate X-rays SpectraPlus Eval2 V2.2.454 software was used for the interpretation of obtained data
2.4 Brunauer-Emmett-Teller (BET) analysis
A Micromeritics ASAP 2020 was utilized to determine the surface area and pore volume from the N2 adsorption isotherms obtain at 196
�C for pristine ZIF-8 and IL/ZIF-8 composites Prior to each measure-ment, 150 mg of sample was degassed at 125 �C for 12 h under vacuum
N2 adsorption isotherm was obtained between the pressure range of
10 6 and 1 bar The BET equation and the t-plot method were used to
calculate surface area and pore volume of the samples in the relative pressure range 0.05–0.65
2.5 Scanning electron microscopy (SEM)
SEM images of the pristine ZIF-8 and its composites with the ILs were obtained with a Zeiss Evo LS 15 using an accelerating voltage of 3 kV under vacuum The sample surfaces were sputtered with gold prior to each measurement The SEM images were obtained at two different magnifications (100 k� and 25 k� )
Trang 32.6 X-ray diffraction (XRD) spectroscopy
XRD pattern of pristine ZIF-8 and IL/ZIF-8 composites were obtained
using a Bruker D8 Advance instrument with Cu-Kα1 radiation (λ ¼
1.5406 Å) operating at a voltage of 30 kV and a current of 10 mA Each
diffraction pattern was collected in a 2θ range of 5–50�, with a step size
of 0.0204�
2.7 Thermal gravimetric analysis (TGA)
TGA of the pristine ZIF-8, bulk ILs, and IL/ZIF-8 composites were
performed on a TA Instruments Q500 thermogravimetric analyzer The
analysis was carried under N2 atmosphere of 40 and 60 mL min 1 for
balance and purge gases, respectively After taring the pan,
approxi-mately 10 mg of each sample was loaded into a platinum pan and
temperature was increased from room temperature to 100 �C at a ramp
rate of 5 �C min 1 After an isothermal treatment for 8 h at 100 �C,
temperature was further increased at a ramp rate of 2 �C min 1 to 700
�C For comparing thermal decomposition temperature of the samples,
the thermal decomposition temperatures, the onset (Tonset) and
deriva-tive onset (T0 onset) temperatures were determined from the
thermogra-vimetric (TG) and derivative TG curves In this study, derivative onset
temperatures (T0onset) were considered for comparison analysis, because
onset temperature values (Tonset) generally overestimate the
decompo-sition temperature, as previously reported [40,41]
2.8 Infrared spectroscopy (IR)
IR spectra of the pristine ZIF-8 and IL/ZIF-8 composites were
recorded using a Bruker Vertex 80v FTIR spectrometer averaging 512
scans collected at a spectral resolution of 2 cm 1 Sample was loaded
between two potassium bromide (KBr) windows in an IR cell, and
ana-lyses were performed under vacuum at room temperature to obtain IR
spectra between 650 and 4000 cm 1 in transmission mode IR bands
deconvolution was performed using Fityk software by employing the
Voigt function [42]
2.9 Conductor-like screening model for realistic solvents (COSMO-RS)
calculations
To predict the CO2, CH4, and N2 solubilities, we used the
COSMO-ThermX software, version C30_160 [43] The gas solubilities were
calculated in a pressure range of 0.1–1 bar at 25 �C These calculations
were performed using the TZVP parameterizations, whereas the
solu-bility values were obtained from the activity coefficients
2.10 Gas adsorption measurements
A High-Pressure Volumetric Analyzer (Micromeritics HPVA II-200)
was used to perform single-component gas adsorption measurements
of samples for CO2, CH4, and N2 gases For each measurement,
approximately 300 mg of sample was loaded into the sample holder and
degassed overnight at 150 �C under vacuum After degassing, the system
was purged with He gas three-times to remove the unwanted residual
gases from the previous measurement Afterward, adsorption isotherm
of CO2, CH4, and N2 gases were obtained in a pressure range of 0.1–1 bar
at 25 �C Gas adsorption isotherms were fitted to the dual-site Langmuir
(DSL), Langmuir-Freundlich (LF), and dual-Site Langmuir-Freundlich
(DSLF) models using Ideal Adsorbed Solution Theory (IAST)þþ[44],
software to calculate the ideal and mixture CO2/CH4, CO2/N2, and
CH4/N2 selectivities Fitting parameters for gas adsorption isotherms are
provided in Table S1 of Supporting Information (SI)
3 Results and discussion
The elemental composition of the composites determined within the
