Int. J. Nanoparticles, Vol. 9, No. 2, 2017 71 Microalgae harvesting of Nannochloropsis sp. using polyethersulphone/lithium chloride/functionalised multiwall carbon nanotube membranes fabricated via temperature induced phase inversion and non-solvent induced phase inversion Nur Farahah Mohd Khairuddin, Ani Idris*, Muhammad Irfan and Teo Chee Loong Department of Bioprocess Engineering, Faculty of Chemical and Energy Engineering, c/o Institute of Bioproduct Development (IBD), Universiti Teknologi Malaysia, 81310 UTM, Johor Bharu, Malaysia Fax: +607-5588166 Email: firstname.lastname@example.org Email: email@example.com Email: firstname.lastname@example.org Email: email@example.com *Corresponding author Abstract: Microalgae are a promising renewable source especially for producing biofuels and other high value products. Biofuel and biomass production involves several steps from cultivation, harvesting and extraction. Recent technology has used ultrafiltration membrane for microalgae harvesting but are faced with adverse effect of microalgae fouling. In the present study, Nannochloropsis sp. harvesting was performed using an ultrafiltration hybrid membrane made of polyethersulphone (PES) blended with the two additives namely lithium chloride (LiCl) and functionalised multiwall carbon nanotube (F-MWCNT). The membranes were prepared via two methods; non-solvent induced phase separation (NIPS) and temperature induced phase separation (TIPS). The membrane performances were evaluated in terms of membrane flux and fouling for the use of microalgae harvesting. The FESEM analysis showed that the morphology of the hybrid PES-MWCNT-LiCl membranes was very much influenced by the phase separation method. Flux rates of the membranes improved dramatically with increasing amount of additives when prepared using TIPS. A 100% microalgae separation from cultivated solution without major fouling (more than 80% flux recovery) was successfully achieved. Keywords: membrane; multiwall carbon nanotube; lithium chloride; nanochloropsis sp. Reference to this paper should be made as follows: Khairuddin, N.F.M., Idris, A., Irfan, M. and Loong, T.C. (2017) ‘Microalgae harvesting of Nannochloropsis sp. using polyethersulphone/lithium chloride/functionalised multiwall carbon nanotube membranes fabricated via temperature induced phase inversion and non-solvent induced phase inversion’, Int. J. Nanoparticles, Vol. 9, No. 2, pp.71–87. Copyright © 2017 Inderscience Enterprises Ltd. 72 N.F. Khairuddin et al. Biographical notes: Nur Farahah Mohd Khairuddin received her Diploma in Environmental Engineering from Faculty of Environmental Engineering, University of Kuala Lumpur, Malaysia in 2008, BEng in Bioprocess Engineering from Faculty of Chemical Engineering, University of Surrey in 2010 and MEng in Safety, Health, and Environment Engineering from Universiti Teknologi Malaysia, Malaysia. She is currently doing her PhD in Bioprocess Engineering in the same university. She is currently in her final stage of research on membrane fabrication for microalgae harvesting. Her major research interest is in membrane separation. Other than that her research interests include polymer and nanoparticle, environmental application of nanotechnology, and water and waste water treatment. Ani Idris is currently a Professor in the Department of Bioprocess Engineering, Faculty of Chemical Engineering at the Universiti Teknologi Malaysia for 23 years and has carried out researches in the field of membrane technology and fermentation, nanotechnology and ferrofluids. She has published over 100 papers relating to her research area. She has also won several awards. Her research centres on the membrane technology includes making of reverse osmosis membranes/ultrafiltration membranes/nanofiltration membranes/thin composite membranes, downstream separation process using membranes for desalination purposes, treatment of waste water, and also biological process streams. Currently, her research area has diversified to biotechnology involving fermentation processes in lactic acid, biomass and microalgae using both free and immobilised cells. Muhammad Irfan had been working in the processing industry in Pakistan for almost nine years before deciding to continue his PhD in Bioprocess Engineering in Universiti Teknologi Malaysia (UTM), Malaysia in 2012. His experience in chemical handling during his working days has helped him a lot in his study. He is currently in his final stage of research. His major research interest is on membrane fabrication. Along his PhD research he has published a few journals on dialysis membrane fabrication in the high