Application of ceramic membranes for microalgal biomass accumulation and recovery of the permeate to be reused in algae cultivation

    Application of ceramic membranes for microalgal biomass accumulation and recovery of the permeate to be reused in algae cultivation Arkadiusz Nedzarek, Arkadiusz Drost, Filip Harasimiuk, Agnieszka T´orz, Małgorzata Bonisławska PII: DOI: Reference:

S1011-1344(15)00294-8 doi: 10.1016/j.jphotobiol.2015.09.009 JPB 10140

To appear in: Received date: Revised date: Accepted date:

15 March 2015 2 September 2015 9 September 2015

Please cite this article as: Arkadiusz Nedzarek, Arkadiusz Drost, Filip Harasimiuk, Agnieszka T´ orz, Malgorzata Bonislawska, Application of ceramic membranes for microalgal biomass accumulation and recovery of the permeate to be reused in algae cultivation, (2015), doi: 10.1016/j.jphotobiol.2015.09.009

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Application of ceramic membranes for microalgal biomass accumulation and recovery of the permeate to be reused in algae cultivation Arkadiusz Nędzarek1,*, Arkadiusz Drost1, Filip Harasimiuk1,*, Agnieszka Tórz1,

T

Małgorzata Bonisławska1

IP

Department of Aquatic Sozology, West Pomeranian University of Technology in Szczecin, ul. Kazimierza Królewicza 4, 71-550 Szczecin, Poland e-mail: First author

SC R

Nędzarek, Arkadiusz ([email protected])

Corresponding author : Harasimiuk, Filip ([email protected]) ABSTRACT

NU

The present study was carried out to investigate the possibility of using ceramic membranes for microalgal biomass densification and to evaluate the qualitative composition

MA

of the permeate as a source of nitrogen and phosphorus for microalgae cultivated in a closed system. The studies were conducted on the microalga Monoraphidium contortum. The microfiltration process was carried out on a quarter-technical scale using ceramic membranes

D

with 1.4 m, 300 and 150 kDa cut-offs. Permeate flux and respective hydraulic resistances

TE

were calculated. Dissolved inorganic nitrogen and phosphorus fractions were measured in the feed and the permeate. It was noted that the permeate flux in the MF process was decreasing

CE P

while the values of reversible and irreversible resistances were increasing as the cut-off of the studied membranes was diminishing. An analysis of the hydraulic series resistance showed that using a 300 kDa membrane would be the most beneficial, as it was characterized by a comparatively high permeate flux (Jv=1.68 10-2 m3/m2s), a comparatively low susceptibility

AC

to irreversible fouling ( 1,72·109 1/m) and a high biomass retention coefficient (91%). The obtained permeate was characterized by high concentrations of dissolved nitrogen and phosphorus forms, which indicated that it could be reused in the process of microalgal biomass production. Keywords microfiltration, Monoraphidium contortum, harvesting microalgae, ceramic membrane 1. Introduction The characteristics of microalgae qualify them as an efficient and valuable source of biomass used for biofuel production, such as biodiesel, bioethanol or biogas [16, 23]. Due to their chemical composition, nutritional value and functional properties the microalgae have also found applications in biotechnology, wastewater pre-treatment, food supplementation and pharmacology [5].

ACCEPTED MANUSCRIPT One of the most serious problems faced by microalgae cultivation is finding an economically justified method of biomass densification [7]. Applied densification methods use centrifugation, sedimentation and membrane processes.

T

Harvesting by centrifugation is generally characterized by high capture productivity

IP

(>90%) under low flux rates and high energy consumption [8]. A comparatively high biomass recovery has been noted resulting from sedimentation using polyelectrolytes, however, the

SC R

process is time-consuming. For instance, according to Granados et. al. [11], the time of microalgae sedimentation using flocculants exceeded 2.10-4 m/s. An alternative for those processes is the application of pressure-driven membrane separation techniques. Cross-flow

NU

membrane techniques are characterized by a high permeate flux accompanied by comparatively low energy outlays. However, the difficulty lies in selecting the material of

MA

which the membranes are made and a cut-off enabling a combination of high microalgae inception and high membrane permeability.

