Graphene oxide incorporated cellulose triacetate/cellulose acetate nanocomposite membranes for forward osmosis desalination

Doaa F. Ahmed; Heba Isawi; Nagwa A. Badway; A.A. Elbayaa; Hosam Shawky

doi:10.1016/j.arabjc.2021.102995

View/Download PDF

Buy Reprints

PDF

Translate this page into:

Original article

2021

:14;

202103

doi:

10.1016/j.arabjc.2021.102995

Graphene oxide incorporated cellulose triacetate/cellulose acetate nanocomposite membranes for forward osmosis desalination

Doaa F. Ahmed^{^a,^c}, Heba Isawi^{^a,^b,⁎}, Nagwa A. Badway^{^c}, A.A. Elbayaa^{^c}, Hosam Shawky^{^b}

a Egyptian Desalination Research Center of Excellence (EDRC), Desert Research Center, Cairo, Egypt

b Desert Research Center, Water Resources and Desert Soils Division, Hydrogeochemistry Dept., Water Desalination Unit, Egyptian Desalination Research Center of Excellence (EDRC), Cairo, Egypt

c Chemistry Department, Faculty of Science, Al-Azhar University, Nasr City, Cairo, Egypt

⁎Corresponding author. hebaessawi@hotmail.com (Heba Isawi)

Received: 2020-9-1, Accepted: 2021-1-4,

Disclaimer:
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.

Abstract

This paper deals with the effect of acetic acid (AA) as a pore-forming additive and as the polymeric improvers in the casting mixture of cellouse triacetate/cellouse acetate (CTA/CA) blend flat sheet membrane using phase inversion procedure. Graphene oxide sheets (GO NSs) were selected to enhance the membrane performance for forwarding osmosis (FO) application in saline water desalination. The surface properties and morphology of synthesized FO membranes were also considered using contact angle, porosity measurements, FTIR spectroscope, mechanical stability, and SEM. The average contact angles for the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes were 71^°± 2, 63^°± 2.5, and 49^°± 1.8 and the porosity displayed about 47 ± 2.5%, 49 ± 3%, and 62 ± 2.8%, respectively. The GO/AA/CTA/CA NC modified membrane with lower content of GO NSs (0.4 wt%) showed a higher water flux (33.6 L/m².h), salt rejection (99.88%), and lower reverse solute flux (1.45 g/m².h), when compared to the neat CTA/CA and AA/CTA/CA membranes using 0.1 M NaCl as feed solution (FS) and 1 M NH₄Cl as a draw solution (DS). The influences of various fertilizer draw solutions are also studied. The results reveal that the GO/AA/CTA/CA NC modified membrane showed the highest water flux using 1 M from different fertilizer draw solutes (DFDS) include KCl and NH₄Cl (20.5 L/m².h) and followed by (NH₄)₂SO₄ (20.2 L/m².h) and K₂HPO₄ (18.1 L/m².h) as draw solutions (DS) under the FO approach and using a natural saline water sample collected from the Mersa Matruh area, North-Western Coast of Egypt with salinity 12760 mg/L and PH (8.5) as feed solution (FS). The reusability test of the synthesized GO/AA/CTA/CA NC modified membrane showed good sustainability during the 1260 min continuous test. The FO application displayed a great potential to be in saline water desalination and enhanced water source sustainability to use in agriculture.

Keywords

Forward osmosis Desalination

CTA/CA polymer blend

Acetic acid pore-forming additive

Graphene oxide nanocomposite

Fertilizer draw solution

Saline water desalination

Show Related Articles from PubMed

1 Introduction

The shortage of fresh and clean water has become the main critical problems for sustainable enlargement to meet the economic, developing social and ecological requests of the society (Isawi, 2019). Membrane-based separation is one of the highest prevalent technologies for desalination of saline/brackish water and water reuse (Isawi et al., 2016). Forward osmosis (FO), a developing desalination procedure, has attracted increasing consideration in both industrial enlargement and scientific research in the latest years (Ahmed et al., 2020; Qasim et al., 2015). FO uses an osmotic-pressure variance among the draw and feed liquid via a semi-penetrable membrane as the driving strength to persuade the freshwater to flow via the membrane to the draw liquid (Xiao et al., 2016). FO has been promoted as an abundant water recovery and low-cost desalination favorite (Chung et al., 2012). Two central keys that affect the competitive and successful FO technologies are the membrane substantial and the draw solution (Su et al., 2012).

The most commonly used commercial membranes for FO applications are cellulose acetate (CA) and cellulose triacetate (CTA). The CA is the most famous FO membranes because of its unique advantages such as good mechanical strength, extensive availability, relatively high hydrophilicity that preferable water flux and low fouling affinity and high resistance to degradation by chlorine and further oxidants. The CTA is an attractive polymeric constituent for desalination applications because it has a wonderful salt rejection potential, practical mechanical strength, relatively low cost, improved oxidant resistance, antifouling tendency and can be produced as a dense film newly; numerous researches works on the structure and performance promotion of FO membrane have been issued. The composite CA/CTA blends the FO membrane is then anticipated to compromise the salt retention and water flux. Consequently, the CA/CTA blend membranes are one of the highest FO membranes obtainable in advertising. Newly, researchers have been using new types of porous and hydrophilic materials as additives to enhance the membrane properties and some wonderful efforts have been motivated on the improvement of the membrane performances. Mostly, utilizing pore-forming improver is an operative strategy to control and optimize the membrane structure such as membrane porosity, pore tortuosity, and thickness so as to increase the separation performance (Shahlol et al., 2020). The membrane modifications using additive can affect its performance, and the improver adding is the easiest and highest economical technique to enhance the anti-fouling of the hydrophobic membrane. Acetic Acid (AA) acts as pore-forming additives where it might form complex with a solvent such as 1.4-dioxide in the casting solution, as well as the complex may release during the stage inversion process. AA was incorporated in the casting solution to be a softener that improved polymer mobility for the development of the casting polymer structure. Besides, AA can also act as a pore-creating factor where it results in a more porous and open-cell sub-layer. (Ong and Chung 2012) combined AA as pore-forming improver into the CTA casting matrix with dioxane/acetone solvent organization and created that the resultant FO membrane had a considerably high porous and open-cell sub-layer structure, leading to the water flux increased from 3.5 to 22.7 L/m².h make use of DI water and 2 M NaCl liquid as FS and DS, respectively.

In the last decades, several hydrophilic nanofillers including aluminum oxide (Al₂O₃) (Ahmed et al., 2020), carbon nanotubes (CNTs) (Shahlol et al., 2020), silica (SiO₂) (Isawi, 2019), zeolite (Isawi, 2020), zinc oxide (ZnO) (Isawi et al., 2016) and TiO₂ (Isawi, 2018) have been employed to be combined into the membranes to enhance its performances. Graphene oxide (GO) is one of the most favorable nanosheets (NSs) that have a two-dimensional structure and atomically thick. GO NSs have an enormously high surface area, feature ratio, resistance to bacteria, and mechanical and chemical stability (Tu et al., 2013; Wang et al. 2016). The superior properties of GO NSs are also reflected in the GO/polymer nanocomposite (Chee et al., 2015). Owing to the oxygen-containing reactive groups in GO NSs (e.g., epoxy groups on the basal level, carboxyl groups at the edge, and hydroxyl), they commonly have better dispensability in polar solvents or water than other nano-materials (Kim et al., 2014). The GO NSs have newly attracted significant consideration in membrane improvement because of their mechanical and physical possessions and unique nano-structures (He et al., 2015). Because of the diverse reactive groups (e.g., OH, COOH, and epoxide) in GO NS, there is high compatibility among the GO NSs with the polymer matrix via covalent or non-covalent connections (Liang et al., 2014). When GO NSs are incorporated into a membrane matrix, the membrane surface hydrophilicity will be improved, which is helpful to increase water permeability and fouling resistance. Furthermore, the incorporated GO NSs in the membrane matrix also enhance the mechanical strength and improve the membrane stability against high trans-membrane pressures (Yin and Deng 2015). The latest studies have used GO as a nano-filler for the creation of CA-FO membranes. Wang et al. modified the CTA/GO membrane to improved anti-biofouling and mechanical possessions for the FO membrane by adding a low amount of GO NSs (Wang et al., 2016). At optimum GO NSs loading, the water flux of the membrane increased to 18.43 L/m².h, and the salt flux was 0.22 g/L when deionized (DI) water was utilized as FS and 0.5 M NaCl was used as DS, and the hydrophilicity improved related to the neat CTA membrane.

The main influence that has hindered the successful improvement of the FO procedure is the obtainability of a perfect DS. Draw solutions have an essential role in instituting osmotic pressure inclines, and therefore the FO technology is affected by the content of draw solution (Johnson et al., 2018). A perfect DS should be economical, commercially obtainable, offer high water flux, and have low fouling prospective, inferior reverse solute dispersion, small or no toxicity to bacteria's, and ease of recovery/reinforcement (Panagopoulos et al., 2019). Researchers have discovered the probable effects of DS features and mass transport resistance on membrane performance. Fertilizer draw solutions (FDS) are good contestants for the FO procedure, and they were used in former studies (Suwaileh et al., 2019). Various inorganic FDSs have been examined in the FO procedure, and the diluted DS was utilized for direct fertigation. The concept of fertigation indicates to the usage of a diluted fertilizer as an irrigation method to provide agricultural lands. One of its benefits is as a cost-effective technique to offer the required nutrients to harvests and plants. After the FO purification, the final diluted FDS becomes less concentrated and utilized for direct irrigation. Due to a low-energy process facet, it is believed that FDFO can be economically used for saline/brackish water desalination to provide useable water to the largest water-consuming farming sector. FO technique could easily be used for plentiful basins of sea/saline water along extensive coastal regions and inland underground brackish water to get low-cost advantageous agricultural water (Majeed et al., 2015).

In this article, synergistic properties of mixed additives will work under the appropriate mixing conditions to improve the membrane assembly and performance of CTA/CA blend membranes with acetic acid (AA) and graphene oxide sheets (GO NSs). The objectives of this paper are developing high-flux CTA/CA FO membranes and study the essential science and engineering toward the fabrication of modified membranes. Membranes are tested in FO processes the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes characterized by FTIR spectroscope, contact angle measurement, membrane porosity, SEM, AFM, and mechanical strength. Then, we conducted the FO experiment to assess water flux and solute reverse flux, in addition to membrane stability for FO water desalination using various kinds of fertilizer draw solutes include; KCl, NH₄Cl, (NH₄)₂SO₄, and (K₂HPO₄). The permeate water flux experimental is conducted using DI waters as FS, 1 M NaCl as DS, active layer faces the FS (AL-FS) at 25^◦C and the membrane performances test was completed using 0.1 M NaCl as FS, 1 M NH₄Cl as DS, AL-FS at 25 °C. To show the novelty of the GO/AA/CTA/CA NC modified membrane for desalination of saline water sample so the applicability of the particular membrane was revealed using saline water sample collected from Mersa Matruh area, North-Western Coast of Egypt with salinity 12760 mg/L as a FS and 1 M of the different types of DSs include KCl, NH₄Cl, (NH₄)2SO₄, and K₂HPO₄ as DS with FS on the active site of the membrane feed solution reservoirs were placed on weighing balances until the constancy of the salt rejection and water flux has been reached.

2 Experimental

2.1

2.1 Materials

The membrane substrate synthesized from a combination of cellulose acetate (CA) with 39.7 wt% acetyl content (Mn = 30,000) and cellulose triacetate (CTA) with an averaged acetyl content of 43–44 wt% purchased from Across Organics (USA). A mixture of 1,4-dioxane and acetone with purity 99.5 and 99.9.5%, respectively, were utilized as a solvent and delivered by Sigma Aldrich. Graphene oxide nanosheets (GO NSs) with nano-size of 15–20 nm were used as an additive and delivered via Sigma Aldrich. Acetic acid (AA) with purity 99% was delivered by Sigma-Aldrich. Sodium chloride (NaCl), ammonium chloride (NH₄Cl), potassium chloride (KCl), and Ammonium sulfate (NH₄)₂SO₄ obtained from Adwick, Nasr Pharma, Egypt and used as draw solutions (DSs). The distilled water (DI) was provided from an ultra-pure H₂O model (pure lab Option-K technique, UK), exhausting a resistivity of greater than 15 mΩ-cm.

