Abstract

Removal of low-concentration ammonia (1–10 ppm) from aquaculture wastewater was investigated via polysulfone (PSf)/zeolite mixed matrix membrane. PSf/zeolite mixed matrix membranes with different weight ratios (90/10, 80/20, 70/30 and 60/40 wt.%) were prepared and characterized. Results indicate that PSf/zeolite (80/20) was the most efficient membrane for removal of low-concentration ammonia. The ammonia elimination by PSf/zeolite (80/20) from aqueous solution for 10, 7, 5, 3 and 1 ppm of ammonia was 100%, 99%, 98.8%, 96% and 95% respectively. The recorded results revealed that pure water flux declined in higher loading of zeolite in the membrane matrix due to surface pore blockage caused by zeolite particles. On the other hand, ammonia elimination from water was decreased in higher contents of zeolite because of formation of cavities and macrovoids in the membrane substructure.

INTRODUCTION

Aquaculture is a fast growing industry that demands quantities of water in the order of 200–600 m3/kg fish produced. Aquaculture commonly is presented as a clean industry (Nora'aini et al. 2009). Fish excretion and uneaten fish feed is the main source of ammonia in aquaculture water. The rate at which fish excrete ammonia is directly related to the feeding rate and the protein level in feed. Ammonia is the major end product of protein catabolism, which remains in the form of un-ionized ammonia (NH3) and ionized ammonia (NH4+). Un-ionized ammonia is a critical water quality parameter and toxic to aquatic life, but the ammonium ion is harmless except in extremely high concentrations (Prabhakar et al. 2015).

Aquaculture water requires ammonia removal at levels of less than 1 mg/L (Jorgensen & Weatherley 2003). If the ammonia concentration gets high enough, the fish will become lethargic and eventually fall into a coma and die. Hence, the reduction of the impact of total ammonia and nitrite on the receiving environment may be essentially obtained upstream by optimizing shrimp/fish farming management practices regarding feeding and water quality (Porrello et al. 2003). Complete removal of ammonia is required due to its extreme toxicity to most fish species. Aquaculture occasionally requires water to be cool (∼10 °C), which is an unsatisfactory temperature for many biological methods. Biological methods (nitrification) do not respond well to shock loads of ammonia, and unacceptable peaks in effluent ammonium concentration may result in such cases. Fluctuating concentrations of ammonium are more toxic to fish than moderate yet constant concentrations (Jorgensen & Weatherley 2003). Zeolite is a promising adsorbent material with high ion exchange capacity for ammonia removal in fishery water. However, the major concern is that zeolite particles will tend to degrade in water. Moreover, other ions and elements tend to leak into the solution, preventing zeolite from being an effective adsorbent for ammonium removal in fishery water. In order to overcome the limitation of the direct adsorption approach, zeolite particles were incorporated within polymeric matrices in three different conformations, namely mixed matrix membranes (MMMs), composite fibers, and pore-filled membranes (Burgess et al. 2004).

Many researchers have studied the ammonia removal from industrial and municipal wastewater by different methods. However, only a few studies are available of ammonia removal from aquaculture wastewater. Ali et al. studied the formation and characterization of an asymmetric nanofiltration membrane for ammonia-nitrogen removal and the effect of shear rate. Based on their results, they could achieve the retention of salt and ammonia-nitrogen at 61% and 68%, respectively (Ali et al. 2010). In another research, Nora'aini et al. investigated the potential of polysulfone (PSf) ultrafiltration membrane as an alternative treatment technology for ammonia removal from aquaculture wastewater. According to their outcomes, the ammonia removal improved in the following order: PSf 23% > PSf 21% > PSf 18%. The smaller pore size in PSf 23% leads to the highest rejection of ammonia with an average removal of 60% (Nora'aini et al. 2009). Zeolites are well known nanoparticles for removing ammonia from wastewater (Burgess et al. 2004; Mazloomi & Jalali 2016). Burgess et al. studied the potential of zeolite for removing ammonia and ammonia-caused toxicity in marine toxicity identification evaluations. There results indicate that use of zeolite in a column-chromatography design is an effective method for decreasing ammonia concentrations in seawater and decreasing ammonia toxicity to marine organisms. They mentioned that, using the column-chromatography, pH did not affect the performance of zeolite in removing total ammonia. The ammonia removal percentage was around 80% in pH range of 4 to 10 (Burgess et al. 2004). In another work, Mazloomi & Jalali investigated the performance of natural Iranian zeolite for ammonium removal from aqueous solutions. They studied the effect of pH, adsorbent dose, contact time, and temperature. The optimum conditions for the effective adsorption of NH4+ ions (∼95%) onto natural zeolite were found to be pH 7.0, temperature 298 K, and contact time 30 min (Mazloomi & Jalali 2016).