Studies on application of membrane separation techniques have been conducted by

D

e.g. Ahmad et al. [4]. The authors tested the process of microfiltrating the suspension of

TE

Chlorella sp. microalgae using a cellulose membrane. They noted that the permeate flux increased together with transmembrane pressure (TMP) and feed flux velocity. The flux was

CE P

higher when the pressure was high, suggesting that the resistance of the membranes to mass transfer increased. The authors suggested that in order to achieve a higher permeate flux, the value of the driving force ought to be increased. Moreover, their study analyzed the values of hydraulic resistances of the membrane itself, the polarization layer and the filtration cake. It

AC

was noted that reversible resistance invoked by the filtration cake layer had a decisive effect on the permeate flux in the filtration process. According to the study conducted to research by Babel and Takizawa [5], hydraulic resistance caused by algal cells is rather complicated due to extracellular release of organic substances which considerably increase fouling. In the presented study we assumed that the application of membrane separation might intensify the process of biomass acquisition. Apart from densified microalgal biomass in the retentate we obtained also the permeate, which might be used in photobioreactors. Thanks to that technical answer, the whole cycle might take place in a closed system in which water loss from the drained retentate would have to be made up for, and nutritional substances for the microalgae would have to be added. In order to test the above assumptions, a study was conducted comparing the permeate flux of ceramic membranes with cut-offs of 150 kDa, 300 kDa and 1.4 μm and their susceptibility to fouling in the process of microalgal biomass densification. Furthermore,

ACCEPTED MANUSCRIPT chemical composition of the permeate was analyzed in order to decide whether it could be reused in the process of microalgal biomass production.

T

2. Materials and Methods

IP

2.1 Microalgae and Culture Conditions

The study used Monoraphidium contortum microalgae. The culture was grown

SC R

aseptically indoors in 200 L photobioreactors at 295±0.5 K, controlled pH 8 and aeration at 0.2 v/v/min. Artificial white light was used (1000 lumen fluorescent lamps) in the cycle of 12 h day / 12 h night. Microalgae were cultivated on F2 culture for the period of 8 days. The

NU

microalgae solution obtained in this way was used as feed in the MF. Due to large feed volumes required for conducting the MF, the microalgae were cultivated periodically,

MA

separately for each of the tested membranes. 2.2 Conducting the MF process

MF tests were conducted using a quarter-scale filtration system (Fig. 1). Ceramic

D

membranes with 1.4 μm, 150 and 300 kDa cut-offs were used. The experiment was conducted

TE

using 23-channel ceramic ultrafiltration membranes (Tami Industries®, France) made of Al2O3/TiO2/ZrO2, 1178 mm long, with the internal diameter of 25 mm, channel diameter of

CE P

3.6 mm, and the filtration area of 0.35 m2. Transmembrane pressure (TMP) amounted to 0.1 MPa, and cross-flow velocity (CFV) equaled 4 m/s. The temperature of the process was fixed and equaled 293±1.0 K. For each of the membranes the filtration process was conducted for 3 hours in an open system, i.e. the permeate was constantly drained. Bearing in mind the scale

AC

of the membrane installation, diversified cut-offs of applied membranes and the measurement of the so-called pseudo-equilibrium of the permeate flux, the following feed volumes were used in the experiments: 50 L (150 kDa membrane), 120 L (300 kDa membrane) and 200 L (1.4 μm membrane). Filtration tests were repeated thrice for each membrane. After each experiment the membranes were submitted to the process of chemical cleaning according to the following procedure: 2% solution of NaOH ((T=363±5 K, t=40 min.), rinsing with demineralized water, 0.5% solution of HNO3 (T=323±5 K, t=30 min.) and thrice rising with demineralized water. Upon the chemical cleaning of the membranes the permeability typical for a clean membrane was obtained. Permeate flux (Jv) and respective hydraulic resistances (R) were calculated according to the equations described by Szmukała and Szaniawska [19] and Kuca and Szaniawska [12] using the following formula:

ACCEPTED MANUSCRIPT JV 

TMP 3 [m / m 2  s ] RT 

[1]

T

TMP stands for transmembrane pressure (Pa), ŋ stands for water viscosity and RT stands for

IP

total resistance. Total resistance consists of the membrane resistance (RM), reversible fouling resistance (RR), and the irreversible fouling resistance (Ri). In order to determine RM, water