2.2

2.2 Preparation of modified nanocomposite flat-sheet GO/AA/CTA/CA based membranes

The schematic chart of the membrane creation procedure offered in Fig. 1. First, the CA and CTA powders dried over 70–80 °C using an oven till constant weight. The neat CTA/CA membrane created via the phase inversion technique. The neat CTA/CA membrane was cast from homogeneous polymer solutions having an identical polymer content having 14 wt% CTA/CA (1:2) powder and 86 wt% of a mixture of 1,4-dioxane/acetone (3:1) as a solvent, as in our previous publication (Ahmed et al., 2020). The membrane modification achieved via AA as a pore-forming additive (1.75 to 8.25 ml per unit fraction of CTA/CA wt.% content) were bled with CTA/CA casting mixture and stirred overnight at 70 ^°C and then degassed to form a homogeneous solution to fabricate AA/CTA/CA membrane. To improve the AA/CTA/CA membrane performances, the graphene oxide (GO NSs) was incorporated into AA/CTA/CA casting polymer mixture, Fig. 1. To prepare GO/AA/CTA/CA NC modified membrane, an appropriate amount of GO NSs ranged from (0.1–0.5 as a wt% of CTA/CA concentration, i.e., 14 wt% CTA/CA (1:2)) mixed with 86 ml of 1,4- dioxane/acetone (60:26) solutions. Next, the mixtures sonicated in an ultrasonic bath at 30–40 °C for at least an hour to allow GO NSs to distribute into the solvent matrix. Then the CTA/CA molar ratio (1:2) powder was added, heated 70 ^°C and stirred at 200 rpm overnight to have the best dispersions of the GO NSs in the polymer matrix then protected overnight for de-gassing at 25 °C, Fig. 1. Afterward, the resultant casting solutions were cast onto a dry and clean horizontal glass plate using a casting knife of 200 μm and a speed of 0.01 m·s⁻¹ using a TQC Automatic film applicator. Formerly, the resulting membrane exposed to air for 120 s followed by immediate immersion in DI water bath for 2 min. The membrane permitted to dry, and gently the solvent evaporated in the air for an adequate time. Then, the plate was dipped into an H₂O bath at 10 °C after exposure in the air for 2 min to complete the phase separation process. After solidification and peel off, the fabricated membranes were dipped in DI water for 24 h before characterization and performance experimental, Fig. 1. The resulting GO/AA/CTA/CA NC membrane was annealed in deionized water at 85 °C for 5 min and finally stored in deionized water before use. The optimized polymer solution composition fixed at (14:60:26) (1CTA:2CTA): dioxane: acetone, 5.25 wt% AA, and 0.4 wt% GO NSs castings on the non-woven support layer (SL) for further application.

Schematic representation of neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes.

2.3

2.3 Membrane characterization and performance:

The membranes characterized by each following FTIR, SEM, porosity, mechanical possessions, and the contact angle. The FTIR examination was estimated using a Genesis Unicam spectrophotometer. The scanning electron microscope (SEM Quanta FEG) was used to characterize the membrane surface and cross-section characteristics. The mechanical possessions of the prepared membranes were estimated by (Universal Analysis Apparatuses, V4.5 A, UTI, and DMA) at 25 °C. The strain and tensile stress were carried out to estimate Young's modulus (Mega pascal, Mpa) were calculated using the following Eq. (1).

(1)

{Y o u n g}^{'} s m o d u l u s (M P a) = \frac{S t r e s s}{S r a i n}

The porosity of the CTA porous membrane was measured by using the gravimetric technique consistent with the equation (Xing et al., 2019).

(2)

p o r o s i t y (ε) = \frac{[\frac{(w_{0} - w_{1})}{ρ_{w a t e r}}]}{[(w_{0} - w_{1}) + (\frac{w_{1}}{ρ_{p}})]}

where ɛ is the porosity of the membrane, W₀ and w₁ are the weights of wet and dried membranes (g), ρ_p and ρ_water are the densities of CTA (1.3 g/cm³) and water (1.0 g/cm³). The final outcome was the mean value of at least five parallel measurements. The contact angle was utilized to estimate the membrane hydrophilicity by a Cam-positive Micro, Tantec Inc. pattern. A water droplet was located onto the membrane surface after air-dried at 25 °C utilizing DI water by a numerical micro-syringe. The contact angle testified with an average 5 measurements were obtained for each membrane sample on different locations.

2.4

2.4 Membrane performance assessment for forward osmosis (FO)

The synthesized membranes were verified using a laboratory-balance FO setup, as presented in Fig. 2. The test cell comprises of two rectangular plastic parallel half cells with the membrane in between. FO performance in expressions of salt rejection and water flux of the resulting neat CTA/CA, AA/CTA/CA and GO/AA/CTA/CA NC modified membranes determined via FO test in FO model with DI water used as feed solution (FS) and 1 M NaCl aqueous solution utilized as a draw solution (DS). Both FS and DS circulated using gear pumps. The test cell includes two rectangular plastic symmetric half-cells with the membrane in between, Fig. 2. The dimensions into the half-cell are 12.7 cm in length, 10 cm in width, and 8.3 cm in height. The membrane was placed in a module with a membrane area of 42 cm² (CF042A-FO), permitting the draw solution and feed solution to flow counter presently on each outline of the membrane with the same velocity of 1.6 L/min as in Fig. 2. The concurrent FS and DS flow were adopted to reduce the strain on the suspended membrane. The flow ratio determined to be 1.6 L/min after an optimization test to reduce the external concentration polarization. A circulator maintained both FS and DSs at 25 °C ± 0.5 °C. FO performance generally was assessed under the FS mode (the FS faced the dense selective layer, and the DS faced the SL) over an hour, by measuring permeate water flux (J_w) and reverse solute flux (RSF, J_s). Calculated data were taken after an initial stabilization period of about 15 min. Water flux via the membrane was simply determined by reading the changes in the volume of the feed solution:

(3)

J_{w} = \frac{Δ V}{A \times Δ t}

where, J_w is the water flux (L/m².h), ΔV is the variation in volume of the feed solution (L), Δt is the time interval (h) for the volume variation of ΔV, and A is the active membrane surface area (m²). In the absence of concentration polarization influences, the ideal or theoretical water flux was calculated by the following equation (Qasim et al., 2017):

(4)

J_{W} = A (π D, b - π F, b)

where, A is the membrane water penetrability factor (ms⁻¹ Pa⁻¹), πD, b is the bulk osmotic pressure of the DS (P_a), and πF, b is the bulk osmotic pressure of the FS (P_a). The reverse DS flux from the DS occurred was assessed using pure DI H₂O as a FS and measuring the initial and final concentrations of the FS. The reverse salt flux (RSF, J_s (g/m².h)) was estimated by the following equation:

(5)

J_{s} = \frac{C_{f} V_{f} - C_{0} V_{0}}{A Δ t}

where, J_s is the RSF (g/m².h, gMH), C₀ and C_f are the initial and final feed concentrations (g/L), respectively, V₀ and V_f are the initial and final volumes of the feed solution (L), respectively, Δt is the time interval (h), and S is the active membrane surface area (m²). The salt rejection (R, %) was calculated as below (Chen et al., 2017).

(6)

R = 1 - \frac{C_{N a c l, d s} \times V_{d s}}{C_{f s 0} \times V_{f s 0}} \times 100

where C _NaCl,ds is the salt content in the DS after a given time, which was determined via standard curve technique by a conductivity meter (DDS-307a), and V_ds is the volume of the DS. C_fs0 and V_fs0 are the initial concentration and volume of the feed solution, respectively.

Forward Osmosis experimental setup for FO membrane test and Sterlitech CF-042A FO membrane Cell.

3 Results and discussion

3.1

3.1 Characterization of the synthesized membranes

3.1.1

3.1.1 FTIR analysis

The FTIR bands were employed to provide information around the chemical configuration of the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes. FTIR presents the band of the membranes from 400 to 4000 cm⁻¹, Fig. 3. As shown in Fig. 3, the FTIR bands of the neat CTA/CA membrane indicate the presence of broadband at 3182–3700 cm⁻¹ associated with the O–H stretching region for the original CTA blend membrane. It is owing to the overlapping of the spectra associated with the O–H groups in cellulose esters with the spectra's associated with O–H stretching vibrations of the sorbet water molecules. The peaks around 2923 and 2877 cm⁻¹ related to elongating vibrations of the C-H band and CH₂ groups. The band showed at 1739 cm⁻¹ depicted the strong band for C = O of carbonyls which consigned to the stretching of the carbonyl group of CA, and the band appears at 1433 cm⁻¹ indicated twisting of C–H subsequently peaks at 1369 and 1224 cm⁻¹ which pronounced wagging and rocking style of C–H bond. The highest band around 1035 cm⁻¹ indicated to the ether functional group, while the peak at 904 cm⁻¹ is linked with the B (1–4) linkages between the glucose monomers and 1112 cm⁻¹, demonstrated the existence of saccharide (Liu et al. 2020). The FTIR band of the AA/CTA/CA membrane was presented in Fig. 3. The FTIR peaks demonstrate that the presence of the broadband at 3262–3700 cm⁻¹ associated with the broadening of O-H groups that intensity decreased after the incorporation of AA to the CTA/CA Polymer matrix. The change in hydrogen-bonded (O–H) band peak shows a change in H–bonding from the AA dimers to the dioxane–AA complexes (Ong and Chung 2012). The decrease of the band relates to the replacement of the –OH groups for the acetyl groups in the cellulose structure (Thi To Nu et al., 2019) and formation of the Lewis acid: base complex between acetic acid and dioxane when adding AA into the casting solution (Ong and Chung 2012). The bands appear around 2947, 1732, 1369, 1225, and 1062 cm⁻¹ which related to stretching vibration of C–H band, C–O stretching vibration, C–O stretching of ester, C–O elongating of carboxylic acid, and C–O stretching of ether, respectively. Besides, the change in peak intensity around 890 cm⁻¹ was also observed for the dioxane/acetic acid mixtures because of the interaction of the oxygen in the C–O–C group of dioxane with the OH group of AA. The chemical characterization of the modified AA/CTA/CA mixture with 0.4 wt% GO NSs carried out using FTIR analysis, Fig. 3.

The FTIR of the neat CTA/CA, AA/CTA/CA, GO/AA/CTA/CA NC modified membranes.

No significant change was observed for FTIR spectrum of GO/AA/CTA/CA NC modified membrane compared to AA/CTA/CA membrane but change intensity stretching vibrations of oxygen reactive groups of GO NSs can be located at about 3205–3700 cm⁻¹ related to the hydroxyl group (O–H). The greater strength of the bands at 2860 and 2910 cm⁻¹ are related to the asymmetric and symmetric stretching of CH₃ groups, 1750 cm⁻¹ related to stretching vibration of the carbonyl group (C = O), 1354 cm⁻¹ stretching of the carboxyl group (C-OH), 1225 cm⁻¹ stretching of the epoxy group (C–O–C) and 1062 cm⁻¹ related to the carbonyl group (C–O) vibration of alkoxy group which can overlap the CA peaks with GO NSs. It was suggested that the H–bonding interaction between the acetyl groups of CTA and the reactive groups of the GO NSs might occur during the membrane fabrication. The variation in band intensity and band shifting were apparent in the spectra of the FTIR when GO was combined into the membrane because of the connections between GO NSs with the AA/CTA/CA polymer matrix. It might be other possible reason for this observation can be no chemical variations in the membrane surface with the incorporation of GO NSs and difficulty of GO NSs detection distributed within the CA matrix.

3.1.2

3.1.2 Morphology analysis of membranes

All the detailed morphologies with higher magnification showed the existence of smooth surfaces with sheet like structures and uniform size of GO. The thickness of GO nanosheets was ranged from 34 to 56 nm, Fig. 4a. A scanning electron microscope (SEM) examination was performed to explore the morphology and dispersion of GO NSs within the composite membranes. The SEM descriptions of the surface and cross-section of the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes were shown in Figs. 4 and 5, respectively. The SEM descriptions in Fig. 4b and c showed the surface structure of neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membrane, respectively. The neat CTA/CA membrane cast from the dioxane/acetone solvent with relatively displays a highly smooth film surface with dense top layers without visible pores, Fig. 4b. Similarly, the membrane cast from dioxane/acetone/acetic acid has a relatively dense top layer as observed from the SEM image, Fig. 4c. SEM descriptions of GO/AA/CTA/CA NC modified membranes showed surfaces were also homogeneous and smooth, Fig. 4d. However, smoothness detected in the SEM pictures is a sign of the fact that the GO NSs were uniformly distributed in the AA/CTA/CA polymer matrix with high homogeneity without visible points of aggregation. Accordingly, the GO NSs oxygenated groups might be interacting with the AA/CTA/CA polymer chains providing good compatibility among the GO NSs and the polymer matrix and perfect distribution of GO NSs (de Moraes et al., 2015). Moreover, the existence of agglomerates has not been noticed. Furthermore, the GO NSs were completely inserted within the AA/CTA/CA polymeric matrix, representing a good adhesion among GO NSs and the polymer chains.

SEM images of Top surface for (a) neat CTA/CA, (b) AA/CTA/CA and (c) GO/AA/CTA/CA NC modified membranes, respectively.

The SEM images of the cross-sections for (a) neat CTA/CA, (b) AA/CTA/CA and (c) GO/AA/CTA/CA NC modified membranes, respectively.