The objective of the present study is to clarify the effect of zeolite (Z) concentration on morphology and performance of PSf/Z MMMs for removal of low-concentration ammonia from aquaculture wastewater. Moreover, the effect of ammonia concentration in the feed solution on the ammonia removal performance of prepared PSf/Z MMMs was analyzed. The novelty of this research compared to other literature is high removal of ammonia (≥ 95%) from aquaculture wastewater containing 1–10 mg/L of ammonia. Furthermore, the pure water flux of PSf/Z (80/20) membrane, which was able to have the highest ammonia removal, was 40, 85 and 95 L/(m2·h) at 1, 2 and 3 bar feed pressure respectively.

EXPERIMENTAL

Materials

PSf resin with the number average molecular weight 22,000 and zeolite with particle size of ≤45 μm were purchased from Sigma-Aldrich. The specific surface area of the zeolite, which was determined using the BET (Brunauer–Emmett–Teller) surface area technique, was 25.8 m2/g. PSf is an amorphous thermoplastic polymer with glass transition temperature of 190 °C. This is a flame retardant polymer, possessing high mechanical, thermal, and oxidative stability, and is soluble in common organic solvents (Moradihamedani & Abdullah 2016). N-N-dimethyl-acetamide (DMAc) from Merck was utilized as a solvent for casting solution preparation. Merck provided ammonia solution with a purity of 25% and Nessler's reagent. Distilled water was used as the coagulation bath. The PSf resin was dried in an oven at 80 °C for 24 h before the usage.

Preparation of membranes

In the present research, flat sheet PSf/Z MMMs were prepared by wet phase inversion technique. Based on this method, the casting solution was cast on a glass plate by a Doctor blade with a thickness of 80 μm. The wet film was immersed in a coagulation bath containing distilled water for 24 h at room temperature. At the final stage, the membranes were dried at room temperature for a day. The casting solutions contained 15 wt.% of PSf and Z with different compositions (PSf/Z: 90/10, 80/20, 70/30 and 60/40) using DMAc as solvent. The compositions of prepared membranes and their designations are listed in Table 1.

Table 1

Compositions of PSf/Z blend membranes

Membrane Casting solutions (wt.%)
 
Composition (15 wt.%)
 
Solvent 
PSf Zeolite DMAc 
Neat PSf 100 85 
PSf/Z (90/10) 90 10 85 
PSf/Z (80/20) 80 20 85 
PSf/Z (70/30) 70 30 85 
PSf/Z (60/40) 60 40 85 
Membrane Casting solutions (wt.%)
 
Composition (15 wt.%)
 
Solvent 
PSf Zeolite DMAc 
Neat PSf 100 85 
PSf/Z (90/10) 90 10 85 
PSf/Z (80/20) 80 20 85 
PSf/Z (70/30) 70 30 85 
PSf/Z (60/40) 60 40 85 

Total mass for casting solution; PSf + Z + DMAc = 15 g.