error range of XRF measurements are presented in Table 1 Data showed that each IL/ZIF-8 composite had an IL loading of 25.5 � 1.5 wt% This amount matches the amount reported to be the highest IL loading on IL/ ZIF-8 composites that can be achieved before overfilling the pores of ZIF-
8 to exceed the wetness point [28,29]
Fig 1 represents the N2 adsorption-desorption isotherms of the pristine ZIF-8 and the IL/ZIF-8 composites measured at 196 �C The results presented in Fig 1 showed typical type-I isotherms without any profound hysteresis loop for pristine ZIF-8 and IL-incorporated com-posites These observations suggest that upon the incorporation of ILs, ZIF-8 retains its microporous feature in each IL/ZIF-8 composite BET surface areas and pore volumes obtained from the correspond-ing N2 isotherms for pristine MOF and IL-incorporated MOF composites are summarized in Table 2 The data showed that the BET surface areas and pore volumes of IL/ZIF-8 composites are notably lower than those of the pristine ZIF-8, as observed in previous reports [33] This difference can be attributed to the successful incorporation of IL molecules into MOF pores, thereby reducing the overall N2 uptake However, it is noteworthy that overall N2 uptake is also dependent on the solubility of
N2 in the ILs Thus, because of the poor solubility of N2 in the IL, espe-cially at the measurement conditions, some IL molecules located at the gate openings of the ZIF-8 might block the accessibility of N2 molecules into the completely/partially available MOF pores [45] Therefore, we note that the BET measurements of IL/ZIF-8 composites are not very reliable even though they consistently present a decrease in surface area upon the incorporation of IL
Fig 2 shows the SEM images of the pristine ZIF-8 and IL/ZIF-8 composites demonstrating the surface morphologies of the materials Accordingly, SEM images of IL-incorporated ZIF-8 composites clearly show that the rhombic dodecahedron structure of pristine ZIF-8 was preserved upon the incorporation of IL
To further evaluate the crystal structure of the samples, XRD patterns
of the pristine ZIF-8 and IL/ZIF-8 composites were obtained as shown in Fig 3 Results showed that all the characteristics peaks of ZIF-8 were intact for each IL/MOF composite, thus, it can be inferred that the crystallinity of ZIF-8 was well-maintained upon the incorporation of ILs However, compared to pristine ZIF-8, the intensities of the diffraction peaks were slightly different in the IL-incorporated MOF composites, implying the presence of possible changes in the electronic structure inside the MOF pores or in the crystal orientation
Next, we investigated the thermal stabilities of the IL/MOF com-posites TGA measurements were performed for the pristine MOF, the bulk ILs, and the IL/MOF composites as shown in Fig 4 The derivative
onset temperatures (T0onset) for pristine ZIF-8, bulk ILs, and IL/ZIF-8 composites were determined from the derivative TG curves
Results presented in Fig 4 showed a typical one-step decomposition for ZIF-8 and bulk ILs, whereas a two-step decomposition mechanism was observed in the TGA curves of IL/ZIF-8 composites The initial weight loss between 100 and 150 �C observed on each TGA curve can be attributed to the evaporation of the moisture content in each sample
Accordingly, the T0 onset of pristine ZIF-8, bulk [BMIM][MeSO3], [BMIM] [CF3SO3], [BMIM][MeSO4], and [BMIM][OcSO4] were found as 375,
278, 324, 302, and 252 �C, respectively Upon incorporation of the ILs into ZIF-8, thermal stability of each IL/ZIF-8 decreased compared to that
of pristine ZIF-8 Accordingly, T0onset for [BMIM][MeSO3]/ZIF-8, [BMIM][CF3SO3]/ZIF-8, [BMIM][MeSO4]/ZIF-8, and [BMIM][OcSO4]/ ZIF-8 composites were found to be 257, 315, 241, and 242 �C,
respec-tively Thus, these changes in T0onset values and in the total weight losses indicate changes in the decomposition mechanisms in the composites, which confirm the presence of IL‒MOF interactions These results are consistent with a comprehensive report on the structural factors deter-mining the thermal stabilities of the IL/MOF composites, reported pre-viously [41] To further identify these IL‒MOF interactions, IR spectra of the pristine MOF, bulk IL, and the IL/ZIF-8 composite were acquired and examined in detail Fig 5 shows the IR spectra of pristine ZIF-8, bulk IL, and IL-incorporated ZIF-8 composites in the spectral regions of