Cross-section images revealed that all samples of the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC membranes had asymmetric structure containing a dense top layer and a sub-layer with macro-void pores. In Fig. 5a, the neat CTA/CA membrane has a typical three-layered sandwich structure: two relatively dense skin layers and a mid-layer with a small number of macro-voids. The cross-section view of the AA/CTA/CA membrane is usually asymmetric and consisted of two distinct layers. Where the active dense layer and the macro-voids containing porous layer, and the sub-layer has a significantly more open-cell and porous structure compared with the neat CTA/CA membrane, Fig. 5b. The membrane cast from acetone/dioxane/acetic acid has a relatively compact top layer and a relatively less dense bottom layer that detected from the SEM images. The acetic acid acts as a pore-forming agent in the AA/CTA/CA membrane Fig. 5b.

Fig. 5c, shows a higher magnification of GO/AA/CTA/CA NC membranes which leads to an improvements in the membrane surface morphology due to a homogeneous dispersion of the GO NSs in the polymer matrix, which enhanced the hydrophilicity of the membrane. The outcomes showed that significant alterations took place with the combination of 0.4 wt% of GO NSs. When GO NSs incorporated with AA/CTA/CA polymer matrix led to increasing macro-voids of pores in addition to sponge-like pores in the sub-layer, Fig. 5c. The development of large-area cavities distributed all over the surface of the sub-layer (Ionita et al., 2016). The existing of –COOH,–OH, and epoxy reactive groups owing to the incorporation of GO NSs that improved the membrane superficial morphology.

In general, the improvement in the membrane surface morphology due to a homogeneous spreading of the GO NSs addicted to the polymer content, which enhanced the membrane hydrophilicity and the rejection percentage. This outcome is inconsistent with the porosity and contact angle that mentioned in Fig. 6.

The Water contact angle and porosity measured of the neat CTA/CA, AA/CTA/CA, GO/AA/CTA/CA NC modified membranes surfaces.

3.1.3

3.1.3 Contact angle and surface porosity of FO membranes.

The contact angles and porosity of the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes revealed in Fig. 6. In Fig. 6, the average contact angles for the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes were 71^°± 2, 63^°± 2.5, and 49^°± 1.8, respectively, demonstrating they have hydrophilic surfaces. It observed that with 5.25 wt% of AA embedded into the membrane matrix, the contact angle of the AA/CTA/CA membrane decreased to 63^°± 2.5 where reveals that AA can improve the hydrophilic nature of CTA/CA membrane. This decrease in the membrane hydrophilicity via combination of AA mainly owing to it acts as a pore-creating agent. Besides, further reduction of the contact angle to 49^°± 1.8 observed as the AA/CTA/CA polymer matrix incorporate with hydrophilic GO NSs. These results matched with the previous reported about hydrophilicity improvement of polysulfone and CA films with the incorporation of GO NSs (Ghaseminezhad et al., 2019). Generally, the high hydrophilicity of the GO/AA/CTA/CA NC modified membrane results from introducing the hydroxyl, carboxylate, and epoxy moieties of the GO NSs. Therefore, when GO was introduced to the membrane hydrophilic substrates tend to have better FO flux performance. The enhanced hydrophilicity of the GO NSs modified AA/CTA/CA membrane may be owing to a superior attraction of H₂O molecules via the GO NSs, and the existence of reactive hydrophilic effective groups onto the matrix of the membranes. The contact angle decreased upon adding the GO NSs producing an improvement in the membrane surface energy. The improvement in membrane surface energy permits water to easy extent onto the membrane surface and improves the capability of the hydrophilic pores to absorb water via capillary possessions.

The porosity of the CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes presented about 47 ± 2.5%, 49 ± 3%, and 62 ± 2.8%, respectively. Consider the information showed in Fig. 6, an increment in the membrane porosity was detected for AA/CTA/CA and GO/AA/CTA/CA NC modified FO membranes compared to the neat CTA/CA membrane. The existence of the functionalized AA as a pore-improver in the casting matrix with the OH and COOH functional group could cause the creation of an AA/CTA/CA membrane with more porosity. It is clear that the porosity of the GO/MACTA/CA NC modified membrane showed higher increment owing to the presence of GO NSs in the casting solution; more pores created on the membrane surface. These OH and COOH reactive groups can influence the hydrophilic possessions of the polymer casting solution and alter the mechanism of membrane creation via growing the rate of the solvent/non-solvent interchange via the stage inversion procedure (Mansourpanah et al., 2011). The improved hydrophilicity when AA as a pore former and GO NSs in the casting solution because of a great attraction of water molecules by the AA and GO NSs and to the occurrence of active OH and COOH hydrophilic reactive groups on the membrane active layer (AL). The increase in the overall porosity of the AA/CTA/CA and GO/AA/CTA/CA NC modified membranes are the main factor beneficial for the smaller structure stricture where leads to improving FO NC modified membranes performance. It can conclude that the more porosity outcomes in the higher permeability and water flux.

3.1.4

3.1.4 Mechanical properties

The mechanical examination is the most significant properties to evaluate the load-bearing proficiency of a component. The mechanical possessions are necessary for the practical applications of the prepared porous membranes (Xing et al. 2019). To analyze the mechanical possessions of the prepared membranes, the typical stress–strain curves of the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes presented in Fig. 7, and the mechanical possessions contain tensile strength, maximum elongation, and the calculated young modulus are shown in Table 1. It is clear that the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes, present the elongation break in the range of 13.4%, 13.5%, and 12.1%, respectively. The membranes revealed tensile strength and Young modulus values in the following order: GO/AA/CTA/CA NCs > AA/CTA/CA > neat CTA/CA membranes. The mechanical properties of the CTA/CA membrane depend mainly on the amount of substitution and the mean molecular weight. Generally, a uniform structure of CTA/CA facilitates to attain more mechanical possessions. The AA/CTA/CA membrane demonstrates enhanced elongation break, tensile strength, and improved Young’s modulus values when compared to the neat CTA/CA membrane. This improvement in tensile strength values for the modified AA/CTA/CA membrane is mainly due to the cross-linked network structure in the alkyl group and COOH hydrophilic functional groups. The improvement in the mechanical stabilities of the GO/AA/CTA/CA NC modified membrane depends mainly on the rate dispersion of GO NSs into the AA/CTA/CA polymer matrix producing a durable interfacial load transfer from the GO NSs to the polymer via H-bonding interactions between GO NSs and the AA/CTA/CA polymer matrix. It was clear that the GO NSs embedding the AA/CTA/CA matrix resulted in improved tensile strength, Young modulus, and mechanical properties of the CTA/CA membrane. These results indicate that GO/AA/CTA/CA NCs membrane showed superior mechanical properties. Furthermore, the linkage bond between the GO NSs and AA into the CTA/CA polymer matrix generates elastic spacers on the nanocomposites membrane interface. It is responsible for the improvement of the stringency of membrane backbone, enhance the inter-molecular forces operating along the membrane chain, and improve the membrane crystallizing where the crystalline membrane is sturdier than the amorphous one. The enhancement of the mechanical strength and Young's modulus of the GO/AA/CTA/CA NC modified membranes is mainly due to the interaction between the carboxylate (COOH) group of AA and the OH, COOH, and epoxy groups on the GO NSs surfaces, were confirmed by FTIR results in Fig. 3, as well as due to the rigid aromatic structure in the membrane backbone. The addition of 0.4 wt% GO NSs has an optimistic effect on the mechanical possessions of the porous membranes might be owing to the uniform distribution of the GO NSs into the membrane matrix. Fundamentally, mechanical possessions are not only related to the penetrability of membranes but also outcomes from their pore structures that confirmed in Fig. 6.

The Stress-Strain curve of the neat CTA/CA, AA/CTA/CA, GO/AA/CTA/CA NC modified membranes.

Table 1 The mechanical properties and the calculated Young’s modulus of neat CTA/CA, AA/CTA/CA, GO/AA/CTA/CA NC modified membranes.

Membrane types	Stress	Strain	Young s modulus (MPa)	Maximum elongation (%)	Tensile strength (MPa)
Neat CTA/CA	29	3.2	911	13.4	53
AA/CTA/CA	28	2.8	1000	13.5	59.8
GO/AA/CTA/CA	39	2.7	1440	12.1	60.1

3.2

3.2 Performance evaluation of FO membranes

3.2.1

3.2.1 Effect of acetic acid concentration on the performance of fabricated membranes

To estimate the effect of acetic acid (AA), as a pore-forming additive, the CTA/CA membrane with a pore-forming agent was synthesized first, and FO performances of the resulting AA/CTA/CA membrane existed in Fig. 8a, and b. By keeping up the total dioxane/acetone percentage at 86 wt% as a solvent and 14 wt% (CTA/CA,1:2) and membrane thickness 200 µm. As a worldwide pore-former, AA incorporated into the CTA/CA casting solution and the effects of AA as a performing agent investigated with the contents of 1.75, 3.5, 5.25, 7 and 8.75 wt%, respectively using 1 M NaCl as a draw solution (DSs) and DI as feed solutions (FSs) in Fig. 8a. The influences of AA content in the creating polymer solution were investigated on water flux (WF, L/m².h) and the RSF (J_s, g/m².h), as revealed in Fig. 8a. The incorporation of the AA as a pore-former improves the water flux from 17.14 to 20 L/m².h with the AA content ranged from 1.75 to 5.25 wt%, demonstrating the pore-creating influence on the FO performance. It detected that the flux passes through a maximum at 5.25 wt% then further increase in AA content cause decrease in water flux to reach 17.38 L/m².h at 8.25 wt% AA content and RSF increase from 2 to 6.6 g/m².h using 1 M NaCl as DSs and DI as FSs, Fig. 8a. The membrane surfaces pore-blocking might be owing to the higher contents of AA, and the dense polymer mixture might take place and interrupt the polymer backbone filling. The development of water flux is owing to the higher porosity and hydrophilicity of membrane structure using AA as a pore-former and improve the number of pores onto membrane surfaces owing to the presenting of polar reactive (COOH and OH) groups to the membrane polymer chain (Shahlol et al., 2020). Additional increasing the AA concentration, beyond 5.25 wt%, causes a decline in water flux to reach 17.38 L/m².h, where penetrating AA content produces a thick membrane with little porosity. The decrease in water flux beyond 5.25 wt% it might be due to the strong hydrophilic-hydrophilic interactions among AA and CTA/CA polymer matrix via high hydrogen bonding, which might hinder the passing of H₂O molecules via the membrane. In the casting dope with the low AA content, CTA/CA molecules remain twisted and cannot accommodate the formation of sufficient network pores within the polymer molecules. Consequently, the sparse pores and the lowest operational surface area are typical for these membranes. The improvement in water flux might be due to the complex created via the COOH group of AA, 1,4-dioxane, and CTA/CA reactive functional groups, Fig. 9.

(a) The effect of different concentration of AA on permeate water flux and RSF of (CTA/CA) membrane using DI water as FS, 1 M NaCl as DS, AL-FS at 25◦C; (b) The effect of AA concentration on water flux and salt rejection using 0.1 M NaCl as FS, 1 M NH4Cl as DS, AL-FS at 25◦C, 1.6 L/min flow rate, with cells vertically oriented, and membrane thickness 200 µm.

The creation of the hydrogen bond (a) between dimer acetic acid; (b) between acetic acid and dioxane.

Fig. 8b shows the effect of AA content on membrane performances such as salt retention and water flux using 0.1 M NaCl as FS, 1 M NH₄Cl as DS, 1.6 L/min flow rate, AL-FS at 25 °C and membrane thickness 200 µm. With the increase of AA content, the membrane flux increased rapidly from 16.7 to 19.5 L/m².h when AA increase from 1.75 to 5.25 wt%, then further increase in AA content causes a decrease in flux to 16.9 L/m².h. The salt retention starts to decrease with increasing the AA concentration from 99.85 to 99.5%, where intense polymer matrix creates a thick membrane with small porosity. An increased outgrowth of the aggregates might lead to a more dense structure that, mostly using its thickness, leads to a flux decrease (Isawi 2019). From the point of opinion, the modified FO membranes revealed increased remarkably in both salt rejection and water flux related to the neat CTA/CA membrane. However, the water flux decreased obviously when the AA content further increased to 8.25 wt%, which might be explained by the incorporation of AA/4-dioxane complex with the polymer matrix. The constructions of carboxylic acid dimers and dioxane-carboxylic acid compounds are revealed in Fig. 9. To confirm the creation of the Lewis acid: base complex between AA and dioxane when adding AA into the casting solution. Therefore, based on the above results, the optimum contented of AA was 5.25 wt% for the CTA/CA membrane.