Ammonia removal

Ammonia removal was carried out with aqueous solutions containing 1, 3, 5, 7 and 10 mg/L of ammonia. All experiments were performed three times at room temperature (25 ± 2 °C) in a batch type and dead-end ultrafiltration (UF) cell (Merck) with an effective membrane filtration area of 13.8 cm2. In this work, the collection time was recorded for 50 cm3 of the permeate solution and then the permeate flux, ammonia rejection and morphology of each membrane were analyzed (Moradihamedani & Abdullah 2016). Figure 1 shows the schematic diagram of the dead-end UF system used in this experiment.

Figure 1

Schematic diagram of the dead-end ultrafiltration system for removal of ammonia from water.

Figure 1

Schematic diagram of the dead-end ultrafiltration system for removal of ammonia from water.

The performance of the prepared membranes in the ammonia removal process was evaluated based on the flux measurement and the rejection percentage. The pure water flux was measured at feed pressure range of 1 to 3 bar. The following equation was used for calculation of the flux:  
formula
(1)
where Jw is the flux (L/m2.h), Q is the amount of collected permeate (L), A is the effective membrane area in cm2 and Δt is the sampling time (h) (Moradihamedani et al. 2016).

The concentration of ammonia was evaluated using Nesslerization. In this method, the ammonia reacts with Nessler reagent to form a colored complex that can change from yellow to deep amber. Absorbance of ammonia was measured by a UV-spectrophotometer (Cam-spec M-350 model) at 425 nm (Jorgensen & Weatherley 2003; Ashrafizadeh & Khorasani 2010; Rezakazemi et al. 2012).

Characterization of the membrane

The surface functional groups of the neat PSf and PSf/Z MMMs were analyzed by Fourier transform infrared spectroscopy (FTIR: Series 100 PerkinElmer FT-IR 1650) in the scanning range of 500–4,000 cm−1. The cross-sectional images of the PSf/Z MMMs were observed by scanning electron microscope (SEM, Philips XL-30). The samples were frozen under liquid nitrogen for 3 minutes, and then frozen fragments were broken and coated with gold by sputtering for producing electric conductivity. Variations in surface roughness parameters of prepared MMMs were studied by atomic force microscopy (AFM) (Ambios Q-scope, Linthi cum Heights, MD, USA) in tapping mode.

The membrane water content was measured by soaking the membrane in water for 24 h and weighing it after wiping off the excess water with tissue paper. The membranes were dried in an oven at 40 °C for 24 h and weighed. Based on wet and dry weights of the membranes, the water content was calculated using the following equation (Abedini et al. 2011; Han et al. 2013):  
formula
(2)
where Ww and Wd are the weights of wet and dry membranes respectively.
The porosity of prepared membranes was calculated using the following equation (Bai et al. 2012; Abdullah et al. 2016):  
formula
(3)
where Ww is the weight of wet membrane, Wd is the weight of dry membrane, d is the density of water (g/cm3), A is the membrane effective area (cm2) and L is the thickness of membrane (cm).

Results and discussion

FTIR spectroscopy was used to investigate the interactions between PSf and zeolite particles. In this respect, the following three samples were compared: neat PSf, PSf/Z (80/20) and PSf/Z (60/40) (Figure 2). The peak at 3,650 cm−1 corresponds to the O–H bond in zeolite. This peak describes the existence of an interaction between polymer and zeolite. In high zeolite loadings, the peak intensity increases and the bonds between the two phases become stronger (Ismail et al. 2008; Dorosti et al. 2011). The asymmetric Ar-O-Ar vibration of the aromatic ether band appears at 1,245 cm−1. The SO2 group exhibits the asymmetric and symmetric O = S = O stretching vibrations at 1,325 and 1,020 cm−1 respectively. Note that the asymmetric O = S = O vibrations are split into two bands at 1,324 and 1,294 cm−1. These bands show the existence of PSf.

Figure 2

FTIR spectra of (a) neat PSf, (b) PSf/Z (80/20) and (c) PSf/Z (60/40).

Figure 2

FTIR spectra of (a) neat PSf, (b) PSf/Z (80/20) and (c) PSf/Z (60/40).