Trang 42800–3200 cm 1 and 650–1800 cm 1
Appearance of all of the characteristic peaks of each IL in the IR
spectra of the IL/ZIF-8 composites in Fig 5 further confirms the
suc-cessful incorporation of ILs into framework To examine the influence of
the fluorination and size change of the IL’s anion on the IL‒MOF
in-teractions, we thoroughly analyzed the changes in the band positions of
the IR features related to the corresponding anion First, we investigated
the influence of fluorination of IL’s anion on the IL‒MOF interactions,
comparing the data related with [BMIM][MeSO3] and its counterpart
with the fluorinated anion, [BMIM][CF3SO3] In the lower region of the
IR spectrum of bulk [BMIM][MeSO3], the peaks at 1037 and 1170 cm 1
correspond to asymmetric νas (—SO3) and symmetric νs (—SO3)
stretching modes of the anion of IL, respectively; whereas the
corre-sponding peaks for [BMIM][CF3SO3] were located at 1224 and 1250
cm 1 [37] In the case of non-fluorinated anion, both νas (—SO3) and νs
(—SO3) modes presented red-shifts of 3 and 4 cm 1 in the IR spectra of
[BMIM][MeSO3]/ZIF-8 composite, respectively; whereas no shifts were
observed in the corresponding IR bands for the composite containing the
fluorinated IL However, the band at 1154 cm 1 corresponding to νas
(—CF3) in [BMIM][CF3SO3] presented a major blue-shift of 12 cm 1 in
the IR spectra of [BMIM][CF3SO3]/ZIF-8 composite From these
obser-vations, we infer that incorporation of IL with a non-fluorinated anion,
the interionic interaction between the sulfonate groups of IL’s anion
becomes weaker as indicated by red-shifts, as the electrons involving in
these interactions were shared with the MOF However, in the case of the composite involving the IL with a fluorinated anion, the interionic in-teractions between the sulfonate groups become stronger as evidenced
by a major blue-shift This increase in strengthening of the interionic interaction can be attributed to the presence of highly electronegative character of fluorine atoms, which probably attracts electrons from the MOF Furthermore, we note that these shifts in the IR features of the bulk ILs upon the incorporation of ILs into ZIF-8 are consistent with a pre-vious report, where [BMIM][CF3SO3] was incorporated into MIL-53(Al) [38]
Likewise, to demonstrate the influence of IL’s anion size on the IL‒ MOF interactions, we examined the IR spectra of bulk [BMIM][MeSO4], [BMIM][OcSO4], and the corresponding IL/ZIF-8 composites in detail The peaks at 1009 and 1218 cm 1 correspond to νas (—SO3) and νs
(—SO3) stretching modes of bulk ILs, respectively The peak at 1009
cm 1 red-shifted to 1003 and 1006 cm 1 in the IR spectra of [BMIM] [MeSO4]/ZIF-8 and [BMIM][OcSO4]/ZIF-8 composites, respectively However, the peak at 1218 cm 1 did not exhibit any shifts Here, we infer that the change in the anion size of imidazolium ILs have no sig-nificant impact on IL‒MOF interactions Furthermore, to investigate any evidence of the IL‒MOF interactions in the higher IR region, we considered ν(C2H) band related to the ring structure of IL’s cation The corresponding bands were located at 3109, 3117, 3105, and 3107 cm 1
in the bulk IR spectra of [BMIM][MeSO3], [BMIM][CF3SO3], [BMIM] [MeSO4], and [BMIM][OcSO4], respectively [46] These bands exhibi-ted blue-shifts of 7, 12, 8, and 10 cm 1 in the corresponding IR spec-trums of IL/ZIF-8 composites, respectively As the interionic interaction energies between cation and anion of the bulk IL is probed by the ν(C2H) band, a major blue-shift in the band position of this feature implies the weak interactions between the cation and anion of the IL when it is confined in the MOF cage [47,48] As most of these shifts were observed
in the IR spectra of IL’s anion, we infer that incorporation of the IL into MOFs’ pores leads to the direct interactions between IL’s anion and MOF surface These shifts in the IR features of bulk IL indicate the possibility
of electron sharing between the IL and MOF, defined the nature of IL‒
Table 1
Zn and S amount in IL/ZIF-8 composites determined from XRF analysis
Concentration [wt%] S Concentration [wt%] Corresponding IL Loading [wt%]
Fig 1 N2 isotherms of pristine ZIF-8 and IL/ZIF-8 composites at 196 �C
Table 2