SC R

was recirculated through the membrane module at the temperature of 293 K, at the cross-flow velocity of 4 m/s. RM was calculated according to the formula: TMP 1 [ ] JW m

[2]

NU

RM 

MA

JW stands for volume velocity of demineralized water. Total resistance may be measured as TMP to permeate flux ratio in the MF process of true solutions (Jv):

TMP 1 [ ] JV m

D

[3]

TE

RT 

Rinsing the membrane with demineralized water after the completion of the MF process of

CE P

true solutions provides an opportunity for calculating the sum of membrane resistances and irreversible fouling:

TMP 1 [ ] JP m

[4]

AC

RM  Ri 

JP stands for volume velocity of the permeate for the water used in the rinsing process. As for the calculable value of reversible fouling (RR), it can be calculated using the following formula:

1 RR  RT  ( RM  Ri )[ ] m

[5]

2.3 Analyses The hydrochemical tests were conducted according to the methodology recommended by the APHA [3]. Concentrations of the dissolved inorganic nitrogen and the dissolved reactive phosphorus were measured in the feed and the permeate after preliminary filtration through a glass fiber filter GA-55 manufactured by Toyo Roshi Kaisha. Particular nitrogen and phosphorus forms were determined by colorimetric methods. UV-VIS spectrophotometer

ACCEPTED MANUSCRIPT Pharo 300 Spectroquant (by Merck) was used in the colorimetric methods, measuring the absorbance at the recommended wavelengths. Nitrite nitrogen was determined with sulphanilamide (=543 nm); nitrate nitrogen was determined as nitrite nitrogen after the

T

reduction on the Cu-Cd column; ammonium nitrogen was determined by the indophenol blue

IP

(=630 nm). Total dissolved inorganic nitrogen (TDIN) was calculated as the sum of the nitrite nitrogen, nitrate nitrogen and the ammonium nitrogen. Total dissolved reactive

SC R

phosphorus (TDRP) was determined by the method using ammonium molybdate and ascorbic acid as a reducing agent (=882 nm). Chlorophyll a was determined by applying the method of extraction with acetone (=665 nm). Water pH was measured with the pH-meter CP-103

NU

(manufactured by Elmetron). Turbidity was measured by the nephelometric method according to the methodology recommended by AOAC [2], using a nephelometer manufactured by

MA

Eutech Instruments. 3. Results and Discussion 3.1 Permeate Flux and Fouling

D

Among the studied membranes the highest permeate flux was noted for the membrane

TE

with the 1.4m cut-off. After 3 h of filtration, the flux equaled on average ca. 4.48.10-2 m3/m2s . After the same duration of filtration, permeate flux for membranes with the 150 cut-off and

CE P

300 kDa equaled 4.06.10-3 and 1.68.10-2 m3/m2s respectively (Fig. 2). Such differences in the permeate flux were connected with morpho-mechanical characteristics of membranes, since permeability increased together with the cut-off [14].

AC

The most significant drop of the permeate flux (Jv) was noted during the first 30 minutes of the experiment (Fig. 2). For membranes with 150 and 300 kDa it respectively amounted to 45 and 43% of its initial value, and for the membrane with 1.4 μm it equaled merely 28% of the initial value. Such a drop of the permeate flux is characteristic for membrane separation processes. For instance, to Kuca and Szaniawska [12] as well as Almécija et. al [1], noted the highest drop of the permeate flux over a similar period of time as was noted in the presently discussed experiments. The phenomenon could be explained by the fact that the largest deposition of organic matter on the membrane surface and inside membrane pores was noted during the first minutes of the MF process, leading to an increase of hydraulic resistance of the system [4]. Pseudo-equilibrium of the permeate flux, indicating that the system equilibrium had been achieved, was noted for each of the studied membranes after the lapse of ca. 1.5 h (Fig. 2). After that time, volume velocity of the permeate for the 1.4m, 150 and 300 kDa

ACCEPTED MANUSCRIPT membranes constituted on average only 9, 17 and 28% of the initial value, respectively. Achievement of the state of equilibrium is a characteristic feature of membrane processes performed by the cross-flow method. The mechanisms of achieving the pseudo-equilibrium

T

permeate velocity is tightly connected with the phenomena of concentration polarization and

IP

fouling. Internal fouling occurs at the first stage of filtration and then an additional layer with a heightened organic matter concentration forms on the membrane surface [4]. It results in

SC R

substantial drop in the flux over time. The time necessary to obtain such a state depends on the separation limit of the membrane used and on chemical constitution of the fluid that is being filtrated. This phenomena can be easily explained on the chart describing the diversity

NU

of permeate flow in time. After an intensive decrease of permeation rate in the initial period of filtration, stabilisation of the flow can be observed. Taking the example of Castaing et al.

MA

[6] depending on the membrane put on test a sudden drop in the flow within initial 10-30 min of filtration as well as reaching pseudo-established maximal flow after 2 hours upon commencing the filtration was noticed (permeate flux drops rapidly over the first 10-30 min

D

of filtration and reaches the same pseudo-steady state value approximately 2 h after the

TE

beginning of filtration).

In the present study, the UF process was conducted for 1.5 h after the moment of achieving

CE P

the so-called pseudo-equilibrium state of the permeate flow. After 3 h of filtrating the respective feed volumes, i.e. 50 L (150 kDa membrane), 120 L (300 kDa membrane) and 200 L (1.4 μm membrane), a similar retentate volume (ca. 9 L) was obtained for each membrane. The retentates were characterized by various chlorophyll concentrations. The highest average

AC

chlorophyll a concentration was noted in the retentate after the UF using the 150 kDa membrane (450 mg/m3) and lower for the 1.4m and 300 kDa membranes (347 and 324 mg/m3 respectively) (Tab. 1). However, concentration coefficient values were similar: 3.533.62. Similar correlations were observed for turbidity. The turbidity of solutions used in the processes of filtration was on similar level and equalled, on average, 24.6 NTU. A high index in the retentates was observed, ranging from 540% (membrane with cut-off of 300 kDa) up to approx. 770% (membrane with cut-off of 150 kDa) (Table 1). The high retention of microalgae is characteristic for the UF and MF processes. For instance Castaing et al. [6] while filtrating sea water containing Heterocapsa triquetra microalgae proved over 90% retention of biomass for the used hollow-fibre membrane with the following cut-offs 0.2 µm, 300 kDa and 10 kDa. At the same time a turbidity retention was proved to reach the level of 98%.

ACCEPTED MANUSCRIPT In spite of the high chlorophyll a retention in the MF process we noted its partial transmission for all studied membranes. As can be seen in Fig. 3, the lowest permeability was noted for the 150 kDa membrane and the highest for the 1.4 m membrane (2.5 and 13.6%

T

respectively). Also the permeability against the turbidity increased together with the

IP

separation limit of the membranes put on test and oscillated from approx. 4% for the membrane with cut-off kDa 150 to approx. 18% for the membrane with cut-off of 1.4 m

SC R

(Fig. 3). For instance Castaing et al. [6] proved the permeability at the level of approx. 2% regardless the cut-off of the membranes put on test. Such differences could result from the discrepancies of transmembrane pressure (they used over three times lower pressure than we

NU

did). The increase of transmembrane pressure can influence the destruction of microalgae and as a result of extracellular matrix release to the solution, including chlorophyll a [9]. Also the

MA

morphologic construction of filtrated microalgae could influence the results. The microalgae M. contortum we used has elongated shape, with diameter ranging from 1.3.-4.5µm while the one tested by Castaing et al. [6] H. triquetra microalgae is oval with diameter of 8-22µm.

D

The elongated shape of M. contortum can make it easier for the microalgae to go through a

TE

membrane, especially the one with cut-off of 1.4 m. The oval shape of H. triquetra can favour high retention, similarly to Chlorella algae (oval shape) that was proved by Babel and

CE P

Takizawa [5]. It indicates the necessity to take into consideration those parameters while constructing membrane systems for alga biomass thickening in order to receive the appropriate turnout.

AC

Factors determining membrane efficiency in the MF process include also the size and type of fouling [22]. Having in view the application of membrane separation processes for the industrial scale algae treatment it is important to select membranes yielding good results with respect to washing off reversible fouling, thus determining high MF efficiency and allowing for less frequent chemical cleaning of the membranes. Average permeate flux values obtained in the MF process and flux values for ultrapure water as well as the degrees of fouling wash-off after conducting the MF process of true solutions are given in Tab. 2. The included data were used for calculating the hydraulic series resistance of which the total resistance was comprised, as shown in Fig. 4. It was noted that the smaller the pore diameters of the used membranes were, the higher was the membrane resistance against distilled water (Fig. 4). This phenomenon is characteristic for MF processes and connected with the morphological structure of membranes [17]. It was also noted that the smaller the pore diameters of the used membranes were, the higher were reversible and irreversible resistance values for the microalgae solutions (Fig. 4).

ACCEPTED MANUSCRIPT The highest reversible resistance was noted for the 150 kDa membrane. For 1.4 μm and 300 kDa membranes, the reversible resistances were 18 and 6 times lower respectively. The value of irreversible resistance was the highest for the 150 kDa membrane and close to the initial

T

resistance value for that membrane. Irreversible resistance values for the 1.4 μm and 300 kDa

IP

membranes were similar and amounted to 1.27·109 and 1.72·109 1/m respectively. The 150

SC R

kDa membrane was characterized by the highest total resistance value (RT=2.46·1010). For the 1.4 μm and 300 kDa membranes, RT values equaled 2.23·109 and 5.94·109 1/m respectively. The analysis of percentage proportions of the respective resistances in the total

NU

resistance is shown in Fig. 5. The lowest percentage proportion of the initial resistance was noted for the 1.4 μm membrane (15.3%). Simultaneously, that membrane was also

MA

characterized by the highest proportion of the irreversible resistance in the total resistance (it constituted 57% of the total resistance). In the case of the 150 and 300 kDa membranes, the proportions of the irreversible resistance in the total resistance amounted to 27 and 29%

D

respectively. The reversible resistance for the 150 kDa constituted as much as 47% of the total

TE

resistance, whereas for the 1.4 μm and 300 kDa membranes the proportions were lower and constituted 28 and 34% respectively.

CE P

Similar correlations between resistance and membrane cut-off were noted e.g by Piron et. al [17]. Studying the phenomenon of fouling while subjecting alage to the MF process the authors noted that the 0.22 μm membrane they used was more susceptible to pore blocking by microalgae than the 50 kDa membrane. The highest irreversible resistance noted in the

AC

present study for the 1.4 m membrane implied that the membrane was the most susceptible to permanent pore blocking by the pressed microalgae and polymeric subtances released by them [5]. As for the highest reversible resistance value, it was noted for the 150 kDa membrane. That kind of resistance resulted from reversible fouling formation, i.e. the formation of an additional filtration layer on the membrane surface [4]. 3.2 Results of Hydrochemical Analyses Concentrations of the dissolved inorganic nitrogen and dissolved reactive phosphorus measured in the feed and in the permeate are given in Tab. 3. Biogenic element concentrations in the respective feeds were similar. Only the feed used for the 150 kDa membrane was characterized by a twice higher nitrate nitrogen concentration than the feeds used for the remaining membranes. Simultaneously, in comparison with the feeds used for the remaining membranes, the feed used for the 300 kDa membrane was characterized by a higher concentration of the dissolved reactive phosphorus by 0.022 mgP dm-3 on average.

ACCEPTED MANUSCRIPT It was noted that as a result of the conducted microfiltration, concentrations of the analyzed nitrogen and phosphorus forms in the permeate were reduced in comparison to their concentrations in the feed. The highest reductions were noted for the nitrate nitrogen and

T

ammonium nitrogen when the 1.4 μm membrane was used (78 and 65% respectively). A high

IP

ammonium nitrogen reduction (55%) was also noted for the 300 kDa membrane. The remaining reduction values ranged from 7% (N-NO3-, 300kDa membrane) to 27% (TDRP,

SC R

150 kDa membrane) (Tab. 3).

The electrostatic interactions between the ions themselves and between the ions and the membrane surface may produce an active layer appearing on the membrane surface that

NU

cause an increase of retention degree, including stoppage of ions with the radius lower than the size of the membrane's pore. Therefore reductions of the determined ions proved in our

MA

tests should be accepted as a typical phenomenon in the membrane processes, despite the sieving character of the filtration process. For example Waeger et al. [21], by filtrating fermentative waste water on ceramic membranes with the cut-off of 0.2µm, 50 kDa and of 20

D

kDa, they received the following retention level of ammonium ion: 12.7, 4.42 and 3.57%.

TE

While Šostar-Turk et al. [20] while conducting the UF process of laundry wastewater on ceramic membranes received the reduction of ammonium ion of approx. 90%. The nominal

CE P

molecular cut-off size of the membrane in their study was from 20-400 kDa. The conducted analyses revealed that in the permeate, in spite of the noted reduction during the MF, concentrations of inorganic nitrogen and phosphorus forms were high and characteristic for surface waters with a high trophy level [18]. Thus, the permeate obtained

AC

from the microfiltration process may be reused in the process of algae biomass production. A comparatively high ammonium nitrogen reduction noted in the present study should be perceived as beneficial. As noted to Xin et. al [22], a high concentration of ammonium ions may be a factor hindering microalgal growth. Furthermore, according to Drost et. al [10], microfiltration is a process effectively removing pathogenic microorganisms and thus may reduce the risk of a secondary infection of recirculation systems.

4. Conclusions According to study results, the permeate flux in the MF process was decreasing and the values of reversible and irreversible resistances were increasing as the cut-off of the studied membranes was diminishing. An analysis of the hydraulic series resistance showed that using the 300 kDa membrane may be the most beneficial, as it was characterized by a

ACCEPTED MANUSCRIPT comparatively high permeate flux accompanied by a comparatively low susceptibility to irreversible fouling and a high selectivity with respect to biomass recovery. In spite of the noted reductions of dissolved inorganic nitrogen and phosphorus forms,

IP

T

the obtained permeate could be reused in the process of microalgal biomass production.

Acknowledgments

SC R

We would like to thank very much for making available for study the microalgae cultures from the Culture Collection of Baltic Algae, Institute of Oceanography, University of Gdańsk.

1. M. C. Almécija, R. Ibáne,

NU

References

A. Guadix , M. Guadix , Effect of pH on the

MA

fractionation of whey proteins with a ceramic ultrafiltration membrane, J. Membrane Sci. 288 (2007) 28-35.

2. AOAC. 1984. Official methods of analysis of the Association of Official Chemists,

D

Washington 431-443.

TE

3. APHA. 1995. American Public Health Association. Standard Methods for Examination of Water and Wastewater, APHA-AWWA-WEF Washington, 1200.

CE P

4. A.L. Ahmad, M. Yasin , N.H. Derek , J.K Lim, Crossflow microfiltration of microalgae biomass for biofuel production, Desalination 302 (2012) 65-70. 5. S. Babel, S. Takizawa, Microfiltration membrane fouling and cake behavior during algal filtration, Desalination 261 (2010) 46-51.

AC

6. J.-B. Castaing, A. Massé, M. Pontié, V. Séchet, J. Haure, P. Jaouen, Investigating submerged ultrafiltration (UF) and microfiltration (MF) membranes for seawater pretreatment dedicated to total removal of undesirable micro-algae, Desalination 253 (2010) 71-77. 7. L. Christenson, R. Sims , Production and harvesting of microalgae for wastewater treatment, biofuels and bioproducts, Biotechnol Adv 29 (2011) 686-702. 8. A.J. Dassey, Ch. S. Theegala, Harvesting economics and strategies using centrifugations for cost effective separation of microalgae cells for biodiesel applications, Bioresource Technol. 128 (2013), 241-245. 9. A. Drews, C.H. Lee, M. Kraume, Membrane fouling – a review on the role of EPS, Desalination 200 (2006) 186-188.

ACCEPTED MANUSCRIPT 10. A. Drost, A. Nędzarek, E. Bogusławska-Wąs, A. Tórz, M. Bonisławska , UF application for innovative reuse of fish brine: product quality, CCP management and the HACCP system. J. Food Process Eng. 37 (2014) 396-401.

T

11. M.R. Granados, F.G. Acién , T. Gómez, J.M. Fernández-Sevilla, E. Molina

IP

Grima, Evaluation of flocculants for the recovery of freshwater microalgae, Bioresource Technol. 118 (2012) 102-110.

SC R

12. Kuca M., Szaniawska D, Application of microfiltration and ceramic membranes for treatment of salted aqueos effluents from fish processing, Desalination 2009 (241) 227-235.

Membrane Sci.438 (2013) 18-28.