The surface roughness changes of the prepared MMMs in different zeolite loadings were analyzed by AFM technique (Figure 3). AFM images present bulges (bright points) and valleys (dark points) by scanning a sharp tip on the membrane surface. The more ups and downs result in higher roughness of membrane surface. Surface pore size can control the intensity of these ups and downs. Comparing the roughness parameters obtained from AFM images (Table 2) confirmed that the surface roughness of the membrane increased with higher amounts of zeolite in the membrane matrix. Membrane surface roughness enhancement in higher zeolite contents might be due to the accumulation and less dispersion of zeolite particles. The higher content of zeolite can produce the bulges because of decreasing the distance between nanoparticles in casting solution.

Table 2

Surface roughness parameters with different zeolite loadings

Membranes RMS rough (nm) Mean rough (nm)% Water content Porosity % 
Neat PSf 10.30 8.80 74.5 65.3 
PSf/Z (80/20) 12.50 10.20 80.0 76.0 
PSf/Z (70/30) 14.30 12.60 81.3 86.9 
PSf/Z (60/40) 16.60 14.50 82.6 95.1 
Membranes RMS rough (nm) Mean rough (nm)% Water content Porosity % 
Neat PSf 10.30 8.80 74.5 65.3 
PSf/Z (80/20) 12.50 10.20 80.0 76.0 
PSf/Z (70/30) 14.30 12.60 81.3 86.9 
PSf/Z (60/40) 16.60 14.50 82.6 95.1 

RMS: root mean square.

Figure 3

Three-dimensional AFM images of MMMs surface layer with different zeolite contents: (a) neat PSf, (b) PSf/Z (80/20) and (c) PSf/Z (60/40).

Figure 3

Three-dimensional AFM images of MMMs surface layer with different zeolite contents: (a) neat PSf, (b) PSf/Z (80/20) and (c) PSf/Z (60/40).

SEM micrographs from the cross-section of membranes show the changes in membrane sublayer porosity, pore shape and distribution of zeolite particles in membrane morphology (Figure 4). According to Figure 4, sublayer macrovoids have grown by increasing the zeolite amount in the membrane matrix. It can be seen that the highest quantity of zeolite causes the creation of largest macrovoids in the PSf/Z (60/40) membrane (see Figure 4(c)). Therefore, the membrane with higher content of zeolite can absorb and save more water into the macrovoids (Casado-Coterillo et al. 2012) leading to higher water content of the MMM (but less water passage due to surface pore blockage).

Figure 4

SEM micrographs from cross-section of (a) neat PSf, (b) PSf/Z (80/20) and (c) PSf/Z (60/40).

Figure 4

SEM micrographs from cross-section of (a) neat PSf, (b) PSf/Z (80/20) and (c) PSf/Z (60/40).

The data given in Table 2 verify that a membrane with higher zeolite content offers greater void capacity and swells to a higher degree. This is attributed to the presence of zeolite and its effect on phase inversion kinetics, i.e., the rate of membrane precipitation during replacement of water and DMAc (Arthanareeswaran et al. 2009). Zeolite–polymer interfacial incompatibility, brittleness and particle agglomeration are other important factors leading to void formation at higher zeolite loadings. The main reason for this structure is weak interactions between polymers and zeolite surfaces as well as incompatibility of the organic and inorganic nature of polymer and zeolite along with low flexibility of glassy polymers. These phenomena cause stresses at the interface during casting of the membrane or evaporation of the solvent. These stresses may in turn separate the polymers from zeolite particles at higher zeolite loadings.

On the other hand, the SEM images of the surface of membranes show the changes in membrane top layer in different zeolite contents (Figure 5). Despite the agglomeration of zeolites near surface pores, which is the reason for pore blockage, the surface SEM micrographs proved good distribution of zeolite particles throughout of the membrane surface layer.

Figure 5

SEM micrographs from surface of (a) neat PSf, (b) PSf/Z (80/20) and (c) PSf/Z (60/40).