impact journals. His interest on membrane fabrication process has led him to work with UTM research centre producing membrane for water filtration for water processing company. Teo Chee Loong received his degree in Biology from Faculty of Science in 2012 and PhD in Bioprocess Engineering from Faculty of Chemical Engineering, Universiti Teknologi Malaysia in 2015. He is currently doing his post-doctoral fellowship in Universiti Teknologi Malaysia under Faculty of Chemical Engineering. His major research interest is in biodiesel production from microalgae. He has an excellent PhD journey due to his few successful publications in high impact journals including a short-communication article. He has invented the fastest transesterification process of lipid to biodiesel via simultaneous cooling and heating of microwave and had won gold medal in UTM innovation competition in 2015. 1 Introduction Microalgae are a promising renewable source that can help world to reduce pollution, carbon footprint and serve better environment for future needs. The first algae-based biofuel in action has been reported in Hamburg, Germany where a system named Microalgae harvesting of Nannochloropsis sp. 73 ‘Algenol’s technology’ has transformed the algae in ethanol to heat up the buildings in Germany. This proves that there is a bright future ahead awaiting for the algae industry. Production of biofuel or other high value added from microalgae involves several steps from selecting microalgae species, cultivating, harvesting, lipid extraction and further downstream processes before obtaining the required final product. Selection of an appropriate microalgae species will result in higher yields and reduce cost of downstream processes. Nannochloropsis sp. and Chlorella sp. are among the microalgae that have high lipid content (Ahmad et al., 2011; Amaro et al., 2011). Microalgae can be harvested using several methods including membrane filtration. The membrane is the heart to the harvesting process that can provide purified products. Membrane filtration has an advantage over centrifugation due to its cost effectiveness, energy efficiency and ease in scaling up (Ahmad et al., 2014). Membrane filtration can remove the majority of water from the media and leave a minimal amount of water for the centrifugation process (Bilad et al., 2014b). Furthermore, membrane process can be applied as a standalone process or a hybrid process by coupling membrane process to a photobioreactor to either supply carbon dioxide into a microalga broth or to retain the biomass (Bilad et al., 2014a). These advantages have made membrane as an effective method for microalgae harvesting. However membrane filtration can be uneconomical if there is fouling by organic/inorganic substances which requires high energy for pumping and frequent membrane replacement (Lam and Lee, 2012; Cheng and Timilsina, 2011). Furthermore, cleaning membrane can be time consuming especially when there is a major fouling and most cleaning processes involves chemical agents. In reality, fouling phenomena is inevitable and occurred in many membrane processes. Thus, in order to make membrane as an effective separation system for future harvesting, the fouling issue needs to be overcome. Recently researchers have modified the membrane filtration system to reduce fouling during filtration. Among these modifications involved the implementation of submerged and vibrating system in the filtration system. A vibration generated from an engine connected to a membrane house would create shear-stress on the liquid-membrane interface thus reduced fouling occurrence (Bilad et al., 2013). Besides that, a control of hydrodynamic condition particularly on the filtration velocity and transmembrane pressure could also improve anti-fouling performance by membrane. It was reported that high filtration velocity and low pressure have reduced the membrane fouling by reducing the chance of microalgae cells to aggregate on the membrane surface (Ahmad et al., 2012). Recent researches have also involved modification of polymer membrane by introducing hydrophilic substances into the membrane matrices for harvesting (Drexler and Yeh, 2014). Such as a hydrophilic PVDF membrane has been successfully embedded with some hydrophilic additives namely pluronic, polyethyleneimine (PEI) and polyethylene glycol (PEG). Result has revealed that 100% Chlorella sp. had been separated using the fabricated membrane with harvesting rate of 96 L/m2h and minor fouling (Hwang et al., 2015). Polyethersulphone (PES) is one of the common polymers used for membrane making due to its high chemical resistant and