Abstract
Hydrophobic metakaolin-based flat sheet membrane was developed via phase inversion and sintering technique and modified through 1H,1H,2H,2H-perfluorooctyltriethoxysilane grafting agents. The prepared membrane was characterized by different techniques such as XRD, FTIR, SEM, contact angle, porosity, and mechanical strength. Their results indicated that the wettability, structural, and mechanical properties of the prepared membrane confirm the suitability of the material for membrane distillation (MD) application. The prepared metakaolin-based flat sheet membrane acquired hydrophobic properties after surface modification with the water contact angle values of 113.2° to 143.3°. Afterward, the membrane performance was tested for different sodium chloride aqueous solutions (synthetic seawater) and various operating parameters (feed temperature, feed flow rate) using direct contact membrane distillation (DCMD). Based on the findings, the prepared membrane at metakaolin loading of 45 wt.% and sintered at 1,300 °C was achieved the best performance with >95% salt rejection and permeate flux of 6.58 ± 0.3 L/m2 · h at feed temperature of 80 °C, feed concentration of 35 g/L, and feed flow rate of 60 L/h. It can be concluded that further optimization of membrane porosity, mechanical, and surface properties is required to maximize the permeate flux and salt rejection.
HIGHLIGHTS
The flat sheet ceramic membrane was synthesized for DCMD.
The sintering process was a much more influential factor in the membrane shrinkage.
The flat sheet ceramic membrane was successfully grafted with PFAS molecules.
Effect of operating variable on the permeate flux and salt rejection.
It was found that the feed temperature was the most significant operating variable affecting the performance of the DCMD.
Graphical Abstract
INTRODUCTION
Water is essential for all socio-economic development in many areas (Brusseau et al. 2019; Gomez et al. 2019). In recent years, the scarcity of safe and clean drinking water is among the main global challenges. This is due to higher standards of living, population growth, rapid industrial growth and advancement, and climate change (Brusseau et al. 2019; Kumari et al. 2021; Zulfiqar et al. 2021). To provide a sufficient amount of safe and clean drinking water, both thermal (non-membrane) and membrane-based water desalination techniques are required (Zewdie et al. 2021a). Thermal-based water desalination technologies include multi-effect distillation (MED), multistage flash (MSF) distillation, and vapor compression distillation (VCD) (Nafey et al. 2006; Baten & Stummeyer 2013; Gude 2018; Feria-Díaz et al. 2021). Membrane-based water desalination technologies include reverse osmosis (RO), electrodialysis (ED), pervaporation (PV), forward osmosis (FO), membrane distillation (MD), and hybrid membrane system (Sanmartino et al. 2016; Aliyu et al. 2018; Zewdie et al. 2021a). Membrane-based water desalination technologies are more energy-efficient, economic, and sustainable than thermal-based water desalination technologies (Baten & Stummeyer 2013; Roy & Ragunath 2018; Zewdie et al. 2021a). Among membrane-based water purification techniques, membrane distillation represents an emerging and promising approach for desalination and wastewater treatment application (Sanmartino et al. 2016; Alkhudhiri & Hilal 2018; Ashoor et al. 2018; Hussain et al. 2021; Saavedra et al. 2021; Tai et al. 2021). Membrane distillation is a hybrid separation technique combining thermal-driven distillation and membrane separation processes (Belessiotis et al. 2016; Ghaffour et al. 2019). Based on the type of configuration, membrane distillation is classified into direct contact membrane distillation (DCMD), sweeping gas membrane distillation (SGMD), vacuum membrane distillation (VMD), air gap membrane distillation (AGMD), and permeate gap membrane distillation (PGMD) (Gugliuzza & Basile 2013; Eykens et al. 2016a; Mahmoudi et al. 2018; Parani & Oluwafemi 2021).
Polymeric membranes (polypropylene (PP), polysulfone (PSF), polytetrafluoroethylene (PTFE), and polyvinylidene fluoride (PVDF)) (Shirazi et al. 2013; Ravi et al. 2020; Parani & Oluwafemi 2021) and ceramic membranes (alumina, titania, zirconia) have been widely studied for membrane distillation application (Hubadillah et al. 2019a; Bandar et al. 2021). As compared to polymeric membranes, ceramic membranes can withstand extreme conditions/harsh environments due to their excellent mechanical strength, biocompatibility, high thermal stability, high chemical stability, long lifetime, energy efficiency, availability, and sustainability (Fang et al. 2012; Arumugham et al. 2021). Ceramic membranes are known to have better antifouling and easy cleaning properties, which allows for repeated reuse (multiple uses) (Wang et al. 2018; Yang & Tang 2018). Ceramic membranes have the possibility to remove part of contamination (deposition and/or adsorption of solutes) from the pores by regeneration using various chemical agents (HCl, NaOH, NaClO, H2O2) and nano air bubbles (Hakami et al. 2020), however, the initial membrane performance and separation efficiency may not be restored (Kim & Jang 2016).
In this study, locally available and inexpensive Ethiopian kaolin was used as the primary raw material for the preparation of metakaolin-based flat sheet membranes. Some researchers prepared kaolin-based flat sheet membranes by various fabrication methods such as combining phase inversion and sintering (Hubadillah et al. 2016a, 2018a), and dry pressing and sintering (Sahnoun & Baklouti 2013). Their results indicated that a thin flat sheet of kaolin-based membranes was difficult to handle during installation and application (Hubadillah et al. 2016a, 2016b, 2018a, 2018b). These negative consequences include easy break (brittleness) and high crack sensitivity features which have hindered a thin flat sheet of kaolin-based membranes for membrane distillation application. Thus, for addressing some of these shortcomings, the current study was proposed to develop metakaolin-based flat sheet membranes by phase inversion and sintering technique. Metakaolin is one type of calcined clay mineral and a well-known material for producing high mechanical strength (compressive and tensile strength) of silica-based ceramic membrane and, high-performance concretes (Dinakar et al. 2013; Hubadillah et al. 2016c; Khatib et al. 2018). A metakaolin-based flat sheet membrane is expected to be hydrophilic due to the presence of hydroxyl groups (OH) on its surface. This hydrophilicity is not suitable for the membrane distillation process (Kujawa et al. 2014a, 2016; Hubadillah et al. 2019a; Twibi et al. 2021). To improve the hydrophobicity of a ceramic membrane, it is necessary to modify the surface of membranes. There are three ways commonly used methods of surface modification of polymeric and ceramic membrane, which are chemical modification, surface morphology modification, and combination of chemical and morphology modifications (Khemakhem et al. 2013, 2014; Usman et al. 2021). Among these methods, grafting with perfluoroalkylsilanes (PFAS) is the most commonly applied, especially in membrane distillation applications, because the molecules of the modifiers are covalently attached to the surface as well as inside the porous structure of the ceramic membranes/materials (Kujawa et al. 2014a, 2014b, 2016; Abu-Zeid et al. 2015; Zuo & Chung 2016; Hubadillah et al. 2019a; Twibi et al. 2021).