Figure 5

SEM micrographs from surface of (a) neat PSf, (b) PSf/Z (80/20) and (c) PSf/Z (60/40).

Pure water fluxes of prepared membranes in pressure range of 1 to 3 bar are indicated in Figure 6. Accordingly, neat PSf and PSf/Z (90/10) membranes did not have any water flux until 3 bar feed pressure. The AFM (Figure 3(a)) and SEM (Figure 5(a)) images of neat PSf membrane clearly show a dense structure for the skin layer that does not allow water molecules to pass through the membrane. The pure water flux increased significantly by addition of 20% of zeolite to the membrane matrix. The pore formation on the surface layer of PSf/Z (80/20) membrane (Figures 3(b) and 5(b)) is the main reason for water flux enhancement in PSf/Z (80/20) membrane. Figure 6 obviously indicates that addition of higher amounts of zeolite particles (>20%) to the membrane matrix leads to reduction of water flux, while the water swelling tests indicated that water content of membranes was enhanced by increasing the zeolite loading of the membranes (Table 2). This can be described by the mutual effect of zeolite particles on membrane structure (porosity increment) against facial pore blockage by these particles. It should be noted that the zeolite particles tend to be accumulated in the membrane surface and its superficial pores, as the membrane surface is the first place contacting with water (coagulation bath) during the phase inversion process (Vatanpour et al. 2011). The migration of zeolite particles to the top of the membrane leads to more changes in membrane surface pores (Vatanpour et al. 2011). Decreasing the pure water flux in PSf/Z (70/30) and PSf/Z (60/40) membranes is because of pore blockage in the top layer of prepared membranes.

Figure 6

Pure water flux of prepared membranes.

Figure 6

Pure water flux of prepared membranes.

The performance of prepared PSf/Z membranes in terms of ammonia removal from aquaculture wastewater is recorded in Table 3. The ammonia removal was analyzed in different concentrations of ammonia in feed solution, different zeolite loadings in membrane matrix and different feed pressures. Based on the results obtained, the ammonia removal increased gradually in higher concentrations of ammonia in feed solution. PSf/Z (80/20) and PSf/Z (70/30) membranes were able to remove 100% of ammonia from feed solution containing 10 ppm of ammonia. PSf/Z (80/20) had a great performance in different ammonia concentrations in the feed solution. This membrane was able to remove 95% of ammonia from aqueous solution containing the lowest concentration of ammonia (1 ppm). As provided in Table 3, the ammonia removal from aquaculture wastewater decreased in higher loadings of zeolite in the membrane matrix. In spite of lower water flux in higher concentrations of zeolite in the membrane matrix (due to surface pore blockage as mentioned before), creation of cavities and macrovoids in PSf/Z (70/30) and PSf/Z (60/40) are the reasons for reduction of ammonia removal percentage in higher contents of zeolite in membrane matrix. SEM images of cross-section views of prepared membranes (Figure 4) clearly show the variation of pore size in the membrane structure in different zeolite concentrations. As shown in Figure 4(c), PSf/Z (60/40) contains macrovoids starting from top layer and continuing to the support layer of the membrane. Existence of these large cavities causes the ammonia removal reduction by the membrane. The results recorded in Table 3 show that PSf/Z (80/20) had a great performance in term of ammonia removal in different concentrations of ammonia in the feed solution. This membrane removed 100%, 99%, 98.8%, 96% and 95% of ammonia from feed solutions containing of 10, 7, 5, 3 and 1 mg/L of ammonia respectively. Accordingly, the existence of the right amount of zeolite accompanied with good dispersion in the PSf membrane matrix can lead to high elimination of ammonia from aquaculture wastewater. Feed pressure affects both the pure water flux (Figure 6) and ammonia removal (Table 3). As shown in Figure 6, water flux across the membrane rises in direct relationship to increases in feed pressure. In contrast, ammonia removal decreases with increasing the feed pressure. This is due to the compression of the deposited layer on the membrane surface. Hence, ammonia passage is increasingly overcome as water is pushed through the membrane (Rahimpour et al. 2007).