3.2.2

3.2.2 Effect of GO content on the performance of fabricated membranes

The effect of the GO NSs content on the performance of the AA/CTA/CA membrane represented in Fig. 10a and resulting GO/AA/CTA/CA nanocomposite (NC) modified membrane. Fig. 10a shows the FO performance of water flux and RSF for the GO/AA/CTA/CA hybrid membrane with various GO NSs concentrations tested by the FO test system using DI water as FS, 1 M NaCl as DS at 25^◦C, and membrane thickness 200 µm. Different contents of GO NSs combined onto AA/CTA/CA composite polymer matrix where these concentrations ranged from 0.1 to 0.5 wt% related to (CTA/CA) content (i.e., 14 wt%). Fig. 10a showed that the water flux of the GO/AA/CTA/CA NC modified membrane substantially increased from 26 to 33.6 L/m².h after the incorporation of a small quantity of GO from 0.1 to 0.4 wt%. It is identified that H₂O molecules could easily be drawn into the membrane surface and bulk by improving the hydrophilicity of the membrane. The GO/AA/CTA/CA NC modified membrane with higher GO NSs concentration over 0.4 wt%, the water flux of the membrane deceased as GO NSs concentration increased to reach 31 L/m².h using 0.5 wt% GO NSs content. The existence of GO NSs into the casting of the membrane develops water transport via the membrane matrix (Gao et al., 2014). Besides, the H₂O molecules can transport via the channels of GO NSs, and the hydrophilic reactive functional groups (epoxy, carboxyl, and hydroxyl) in GO NSs facilitate the absorption of H₂O molecules on the membrane reactive layer. These properties favor water passage via the membrane. Conversely, the decline in water flux could result from two reasons; firstly, the two dimensional GO NSs in the membrane matrix would act as a barrier to water flux. In general, water flux influenced by both the hydrophilicity and the hindrance effect to GO NSs. The hydrophilicity could be the dominant factor at relatively low GO NSs loadings, whereas it became a hindrance of GO NSs when the filling was more than 0.4 wt%. Secondly, the addition of excess GO increased the viscosity of the AA/CTA/CA polymer casting solution. It could restrain the creation of big pores, which was of ease for H₂O transportation (Wang et al., 2016). Fig. 10a reveals the RSF as a function of GO NSs content, where it decreased from 4.24 to 1.45 g/m².h by the increase of the GO NSs content from 0.1 to 0.5 wt%. When GO NSs loading increased, the decease trend of RSF became slow. The overall decrease trend mainly resulted from the hindrance effect of GO NSs in the membrane. The RSF of the hybrid membranes was very low, indicating a high salt retention. The GO/AA/CTA/CA NC modified membrane possessed both high water permeation and low RSF. The performance of fabricated membranes in terms of water flux and salt retention performed for the GO/AA/CTA/CA NC modified membrane with various GO NSs concentration using 0.1 M NaCl as the FS and 1 M NH₄Cl as the DSS. The water flux of the GO/AA/CTA/CA NCs modified membrane increased from 22.6 to 32.4 L/m².h with the increase of GO NSs from 0.1 to 0.4 wt%, Fig. 10b. The GO/AA/CTA/CA NCs modified membrane with higher GO NSs concentration over 0.4 wt%, the water flux of the membrane deceased as GO NSs content increased. The higher water flux of the modified membrane attributed to the increase in porosity, surface hydrophilicity, and surface certainly passes through waterways in the membrane structure that eventually leads to enhance water flux. To get better membrane performance, the content of GO NSs fixed at 0.4 wt% in the membrane fabrication experiments. At 0.5 wt% of GO NSs, the aggregation of GO NSs on the membrane active layer leads to the lower decline of water flux. We can see that GO NSs content caused a positive effect on salt rejection, Fig. 10b during the FO test using 0.1 M NaCl as the FS and 1 M NH₄Cl as the DS, the rejection increases from 99.7 to 99.88% using GO NSs concentration ranged from 0.1 to 0.5 wt%. In general, there was a slight decrease in the water flux with further GO NSs concentration were incorporated into the membrane, probably owing to the block of the passage of H₂O molecule and allowing a small amount of it to permeate. Furthermore, too much GO NSs in the membrane will aggregate, which will also affect the dispersion of H₂O molecules where tend to lower water flux and higher salt rejection. To get better membrane performance, the content of GO NSs fixed at 0.4 wt% in membrane preparation experiments. Table 2 summarizes the comparison study of FO performance among the present work and FO membranes indicated in literature under AL-FS. Obviously; the GO/AA/CTA/CA NCs modified membrane has obtained a higher water flux and lower reverse solute flux compared to the neat CTA/CA, AA/CTA/CA and the previous membranes. This result shows that GO NSs would be a potential nanomaterial to fabricate high performance FO membranes for water desalination.

(a) The effect of different concentration of GO NSs on permeate water flux and RSF of AA/CTA/CA membrane using DI water as FS, 1 M NaCl as DS, AL-FS at 25◦C; (b) The effect of different concentration of GO NSs on performances of AA/CTA/CA membrane using 0.1 M NaCl as FS and 1 M NH4Cl as DS, AL-FS at 25◦C, 1.6 L/min flow rate, with cells vertically oriented, and membrane thickness 200 µm.

Table 2 Comparison of FO performances between present work and GO-based FO membranes reported in literature.

FO membrane	FS	DS	Water flux AL-FS (L/m².h)	Reverse salt flux (g/m²·h)	REF
Neat CTA/CA	DI water	1 M NaCl	12.9	0.68	(this work)
AA/CTA/CA	DI water	1 M NaCl	20	3.1	(this work)
GO/AA/CTA/CA	DI water	1 M NaCl	33.6	1.5	(this work)
HTI-CTA	10 mM NaCl	0.5 M NaCl	9	5.3	(Wei et al. 2011)
CTA/dioxane/acetone/acetic Acid	DI water	2 M NaCl	9	6.2	(Ong and Chung 2012)
Free-standing cellulose triacetate (F-CTA)/graphene oxide (GO) membranes	DI water	1 M NaCl	18.3	3.5	(Wang et al. 2016)
GO incorporated TFC membranes	DI water	1 M NaCl	27.0	4.6	(Shen et al. 2016)
graphene laminate membrane (GLM).	DI water	1 M NaCl	10.5	6	(Rastgar et al. 2019)
PSF/GO	DI water	1 M NaCl	33.8	6.9	(Lim et al. 2017)
GO incorporated into the TFC membrane active layer	DI water	1 M NaCl	14.5	2.6	(Shokrgozar Eslah et al. 2018)
PSf–PDA/GO (ultrathin polydopamine/graphene oxide)	DI water	1 M NaCl	24.3	3.8	(Choi et al. 2019)

3.2.3

3.2.3 Effects of membrane orientation on the performance of fabricated membranes

FO mode of operation refers to the membrane orientation in which the support layer (SL) faces the DSs, and the active layer (AL) faces the FSs. The other face to FO mode is the pressure-related osmosis (PRO) mode of membrane orientation, where the SL faces the FSs, and the AL faces the DSs. The FO water flux (J_w) and RSF (J_s) of the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes evaluated in both AL-DS and AL-FS orientations using DI as the FSs and1 M NaCl as a DSs and flow rates 1.6 L/min, Fig. 11. Comparing the water fluxes in the FO mode and the PRO mode, the water flux in PRO mode for neat CTA/CA, AA/CTA/CA and GO/AA/CTA/CA NC modified membranes were about 17.6, 27.1 and 40.5 L/m².h higher than the fluxes in FO mode 12.9, 20 and 33.6 L/m².h respectively. The water flux in PRO style of setup observed to be relatively more than the FO style of setup for all three membranes. The water flux achieved in the AL-DS procedure was expressively higher related to the AL-FS membrane orientation owing to the absence of internal concentration polarization (ICP) when DI water used as FSs on the SL part of the prepared membrane. As showed in various previous studies, dilutive ICP expressively decreases the operational osmotic driving strength at the membrane SL and AL interface in AL-FS direction (Akther et al., 2020). The higher water flux obtained in PRO style than in the FO style of operation where the variance in water fluxes in the FO style and PRO style is due to the change in the membrane structural resistance relying on which sides of the solutions the SL of the synthesized membrane is oriented. In the FO style of operation, the DS faces the permeable SL of the prepared membrane, whereas the feed faces the AL (Phuntsho et al. 2013). The received water flux from the feed part dilutes the DS at the membrane’s surface and inside the membrane’s porous SL, decreasing the mesh osmotic pressure at the membrane’s surface and decreasing the water flux. Since the osmotic pressure of the DS at the membrane boundary layer is crucial to the FO process, dilutive ICP severely affects water flux in the FO process. When DI utilized as the FS in the FO style, concentrative external concentration polarization (ECP) is absent on the AL side of the membrane. When the process operated in PRO mode, the ICP occurrence reverses. Since the DS in PRO style faces the membrane AL, the occurrence is dilutive ECP, which can be mitigated via the cross-flow shear delivered on the membrane superficial. Although concentrative ICP occurs on the feed part of the membrane, its effect is less severe than dilutive ICP, and this is the reason why water flux in PRO mode is higher than in the FO mode. Although PRO mode gives a higher water flux in the laboratory scale test under controlled conditions, it obvious that the PRO mode is not suitable for desalination because the membrane is disposed to severe fouling as the porous SL viewing to the FS containing scaling and fouling species (Phuntsho et al. 2013). The RSF (Js) of the fabricated membranes evaluated as following the neat CTA/CA membrane showed a Js of 0.68 and 3.1 g/m².h in AL-FS and AL-DS styles, respectively, the AA/CTA/CA membrane showed a Js of 3.1 and 5.5 g/m².h in AL-DS mode and AL-DS mode, respectively, and the GO/AA/CTA/CA NC modified membrane showed a Js of 1.5 and 5.5 g/m².h in AL-FS and AL-DS modes, respectively. In the PRO style of operation, the draw solute concentricity at the membrane surface is higher than it is in FO mode operation, which generates a higher driving strength and higher water flux. Conversely, the occurrence of higher draw solute concentricity at the membrane surface in PRO style also leads to greater RSF with the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes, Fig. 11. The results indicate that greater Js occurred through the PRO style compared to the FO style, which is in covenant with the literature (Lambrechts and Sheldon 2019).

Effect FO membrane orientations on comparison between water fluxes and revers solute fluxes of neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes; 1 M NaCl on AL-FS mode and 1 M NaCl on SL (AL-DS mode). 1.6 L/min flow rate, with cells vertically oriented.

3.2.4

3.2.4 The effect of different concentration of NaCl as a draw solution

The FO performance of the fabricated neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes were evaluated in form of water flux and RSF experiments were performed using DI as FSs and different concentration various from 0.5 and 3 M of NaCl as DSs, flow rate 1.6 L/min, operation temperature 25 °C and the membrane operational area of 42 cm², Fig. 12. Fig. 12, show the water flux and the RSF of neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes as a function of the DSs concentration (NaCl). The water flux and RSF increased gradually with increasing content from 0.5 to 3 M. The water fluxes increase to 30.5, 32.4 and 41 L/m².h, and the RSF increased to 1.04, 6.3 and 2.2 g/m².h for neat CTA/CA, AA/CTA/CA and GO/AA/CTA/CA NC modified membranes, respectively at a constant operating conditions. As can be shown, the water flux increases for each membrane with increasing DSs concentration due to increased osmotic driving force. The reason is that the net driving strength via the membrane enhanced for both water and NaCl at a high content of DSs (Lia et al., 2018). Furthermore, the higher water flux was consistently observed in the GO/AA/CTA/CA NC modified membrane compared with the neat CTA/CA and AA/CTA/CA membranes. As can be gotten from Fig. 12 both AA/CTA/CA and GO/AA/CTA/CA NC modified membranes experienced slightly higher RSF than the neat CTA/CA membrane at different DSs concentration. Approving with Fig. 12, RSF increased as DS concentration increased due to the salt concentricity gradient growth over the membrane sides that provide a higher driving force for the migration of NaCl ions from the DS side to the FS side. A similar observation made in previous studies (Sahebi et al., 2020).

The effect of different NaCl concentration as a DSs on water flux and revers solute flux of neat CTA/CA, AA/CTA/CA and GO/AA/CTA/CA NC modified membranes; DI water as FSs on AL- FS mode.