mechanical strength including good miscible property with hydrophilic additives. Either inorganic or organic additives both can mix well in PES polymer solution (Zhao et al., 2013). Generally, a membrane can be prepared 74 N.F. Khairuddin et al. via non-solvent induced phase separation (NIPS) and temperature induced phase separation (TIPS). Membrane preparation via NIPS are based on compositional change while membrane prepare via TIPS are based on temperature change. In comparison to NIPS, the TIPS process has many advantages because it can be applied to a wide range of polymers mixture and can be used to fabricate any type of membrane since it depends primarily on heat transfer (Dongmei et al., 2006). Based on our knowledge, no research was reported on the use of TIPS for fabrication of hybrid PES-functionalised multiwall carbon nanotube (F-MWCNT) lithium chloride (LiCl) membranes for microalgae harvesting. Thus, in this study PES was mixed with F-MWCNT and then hybridised with LiCl additive. The polymer solution was then cast to form membranes via TIPS and its performance in terms of flux rate and fouling was then compared with membranes fabricated via NIPS process. 2 Materials and methods 2.1 Materials Ultrason PES was purchased from BASF in flakes form and LiCl was purchased from Merck Milipore with grade ACS Reagent, Ph Eur. F-MWCNT, an acid functionalised multiwall carbon nanotube was obtained from a local supplier. Dimethyl acetamide (DMAC) was obtained from QReC, grade AR. Distilled water was used as the non-solvent in the coagulation bath. Nannochlropsis sp. was the chosen microalgae which were also been cultivated in our laboratory. 2.2 Membrane preparation The membrane was prepared by blending 18 wt% of PES, 4 wt% of LiCl in two different concentrations of F-MWCNT (0.2 wt% and 1 wt%). According to the literature, TIPS process normally involves high polymer concentration (40–50 wt%) because it is capable of dissolving high polymer concentrations at high temperature meanwhile NIPS process is usually limited to 20 wt% of polymer (Xiao et al., 2015). However in this experiment, the PES concentration was fixed at 18 wt% for both TIPS and NIPS process so that comparison can be made. Both TIPS and NIPS membranes were identified as T and N respectively and the detail compositions are listed in Table 1. Prior to dope preparation, PES and F-MWCNT were first dried in an oven at 60°C overnight to remove any moisture present. Then, the PES and the additives were dissolved in DMAC at temperature above 90°C in a 3-necked flask equipped with a stirrer and a heater. The stirrer speed was kept at 300 rpm so as not to create bubbles as they were difficult to remove from the homogenous solution. For TIPS preparation, the solution was casted immediately on a glass plate upon complete dissolution of polymer and additives and then immersed inside the coagulation bath at room temperature. For NIPS, the homogenous solution was allowed to cool to room temperature first before casting and then immersed in the coagulation bath. The prepared membranes were stored in distilled water until use. Microalgae harvesting of Nannochloropsis sp. Table 1 75 Composition of hybrid PES-FMWCNT-LiCl membranes and their porosity and pore size radius Membrane no. Polymer and additives composition Porosity (ε) Mean pore size radius (µm) 0 0.68 0.0051 4 0.2 0.83 0.013 4 1 0.97 0.019 PES LiCl MWCNT TM0 18 4 TM0.2 18 TM1 18 NM0 18 4 0 0.77 0.0076 NM0.2 18 4 0.2 0.88 0.015 NM1 18 4 1 0.67 0.0094 2.3 Membrane characterisation The membrane morphologies were characterised using scanning electron microscope (SEM) (Philips XL 40). Samples were immersed into liquid nitrogen and then fractured using forceps so as to obtain clean-cut cross sections of samples. The samples were sputtered with gold before putting inside SEM. Compositions of the membrane were determined using the same SEM instrument through energy dispersive x-ray (EDX) software. The FTIR spectra of functionalised and pristine MWCNT were obtained using a Perkin Elmer FTIR spectrometer. Surface structure of the membrane was measured using atomic force microscopy (AFM) (Park XE) instrument. Contact mode was used in AFM measurement and the scanning area of the membrane was approximately 10 µm × 10 µm. The AFM measurements were depicted in terms of maximum height of the surface (Rz), root mean square roughness (Rq) and maximum depth of valleys (Rpv). Wettability of the membranes was determined using contact angle (CAM 200). The contact angle measurement was performed using dynamic sessile drop. 2 µL of distilled water was dropped onto the dry membrane surface