Study on water and gas permeation characteristics with ZIF-8 mixed matrix membranes

The membrane separation process lacks intrinsic permeation characteristics to compete with other separation technologies like adsorption, sedimentation, coagulation, skimming, and distillation. A mixed matrix membrane (MMM) is one of the strategies to improve the separation characteristics with embedded nanofillers particles. Zeolite imidazolate framework (ZIF) has a new subclass of inorganic–organic hybrid materials that are being introduced as new fillers for incorporation into the polymer matrix for various applications such as oily wastewater separation, wastewater treatment, natural gas dehydration, landfill gas upgrading, and mixed gas separation. In this experimental work, a metal-organic framework called ZIF-8 was synthesized and used as a filler for modification of MMMs and characterized with FTIR and SEM. ZIF-8 nanoparticles up to 5 wt% loading were added to PSF casting solution then the permeation characteristics of MMMs showed an improved result like the pure water flux of the modified membrane at 2.5 bar was increased up to 456.38 L/m²h. In the case of pure gas separation, at 5 wt% ZIF-8 loading in PSF, the pure gas CO₂ permeability at 9 bar pressure had increased to 10.54 barrer.

HIGHLIGHTS

Listen

We have studied water and gas permeation characteristics incorporated with ZIF-8 containing mixed matrix membranes.
ZIF-8 was made as a gateway for quick transport of CO₂ gas molecules and water molecules through the polymer matrix.
As per the observed results, higher permeability of the MMMs can be possible with higher loading of ZIF-8.

gas permeation, mixed matrix membranes, selectivity, water flux, ZIF-8

ABBREVIATIONS

Listen

PSF: Polysulfone
ZIF-8: Zeolitic Imidazolate Frameworks-8
NMP: N-Methyl 2-pyrrolidone
NPs: Nanoparticles
PEG: Polyethylene glycol
PWS: Pure water flux
MMMs: Mixed matrix membranes
MFC: Mass flow controller
15M0: Plain PSF membrane
15M1: 1 wt% ZIF-8 mixed matrix membrane
15M5: 5 wt% ZIF-8 mixed matrix membrane
J_pw: Pure water flux
A: Surface area (cm²)
Δt: Time (min)
N_A: Normal flux
P₁: Feed pressure
P₂: Back pressure
L: Thickness
P_A: Permeability of gas A
P_B: Permeability of gas B
: Selectivity of gas
TMP: Transmembrane pressure
FTIR: Fourier transform infrared radiation
TGA: Thermal gravimetric analysis
SEM: Scanning electronic microscope
WCA: Water contact analysis
XRD: X-ray diffraction

INTRODUCTION

Listen

No living person can live without pure water. Freshwater availability on earth is meagre, and demand increases as the world's population grows. Water is also contaminated by industrial and agricultural activities, medications, technological civilization, pesticides, clothing, and worldwide changes (Gnanasekaran & Balaguru 2019). Furthermore, greenhouse and toxic gases produced by the dumping and burning of fossil fuels are increasing environmental damage and global warming. This directly affects an ecosystem's biological cycle, causing skin and respiratory diseases from air pollution. Gas separation is essential, particularly in chemical and petrochemical plants that purify landfill gas, upgrade biogas, and sweeten natural gas. Natural gas has a significant amount of energy that can be utilized for cooking, heating, power, fuel in cars, and chemical feedstock. Crude natural gas includes corrosive gases such as CO₂ and H₂S; it is corrosive, toxic, and flammable due to this problem, creating corrosion in gas transportation pipelines. Hence, it is necessary to remove the impurity from the natural gas. To address these issues, less use of processing water and fossil fuels is impossible for economic development. Thus, it is essential to get fresh water and capture CO₂ to reduce waste emissions to protect the environment. So, it requires cost-effective and environmentally friendly techniques for purifying contaminated water and air (Le et al. 2021).

Water and carbon dioxide purification has been done by equilibrium separation and rates governing process. Equilibrium processes such as skimming, sedimentation, filtration, adsorption, absorption, and cryogenic distillation required solvent or adsorbent for product separation (Vatanpour & Khorshidi 2020). So, it has a costly process for operation and maintenance. Other disadvantages are that an open loop requires considerable space, demands more energy, and is challenging to scale up. However, the rate governing processes like membrane-based separation does not require a solvent for water and gas separation (Lin et al. 2019; Salahshoori et al. 2021). So, there are low operation and maintenance costs, greater process simplicity, closed loop, greater ease of operation, less space, requirement easy scale-up, and greater energy efficiency compared to the equilibrium separation process (Kumar et al. 2018; Nabipour et al. 2020).

In the last few decades, polymeric membranes such as polysulfone, polyamide, cellulose acetate, polyethersulfone, and Pebax have been used because of easy scale-up and high-performance gas separation. Most polymeric membranes have a fundamental problem, such as a limitation between permeability and selectivity (Maghami et al. 2021). So, the polymeric membrane is incorporated with highly porous solid material as filler and fabricated mixed matrix membranes (MMMs) on the geometry base. Many researchers worked on the different types of fillers like zeolites (Susanti 2019), nano-silica (Salahshoori et al. 2021), graphene oxide (Sainath et al. 2021), SiO₂ (Yang et al. 2021), and CNTs (Singh et al. 2021). However, due to the surpassing trend of the limit between permeability and selectivity, the previous work has reported metal-organic frameworks (MOFs) having excellent results for water and gas separation performance. For example, incorporating 3 wt% MOF in polyethersulfone MMMs improved the water flux by 121.5 L/m² h (Zhang et al. 2020). In another work, authors reported that adding 20 wt% ZIF-68 in a Matrimid MMM enhanced the permeability of CO₂ by 122% (Essen et al. 2021). Furthermore, 5 wt% ZIF-67/Pebax-1657 membrane has a higher permeability of CO₂ compared to the plain Pebax membrane (Meshkat et al. 2019). Whereas on the addition of 20 wt% ZIF-301 with polyimide, the permeability of CO₂ was found to be around 899 barrer and the selectivity CO₂/CH₄ of 29.3 (Wang et al. 2021).