This study aims to fabricate a metakaolin-based flat sheet membrane by combining phase inversion and sintering techniques for water desalination via DCMD. The main purpose of this study was to investigate the effect of metakaolin content and sintering temperature on the surface morphology, porosity, hydrophobicity property, and mechanical strength of the eventual metakaolin-based flat sheet membrane. This is the first broader study on the use of modified metakaolin-based flat sheet ceramic membranes in the membrane distillation process.
MATERIALS AND METHODS
Raw materials
Beneficiated and calcined Ethiopia kaolin (metakaolin) powder from our previous study (Zewdie et al. 2021b) was used as the ceramic material. Analytical grade polyethersulfone (PES; Radel 3100P, Solvay Advanced Polymer), polyethyleneglycol-30 dipolyhydroxystearate (Arlacel P135, CRODA, Belgium), and N-methyl-2-pyrrolidone (NMP; VWR International bvba, Belgium) were purchased and used as a polymer binder, dispersant, and solvent, respectively. Deionized water was used as the non-solvent coagulation bath. 1H,1H,2H,2H-perfluorooctyltriethoxysilane (PFAS; 97%, VWR International bvba, Belgium) was used as a grafting agent, and ethanol (>99.8%, Sigma-Aldrich, Belgium), acetone, and alkaline solution (NaOH at pH 10).
Fabrication of metakaolin flat sheet membrane
The preparation of metakaolin-based flat sheet membrane was conducted at four different metakaolin contents with four different sintering temperatures (1,200, 1,300, 1,400, and 1,500 °C) (Figure 1). The detailed compositions of the dope solutions are listed in Table 1.
Membranes . | 1 . | 2 . | 3 . | 4 . |
---|---|---|---|---|
Metakaolin (wt.%) | 37.5 | 40 | 42.5 | 45 |
NMP (wt.%) | 56.5 | 54 | 51.5 | 49 |
PES (wt.%) | 5 | 5 | 5 | 5 |
Arlacel P135 (wt.%) | 1 | 1 | 1 | 1 |
Membranes . | 1 . | 2 . | 3 . | 4 . |
---|---|---|---|---|
Metakaolin (wt.%) | 37.5 | 40 | 42.5 | 45 |
NMP (wt.%) | 56.5 | 54 | 51.5 | 49 |
PES (wt.%) | 5 | 5 | 5 | 5 |
Arlacel P135 (wt.%) | 1 | 1 | 1 | 1 |
Beneficiated and calcined kaolin (metakaolin) powder and PES were dried to ensure that no moisture was trapped in the particle. Then, the required quantity of NMP was taken in a 250 ml glass bottle and PES was slowly added and stirred by a hot plate magnetic stirrer for 4 h under continuous moderate stirring (300 rpm) to form the polymer solution. Arlacel P135 was then added as a dispersant into a polymer solution and stirred by a hot plate magnetic stirrer for 24 h under continuous moderate stirring (300 rpm). After the polymer solution was formed, metakaolin powder was then added into a polymer solution slowly and then stirred by a hot plate magnetic stirrer for 72 h under continuous moderate stirring (300 rpm) at 60 °C to ensure that the metakaolin powder and polymer solution were mixed well. The resulting dope solution was degassed in an ultrasonic bath for 30 min at room temperature to eliminate the air bubbles. The casting dope solution was cast on a casting machine and left for the evaporation process to occur for 30 s before the solvent exchange in the coagulation bath. The cast slurry was left in the water bath (2 L deionized water) for 24 h to let the phase inversion process be completed. Afterward, the membrane precursors were dried at room temperature for 24 h. Before the sintering process, the membrane precursors were cut in a square form (100 mm × 100 mm). The membrane precursors were then placed in an electric furnace and sintered in a controlled furnace to 600 °C at a rate of 2 °C min−1 and held for 2 h, then to target temperature (1,200, 1,300, 1,400, and 1,500 °C) at a rate of 5 °C min−1 and held for 4 h. Lastly, the furnace was cooled down at a rate of 5 °C min−1 to room temperature.
Preparation of hydrophobic metakaolin-based flat sheet membrane
Metakaolin-based membranes are hydrophilic. Hydrophilic membranes are not suitable to be used for membrane distillation applications (Hubadillah et al. 2019a; Twibi et al. 2021). Thus, surface modification/grafting is required to obtain hydrophobic properties. Four equal size metakaolin-based flat sheet membranes prepared at various sintering temperatures (1,200, 1,300, 1,400, and 1,500 °C) were used for this treatment. The membranes were cleaned using ethanol, acetone, and distilled water for 10 min, respectively. The samples were dried in an oven at 105 °C for approximately 12 h and have been treated by using an alkaline solution (NaOH at pH 10) for 8 h.
Characterization of membranes
DCMD test
An experimental laboratory-scale DCMD setup was constructed with an effective membrane area of 0.0025 m2 (50 mm × 50 mm) located in a plate and frame (flat sheet membranes) module made of Plexiglas™ with 90 mm × 90 mm dimensions. The desalination performance of metakaolin-based flat sheet membranes in different operating conditions including feed inlet temperature, feed inlet flow rate, feed inlet concentration under counter-current flow pattern was investigated. The membrane had an approximate thickness of 400 μm and exhibit low values of effective thermal conductivity. The membrane cell consisted of two compartments, the feed side and the permeate side. Figure 2 shows the experimental setup.
The synthetic solutions were prepared by dissolving a reagent grade NaCl salt (supplied by Fluka) in distilled water to obtain different salt concentrations (0, 5, 15, 25, and 35 g/L) and served as the feed solution on the hot side of the module. The feed container was immersed in a water bath and heated until it reaches a predetermined operating temperature (50, 60, 70, and 80 °C). The hot feed was circulated with a peristaltic pump (Model, Sci-Q 323) into the membrane module with a flow rate of 30, 40, 50, and 60 L/h. A coiled heater was used to regulate the temperature of the hot stream. Deionized water was used as a cooling liquid on the permeate side of the module from a double-walled cooling water container by a peristaltic pump (Model, Sci-Q 323) at a constant flow rate of 40 L/h. A cooling thermostat was used to regulate the temperature of the cold stream.