3.2.5

3.2.5 The influence of different fertilizer draws solution on the performance of FO process

This study examined the sustainable source of freshwater from desalinating saline water utilizing fertilizer drawn forward osmosis (FDFO) technique via osmotic dilution of marketable nutrient solution. Using the FDFO procedure for the desalination of saline H₂O is an effective technique to be applicable in agriculture without applying any hydraulic pressure. The diluted fertilizers can be applied straightway without any further treatment as a source of nutrient (NPK) for irrigation. FDFO is a cost-operative technology for saline water desalination. Four different types of fertilizer draw solution (FDS) include KCl, NH₄Cl as monovalent ions, (NH₄)₂SO₄ as a monovalent cation and divalent anion, dipotassium hydrogen orthophosphate (K₂HPO₄) as monovalent cation trivalent anion and 1 M from each different fertilizer DSs used at flow rate 1.6 L/min, temperature 25 °C and the membrane effective area of 42 cm². Draw solutions synthesized by dissolving fertilizer powder in the DI water. Details information of the selected fertilizers showed in Table 3. Fig. 13a shows the performance of the four different FDS in terms of water flux and RSF using neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes and DI as a FSs. The results show that the KCl, NH₄Cl, and (NH₄)₂SO₄ have the highest water flux (12.6 L/m².h) while K₂HPO₄ has the lowest value (11.9 L/m².h) using the neat CTA/CA. While in the case of the AA/CTA/CA modified membrane, the KCl and NH₄Cl showed the highest water flux (19.8 L/m².h) compared to the (NH₄)₂SO₄ and K₂HPO₄ FDS which have 19.5 and 17.9 L/m².h, respectively. The GO/AA/CTA/CA NC modified membrane showed the highest water flux using KCl and NH₄Cl (33.5 L/m².h) followed by (NH₄)₂SO₄ (31.7 L/m².h) and K₂HPO₄ (28.5 L/m².h) using DI as a FSs. This outcomes showed that the GO/AA/CTA/CA NC modified membranes have a greater water flux compared to neat CTA/CA and AA/CTA/CA membranes using the selected various FDSs. The higher water flux using KCl and NH₄Cl than the (NH₄)₂SO₄ and K₂HPO₄ attributed to the higher diffusion coefficient of monovalent electrolytes as compared to divalent electrolytes and trivalent electrolyte (Corzo et al., 2017). Theoretically, as the osmotic pressure variance across the membrane is the main driving strength in the FO procedure, the water flux tendency between the fertilizers should follow the similar trend as the osmotic pressure. However, the result in Table 3 and Fig. 13b shows that there is no direct relation between the osmotic pressure of the FDS and the water flux and this is due to the concentration polarization (CP) effects of the high amount of ICP effects induced by the solute resistance inside the membrane SL facing the DS (Chekli et al., 2017).

Table 3 Chemical characteristics for the selected draw solutions.

Name of fertilizers	Chemical formula	MW	π at 1.0 M (atm)	π at 1 M calculated π = n. C.R.T (atm)
Ammonium chloride	NH₄Cl	53.5	43.5	49.26
Potassium chloride	KCl	74.6	44.0	49.26
Ammonium sulphate	(NH₄)₂SO₄	132.1	46.1	73.89
Di potassium hydrogen ortho phosphate	K₂HPO₄	174	36.5	73.89

The bulk osmotic pressure (π) value according to Phuntsho et al., 2012.

(a) Influence of the different types of fertilizer DS on the performances of FO process in terms of water flux and RSF using DI as FS and 1 M from each selected DSs; (b) Water flux and salt rejection using 0.1 M NaCl as FS and 1 M from each selected DSs as KCl, NH4Cl, (NH4)2SO4 and K2HPO4, (AL-FS), temperature 25 °C, flow rate 1.6 L/min, using neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes.

The performance of FDSs in terms of RSF varied extensively according to the type of fertilizers, also presented in Fig. 13a. The RSF can be a useful indicator of the extent of draw solute that can be absent during the FO process. The K₂HPO₄ and (NH₄)₂SO₄ showed the lowest RSF compared to the other selected fertilizers KCl and NH₄Cl. The results indicate that the fertilizer DSs containing monovalent elements have higher RSF than divalent and trivalent elements (NH₄)₂SO₄ and K₂HPO₄. The divalent and trivalent fertilizer draw solutions (NH₄)₂SO₄ and K₂HPO₄ have ionic species with a hydrated diameter of SO₄ and PO₄ relatively much higher than the hydrated diameter of the monovalent fertilizer species (NH₄, Cl and K) and then might be unique of the aims for lower RSF, Table 3. Furthermore, the RSF might be particularly important when phosphorus and nitrogen include DSs utilized as these composites identified to cause eutrophication in the getting water environment. Consequently, it is essential to calculate the performance of fertilizer DSs in terms of RSF. All of the prepared membranes verified were oriented to the AL faced the FS however, a greater water flux attained when the DS is onto the AL side, as ICP is low severe, operative with the active part facing the feed is recommended when the FS has a high fouling potential (Corzo et al., 2017). This trend clarified via the fact that an initial higher water flux level can mostly be coupled with an elevated level of RSF result in more severe fouling. Besides, both KCl and NH₄Cl have ionic species with the small hydrated diameter (i.e., K, Cl, and NH₄) that will readily diffuse through the membrane compared to fertilizers having larger-sized hydrated anions (i.e., SO₄, and PO₄) regardless of the paired cations. The RSF in the FDS FO process has also evaluated in terms of loss of essential nutrients (i.e., N, P, and K) per unit volume of water extracted from the FS as substantive in (Phuntsho et al., 2012). Results in Fig. 13a showed that NH₄Cl, KCl, and had the highest loss of nutrient which correlates with the RSF data for these fertilizers. The (NH₄)₂SO₄ and KH₂PO₄ exhibited the lowest loss of nutrients by reverse diffusion for N, P, and K, respectively. Actually, these fertilizers have divalent ions (i.e. SO₄, PO₄) which show significantly lower loss via RSF as a result of their larger hydrated ions, Table 3.

Fig. 13b showed the water flux and salt rejection of neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membrane using 1 M concentration of each selected FDSs and 0.1 M NaCl as FSs. The membrane has water flux values in the following order: KCl = NH₄Cl (12.6 L/m².h)> (NH₄)₂SO₄ (12.4 L/m².h) > K₂HPO₄ (11.9 L/m².h) using neat CTA/CA membrane. While in the case of AA/CTA/CA membrane the NH₄Cl and KCl (19.1 L/m².h) has the highest water flux followed by (NH₄)₂SO₄ (18.6 L/m².h) and K₂HPO₄ (16.7 L/m².h). The GO/AA/CTA/CA NC modified membrane showed the highest water flux using KCl and NH₄Cl (33.1 L/m².h) and followed by (NH₄)₂SO₄ (31 L/m².h) and K₂HPO₄ (28.1 L/m².h). The water flux varied slightly when 0.1 M NaCl was used as feed Fig. 13b. This difference in water flux is due to the coupled influences of dilutive ICP on the DS and concentrative ECP on the FS. Furthermore, the dilutive and concentrative CP moduli are lower when low DS concentrations are utilized because of the comparatively lower flux achieved (Phuntsho et al., 2013).

Fig. 13b showed the salt rejection for each FDS as a function using neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes. K₂HPO₄ has the highest salt rejection compared to NH₄Cl, (NH₄)₂SO₄, and KCl using all membranes. These results might be owing to the importance of FDS molecular weight that plays an essential role in the FO process. Smaller fertilizers draw molecules, KCl and NH₄Cl exhibit identical separation styles than (NH₄)₂SO₄ and K₂HPO₄. The (NH₄)₂SO₄ and K₂HPO₄ were rejected higher than the KCl and NH₄Cl. The molecular size and the chemical origin of natural inorganic matter are valuable tools for evaluating their effects on the performances of FO desalination organizations. The FO separation efficiency depends on the charges on the membrane surfaces and the natural inorganic matter that plays an important role in the rejection process. The (NH₄)₂SO₄ and K₂HPO₄ have the same trend for FO rejection due to the higher molecular weight of their chemical structure where the divalent and trivalent fertilizer draws solutions (NH₄)₂SO₄ and K₂HPO₄) have higher retention than the monovalent one. The results indicate that the K₂HPO₄ fertilizer comprises the important groups of the more concentrated soluble fertilizer materials comprising both k and P elements. These essential fertilizer components wanted by the vegetation and can be either alone as fertilizer or is favorable to be used as mixed fertilizer with several other fertilizers. Even though their water flux is lower than the other DSs however their low RSF creates them a promising candidate for FO desalination for fertigation using GO/AA/CTA/CA NC modified membrane.

3.3

3.3 Effect of time on activity of synthesized FO membranes

In the membrane procedure, fouling leads to a drop in the water flux and reinforces the conveyance of solutes via the membrane due to an increase in the diffusion and convection factors of the membrane surface which lead to increases energy consumption and results in high operating cost (Im et al., 2020). Fig. 14 shows the mean permeate flux of FO operation using neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes with the AL facing the FS mode using DI water as the FS and 1 M NaCl as the DS at flow rate 1.6 L/min as a function of time (21 h). The water flux decreased gradually with contact time from 14.2 to 10.7 L/m².h for neat CTA/CA and from 21.4, 34.3 L/m².h to 14.3, 20.5 L/m².h for AA/CTA/CA and GO/AA/CTA/CA NC modified membranes, respectively, during the operational period (it notes that no physical/chemical cleaning applied during the operational cycles). It found that the water flux for the GO/AA/CTA/CA NC modified membrane was higher than the water fluxes of neat CTA/CA, and AA/CTA/CA membranes. The water flux kept declining as a result of the dilution of DS and concentration of FS. So, the flux decrease in the fouling examinations is produced not only by membrane fouling but also by the decline in osmotic driving strength (Mi and Elimelech 2008). The fouling reversibility of FO ascribed to the less compact organic fouling layer created in the FO style because of the deficiency of hydraulic pressure or the back diffusion of ions from the DS to the membrane surface on the feed sideway. As described related observation was made by the same author in the gypsum scaling tests (Mi and Elimelech 2010). As a universal pore-improver, AA incorporated into the casting solution and it observed that the CTA/CA membrane exhibited an increasing water flux from 10.7 to 14.3 L/m².h and an increase in antifouling properties. The enhancement of the fouling resistance of GO/AA/CTA/CA NC modified membrane is mainly because of the advanced membrane hydrophilicity in addition to the higher surface negative charges after incorporation of GO NSs into the membrane matrix. The GO/AA/CTA/CA NC modified membrane has high fouling resistance with final water flux 20.5 L/m².h, about 59.1% from its initial value when compared to the neat CTA/CA and AA/CTA/CA membranes. The neat CTA/CA and AA/CTA/CA membranes have final water flux 10.7 and 14.3 L/m².h, about 74.6% and 66.4% from their initial values, respectively. It indicated that the GO NSs embedded membrane exhibits an enhancement in AA/CTA/CA membrane hydrophilicity and may hinder the solute deposition and adsorption on the membrane matrix. After introducing the AA as a pore-improver and incorporation of hydrophilic GO NSs to the CTA/CA membrane, thus the membrane became more hydrophilic so, the membrane fouling was delayed owing to the improvement of the membrane surfaces with the hydrophilic AA and GO NSs.

Permeate flux versus filtration time for neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes. 1.6 L/min flow rate, with cells vertically oriented, and AL-FS using DI water as the FS and 1 M NaCl as the DS.

3.4

3.4 Application of FO synthetic membranes in a natural saline water sample

The productions of freshwater from saline water are one of the extremely substantial challenges facing Egypt at present, as Egypt does not only face a water scarcity problem but also an inevitable energy crisis. The neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes were selected for desalination of a natural saline water sample collected from the Mersa Matruh area, North-Western Coast of Egypt and the selected four types of fertilizer as the DS. The Mersa Matruh area has a dry climate where the freshwater is insufficient to meet the probable increases in water demands. Mersa Matruh region deliberated to be one of the greatest promising localities for land reclamation in Egypt. Mersa Matruh is located in the southern portion of the Mediterranean coastal semi-arid area and has been exposed to an intensive estimation by governmental authorities for the establishing of the new communities, tourist societies, and land reclamation developments. The sample properties, prior FO process, are showed in Table 4. The FS contains a natural saline water sample with the total dissolved solids 12760 mg/L and 1 M from different types of FDSs include KCl, NH₄Cl, (NH₄)₂SO₄, and K₂HPO₄ as DSs, flow rate 1.6 L/min, at 25 °C and the membrane operation area as 42 cm² with FS-AL mode. The FS reservoirs placed on weighing balances until the stability of the salt rejection and water flux has reached. An evaluation study was evaluated in Fig. 15, where the water flux and salt rejection of the selected natural saline water sample collected from the Mersa Matruh area as FS. The amount of measured water flux of the neat CTA/CA membrane showed that KCl and NH₄Cl have the highest water flux (9.1 L/m².h) followed by (NH₄)₂SO₄ (8.8 L/m².h) and K₂HPO₄ (7.1 L/m^2.h). In case of the AA/CTA/CA membrane, the water flux showed the highest value as in order KCl and NH₄Cl (15.2 L/m².h)> (NH₄)₂SO₄ (15 L/m².h) > K₂HPO₄ (13.1 L/m².h) followed the same trend in the case of the GO/AA/CTA/CA NC modified membranes that displays the highest water flux as in order KCl and NH₄Cl (20.5 L/m².h)> (NH₄)₂SO₄ (20.2 L/m².h) > K₂HPO₄ (18.1 L/m².h), Fig. 15. Flux in an FO experiment can be influenced by the concentration of DS and FSs, as well as by the overall osmotic pressure variance (Δπ) among the DS and the FS and fouling of membrane (Liden et al., 2019). The primary driving force for flux in FO can consider as a variance in osmotic pressure between the FS and DS. Therefore, the osmotic pressure variance between the 0.1 M NaCl and the DS was lower than the saline water sample as an FS and DS, as represented in Fig. 15. Owing to the combined high TDS altitudes and the organics, the activity or “effective concentration” of constituents in the natural saline water sample was higher than that for the salt copy. Moreover, the natural saline water sample contains a large number of ions with high TDS shown in Table 4. The membrane fouling includes both organic fouling and inorganic fouling (i.e., scaling) anticipated being a key reason for the severe flux decrease shown in Fig. 15. In this work, the FO rejections (R %) of the FS were evaluated by taking the DS sample at the finale of every test and analyzing it for Na⁺ ions Fig. 15 and Table 5. All desalted permeated water samples collected for chemical analyses after one hour. Fig. 15, showed the salt retention for each fertilizer DS as a function using neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes. K₂HPO₄ has the highest salt rejection compared to NH₄Cl, (NH₄)₂SO₄, and KCl for all membranes. These results might be owing to the impact of fertilizer DS molecular weight that plays an essential role in the FO process. Smaller fertilizers draw molecules, KCl and NH₄Cl exhibit identical separation styles than (NH₄)₂SO₄ and K₂HPO₄ where (NH₄)₂SO₄ and K₂HPO₄ were rejected higher than the KCl and NH₄Cl. The full chemical analyses of the feed and the product (permeate) water samples were revealed in Table 4. When using NH₄Cl as DS, it is obvious that the concentricity's of the cations and anions decrease after the desalination procedure in Table 4. The rejection percent (R%) has measured using the saline water sample as FS, Table 5. The GO/AA/CTA/CA NC modified membranes shows higher water flux compared to neat CTA/CA, and AA/CTA/CA membranes. The highest water flux of GO/AA/CTA/CA NC modified membranes achieved due to the OH, COOH, and epoxy functionalized modified membrane. This high performance anticipated because the H₂O molecules could be simply drawn into the membrane surface and bulk by enhancing the hydrophilicity of the membrane. When DI water was used as the FS, water flux in FO style is higher than 0.1 M NaCl, and the saline water sample used as the FS under the same flow rate (1.6 L/min) and DS (1 M) concentration Table 6. These outcomes propose that fouling plays a more pronounced role in the flux pattern when a real saline water sample used. Both external concentration polarization (ECP) and internal concentration polarization (ICP) also vary with the changing DS/FS concentrations. High FS concentricity also contributed to the increase in ECP that directly caused the high reduction of the resultant water flux (Majeed et al. 2015). This result illustrated that the concentration polarization (CP) affects the DS and FS concentrations at the membrane AL and causes a lower actual flux in the FO, as revealed in Table 6. ICP and ECP, along with the dilution and concentration of DS/FS, participated to the reduced available osmotic pressure across the membrane’s AL, which resulted in a lower flux outcome compared with the high-theoretical flux potential.