We focused on ZIF-8 as a subclass MOF. MOFs are crystalline organic–inorganic hybrid complexes of bivalent or trivalent metal clusters or ions linked by organic linkers (Nuhnen et al. 2018). The presence of organic linkers is suitable for various strategies to fine pore structure, aperture, and pore polarity. Zeolitic imidazolate frameworks (ZIFs) have high crystalline nature, high surface area, porous structure, high thermal stability and chemical stability (Essen et al. 2021), and high adsorption capacity (Furukawa et al. 2010). On the basis of microporosity, ZIFs were suitable for CO₂ gas separation and water purification.

Nevertheless, despite the wide variety in the unravelled structure of ZIFs, they are not used in MMMs to apply the water and gas separation process (Chen et al. 2014; Qian et al. 2020). ZIF-8 is synthesized by 2-methyl imidazolate, linked with Zn²⁺ metal ions, forming a cub octahedral structure. The presence of 2-methyl imidazolate made the ZIF-8 framework structure dynamic, where the crystallographic size of the pore aperture 3.4A⁰ (Banerjee et al. 2009). In the case of gas separation, the molecular size of CO₂ has 3.3A⁰ and CH₄ of 3.8A⁰ (Furukawa et al. 2014), which is close to ZIF-8 pore size of 3.4A⁰ and becomes the flexible structure to diffuse the polar CO₂ gas, which was not attributed non-selective flow channels for CH₄ gas (Deng et al. 2020). In the case of water, the molecular size is 2.8A⁰, smaller than the ZIF-8 pore size, making it very easy for water molecules to pass through ZIF-8. So, they have emerged as a promising material for water purification and CO₂ gas separation processes (Aframehr et al. 2020). The polymer used as PSF has an inherent material, and high mechanical and chemical stability (Wu et al. 2019).

In the current work, ZIF-8 filler was synthesized and it was utilized for the modification of porous and non-porous mixed matrix PSF membranes. The weight percentage of ZIF-8 varied from 0.5 to 5%. The plain and modified membranes were characterized with the help of FTIR, XRD, water contact angle, and SEM. Finally, the fabricated MMMs were investigated for water and gas permeation studies.

EXPERIMENTAL WORK

Listen

Chemicals

Listen

Zinc nitrate hexahydrate, polyethylene glycol (4000), 2-methylimidazole (C₄H₆N₂ > 99%), and polysulfone (average Mw 35,000) were purchased from Sigma-Aldrich. Methanol (HPLC grade) and N-methyl 2-pyrrolidone (98%) were purchased from Loba Chemical.

Synthesis of ZIF-8

Listen

ZIF-8 was synthesized, as shown in Figure 1 (Anastasiou et al. 2018; Yang et al. 2021). 0.94 g of zinc nitrate hexahydrate was mixed with 20 ml of methanol and 20 ml of DI water. Furthermore, 2 g of 2-methylimidazole was mixed with 20 ml of methanol. These two reaction mixture solutions were mixed in a single beaker and stirred for 3 h at room temperature. Then finally, a white milky solution was formed. This solution was centrifuged for 20 min at 6,000 rpm to separate ZIF-8 nanoparticles and washed with methanol for three cycles to remove excess reactants. The wetted ZIF-8 was put in an oven at 110 °C for 2 h to dry and remove moisture; after that, a dry ZIF-8 product was obtained.

Figure 1

Synthesis process of ZIF-8.

Fabrication of membranes

Listen

Phase inversion method (for water application)

Listen

Figure 2 shows the fabrication of the membrane process. Take 0.5 g of PEG4000 and 1.5 g of PSF crystal mixed in 8.0 g of N-methyl 2-pyrrolidone; the solution was agitated for 12 h to obtain the homogeneous solution. The solution was put for degasification for 6 h without stirring. The polymer solution was then cast onto the glass plate before maintaining the air gap thickness of 255 μm. The thin wet film was obtained on the glass plate and put in a water bath for 24 h to remove the solvent. The film was dried at room temperature overnight (Saini et al. 2019).

Figure 2

Fabrication of mixed matrix membrane by the phase inversion method.

For MMMs, the desired quantity of a predetermined ZIF-8 was mixed with PSF and PEG4000 homogeneous solution, which was then put in a sonication water bath for an hour. The polymer solution was then cast onto the glass plate before maintaining the air gap thickness of 255 μm. The thin wet film was obtained on the glass plate and put in a water bath for 24 h to remove the solvent. The film was dried at room temperature overnight. The membranes obtained were 15M0, 15M0.5, 15M1, 15M2, 15M3, and 15M5.

Solution casting method (for gas application)

Listen

To fabricate a pure PSF membrane, as shown in Figure 3, 2 g of polysulfone crystal was dissolved in 8 g of NMP solvent at 55 °C. The solution was then agitated for 12 h to achieve a homogeneous solution. The solution was then cast onto a glass plate before maintaining the air gap between the digital applicator 255 μm. The thin film was obtained on the glass plate and allowed to dry for 24 h at a temperature of 180 °C in an oven (Deng et al. 2021).

Figure 3

Fabrication of mixed matrix membrane by the solution casting method.

For MMMs, the desired quantity of a predetermined ZIF-8 was mixed with PSF homogeneous solution, which was then put in a sonication water bath for an hour. The resultant solutions were cast onto a glass plate before maintaining the air gap between the digital applicator 255 μm. The thin film was obtained on the glass plate and allowed to dry for 24 h at a temperature of 180 °C in an oven. The membranes obtained were PSF, 1, 3, and 5 wt% ZIF-8/PSF MMMs.