The inlet temperatures of the hot feed and the cold water were measured by two digital thermometers (Model, HI98509 Checktemp®1). The electrical conductivity or the salt concentration of the feed and permeate solution was measured by a conductivity meter (Radiometer CDM230, Sweden) inserted into the vessel. After the flow rates of the hot solution, cold distillate water, and the two inlet and outlet temperatures were stabilized, it was assumed that the experimental conditions had reached a steady state; the permeated liquid was circulated through a double-walled cooling water reservoir, and the volume measured at regular intervals was used to calculate the water vapor flux through the membrane under the given experimental conditions.
RESULTS AND DISCUSSION
Characterization of starting materials and membranes
The physical properties of beneficiated Ethiopian kaolin and beneficiated and calcined kaolin (metakaolin) were reported in previous work (Zewdie et al. 2021b). Powder particle size and the distribution of the starting material (raw material) are critical factors in the fabrication of ceramic membranes (Hubadillah et al. 2016a; Al-Naib 2018). According to the literature, the powder particle size used in the fabrication of ceramic membrane is mainly in the range of 1 μm (Harabi et al. 2014; Hubadillah et al. 2019b) to 15 μm (Zhou et al. 2010). In this study, the particle size of beneficiated and calcined kaolin (metakaolin) is sufficiently small suggesting its suitability for ceramic membrane fabrication. The powder particle size and the distribution affected the viscosity of the ceramic suspension, the movement of the particles in ceramic suspension during phase inversion, and the particle migration and bindings during sintering at high temperatures (Konijn et al. 2014; Renteria et al. 2019; Li et al. 2020a). Moreover, powder particle size and the distribution of the starting material (raw material) affected the morphology characteristic and the porosity of the fabricated ceramic membrane (Hubadillah et al. 2016a, 2021; Ji et al. 2020).
XRF analyses
The chemical composition of beneficiated Ethiopian kaolin and beneficiated and calcined kaolin (metakaolin) were reported in previous work (Zewdie et al. 2021b). The analysis showed that the composition of SiO2 and Al2O3 varied between 60.10 and 60.30%, and 27.60 and 29.40%, respectively. On the other hand, the loss on ignition varied between 10.7 and 0.98%. These amounts were close to the values determined for theoretical kaolin and metakaolin. Due to its lower iron oxide content and SiO2/Al2O3 mass ratio, and higher Al2O3/Fe2O3 mass ratio, beneficiated and calcined kaolin (metakaolin) is a suitable cheap raw material for developing ceramic membranes for water purification.
XRD measurements
Figure 3 presents the XRD patterns of beneficiated Ethiopia kaolin, beneficiated and calcined kaolin (metakaolin), and metakaolin-based flat sheet membrane. It can be noted that the XRD patterns of beneficiated Ethiopia kaolin, beneficiated and calcined kaolin (metakaolin) powder, and metakaolin-based flat sheet membrane reported in Figure 3 depict a clear difference. During the calcination process, the kaolinite structure is disordered due to dihydroxylation (Cheng et al. 2019; Izadifar et al. 2020). Due to this, the complex structural transformation of kaolinite mullite is formed. As shown in the previous study (Zewdie et al. 2021b), XRD peaks at Bragg's angles of 12.4°, 25.1°, 32.2°, 40.5°, and 44.1° represented the presence of kaolinite minerals in beneficiated Ethiopian kaolin powder. Similar trends have been reported elsewhere (Djobo et al. 2014; Douiri et al. 2017; Khan et al. 2017; Obada et al. 2017b) for other types of clay used in ceramic membrane fabrication. During the sintering process, kaolinite peaks disappeared whereas mullite ceramic peaks were formed in all XRD patterns for metakaolin-based flat sheet membrane (Sahnoun & Baklouti 2013; Abdulhameed et al. 2017a, 2017b; Obada et al. 2017b). The XRD patterns showed that by increasing the sintering temperature from 1,200 to 1,500 °C, the peak intensity of mullite ceramic became stronger which indicated that the amount of mullite increased with increasing sintering temperature (Chen et al. 2008; Abdulhameed et al. 2017a, 2017b; Mohtor et al. 2017; Liu et al. 2021; Ndjigui et al. 2021). Besides that, XRD peaks at Bragg's angles of 27.3°, 27.5°, and 60.6° were also observed in beneficiated Ethiopian kaolin, beneficiated and calcined kaolin (metakaolin) powder, and metakaolin-based flat sheet membrane, respectively, which indicated the presence of non-clay mineral (quartz). XRD analysis revealed a significant predominance of quartz and kaolinite in the beneficiated Ethiopian kaolin clay. These observations are in agreement with the X-ray fluorescence (XRF) results in the previous study (Zewdie et al. 2021b). It has also been noticed that the beneficiated Ethiopian kaolin powder used in this study consisted of quartz as the non-clay mineral in which the XRD peaks remained for metakaolin-based flat sheet membranes sintered at different target temperatures (1,200–1,500 °C) (Sahnoun & Baklouti 2013; Mohtor et al. 2017; Magalhaes et al. 2020). According to the XRD analysis, it was also proved that the higher sintering temperature has no dramatic change on the mullite phase presented in metakaolin-based flat sheet membrane (Figure 3). Additionally, the intensity of the XRD peaks of quartz decreased with temperature (Mohamed Bazin et al. 2019; Magalhaes et al. 2020).
TGA/DSC analyses
The TGA/DSC is an important analysis that needs to be performed on the powder samples. As shown in the previous study (Zewdie et al. 2021b), TGA/DSC curves were obtained from the simultaneous thermal analysis of the beneficiated Ethiopian kaolin and beneficiated and calcined kaolin (metakaolin) powder. The total weight loss of mass was 9.12 wt.% for beneficiated Ethiopian kaolin and 0.55 wt.% for beneficiated and calcined kaolin (metakaolin). The TGA/DSC analysis has proven that the thermal properties of beneficiated Ethiopian kaolin and beneficiated and calcined kaolin (metakaolin) powders are suitable for possible industrial use, especially for ceramic membrane fabrication.
