- Email: [email protected]
A comparative study of microfiltration and ultrafiltration for algae harvesting
Algal Research 2 (2013) 437–444
Contents lists available at ScienceDirect
Algal Research journal homepage: www.elsevier.com/locate/algal
A comparative study of microfiltration and ultrafiltration for algae harvesting Xuefei Sun a, Cunwen Wang a,⁎, Yanjie Tong a, Weiguo Wang a, Jiang Wei b,⁎ a b
Wuhan Institute of Technology, Xiongchu Avenue 693, Wuhan, PR China Alfa Laval Nakskov A/S, Stavangervej 10, DK-4900 Nakskov, Denmark
a r t i c l e
i n f o
Article history: Received 19 March 2012 Received in revised form 26 June 2013 Accepted 20 August 2013 Available online 13 September 2013 Keywords: Algae Microfiltration Ultrafiltration Harvesting Fouling
a b s t r a c t The present work deals with the filtration and concentration of algae (Chlorella) from a diluted culture medium using six commercial microfiltration membranes (MFP2, MFP5 and MFP8 with different pore sizes) and ultrafiltration membranes (FS40PP, FS61PP and ETNA10PP with different Molecular Weight Cut-Off (MWCO)). The effects of the operating conditions, e.g. feed solution temperature, TMP (transmembrane pressure), VCF (volume concentration factor) and cross-flow velocity on the filtration performance were investigated. The results showed that permeate fluxes increased with the increase in feed solution temperature, and the fluxes were probably limited by released extracellular polymeric substances (EPS) at higher temperatures. The permeate fluxes increased slowly with increasing TMP up to a certain limit, and after that the fluxes were stable or even decreased. The higher cross-flow velocity can significantly decrease particles accumulating on the surface of membrane, and thus leading to higher permeate flux. Although ETNA10PP exhibited much less fouling than other membranes, the permeate flux of this membrane was not higher than other membranes most likely due to the fact that this membrane is the ‘tightest’ membrane with MWCO 10,000. The performance of UF and MF membranes was compared for this application. The interesting finding of our work is that microfiltration and ultrafiltration showed very similar performance in terms of permeate flux under the same operation conditions at low TMP. © 2013 Elsevier B.V. All rights reserved.
1. Introduction In recent years, there has been increasing interest for the production of biofuels recognizing algae biomass as the raw material [1,2]. The production of biofuels through microalgae has not only attended to the quest for renewable energy source, it also has enormous commercial potential due to the growth rates of microalgae [3]. Microalgae can be cultivated in seawater [4], saline–alkali water [5], agricultural sewage [6] and industrial wastewater [7–9]. More recently, sources of woody material (Lignocellulose hydrolysates) have been considered to be an attractive feedstock for microalgae cultivation, which are the most widespread sources of carbon in nature. However, the harvest of microalgae biomass is still a major problem because of the small size of algae cells and low biomass concentration. Although conventional methods, such as flocculation, flotation and centrifugation have been used as processes for effective removal of microalgae biomass from culture medium, there are still some problems remaining during practical operations. For example, chemical flocculents like alum and ferric chloride were used to harvest microalgae. However, chemical flocculation has not been used for large operations ⁎ Corresponding authors. E-mail addresses: [email protected] (C. Wang), [email protected] (J. Wei). 2211-9264/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.algal.2013.08.004
[10]. Usually, flotation was used in combination with flocculation for algae harvesting, but the cost of front flotation was estimated to be too high for commercial use [11]. Centrifugation and drying are currently considered too expensive due to low content biomass of the culture media. Membrane technologies have been used for the removal of bacteria, viruses and other microorganisms [12]. As manufacturing techniques improve and the range of applications expands, the cost of membranes and membrane systems have steadily decreased, which may make it possible to use membrane technology for microalgae harvesting. Most importantly, membrane filtration can achieve complete removal of algae from the culture media [12]. Different membrane filtration technologies have been used for the removal or concentration of microalgae. Zhang [13] evaluated the feasibility of using a cross-flow membrane ultrafiltration process to harvest and dewater algae suspension, and the microalgae was concentrated 150 times and final algae concentration reached 154.85 g/L. Hung [14] studied how operating parameters affect microfiltration and examined the effect of preozonation on flux behavior when using hydrophobic and hydrophilic membranes. Zou [15] investigated the effect of physical and chemical parameters on forward osmosis (FO) fouling during algae separation. In addition, the effect of solute reverse diffusion on FO fouling was systematically studied. Pressure-driven microfiltration (MF) and ultrafiltration (UF) membrane processes are prone to fouling and are relatively energy intensive,
438
X. Sun et al. / Algal Research 2 (2013) 437–444
Table 1 Membrane type and characteristics. Membrane process
Type
Pore size
pH
Pressure, (bar)
Temperature (°C)
Material
MF
MFP2 MFP5 MFP8 FS40PP FS61PP ETNA10PP
0.2 0.45 0.8 MWCO = 100,000 MWCO = 20,000 MWCO = 10,000
1–12 1–12 1–12 1–11 1–11 1–11
1–10 1–10 1–10 1–10 1–10 1–10
0–75 0–75 0–75 0–75 0–75 0–75
Fluoro polymer Fluoro polymer Fluoro polymer Fluoro polymer Fluoro polymer Composite Fluoro polymer
UF
while the FO membrane process showed a very low permeate flux [16]. There were a few reports concerning comparison of MF and UF for microalgae filtration. Chow et al. [16]. compared microfiltration and ultrafiltration methods and found both techniques attractive for removal of cyanobacterial cells. Rossignol [17] compared MF and UF technologies for continuous filtration of microalgae. The results showed that, although the pure water fluxes of microfiltration membrane were higher, during separation of microorganisms, fluxes of the ultrafiltration membrane became higher than microfiltration membrane. The effectiveness of membrane separation is greatly affected by fouling. It can be further explained that the accumulation of microorganisms on membrane surface or in membrane pores causes decline in permeate flux [18]. Many efforts have been made to understand and reduce fouling, including membrane surface modification and new membrane material development [19,20]. Conventional polymeric materials membranes have been widely used in filtration and concentration of microalgae [13,21–23]. Rossignol [24] evaluated the performances of inorganic filtration membranes. Liu [25] utilized a thin, porous metal sheet membrane to harvest microalgae, which exhibited high properties of membrane area packing density, chemical stability, thermal stability, mechanical strength, high permeability and low cost. The purpose of our work is to compare the performance of microfiltration and ultrafiltration for algae harvesting by using microfiltration (MF) membranes with different pore size and ultrafiltration (UF) membranes with different MWCO. All 6 types of the membranes used are Polyvinylidene Fluoride (PVDF) based, and ETNA10PP is a surface modified PVDF membrane. ETNA10PP is the only membrane with hydrophilic surface [26], which is supposed to show lower fouling tendency. Our intention is to investigate the influence of membrane materials (hydrophobic versus hydrophilic), membrane pore size, and porosity on performance. We have studied how operating parameters affect MF and UF filtration. MF and UF experiments were carried out separately including 3 kinds of membranes in each test. Then, the performance of the microfiltration membrane (MFP8) and ultrafiltration membrane (FS40PP) were compared in the same test for the filtration of Chlorella. The effect of VCF (Volume Concentration Factor = Total starting feed volume / retentate volume) on permeate flux was also studied during the concentration process of Chlorella.
membrane module M10 (a small lab-scale membrane module). Performance of different membranes can be compared according to the permeate flux and cell retention. The membrane characteristics are shown in Table 1. 2.3. Experimental set-up The schematic diagram of the membrane module is shown in Fig. 1. The membrane module consists of four plates kept together with four bolts. The module contains four flat-sheet membrane samples operating in series, with each having an effective filtration area of 0.0084 m2. Inlet (Pin) and outlet pressures (Pout) are measured with pressure transducers (D) and (F) mounted on the inlet and outlet of the membrane module. The transmembrane pressure (TMP) was calculated as TMP = (Pin + Pout) / 2-Ppermeate. A diluted Chlorella culture medium was kept in the feed tank (G). The membrane filtration was performed in a batch mode operation with recycling of permeate and retentate back to the feed tank to simulate a continuous operation. The permeate flow rate was measured by measuring the collected permeate in a 500 ml cylinder over a time of 60 s. The flux data were measured 2 times to get an average value for each measurement. The total test time for each membrane test was 4.5 h. After each experiment, the M10 module was cleaned with cleaning agents Ultrasil 10 (from Ecolab) for approximately half an hour at 55 °C. 3. Results and discussion 3.1. Effect of temperature In most microfiltration and ultrafiltration processes, permeate flux increases with increasing feed solution temperature [27]. The effect of temperature on permeate flux may be attributed not only to the effect of temperature on the physical properties (viscosity, solubility, etc.) of
2. Materials and methods 2.1. Microalgal suspensions Chlorella pyrenoidosa FACHB-9 cells were cultivated in an open cultivation system, provided by Algae Innovation Center of Denmark. The fresh cultures were taken in the middle of the exponential growth phase. Then algae cells were placed in a refrigerator and stored under darkness at 4 °C. The pH of the culture was 9.0 ± 0.5. In order to compare the performance of the tested membranes, all comparative experiments have been carried out with the same cell concentration level, 0.68 g/L. 2.2. Membrane characteristics Different commercial MF and UF membranes from Alfa Laval Nakskov A/S were used in the experiments, using Alfa Laval's cross-flow
Fig. 1. Schematic diagram of experimental system, showing feed (A), cooling/healing (B), pump (C), pressure (D), permeate (E), pressure (F), retentate (G), control value (H).
X. Sun et al. / Algal Research 2 (2013) 437–444
feed suspension [28], but also to the complex physical changes that may be occurring in the membrane as the temperature is changed [29]. For the solution of Chlorella with relatively high cell density, the impact of temperature on the permeate flux becomes particularly complex. Temperature plays an important role in the release of EPS (extracellular polymeric substances), which accumulates on the membrane surface and causes the flux to decline [30]. Fig. 2 shows the effect of temperature on the permeate flux in microfiltration and ultrafiltration of Chlorella solution. The temperature of the feed suspension was varied while transmembrane pressure and cross-flow were kept constant at 1.3 bar, 3.86 m/s (microfiltration), and 2.3 bar, 7.72 m/s (ultrafiltration), respectively. In this process, the temperature of the feed suspension ranges from 20 °C to 28 °C, which is within the normal temperature range of the growth of Chlorella. As Fig. 2 demonstrates, membrane permeate flux is sensitive to changes in feed solution temperature. When the solution temperature is 20 °C, the viscosity is higher and the diffusion coefficient is lower, resulting in a relatively low permeate flux. With increasing temperature, the flux of the MF and UF membranes also increases. However, as the temperature increases from 24 °C to 28 °C, the permeate fluxes of all MF
439
membranes were similar to each other. It is possible that higher temperatures favor the metabolism of the Chlorella, and thus concentrations of extracellular polymeric substances (EPS), e.g., proteins and nucleic acids increase in the feed solution [31]. These substances could adsorb the membrane surface, leading to the permeate flux decreasing. The optimum temperature for filtration was found to be 24 °C, at which point the best growth state of Chlorella was observed. The change in permeate flux for all membranes shows similar patterns. Typically flux versus time curves show a relatively rapid flux decline in the first 2 h of the process, followed by a more gradual decrease, until a steady-state flux has been reached. 3.2. Effect of transmembrane pressure (TMP) Fig. 3 shows the variation of flux with time under different transmembrane pressures. In most cases, an increase in pressure leads to an increase of the permeate flux. However, with the microfiltration membranes only a slight increase was observed as the transmembrane pressure increased from 1.3 bar to 1.8 bar. Similar results can be seen for ultrafiltration. The fluxes increased significantly from TMP of
Fig. 2. Effect of temperature on permeate flux in cross-flow microfiltration (MFP2, MFP5 and MFP8) and ultrafiltration (FS40PP, FS61PP and ETNA10PP). Filtration conditions: TMP = 1.3 bar, cross-flow = 3.86 m/s (microfiltration); and TMP = 2.3 bar, cross-flow = 7.72 m/s (ultrafiltration).
440
X. Sun et al. / Algal Research 2 (2013) 437–444
Fig. 3. Effect of transmembrane pressure on permeate flux in cross-flow microfiltration (MFP2, MFP5 and MFP8) and ultrafiltration (FS40PP, FS61PP and ETNA10PP). Filtration conditions: T = 20 °C, cross-flow = 5.79 m/s (microfiltration); and T = 24 °C, cross-flow = 7.72 m/s (ultrafiltration).
1.3 bar to 1.8 bar for UF membranes, but not from 1.8 bar to 2.3 bar. Therefore we can assume that there is an optimal pressure, after which further increase in transmembrane pressure will not improve flux. Higher permeate fluxes were observed at the beginning of both microfiltration and ultrafiltration processes, but then the permeate fluxes declined rapidly. The permeate fluxes declined more rapidly with increasing transmembrane pressure. As shown in Fig. 3, although the initial permeate fluxes of the ultrafiltration membrane at 2.3 bar was higher than the flux at 1.8 bar, the decline rate of the permeate flux was faster at 2.3 bar. The permeate flux at 2.3 bar decreased even further below the flux at 1.8 bar. Such a phenomenon has already been observed with other biological suspensions (bacteria, apple juice, etc.) due to the presence of polysaccharides in the feed solution [32–34]. Chlorella cells can attach to the membrane surface, which can be seen by visually checking the fouled membrane surfaces. The attached cells could release a secretion and EPS [30], which might be enhanced at higher transmembrane pressure. The higher pressure can add additional resistance to permeation by compressing the Chlorella cells and EPS into a thicker and denser fouling layer. Further, according to Makardij [35], at high transmembrane pressure, the membrane pore
size and EPS layer porosity decrease, resulting in increase of the cake layer and hence, more rapid flux decline. 3.3. Effect of cross-flow velocity The cross-flow velocity is another important parameter which has an influence on microfiltration and ultrafiltration performance. Fig. 4 indicates the effects of cross-flow on performance of the membranes. As the cross-flow velocity increased, the permeate fluxes increased, suggesting that Chlorella and other particles were prevented from accumulating on the surface of membrane. During the initial stage of Chlorella filtration, the permeate fluxes are seen to be independent of cross-flow velocity. As shown in Fig. 4, the initial permeate fluxes of microfiltration and ultrafiltration (FS40PP and FS61PP) are nearly identical under different cross-flow velocities. However, as the cake resistance increased after the first two-hour process, cross-flow velocity showed a more pronounced effect on permeate fluxes, whereby the steady-state fluxes increased with the cross-flow velocity. This can be explained by less particles depositing onto the membrane surface at high cross-flow velocity. ETNA10PP is a surface-modified PVDF
X. Sun et al. / Algal Research 2 (2013) 437–444
441
Fig. 4. Effect of cross-flow velocity on permeate flux in cross-flow microfiltration (MFP2, MFP5 and MFP8) and ultrafiltration (FS40PP, FS61PP and ETNA10PP). Filtration conditions: T = 20 °C, TMP = 1.3 bar (microfiltration); and T = 24 °C, TMP = 2.3 bar (ultrafiltration).
membrane, which can reduce fouling by rendering the membrane surface hydrophilic whereby it can be cleaned without using cleaning agents [26]. The results in Fig. 4 demonstrate that, after a 4-hour filtration run, an average drop of 10.5% in permeate flux for ETNA10PP membrane was observed, while flux drops for FS40PP and FS61PP were 40.7% and 33.3%, respectively. The membrane fouling degree of ultrafiltration is shown in Fig. 5. It is very obvious that the accumulated material (green in the photos) on the surface of ETNA10PP was much less than for FS40PP and FS61PP. 3.4. Direct comparison between microfiltration and ultrafiltration To compare the performance between microfiltration and ultrafiltration techniques, a longer period experiment (in recirculation mode) with total 48-hour run was carried out. The highest flux microfiltration
membrane (MFP8) and the highest flux ultrafiltration membrane (FS40PP) and ETNA10PP were tested under optimal operating conditions (5.79 m/s, 1.8 bar and 24 °C). These operating conditions were chosen to reduce Chlorella cell damage due to filtration and provide suitable surviving conditions for Chlorella. The surface-modified PVDF membrane ETNA10PP was used here to determine whether it has higher permeate flux compared with MFP8 and FS40PP in longer period tests due to its low fouling properties. Fig. 6 shows performance of this experiment during a 48-hour run. Significant differences were observed between the microfiltration and ultrafiltration membranes. The MF membrane exhibited a higher initial permeate flux than the UF membranes. However, a sharp decline in flux was observed for the MF membrane in the first three hours of the process, resulting in the permeate flux of MFP8 being similar or only slightly below FS40PP. Such a rapid drop of initial permeate flux for the MF membrane may be due to the higher initial
442
X. Sun et al. / Algal Research 2 (2013) 437–444
In this study, the volume concentration factor (VCF) is defined as: VCF ¼ V0 =ðV0 −Vt Þ
Fig. 5. Photos of used membranes after 4.5 h run: a. FS40PP; b. FS61PP; c. ETNA10PP. Green seen on the photos are foulant material. Red seen on ETNA10PP (c) is the color of membrane itself.
flux of the MF membranes which leads to higher fouling tendency. However, our results shown in Figs. 2, 3 and 4 do not indicate better performance of UF membranes over MF membranes as reported by Rossignol [17]. The discrepancy observed in the two studies could be due to different algal types, different membrane types and different õoperating conditions used by two groups. Membrane flux could be influenced by many parameters. All experiments we conducted showed very similar flux level for MF and UF membranes. Although ETNA10PP exhibited much lower membrane fouling because of its anti-fouling properties, the steady state flux was in fact still lower than MFP8 and FS40PP. It was hypothesized that the flux of this membrane in long period tests may be higher than other membranes due to its low fouling properties. However, the hypothesis was not proven in this experiment. 3.5. Concentration of Chlorella suspensions The above filtration experiments in recirculation mode were carried out to find the MF or UF membrane with the best performance (highest permeate flux and best rejection of Chlorella). The final goal of this work is to use MF or UF membranes for the concentration of Chlorella suspensions.
where V0 and Vt are the initial feed volume (32 L) and feed volume at time t, respectively. The relationship between VCF and permeation flux was studied. Experiments were performed with the microfiltration membrane (MFP8) and ultrafiltration membranes (FS40PP and ETNA10PP). The results in Fig. 7 demonstrated that permeate fluxes declined much faster at low Chlorella cell concentrations during the initial stages of the experiment. As the process continued, a stable flux was reached at higher Chlorella cell concentrations. It is believed that the cake layer (or fouling layer) probably became thicker and denser with increasing Chlorella concentration at high cell concentration range, and thus reduced permeate flux. A stable flux was observed when the cake layer did not change at higher cell concentrations. There is a larger flux drop for MFP8 than the other membranes at the early stages of the concentration, which can be explained by the larger pore size of the microfiltration membrane However, the steady-state flux seems to be similar for MFP8 and FS40PP, and the flux is slightly higher than for ETNA10PP. After 130 min of concentration, the feed volume in the batch tank was decreased from the initial 32 L to 2.8 L, resulting in a VCF of 11.4. Since MF membranes have much more open pore structure and much higher porosity than UF membranes, one would expect that MF membranes show much higher permeate flux for algae harvesting. The results in Figs. 6 and 7 show very similar performance of MFP8 (pore size of 0.8 μm) and FS40PP (MWCO 100,000), indicating that the pore size of membranes is not as an important parameter for this kind of application as originally hypothesized. A possible explanation for these results is that the fouling layer caused by deposition of algae cells and EPS was acting as a membrane selective layer [36]. The fouling layer can be clearly seen in Fig. 5 for FS40PP. The results in this work further suggest that the fouling layer, as a membrane selective layer, is similar for MF and UF membranes, implying that pore size and porosity have little influence on the formation of the fouling layer. Although ETNA10PP showed lower permeate flux than MFP8 and FS40PP in Figs. 6 and 7, the membrane exhibited very low fouling tendency as the permeate flux was rather stable during the whole filtration period. The low fouling tendency was further shown by very little deposition of foulants on the membrane surface, as seen in Fig. 5. In fact, the low fouling tendency of this membrane can be expected, owing to its hydrophilic membrane surface. The lower permeate flux of ETNA10PP can be considered to be due to the higher flow resistance of the membrane as this is a ‘tight’ (much smaller pore size) UF membrane. It may be
Fig. 6. Direct comparison between microfiltration (MFP8 and ultrafiltration membranes (FS40PP, ETNA10PP). Filtration conditions: T = 24 °C, TMP = 1.8 bar and cross-flow = 5.79 m/s.
X. Sun et al. / Algal Research 2 (2013) 437–444
443
Fig. 7. Permeate flux versus VCF for microfiltration (MFP8) and ultrafiltration (FS40PP, ETNA10PP) membranes. Filtration conditions: T = 24 °C, TMP = 1.8 bar and cross-flow = 5.79 m/s.
concluded from our work that membrane materials play a very important role for this application. Membranes with hydrophilic surfaces will most likely show better performance from the viewpoint of reducing fouling, whereas hydrophobic membranes will experience severe fouling for algae filtration.
thank Gary Lloyd for proof reading and corrections of the manuscript. This work was supported by the National Natural Science Foundation of China (Grant No. 20976140).
References 4. Conclusion The filtration and concentration of Chlorella from dilute culture media and the performance of several commercial MF and UF membranes were evaluated and compared. Flux increased with increasing temperature of the feed solution. However, above a certain temperature, further increase in temperature didn't improve permeate flux, most likely because of the release of EPS by Chlorella cells and/or deposition of Chlorella cells. A general trend of increased permeate flux with increasing TMP was observed. At higher TMP, however, permeate flux gradually leveled off or even dropped due to the effect of a fouling (cake) layer development. It was also seen that higher cross-flow velocity can significantly decrease particles accumulating on the membrane surface. Although ETNA10PP exhibited much better anti-fouling properties, the steady-state permeate flux of this membrane was not higher than for MFP8 and FS40PP in a direct comparison test. However, this is believed to be due to the much lower pore size i.e. MWCO (10,000) of ETNA10PP rather than limitation due to fouling layer buildup (see below). Furthermore, the concentration experiments indicate that the MF membrane did not show higher permeate flux than the UF membranes under the same operation conditions. MF membranes and UF membranes show similar flux in this work, indicating that pore size and porosity are not important for this application. This suggests that the permeate flux of different membranes is controlled by the fouling layer that acts as the membrane selective layer. Our work also demonstrated that a membrane with hydrophilic surface shows very little fouling for algae harvesting. The results of ETNA10PP suggest that membrane materials are the most important parameters for reducing fouling tendency. In future work, we will make analysis on the deposition layer (or EPS) attached to membrane surfaces to understand the membrane fouling better. Acknowledgments Xuefei Sun wishes to acknowledge the Alfa Laval Nakskov A/S for the support of the work. Authors would like to thank Jørgen Enggaard Boelsmand at the Algae Innovation Center of Denmark for providing algae suspensions and helpful discussions. We would also like to
[1] Q. Hu, C.W. Zhang, M. Sommerfeld, Biodiesel from algae: lessons learned over the past 60 years and future perspectives, Psychological Society of America Annual Meeting, University of Alaska Southeast, Juneau, Alaska, USA, 2006. [2] Q. Hu, M. Sommerfeld, E. Jarvis, M. Ghirardi, M. Posewitz, M. Seibert, A. Darzins, Microalgal triacylglycerols as feedstocks for biofuel production: perspectives and advances, Plant J. 54 (2008) 621. [3] Y.-T. Hung, Q.S. Amuda, A.O. Alade, Algae harvest energy conversion, Handb. Environ. Eng. 11 (723) (2010). [4] M. Chateaudegat, M. Chinain, N. Cerf, S. Gingras, B. Hubert, E. Dewaillyii, Seawater temperature, Gambierdiscus spp. variability and incidence of ciguatera poisoning in French Polynesia, Harmful Algae 4 (2005) 1053. [5] L.C. Bruce, J. Imberger, The role of zooplankton in the ecological succession of plankton and benthic algae across a salinity gradient in the Shark Bay solar salt ponds, Hydrobiologia 626 (2009) 111. [6] N.K. Singh, D.W. Dhar, Sewage effluent: a potential nutrient source for microalgae, Proc. Indian Natl. Sci. Acad. 72 (2006) 113. [7] S. Chinnasamy, A. Bhatnagar, R.W. Hunt, K.C. Das, Microalgae cultivation in a wastewater dominated by carpet mill effluents, Bioresour. Technol. 101 (2009) 3097. [8] T. Hirooka, Y. Akiyama, N. Tsuji, H. Naqase, K. Hirata, K. Miyamoto, Removal of hazardous phenols by microalgae under photoautotrophic conditions, J. Biosci. Bioeng. 95 (2003) 200. [9] E.B. Sydney, T.E. Da Silva, A. Tokarski, A.C. Novak, J.C. De Carvalho, A.L. Woiciecohwski, C. Larroche, C.R. Soccol, Screening of microalgae with potential for biodiesel production and nutrient removal from treated domestic sewage, Appl. Energy 88 (2011) 3291. [10] S. Thomas, The economics of micro-algae production and processing into biofuel faming systems, Department of Agriculture, Western Australia, 2006. [11] L.C. Neves, L.F.M. Silva, L.M.C. Cabral, S.G.F. Leite, V.M. Matta, Effect of flow rate and transmembrane pressure on permeate flux during ultrafiltration of residual brine from squid processing, 2nd Mercosur Congress on Chemical Engineering, 4th Mercosur Congress on Process Systems Engineering, 2005. [12] B. Sandhya, T. Satoshi, Microfiltration membrane fouling and cake behavior during algal filtration, Desalination 261 (2010) 46. [13] X.Z. Zhang, Q. Hu, M. Sommerfeld, E. Puruhito, Y.S. Chen, Harvesting algal biomass for biofuels using ultrafiltration membranes, Bioresour. Technol. 101 (2010) 5297. [14] M.T. Hung, C.J. Liu, Microfiltration for separation of green algae from water, Biointerfaces 51 (2006) 157. [15] S. Zou, Y.S. Gu, D.Z. Xiao, C.Y. Tang, The role of physical and chemical parameters on forward osmosis membrane fouling during algae separation, J. Membr. Sci. 366 (356) (2011). [16] C.K.W. Chow, S. Panglisch, J. House, M. Drikas, M.D. Bruch, R. Gimbeld, A study of membrane filtration for the removal of cyanobacterial cells, J. Water Supply Res Technol. 46 (1997) 324. [17] N. Rossignol, L. Vandanjon, P. Jaouen, F. Quemeneur, Membrane technology for the continuous separation microalgae/culture medium: compared performances of cross-flow microfiltration and ultrafiltration, Aquac. Eng. 20 (1991) 191. [18] A.L. Wei, G.M. Zeng, G.H. Huang, J. Liang, X.D. Li, Modeling of a permeate flux of cross-flow membrane filtration of colloidal suspensions: a wavelet network approach, Int. J. Environ. Sci. Technol. 6 (2009) 395.
444
X. Sun et al. / Algal Research 2 (2013) 437–444
[19] F.G. Meng, S.R. Chae, A. Drews, M. Kraume, H.S. Shin, F.L. Yang, Recent advances in membrane bioreactors (MBRs): membrane fouling and membrane material, Water Res. 43 (2009) 1489. [20] C.H. Zhang, F.I. Yang, W.J. Wang, B. Chen, Preparation and characterization of hydrophilic modification of polypropylene non-woven fabric by dip-coating PVA (polyvinyl alcohol), Sep. Purif. Technol. 61 (2008) 272. [21] L. Heng, Y.L. Yang, W.J. Gong, X. Li, G.B. Li, Effect of pretreatment by permanganate/ chlorine on algae fouling control for ultrafiltration (UF) membrane system, Desalination 222 (2008) 74. [22] J.B. Castaing, A. Massé, M. Pontié, V. Séchet, J. Haure, P. Jaouen, Investigating submerged ultrafiltration (UF) and microfiltration (MF) membranes for seawater pre-treatment dedicated to total removal of undesirable micro-algae, Desalination 253 (2010) 71. [23] S. Babel, S. Takizawa, A study on membrane fouling due to algal deposition, Water Sci. Technol. 41 (2000) 327. [24] N. Rossi, I. Petit, P. Jaouen, P. Legentilhomme, M. Derouiniot, Harvesting of Cyanobacterium Arthrospira platensis using inorganic filtration membranes, Sep. Sci. Technol. 40 (2005) 3033. [25] W. Liu, Harvesting of Micro-algae by Membrane Filtration, The International Congress on Membranes and Membrane Processes, Amsterdam, 2011. [26] J. Wei, S.H. Gitte, C.W. Nick, Characterization of a non-fouling ultrafiltration membrane, Desalination 192 (2006) 252. [27] M. Kallioinen, M. Pekkarinen, M. Mänttäri, M. Nyström, N.J. Jutta, Stability of two different regenerated cellulose ultrafiltration membrane under varying solution temperature, Desalination 199 (2006) 204.
[28] Y. Magara, M. Itoh, The effect of operational factors on solid/liquid separation by ultra-membrane filtration in a biological denitrification system for collected human excreta treatment plants, Water Sci. Tech. 23 (1991) 1583. [29] F.A.G. Mattheus, S.S. Shyam, S.A.M. Salha, H.A.B. Rashid, W. Mark, Effect of feed temperature on permeate flux and mass transfer coefficient in spiral-wound reverse osmosis systems, Desalination 144 (2002) 367. [30] A. Drews, C.H. Lee, M. Kraume, Membrane fouling—a review on role of EPS, Desalination 200 (2006) 186. [31] S. Tsuneda, H. Aikawa, H. Hayashi, A. Yuasa, A. Hirata, Extracellular polymeric substances responsible for bacterial adhesion onto solid surface, FEMS Microbiol. Lett. 223 (287) (2003). [32] N. Rossi, M. Derouiniot-Chaplain, P. Jaouen, P. Legentilhomme, I. Petit, Arthrospira platensis harvesting with membranes: fouling phenomenon with limiting and critical flux, Bioresour. Technol. 99 (2008) 6161. [33] M.K. Darko, S.L. Markov, N.T. Miodrag, Membrane fouling during cross-flow microfiltration of Polyporus squamosus fermentation broth, Biochem. Eng. J. 9 (2001) 103. [34] G.T. Vladisavljevic, P. Vukosavljevic, B. Bukvic, Permeate flux and fouling resistance in ultrafiltration of depectinized apple juice using ceramic membranes, J. Food Eng. 60 (2003) 241. [35] A.A. Makardij, Impact of operating parameters on flux decline in microfiltration and ultrafiltration. Thesis The University of Auckland, Auckland, New Zealand, 2002. [36] N. Thang, F.A. Roddick, L.H. Fan, Biofouling of water treatment membranes: a review of the underlying causes, monitoring techniques and control measures, Membranes 2 (2012) 804.
Contents lists available at ScienceDirect
Algal Research journal homepage: www.elsevier.com/locate/algal
A comparative study of microfiltration and ultrafiltration for algae harvesting Xuefei Sun a, Cunwen Wang a,⁎, Yanjie Tong a, Weiguo Wang a, Jiang Wei b,⁎ a b
Wuhan Institute of Technology, Xiongchu Avenue 693, Wuhan, PR China Alfa Laval Nakskov A/S, Stavangervej 10, DK-4900 Nakskov, Denmark
a r t i c l e
i n f o
Article history: Received 19 March 2012 Received in revised form 26 June 2013 Accepted 20 August 2013 Available online 13 September 2013 Keywords: Algae Microfiltration Ultrafiltration Harvesting Fouling
a b s t r a c t The present work deals with the filtration and concentration of algae (Chlorella) from a diluted culture medium using six commercial microfiltration membranes (MFP2, MFP5 and MFP8 with different pore sizes) and ultrafiltration membranes (FS40PP, FS61PP and ETNA10PP with different Molecular Weight Cut-Off (MWCO)). The effects of the operating conditions, e.g. feed solution temperature, TMP (transmembrane pressure), VCF (volume concentration factor) and cross-flow velocity on the filtration performance were investigated. The results showed that permeate fluxes increased with the increase in feed solution temperature, and the fluxes were probably limited by released extracellular polymeric substances (EPS) at higher temperatures. The permeate fluxes increased slowly with increasing TMP up to a certain limit, and after that the fluxes were stable or even decreased. The higher cross-flow velocity can significantly decrease particles accumulating on the surface of membrane, and thus leading to higher permeate flux. Although ETNA10PP exhibited much less fouling than other membranes, the permeate flux of this membrane was not higher than other membranes most likely due to the fact that this membrane is the ‘tightest’ membrane with MWCO 10,000. The performance of UF and MF membranes was compared for this application. The interesting finding of our work is that microfiltration and ultrafiltration showed very similar performance in terms of permeate flux under the same operation conditions at low TMP. © 2013 Elsevier B.V. All rights reserved.
1. Introduction In recent years, there has been increasing interest for the production of biofuels recognizing algae biomass as the raw material [1,2]. The production of biofuels through microalgae has not only attended to the quest for renewable energy source, it also has enormous commercial potential due to the growth rates of microalgae [3]. Microalgae can be cultivated in seawater [4], saline–alkali water [5], agricultural sewage [6] and industrial wastewater [7–9]. More recently, sources of woody material (Lignocellulose hydrolysates) have been considered to be an attractive feedstock for microalgae cultivation, which are the most widespread sources of carbon in nature. However, the harvest of microalgae biomass is still a major problem because of the small size of algae cells and low biomass concentration. Although conventional methods, such as flocculation, flotation and centrifugation have been used as processes for effective removal of microalgae biomass from culture medium, there are still some problems remaining during practical operations. For example, chemical flocculents like alum and ferric chloride were used to harvest microalgae. However, chemical flocculation has not been used for large operations ⁎ Corresponding authors. E-mail addresses: [email protected] (C. Wang), [email protected] (J. Wei). 2211-9264/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.algal.2013.08.004
[10]. Usually, flotation was used in combination with flocculation for algae harvesting, but the cost of front flotation was estimated to be too high for commercial use [11]. Centrifugation and drying are currently considered too expensive due to low content biomass of the culture media. Membrane technologies have been used for the removal of bacteria, viruses and other microorganisms [12]. As manufacturing techniques improve and the range of applications expands, the cost of membranes and membrane systems have steadily decreased, which may make it possible to use membrane technology for microalgae harvesting. Most importantly, membrane filtration can achieve complete removal of algae from the culture media [12]. Different membrane filtration technologies have been used for the removal or concentration of microalgae. Zhang [13] evaluated the feasibility of using a cross-flow membrane ultrafiltration process to harvest and dewater algae suspension, and the microalgae was concentrated 150 times and final algae concentration reached 154.85 g/L. Hung [14] studied how operating parameters affect microfiltration and examined the effect of preozonation on flux behavior when using hydrophobic and hydrophilic membranes. Zou [15] investigated the effect of physical and chemical parameters on forward osmosis (FO) fouling during algae separation. In addition, the effect of solute reverse diffusion on FO fouling was systematically studied. Pressure-driven microfiltration (MF) and ultrafiltration (UF) membrane processes are prone to fouling and are relatively energy intensive,
438
X. Sun et al. / Algal Research 2 (2013) 437–444
Table 1 Membrane type and characteristics. Membrane process
Type
Pore size
pH
Pressure, (bar)
Temperature (°C)
Material
MF
MFP2 MFP5 MFP8 FS40PP FS61PP ETNA10PP
0.2 0.45 0.8 MWCO = 100,000 MWCO = 20,000 MWCO = 10,000
1–12 1–12 1–12 1–11 1–11 1–11
1–10 1–10 1–10 1–10 1–10 1–10
0–75 0–75 0–75 0–75 0–75 0–75
Fluoro polymer Fluoro polymer Fluoro polymer Fluoro polymer Fluoro polymer Composite Fluoro polymer
UF
while the FO membrane process showed a very low permeate flux [16]. There were a few reports concerning comparison of MF and UF for microalgae filtration. Chow et al. [16]. compared microfiltration and ultrafiltration methods and found both techniques attractive for removal of cyanobacterial cells. Rossignol [17] compared MF and UF technologies for continuous filtration of microalgae. The results showed that, although the pure water fluxes of microfiltration membrane were higher, during separation of microorganisms, fluxes of the ultrafiltration membrane became higher than microfiltration membrane. The effectiveness of membrane separation is greatly affected by fouling. It can be further explained that the accumulation of microorganisms on membrane surface or in membrane pores causes decline in permeate flux [18]. Many efforts have been made to understand and reduce fouling, including membrane surface modification and new membrane material development [19,20]. Conventional polymeric materials membranes have been widely used in filtration and concentration of microalgae [13,21–23]. Rossignol [24] evaluated the performances of inorganic filtration membranes. Liu [25] utilized a thin, porous metal sheet membrane to harvest microalgae, which exhibited high properties of membrane area packing density, chemical stability, thermal stability, mechanical strength, high permeability and low cost. The purpose of our work is to compare the performance of microfiltration and ultrafiltration for algae harvesting by using microfiltration (MF) membranes with different pore size and ultrafiltration (UF) membranes with different MWCO. All 6 types of the membranes used are Polyvinylidene Fluoride (PVDF) based, and ETNA10PP is a surface modified PVDF membrane. ETNA10PP is the only membrane with hydrophilic surface [26], which is supposed to show lower fouling tendency. Our intention is to investigate the influence of membrane materials (hydrophobic versus hydrophilic), membrane pore size, and porosity on performance. We have studied how operating parameters affect MF and UF filtration. MF and UF experiments were carried out separately including 3 kinds of membranes in each test. Then, the performance of the microfiltration membrane (MFP8) and ultrafiltration membrane (FS40PP) were compared in the same test for the filtration of Chlorella. The effect of VCF (Volume Concentration Factor = Total starting feed volume / retentate volume) on permeate flux was also studied during the concentration process of Chlorella.
membrane module M10 (a small lab-scale membrane module). Performance of different membranes can be compared according to the permeate flux and cell retention. The membrane characteristics are shown in Table 1. 2.3. Experimental set-up The schematic diagram of the membrane module is shown in Fig. 1. The membrane module consists of four plates kept together with four bolts. The module contains four flat-sheet membrane samples operating in series, with each having an effective filtration area of 0.0084 m2. Inlet (Pin) and outlet pressures (Pout) are measured with pressure transducers (D) and (F) mounted on the inlet and outlet of the membrane module. The transmembrane pressure (TMP) was calculated as TMP = (Pin + Pout) / 2-Ppermeate. A diluted Chlorella culture medium was kept in the feed tank (G). The membrane filtration was performed in a batch mode operation with recycling of permeate and retentate back to the feed tank to simulate a continuous operation. The permeate flow rate was measured by measuring the collected permeate in a 500 ml cylinder over a time of 60 s. The flux data were measured 2 times to get an average value for each measurement. The total test time for each membrane test was 4.5 h. After each experiment, the M10 module was cleaned with cleaning agents Ultrasil 10 (from Ecolab) for approximately half an hour at 55 °C. 3. Results and discussion 3.1. Effect of temperature In most microfiltration and ultrafiltration processes, permeate flux increases with increasing feed solution temperature [27]. The effect of temperature on permeate flux may be attributed not only to the effect of temperature on the physical properties (viscosity, solubility, etc.) of
2. Materials and methods 2.1. Microalgal suspensions Chlorella pyrenoidosa FACHB-9 cells were cultivated in an open cultivation system, provided by Algae Innovation Center of Denmark. The fresh cultures were taken in the middle of the exponential growth phase. Then algae cells were placed in a refrigerator and stored under darkness at 4 °C. The pH of the culture was 9.0 ± 0.5. In order to compare the performance of the tested membranes, all comparative experiments have been carried out with the same cell concentration level, 0.68 g/L. 2.2. Membrane characteristics Different commercial MF and UF membranes from Alfa Laval Nakskov A/S were used in the experiments, using Alfa Laval's cross-flow
Fig. 1. Schematic diagram of experimental system, showing feed (A), cooling/healing (B), pump (C), pressure (D), permeate (E), pressure (F), retentate (G), control value (H).
X. Sun et al. / Algal Research 2 (2013) 437–444
feed suspension [28], but also to the complex physical changes that may be occurring in the membrane as the temperature is changed [29]. For the solution of Chlorella with relatively high cell density, the impact of temperature on the permeate flux becomes particularly complex. Temperature plays an important role in the release of EPS (extracellular polymeric substances), which accumulates on the membrane surface and causes the flux to decline [30]. Fig. 2 shows the effect of temperature on the permeate flux in microfiltration and ultrafiltration of Chlorella solution. The temperature of the feed suspension was varied while transmembrane pressure and cross-flow were kept constant at 1.3 bar, 3.86 m/s (microfiltration), and 2.3 bar, 7.72 m/s (ultrafiltration), respectively. In this process, the temperature of the feed suspension ranges from 20 °C to 28 °C, which is within the normal temperature range of the growth of Chlorella. As Fig. 2 demonstrates, membrane permeate flux is sensitive to changes in feed solution temperature. When the solution temperature is 20 °C, the viscosity is higher and the diffusion coefficient is lower, resulting in a relatively low permeate flux. With increasing temperature, the flux of the MF and UF membranes also increases. However, as the temperature increases from 24 °C to 28 °C, the permeate fluxes of all MF
439
membranes were similar to each other. It is possible that higher temperatures favor the metabolism of the Chlorella, and thus concentrations of extracellular polymeric substances (EPS), e.g., proteins and nucleic acids increase in the feed solution [31]. These substances could adsorb the membrane surface, leading to the permeate flux decreasing. The optimum temperature for filtration was found to be 24 °C, at which point the best growth state of Chlorella was observed. The change in permeate flux for all membranes shows similar patterns. Typically flux versus time curves show a relatively rapid flux decline in the first 2 h of the process, followed by a more gradual decrease, until a steady-state flux has been reached. 3.2. Effect of transmembrane pressure (TMP) Fig. 3 shows the variation of flux with time under different transmembrane pressures. In most cases, an increase in pressure leads to an increase of the permeate flux. However, with the microfiltration membranes only a slight increase was observed as the transmembrane pressure increased from 1.3 bar to 1.8 bar. Similar results can be seen for ultrafiltration. The fluxes increased significantly from TMP of
Fig. 2. Effect of temperature on permeate flux in cross-flow microfiltration (MFP2, MFP5 and MFP8) and ultrafiltration (FS40PP, FS61PP and ETNA10PP). Filtration conditions: TMP = 1.3 bar, cross-flow = 3.86 m/s (microfiltration); and TMP = 2.3 bar, cross-flow = 7.72 m/s (ultrafiltration).
440
X. Sun et al. / Algal Research 2 (2013) 437–444
Fig. 3. Effect of transmembrane pressure on permeate flux in cross-flow microfiltration (MFP2, MFP5 and MFP8) and ultrafiltration (FS40PP, FS61PP and ETNA10PP). Filtration conditions: T = 20 °C, cross-flow = 5.79 m/s (microfiltration); and T = 24 °C, cross-flow = 7.72 m/s (ultrafiltration).
1.3 bar to 1.8 bar for UF membranes, but not from 1.8 bar to 2.3 bar. Therefore we can assume that there is an optimal pressure, after which further increase in transmembrane pressure will not improve flux. Higher permeate fluxes were observed at the beginning of both microfiltration and ultrafiltration processes, but then the permeate fluxes declined rapidly. The permeate fluxes declined more rapidly with increasing transmembrane pressure. As shown in Fig. 3, although the initial permeate fluxes of the ultrafiltration membrane at 2.3 bar was higher than the flux at 1.8 bar, the decline rate of the permeate flux was faster at 2.3 bar. The permeate flux at 2.3 bar decreased even further below the flux at 1.8 bar. Such a phenomenon has already been observed with other biological suspensions (bacteria, apple juice, etc.) due to the presence of polysaccharides in the feed solution [32–34]. Chlorella cells can attach to the membrane surface, which can be seen by visually checking the fouled membrane surfaces. The attached cells could release a secretion and EPS [30], which might be enhanced at higher transmembrane pressure. The higher pressure can add additional resistance to permeation by compressing the Chlorella cells and EPS into a thicker and denser fouling layer. Further, according to Makardij [35], at high transmembrane pressure, the membrane pore
size and EPS layer porosity decrease, resulting in increase of the cake layer and hence, more rapid flux decline. 3.3. Effect of cross-flow velocity The cross-flow velocity is another important parameter which has an influence on microfiltration and ultrafiltration performance. Fig. 4 indicates the effects of cross-flow on performance of the membranes. As the cross-flow velocity increased, the permeate fluxes increased, suggesting that Chlorella and other particles were prevented from accumulating on the surface of membrane. During the initial stage of Chlorella filtration, the permeate fluxes are seen to be independent of cross-flow velocity. As shown in Fig. 4, the initial permeate fluxes of microfiltration and ultrafiltration (FS40PP and FS61PP) are nearly identical under different cross-flow velocities. However, as the cake resistance increased after the first two-hour process, cross-flow velocity showed a more pronounced effect on permeate fluxes, whereby the steady-state fluxes increased with the cross-flow velocity. This can be explained by less particles depositing onto the membrane surface at high cross-flow velocity. ETNA10PP is a surface-modified PVDF
X. Sun et al. / Algal Research 2 (2013) 437–444
441
Fig. 4. Effect of cross-flow velocity on permeate flux in cross-flow microfiltration (MFP2, MFP5 and MFP8) and ultrafiltration (FS40PP, FS61PP and ETNA10PP). Filtration conditions: T = 20 °C, TMP = 1.3 bar (microfiltration); and T = 24 °C, TMP = 2.3 bar (ultrafiltration).
membrane, which can reduce fouling by rendering the membrane surface hydrophilic whereby it can be cleaned without using cleaning agents [26]. The results in Fig. 4 demonstrate that, after a 4-hour filtration run, an average drop of 10.5% in permeate flux for ETNA10PP membrane was observed, while flux drops for FS40PP and FS61PP were 40.7% and 33.3%, respectively. The membrane fouling degree of ultrafiltration is shown in Fig. 5. It is very obvious that the accumulated material (green in the photos) on the surface of ETNA10PP was much less than for FS40PP and FS61PP. 3.4. Direct comparison between microfiltration and ultrafiltration To compare the performance between microfiltration and ultrafiltration techniques, a longer period experiment (in recirculation mode) with total 48-hour run was carried out. The highest flux microfiltration
membrane (MFP8) and the highest flux ultrafiltration membrane (FS40PP) and ETNA10PP were tested under optimal operating conditions (5.79 m/s, 1.8 bar and 24 °C). These operating conditions were chosen to reduce Chlorella cell damage due to filtration and provide suitable surviving conditions for Chlorella. The surface-modified PVDF membrane ETNA10PP was used here to determine whether it has higher permeate flux compared with MFP8 and FS40PP in longer period tests due to its low fouling properties. Fig. 6 shows performance of this experiment during a 48-hour run. Significant differences were observed between the microfiltration and ultrafiltration membranes. The MF membrane exhibited a higher initial permeate flux than the UF membranes. However, a sharp decline in flux was observed for the MF membrane in the first three hours of the process, resulting in the permeate flux of MFP8 being similar or only slightly below FS40PP. Such a rapid drop of initial permeate flux for the MF membrane may be due to the higher initial
442
X. Sun et al. / Algal Research 2 (2013) 437–444
In this study, the volume concentration factor (VCF) is defined as: VCF ¼ V0 =ðV0 −Vt Þ
Fig. 5. Photos of used membranes after 4.5 h run: a. FS40PP; b. FS61PP; c. ETNA10PP. Green seen on the photos are foulant material. Red seen on ETNA10PP (c) is the color of membrane itself.
flux of the MF membranes which leads to higher fouling tendency. However, our results shown in Figs. 2, 3 and 4 do not indicate better performance of UF membranes over MF membranes as reported by Rossignol [17]. The discrepancy observed in the two studies could be due to different algal types, different membrane types and different õoperating conditions used by two groups. Membrane flux could be influenced by many parameters. All experiments we conducted showed very similar flux level for MF and UF membranes. Although ETNA10PP exhibited much lower membrane fouling because of its anti-fouling properties, the steady state flux was in fact still lower than MFP8 and FS40PP. It was hypothesized that the flux of this membrane in long period tests may be higher than other membranes due to its low fouling properties. However, the hypothesis was not proven in this experiment. 3.5. Concentration of Chlorella suspensions The above filtration experiments in recirculation mode were carried out to find the MF or UF membrane with the best performance (highest permeate flux and best rejection of Chlorella). The final goal of this work is to use MF or UF membranes for the concentration of Chlorella suspensions.
where V0 and Vt are the initial feed volume (32 L) and feed volume at time t, respectively. The relationship between VCF and permeation flux was studied. Experiments were performed with the microfiltration membrane (MFP8) and ultrafiltration membranes (FS40PP and ETNA10PP). The results in Fig. 7 demonstrated that permeate fluxes declined much faster at low Chlorella cell concentrations during the initial stages of the experiment. As the process continued, a stable flux was reached at higher Chlorella cell concentrations. It is believed that the cake layer (or fouling layer) probably became thicker and denser with increasing Chlorella concentration at high cell concentration range, and thus reduced permeate flux. A stable flux was observed when the cake layer did not change at higher cell concentrations. There is a larger flux drop for MFP8 than the other membranes at the early stages of the concentration, which can be explained by the larger pore size of the microfiltration membrane However, the steady-state flux seems to be similar for MFP8 and FS40PP, and the flux is slightly higher than for ETNA10PP. After 130 min of concentration, the feed volume in the batch tank was decreased from the initial 32 L to 2.8 L, resulting in a VCF of 11.4. Since MF membranes have much more open pore structure and much higher porosity than UF membranes, one would expect that MF membranes show much higher permeate flux for algae harvesting. The results in Figs. 6 and 7 show very similar performance of MFP8 (pore size of 0.8 μm) and FS40PP (MWCO 100,000), indicating that the pore size of membranes is not as an important parameter for this kind of application as originally hypothesized. A possible explanation for these results is that the fouling layer caused by deposition of algae cells and EPS was acting as a membrane selective layer [36]. The fouling layer can be clearly seen in Fig. 5 for FS40PP. The results in this work further suggest that the fouling layer, as a membrane selective layer, is similar for MF and UF membranes, implying that pore size and porosity have little influence on the formation of the fouling layer. Although ETNA10PP showed lower permeate flux than MFP8 and FS40PP in Figs. 6 and 7, the membrane exhibited very low fouling tendency as the permeate flux was rather stable during the whole filtration period. The low fouling tendency was further shown by very little deposition of foulants on the membrane surface, as seen in Fig. 5. In fact, the low fouling tendency of this membrane can be expected, owing to its hydrophilic membrane surface. The lower permeate flux of ETNA10PP can be considered to be due to the higher flow resistance of the membrane as this is a ‘tight’ (much smaller pore size) UF membrane. It may be
Fig. 6. Direct comparison between microfiltration (MFP8 and ultrafiltration membranes (FS40PP, ETNA10PP). Filtration conditions: T = 24 °C, TMP = 1.8 bar and cross-flow = 5.79 m/s.
X. Sun et al. / Algal Research 2 (2013) 437–444
443
Fig. 7. Permeate flux versus VCF for microfiltration (MFP8) and ultrafiltration (FS40PP, ETNA10PP) membranes. Filtration conditions: T = 24 °C, TMP = 1.8 bar and cross-flow = 5.79 m/s.
concluded from our work that membrane materials play a very important role for this application. Membranes with hydrophilic surfaces will most likely show better performance from the viewpoint of reducing fouling, whereas hydrophobic membranes will experience severe fouling for algae filtration.
thank Gary Lloyd for proof reading and corrections of the manuscript. This work was supported by the National Natural Science Foundation of China (Grant No. 20976140).
References 4. Conclusion The filtration and concentration of Chlorella from dilute culture media and the performance of several commercial MF and UF membranes were evaluated and compared. Flux increased with increasing temperature of the feed solution. However, above a certain temperature, further increase in temperature didn't improve permeate flux, most likely because of the release of EPS by Chlorella cells and/or deposition of Chlorella cells. A general trend of increased permeate flux with increasing TMP was observed. At higher TMP, however, permeate flux gradually leveled off or even dropped due to the effect of a fouling (cake) layer development. It was also seen that higher cross-flow velocity can significantly decrease particles accumulating on the membrane surface. Although ETNA10PP exhibited much better anti-fouling properties, the steady-state permeate flux of this membrane was not higher than for MFP8 and FS40PP in a direct comparison test. However, this is believed to be due to the much lower pore size i.e. MWCO (10,000) of ETNA10PP rather than limitation due to fouling layer buildup (see below). Furthermore, the concentration experiments indicate that the MF membrane did not show higher permeate flux than the UF membranes under the same operation conditions. MF membranes and UF membranes show similar flux in this work, indicating that pore size and porosity are not important for this application. This suggests that the permeate flux of different membranes is controlled by the fouling layer that acts as the membrane selective layer. Our work also demonstrated that a membrane with hydrophilic surface shows very little fouling for algae harvesting. The results of ETNA10PP suggest that membrane materials are the most important parameters for reducing fouling tendency. In future work, we will make analysis on the deposition layer (or EPS) attached to membrane surfaces to understand the membrane fouling better. Acknowledgments Xuefei Sun wishes to acknowledge the Alfa Laval Nakskov A/S for the support of the work. Authors would like to thank Jørgen Enggaard Boelsmand at the Algae Innovation Center of Denmark for providing algae suspensions and helpful discussions. We would also like to
[1] Q. Hu, C.W. Zhang, M. Sommerfeld, Biodiesel from algae: lessons learned over the past 60 years and future perspectives, Psychological Society of America Annual Meeting, University of Alaska Southeast, Juneau, Alaska, USA, 2006. [2] Q. Hu, M. Sommerfeld, E. Jarvis, M. Ghirardi, M. Posewitz, M. Seibert, A. Darzins, Microalgal triacylglycerols as feedstocks for biofuel production: perspectives and advances, Plant J. 54 (2008) 621. [3] Y.-T. Hung, Q.S. Amuda, A.O. Alade, Algae harvest energy conversion, Handb. Environ. Eng. 11 (723) (2010). [4] M. Chateaudegat, M. Chinain, N. Cerf, S. Gingras, B. Hubert, E. Dewaillyii, Seawater temperature, Gambierdiscus spp. variability and incidence of ciguatera poisoning in French Polynesia, Harmful Algae 4 (2005) 1053. [5] L.C. Bruce, J. Imberger, The role of zooplankton in the ecological succession of plankton and benthic algae across a salinity gradient in the Shark Bay solar salt ponds, Hydrobiologia 626 (2009) 111. [6] N.K. Singh, D.W. Dhar, Sewage effluent: a potential nutrient source for microalgae, Proc. Indian Natl. Sci. Acad. 72 (2006) 113. [7] S. Chinnasamy, A. Bhatnagar, R.W. Hunt, K.C. Das, Microalgae cultivation in a wastewater dominated by carpet mill effluents, Bioresour. Technol. 101 (2009) 3097. [8] T. Hirooka, Y. Akiyama, N. Tsuji, H. Naqase, K. Hirata, K. Miyamoto, Removal of hazardous phenols by microalgae under photoautotrophic conditions, J. Biosci. Bioeng. 95 (2003) 200. [9] E.B. Sydney, T.E. Da Silva, A. Tokarski, A.C. Novak, J.C. De Carvalho, A.L. Woiciecohwski, C. Larroche, C.R. Soccol, Screening of microalgae with potential for biodiesel production and nutrient removal from treated domestic sewage, Appl. Energy 88 (2011) 3291. [10] S. Thomas, The economics of micro-algae production and processing into biofuel faming systems, Department of Agriculture, Western Australia, 2006. [11] L.C. Neves, L.F.M. Silva, L.M.C. Cabral, S.G.F. Leite, V.M. Matta, Effect of flow rate and transmembrane pressure on permeate flux during ultrafiltration of residual brine from squid processing, 2nd Mercosur Congress on Chemical Engineering, 4th Mercosur Congress on Process Systems Engineering, 2005. [12] B. Sandhya, T. Satoshi, Microfiltration membrane fouling and cake behavior during algal filtration, Desalination 261 (2010) 46. [13] X.Z. Zhang, Q. Hu, M. Sommerfeld, E. Puruhito, Y.S. Chen, Harvesting algal biomass for biofuels using ultrafiltration membranes, Bioresour. Technol. 101 (2010) 5297. [14] M.T. Hung, C.J. Liu, Microfiltration for separation of green algae from water, Biointerfaces 51 (2006) 157. [15] S. Zou, Y.S. Gu, D.Z. Xiao, C.Y. Tang, The role of physical and chemical parameters on forward osmosis membrane fouling during algae separation, J. Membr. Sci. 366 (356) (2011). [16] C.K.W. Chow, S. Panglisch, J. House, M. Drikas, M.D. Bruch, R. Gimbeld, A study of membrane filtration for the removal of cyanobacterial cells, J. Water Supply Res Technol. 46 (1997) 324. [17] N. Rossignol, L. Vandanjon, P. Jaouen, F. Quemeneur, Membrane technology for the continuous separation microalgae/culture medium: compared performances of cross-flow microfiltration and ultrafiltration, Aquac. Eng. 20 (1991) 191. [18] A.L. Wei, G.M. Zeng, G.H. Huang, J. Liang, X.D. Li, Modeling of a permeate flux of cross-flow membrane filtration of colloidal suspensions: a wavelet network approach, Int. J. Environ. Sci. Technol. 6 (2009) 395.
444
X. Sun et al. / Algal Research 2 (2013) 437–444
[19] F.G. Meng, S.R. Chae, A. Drews, M. Kraume, H.S. Shin, F.L. Yang, Recent advances in membrane bioreactors (MBRs): membrane fouling and membrane material, Water Res. 43 (2009) 1489. [20] C.H. Zhang, F.I. Yang, W.J. Wang, B. Chen, Preparation and characterization of hydrophilic modification of polypropylene non-woven fabric by dip-coating PVA (polyvinyl alcohol), Sep. Purif. Technol. 61 (2008) 272. [21] L. Heng, Y.L. Yang, W.J. Gong, X. Li, G.B. Li, Effect of pretreatment by permanganate/ chlorine on algae fouling control for ultrafiltration (UF) membrane system, Desalination 222 (2008) 74. [22] J.B. Castaing, A. Massé, M. Pontié, V. Séchet, J. Haure, P. Jaouen, Investigating submerged ultrafiltration (UF) and microfiltration (MF) membranes for seawater pre-treatment dedicated to total removal of undesirable micro-algae, Desalination 253 (2010) 71. [23] S. Babel, S. Takizawa, A study on membrane fouling due to algal deposition, Water Sci. Technol. 41 (2000) 327. [24] N. Rossi, I. Petit, P. Jaouen, P. Legentilhomme, M. Derouiniot, Harvesting of Cyanobacterium Arthrospira platensis using inorganic filtration membranes, Sep. Sci. Technol. 40 (2005) 3033. [25] W. Liu, Harvesting of Micro-algae by Membrane Filtration, The International Congress on Membranes and Membrane Processes, Amsterdam, 2011. [26] J. Wei, S.H. Gitte, C.W. Nick, Characterization of a non-fouling ultrafiltration membrane, Desalination 192 (2006) 252. [27] M. Kallioinen, M. Pekkarinen, M. Mänttäri, M. Nyström, N.J. Jutta, Stability of two different regenerated cellulose ultrafiltration membrane under varying solution temperature, Desalination 199 (2006) 204.
[28] Y. Magara, M. Itoh, The effect of operational factors on solid/liquid separation by ultra-membrane filtration in a biological denitrification system for collected human excreta treatment plants, Water Sci. Tech. 23 (1991) 1583. [29] F.A.G. Mattheus, S.S. Shyam, S.A.M. Salha, H.A.B. Rashid, W. Mark, Effect of feed temperature on permeate flux and mass transfer coefficient in spiral-wound reverse osmosis systems, Desalination 144 (2002) 367. [30] A. Drews, C.H. Lee, M. Kraume, Membrane fouling—a review on role of EPS, Desalination 200 (2006) 186. [31] S. Tsuneda, H. Aikawa, H. Hayashi, A. Yuasa, A. Hirata, Extracellular polymeric substances responsible for bacterial adhesion onto solid surface, FEMS Microbiol. Lett. 223 (287) (2003). [32] N. Rossi, M. Derouiniot-Chaplain, P. Jaouen, P. Legentilhomme, I. Petit, Arthrospira platensis harvesting with membranes: fouling phenomenon with limiting and critical flux, Bioresour. Technol. 99 (2008) 6161. [33] M.K. Darko, S.L. Markov, N.T. Miodrag, Membrane fouling during cross-flow microfiltration of Polyporus squamosus fermentation broth, Biochem. Eng. J. 9 (2001) 103. [34] G.T. Vladisavljevic, P. Vukosavljevic, B. Bukvic, Permeate flux and fouling resistance in ultrafiltration of depectinized apple juice using ceramic membranes, J. Food Eng. 60 (2003) 241. [35] A.A. Makardij, Impact of operating parameters on flux decline in microfiltration and ultrafiltration. Thesis The University of Auckland, Auckland, New Zealand, 2002. [36] N. Thang, F.A. Roddick, L.H. Fan, Biofouling of water treatment membranes: a review of the underlying causes, monitoring techniques and control measures, Membranes 2 (2012) 804.