Membrane characterization

Listen

The ATR-FTIR characterization was carried out with the help of an FTIR spectrometer (Perkin Elmer spectrum) over a wave number of 4000–400 cm⁻¹ to study the different functional groups of ZIF-8 modified MMM. Thermogravimetric analyses characterized the thermal decomposition temperature of MMMs by Hitachi STA7200. The crystalline structure MOFs and mix matrix membranes were characterized with the help of an X-ray diffractometer (XRD) Model D8 DISCOVER (Bruker). The morphology of all MMMs was observed with the help of a scanning electron microscope (FE-SEM) Model JSM 7600F (Jeol). The DFT-based investigations were studied using the Gaussian 09 computational package and Goniometer (APEX S/N: ACAMNSC 34, model Acam-D2) was analysed for the measurement of static water contact angles of unmodified and modified water membranes.

THEORY

Listen

Evaluation of water flux through membrane performance

Listen

The pure water flux was evaluated by measuring permeate through the membrane. Hence, this experiment was carried out in the dead-end batch filtration apparatus seen in Figure 4. A nitrogen cylinder and pressure gauge are included in the setup. Porous base support was used to place a flat circular membrane with a diameter of 5 × 10⁻² m and an effective area of 19.625 × 10⁻⁴ m² inside the footprint of the O-ring. Pure water was collected from the other side of the membrane after nitrogen gas pressure was utilized as a driving force to the permeate of the water during the filtration process. The pure water flow (J_pw) was measured by measuring the volume (V) of permeate water that passed through the membrane per unit surface area (A) at intervals of 5 min (Δt) for 60 min. The pure water flux was calculated by the following equation (Saini & Sinha 2019):

(1)

Figure 4

Membrane experimental setup for water.

Pure CO₂ gas permeability studies

Listen

In gas permeation, the feed gas acts on the membrane, and due to pressure differences on both sides of the membrane, a selective permeate was obtained due to the solution diffusion mechanism. It is the slowest process and rate-determining step in the solution diffusion model. In general, the relationship between the linear flux J and the driving force is the following equation:

(2)

where D is the diffusion coefficient.

The concentration (C) percentage of permeate is proportional to the vapour pressure (p) of the permeate, which means the solubility (s) of permeate is sufficiently low in the polymer. Henry's law expressed this relationship for calculating the gas solubility of the membrane as the following equation:

(3)

where s and p are the solubility coefficient and partial pressure of ‘A’ in gas, at a steady state, the gas permeates through the membrane of thickness L (cm), which is caused due to pressure difference between the feed pressure (P₁) and back pressure (P₂). So, the permeability coefficient (P_A) at normalized flux (N_A) can be calculated as follows:

(4)

The permeability coefficient (P_B) at normalized flux (N_B) can be calculated as follows:

(5)

For binary gas mixture, the selectivity of the mixed gases depends on the permeation (A and B gas components), which is known as ideal selectivity or perm selectivity. The selectivity (⁠

⁠) can be calculated as follows:

(6)

As shown in Figure 5, the thin film of the membrane was put in a membrane module with the support of a circular microporous ceramic disk 3.14 cm² and 20 μm pore size. The membrane is held with a silicon rubber O-ring in a module. For single gas performance, the feed line pressure of CO₂ or CH₄ gas was increased from 7 to 9 bar pressure with a mass flow rate (MFC) of 20 ml/min to supply the membrane module. The helium as a carrier gas is used to push permeate from the bottom side of the membrane module towards online gas-chromatography (Nucon-5700) and in which the peak area of CO₂ or CH₄ gas was analysed and recorded in the computer. In case of mixed gas (CO₂/CH₄) separation, the MFC of CO₂:CH₄ gas (30:70) ml/min at 9 bar pressure supplied to membrane module, then permeate carried towards the online gas-chromatography in which the peak area of gas analysed and those areas recorded in the computer. The permeability data are reported in barrer [1 barrer = 10⁻¹⁰ cm² (STP)/(cm s cm Hg)].

Figure 5

Membrane experimental setup for gas separation.

RESULTS AND DISCUSSION

Listen

Characterization of membranes

Listen

FTIR analysis

Listen

FTIR was analysed to identify the different functional groups of ZIF-8 on each peak of water (15M3, 15M5) and gas (1, 3, 5 wt%) MMMs as shown in Figure 6. In spectra of both water and gas MMMs, we observed the absorption bands at 3189.47 and 2926.43 cm⁻¹ represented the aromatic C–H stretching and aliphatic C–H of the imidazole, respectively (Uribe-romo et al. 2010) and showed with increasing wt% ZIF-8 in both water and gas MMMs then the absorption bond becomes stronger (Barooah & Mandal 2018). The peak 1421.34 cm⁻¹ corresponds to the C = N stretching vibrations and the peak 1310.97 cm⁻¹ corresponds to C–N stretching vibrations. The lower wavelength spectra around 753.22 and 838.67 cm⁻¹ indicate the presence of Zn–N and Zn–O bonds in water and gas MMMs, respectively (Karimi et al. 2019a, 2019b). From the observation, we confirmed the functional groups of ZIF-8 at each peak of MMMs.

Figure 6

FTIR spectra of water and gas mixed matrix membranes.

TGA analysis

Listen

Figure 7 shows the TGA analysis of ZIF-8 and MMMs. ZIF-8 shows two steps of weight loss at temperatures 200 and 510 °C due to the decomposition of imidazolate and carbon species, respectively. From the TGA graph, we observed that both water and gas MMMs 15M5, 1 wt% ZIF-8/PSF and 5 wt% ZIF-8/PSF show the weight loss starts in two steps at 200 °C due to decomposition imidazolate species of ZIF-8 and at 450 °C decomposition of carbon species of polymer (Aframehr et al. 2020; Sasi et al. 2020). The zinc (Zn) element was not decomposed at 1000 °C in all MMMs and ZIF-8 powder. PSF is exposed to decompose with nitrogen at 150 °C. This result showed that the presence of ZIF-8 improved the thermal stability of MMMs. On the base of increasing residual wt%, we identified the different wt% of MMMs because zinc elements do not decompose.