FTIR measurements
FTIR spectra of the beneficiated Ethiopia kaolin powder, beneficiated and calcined kaolin (metakaolin) powder, and metakaolin-based flat sheet membrane are shown in Figure 4. According to the result presented in Figure 4(a), beneficiated Ethiopia kaolin powder has OH-bending modes at 1,122 cm−1 and OH-stretching modes at 3,619 and 3,688 cm−1 (Khan et al. 2017; Mohtor et al. 2017; Boussemghoune et al. 2020). Moreover, the absorption band at around 1,626 cm−1 was observed in beneficiated Ethiopia kaolin powder due to its deformation vibration of physisorbed water molecules at the surface (Khan et al. 2017; Kljajević et al. 2017; Aragaw & Angerasa 2020). The absorption band for beneficiated Ethiopia kaolin powder at 920 cm−1 was attributed to Al–OH bending vibration (Aragaw & Angerasa 2020). This sharp band disappeared in all spectra of beneficiated and calcined kaolin (metakaolin) powder and metakaolin-based flat sheet membranes as a result of the bonds breaking between the octahedral sheet and tetrahedral sheet of kaolinite structure due to the mullitization process of kaolinite at a high sintering temperature and for a long sintering time (Mohtor et al. 2017). Beneficiated and calcined kaolin (metakaolin) powder and metakaolin-based flat sheet membranes sintered at different temperatures do not contain OH because of the loss of structural hydroxyl group due to the sintering process. As a result of the sintering process, these characteristic bands for kaolinite structure diminished in all spectra of metakaolin powder and metakaolin-based flat sheet membranes, showing that the sintering process was adequate for complete dehydroxylation and structural distortion of kaolinite and its transformation into metakaolin (Mohtor et al. 2017; Merabtene et al. 2019; Magalhaes et al. 2020).
The band between 900 and 1,150 cm−1 is relatively wide in beneficiated and calcined kaolin (metakaolin) powder, but in metakaolin membranes sintered at different temperatures, it becomes a relatively narrow band which explains the strong bond of aluminosilicate molecules in metakaolin membranes resulting in the increase of its mechanical strength.
Morphological properties of membranes
During the phase inversion process, shrinkage of the flat sheet precursor occurred (Bikel et al. 2010; dan Sinteran 2017). This is because the rate of solvent (NMP) diffusion from the suspension is always faster than the rate of diffusion of water into the suspension (Bonyadi et al. 2007; dan Sinteran 2017). Similarly, during the sintering process, organic components (polymer binders) were removed from the precursors, and only inorganic mineral was left in the metakaolin-based membranes after the sintering process (Wang et al. 2009; Paiman et al. 2015). Then, the shrinkage of the metakaolin-based flat sheet membrane was determined using dimensions (volume) of the membrane before and after sintering. Figure 5 depicts the percentage of shrinkage volume for the metakaolin-based flat sheet membrane sintered at a targeted temperature (1,200–1,500 °C) in different beneficiated and calcined kaolin (metakaolin) contents (37.5–45 wt.%). The maximum shrinkage of the volume for the metakaolin-based flat sheet membrane sintered at a targeted temperature of 1,200, 1,300, 1,400, and 1,500 °C in 45 wt.% metakaolin is found to be 10.31 ± 4.25%, 31.24 ± 3.74%, 46.97 ± 1.42%, and 53.93 ± 0.21%, respectively.
The more the shrinkage, the higher the internal stress generation, which results in more shape distortions and severe cracks or warps in the final sintered ceramic membrane (Green et al. 2008; Ni et al. 2013). This is due to the internal rearrangement (densification process) of the membrane at a higher temperature. It can be concluded that the shrinkage of the metakaolin-based flat sheet membrane increases with the sintering temperature. Similar results have been reported in previous studies (Qiu et al. 2009; Vasanth et al. 2011; Mankai et al. 2018; Elgamouz et al. 2019; Jiang et al. 2019; Li et al. 2020b; Ndjigui et al. 2021; Souza et al. 2021) for clay materials used in ceramic membrane fabrication. These results indicate that the metakaolin-based flat sheet membrane is suitable for application in DCMD for water desalination (Eykens et al. 2016b).
Evaluating the grafting efficiency
After chemical modification, the grafting efficiency of the flat sheet metakaolin-based ceramic membranes was determined by measurement of grafting degree and contact angle measurement.
Degree of grafting
The dependence of grafting time on grafting degree with membrane sintered at different sintering temperatures (1,200, 1,300, 1,400, and 1,500 °C) was evidenced in Figure 6. The grafting process was carried out at room temperature at different grafting times (12, 24, 48, 72, and 96 h). The grafting degree of the reaction system could change for membranes sintered at different sintering temperatures. The grafting degree of metakaolin-based membrane sintered at 1,200 °C was 3.25 ± 0.1%, 3.43 ± 0.25%, 3.94 ±0.13%, 4.6 ± 0.06%, and 4.03 ± 0.14% corresponding to a reaction time of 12, 24, 48, 72, and 96 h, respectively. As shown in Figure 6, the grafting efficiency is proportional to the grafting time, where the modified membranes presented the highest grafting degree at 72 h. The determined grafting degree values for metakaolin-based membrane increased with grafting time from 12 to 72 h. Thus, the longer membrane hydrophobization leads to the creation of a smoother surface (lower surface roughness). The membrane also has a high contact angle due to a higher level of PFAS grafting agent concentration and covering all active sites on the membrane surface (Kujawa et al. 2016; Li et al. 2021). For a grafting time longer than 72 h, the value of grafting degree was slightly lower. This is due to a long-time interaction between the hydroxyl group on the metakaolin-based ceramic membrane surface and Si–O-alkyl groups of the PFAS grafting agent, which contributes to a decline in the amount of hydroxyl group. This is associated with the fact that contact angle increases with increasing the grafting time from 12 to 72 h, while it decreases with increasing grafting time longer than 72 h (Lu et al. 2009). Furthermore, it can be seen that a membrane sintered at a lower temperature leads to a higher grafting degree (Figure 6). A similar grafting process was also found in the literature for ceramic membrane (Lu et al. 2009; Fang et al. 2012; Kujawa et al. 2013, 2014b, 2016; Hubadillah et al. 2019c; Saud et al. 2021) and clay powder materials (Kujawa et al. 2014b, 2014c).