BET surface area and pore volume of pristine ZIF-8 and IL/ZIF-8 composites
Sample SBET [m 2 g 1 ] Vpore [cm3 g 1 ]
Trang 5MOF interactions in the composites To examine the influence of these
interactions on the gas adsorption and separation performance of the
materials, single-component gas adsorption isotherms for CO2, CH4, and
N2 were measured in a pressure range of 0.1–1 bar for pristine ZIF-8 and
IL/ZIF-8 composites at 25 �C The gas adsorption isotherms for pristine
ZIF-8 and IL/ZIF-8 composites are presented in Fig 6
As demonstrated in Fig 6, the gas uptake capacity of each IL/MOF
composite reduced compared to that of pristine ZIF-8 This decrease in
the uptake capacity of the IL-incorporated ZIF-8 composites can be
attributed to the reduced available surface area and pore volume by the
presence of the IL molecules in the cages of the ZIF-8 Furthermore, the
data showed that the IL/MOF composites with ILs having a fluorinated
([BMIM][CF3SO3]) and non-fluorinated ([BMIM][MeSO3]) anions have similar CO2 uptakes; however, significant differences were observed in their CH4 and N2 uptakes CH4 uptake significantly decreased in the case
of the composite having the IL with the fluorinated anion, whereas the composite having the IL with the non-fluorinated anion was measured to have a lower N2 uptake Such differences in gas uptakes could be attributed to the higher affinity of the fluorinated anion towards CO2
and N2 molecules, which have quadrupole moments, while having a comparatively weak attraction towards non-polar CH4 Furthermore, it
is well-known that CO2 has a great affinity towards fluorine moieties, thus the presence of C–F bond in a highly fluorinated anion may act as Lewis base to interact with acidic carbon atom of CO2 Such interactions improves the CO2-philicity by providing preferential adsorption sites for
CO2 molecules compared to CH4 [49,50] We also calculated CO2, CH4, and N2 solubilities of bulk ILs using COSMO-RS, which is widely used to estimate the solubilities of various hydrocarbons and gases in ILs [51–53] The stronger interactions between CO2 and N2 with the fluo-rinated anion are in agreement with the gas solubilities estimated by COSMO-RS calculations, where bulk [BMIM][CF3SO3] has higher CO2
and N2 solubilities compared to the bulk IL having a non-fluorinated anion ([BMIM][MeSO3]) as shown in Fig S1 Thus, the lower CH4 up-take of [BMIM][CF3SO3]/ZIF-8 composite can be attributed to the weak interactions between the fluorinated anion and CH4, which is also consistent with the poor solubility of CH4 in the bulk ([BMIM][CF3SO3])
as predicted by COSMO-RS calculations
In addition to the IL‒MOF surface interactions with the adsorbate molecules, the size of the IL’s anion significantly influences the adsorption capacity of the corresponding IL/ZIF-8 composite The data showed that incorporation of the IL having a bulky anion ([OcSO4] ) into ZIF-8 led to the lowest uptakes for each gas compared to the gas uptakes of [BMIM][MeSO4]/ZIF-8 composite Here, we note that the gas solubilities in bulk ILs generally increase with the increase in alkyl chain length or electronic environment of cation/anion [54,55] However, when the anion size of the IL increased, we observed a different trend in the gas uptakes of the corresponding IL/ZIF-8 composite This opposite trend between the bulk ILs’ gas solubilities and their gas uptakes in the corresponding IL/ZIF-8 composite can be attributed to the change in the affinity of the corresponding IL towards the adsorbate molecules [38] Furthermore, we note that when a bulky IL is incorporated into ZIF-8, less pore volume is available for the adsorbate molecule in the corre-sponding composite compared to the composite having an IL with a small anion Thus, [BMIM][OcSO4]/ZIF-8 composite having an IL with a
Fig 2 SEM images of (a) ZIF-8, (b) [BMIM][MeSO3]/ZIF-8, (c) [BMIM]
[CF3SO3]/ZIF-8, (d) [BMIM][MeSO4]/ZIF-8, and (e) [BMIM][OcSO4]/ZIF-8 at
magnifications of 100 k� and 25 k�
Fig 3 XRD patterns of pristine ZIF-8 and IL/ZIF-8 composites
Trang 6bulky anion showed the lowest uptake for each gas compared to [BMIM]
[MeSO4]/ZIF-8 composite
Finally, to assess the influence of the differences in the gas uptakes on
the gas separation performance of the IL/MOF composites, we fitted the
individual single-component gas adsorption isotherms of pristine ZIF-8