Figure 7

TGA profile of water and gas mixed matrix membranes.

SEM analysis

Listen

The cross-section SEM images of both water and gas membranes analysed the internal structure of membranes. As shown in Figure 8(a)–8(d) water membranes, the cross-section view of 15M0, 15M1, 15M3, and 15M5 MMMs shows asymmetric, dense top skin layer, and porous sublayer (Saini et al. 2019). The porous sublayer was observed to consist of a finger-like structure showing the ideal morphology of ultra-filtration membrane (UF) characteristics for archiving better flux (Furukawa et al. 2010; Azizah et al. 2021). The wt% ZIF-8 (1, 3, and 5) increases in MMMs, and then the finger-like structure made a sponge-like structure, which affects the permeability of water flux, supporting the increase of the water flux as shown in Figure 8(b)–8(d). Figure 9 shows the cross-section view of the gas separation membranes, and one can see that the cross-section view of plain PSF membrane is dense and smooth (Figure 9(a)). Incorporating ZIF-8 (1, 3 and 5 wt%) in MMMs observed good compatibility between ZIF-8 and PSF shows roughly unsymmetrical and dense (Figure 9(b)–9(d)). From the observation, we see that the filler materials internally tightly adhere together; no visible voids observed that means ZIF-8 filler uniformly dispersed in MMMs as shown in Figure 9(d) (Jiang et al. 2021) which indicated the excellent adhesion properties between PSF and ZIF-8 (Nuhnen et al. 2018). The thickness of all MMMs was around 41.87 μm. From the observation of SEM images, we confirmed that the morphology of water and gas membranes is in accordance with the previous studies and that's why it had been continued for the study of water and gas permeation characteristics.

Figure 8

Cross-sectional SEM images of water separation membranes.

Figure 9

Cross-sectional SEM images of gas separation membranes.

Water contact angle measurement and wettability study

Listen

The goniometer (APEX S/N: ACAMNSC 34, model Acam-D2) was used for the measurement of dynamic water contact angles of modified and unmodified water membranes, which shows the hydrophilicity of each membrane. We investigated the effects of incorporated ZIF-8 MMMs (M0, M0.5, M1, M2, M3, M4, and M5) as shown in Figure 10. The plain PSF had a water contact angle of 75.73°, which was reduced to 62.05° for the modified membrane. It demonstrated that the incorporation of ZIF-8 enhanced the hydrophilicity of the MMMs. A few structural defects can be found in the framework of ZIF-8 materials. They may be caused by inadequate coordination of zinc atoms by imidazole linkers, which results in open metal sites or zinc hydroxide defect sites (this process could also occur at crystal surfaces) (Karimi et al. 2019a, 2019b). The unconjugated lone pairs of neutral imidazole linkers attack these open sites to maintain charge balance, forming imidazolate linker terminated crystals. The terminal Imidazole linkers will introduce the –NH functional groups to the crystal surface (Choi et al. 2006). The different wt% of MMMs become more hydrophilic due to hydrogen bonding between N atoms (–NH) and water molecules, which causes a monolayer covering of water molecules to form on the crystal surface. The MMMs decreased contact angle can be attributed to increased surface roughness and rougher surface results due to the lower contact angle (Zhang et al. 2020).

Figure 10

WCA analysis for porous mixed matrix membranes.

Density functional theory (DFT)

Listen

Complimentary to the experimental investigations, density functional theory (DFT) calculations were carried out to obtain the interaction energies between ZIF-8 and CO₂ and CH₄ gases. All the DFT-based investigations were performed using the Gaussian 09 computational package. Becke's three-parameter exchange, coupled with Lee et al.'s (B3LYP) correlation functional (Becke 1988; Lee et al. 1988), together with Grimme's D3 dispersion correction, was employed for the calculations. A 6-31G ++ (d, p) basis set was exploited. Firstly, ZIF-8 and the gas molecules were optimized, followed by the optimization of the complexes. No geometrical constraints were applied during the optimization calculations. Frequency calculations verified the obtained stationary point. Gauss View package was used for the visualization of the complexes. The interaction energies for the complexes were calculated.

(7)

Here, E_complex is the energy of the ZIF-8 and gas complex, E_ZIF-8 is the energy of the ZIF-8, and is the energy of a gas molecule in their optimized geometries. The energy of the complexes includes the basis set superimposition errors (BSSE) by performing counterpoise corrections.

The interaction energy between ZIF-8 and gas molecules is calculated using the DFT simulation results. The interaction energy reflects the strength of interactions between gas and ZIF-8. The negative value of interaction energy indicates a stable complex. The minimum energy geometries of ZIF-8, CO₂, and CH₄ are displayed in Figure 11 and the stable structures of the complexes are shown in Figure 12. An intermolecular hydrogen bond between the O atom of CO₂ and the H atom of ZIF-8 has a length of 2.85 Å. A weak CH-π interaction in ZIF-8-CH₄ is also observed. The interaction energies for the complex of ZIF-8 with CO₂ and CH₄ are determined. The interaction energy for ZIF-8-CO₂ and ZIF-8-CH₄ are −10.30 and 13.17 kJ/mol, respectively. The interaction energy for ZIF-8-CO₂ is negative, suggesting a spontaneous complexation, while the ZIF-8-CH₄ complexation process is not spontaneous. This indicates that ZIF-8 has more selectivity towards CO₂ than CH₄. This work theoretically calculated the complexation energy between ZIF-8 and gas molecules using DFT calculations, indicating that ZIF-8 is selective towards CO₂ (Lee et al. 1988).