Table 3

Ammonia removal in different feed concentrations and pressures

    Ammonia removal (%)
 
Membrane Ammonia concentration Pressure (bar)
 
(ppm) 
PSf/Z (80/20) 10 99.1 99.7 100 
PSf/Z (70/30) 10 98.6 99.6 100 
PSf/Z (60/40) 10 88.0 86.0 81.0 
PSf/Z (80/20) 99.0 97.6 97.6 
PSf/Z (70/30) 95.0 95.0 93.6 
PSf/Z (60/40) 84.3 83.0 83.0 
PSf/Z (80/20) 98.8 97.3 97.0 
PSf/Z (70/30) 88.4 88.2 88.2 
PSf/Z (60/40) 82.8 94.8 95.0 
PSf/Z (80/20) 96.0 88.5 84.0 
PSf/Z (70/30) 68.6 64.1 59.0 
PSf/Z (60/40) 57.2 52.0 48.4 
PSf/Z (80/20) 75.6 87.3 95.0 
PSf/Z (70/30) 52.0 63.0 61.0 
PSf/Z (60/40) 48.0 45.0 44.0 
    Ammonia removal (%)
 
Membrane Ammonia concentration Pressure (bar)
 
(ppm) 
PSf/Z (80/20) 10 99.1 99.7 100 
PSf/Z (70/30) 10 98.6 99.6 100 
PSf/Z (60/40) 10 88.0 86.0 81.0 
PSf/Z (80/20) 99.0 97.6 97.6 
PSf/Z (70/30) 95.0 95.0 93.6 
PSf/Z (60/40) 84.3 83.0 83.0 
PSf/Z (80/20) 98.8 97.3 97.0 
PSf/Z (70/30) 88.4 88.2 88.2 
PSf/Z (60/40) 82.8 94.8 95.0 
PSf/Z (80/20) 96.0 88.5 84.0 
PSf/Z (70/30) 68.6 64.1 59.0 
PSf/Z (60/40) 57.2 52.0 48.4 
PSf/Z (80/20) 75.6 87.3 95.0 
PSf/Z (70/30) 52.0 63.0 61.0 
PSf/Z (60/40) 48.0 45.0 44.0 

Combining ion exchange and filtration provides clear benefits for membranes in wastewater treatment plants. A membrane-covered zeolite maintains the advantages of pure zeolite of being a rapid process and having a good response to shock loads of ammonia. Additionally, the active membrane matrix, which removes part of the incoming ammonia, delays the saturation point and therefore less regeneration is necessary (Santiago et al. 2016). In this study, PSf/Z (80/20), which had the best performance among prepared membranes, was regenerated by filtering 1 M NaCl solution through the membranes followed by rinsing with deionized water (Ahmadiannamini et al. 2017). The regenerated membranes exhibit reproducible ammonia removal capacities. Figure 7 confirms that PSf/Z (80/20) can be reused even after four cycles of filtration process with only 4.2% reduction in ammonia removal.

Figure 7

Reusability of PSf/Z (80/20) membrane for four sequential runs. (Feed: 1 ppm of ammonia, pressure: 3 bar).

Figure 7

Reusability of PSf/Z (80/20) membrane for four sequential runs. (Feed: 1 ppm of ammonia, pressure: 3 bar).

In Table 4, the optimum results obtained in previous studies relevant to this research are summarized. The recorded results show that the current research achieved a great improvement in terms of ammonia removal percentage in aquaculture wastewater. As provided in Table 4, PSf/Z (80/20) was able to achieve ≥95% ammonia removal from aquaculture wastewater containing 1 to 10 ppm of ammonia.