Contact angle measurements
It was shown that the hydrophilic property of a metakaolin-based flat sheet membrane could be changed into a hydrophobic one by grafting 1H,1H,2H,2H-perfluorodecyltriethoxysilane on the surface of a metakaolin-based flat sheet membrane. The grafting process can be performed by a reaction between OH− groups on the ceramic membrane surface and Si-O-alkyl groups of the silane (Krajewski et al. 2006; Khemakhem & Amar 2011; Khemakhem et al. 2014; Wang et al. 2016; Zuo & Chung 2016; Yang et al. 2017; Twibi et al. 2021). The surface modifying process can decrease the surface free energy and increase the contact angle of the membranes (Kujawa et al. 2017; Shahabadi et al. 2017; Yang et al. 2017; Hubadillah et al. 2018c; Dong et al. 2020). During the sintering process at temperatures between 400 and 800 °C, the hydroxyl groups (OH−) can be suppressed from the membrane surface. Therefore, alkaline pretreatment is required to restore hydroxyl groups (OH−) on the membrane surface and allow more coupling reactions with PFAS. Water drops deposited on the nongrafted metakaolin-based flat sheet membrane form a contact angle of 3.6 ± 0.37° (super hydrophilic) (Figure 7(a)). This is due to the uniform distribution of hydroxyl groups on the membrane surface. After grafting, the contact angle for grafted metakaolin-based flat sheet membrane sintered at a temperature of 1,200 °C was 143.3 ± 0.5°, proving that the metakaolin-based flat sheet membrane surface was successfully changed into superhydrophobic (Figure 7(b)) by grafting with PFAS. These results adequately demonstrate that the modified metakaolin-based flat sheet membrane is hydrophobic and sufficient for application in the DCMD for water desalination (Khayet 2011; Liu et al. 2012; Ren et al. 2015; Yang et al. 2017). Contact angle values of modified membranes were higher than for unmodified membranes, which indicate that grafting the membrane surface revealed a highly hydrophobic character. Similar phenomena have been observed in previous studies (Kujawa et al. 2014a; Wang et al. 2016; Yang et al. 2017; Hubadillah et al. 2018c, 2019c; Polak et al. 2021; Saud et al. 2021; Twibi et al. 2021).
The contact angle measurement of metakaolin-based membranes sintered at different targeted temperatures (1,200, 1,300, 1,400, and 1,500 °C) before and after chemical modification is shown in Figure 8. After grafting, the pore size of the membrane sintered at 1,200 °C became smaller due to PFAS grafted on the metakaolin-based flat sheet membrane surface. The highest hydrophobicity value was obtained by grafted metakaolin-based flat sheet membrane sintered at 1,200 °C with a contact angle value of 143.3 ± 0.5°. This was subsequently described as grafted metakaolin-based flat sheet membrane sintered at 1,200 °C that obtained a microtextured or micropatterned surface as shown in the SEM images (Figure 9).
SEM measurements
The effect of sintering temperature (ranging from 1,200 to 1,500 °C) on the membrane surface morphology of metakaolin-based flat sheet membrane before and after grafting was investigated by SEM, and corresponding images are shown in Figure 9. The SEM results indicated that no cracks were present in any membranes. The surface morphology of the membrane was found to be uniform and free of any defect. The porous structure and uniformly distributed pore were observed in the sintered membrane at 1,200 °C. The SEM results indicated that the membranes were cast from 45 wt.% metakaolin loading and sintered at 1,200 °C have a higher porosity than the membranes sintered at 1,500 °C. The increase in sintering temperature of the membranes has decreased the porosity of the membrane, from 40.28 ± 0.93% to 14.50 ± 1.89% for temperatures from 1,200 to 1,500 °C, respectively. Thus, the SEM micrographs reported in Figure 9 show that the effect of sintering is very marked; a progressive reduction of porosity can be observed when temperature increases. The number of pores decreased when the sintering temperature increased. At higher sintering temperatures, small pores disappeared and a less porous structure was produced, where sintering at 1,500 °C produced the strongest membrane with a highly dense structure and limited grain and non-interconnected pores. This may be due to the particles binding/aggregating together and it forms a dense structure. The sintered membranes at 1,200, 1,300, 1,400, and 1,500 °C for 4 h showed a regular surface morphology as can be seen in Figure 9. The above observation indicated that the sintering temperature plays an important role in controlling the ceramic membrane porosity and structure. A similar observation was reported in the literature (Ghouil et al. 2015; Das et al. 2016a, 2016b; Wang et al. 2016; Mohtor et al. 2017; Mouiya et al. 2019).
In addition, it can be observed in Figure 9, that the structural and surface morphology was unchanged after PFAS grafting: this is due to the limited number of hydroxyl groups (OH−) left on the metakaolin-based ceramic membrane surface after sintering at high temperatures (1,200–1,500 °C) (Fang et al. 2012; Hubadillah et al. 2018b). It can be concluded that the structure and surface morphology changed gradually with changing composition and sintering temperature (Yang et al. 2017; Twibi et al. 2021).
Chemical stability test
Chemical resistance tests of the unmodified kaolin-based ceramic membrane in an acidic (hydrochloric acid, pH = 1) and basic (sodium hydroxide, pH = 14) environment were reported in the previous study (Zewdie et al. 2021b). The results showed that the prepared flat ceramic specimen offers good chemical stability in acidic (<3% mass loss in acid solution) and an excellent chemical stability in basic media (<1% mass loss in alkali solution). In this study, the grafted metakaolin-based flat sheet membranes were then immersed into hexane for 120 h at room temperature. After contacting the samples with hexane, the water contact angle of the grafted metakaolin-based ceramic membrane surface was found to be 142.7 ± 0.86° (Figure 10). This shows that the modified membranes demonstrate good chemical stability when treated with a harsh solvent. Besides, insignificant changes in their surface hydrophobicity/wetting properties after being in contact with hexane for 120 h were observed. The stability of the modified membranes in a harsh solvent is due to the strong structure and morphology of the ceramic material. In addition, this confirms that 1H,1H,2H,2H-perfluorodecyltriethoxysilane molecules were covalently attached to the surface as well as inside the porous structure of the metakaolin-based membrane and were stable in hexane (Kujawa et al. 2014b). The stability test has shown that the modified metakaolin-based membrane shows good stability in hexane for the PFAS grafting agent.
Porosity measurements
The porosity of the metakaolin-based flat sheet membranes was determined by the Archimedes method (immersion) using distilled water. In this study, the metakaolin loading and sintering process play an important role in controlling the porosity of the metakaolin-based flat sheet membrane. The experimental data shown in Figure 11 indicate the effect of metakaolin loading and sintering temperature on the porosity of a metakaolin-based flat sheet membrane. According to Figure 11, the porosity of the ceramic membrane decreased with increasing temperature from 1,200 to 1,500 °C, resulting in higher flexural strength (Paiman et al. 2015; Mohtor et al. 2017; Li et al. 2020b). This may be due to the close packing of the particles and the denser texture of the ceramic membrane because, at high temperatures, the particles aggregate and form a cross-linked network structure to attain a further solidified structure.