using a micro-syringe, and the contact angle was measured. At least five readings of contact angles were taken to obtain the average value. The mechanical strength of the membranes was analysed using tensile testing machine (LRX Tensile Testing Machine – 2.5 kN) with an operating head load of 0.5 N. The samples were placed between the grips of the testing machine. The grip length was 5 cm and the speed of testing was set at the rate of 5 mm/min. 2.4 Pure water flux, porosity and pore size of the membranes The flat sheet ultrafiltration module was used to filter pure water and harvest the microalgae. The flux (J) of the membrane was calculated using equation (1). J= V At (1) where V is the volume of permeate collected (L), A is the effective area of the membrane (m2) and t is the permeation time (h). The porosity (ε) of the prepared membrane was determined using equation (2). 76 N.F. Khairuddin et al. ε= w1 − w2 A × l × dw (2) where w1 is the weight of the wet membrane (g), w2 is the weight of the dry membrane (g), l is the thickness of the membrane (m) and dw is the density of water (0.998 g/cm3). The membrane mean radius pore size was determined using Geurout-Elford-Ferry equation. rm = (2.9 − 1.75ε ) × 8 μlQ ε × A × ΔP (3) where µ is the water viscosity (8.9 × 10–4 Pa·s), Q is the pure water permeation rate (m3/s), ΔP is trans membrane pressure (150 kPa). Both membrane radius pore size and porosity information are included in Table 1. 2.5 Preparation of Nannochloropsis sp. culture Nannochloropsis sp. was obtained from the culture collection of Borneo Marine Research institute (BMRI), Universiti Malaysia Sabah, Malaysia. The Nannochloropsis sp. was cultivated in sterilised seawater enriched with Walne’s medium which contains: 100 g NaNO3, 1.3 g FeCl3·6H2O; 0.36 g MnCl2·4H2O; 33.6 g H3BO3; 45 g Na2·EDTA; 20 g NaH2PO4·2H2O; 2.1 g ZnCl2; 2 g CoCl2·6H2O; 0.9 g (NH4)6Mo7O24·4H2O; 2 g CuSO4·5H2O; 0.001 g vitamin B12; 0.001 g vitamin B1 and 0.2 µg biotin per litre. 500 ml of Nannochloropsis sp. was cultivated in 1,000 mL flasks at 23 ± 0.5°C, pH 8 ± 0.2 under a light intensity of 100 µmol m–2 s–1. 2.6 Ultrafiltration of Nannochloropsis sp. The Nannochloropsis sp. culture was used as a feed without any pretreatment. A cross flow filtration was used to filter the Nannochloropsis sp. culture at a constant pressure of 300 kPa to force the feed solution through the membrane. Flux and samples were measured and collected at a time intervals of 10 min. Initial and final concentration of microalgae was determined using UV-Vis. Cleaning experiments were conducted to determine membrane fouling. The fouling membranes was backwashed with water for 10 minutes. Then the membranes were subjected again to the same feed solution containing nannochloropsis sp. Flux recovery ratio (FRR), total fouling ratio (Ft), reversible fouling ratio (Fr) and irreversible fouling ratio (Fir) of the microalgae harvesting were calculated using equations (4), (5), (6) and (7). J w2 × 100 J w1 (4) Ft (%) = J w1 − J h × 100 J w1 (5) Fr (%) = J w2 − J h × 100 J w1 (6) FRR (%) = Microalgae harvesting of Nannochloropsis sp. Fir = Ft − Fr 77 (7) where Jw1 and Jw2 are the initial and final pure water fluxes and Jh is the flux of harvesting. 3 Results and discussion 3.1 Membrane properties 3.1.1 Surface characteristics The membrane surface characteristics of membranes have been determined by membrane contact angle and AFM. The smaller contact angle value indicates higher hydrophilicity. Figure 1 shows that all PES membranes were hydrophilic because the contact angle values are smaller than 90°. The contact angle of membranes reduces when amount of F-MWCNT was increased indicating that its presence improves the hydrophilic property of the membrane. Moreover, membrane prepared by only LiCl additive was also hydrophilic supported by Idris et al. (2010) but its degree of hydrophilicity was further improved with the presence of F-MWCNT. This is due to the effect of hydrophilic functional groups of the acid F-MWCNT namely, C=O and O-H which creates hydrogen bonds and attracts water molecules (Vatanpour et al., 2011). Figure 1 Relationship between contact angle and surface roughness measurements of PES/LiCl membranes with different MWCNT additive amount as a function of TIPS and NIPS (see online version for colours) 78 Figure 2 N.F. Khairuddin et al. AFM images of membrane surface roughness, (a) NM0 (b) TM0 (c) NM0.2 (d) TM0.2 (e) NM1 (f) TM1 (see online version for colours) (a) (b) (c) (d) (e) (f) The results of AFM are depicted in Figure 2 and Table 2. Results showed that the higher F-MWCNT concentration the rougher the membrane