Figure 11

Optimized geometries of (a) ZIF-8, (b) CO2, and (c) CH4.

Optimized geometries of (a) ZIF-8, (b) CO₂, and (c) CH₄.

Figure 12

Optimized geometries of complex (a) ZIF-8-CO2 and (b) ZIF-8-CH4.

Optimized geometries of complex (a) ZIF-8-CO₂ and (b) ZIF-8-CH₄.

X-ray diffraction (XRD)

Listen

The crystalline property was analysed using X-ray diffraction as shown in Figure 13. It represented the central diffraction peak of ZIF-8 and MMMs. The ZIF-8 sample shows a strong peak at two thetas, 7.25, 10.25, 12.61, 14.20, 16.41, 17.84, and 18.23°, corresponding to the intensity of planes (Miralda et al. 2012; Xian et al. 2015). The ZIF-8 obtained phases indicating high crystalline structure. The XRD pattern of the 5 wt% ZIF-8/PSF membrane was similar to the 15M5 membranes. The peak strength for these membranes is strong at 10–23°, which corresponds to the characteristic peak of ZIF-8 at 5–23°. This means that the ZIF-8 was incorporated into PSF through interfacial polymerization.

Figure 13

XRD of ZIF-8 sample and mixed matrix membranes.

Experimental performance

Listen

Pure water flux (PWF) permeation characteristics studies

Listen

Incorporating ZIF-8 NPs into the casting solution can affect the PWF in two main ways. First, it increases the hydrophilicity of the PSF membrane. Second, it affects the permeation performance of the membrane by changing the membrane morphology (Dasgupta et al. 2014). PSF membranes prepared (using the phase inversion method) with PEG as the pore former and NMP as the solvent are tested to observe the impact of different concentrations of ZIF-8 NPs on their permeation behaviour. The permeation features of the membranes were defined in terms of a densification study that includes PWF and compaction factor, PWF at different transmembrane pressure (TMP), and hydraulic permeability. Compaction is the densification of the membrane structure at TMP, resulting in mechanical deformation and reduced hydraulic permeability. This is common in membrane applications (Saini et al. 2019). In this investigation, all membranes were pressurized at a constant pressure of 250 kPa for 1 h until steady-state flux was achieved. The PWF was obtained from the value of the experimental permeate volume quantified every 5 min. Figure 14 shows the plot of PWF versus time during the densification of all membranes. From M0 to M5 membranes, it was observed that a gradual decrease in flux at the initial densification time duration and PWF reached a steady-state value after about 30 min of densification. This was because all membranes had an asymmetric structure; high-pressure compaction generates compression of the porous support layer, resulting in the thickening of the skin layer, leading to a reduction in pore size and flux (Saini et al. 2019; Zhang et al. 2020). To observe the influence of the ZIF-8 mixture on the permeation properties of the MMMs, the pure water permeation of the membrane was determined and observed, and the pure water flux of the membrane with immersed ZIF-8 nanoparticles increased up to 5 wt%. Further increasing the concentration of ZIF-8 above 5 wt% PWF may decrease due to agglomerated nanoparticles in the membrane (Karimi et al. 2019a, 2019b). Additionally, it can be seen in SEM images of the membrane, where the membranes have a finger-like structure that initially gradually changed into sponge-like structures from M0 to M5. Beyond 5 wt%, increasing the weight percentage of ZIF-8 in the polymeric solution, it became dense and decreased PWF (Zhang et al. 2013).

Figure 14

Pure water flux studies for mixed matrix membranes.

Gas permeation studies

Listen

Pure gas studies

Listen

The plain PSF and ZIF-8/PSF MMMs have been fabricated using the solution casting method and investigated to study the effect of operating pressure and different loading wt% of MMMs, as shown in Figure 15. The plain PSF membrane has low permeability at pressure 7, 8, and 9 bars. However, the selectivity increases at each pressure because the PSF matrix was attributed to minimum selective CO₂ polar gas molecules through the membrane. But the membrane has two characteristics: permeability and selectivity. It was needed to improve the permeability. So, PSF was incorporated with different wt% (1, 3, and 5) of ZIF-8, the fabricated ZIF-8/PSF MMMs. Furthermore, with the increasing pressure at 7, 8, and 9 bars, more and more polar CO₂ gas molecules are attributed to the surface matrix. The molecular size of CO₂ is 3.3A⁰, and CH₄ is 3.8A⁰ (Furukawa et al. 2014). Hence, ZIF-8 has a pore size of 3.4A⁰ and is close to the molecular size of polar CO₂ gas, which was not attributed to non-selective flow channels for CH₄ gas (Deng et al. 2020). We observed the pressure increases, and then the permeability of CO₂ increases, as shown in Figure 15(a). Also, the loading 1, 3, and 5 wt% of ZIF-8 in PSF MMMs was increased effect on enhanced CO₂ gas diffusion mechanism in the MMMs because ZIF-8 filler has strong adsorption ability, as shown in Figure 15(a) (Samarasinghe et al. 2018). This result was heightened to improve polar CO₂ gas molecules diffuse through MMMs; the results of ZIF-8/PSF MMMs show the permeability of CO₂ increases with increasing the pressure and loading wt% ZIF-8. However, in the case of CH₄ gas permeation studies, the permeability of CH₄ was greatly increased when increasing the pressure and loading wt% ZIF-8 in MMMs, as shown in Figure 15(b). From the permeation result, we observed that the permeability of CO₂ of 5 wt% ZIF-8/PSF MMMs (10.54 barrer) increased compared to plain PSF membrane (7.02 barrer). However, the selectivity of CO₂/CH₄ decreases, as shown in Figure 15(c). Beyond 9 bars of pressure, the plasticization effects occurred on MMMs (Karimi et al. 2019b; Ying et al. 2019); this affects the excess swelling on MMMs and decreases the selectivity of membranes due to segmental mobilization, more polar and non-polar gas molecules diffuse through MMMs.

Figure 15

Pure gas permeation studies of mixed matrix membranes.

Mixed gas studies

Listen

The permeability studies of mixed gas (CO₂:CH₄ = 30:70 m L/min) are evaluated at 9 bar pressure. The graphical representations of mixed gas performance are shown in Figure 16. The permeability of CO₂ was increased in all MMMs, and in the case of CH₄, the permeability of CH₄ was increased as increasing wt% ZIF-8 in MMMs as shown in Figure 16(a), its effect on the selectivity of mixed gases CO₂/CH₄. The permeability of CO₂ increases but the selectivity of mixed gases decreases as shown in Figure 16(b). The 5 wt% ZIF-8 MMM has the highest permeability and selectivity with 12.30 barrer and 9.09, respectively, compared to plain PSF membrane (7.17 barrer and 6.07). The permeability of CO₂ and selectivity of CO₂/CH₄ (MMMs) of the mixed gas is higher than the result of pure gas separation studies. This is often because each gas in the mixture has an individual competing for adsorption and plasticization. However, the CO₂ pressure is much higher in mixed gas separation than the plasticization pressure, which suggests that competitive sorption is a significant factor in the deterioration of mixed gas CO₂/CH₄ selectivity (Ying et al. 2019; Deng et al. 2020).

Figure 16

Mixed gas permeation studies of mixed matrix membranes.

CONCLUSION

Listen

In this work, we synthesized ZIF-8, which was the primary key to improving the permeability of gas and water flux. ZIF-8 was incorporated into PSF (using phase inversion and solution casting method) and fabricated MMMs (porous and non-porous) in which ZIF-8 made a barrier through which the selective component of molecules diffuse through it. These membranes were characterized with FTIR, TGA, SEM, WCA, DFT, and XRD, to study the functional group of filler, thermal decomposition temperature, cross-sectional morphology, hydrophilicity, complexation energy studies (CO₂, CH₄), and crystalline structure of the MMMs, respectively. We studied water and gas permeation characteristics based on the ZIF-8 mechanism in MMMs. The water flux of the 15M1, 15M3, and 15M5 membranes was increased by 152, 248, and 456 L/m²h, respectively, compared to the 15M0 (Plain PSF) membranes. We observed a higher water flux of 456 L/m²h at 15M5 membranes. Similarly, in pure gas (CO₂) and mixed gas (CO₂/CH₄) permeation studies, the permeability of CO₂ was increased 10.57 and 12.30 barrer, respectively, as compared to plain PSF (7.02 barrer). From experimental results, the CO₂ permeability was higher in mixed gas permeation studies and has a novelty for mixed gas studies. In this work, we calculated the complexation energy between ZIF-8 and CO₂ gas molecules as −10.30 kJ/mol using DFT studies, indicating that ZIF-8 is selective towards CO₂ molecules. From both experimental and theoretical results, we observed that ZIF-8 is an excellent and promising filler for studying the permeation characteristics with MMMs.

ACKNOWLEDGEMENT

Listen

We thank Pandit Deendayal Energy University and the Indian Institute of Technology (IIT) Gandhinagar (India) for supporting the central instrumental facilities for materials characterization.

DATA AVAILABILITY STATEMENT

Listen

All relevant data are included in the paper or its Supplementary Information.

CONFLICT OF INTEREST

Listen

The authors declare there is no conflict.

REFERENCES

Aframehr

W. M.

,

Rouhollah

B. M.

&

Davoodi

S. M.

2020

Characterization and enhancement of the gas separation properties of mixed matrix membranes: polyimide with nickel oxide nanoparticles

.

Chemical Engineering Research and Design

.

doi:10.1016/j.cherd.2019.11.006

.

Metal-organic framework/graphene oxide composite fillers in mixed-matrix membranes for CO₂ separation

Anastasiou

S.

,

Bhoria

N.

&

Porkhrel

J.

2018

.

Materials Chemistry and Physics

212

,

513

–

522

.

doi:10.1016/j.matchemphys.2018.03.064

.

Azizah

N.

,

Norhaniza

Y.

&

Woei

J.

2021

Polyethersulfone ultrafiltration membrane incorporated with ferric-based metal-organic framework for textile wastewater treatment

.

Separation and Purification Technology

270

(

April

),

118819

.

doi:10.1016/j.seppur.2021.118819

.

Banerjee

R.

,

Hiroyasu

F.

&

David

B.

2009

Control of pore size and functionality in isoreticular zeolitic imidazolate frameworks and their carbon dioxide selective capture properties

.

Journal of the American Chemical Society

131

(

11

),

3875

–

3877

.

Barooah

M.

&

Mandal

B.

2018

Synthesis, characterization and CO₂ separation performance of novel PVA/PEG/ZIF-8 mixed matrix membranes

.

Journal of Membrane Science

.

doi:10.1016/j.memsci.2018.11.001

.

Density-functional exchange-energy approximation with correct asymptotic behavior

Becke

A. D.

1988

.

Physical Review A

38

(

1988

),

3098

–

3100

.

https://doi.org/10.1103/PhysRevA.38.3098

Zeolitic imidazolate framework materials: recent progress in synthesis and applications

Chen

B.

,

Yang

Z.

&

Zhu

Y.

2014

.

Journal of Materials Chemistry A: Materials for Energy and Sustainability

2

,

16811

–

16831

.

doi:10.1039/C4TA02984D

.

Choi

J. Y.

,

Huang

R.

&

Uriberomo

F.

2006

Exceptional chemical and thermal stability of zeolitic imidazolate frameworks

.

Science

103

(

27

),

8

–

13

.

Dasgupta

J.

,

Chakraborty

S.

&

Sikder

J.

2014

The effects of thermally stable titanium silicon oxide nanoparticles on structure and performance of cellulose acetate ultrafiltration membranes

.