The values obtained for the porosity of the metakaolin flat sheet membranes (45 wt.% metakaolin loading) slightly decreased from 40.28 ± 0.93% to 36.61 ± 1.07% when sintered at 1,200 and 1,300 °C, respectively. Furthermore, increasing the temperature up to 1,400 and 1,500 °C, the porosity sharply decreases from 34.70 ± 0.8% to 14.50 ± 1.89%, respectively, the latter is due to the significant sintering shrinkage and densification. This decline simultaneously enhances the flexural strength of the membranes. As expected, the membrane porosity decreased when the sintering temperature increased (Guechi et al. 2016; Hubadillah et al. 2016c, 2019d, 2020; Wang et al. 2016; Abdulhameed et al. 2017a, 2017b; Mankai et al. 2018; Kadiri et al. 2020; Li et al. 2020b). The micrographs confirm that increasing the sintering temperature decreases the porosity and increases the pore size of the sintered metakaolin-based membrane.
Furthermore, the experimental results indicated that the porosity of the ceramic membrane decreased with increasing metakaolin loading from 37.5 to 45 wt.% (Das et al. 2016a, 2016b; Hubadillah et al. 2016a, 2016c; Figure 11). Porosity is crucial for a metakaolin-based flat sheet membrane to be used for membrane distillation while providing enough pores for the water permeation (Fang et al. 2012; Hubadillah et al. 2016a, 2016c). Hence, controlling the metakaolin loading and sintering temperature is essential for controlling the properties of the metakaolin-based flat sheet membrane (Hubadillah et al. 2016a, 2016c). It can be concluded that the porosity of the metakaolin-based flat sheet membranes ranges from 34.70 ± 0.8% to 57.89 ± 0.43%, indicating that the membrane is sufficient for application in the DCMD for water desalination (Khayet 2011; Eykens et al. 2016c).
Mechanical properties
Organic polymer materials such as the polyethersulfone (PES) as a binder and polyethyleneglycol-30 dipolyhydroxystearate (Arlacel P135) as a dispersant in ceramic suspension have to be removed completely from ceramic precursors during the sintering process (Wang et al. 2009; Paiman et al. 2015), before forming the metakaolin-based flat sheet membrane. Therefore, the metakaolin content in the suspension plays a significant contribution to the flexural strength of the metakaolin-based flat sheet membrane. The increase of the metakaolin content in the ceramic suspension would increase the suspension viscosity. It is a determinantal factor to the flexural strength of metakaolin-based flat sheet membrane. Therefore, to effectively fabricate and produce a low-cost metakaolin-based flat sheet membrane with excellent flexural strength, the higher metakaolin content in the ceramic suspension must be maintained during ceramic suspension preparation (Hubadillah et al. 2016a, 2016c). In addition to that, an increase in sintering temperature would enhance the flexural strength of the ceramic membrane. Where, at the higher temperature, the ceramic particles fused and developed larger grains, which gave a higher flexural strength.
The increase in sintering temperature of the metakaolin-based flat sheet membranes (45 wt.% metakaolin loading) raised the flexural strength from 3.4 ± 0.86 MPa to 24.39 ± 1.9 MPa but decreased the porosity of the metakaolin-based flat sheet ceramic membrane from 40.28 ± 0.93% to 14.50 ± 1.89% for temperatures from 1,200 to 1,500 °C, respectively. Findings indicate that the porosity and flexural strength of the metakaolin-based flat sheet membranes can be controlled by varying the metakaolin loading and the sintering temperature.
In general, the flexural strength of ceramic membrane tends to enhance with increasing sintering temperature and decreasing porosity (Sahnoun & Baklouti 2013; Guechi et al. 2016; Wang et al. 2016; Hubadillah et al. 2016c, 2019d, 2020; Abdulhameed et al. 2017a, 2017b; Mohamed Bazin et al. 2019; Li et al. 2020b; Ndjigui et al. 2021; Souza et al. 2021). Based on the above-observed results, it can be said that a metakaolin-based flat sheet membrane sintered at 1,300, 1,400, and 1,500 °C has a sufficient flexural strength compared to a membrane sintered at 1,200 °C. This is in good agreement with the literature (Feng et al. 2004; Essalhi & Khayet 2013; Abdulhameed et al. 2017a, 2017b; Mohtor et al. 2017). It is seen from Figure 12 that a rapid increase in the flexural strength for 1,500 °C sintered membrane is due to the larger grain growth and densification of the membrane. Thus, it is concluded that the flexural strength of the membranes ranges from 7.3 ± 0.92 to 24.39 ± 1.9 MPa, indicating that the membrane is suitable for application in the DCMD for water desalination (Eykens et al. 2016c) and in the filtration process (Sahnoun & Baklouti 2013). Moreover, a metakaolin-based ceramic membrane could be used for a high-strength application at a low cost (Hubadillah et al. 2019d).
Performance of ceramic membrane in DCMD
Previous experimental studies have investigated the effect of flow patterns on permeate flux and salt rejection in desalination by DCMD under countercurrent-flow and concurrent-flow operations. Based on the obtained results, a higher permeate flux was observed in the counter-current flow arrangement. Thus, counter-current flow is preferable to co-current flow (He et al. 2011; Hwang et al. 2011; Duong et al. 2017). Therefore, in this study, a counter-current flow arrangement was applied in all experiments. Moreover, the operational conditions can significantly influence the performance of the membrane distillation process for desalination (Gryta 2012; Zhang et al. 2013; Francis et al. 2014; Duong et al. 2016; Ameen et al. 2020). In order to examine the performance of metakaolin-based flat sheet membrane in DCMD the effect of processing time on permeate flux for membranes was investigated. The effect of operating parameters on the permeate flux and salt rejection was assessed. Three parameters at different levels were studied, namely feed temperature in the 50 to 80 °C range, feed flow rate of 30 to 60 L/h, and feed concentration ranging from 0 to 35 g/L.