surface. TM1 membranes have slightly rougher surface than NM1 membranes. The results of membrane contact angle and AFM were then correlated and are shown in Figure 1. It was observed that the higher contact angle has lower surface roughness. The results were in line with Wenzel theory (Wenzel, 1936). The author stated that increasing surface roughness will increase the Microalgae harvesting of Nannochloropsis sp. 79 surface wettability based on the assumption that the liquid penetrates into the valley of chemically homogenous surfaces. His work was apparently agreed later by a few findings (Hirose et al., 1996; Yana et al., 2006; Rahimpour et al., 2010). In this experiment, both TIPS and NIPS membranes with 1 wt% F-MWCNT exhibited high surface roughness and low contact angle (high wettability) which reflected the Wenzel theory. The NIPS and TIPS membranes of 1 wt% F-MWCNT have the highest wettability and were also the roughest. The smoothest surface was demonstrated by membranes without F-MWCNT which has the lowest wettability (less hydrophilic). Table 2 Surface roughness of the fabricated membranes Surface roughness parameters Membrane NM0 NM0.2 NM1 TM0 TM0.2 TM1 Rz (nm) 536.93 590.00 806.88 209.34 260.25 824.30 Rq (nm) 46.56 103.01 123.32 27.91 55.56 107.05 Rpv (nm) 541.93 598.54 816.26 216.98 268.31 844.12 However, it seems that 1 wt% F-MWCNT membrane fabricated using TIPS has an advantage over the 1 wt% F-MWCNT membrane fabricated using NIPS when it demonstrated smoother surfaces and yet higher hydrophilic properties; (50.83°, 82.38 nm and 57.61°, 88.12 nm respectively). In this case, the higher hydrophilicity of TIPS membranes was believed due to the rapid demixing during phase separation process that forces more F-MWCNT to move up the membrane layer. Meanwhile, smoother surface was associated with more nucleation and growth during the phase separation. The rapid cooling rates involved in TIPS resulted in a decrease in spherulite diameter owing to an increased nucleation density and thus a smoother surface was established (Dongmei et al., 2006). The advantage of having smooth surface is to reduce chances of fouling (Li et al., 2007). 3.1.2 FTIR and EDX analysis The FTIR results obtained in Figure 3 proved that the F-MWCNT nanoparticles used contains hydrophilic functional groups. The new peaks developed between 3,600 to 3,200 cm–1 in both spectra reading of F-MWCNT nanoparticle and F-MWCNT membranes confirmed the presence of strong O-H bonds. Moreover, the new peak developed at 600 cm–1 indicating C-Cl bonds showed that F-MWCNT also assist in retaining the chlorine ion. LiCl is a hygroscopic material that has strong interaction with water and would tend to diffuse out from membrane matrices during phase inversion (Pabby et al., 2015). However, EDX results in Table 3 shows that membrane with F-MWCNT can entrap the salt additive. The membranes with higher F-MWCNT concentration produced via TIPS contain more Cl– (0.84 atom wt%) and it is believed that rapid nucleation and aggregation of molecules inside polymer have entrapped more Cl– atoms inside the membrane matrices. The EDX results are in agreement with FTIR results because it also determined the presence of Cl– inside the membranes with F-MWCNT. 80 N.F. Khairuddin et al. Figure 3 FTIR spectra of pristine PES membrane, PES/F-MWCNT/LiCl membrane, pristine MWCNT nanoparticle and F-MWCNT nanoparticle (see online version for colours) Table 3 Chemical components of membranes by EDX analysis Cross-section atom (wt%) Chemical component NM0 NM0.2 NM1 TM0 TM0.2 TM1 Carbon 73.29 73.22 74.90 73.38 74.42 74.28 Oxygen 19.20 19.51 18.17 19.39 17.63 17.40 Sulphur 7.48 7.15 6.82 6.71 7.60 7.48 Chlorine 0.03 0.11 0.11 0.02 0.35 0.84 3.1.3 Membrane morphology SEM images of the cross section of the PES membrane containing 0.2 and 1 wt% F-MWCNT prepared via TIPS and NIPS processes are shown in Figure 4. The images confirmed that the membranes are asymmetric with skin layer and macrovoids sub-layer. The membranes without F-MWCNT showed different structures where TIPS membrane (TM0) possessed longer well defined finger-like structure and smaller macrovoids while the NIPS membrane (NM0) has shorter finger like structure and larger macrovoids. Membranes containing of 0.2 wt% F-MWCNT fabricated by TIPS exhibited a slightly different structure with very fine finger-like structures and fine voids form (almost a sponge like structure). However when 1 wt% F-MWCNT was introduced, the membrane cross-section structure exhibited very fine well-arranged long finger-like structure with little macrovoids. Meanwhile, the membranes produced using NIPS