Table 4 Characteristics of a natural saline water sample selected from Mersa Matruh area, before and after FO desalination using Neat CTA/CA, AA/CTA/CA and GO/AA/CTA/CA NC modified membranes and NH₄Cl as a draw solution.

Ion	Concentrations (mg/L) of brine water before FO system.	Concentrations (mg/L) of the brine wastewater effluent was collected from Mersa Matruh area, after FO system.
Ion	Concentrations (mg/L) of brine water before FO system.	CTA/CA	AA/CTA/CA	GO/AA/CTA/CA
TDS	12,760	12,880	13,003	12,892
Ca⁺²	296	297	299	298
Mg⁺²	708	709	711	710
Na⁺	3366	3465	3564	3465
K ⁺	107	108	109	108
CO₃^{– 2}	120	120	120	120
HCO₃^–	134	134	134	134
SO₄^-2	1049	1050	1052	1051
Cl^-	7046	7066	7085	7076

Influence of the different types of DSs on the performances of FO process in terms of water flux and salt rejection using a natural saline water sample collected from Mersa Matruh desalination plant North-Western Coast of Egypt, using different draw solutions as KCl, NH4Cl, (NH4)2SO4 and K2HPO4 at 25 °C, flow rate1.6 L/min, FO mode: with cells vertically oriented, and (AL-FS), using Neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes.

Table 5 The revers ions flux and salt rejection of saline water sample selected from Mersa Matruh area, before and after FO desalination using Neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes using NH₄Cl as a DSs.

Ions	Saline water sample (mg/L) before	The revers ions Flux (g/m².h)			R%
Ions	Saline water sample (mg/L) before	Neat CTA/CA	AA/CTA/CA	GO/AA/CTA/CA	Neat CTA/CA	AA/CTA/CA	AA/GO/CTA/CA
Ca⁺²	296	6	30	6	99.82	99.10	99.82
Mg⁺²	708	2.8	3.9	2.9	99.05	98.68	99.02
Na⁺	3366	6	30	6	99.82	99.10	99.82
K⁺	107	4	6	5	98.64	97.97	98.31
CO₃^-2	120	3	6	5	99.57	99.15	99.29
HCO₃^–	134	10	20	16	99.85	99.71	99.77
SO₄^-2	1049	1	3	2	99.06	97.19	98.13
Cl^-	7046	22	25	24	97.90	97.61	97.71

Table 6 The water flux of Neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes using different fertilizer as DSs and DI water, 0.1 M NaCl, and the saline water sample selected from Mersa Matruh area, respectively as a FSs, after FO desalination.

Membrane types	Neat CTA/CA	AA/CTA/CA	GO/AA/CTA/CA
Draw solution types	Water Flux (L/m².h)
NH₄Cl	Feed (DI)	12.6	19.8	33.6
	Feed (0.1 M NaCl)	12.6	19.1	33.2
	Feed (Saline)	9.1	15.2	20.5
KCl	Feed (DI)	12.6	19.8	33.6
	Feed (0.1 M NaCl)	12.6	19.1	33.1
	Feed (Saline)	9.1	15.2	20.5
(NH₄)₂SO₄	Feed (DI)	12.6	19.5	31.7
	Feed (0.1 M NaCl)	12.4	18.6	31
	Feed (saline)	8.8	15	20.2
K₂HPO₄	Feed (DI)	11.9	17.9	28.6
	Feed (0.1 M NaCl)	11.9	16.7	28.1
	Feed (saline)	7.1	13.1	18.1

It revealed that the GO/AA/CTA/CA NC modified membranes show higher results compared to neat CTA/CA, and AA/CTA/CA membranes for saline water desalination using the FO technique. Fertilizers were proposed as DS to extract water from natural saline water samples for direct irrigation. The diluted fertilizers can be applied directly without any further treatment as a source of nutrients for irrigation. Utilizing the FDFO for the desalination of the saline water sample is an efficient technique that reduces the amount of salt from the saline water without applying any hydraulic pressure. Thus, this is an ecologically friendly technique that saves energy and keeps the environs and lower desalination cost. FO is a promising alternative sustainable technology promoting saline water desalination. The FDFO technique is in not requiring a recovery step to re-concentrate the DS so, the alternative using the diluted DS as a supplement for irrigation H₂O by fertigation.

4 Conclusion

Membranes were successfully synthesized by blending CTA/CA polymer mixture in Dioxane/acetone used as a solvent, and deionized water utilized as the coagulant bath. Improvement of CTA/CA blends flat sheet membranes via incorporating various amounts of AA as a pore-forming agent and polymer improver ranging from1.75 to 8.75 wt% into the membrane matrix. The GO NSs concentration ranged from 0.1 to 0.5 wt%, related to (CTA/CA) concentration, (i.e., 14 wt%) and used as additives with AA/CTA/CA blend polymer matrix to fabricate highly permeable GO/AA/CTA/CA NC modified FO membrane. By incorporation of GO NSs into the AA/CTA/CA substrate, the hydrophilicity, overall porosity, mechanical strength also enhanced, and a further improvement in water flux significantly achieved compared to neat CTA/CA and AA/CTA/CA membranes. The average contact angles for the neat CTA/CA, AA/CTA/CA, and GO/AA/CTA/CA NC modified membranes were 71^°± 2, 63^°± 2.5, and 49^°± 1.8, and the porosity displayed about 47 ± 2.5%, 49 ± 3%, and 62 ± 2.8%, respectively. The GO/AA/CTA/CA NC modified membrane with lower content of GO NSs (0.4 wt%) showed higher water flux 33.6 L/m².h, salt rejection 99.88%, and lower RSF 1.45 g/m².h when compared to neat CTA/CA and AA/CTA/CA membranes using 0.1 M NaCl as FS and 1 M NH₄Cl as DS under the AL-FS mode. FDFO is a cost-operative technology for saline water desalination. The diluted fertilizer DS can be used to supply nutrients to harvest in place of extricating it from the desalinated water. The GO/AA/CTA/CA NC modified membrane showed the highest water flux using KCl and NH₄Cl (33.5 L/m².h) followed by (NH₄)₂SO₄ (31.7 L/m².h) and K₂HPO₄ (28.5 L/m².h) using DI as a FSs. The results reveal that the water flux of GO/AA/CTA/CA NC modified membrane was 20.5, 20.5, 20.2 and 18.1 L/m².h using KCl, NH₄Cl, (NH₄)₂SO₄ and K₂HPO₄ as DS, respectively using the natural saline water with TDS (12760 mg/L) and PH (8.5) collected from Mersa Matruh area, North-Western Coast of Egypt as FS. The results indicate that the fertilizer DSs containing monovalent elements have higher RSF than divalent elements ((NH₄)₂SO₄ and trivalent elements K₂HPO₄). The reusability test of the synthesized GO/AA/CTA/CA NC modified membrane showed good sustainability during the 1260 min continuous test and it displayed a great potential to be interested in saline water desalination and enhanced water source sustainability to use in agriculture. Using the FDFO procedure for the desalination of saline water is an effective technique to be applicable in agriculture without applying any hydraulic pressure and it is an ecologically friendly technique that saves energy and lowers desalination costs. The diluted fertilizers can be applied in a straight line without any further treatment as a source of nutrient (N.P.K.) for irrigation. In conclusion, the GO/AA/CTA/CA NC modified flat sheet membrane is efficient for the FDFO process using different draw solution containing (N.P.K) and exhibit a great water flux, low RSF, and high rejection.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