and IL/ZIF-8 composites to dual-site Langmuir, Langmuir-Freundlich,
and dual-site Langmuir-Freundlich models and calculated their ideal
and mixture selectivities Ideal CO2/CH4, CO2/N2, and CH4/N2
selec-tivities were calculated from the fitted adsorption isotherms, whereas
mixture selectivities were calculated using the Ideal Adsorption Solution
Theory (IAST) for pristine ZIF-8 and IL/ZIF-8 composites [38] IAST is
an effective method to predict the gas mixture adsorption data by using
experimentally measured single-component gas adsorption isotherms
Fig 7 shows the ideal and mixture (CO2/CH4:50/50, CO2/N2:15/85, and
CH4/N2:50/50) selectivities of IL/ZIF-8 composites normalized by their
corresponding values on the pristine ZIF-8 at the same condition Thus,
having a normalized value higher than unity for any selectivity value
indicates an improvement in the corresponding selectivity of ZIF-8 upon
the incorporation of IL
According to Fig 7(a–c), [BMIM][CF3SO3]/ZIF-8, having an IL with
the fluorinated anion, exhibited 3-times higher ideal selectivity than
that of [BMIM][MeSO3]/ZIF-8 at a low pressure range for CO2/CH4
separation In addition, IL/ZIF-8 composites with both fluorinated and
non-fluorinated anions showed 2.5- and 2-times higher ideal CO2/CH4
selectivities, respectively, than those of the pristine ZIF-8 at 1 bar
However, an opposite trend was observed for the ideal CO2/N2 and CH4/
N2 selectivities of the composites At low pressures, [BMIM][MeSO3]/
ZIF-8 exhibited 3-times higher ideal CO2/N2 selectivity compared to
[BMIM][CF3SO3]/ZIF-8 Furthermore, as the pressure increases, ideal
CO2/N2 selectivity of [BMIM][MeSO3]/ZIF-8 decreased; however, the
selectivity remains 2-times higher compared to that of [BMIM]
[CF3SO3]/ZIF-8 at 1 bar Similarly, [BMIM][MeSO3]/ZIF-8 having a
non-fluorinated anion exhibited 4.3-times higher ideal CH4/N2
selec-tivity than that of [BMIM][CF3SO3]/ZIF-8 at 0.01 bar Here, we
conclude that high electronegativity of the fluorinated anion promotes
the IL interactions with the surface electrons of ZIF-8 cage as discussed
in the IR analysis, where a major blue-shift was observed for νas(—CF3)
band in the IR spectra of [BMIM][CF3SO3]/ZIF-8 composite Presence of such strong interactions between the IL and MOF cage favors the pref-erential adsorption of CO2 compared to CH4, leading to an improvement
in the CO2/CH4 separation performance On the other hand, because of the very poor solubility of N2 compared to CO2 and CH4 in a non- fluorinated bulk IL ([BMIM][MeSO3]), the corresponding IL/ZIF-8 composite showed significantly improved CO2/N2 and CH4/N2 separa-tion performances compared to those of the [BMIM][CF3SO3]/ZIF-8 composite Furthermore, it is noteworthy here that when ZIF-8 was incorporated with an IL having either a fluorinated or bulky anion, the corresponding IL/ZIF-8 composite becomes N2 selective over CH4, which is the opposite of the separation performance of the pristine ZIF-8 Thus, we infer that by changing the electronic structure or size of the IL anion, ZIF-8 separation characteristics can be switched from being CH4
selective to N2 selective in the IL/ZIF-8 composite This observation further demonstrates the broad potential of incorporating ILs into MOFs
in tuning the adsorption and separation characteristics of MOF Next, we compared the selectivities of [BMIM][MeSO4]/ZIF-8 with those of [BMIM][OcSO4]/ZIF-8 to investigate the impact of the changes
in the anion size of IL on the corresponding CO2 separation performance
of the composites At low pressures, [BMIM][OcSO4]/ZIF-8 composite presented 2-times higher CO2/CH4 separation performance than [BMIM][MeSO4]/ZIF-8 composite Moreover, both [BMIM][MeSO4]/ ZIF-8 and [BMIM][OcSO4]/ZIF-8 composites showed approximately 1.5-times higher CO2/CH4 separation performance compared to pristine ZIF-8 at 1 bar Furthermore, [BMIM][MeSO4]/ZIF-8 showed 1.5- and 3.3-times higher ideal CO2/N2 selectivity than that of the [BMIM] [OcSO4]/ZIF-8 composite at 0.01 and 1 bar Likewise, [BMIM][MeSO4]/ ZIF-8 composite having a small anion showed approximately 1.8-times higher CH4/N2 separation performance compared to [BMIM][OcSO4]/ ZIF-8 in the whole pressure range (0.01–1 bar) These results demon-strate that incorporation of IL with a small anion ([MeSO4] ) into ZIF-8 significantly improved CO2/N2 and CH4/N2 separation performance of the composite