Separation and Purification of Technology

133

,

55

–

68

.

doi:10.1016/j.seppur.2014.06.035

.

Effects of the morphology of the ZIF on the CO₂ separation performance of MMMs

Deng

J.

,

Dai

Z.

&

Deng

L.

2020

.

Industrial and Engineering Chemistry Research

59

(

32

),

14458

–

14466

.

doi:10.1021/acs.iecr.0c01946

.

Fabrication of high-performance mixed-matrix membranes via constructing an in-situ crosslinked polymer matrix for gas separations

Deng

G.

,

Luo

J.

,

Liu

X.

&

Liu

S.

2021

.

Separation and Purification Technology

271

(

April

),

118859

.

doi:10.1016/j.seppur.2021.118859

.

The influence of pore aperture, volume and functionality of isoreticular gmelinite zeolitic imidazolate frameworks on the mixed gas CO₂/N₂ and CO₂/CH₄ separation performance in mixed matrix membranes

Essen

M. V.

,

Akher

L. D.

&

Thur

R.

2021

.

Separation and Purification Technology

260

(

October 2020

),

118103

.

doi:10.1016/j.seppur.2020.118103

.

Ultrahigh porosity in metal-organic frameworks

Furukawa

H.

,

Nakeun

K.

,

Aratani

N.

&

Yaghi

O.

2010

.

Science

329

(

5990

),

424

–

428

.

doi:10.1126/science.1192160

.

Furukawa

H.

,

Furukawa

H.

&

Cordora

K.

2014

The chemistry and applications of metal-organic frameworks

.

Science

341

,

1230444

.

Removal of hazardous material from wastewater by using metal organic framework (MOF) embedded polymeric membranes

Gnanasekaran

G.

&

Balaguru

S.

2019

.

Separation Science and Technology (Philadelphia)

54

(

3

),

434

–

446

.

doi:10.1080/01496395.2018.1508232

.

Aqueous one-step modulation for synthesizing monodispersed ZIF-8 nanocrystals for mixed-matrix membrane

Jiang

X.

,

He

S.

,

Han

G.

&

Long

J.

2021

.

ACS Applied Materials and Interfaces

13

(

9

),

11296

–

11305

.

doi:10.1021/acsami.0c22910

.

High-flux PVDF mixed matrix membranes embedded with size-controlled ZIF-8 nanoparticles

Karimi

A.

,

Khataee

A.

&

Vatanpour

V.

2019a

.

Separation and Purification Technology

229

(

July

),

115838

.

doi:10.1016/j.seppur.2019.115838

.

Karimi

S.

,

Firouzfar

E.

&

Khoshchehreh

M. R.

2019b

Potental adsorb

.

Journal of Petroleum Science and Engineering

173

(

June 2018

),

13

–

19

.

doi:10.1016/j.petrol.2018.10.012

.

Metal-organic frameworks (MOFs) as futuristic options for wastewater treatment

Kumar

P.

,

Bansal

V.

&

Kim

K. H.

2018

.

Journal of Industrial and Engineering Chemistry

62

,

130

–

145

.

doi:10.1016/j.jiec.2017.12.051

.

Le

T.

,

Chen

X.

,

Dong

H.

&

Tarpeh

W.

2021

An evolving insight into metal organic framework-functionalized membranes for water and wastewater treatment and resource recovery

.

Industrial & Engineering Chemical Research

60

(

19

),

6869

–

6907

.

doi:10.1021/acs.iecr.1c00543

.

Development of the Colle-Salvetti correlation energy formula into a functional of the electron density

Lee

C.

,

Yang

W.

,

Parr

R. G.

,

Hill

C.

&

Carolina

N.

1988

.

Physical Review B

37

(

1988

),

785

–

789

.

Exploration of porous metal–organic frameworks for gas separation and purification

Lin

R. B.

,

Xiang

S.

&

Xing

H.

2019

.

Coordination Chemistry Reviews

378

,

87

–

103

.

doi:10.1016/j.ccr.2017.09.027

.

Influence of solvent, Lewis acid–base complex, and nanoparticles on the morphology and gas separation properties of polysulfone membranes

Maghami

S.

,

Sadeghi

M.

&

Baghersad

S.

2021

.

Polymer Engineering and Science

61

(

7

),

1931

–

1942

.

doi:10.1002/pen.25708

.

Meshkat

S.

,

Kaliaguine

S.

&

Rodrigue

D.

2019

Comparison between ZIF-67 and ZIF-8 in Pebax MH-1657 mixed matrix membranes for CO₂ separation

.

Separation and Purification Technology

235

,

116150

.

doi:10.1016/j.seppur.2019.116150

.

Zeolitic imidazole framework-8 catalysts in the conversion of CO₂ to chloropropene carbonate

Miralda

C. M.

,

Macias

E.

,

Zhu

M.

&

Carreon

M.

2012

.

ACS Catalysis

2

(

1

),

180

–

183

.

doi:10.1021/cs200638h

.

Highly flame retardant zeolitic imidazole framework-8@cellulose composite aerogels as absorption materials for organic pollutants

Nabipour

H.

,

Nie

S.

,

Wang

X.

&

Hu

Y.

2020

.

Cellulose

27

(

4

),

2237

–

2251

.

doi:10.1007/s10570-019-02860-9

.

Role of filler porosity and filler/polymer interface volume in metal-organic framework/polymer mixed-matrix membranes for gas separation

Nuhnen

A.

,

Dietrich

D.

&

Millan

S.

2018

.

ACS Applied Materials and Interfaces

10

(

39

),

33589

–

33600

.

doi:10.1021/acsami.8b12938

.

Qian

Q.

,

Asinger

P.

&

Lee

M.

2020

MOF-based membranes for gas separations

.

120

(

16

),

8161

–

8266

.

doi:10.1021/acs.chemrev.0c00119

.