Isawi H., . Evaluating the performance of different nano-enhanced ultrafiltration membranes for the removal of organic pollutants from wastewater. J. Water Process Eng.. 2019;31:100833
[Google Scholar]
Isawi H., El-Sayed M.H., Feng X., Shawky H., Abdel Mottaleb M.S., . Surface nanostructuring of thin film composite membranes via grafting polymerization and incorporation of ZnO nanoparticles. Appl. Surf. Sci.. 2016;385:268-281.
[Google Scholar]
D.F. Ahmed, H. Isawi, N.A. Badway, A. Elbayaa, H. Shawky, Highly porous cellulosic nanocomposite membranes with enhanced performance for forward osmosis desalination, Under Review, Iranian Polymer Journal.
Qasim M., Darwish N.A., Sarp S., Hilal N., . Water desalination by forward (direct) osmosis phenomenon: A comprehensive review. Desalination. 2015;374:47-69.
[Google Scholar]
Xiao P., Li J., Ren Y., Wang X., . A comprehensive study of factors affecting fouling behavior in forward osmosis. Colloids Surf., A. 2016;499:163-172.
[Google Scholar]
Chung T.-S., Zhang S., Wang K.Y., Su J., Ling M.M., . Forward osmosis processes: yesterday, today and tomorrow. Desalination. 2012;287:78-81.
[Google Scholar]
Su J., Zhang S., Ling M.M., Chung T.-S., . Forward osmosis: an emerging technology for sustainable supply of clean water. Clean Technol. Environ. Policy. 2012;14(4):507-511.
[Google Scholar]
Shahlol O.M.A., Isawi H., El-Malky M.G., Al-Aassar A.-H., zwai A.E., . Performance evaluation of the different nano-enhanced polysulfone membranes via membrane distillation for produced water desalination in Sert Basin-Libya. Arabian J. Chem.. 2020;13(4):5118-5136.
[Google Scholar]
Ong R.C., Chung T.-S., . Fabrication and positron annihilation spectroscopy (PAS) characterization of cellulose triacetate membranes for forward osmosis. J. Membr. Sci.. 2012;394-395:230-240.
[Google Scholar]
Isawi H., . Using Zeolite/Polyvinyl alcohol/sodium alginate nanocomposite beads for removal of some heavy metals from wastewater. Arabian J. Chem.. 2020;13(6):5691-5716.
[Google Scholar]
Isawi H., . Development of thin-film composite membranes via radical grafting with methacrylic acid/ZnO doped TiO2 nanocomposites. React. Funct. Polym.. 2018;131:400-413.
[Google Scholar]
Tu Y., Lv M., Xiu P., Huynh T., Zhang M., Castelli M., Liu Z., Huang Q., Fan C., Fang H., . Destructive extraction of phospholipids from Escherichia coli membranes by graphene nanosheets. Nat. Nanotechnol.. 2013;8:594.
[Google Scholar]
Wang X., Wang X., Xiao P., Li J., Tian E., Zhao Y., Ren Y., . High water permeable free-standing cellulose triacetate/graphene oxide membrane with enhanced antibiofouling and mechanical properties for forward osmosis. Colloids Surf., A. 2016;508:327-335.
[Google Scholar]
Chee W.K., Lim H.N., Huang N.M., Harrison I., . Nanocomposites of graphene/polymers: a review. RSC Adv.. 2015;5(83):68014-68051.
[Google Scholar]
Kim S., Yoo J.-B., Yi G.-R., Lee Y., Choi H.R., Koo J.C., Oh J.-S., Nam J.-D., . An aggregation-mediated assembly of graphene oxide on amine-functionalized poly (glycidyl methacrylate) microspheres for core–shell structures with controlled electrical conductivity. J. Mater. Chem. C. 2014;2(32):6462-6466.
[Google Scholar]
He L.i., Dumée L.F., Feng C., Velleman L., Reis R., She F., Gao W., Kong L., . Promoted water transport across graphene oxide–poly (amide) thin film composite membranes and their antibacterial activity. Desalination. 2015;365:126-135.
[Google Scholar]
Liang JiaJie, Huang Y.i., Zhang F., Zhang Y.i., Li N., Chen YongSheng, . The use of graphene oxide membranes for the softening of hard water. Science China Technological Sciences. 2014;57(2):284-287.
[Google Scholar]
Yin J., Deng B., . Polymer-matrix nanocomposite membranes for water treatment. J. Membr. Sci.. 2015;479:256-275.
[Google Scholar]
Johnson D.J., Suwaileh W.A., Mohammed A.W., Hilal N., . Osmotic's potential: An overview of draw solutes for forward osmosis. Desalination. 2018;434:100-120.
[Google Scholar]
Panagopoulos A., Haralambous K.-J., Loizidou M., . Desalination brine disposal methods and treatment technologies-A review. Sci. Total Environ.. 2019;693:133545
[Google Scholar]
Suwaileh W., Johnson D., Jones D., Hilal N., . An integrated fertilizer driven forward osmosis-renewables powered membrane distillation system for brackish water desalination: A combined experimental and theoretical approach. Desalination. 2019;471:114126
[Google Scholar]
Majeed T., Sahebi S., Lotfi F., Kim J.E., Phuntsho S., Tijing L.D., Shon H.K., . Fertilizer-drawn forward osmosis for irrigation of tomatoes. Desalin. Water Treat.. 2015;53(10):2746-2759.
[Google Scholar]
Xing X.-Y., Gu L., Jin Y.u., Sun R., Xie M.-Y., Wu Q.-Y., . Fabrication and characterization of cellulose triacetate porous membranes by combined nonsolvent-thermally induced phase separation. Cellulose. 2019;26(6):3747-3762.
[Google Scholar]
Qasim M., Mohammed F., Aidan A., Darwish N.A., . Forward osmosis desalination using ferric sulfate draw solute. Desalination. 2017;423:12-20.
[Google Scholar]
Chen X., Xu J., Lu J., Shan B., Gao C., . Enhanced performance of cellulose triacetate membranes using binary mixed additives for forward osmosis desalination. Desalination. 2017;405:68-75.
[Google Scholar]
Liu X., Wu J., Hou L.-A., Wang J., . Performance and deterioration of forward osmosis membrane exposed to various dose of gamma-ray irradiation. Ann. Nucl. Energy. 2020;135:106950
[Google Scholar]
D. Thi To Nu, N. Phi Hung, C. Van Hoang, B. Van der Bruggen, Preparation of an Asymmetric Membrane from Sugarcane Bagasse Using DMSO as Green Solvent, Applied Sciences 9 (2019) 3347.
de Moraes Ana Carolina Mazarin, Andrade Patricia Fernanda, de Faria Andreia Fonseca, Simões Mateus Batista, Salomão Francisco Carlos Carneiro Soares, Barros Eduardo Bedê, Gonçalves Maria do Carmo, Alves Oswaldo Luiz, . Fabrication of transparent and ultraviolet shielding composite films based on graphene oxide and cellulose acetate. Carbohydr. Polym.. 2015;123:217-227.
[Google Scholar]
Ionita Mariana, Crica Livia Elena, Voicu Stefan Ioan, Pandele Andreea Madalina, Iovu Horia, . Fabrication of cellulose triacetate/graphene oxide porous membrane. Polym. Adv. Technol.. 2016;27(3):350-357.
[Google Scholar]
Ghaseminezhad S.M., Barikani M., Salehirad M., . Development of graphene oxide-cellulose acetate nanocomposite reverse osmosis membrane for seawater desalination. Compos. B Eng.. 2019;161:320-327.
[Google Scholar]
Mansourpanah Y., Madaeni S., Rahimpour A., Adeli M., Hashemi M., Moradian M., . Fabrication new PES-based mixed matrix nanocomposite membranes using polycaprolactone modified carbon nanotubes as the additive: property changes and morphological studies. Desalination. 2011;277:171-177.
[Google Scholar]
Gao Yong, Hu Meng, Mi Baoxia, . Membrane surface modification with TiO2–graphene oxide for enhanced photocatalytic performance. J. Membr. Sci.. 2014;455:349-356.
[Google Scholar]
Wei Jing, Qiu Changquan, Tang Chuyang Y., Wang Rong, Fane Anthony G., . Synthesis and characterization of flat-sheet thin film composite forward osmosis membranes. J. Membr. Sci.. 2011;372(1-2):292-302.
[Google Scholar]
Shen Liang, Xiong Shu, Wang Yan, . Graphene oxide incorporated thin-film composite membranes for forward osmosis applications. Chem. Eng. Sci.. 2016;143:194-205.
[Google Scholar]
Rastgar Masoud, Bozorg Ali, Shakeri Alireza, Sadrzadeh Mohtada, . substantially improved antifouling properties in electro-oxidative graphene laminate forward osmosis membrane. Chem. Eng. Res. Des.. 2019;141:413-424.
[Google Scholar]
Lim Sungil, Park Myoung Jun, Phuntsho Sherub, Tijing Leonard D., Nisola Grace M., Shim Wang-Geun, Chung Wook-Jin, Shon Ho Kyong, . Dual-layered nanocomposite substrate membrane based on polysulfone/graphene oxide for mitigating internal concentration polarization in forward osmosis. Polymer. 2017;110:36-48.
[Google Scholar]
Shokrgozar Eslah S., Shokrollahzadeh S., Moini Jazani O., Samimi A., . Forward osmosis water desalination: Fabrication of graphene oxide-polyamide/polysulfone thin-film nanocomposite membrane with high water flux and low reverse salt diffusion. Sep. Sci. Technol.. 2018;53:573-583.
[Google Scholar]
Choi Hyeon-gyu, Shah Aatif Ali, Nam Seung-Eun, Park You-In, Park Hosik, . Thin-film composite membranes comprising ultrathin hydrophilic polydopamine interlayer with graphene oxide for forward osmosis. Desalination. 2019;449:41-49.
[Google Scholar]
Akther N., Yuan Z., Chen Y., Lim S., Phuntsho S., Ghaffour N., Matsuyama H., Shon H., . Influence of graphene oxide lateral size on the properties and performances of forward osmosis membrane. Desalination. 2020;484:114421
[Google Scholar]
Phuntsho Sherub, Sahebi Soleyman, Majeed Tahir, Lotfi Fezeh, Kim Jung Eun, Shon Ho Kyong, . Assessing the major factors affecting the performances of forward osmosis and its implications on the desalination process. Chem. Eng. J.. 2013;231:484-496.
[Google Scholar]
Lambrechts R., Sheldon M.S., . Performance and energy consumption evaluation of a fertiliser drawn forward osmosis (FDFO) system for water recovery from brackish water. Desalination. 2019;456:64-73.
[Google Scholar]
Li Chunxia, Song Jianfeng, Dou Pengjia, Yin Yong, He Tao, He Weina, . Effects of low-toxicity solvent and binary mixed additives on the microstructure and performance of cellulose triacetate (CTA) membranes for forward osmosis. Desalin. Water Treat.. 2018;130:44-53.
[Google Scholar]
Sahebi Soleyman, Sheikhi Mohammad, Ramavandi Bahman, Ahmadi Mehdi, Zhao Shuaifei, Adeleye Adeyemi S., Shabani Zhara, Mohammadi Toraj, . Sustainable management of saline oily wastewater via forward osmosis using aquaporin membrane. Process Saf. Environ. Prot.. 2020;138:199-207.
[Google Scholar]
Corzo Beatriz, de la Torre Teresa, Sans Carmen, Ferrero Enrique, Malfeito Jorge J., . Evaluation of draw solutions and commercially available forward osmosis membrane modules for wastewater reclamation at pilot scale. Chem. Eng. J.. 2017;326:1-8.
[Google Scholar]
Chekli Laura, Kim Youngjin, Phuntsho Sherub, Li Sheng, Ghaffour Noreddine, Leiknes TorOve, Shon Ho Kyong, . Evaluation of fertilizer-drawn forward osmosis for sustainable agriculture and water reuse in arid regions. J. Environ. Manage.. 2017;187:137-145.
[Google Scholar]
Phuntsho Sherub, Shon Ho Kyong, Majeed Tahir, El Saliby Ibrahim, Vigneswaran Saravanamuthu, Kandasamy Jaya, Hong Seungkwan, Lee Sangyoup, . Blended fertilizers as draw solutions for fertilizer-drawn forward osmosis desalination. Environ. Sci. Technol.. 2012;46(8):4567-4575.
[Google Scholar]
Im S.-J., Jeong G., Jeong S., Cho J., Jang A., . Fouling and transport of organic matter in cellulose triacetate forward-osmosis membrane for wastewater reuse and seawater desalination. Chem. Eng. J.. 2020;384:123341
[Google Scholar]
Mi Baoxia, Elimelech Menachem, . Chemical and physical aspects of organic fouling of forward osmosis membranes. J. Membr. Sci.. 2008;320(1-2):292-302.
[Google Scholar]
Mi Baoxia, Elimelech Menachem, . Gypsum scaling and cleaning in forward osmosis: measurements and mechanisms. Environ. Sci. Technol.. 2010;44(6):2022-2028.
[Google Scholar]
Liden Tiffany, Carlton Doug D., Miyazaki Shinji, Otoyo Takehiko, Schug Kevin A., . Forward osmosis remediation of high salinity Permian Basin produced water from unconventional oil and gas development. Sci. Total Environ.. 2019;653:82-90.
[Google Scholar]

Introduction

Experimental

Materials
Preparation of modified nanocomposite flat-sheet GO/AA/CTA/CA based membranes
Membrane characterization and performance:
Membrane performance assessment for forward osmosis (FO)

Results and discussion

Characterization of the synthesized membranes
Performance evaluation of FO membranes
Effect of time on activity of synthesized FO membranes
Application of FO synthetic membranes in a natural saline water sample

Conclusion

Fulltext Views
1,247

PDF downloads
663

View/Download PDF
Download Citations

BibTeX
RIS

Show Sections

[1] Isawi H., . Evaluating the performance of different nano-enhanced ultrafiltration membranes for the removal of organic pollutants from wastewater. J. Water Process Eng.. 2019;31:100833
[Google Scholar]

[2] Isawi H., El-Sayed M.H., Feng X., Shawky H., Abdel Mottaleb M.S., . Surface nanostructuring of thin film composite membranes via grafting polymerization and incorporation of ZnO nanoparticles. Appl. Surf. Sci.. 2016;385:268-281.
[Google Scholar]

[3] D.F. Ahmed, H. Isawi, N.A. Badway, A. Elbayaa, H. Shawky, Highly porous cellulosic nanocomposite membranes with enhanced performance for forward osmosis desalination, Under Review, Iranian Polymer Journal.