showed very little structural changes when F-MWCNT up to 1 wt% was added. All the NIPS membranes possess short finger-like structure. The finger-like structure was not elongated further although 1 wt% F-MWCNT was added probably due to the effect of LiCl that increased Microalgae harvesting of Nannochloropsis sp. 81 the viscosity of membrane solution combined with no temperature difference to induce separation. Figure 4 SEM cross-section images (400x magnificence) of TIPS and NIPS membranes with various concentrations of F-MWCNT, (a) NM0 (b) TM0 (c) NM0.2 (d) TM0.2 (e) NM1 (f) TM1 (see online version for colours) (a) (b) (c) (d) (e) (f) Generally, addition of LiCl in polymer solution would produce membrane with macrovoids structure due to liquid-liquid (L-L) and solid liquid (S-L) separation between homogenous LiCl, polymer and solvent (Zheng et al., 2016). Thus, the separation mechanism has created macropores in the membrane structure exhibited by membrane with only LiCl additive. As can be observed, small amount of F-MWCNT as such 82 N.F. Khairuddin et al. 0.2 wt% has hindered the growth of finger-like structure in TIPS membrane; whereby the small amount of F-MWCNT particles has slowed down the rate of L-L demixing. However further addition of 1 wt% of F-MWCNT has promoted and elongated the formation of finger-like structure due to the significant effect from hydrophilic F-MWCNT combined with temperature effect that accelerated the L-L and S-L separation rate in the coagulation bath. It has been reported that the acceleration of separation that resulted in the formation of macropores and high porosity was strongly influenced by TIPS process instead of the hydrophilic effect of additives. TIPS process causes rapid nucleation that attracts polymer molecules to aggregate to each other thus lead to pore formation (Pan et al., 2015). 3.1.4 Membrane mechanical properties The mechanical property of the membranes was determined at the tensile rate of 5 mm/min using LRX tensile testing machine and the tensile test parameters which include load, strength and extension at yield are shown in Table 4. These parameters initially increased with the presence of F-MWCNT. For NIPS membranes, the mechanical properties improved when the F-MWCNT amount was increased. This behaviour indicated that the F-MWCNT nanoparticle can improve the strength of the membrane. However, the membrane strength reached the maximum value when F-MWCNT was 0.2 wt% for TIPS membranes and declined as the amount of F-MWCNT increased. The reason can be explained by the structure of TM1 which consisted of many fine long finger like structure throughout the membrane thickness. Thus it was much easier to break compared to the TM0.2 membrane structure that was spongy. Table 4 Mechanical strength of the membranes Tensile test 1 Load at yield (N) 2 Strength at yield (N/m) 3 Extension at yield (mm) NIPS TIPS NM0 NM0.2 NM1 TM0 TM0.2 TM1 3.18 4.16 8.24 3.56 5.90 3.37 318.30 416.29 823.68 355.77 589.92 337.39 0.28 0.41 3.34 0.28 3.80 0.34 3.2 Membrane performances 3.2.1 Membrane water permeation Figure 5 shows pure water permeation of TIPS and NIPS membranes. As illustrated, membranes consisting of F-MWCNT nanoparticles have higher pure water flux than membrane without the nanoparticles. Among these fabricated membranes, the 1 wt% F-MWCNT TIPS membrane showed the highest pure water permeation. For TIPS membranes, the presence of 0.2 wt% and 1 wt% in polymer solution increased the water flux gradually. However, for NIPS membrane, the maximum flux achieved was at 0.2 wt% and then decreased as amount of F-MWCNT increased to 1 wt%. Microalgae harvesting of Nannochloropsis sp. Figure 5 83 Water permeation of membrane with different F-MWCNT nanoparticles content (see online version for colours) According to the results, it was observed that at 1 wt% F-MWCNT for both the NIPS and TIPS membranes, porosity and membrane structure are dominant factors which influenced the water permeation. The NIPS membrane containing 1 w% F-MWCNT (NM1) was the most hydrophilic yet it did not exhibit good permeation rates. Meanwhile for both NIPS and TIPS membrane containing 0.2 wt F-MWCNT, both membrane structure and hydrophilic factors influenced the water permeation. Therefore it can be reported that membrane structure is also an influential factor that can facilitate water to pass through the membrane. Although the hydrophilic properties of F-MWCNT nanoparticles improved the hydrophilicity of the membrane surface and increased the membrane permeability by attracting water molecules into it but yet water permeation depends on the morphological property of the membrane. 