Next, we considered another structural change in the anion by comparing the ILs having sulfite and sulfate groups in their anions to demonstrate their impact on CO2 separation of the composites
[BMIM][OcSO4]/ZIF-8
Trang 7Fig 5 IR spectra of pristine ZIF-8, bulk IL, and IL-incorporated ZIF-8 composite: (a) [BMIM][MeSO3]/ZIF-8, (b) [BMIM][CF3SO3]/ZIF-8, (c) [BMIM][MeSO4]/ZIF-8, and (d) [BMIM][OcSO4]/ZIF-8
Trang 8Accordingly, IL/ZIF-8 composites ([BMIM][MeSO3]/ZIF-8 and [BMIM]
[MeSO4]/ZIF-8) having ILs with sulfite and sulfate groups in their
an-ions, respectively, have almost similar CO2/CH4 and CH4/N2 separation
performances; however, the IL/ZIF-8 composite with an IL having a
sulfate group in its anion presented 2-times higher ideal CO2/N2
selec-tivity than IL/ZIF-8 composite with sulfite anion at 1 bar
Furthermore, for a qualitative comparison between the separation
performances of the bulk ILs and their corresponding IL/ZIF-8
com-posites, CO2/CH4, CO2/N2, and CH4/N2 selectivities of bulk ILs based on
the ratios of the corresponding gas solubilities determined from the
COSMO-RS calculations were estimated as presented in Fig S2
Accordingly, fluorination of the anion led to an improvement in CO2/
CH4 separation performance, whereas the corresponding CO2/N2 and
CH4/N2 selectivities were lower than those of composites having the IL
with a non-fluorinated anion On the other hand, an increase in the
anion size of the IL leads to a decrease in both ideal CO2/CH4 and CO2/
N2 selectivities; however, the ideal CH4/N2 selectivity was improved
Similarly, the bulk IL with a sulfate anion has higher CO2/CH4 and CO2/
N2 selectivities compared to those of the IL with a sulfite anion These
trends in CO2/CH4, CO2/N2, and CH4/N2 separation performances of
bulk ILs are consistent with their corresponding IL/ZIF-8 composites
Gases exist as mixtures in real processes, therefore, we performed
IAST calculations to predict the corresponding mixture selectivities for
CO2/CH4:50/50, CO2/N2:15/85, and CH4/N2:50/50 separations as
presented in Fig 7(d–f) Accordingly, the highest improvement in CO2/
CH4:50/50 separation was observed for [BMIM][MeSO4]/ZIF-8
fol-lowed by [BMIM][CF3SO3]/ZIF-8, [BMIM][MeSO3]/ZIF-8, and [BMIM]
[OcSO4]/ZIF-8 At 1 bar, the corresponding mixture selectivities were 3.7-, 2.7-, 2-, and 1.3-times higher than the CO2/CH4 separation per-formance of pristine ZIF-8, respectively Likewise, CO2/N2 mixture se-lectivities of [BMIM][MeSO4]/ZIF-8 and [BMIM][MeSO3]/ZIF-8 were calculated to be 2.3- and 1.7-times higher than those of pristine ZIF-8 at
1 bar, respectively Furthermore, at low pressure (0.01 bar), [BMIM] [MeSO3]/ZIF-8 exhibited 5.5-times higher CH4/N2:50/50 separation performance compared to its fluorinated counterpart composite ([BMIM][CF3SO3]/ZIF-8) Similarly, a 1.4-times improvement in CH4/
N2:50/50 separation performance was observed in the whole pressure range of 0.01–1 bar for IL/ZIF-8 composite with small anion ([MeSO4] ) compared to [BMIM][OcSO4]/ZIF-8 As a result, we can conclude that IL/MOF composites with a fluorinated anion offer significantly improved the CO2/CH4 mixture separation performances especially at low pressures Whereas the IL/MOF composite having an IL with a non- fluorinated anion led to improvements in CO2/N2:15/85 and CH4/
N2:50/50 separation performances On the other hand, IL/ZIF-8 com-posite with a small anion ([MeSO4] ) demonstrated improvements in
CO2/CH4, CO2/N2, and CH4/N2 mixture selectivities compared to pris-tine ZIF-8 In contrast, an increase in the anion size improved the CO2/
CH4 mixture selectivity of IL/ZIF-8 composite at low pressures; however,
CO2/N2 and CH4/N2 selectivities of the composite were adversely affected compared to pristine ZIF-8 Furthermore, IL/ZIF-8 composite having a sulfate anion ([BMIM][MeSO4]/ZIF-8) showed approximately 1.5-times higher CO2/CH4:50/50 separation performance in the whole pressure range compared to IL/ZIF-8 composite having an IL with a sulfite anion ([BMIM][MeSO3]/ZIF-8) Likewise, CO2/N2:15/85 and
Fig 6 Excess adsorption isotherms of (a) CO2, (b) CH4, and (c) N2 in pristine ZIF-8 and IL/ZIF-8 composites at 25 �C
Trang 9CH4/N2:50/50 separation performance of IL/ZIF-8 composite having a
sulfate anion was only 0.6- and 0.4-times of those of the IL/ZIF-8