CO₂/CH₄ mixed gas separation using graphene oxide nanosheets embedded hollow fiber membranes: evaluating effect of filler concentration on performance

Sainath

K.

,

Modi

A.

&

Bellare

J.

2021

.

Chemical Engineering Journal Advances

5

(

October 2020

),

100074

.

doi:10.1016/j.ceja.2020.100074

.

Effect of hydrophilic poly(ethylene glycol) methyl ether additive on the structure, morphology, and performance of polysulfone flat sheet ultrafiltration membrane

Saini

B.

&

Sinha

M. K.

2019

.

Journal of Applied Polymer Science

136

(

10

),

1

–

14

.

doi:10.1002/app.47163

.

Incorporation of cross-linked poly(AA-co-ACMO) copolymer with pH responsive and hydrophilic properties to polysulfone ultrafiltration membrane for the mitigation of fouling behaviour

Saini

B.

,

Khuntia

S.

&

Sinha

M. K.

2019a

.

Journal of Membrane Science

572

(

October 2018

),

184

–

197

.

doi:10.1016/j.memsci.2018.11.017

.

Saini

B.

,

Sinha

M. K.

&

Dash

S. K.

2019b

Mitigation of HA, BSA and oil/water emulsion fouling of PVDF ultrafiltration membranes by SiO₂-g-PEGMA nanoparticles

.

Journal of Water Process Engineering

30

(

March

),

100603

.

doi:10.1016/j.jwpe.2018.03.018

.

Salahshoori

I.

,

Nasirian

D.

&

Rashidi

N.

2021

The effect of silica nanoparticles on polysulfone–polyethylene glycol (PSF/PEG) composite membrane on gas separation and rheological properties of nanocomposites

.

Polymer Bulletin

78

,

3227

–

3258

.

doi:10.1007/s00289-020-03255-8

.

Tailoring CO₂/CH₄ separation properties of mixed-matrix membranes via combined use of two- and three-dimensional metal-organic frameworks

Samarasinghe

S. A. S. C.

,

Chuah

C. Y.

&

Yang

Y.

2018

.

Journal of Membrane Science

557

(

April

),

30

–

37

.

doi:10.1016/j.memsci.2018.04.025

.

Sasi

S.

,

Sinha

M. K.

&

Vanshpati

R.

2020

Adsorption and recyclability aspects of humic acid using nano ZIF-8 adsorbent

.

Environmental Technology & Innovation

19

,

100927

.

doi:10.1016/j.eti.2020.100927

.

Polysulfone mixed-matrix membranes comprising poly(ethylene glycol)-grafted carbon nanotubes: mechanical properties and CO₂ separation performance

Singh

S.

,

Varghese

A.

&

Reddy

K.

2021

.

Industrial and Engineering Chemistry Research

60

(

30

),

11289

–

11308

.

doi:10.1021/acs.iecr.1c02040

.

Susanti

I

.

2019

Polysulfone membrane with zeolite filler for CO₂/CH₄ gas separation: a review

.

Science Education and Application Journal

1

(

1

),

1

–

10

.

doi:10.1002/chem.201201563

.

Uribe-romo

F. J.

,

Knobler

M.

,

Keeffe

M.

,

Phan

A.

,

Doonan

C. J.

&

Yaghi

A. M.

2010

Capture properties of zeolitic imidazolate frameworks

.

Science

43

(

1

),

58

–

67

.

Surface modification of polyvinylidene fluoride membranes with ZIF-8 nanoparticles layer using interfacial method for BSA separation and dye removal

Vatanpour

V.

&

Khorshidi

S.

2020

.

Materials Chemistry and Physics

241

,

122400

.

doi:10.1016/j.matchemphys.2019.122400

.

ZIF-301 MOF/6FDA-DAM polyimide mixed-matrix membranes for CO₂/CH₄ separation

Wang

Z.

,

Yuen

J.

&

Zhu

H.

2021

.

Separation and Purification Technology

264

(

January

),

118431

.

doi:10.1016/j.seppur.2021.118431

.

Polydopamine-modified metal-organic framework membrane with enhanced selectivity for carbon capture

Wu

W.

,

Zhanjun

L.

&

Chen

Y.

2019

.

Environmental Science and Technology

53

(

7

),

3764

–

3772

.

doi:10.1021/acs.est.9b00408

.

Vapor-enhanced CO₂ adsorption mechanism of composite PEI @ ZIF-8 modified by polyethyleneimine for CO₂/N₂ separation

Xian

S.

,

Xu

F.

,

Ma

C.

&

Wu

Y.

2015

.

Chemical Engineering Journal

280

,

363

–

369

.

doi:10.1016/j.cej.2015.06.042

.

Yang

Z.

,

Ao

D.

,

Guo

X.

&

Nil

L.

2021

Preparation and characterization of small-size amorphous MOF mixed matrix membrane

.

Separation and Purification Technology

272

(

399

),

1

–

8

.

doi:10.1016/j.seppur.2021.118860

.

Plasticization resistance-enhanced CO₂ separation at elevated pressures by mixed matrix membranes containing flexible metal-organic framework fillers

Ying

Y.

,

Youdong

C.

,

Peh

S. B.

&

Liu

G.

2019

.

Journal of Membrane Science

582

,

103

–

110

.

doi:10.1016/j.memsci.2019.03.088

.

Zhang

K.

,

Lively

R.

&

Zhange

C.

2013

Investigating the intrinsic ethanol/water separation capability of ZIF-8: an adsorption and diffusion study

.

Journal of Physical Chemistry C

117

(

14

),

7214

–

7225

.

Water-soluble MOF nanoparticles modified polyethersulfone membrane for improving flux and molecular retention

Zhang

S.

,

Liu

D.

&

Li

D.

2020

.

Applied Surface Science

505

,

144553

.

doi:10.1016/j.apsusc.2019.144553

.