3.2.2 Harvesting performance of the membrane In this study, all membranes achieved 100% rejection indicated by the nil concentration of microalgae in permeates. Figure 6 and Table 5 demonstrate membrane permeate rate and fouling properties. Membranes fabricated using TIPS containing 0 to 1 wt% F-MWCNT showed increased in permeation rates as F-MWCNT increased. NIPS membranes on the other hand showed an increase in permeation rate when F-MWCNT was increased to 0.2 wt% and then decreased upon further addition of F-MWCNT to 1 wt%. The profile of the microalgae permeation rates follows a similar trend to that of pure water permeation. In addition, Figures 6 depicts that as permeation flux increased, the FRR decreased. Thus this explains that the high flux has led to higher fouling. Antifouling properties can be assessed by considering the Ft, Fr and Fir in Table 5. Lower Ft means better antifouling property of the membrane. The Ft of membranes without F-MWCNT were lower than the other membranes. Less and no fouling occurred on these membranes was believed due to the persistent lower flux within tested time frame that prevented it from further microalgae fouling. Evident for this statement was shown by the photo images in Figure 7 which demonstrates that membrane with higher flux (TM1) is fully covered whilst NM0 and TM0 membranes are not fully covered by microalgae sludge after ten minutes of harvesting. 84 Figure 6 N.F. Khairuddin et al. Membrane flux and flux recovery rate of microalgae harvesting (see online version for colours) Table 5 Fouling properties; total fouling (Ft), reversible fouling (Fr) and irreversible fouling (Fir) of NIPS and TIPS membranes Fouling properties Membrane NM0 NM0.2 NM1 TM0 TM0.2 TM1 Figure 7 Ft (%) Fr (%) Fir (%) 0.00 71.34 16.61 8.87 44.14 69.78 0.00 50.92 16.61 8.87 17.71 52.54 0.00 20.42 0.00 0.00 56.34 17.24 Microalgae fouling on a flat sheet membrane with area of 5.31 cm2 after 10 min of filtration of (a) 0 wt%, (b) 2 wt%, (c) 5 wt% and (d) 1 wt% membranes (see online version for colours) TIPS NIPS (a) (b) (c) (d) Microalgae harvesting of Nannochloropsis sp. 85 As fouling is inevitable, the best membrane was the one that has high permeation rates and minimal fouling propensity. Thus in this study, the excellent antifouling and permeation rates was exhibited by the 1 wt% F-MWCNT TIPS (TM1) membrane with more than 80% FRR. The addition of 1 wt% hydrophilic F-MWCNT and the use of TIPS process have increased the hydrophilicity and antifouling properties of the membrane. It was believed that hydrogen bonding between F-MWCNT with water layer has reduced the microalgae fouling on the surface of the membrane. It was proven that both 0.2 wt% TIPS and NIPS membrane that consist minimal amount of F-MWCNT experienced high irreversible flux due to the very small amount of hydrophilic F-MWCNT. 4 Conclusions In summary, the hybrid PES-LiCl-F-MWCNT membranes are suitable for microalgae harvesting indicated by the 100% rejection of microalgae in permeate. The presence of F-MWCNT inside the polymer solution has enhanced the attachment of Cl– into the polymer structure and would be the reason of low fouling. Cl atom has negative ion charges that has probably made the membrane negatively charged and thus favour anti-fouling. However, this claim needs for further research in our future work. The combination of temperature and additive effects have further increased the permeation rate and reduced fouling. TIPS which involved temperature change governed the membrane properties and simultaneously improved the performance of hydrophilic PES-F-MWCNT-LiCl membranes. Besides improving the permeation flux the presence of 1 wt% F-MWCNT in TIPS membrane exhibited antifouling behaviour indicated by the more than 80 wt% flux recovery. Acknowledgements The authors are grateful to the Universiti Teknologi Malaysia and funding source (Vote No. 4F236) Ministry of Higher Education (MOHE) Malaysia. References Ahmad, A.L., Yasin, N.H.M., Derek, C.J.C. and Lim, J.K. (2011) ‘Microalgae as a sustainable energy source for biodiesel production: a review’, Renewable and Sustainable Energy Reviews, Vol. 15, No. 1, pp.584–593. Ahmad, A.L., Yasin, N.H.M., Derek, C.J.C. and Lim, J.K. (2012) ‘Crossflow microfiltration of microalgae biomass for biofuel production’, Desalination, Vol. 302, pp.65–70. Ahmad, A.L., Yasin, N.H.M., Derek, C.J.C. and Lim, J.K. (2014) ‘Comparison of harvesting methods for microalgae Chlorella sp. and its potential use as a biodiesel feedstock’, Environmental Technology, Vol. 35, No. 17, pp.244–253. Amaro, M., Guedes, A.C. and Malcata, F.X. (2011) ‘Advances and perspectives in using microalgae to produce biodiesel’, Applied Energy, Vol. 88, No. 10, pp.3402–3410. 