com-posite with a sulfite anion at 1 bar These results suggest that changes in
both electronic environment and the size of the IL’s anion have a
sig-nificant impact on both the ideal and mixture selectivities of IL/ZIF-8
composites
4 Conclusions
In this study, four different imidazolium ILs having the same cation, [BMIM]þ, and different anions ([MeSO3] ; [CF3SO3] ; [MeSO4] ; and [OcSO4] ) were incorporated into ZIF-8 to demonstrate the impact of changes in the electronic structure and the size of the anion on the gas adsorption and separation performance of the corresponding IL/ZIF-8 composites The resultant IL/MOF composites were characterized in detail by using various techniques The characterization results of the
Fig 7 Normalized ideal and mixture selectivities of IL/ZIF-8 composites at 25 �C
Trang 10composites illustrated the successful incorporation of ILs in ZIF-8, while
the crystal structure and morphology of the ZIF-8 were well-maintained
TGA and IR data confirmed the presence of IL‒MOF interactions that
were accompanied by the changes in the decomposition temperature
and shifts in IR features of bulk ILs in composite samples Finally, to
investigate the gas adsorption and separation performance, CO2, CH4,
and N2 adsorption isotherms were measured for pristine ZIF-8 and IL/
ZIF-8 composites and their corresponding ideal and mixture
selectiv-ities were determined Accordingly, [BMIM][CF3SO3]/ZIF-8 and
[BMIM][MeSO4]/ZIF-8 composites exhibited 2.5- and 3.3-times higher
ideal CO2/CH4 and CO2/N2 selectivities compared to pristine ZIF-8 at 1
bar [BMIM][MeSO3]/ZIF-8 composite showed a 4.3-times higher ideal
CH4/N2 selectivity than that of the pristine ZIF-8 at 0.01 bar, which was
the highest level of improvement among all the IL/ZIF-8 composites
examined in this work Similarly, CO2/CH4:50/50 mixture selectivities
of [BMIM][CF3SO3]/ZIF-8 and [BMIM][MeSO4]/ZIF-8 were improved
3.7- and 2.7-times compared to those of pristine ZIF-8 at 1 bar,
respec-tively CO2/N2:15/85 and CH4/N2:50/50 mixture selectivities of
[BMIM][MeSO4]/ZIF-8 improved 2.3- and 1.8-times compared to those
of pristine ZIF-8 at 1 bar, respectively In summary, we demonstrate that
the ILs with a fluorinated anion significantly improved the CO2/CH4
separation performance, owing to a stronger affinity of C–F bond
to-wards CO2 compared to that of C–H The poor solubility of N2 in the IL
with a non-fluorinated anion led to improved CO2/N2 and CH4/N2
separation performance of the corresponding IL/ZIF-8 composite On
the other hand, change in IL anion size did not have a significant impact
on CO2/CH4 separation performance, however, remarkable
improve-ments in the CO2/N2 and CH4/N2 separation performance were
observed for the composites having ILs of small anion These results
demonstrated that the change in electronic environment and anion size
of ILs alter the IL-MOF interactions, which have significant impacts on
the gas separation performances of the IL/MOF composites
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper
CRediT authorship contribution statement
Muhammad Zeeshan: Methodology, Investigation, Writing -
orig-inal draft, Writing - review & editing Harun Kulak: Investigation,
Validation Safiyye Kavak: Investigation, Validation H Mert Polat:
Investigation, Validation Ozce Durak: Investigation, Validation Seda
Keskin: Conceptualization, Supervision, Methodology, Writing -
orig-inal draft, Writing - review & editing Alper Uzun: Conceptualization,
Supervision, Methodology, Writing - original draft, Writing - review &
editing
Acknowledgments
This work is supported by the Scientific and Technological Research
Council of Turkey (TUBITAK) under 1001-Scientific and Technological
Research Projects Funding Program (Project Number 114R093) and by
Koç University Seed Fund Program S.K acknowledges ERC-2017-
Starting Grant This study received funding from the European Research
Council (ERC) under the European Union’s Horizon 2020 research and
innovation programme (ERC-2017-Starting Grant, grant agreement no
756489-COSMOS) M.Z acknowledges HEC-Pakistan Scholarship The
authors thank Koç University Surface Science and Technology Center
(KUYTAM) for providing help with the sample characterization The
authors thank TARLA for the collaborative research support Support
provided by the Koç University TÜPRAŞ Energy Center (KUTEM) is