Effect of processing time
Based on the characterization results, the metakaolin-based flat sheet membrane sintered at 1,200 °C has inadequate flexural strength (3.4 ± 0.86 MPa). Due to that, the membrane could not withstand the operating conditions (not tolerate higher pressure/feed flow velocity) during the operations, which does not allow for extended process runs (running time: 3 h). Therefore, a preliminary membrane desalination experimental study with a metakaolin-based membrane sintered at 1,200 °C shows a negligible permeate flux (not predict the permeate flux) for the range of feed flow rate tested. Thus, it is not suitable for membrane distillation (Eykens et al. 2016c). Figure 13 shows the change in the permeate flux as a function of process time for metakaolin-based membranes sintered at different temperatures (1,300, 1,400, and 1,500 °C). The results demonstrate that the behavior of the permeate flux as a function of process time consists of two distinct stages: In the first stage, the permeate flux declined. In the second stage of the process, a constant permeate flux was observed. Permeate flux decreased with process time in all experimental runs for all membranes. A slight decline in the permeate flux was observed during the first 60 min for all membranes. The permeate flux remained constant after 60 min for time on stream of 180 min. The obtained results showed that the metakaolin-based flat sheet membrane sintered at 1,300 °C has the highest water permeation of 6.58 ± 0.3 kg/m2h, followed by the membranes sintered at 1,400 and 1,500 °C with a water permeation of 4.33 ± 0.65 and 1.93 ± 0.43 kg/m2h, respectively. This is due to the slightly higher porosity of the metakaolin-based membrane sintered at 1,300 °C (about 36.61 ± 1.07%) compared to membranes sintered at 1,400 °C (about 34.70 ± 0.8%) and 1,500 °C (about 14.50 ± 1.89%). The metakaolin-based flat sheet membranes showed a comparable performance (permeate flux) results reported elsewhere (Kujawa et al. 2014a; Zhang et al. 2014; Wang et al. 2016; Hubadillah et al. 2019e).
Effect of feed temperature
Feed temperature is a key operational parameter for DCMD. The permeate flux and salt rejection of metakaolin-based flat sheet membrane for artificial saline water during 3 h running time at different feed temperatures in counter-current flow operation are presented in Figure 14. Typically, the temperature at the feed stream is adjustable and the temperature at the coolant inlet stream is fixed to study the effect of temperature on permeate flux and salt rejection. The feed temperature was investigated in the range of 50 to 80 °C at 10 °C intervals with a feed solution of 35 g/L; the feed inlet flow rate, coolant inlet flow rate, and temperature were maintained at 60 L/h, 60 L/h, and 20 ± 0.5 °C, respectively.
At feed inlet temperatures 50, 60, 70, and 80 °C, the permeate flux increases were 2.58 ± 0.21, 3.17 ± 0.09, then 5.67 ± 0.11, and 6.58 ± 0.3. These results reflected the increase of the permeate flux, when the feed inlet temperature increased from 50 to 80 °C for a fixed feed inlet solution of 35 g/L, and permeate inlet temperature of 20 ± 0.5 °C. The vapor pressure of a liquid increases exponentially with feed inlet temperature, as described in the Antoine equation. Therefore, at high feed inlet temperatures, the permeate flux increased exponentially. The obtained results are in good agreement with the previously reported results (Fang et al. 2012; Singh & Sirkar 2012; Kujawa et al. 2014a; Lee et al. 2015; Shim et al. 2015; Khalifa et al. 2017; Luo & Lior 2017; Ameen et al. 2020; Twibi et al. 2021).
Interestingly, the salt rejections remained almost constant as the feed inlet temperature increased from 50 °C (salt rejection: 95.17 ± 0.2%) to 80 °C (salt rejection: 95.0 ± 0.17%). The high salt rejection (>95%), as shown in Figure 14, indicates the good desalination performance in the DCMD process.
Effect of feed flow rate
The influence of feed inlet flow rate on the permeate flux and salt rejection in the DCMD process in counter-current flow is illustrated in Figure 15. The increase in feed inlet flow rate in the DCMD process ranging from 30 to 60 L/h, leads to an increase in the permeate flux from 2.03 ± 0.14 to 6.58 ± 0.3 kg/m2 h for a feed concentration of 35 g/L, at a feed inlet temperature of 80 °C and a constant coolant inlet flow rate and temperature of 60 L/h and 20 ± 0.5 °C, respectively. As displayed in Figure 15, the permeate flux increased linearly with increasing feed flow rate. In most previous studies, the permeate fluxes increased with the feed flow rate (Shirazi et al. 2014; Shim et al. 2015). Generally, a higher feed flow rate leads to higher turbulence, which results in the better mixing of the feed solution, which in turn enhanced the mass and heat transfer coefficient (Qusay et al. 2017; Chen et al. 2020). This is due to the reduction in the temperature and concentration polarization boundary layer thickness. Thus, the temperature difference across the membrane sides increased and resulted in an improved permeate flux (Qusay et al. 2017).
Furthermore, it can be seen from Figure 15 that the salt rejection is slightly lower at a higher feed flow rate. This is due to the lower residence time inside the membrane distillation module with high feed flow rates and therefore enhanced heat loss by conduction and poor heat recovery (Guillén-Burrieza et al. 2015; Subrahmanya et al. 2021).
Effect of feed concentration
The experiments were carried out at various feed concentrations (i.e., 5, 15, 25, and 35 g/L NaCl) in counter-current flow keeping the feed inlet temperature, coolant inlet temperature, and feed and coolant flow rate at constant values of 80 °C, 20 °C, and 60 L/h, respectively. Figure 16 shows the effect of the feed concentration on the permeate flux and salt rejection at constant feed flow rates.
The results demonstrated that the permeate flux slightly decreased with increasing the feed salt concentration from 5 to 35 g/L. Many studies reported that permeate flux in the DCMD process decrease with increasing feed salt concentration because the water vapor pressure decreases at higher feed salt concentration based on Raoult's law (Fang et al. 2012; Kujawa et al. 2014a, 2016; Hubadillah et al. 2018b; Twibi et al. 2021). This is because more salt molecules are deposited and accumulated on the membrane surface at the feed side, formation of fouling/scaling layer on the membrane surface, an increase of resistance in transfer, and induced wetting of the membrane on the feed side which ultimately leading in a vapor pressure reduction. Moreover, it causes a decrease of the driving force across the membrane and an increase of salt concentration of permeate flux. Thus, this study and others (Qusay et al. 2017; Ameen et al. 2020) all confirmed that salt rejection decreases with increasing salt concentration.