Table 4

Comparison of ammonia removal performance of prepared membranes in current work with the literature

Membrane Ammonia concentration Ammonia removal (%)  Reference 
PSf/Z (80/20) 10 ppm 100 (3 bar) Present study 
PSf/Z (80/20) 7 ppm 99 (1 bar) 
PSf/Z (80/20) 5 ppm 98.8 (1 bar) 
PSf/Z (80/20) 3 ppm 96 (1 bar) 
PSf/Z (80/20) 1 ppm 95 (3 bar) 
PSf 0.432 ppm 68 (10 bar) Nora'aini et al. (2009)  
PES 10 ppm 68 (8 bar) Ali et al. (2010)  
PES 0.432 ppm 85.7 (4 bar) Ali et al. (2005)  
PSf/Z (50/50) 7 ppm 73 (15 LMH/bar) Ahmadiannamini et al. (2017)  
Membrane Ammonia concentration Ammonia removal (%)  Reference 
PSf/Z (80/20) 10 ppm 100 (3 bar) Present study 
PSf/Z (80/20) 7 ppm 99 (1 bar) 
PSf/Z (80/20) 5 ppm 98.8 (1 bar) 
PSf/Z (80/20) 3 ppm 96 (1 bar) 
PSf/Z (80/20) 1 ppm 95 (3 bar) 
PSf 0.432 ppm 68 (10 bar) Nora'aini et al. (2009)  
PES 10 ppm 68 (8 bar) Ali et al. (2010)  
PES 0.432 ppm 85.7 (4 bar) Ali et al. (2005)  
PSf/Z (50/50) 7 ppm 73 (15 LMH/bar) Ahmadiannamini et al. (2017)  

PES: Polyethersulfone, LMH: L/(m2·h).

CONCLUSION

The high performance MMMs were prepared by PSf and zeolite particles with different weight ratios in order to remove low-concentration ammonia from aquaculture wastewater. FTIR spectroscopy was used to investigate the interactions between PSf and zeolite particles. The surface roughness changes of the prepared MMMs in different zeolite loadings were analyzed by AFM technique. The AFM results indicate that the surface roughness of membranes increased in higher concentrations of zeolite particles in the membrane matrix. The SEM images from the top layer of membranes demonstrated that the surface pore blockage has happened in higher contents of zeolite particles. Meanwhile, the cross-section view of prepared membranes taken by SEM revealed that the big cavities and macrovoids were formed in higher loadings of zeolite in the membrane matrix. The pure water flux declined in higher zeolite contents in the membrane matrix, caused by surface pore blockage. Furthermore, the ammonia elimination from aqueous solution was decreased in higher zeolite concentrations due to formation of macrovoids in the membrane substructure.