86 N.F. Khairuddin et al. Bilad, M.R., Arafat, H.A., Vankelecom, I.F.J. (2014a) ‘Membrane technology in microalgae cultivation and harvesting: a review’, Biotechnology Advances, Vol. 32, No. 7, pp.1283–1300. Bilad, M.R., Discart, V., Vandamme, D., Foubert, I., Muylaert, K. and Vankelecom, I.F.J. (2014b) ‘Coupled cultivation and pre-harvesting of microalgae in a membrane photobioreactor (MPBR)’, Bioresource Technology, Vol. 155, pp.410–417. Bilad, M.R., Discart, V., Vandamme, D., Foubert, I., Muylaert, K. and Vankelecom, F.J. (2013) ‘Harvesting microalgal biomass using a magnetically induced membrane vibration (MMV) system: filtration performance and energy consumption’, Bioresource Technology, Vol. 138, pp.329–338. Cheng, J.J. and Timilsina, G.R. (2011) ‘Status and barriers of advanced biofuel technologies: a review’, Renewable Energy, Vol. 36, No. 12, pp.3541–3549. Dongmei, L., Krantz, W.B., Greenberg, A.R. and Sania, R.L. (2006) ‘Membrane formation via thermally induced phase separation (TIPS): model development and validation’, Journal of Membrane Science, Vol. 279, Nos. 1–2, pp.1–2. Drexler, I.L.C. and Yeh, D.H. (2014) ‘Membrane applications for microalgae cultivation and harvesting: a review’, Review Environmental Science Biotechnology, Vol. 13, No. 4, pp.487–504. Hirose, M., Ito, H. and Kamiyama, Y. (1996) ‘Effect of skin layer surface structures on the flux behaviour of RO membranes’, Journal of Membrane Science, Vol. 121, No. 2, pp.209–215. Hwang, T., Kotte, M.R., Han, J.I., Oh, Y.K. and Diallo, M.S. (2015) ‘Microalgae recovery by ultrafiltration using novel fouling-resistant PVDF membranes with in situ PEGylated polyethyleneimine particles’, Water Research, Vol. 73, pp.181–192. Idris, A., Ahmed, I. and Limin, M.A. (2010) ‘Influence of lithium chloride, lithium bromide and lithium fluoride additives on performance of polyethersulfone membranes and its application in the treatment of palm oil mill effluent’, Desalination, Vol. 250, No. 2, pp.805–809. Lam, M.K. and Lee, M.K. (2012) ‘Microalgae biofuels: a critical review of issues, problems and the way forward’, Biotechnology Advances, Vol. 30, No. 3, pp.673–690. Li, Q., Xu, Z. and Pinnau, I. (2007) ‘Fouling of reverse osmosis membranes by biopolymers in wastewater secondary effluent: role of membrane surface properties and initial permeate flux’, Journal of Membrane Science, Vol. 290, Nos. 1–2, pp.173–181. Pabby, A.K., Rizvi, S.S.H. and Requen, A.M.S. (2015) Handbook of Membrane Separations: Chemical, Pharmaceutical, Food, and Biotechnological Applications, 2nd ed., RCS Publication, New York. Pan, J., Xiao, C., Huang, Q., Wang, C. and Liu, H. (2015) ‘Fabrication and properties of poly(ethylene chlorotrifluoroethylene) membranes via thermally induced phase separation (TIPS)’, Royal Society of Chemistry, Vol. 5, No. 56, pp.45249–45257. Rahimpour, A., Jahanshahi, M., Mortazavian, N., Madaeni, S.S. and Mansourpanah, Y. (2010) ‘Preparation and characterization of asymmetric polyethersulfone and thin-film composite polyamide nanofiltration membranes for water softening’, Applied Surface Science, Vol. 256, No. 6, pp.1657–1663. Vatanpour, V., Madaeni, S., Moradian, R., Zinadini, S. and Astinchap, B. (2011) ‘Fabrication and characterization of novel antifouling nanofiltration membrane prepared from oxidized multiwalled carbon nanotube/polyethersulfone nanocomposite’, Journal of Membrane Science, Vol. 375, Nos. 1–2, pp.284–294. Wenzel, R.N. (1936) ‘Resistance of solid surfaces to wetting by water’, Industrial and Engineering Chemistry, Vol. 28, No. 8, pp.988–994. Xiao, T., Wang, P., Yang, X., Caia, X. and Lu, J. (2015) ‘Fabrication and characterization of novel asymmetric polyvinylidene fluoride (PVDF) membranes by the nonsolvent thermally induced phase separation (NTIPS) method for membrane distillation applications’, Journal of Membrane Science, Vol. 489, pp.160–174. Microalgae harvesting of Nannochloropsis sp. 87 Yana, L., Li, Y.S., Xiang, C.B. and Xianda, S. (2006) ‘Effect of nano-sized Al2O3-particle addition on PVDF ultrafiltration membrane performance’, Journal of Membrane Science, Vol. 276, Nos. 1–2, pp.162–167. Zhao, C., Xue, J., Ran, F. and Sun, S. (2013) ‘Modification of polyethersulfone membranes of methods – a review’, Progress in Materials Science, Vol. 58, No. 1, pp.76–150. Zheng, L., Wu, Z., Wei, Y., Zhang, Y., Yuan, Y. and Wang, Y. (2016) ‘Preparation of PVDF-CTFE hydrophobic membranes for MD application: effect of LiCl-based mixed additives’, Journal of Membrane Science, Vol. 506, pp.71–85.