Table 2 lists the comparison between several studies reporting on ceramic membrane properties (hydrophobicity) and performances (permeate flux and salt rejection) along with study results. The permeate flux and salt rejection obtained in this study were low (permeate flux: 1.9–6.58 kg/m2 h, salt rejection: >95%) compared to previous studies on hollow fiber and tubular membrane module for a feed temperature of 80 °C (see Table 2). The obtained results for the prepared metakaolin-based flat sheet membrane sintered at 1,300 °C (45 wt.% metakaolin loading) can be explained by the lower porosity (36.61 ± 1.07%) and flexural strength (14.88 ± 1.5 MPa). Despite the relatively low performance, there is no study yet on the performance of metakaolin-based flat sheet membrane in terms of water permeate flux and salt rejection using DCMD to directly compare with our findings. This work shows that the properties of metakaolin-based flat sheet membranes can be further fine-tuned to obtain a better membrane performance. Therefore, metakaolin-based flat sheet membrane fabrication needs further investigation under different membrane preparation parameters (metakaolin loading, sintering temperature, the concentration of the additives, and mixing practice) to obtain desirable separation performances (permeate flux and salt rejection) as well as membrane properties (porosity and flexural strength). Nevertheless, considering the abundant availability and further refined the preparation methods. It will be an attractive alternative for seawater desalination in the membrane distillation process and can achieve comparable membrane performance with hollow fiber and tubular membrane modules.
No . | Membrane materials . | Membrane module type . | Modifying agent (PFAS) . | Application . | Contact angle (°) . | Porosity (%) . | Permeate flux (kg/m2 · h) . | Salt rejection (%) . | Reference . |
---|---|---|---|---|---|---|---|---|---|
1 | Titania | Tubular | 1H,1H,2H,2H-perfluorooctyltriethoxysilane | Desalination (NaCl) | 135–145 | – | 0.9–3.0 | >99 | Kujawa et al. (2014a) |
2 | Silicon nitride | Hollow fiber | 1H,1H,2H,2H-perfluorooctyltriethoxysilane | Desalination (NaCl) | 136 | 50 | 9.2 | >99 | Zhang et al. (2014) |
3 | β-Sialon | Hollow fiber | 1H,1H,2H,2H-perfluorooctyltriethoxysilane | Desalination (NaCl) | 125 | 48 | 12.2 | >99 | Wang et al. (2016) |
4 | Green Silica | Hollow fiber | 1H,1H,2H,2H-perfluorodecyltriethoxysilane | Desalination (NaCl) | 157–161 | 35.86–54.12 | 38.2–52.4 | 99.9 | Hubadilla et al. (2018b) |
5 | Metakaolin | Hollow fiber | 1H,1H,2H,2H-perfluorodecyltriethoxysilane | Deionized water | 120 | – | 5.3–16.5 | – | Hubadillah et al. (2019e) |
6 | Mullite | Hollow fiber | 1H,1H,2H,2H-perfluorodecyltriethoxysilane | Desalination (NaCl) | 139 | 43 | 30.55 | 99.99 | Twibi et al. (2021) |
7 | Metakaolin | Flat sheet | 1H,1H,2H,2H-perfluorooctyltriethoxysilane | Desalination (NaCl) | 113.2–143.3 | 14.88–36.61 | 1.9–6.58 | >95 | This study |
No . | Membrane materials . | Membrane module type . | Modifying agent (PFAS) . | Application . | Contact angle (°) . | Porosity (%) . | Permeate flux (kg/m2 · h) . | Salt rejection (%) . | Reference . |
---|---|---|---|---|---|---|---|---|---|
1 | Titania | Tubular | 1H,1H,2H,2H-perfluorooctyltriethoxysilane | Desalination (NaCl) | 135–145 | – | 0.9–3.0 | >99 | Kujawa et al. (2014a) |
2 | Silicon nitride | Hollow fiber | 1H,1H,2H,2H-perfluorooctyltriethoxysilane | Desalination (NaCl) | 136 | 50 | 9.2 | >99 | Zhang et al. (2014) |
3 | β-Sialon | Hollow fiber | 1H,1H,2H,2H-perfluorooctyltriethoxysilane | Desalination (NaCl) | 125 | 48 | 12.2 | >99 | Wang et al. (2016) |
4 | Green Silica | Hollow fiber | 1H,1H,2H,2H-perfluorodecyltriethoxysilane | Desalination (NaCl) | 157–161 | 35.86–54.12 | 38.2–52.4 | 99.9 | Hubadilla et al. (2018b) |
5 | Metakaolin | Hollow fiber | 1H,1H,2H,2H-perfluorodecyltriethoxysilane | Deionized water | 120 | – | 5.3–16.5 | – | Hubadillah et al. (2019e) |
6 | Mullite | Hollow fiber | 1H,1H,2H,2H-perfluorodecyltriethoxysilane | Desalination (NaCl) | 139 | 43 | 30.55 | 99.99 | Twibi et al. (2021) |
7 | Metakaolin | Flat sheet | 1H,1H,2H,2H-perfluorooctyltriethoxysilane | Desalination (NaCl) | 113.2–143.3 | 14.88–36.61 | 1.9–6.58 | >95 | This study |
CONCLUSION
This study was concerned with the development and characterization of hydrophobic metakaolin-based flat sheet ceramic membranes for water desalination by DCMD. Based on the findings or outcomes of the present studies, the following conclusions can be drawn:
Based on the findings, the metakaolin loading of 45 wt.% and the sintering temperature of 1,300 °C could be selected as the best parameters in preparing the metakaolin flat sheet membrane and provided desirable membrane performance in terms of porosity, flexural strength, hydrophobicity, water permeate flux, and salt rejection.
It was noted that the most significant operating variable affecting the performance of the DCMD process was the feed temperature. To a lower extent, the feed flow rate and concentration had a clear effect on the permeate flux. The permeate flux reached 6.58 ± 0.3 kg/m2 h and salt rejection was as high as 95%, the feed inlet temperature was 80 °C and the feed inlet flow rate and concentration were 60 L/h and 35 g/L NaCl, respectively.
The flux and salt rejections reported in this work (flat sheet membrane module) are low compared to those reported in the literature (hollow fiber and tubular membrane modules) (see Table 2). This can be attributed to the low membrane porosity (36.61 ± 1.07%). Thus, this study could provide new insights on the utilization of metakaolin-based flat sheet membranes in the field of the advanced separation process. Nevertheless, further optimization of membrane porosity, mechanical, and surface properties is required to maximize the permeate flux and salt rejection.
ACKNOWLEDGEMENTS
The author would like to thank and appreciate the Ethiopian government/Ministry of Higher Education and the German government for their financial support and special thanks to the KU Leuven and the Bahir Dar University, for providing the necessary facilities for this research.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.