[4] Qasim M., Darwish N.A., Sarp S., Hilal N., . Water desalination by forward (direct) osmosis phenomenon: A comprehensive review. Desalination. 2015;374:47-69.
[Google Scholar]

[5] Xiao P., Li J., Ren Y., Wang X., . A comprehensive study of factors affecting fouling behavior in forward osmosis. Colloids Surf., A. 2016;499:163-172.
[Google Scholar]

[6] Chung T.-S., Zhang S., Wang K.Y., Su J., Ling M.M., . Forward osmosis processes: yesterday, today and tomorrow. Desalination. 2012;287:78-81.
[Google Scholar]

[7] Su J., Zhang S., Ling M.M., Chung T.-S., . Forward osmosis: an emerging technology for sustainable supply of clean water. Clean Technol. Environ. Policy. 2012;14(4):507-511.
[Google Scholar]

[8] Shahlol O.M.A., Isawi H., El-Malky M.G., Al-Aassar A.-H., zwai A.E., . Performance evaluation of the different nano-enhanced polysulfone membranes via membrane distillation for produced water desalination in Sert Basin-Libya. Arabian J. Chem.. 2020;13(4):5118-5136.
[Google Scholar]

[9] Ong R.C., Chung T.-S., . Fabrication and positron annihilation spectroscopy (PAS) characterization of cellulose triacetate membranes for forward osmosis. J. Membr. Sci.. 2012;394-395:230-240.
[Google Scholar]

[10] Isawi H., . Using Zeolite/Polyvinyl alcohol/sodium alginate nanocomposite beads for removal of some heavy metals from wastewater. Arabian J. Chem.. 2020;13(6):5691-5716.
[Google Scholar]

[11] Isawi H., . Development of thin-film composite membranes via radical grafting with methacrylic acid/ZnO doped TiO2 nanocomposites. React. Funct. Polym.. 2018;131:400-413.
[Google Scholar]

[12] Tu Y., Lv M., Xiu P., Huynh T., Zhang M., Castelli M., Liu Z., Huang Q., Fan C., Fang H., . Destructive extraction of phospholipids from Escherichia coli membranes by graphene nanosheets. Nat. Nanotechnol.. 2013;8:594.
[Google Scholar]

[13] Wang X., Wang X., Xiao P., Li J., Tian E., Zhao Y., Ren Y., . High water permeable free-standing cellulose triacetate/graphene oxide membrane with enhanced antibiofouling and mechanical properties for forward osmosis. Colloids Surf., A. 2016;508:327-335.
[Google Scholar]

[14] Chee W.K., Lim H.N., Huang N.M., Harrison I., . Nanocomposites of graphene/polymers: a review. RSC Adv.. 2015;5(83):68014-68051.
[Google Scholar]

[15] Kim S., Yoo J.-B., Yi G.-R., Lee Y., Choi H.R., Koo J.C., Oh J.-S., Nam J.-D., . An aggregation-mediated assembly of graphene oxide on amine-functionalized poly (glycidyl methacrylate) microspheres for core–shell structures with controlled electrical conductivity. J. Mater. Chem. C. 2014;2(32):6462-6466.
[Google Scholar]

[16] He L.i., Dumée L.F., Feng C., Velleman L., Reis R., She F., Gao W., Kong L., . Promoted water transport across graphene oxide–poly (amide) thin film composite membranes and their antibacterial activity. Desalination. 2015;365:126-135.
[Google Scholar]

[17] Liang JiaJie, Huang Y.i., Zhang F., Zhang Y.i., Li N., Chen YongSheng, . The use of graphene oxide membranes for the softening of hard water. Science China Technological Sciences. 2014;57(2):284-287.
[Google Scholar]

[18] Yin J., Deng B., . Polymer-matrix nanocomposite membranes for water treatment. J. Membr. Sci.. 2015;479:256-275.
[Google Scholar]

[19] Johnson D.J., Suwaileh W.A., Mohammed A.W., Hilal N., . Osmotic's potential: An overview of draw solutes for forward osmosis. Desalination. 2018;434:100-120.
[Google Scholar]

[20] Panagopoulos A., Haralambous K.-J., Loizidou M., . Desalination brine disposal methods and treatment technologies-A review. Sci. Total Environ.. 2019;693:133545
[Google Scholar]

[21] Suwaileh W., Johnson D., Jones D., Hilal N., . An integrated fertilizer driven forward osmosis-renewables powered membrane distillation system for brackish water desalination: A combined experimental and theoretical approach. Desalination. 2019;471:114126
[Google Scholar]

[22] Majeed T., Sahebi S., Lotfi F., Kim J.E., Phuntsho S., Tijing L.D., Shon H.K., . Fertilizer-drawn forward osmosis for irrigation of tomatoes. Desalin. Water Treat.. 2015;53(10):2746-2759.
[Google Scholar]

[23] Xing X.-Y., Gu L., Jin Y.u., Sun R., Xie M.-Y., Wu Q.-Y., . Fabrication and characterization of cellulose triacetate porous membranes by combined nonsolvent-thermally induced phase separation. Cellulose. 2019;26(6):3747-3762.
[Google Scholar]

[24] Qasim M., Mohammed F., Aidan A., Darwish N.A., . Forward osmosis desalination using ferric sulfate draw solute. Desalination. 2017;423:12-20.
[Google Scholar]

[25] Chen X., Xu J., Lu J., Shan B., Gao C., . Enhanced performance of cellulose triacetate membranes using binary mixed additives for forward osmosis desalination. Desalination. 2017;405:68-75.
[Google Scholar]

[26] Liu X., Wu J., Hou L.-A., Wang J., . Performance and deterioration of forward osmosis membrane exposed to various dose of gamma-ray irradiation. Ann. Nucl. Energy. 2020;135:106950
[Google Scholar]

[27] D. Thi To Nu, N. Phi Hung, C. Van Hoang, B. Van der Bruggen, Preparation of an Asymmetric Membrane from Sugarcane Bagasse Using DMSO as Green Solvent, Applied Sciences 9 (2019) 3347.

[28] de Moraes Ana Carolina Mazarin, Andrade Patricia Fernanda, de Faria Andreia Fonseca, Simões Mateus Batista, Salomão Francisco Carlos Carneiro Soares, Barros Eduardo Bedê, Gonçalves Maria do Carmo, Alves Oswaldo Luiz, . Fabrication of transparent and ultraviolet shielding composite films based on graphene oxide and cellulose acetate. Carbohydr. Polym.. 2015;123:217-227.
[Google Scholar]

[29] Ionita Mariana, Crica Livia Elena, Voicu Stefan Ioan, Pandele Andreea Madalina, Iovu Horia, . Fabrication of cellulose triacetate/graphene oxide porous membrane. Polym. Adv. Technol.. 2016;27(3):350-357.
[Google Scholar]

[30] Ghaseminezhad S.M., Barikani M., Salehirad M., . Development of graphene oxide-cellulose acetate nanocomposite reverse osmosis membrane for seawater desalination. Compos. B Eng.. 2019;161:320-327.
[Google Scholar]

[31] Mansourpanah Y., Madaeni S., Rahimpour A., Adeli M., Hashemi M., Moradian M., . Fabrication new PES-based mixed matrix nanocomposite membranes using polycaprolactone modified carbon nanotubes as the additive: property changes and morphological studies. Desalination. 2011;277:171-177.
[Google Scholar]

[32] Gao Yong, Hu Meng, Mi Baoxia, . Membrane surface modification with TiO2–graphene oxide for enhanced photocatalytic performance. J. Membr. Sci.. 2014;455:349-356.
[Google Scholar]

[33] Wei Jing, Qiu Changquan, Tang Chuyang Y., Wang Rong, Fane Anthony G., . Synthesis and characterization of flat-sheet thin film composite forward osmosis membranes. J. Membr. Sci.. 2011;372(1-2):292-302.
[Google Scholar]

[34] Shen Liang, Xiong Shu, Wang Yan, . Graphene oxide incorporated thin-film composite membranes for forward osmosis applications. Chem. Eng. Sci.. 2016;143:194-205.
[Google Scholar]

[35] Rastgar Masoud, Bozorg Ali, Shakeri Alireza, Sadrzadeh Mohtada, . substantially improved antifouling properties in electro-oxidative graphene laminate forward osmosis membrane. Chem. Eng. Res. Des.. 2019;141:413-424.
[Google Scholar]

[36] Lim Sungil, Park Myoung Jun, Phuntsho Sherub, Tijing Leonard D., Nisola Grace M., Shim Wang-Geun, Chung Wook-Jin, Shon Ho Kyong, . Dual-layered nanocomposite substrate membrane based on polysulfone/graphene oxide for mitigating internal concentration polarization in forward osmosis. Polymer. 2017;110:36-48.
[Google Scholar]

[37] Shokrgozar Eslah S., Shokrollahzadeh S., Moini Jazani O., Samimi A., . Forward osmosis water desalination: Fabrication of graphene oxide-polyamide/polysulfone thin-film nanocomposite membrane with high water flux and low reverse salt diffusion. Sep. Sci. Technol.. 2018;53:573-583.
[Google Scholar]

[38] Choi Hyeon-gyu, Shah Aatif Ali, Nam Seung-Eun, Park You-In, Park Hosik, . Thin-film composite membranes comprising ultrathin hydrophilic polydopamine interlayer with graphene oxide for forward osmosis. Desalination. 2019;449:41-49.
[Google Scholar]

[39] Akther N., Yuan Z., Chen Y., Lim S., Phuntsho S., Ghaffour N., Matsuyama H., Shon H., . Influence of graphene oxide lateral size on the properties and performances of forward osmosis membrane. Desalination. 2020;484:114421
[Google Scholar]

[40] Phuntsho Sherub, Sahebi Soleyman, Majeed Tahir, Lotfi Fezeh, Kim Jung Eun, Shon Ho Kyong, . Assessing the major factors affecting the performances of forward osmosis and its implications on the desalination process. Chem. Eng. J.. 2013;231:484-496.
[Google Scholar]

[41] Lambrechts R., Sheldon M.S., . Performance and energy consumption evaluation of a fertiliser drawn forward osmosis (FDFO) system for water recovery from brackish water. Desalination. 2019;456:64-73.
[Google Scholar]

[42] Li Chunxia, Song Jianfeng, Dou Pengjia, Yin Yong, He Tao, He Weina, . Effects of low-toxicity solvent and binary mixed additives on the microstructure and performance of cellulose triacetate (CTA) membranes for forward osmosis. Desalin. Water Treat.. 2018;130:44-53.
[Google Scholar]

[43] Sahebi Soleyman, Sheikhi Mohammad, Ramavandi Bahman, Ahmadi Mehdi, Zhao Shuaifei, Adeleye Adeyemi S., Shabani Zhara, Mohammadi Toraj, . Sustainable management of saline oily wastewater via forward osmosis using aquaporin membrane. Process Saf. Environ. Prot.. 2020;138:199-207.
[Google Scholar]

[44] Corzo Beatriz, de la Torre Teresa, Sans Carmen, Ferrero Enrique, Malfeito Jorge J., . Evaluation of draw solutions and commercially available forward osmosis membrane modules for wastewater reclamation at pilot scale. Chem. Eng. J.. 2017;326:1-8.
[Google Scholar]

[45] Chekli Laura, Kim Youngjin, Phuntsho Sherub, Li Sheng, Ghaffour Noreddine, Leiknes TorOve, Shon Ho Kyong, . Evaluation of fertilizer-drawn forward osmosis for sustainable agriculture and water reuse in arid regions. J. Environ. Manage.. 2017;187:137-145.
[Google Scholar]

[46] Phuntsho Sherub, Shon Ho Kyong, Majeed Tahir, El Saliby Ibrahim, Vigneswaran Saravanamuthu, Kandasamy Jaya, Hong Seungkwan, Lee Sangyoup, . Blended fertilizers as draw solutions for fertilizer-drawn forward osmosis desalination. Environ. Sci. Technol.. 2012;46(8):4567-4575.
[Google Scholar]

[47] Im S.-J., Jeong G., Jeong S., Cho J., Jang A., . Fouling and transport of organic matter in cellulose triacetate forward-osmosis membrane for wastewater reuse and seawater desalination. Chem. Eng. J.. 2020;384:123341
[Google Scholar]

[48] Mi Baoxia, Elimelech Menachem, . Chemical and physical aspects of organic fouling of forward osmosis membranes. J. Membr. Sci.. 2008;320(1-2):292-302.
[Google Scholar]

[49] Mi Baoxia, Elimelech Menachem, . Gypsum scaling and cleaning in forward osmosis: measurements and mechanisms. Environ. Sci. Technol.. 2010;44(6):2022-2028.
[Google Scholar]

[50] Liden Tiffany, Carlton Doug D., Miyazaki Shinji, Otoyo Takehiko, Schug Kevin A., . Forward osmosis remediation of high salinity Permian Basin produced water from unconventional oil and gas development. Sci. Total Environ.. 2019;653:82-90.
[Google Scholar]

Graphene oxide incorporated cellulose triacetate/cellulose acetate nanocomposite membranes for forward osmosis desalination

Abstract

Keywords

Forward osmosis Desalination

CTA/CA polymer blend

Acetic acid pore-forming additive

Graphene oxide nanocomposite

Fertilizer draw solution

Saline water desalination

1 Introduction

2 Experimental

2.1 Materials

2.2 Preparation of modified nanocomposite flat-sheet GO/AA/CTA/CA based membranes

2.3 Membrane characterization and performance:

2.4 Membrane performance assessment for forward osmosis (FO)

3 Results and discussion

3.1 Characterization of the synthesized membranes

3.1.1 FTIR analysis

3.1.2 Morphology analysis of membranes

3.1.3 Contact angle and surface porosity of FO membranes.

3.1.4 Mechanical properties

3.2 Performance evaluation of FO membranes

3.2.1 Effect of acetic acid concentration on the performance of fabricated membranes

3.2.2 Effect of GO content on the performance of fabricated membranes

3.2.3 Effects of membrane orientation on the performance of fabricated membranes

3.2.4 The effect of different concentration of NaCl as a draw solution

3.2.5 The influence of different fertilizer draws solution on the performance of FO process

3.3 Effect of time on activity of synthesized FO membranes

3.4 Application of FO synthetic membranes in a natural saline water sample

4 Conclusion

Declaration of Competing Interest

References

Suggested read for related articles: