Evaluating the performance of chitosan and chitosan-palm membrane for water treatment: preparation, characterization and purification study

In this research, the membranes were stemmed from the biopolymer containing quaternary amine moieties (Chitosan and Chitosan-palm) for nanofiltration purposes. The developed membranes were fully featured using different characterization techniques (SEM), (TGA), zeta potential, and contact angle measurement. The membrane’s features were systematically characterized in hydrophilicity contact angle, surface morphology, and charge on the surface, acidity, and water permeability. The permeability of water for the chitosan membrane with palm was 3.04 ± 0.12 L m−2 h−1bar−1 twice as the average permeability of the pristine chitosan membrane 1.68 ± 0.04 L m−2.h−1.bar−1. The salt rejection was enhanced (from 5% for NaCl to 70% for MgCl2 in the same condition). These membranes could endure up to 22 bar. Therefore, the developed Chitosan and chitosan-palm membranes are more noteworthy for water treatment than the other commercially available membranes and costly activated carbons.


Introduction
The membrane technology has arisen as a new approach to regulate drinking water contamination because of its intrinsic features [1,2], in comparison with those other techniques like adsorption; distillation; and extraction. The lack of drinking water has become a significant challenge around the globe in recent years. Due to the rapid progress in humans' population and industrial water pollution, drinking water resources have been reduced [3]. Human actions have contaminated the water with vast quantities of pesticides, minerals, drugs, and various toxic metals [4][5][6][7][8]. Consequently, novel inventions and resources are being taken into account for the purification of water. Several membranes developed from polymers with biomaterials exhibit vast prospective for water treatment [9]. Various membranes are used for separating particles with several sizes through microfiltration, conventional filtration, ultrafiltration, reverse osmosis, and nanofiltration [10]. Nanofiltration (NF) membrane is among the most effective wastewater treatment activities [11]. It is a comparatively modern improvement in membrane application, and it might be aqueous or non-aqueous. Nanofiltration has substituted reverse osmosis in several technologies owing to lesser energy intake and greater flux rates [12,13]. The NF membrane features lie among non-porous reverse osmosis membranes and porous ultrafiltration membranes [14]. This nanofiltration (NF) is a type of separation membrane technology.
Most recently produced NF membranes are composite membranes comprising an active layer and a support layer. Because of their surface charge, the separation properties of this type of membrane can be determined by the Donnan effect's co-effect. However, their surface nano-sized pores and mass transport will allow diffusion and convection [15].
Additionally, NF employs many mechanisms to reject salts and inorganic contaminants. Size exclusion is one of these mechanisms used by NF. Ion transport is considerably affected by hydration strength and hydrated radii due to size variations. NF membranes can reject that ions can pass through the pores of the membrane using diffusion approach [15,16].
In this situation, the addition of biopolymers like pure Chitosan and mixing with other biomaterials greatly improves the membranes' flexibility and performance. Over recent decades, researchers and scientists have found membrane-coated adsorbents to be prevalent, enticing methods for extracting hazardous substances from wastewater, such as volatile organic materials, heavy metals, and dyes. Chitosan-based membranes have evolved to remove numerous pollutants and toxic heavy metals from several water resources among these membrane adsorbents. Chitosan is obtained from chitin deacetylation and is considered to be the second most common natural polysaccharide as a natural polymer. Because of their biodegradability, reactivity, being environmentally-friendly, anti-toxicity, and desirable hydrophilicity chitosan biopolymers, and their compounds have attracted more attention. Using Chitosan, which is vital to the separation characteristics and active surface substances, can enhance the efficiency of NF membranes. The Chitosan contains 2-amino-2deoxy-(1, 4)-D-glucopyranose and it is easily formed from the crabs shells, shrimps and prawns [16].
Moreover, Chitosan can be used for a wide application, such as water and wastewater purification [17], elimination of heavy metal ions from aquatic environment [18,19], biomaterials [20,21], membrane materials [22,23]. Recently Chitosan and its derivatives have been used with nanomaterials for fabrication of NF composite membranes [24][25][26]. Membranes for nanofiltration applications were prepared from Chitosan, which contains quaternary amine groups.
In the present study, the chitosan membrane was prepared using Chitosan's biopolymer material, and modified Chitosan with a local natural source, and a palm tree. These membranes have attained great consideration due to their biodegradability, nontoxic, hydrophilicity, and high performance.

Materials and methods
Small molecular mass chitosan having a deacetylation extent around 82% and viscosity of 200-800 cP (1% in 1% acetic acid) was purchased from Sigma-Aldrich. Glacial acetic acid was obtained from Asia Pacific Specialty (APS), and Polysulphone supports PS-20, Sepro, USA. In addition, sodium chloride, magnesium chloride, sodium sulphate, calcium chloride , and magnesium sulphate were purchased from Sigma-Aldrich, USA. In this analysis, both solutions and dilutions are developed  Figure 1 shows the structure of chitosan material [27].

Palm preparation
Palm was grinded utilizing a crusher machine into different sized particles and then using a sieving shaker machine (FRITSCH, Germany) to use particles with a mesh size of less than 63 µm.

Preparing of chitosan and chitosan-palm solution
The chitosan solution was made by disintegration 0.5 g of Chitosan in 100 ml of diluted acetic acid. The solution was then heated at 70-80°C for 3-4 h and stirred until the Chitosan was fully soluble for 6 h. The prepared solutions were then kept overnight at room temperature to cool down. The solution was homogeneous and then isolated to eliminate any insoluble chitosan by a 5.0 µm hydrophobic polytetrafluoroethylene (PTFE) membrane. The final solution was divided into two groups following the filtration.

Fabrication of coating membrane
The membrane was fabricated using a custom-made unit and a rectangular polysulphone (PS) membrane. The solution Chitosan and Chitosan-Palm was deposited on the surface of the PS membrane (6.0 cm × 12.0 cm) and then was casted by using roll tools to make the  surface more homogenous, which was typically operated at room temperature (21°C). Membranes were washed with sodium hydroxide (NaOH) followed by Milli-Q water to remove the remaining acetic acid, and the surface of membranes becomes neutral. The membranes were then dried for 24-48 h in a temperaturecontrolled at 21°C ( Figure 2).

Surface morphology
Scanning electron microscope (FEI Nova-Nano SEM-600, Netherlands) was adopted to examine membrane surface morphology.

Contact-angle assessment
The membrane's contact angles are normally evaluated via a water droplet (2 µL) on the membrane surface owing to determine the hydrophobicity of the surface of the membrane. Using the sessile fall technique (Data Physics ® SCA20 Goniometer), which is set using a digital camera, the developed membranes contact angles are calculated. All samples were mounted flat on a microscope glass plate, and water droplets of 2-µL were found on the membrane layer.

Thermogravimetric analysis
The thermogravimetric analysis (TGA) of the Chitosan and chitosan-palm membranes was carried out using a Pyris 1 TGA system (PerkinElmer, USA). The TGA experiment conditions are as follow (scan rang room temperature up to 800°C), the heating rate maintained during the analysis was 10°C/min, and this experiment was carried out in the atmosphere of nitrogen gas using a ceramic pan at a rate of flow of 20 ml/min.

Zeta potential
A SurPASS electrokinetic analyzer was applied to estimate the charge of the surface of the membrane. Zeta potential (ZP) analysis was performed employing 10 mM KCl of the ambient electrolyte solution, and automated titration has been used to change the pH 3-10 using HCl and KOH solutions.

Permeability and salt-rejection
The NF/RO cross-flow method ( Figure 3) was used to test water permeability and salt-rejection attitudes for all membranes. At a pressure of 22 bar, Milli-Q ® water has been used for all membranes to compress the membranes before the filtration process. Membrane compression was performed for approximately 1 h until a stable baseline flux was achieved. The deionized water permeates flux were subsequently tested at various imposed pressures to determine water membranes' permeability. The salt rejections were examined, applying five electrolyte salt solutions such as sodium chloride, magnesium chloride, magnesium sulphate, calcium chloride, and sodium sulphate. Salt concentrations were kept 2 g/L individually. The feed solution's temperature was kept at 20 ± 2°C and the crossflow speed was set at 34.7 cm/s (100 L/h crossflow). The membrane thickness is equivalent to 40 cm 2 . The different salt concentrations in feed and water permeability were tested using a calibrated conductivity meter and pH meter. Figure 3 displays a diagram of the cross-flow filtration system. The percentage of salt rejection noted (R%) is determined using the following theory from the feed and permeate samples.
Where C p is permeated salt concentration and C f is feed salt concentrations correspondingly. The experiment was performed primarily at a crossflow rate of 34.7 cm/s by measuring the flux of water permeate, followed by salt aggregation in the feed solution to make the concentration of 2 g/L of salt. This step was evaluated at the various applied pressures i.e. 6 and 22 bar. To estimate the effect of solution in salt rejection of pH, the solution's pH was stepwise raised to pH 10 by the addition of little quantity of 1 M KOH and then a step-by-step reduction in pH by 1 M HCl accumulation.

Surface morphology
The surface morphology of the Chitosan and chitosanpalm and Polysulphone support (PS-20) was carried out by scanning electron microscope (SEM) at a magnification of 50 k in Figure 4. The image of Chitosan and polysulphone's membrane has shown a smooth surface compared to the chitosan-palm membrane. The images have clearly shown that adding palm to Chitosan markedly changes the morphology of the membrane formed. In addition, the developed membranes' pore size increased compared to the images taken for the chitosan membrane. In the production of membranes with large pore size, the flux is expected to be higher as well. Therefore, it is confirming that larger pore size is affecting both the flux rate and salt rejection.

Contact angle assessments
Contact angles of water droplets on the nanofiltration membranes surfaces containing polysulphone, Chitosan, and chitosan-palm were determined. Measured contact angles for the Chitosan and chitosan-palm membranes are displayed in Table 2. As observed from the results, the contact angles were 89.26 o for chitosanpalm and 83.46 o for chitosan membrane. The contact angle of both the developed membranes is less than 90 o , which means that they suffer less from membrane fouling in water treatment. It is ascribed to the membrane surface hydration by water molecules, which avoids foulant from direct interaction with membranes' surface. This hydrophilicity nature offers high polarity to the surface that attracts polar molecules such as water; no extra pressure is essential to initiate permeation. This may explain why the chitosan-palm has hydrophilic surface higher than Chitosan, and both membranes have a higher hydrophilicity surface than polysulphone due to the palm, which also has hydroxyl (OH) groups on the surface as shown in Figure 5 which comes from cellulose and hemicellulose that increase the hydrophilicity [27][28][29]. Figure 6 exhibits the thermogravimetric analysis of both developed membranes using Chitosan and chitosanpalm on a polysulphone as support. It is clearly indicating from Figure 6 that the thermal degradation of both membranes involved two main steps, the first degradation step 59-60% weight loss ensued at temperatures within the range of 375-475°C, and the other decomposition phase was noticed at temperatures in the range of 475-700°C. It is perceived that our attained outcomes are in covenant with those accomplished by other  inventors or researchers [30]. Remarkably, at 100-170°C temperature range in PS a minor degradation phase 3% weight loss was witnessed owing to the elimination of absorbed water. This degradation phase was perceived in PS only, and it was not exhibited in other developed chitosan membranes. In this developed, chitosan and chitosan-palm membranes are relatively more stable than PS because the peaks are produced by sulphone base and amide linkage. At this point, amide and sulphone linkage degradation was initiated only when the temperature was in high circumstances. The fabricated membranes exhibit high stability and the degradation temperatures were around 475 ± 10°C. This stability level in the polymer chain might be accredited to the existence of sulphonic acid groups. These conclusions are reliable with those described by other research scholars in this field [31]. The different degradation phase near 525 ± 3°C was initiated by cross-linking and polymerization processes in the chitosan membrane development [32]. Also, Figure 6(a) displays the residual weights of the membrane after thermal degradation at 800°C. The amount of remained weights was found to be 20-23.1% of the initial sample masses.

Zeta potential analysis
The zeta potential of a membrane is a conceptual model of the electrical potential of its membrane surface. By studying the pH impact on the membrane's zeta potential, it is possible to achieve its membrane surface's overall acidity or fundamentality. The developed chitosan and chitosan-palm membranes surface charges were estimated by using zeta potential. The outcomes are drawn in terms of pH, as displayed in Figure 6(b). The outcomes exhibited that the membranes were neutral around pH from 4 to 8 in both the developed chitosan membranes and polysulphone. The chitosanpalm membrane was marginally negative after pH 6 and marginally positive below pH 4; the negative charge was very high noticed in the polysulphone. Those findings distinctly are shown in Figure 6(b), that they have no ionizable chitosan membrane attached moieties. The considerably positive and negative charge at lower pH than four and pH higher than eight correspondingly are perhaps owed to the dispersant proton gain and proton loss process of the Chitosan, chitosan-palm membranes residuals in the membrane matrix [33][34][35]. These outcomes agree with the earlier studies [36], which exhibited that chitosan isoelectric point is neutral around pH 8.6. Rashid et. al., developed a membrane with charges higher than this isoelectric point of pH 8.6 [24]. Zhan et al. [37] developed CNT-CS BP membranes adopting an identical method and described the isoelectric point at pH amount to 5, within the range of pH 4-8 i.e. conveyed in our present work.

Water permeability
As mentioned in the experimental part, the developed membranes' water permeability is examined and exhibited in Table 2 and Figure 7. The surface area is about 40 cm 2 , where Figure 7 is drawn to permeate flux vs. applied pressure bar of the membrane. The permeability of water was determined from the linear correlation slopes in the permeate flux and the pressure applied. The Chitosan and chitosan-palm membrane water permeate flux significantly increases with higher applied pressure [15]. Chitosan-palm membrane displayed the permeability of water after 8 h of process around 3.04 ± 0.12 L m −2 h −1 bar −1 . This water permeability is nearly two times more than the Chitosan membranes water permeability i.e. 1.68 ± 0.04 L m −2 .h −1 .bar −1 . This confirms that the palm plays a major role in pore size and permeate flux in chitosanpalm membrane. An increase in pore size affects the increase in permeate flux correspondingly [30]. Besides, that palm increases the hydrophilicity on the surface because of carboxylic and hydroxide groups on cellulose's chemical structure [27] as shown in Figure 6.

Salt-rejection capability
Tables 3 and 4 exhibit the salt rejection of the two developed membranes on polysulphone support (Chitosan, chitosan-palm). The salt rejection was examined with five NaCl, MgCl 2 , MgSO 4 , CaCl 2 , and Na 2 SO 4 inorganic electrolyte solutions using the system as displayed in Figure 3. Such tests were performed as a single salt solution, 2 g/L of salt concentrations were maintained, and varying pressures were applied for 8 h at a temperature of 20 ± 2°C, and at pH equal to 7. Generally, an increase in permeate flux causes an increase in salt rejection. In chitosan membrane, MgCl 2 salt rejection (7-90%) was the maximum among the five electrolyte solution  chitosan membranes. This MgCl 2 salt rejection value was nearby more than 42% compared to NaCl rejection and more than 13 and 6 times of NaS 2 O 4 and MgSO 4 correspondingly. On the other hand, chitosan-palm was exhibiting a similar behaviour of chitosan membrane for all salts. It is illustrious that Na 2 SO 4 rejection in two membranes was quite low i.e. 5-16%, as exhibited in Figures 8(a and b). The accomplished divalent salts MgSO 4 and Na 2 SO 4 low salt rejections might be described by the non-existence of membrane surface charge [22]. These results indicated that both developed membranes under the working conditions follow the rank of MgCl 2 > CaCl 2 > NaCl > MgSO 4 > Na 2 SO 4 . These outcomes are associated with the salt rejection efficiency of nanofiltration polyamide membranes, in which the divalent salts salt rejection is more than the monovalent salts [16,38,39]. It is important that the membranes fabricated from polyamide carried a negative charge near pH equal to 7. In the case of salt rejection of cations and anions, they are most affected by the electrostatic interaction, and this plays a huge part in salt rejection. In this study, the chitosan membrane is neutral. Thus, Columbia interaction is not predicted as a main mechanism of rejection. The order of rejection is mentioned here agrees with the descending Kielland [40] rank of Cl − (0.19 nm) > Na + (0.1 nm) > Mg 2+ (0.09 nm) unhydrated ionic radius. The SO 4 2− unhydrated ionic radius does not exist in the literature. The surface charge of the fabricated membranes at pH 7 is neutral, which could explain the size impact for the higher rejection rate of monovalent ions (Cl − ) than for multi-valent ions (SO 4 2− ). These present outcomes are reliable to the earlier investigation by Tongwen et al [41]. Tongwen et al. exhibited that the rejection of various electrolytes was present in the sequence of MgCl 2 > NaCl > MgSO 4 . This mentioned order and surface negative charge are associated similarly with our current work. The separation of MgCl 2 could be higher in their study of a positively charged membrane than that of Na 2 SO 4 and NaCl, as stated by Rios et al. [42]. Correspondingly, the membrane charging offers the separation reliance on electrolyte valence. The salt rejection by both charged membranes, which mostly results from the surface contact between the membrane and the ions.

PH effect
The salt rejection of NaCl, MgSO 4 and MgCl 2 .6H 2 O of Chitosan and chitosan-palm membranes was examined at the temperature of 20°C within the range of pH from 3 to 10, and at a pressure of 22 bar. The salt rejection efficiency as a pH function was shown in Figure 9(a and b) and Table 5. The feed solutions pH turn into acidic caused to be increased in the salt rejection  of all membranes. The NaCl, MgSO 4 , and MgCl 2 .6H 2 O salt rejection increase in both membranes concerning smaller pH might result from the protonation of the available amino moieties in Chitosan [24]. The earlier investigations stated that NH 2 on Chitosan might gain proton at small pH. Consequently, the NH 3 + moiety on the Chitosan might play a significant part because the group is mainly accountable for interactions with anions and negatively charged surfaces [43,44]. Furthermore, Figure 9(a) also exhibits that the chitosan membrane had a smaller higher salt rejection than the chitosan-palm membrane. This might be endorsed to the charge repulsion, which might lead to greater salt rejection rates depending on the steric interactions. Also, it might be evident that MgCl 2 salt rejection for both membranes was greater amongst the other salts types, essentially at low and neutral pH. This might be owing to unhydrated ionic radius of Mg 2+ (0.09 nm), which is lower than the other ionic radius [40]. A comparison of the salt rejection results stemmed from the previous works [45][46][47][48][49] using different membranes with chitosan-palm membrane for nanofiltration is illustrated in Table 6. In summary, the data obtained shows that the manufactured membrane offers new exciting opportunities for nanofiltration membranes.

Conclusion
Chitosan and Chitosan with palm membranes coated on polysulphone as support have been developed and characterized about their properties to investigate their water purification application. The SEM results confirmed that the palm was played a role on Chitosan's surface and the contact angle of the chitosan-palm membrane is high 89.26°compared to the chitosan membrane 83.46°. In comparison, the water permeability for membrane chitosan with palm was 3.04 ± 0.12 L m −2 h −1 bar −1 , i.e. two times greater than the average value of the permeability of water of chitosan membrane 1.68 ± 0.04 L m −2 h −1 bar −1 . These data  indicated that the performances of water permeate flux and salt rejection were greater than the unmodified membrane. Moreover, the salt rejection was improved (from 5% for NaCl to 70% for MgCl 2 at the same condition). Therefore, these findings have shown that the chitosan and chitosan-palm membranes produced are more significant for water treatment implementations.