gratefully acknowledged
Appendix A Supplementary data
Supplementary data to this article can be found online at https://doi org/10.1016/j.micromeso.2020.110446
References
[1] D.Y.C Leung, G Caramanna, M.M Maroto-Valer, An overview of current status of carbon dioxide capture and storage technologies, Renew Sustain Energy Rev 39 (2014) 426–443, https://doi.org/10.1016/J.RSER.2014.07.093
[2] N Bauer, I Mouratiadou, G Luderer, L Baumstark, R.J Brecha, O Edenhofer,
E Kriegler, Global fossil energy markets and climate change mitigation – an analysis with REMIND, Climatic Change 136 (2016) 69–82, https://doi.org/
[3] C Song, Global challenges and strategies for control, conversion and utilization of
CO 2 for sustainable development involving energy, catalysis, adsorption and chemical processing, Catal Today 115 (2006) 2–32, https://doi.org/10.1016/J
combustion carbon dioxide capture technologies using activated carbon,
materials for the adsorption, separation, and catalytic conversion of carbon
[8] K Sumida, D.L Rogow, J.A Mason, T.M McDonald, E.D Bloch, Z.R Herm, T.-
H Bae, J.R Long, Carbon dioxide capture in metal–organic frameworks, Chem Rev 112 (2012) 724–781, https://doi.org/10.1021/cr2003272
[9] J.R Li, R.J Kuppler, H.C Zhou, Selective gas adsorption and separation in metal- organic frameworks, Chem Soc Rev 38 (2009) 1477–1504, https://doi.org/
[10] A Uzun, S Keskin, Site characteristics in metal organic frameworks for gas adsorption, Prog Surf Sci 89 (2014) 56–79, https://doi.org/10.1016/j
[11] H Furukawa, K.E Cordova, M O’Keeffe, O.M Yaghi, The chemistry and applications of metal-organic frameworks, Science 80 (2013) 341, https://doi.org/
[12] A.J Howarth, Y Liu, P Li, Z Li, T.C Wang, J.T Hupp, O.K Farha, Chemical, thermal and mechanical stabilities of metal-organic frameworks, Nat Rev Mater 1 (2016) 1–15, https://doi.org/10.1038/natrevmats.2015.18
[13] A.J Howarth, Y Liu, P Li, Z Li, T.C Wang, J.T Hupp, O.K Farha, Chemical, thermal and mechanical stabilities of metal–organic frameworks : nature Reviews Materials, Nat Rev Mater 1 (2016) 15018 http://www.nature.com/articl
[14] M Vahidi, A.M Rashidi, A Tavasoli, Preparation of piperazine-grafted amine- functionalized UiO-66 metal organic framework and its application for CO 2 over
CH 4 separation, J Iran Chem Soc 14 (2017) 2247–2253, https://doi.org/
[15] Z Qiao, N Wang, J Jiang, J Zhou, Design of amine-functionalized metal-organic frameworks for CO 2 separation: the more amine, the better? Chem Commun 52 (2016) 974–977, https://doi.org/10.1039/c5cc07171b
[16] D.H Hong, M.P Suh, Enhancing CO 2 separation ability of a metal-organic framework by post-synthetic ligand exchange with flexible aliphatic carboxylates, Chem Eur J 20 (2014) 426–434, https://doi.org/10.1002/chem.201303801 [17] L Du, Z Lu, K Zheng, J Wang, X Zheng, Y Pan, X You, J Bai, Fine-tuning pore size by shifting coordination sites of ligands and surface polarization of metal- organic frameworks to sharply enhance the selectivity for CO 2 , J Am Chem Soc
135 (2013) 562–565, https://doi.org/10.1021/ja309992a [18] I Cota, F Fernandez Martinez, Recent advances in the synthesis and applications
of metal organic frameworks doped with ionic liquids for CO 2 adsorption, Coord Chem Rev 351 (2017) 189–204, https://doi.org/10.1016/j.ccr.2017.04.008 [19] F.P Kinik, A Uzun, S Keskin, Ionic liquid/metal–organic framework composites: from synthesis to applications, ChemSusChem 10 (2017) 2842–2863, https://doi
[20] K Fujie, H Kitagawa, Ionic liquid transported into metal-organic frameworks, Coord Chem Rev 307 (2016) 382–390, https://doi.org/10.1016/j
[21] R.D Rogers, K.R Seddon, Ionic liquids - solvents of the future? Science 80 (2003) 792–793, https://doi.org/10.1126/science.1090313 , 302
[23] S Keskin, D Kayrak-Talay, U Akman, €O Hortaçsu, A review of ionic liquids towards supercritical fluid applications, J Supercrit Fluids 43 (2007) 150–180,
[24] A Lewandowski, A �Swiderska-Mocek, Ionic liquids as electrolytes for Li-ion batteries-An overview of electrochemical studies, J Power Sources 194 (2009) 601–609, https://doi.org/10.1016/j.jpowsour.2009.06.089
[26] N.A Khan, Z Hasan, S.H Jhung, Ionic liquids supported on metal-organic frameworks: remarkable adsorbents for adsorptive desulfurization, Chem Eur J 20 (2014) 376–380, https://doi.org/10.1002/chem.201304291