REFERENCES

REFERENCES
Abdullah
,
N.
,
Gohari
,
R. J.
,
Yusof
,
N.
,
Ismail
,
A. F.
,
Juhana
,
J.
,
Lau
,
W. J.
&
Matsuura
,
T.
2016
Polysulfone/hydrous ferric oxide ultrafiltration mixed matrix membrane: preparation, characterization and its adsorptive removal of lead (II) from aqueous solution
.
Chemical Engineering Journal
289
,
28
37
.
Ahmadiannamini
,
P.
,
Eswaranandam
,
S.
,
Wickramasinghe
,
R.
&
Qian
,
X.
2017
Mixed-matrix membranes for efficient ammonium removal from wastewaters
.
Journal of Membrane Science
526
,
147
155
.
Ali
,
N.
,
Mohammad
,
A. W.
,
Jusoh
,
A.
,
Hasan
,
M. R.
,
Ghazali
,
N.
&
Kamaruzaman
,
K.
2005
Treatment of aquaculture wastewater using ultra-low pressure asymmetric polyethersulfone (PES) membrane
.
Desalination
185
(
1–3
),
317
326
.
Arthanareeswaran
,
G.
,
Sriyamuna Devi
,
T. K.
&
Mohan
,
D.
2009
Development, characterization and separation performance of organic–inorganic membranes
.
Separation and Purification Technology
67
(
3
),
271
281
.
Ashrafizadeh
,
S. N.
&
Khorasani
,
Z.
2010
Ammonia removal from aqueous solutions using hollow-fiber membrane contactors
.
Chemical Engineering Journal
162
(
1
),
242
249
.
Burgess
,
R. M.
,
Perron
,
M. M.
,
Cantwell
,
M. G.
,
Ho
,
K. T.
,
Serbst
,
J. R.
&
Pelletier
,
M. C.
2004
Use of zeolite for removing ammonia and ammonia-caused toxicity in marine toxicity identification evaluations
.
Archives of Environmental Contamination and Toxicology
47
(
4
),
440
447
.
Casado-Coterillo
,
C.
,
Soto
,
J.
,
Jimaré
,
M. T.
,
Valencia
,
S.
,
Corma
,
A.
,
Téllez
,
C.
&
Coronas
,
J.
2012
Preparation and characterization of ITQ-29/polysulfone mixed-matrix membranes for gas separation: effect of zeolite composition and crystal size
.
Chemical Engineering Science
73
,
116
122
.
Dorosti
,
F.
,
Omidkhah
,
M. R.
,
Pedram
,
M. Z.
&
Moghadam
,
F.
2011
Fabrication and characterization of polysulfone/polyimide–zeolite mixed matrix membrane for gas separation
.
Chemical Engineering Journal
171
(
3
),
1469
1476
.
Ismail
,
A. F.
,
Rahim
,
R. A.
&
Rahman
,
W. A. W. A.
2008
Characterization of polyethersulfone/Matrimid 5218 miscible blend mixed matrix membranes for O2/N2 gas separation
.
Separation and Purification Technology
63
(
1
),
200
206
.
Mazloomi
,
F.
&
Jalali
,
M.
2016
Ammonium removal from aqueous solutions by natural Iranian zeolite in the presence of organic acids, cations and anions
.
Journal of Environmental Chemical Engineering
4
(
2
),
1664
1673
.
Moradihamedani
,
P.
&
Abdullah
,
A. H.
2016
Phosphate removal from water by polysulfone ultrafiltration membrane using PVP as a hydrophilic modifier
.
Desalination and Water Treatment
57
(
53
),
25542
25550
.
Moradihamedani
,
P.
,
Kalantari
,
K.
,
Abdullah
,
A. H.
&
Morad
,
N. A.
2016
High efficient removal of lead(II) and nickel(II) from aqueous solution by novel polysulfone/Fe3O4–talc nanocomposite mixed matrix membrane
.
Desalination and Water Treatment
57
(
59
),
28900
28909
.
Nora'aini
,
A.
,
Nurbaiti
,
A. H.
&
Ahmad
,
J.
2009
The potential of a polysulfone (PSF) nanofiltration membrane as the end stage treatment technology of aquaculture wastewater
.
Desalination and Water Treatment
32
(
1
),
627
633
.
Porrello
,
S.
,
Ferrari
,
G.
,
Lenzi
,
M.
&
Persia
,
E.
2003
Ammonia variations in phytotreatment ponds of land based fish farm wastewater
.
Aquaculture Engineering
219
(
1–4
),
485
494
.
Prabhakar
,
M.
,
Sarkar
,
S.
,
Sethi
,
L. N.
&
Kunmar
,
A.
2015
Ammonia removal from aquaculture waste water using coconut coir media in lab-scale batch reactors
.
Journal of Energy Research and Environmental Technology
2
(
1
),
30
34
.
Santiago
,
T.
,
Cadiz
,
M.
,
Piedrahita
,
R. H.
&
Simmons
,
C. W.
2016
Ammonia Removal in Simulated Aquaculture Wastewater by Biofilm-Covered Zeolite
.
University of California, Davis, ProQuest Dissertations Publishing
.
Vatanpour
,
V.
,
Madaeni
,
S. S.
,
Moradian
,
R.
,
Zinadini
,
S.
&
Astinchap
,
B.
2011
Fabrication and characterization of novel antifouling nanofiltration membrane prepared from oxidized multiwalled carbon nanotube/polyethersulfone nanocompo-site
.
Journal of Membrane Science
375
(
1–2
),
284
294
.