NU

13. L. Luo, Y. Wan, Effect of pH and salt on nanofiltration – a critical review, J.

MA

14. D. Nanda, K.L.Tung, Y.L. Li , N.J. Lin, C.J. Chuang , Effect of pH on membrane morphology, fouling potential, and filtration performance of nanofiltration membrane for water softening, J. Membrane Sci.349 (2010), 411-420.

D

15. A. Nędzarek, A. Drost, A. Tórz, F. Harasimiuk, D. Kwaśniewski, The impact of

TE

pH and sodium chloride concentration on the efficiency of the process of separating high-molecular compounds, J. Food Process Eng. 38(2) (2015) 115-124.

241-264.

CE P

16. S.S. Oncel ,Microalgae for macroenergy world, Renew. Sust. Energ. Rev. 26 (2013)

17. E. Piron, F. René, E. Latrille , A cross-flow microfiltration model based on integration of the mass transport equation, J. Membrane Sci.108 (1995), 57-70.

AC

18. N.N. Rabalais :Nitrogen in Aquatic Ecosystems, Ambio 31 (2002) 102-112. 19. M. Szmukała, D. Szaniawska, Application of ceramic membranes in water treatment for fish hatchery supplying purposes, Desalination 240 (2009) 117-126. 20. S. Šostar-Turk, I. Petrinić, M. Simonič, Laundry wastewater treatment using coagulation and membrane filtration, Resour. Conserv. Recy. 44 (2005) 185-196. 21. F. Waeger, T. Delhaye, W. Fuchs, The use of ceramic microfiltration and ultrafiltration membranes for particle removal from anaerobic digester effluents, Sep. Purif. Technol. 73 (2010) 271-278. 22. L. Xin, H. Hong-ying , G. Ke , Y. Jia , Growth and nutrient removal properties of a freshwater microalga Scenedesmus sp. LX1 under different kinds of nitrogen sources. Ecol. Eng. 36 (2010), 379-381.

ACCEPTED MANUSCRIPT 23. X. Zhang, Q. Hu, M. Sommerfeld, T. Puruhito, Y. Chen, Harvesting algal biomass for biofuels using ultrafiltration membranes. Bioresource Technol. 101 (2010), 5297-

AC

CE P

TE

D

MA

NU

SC R

IP

T

5304.

ACCEPTED MANUSCRIPT

M R

F

SC R

IP

T

4

M 3

NU

1

P

MA

M

D

2

TE

Fig. 1. Small-scale experimental rig: 1 – feed tank; 2 – pump; 3 – membrane module;

AC

CE P

4 – radiator; F – the feed; R – retentate; P – permeate; M – manometer

SC R

IP

T

ACCEPTED MANUSCRIPT

CE P

TE

D

MA

NU

membrane 1.4 µm

AC

membrane 300 kDa

membrane 150 kDa

Figure 2

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Fig. 2. Variation of permeate flux (JV) during MF and UF process of Monoraphidium contortum suspensions

ACCEPTED MANUSCRIPT

25

T IP

15

SC R

Permeability [%]

20

10

5

NU

0 Chlorophyll

300 kDa

MA

150 kDa

Turbidity NTU

1.4 ?m

AC

CE P

TE

D

Fig. 3. Selectivity of the studies membranes with respect to the measured indicators

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

D

Fig. 4. Hydraulic resistances (initial resistance RM for ultrapure water; reversible resistances RR and irreversible resistances Ri for the filtered microalgal suspensions) measured for the 1.4

AC

CE P

TE

μm, 150 kDa and 300 kDa membranes during the MF and UF process

ACCEPTED MANUSCRIPT

80

IP

T

60 40

SC R

s hare in the total res is tanc e [% ]

100

20 0 Membrane 1.4 um

Membrane 300 kDa

Membrane 150 kDa

R evers ible res is tanc e

Irrevers ible res is tanc e

NU

Initial membrane res is tanc e

MA

Fig. 5. Percentage proportions of hydraulic resistances (initial, reversible and irreversible) in

AC

CE P

TE

D

the total resistance of the analyzed membranes measured during the MF process

ACCEPTED MANUSCRIPT Table 1. Mean values and the standard deviation (SD) for turbidity and chlorophyll a in the feed and the retentate and average increases of those indicators in the retentate as compared to

Turbidity

mg/m3

NTU

124.7

26.9

(15.2)

(2.1)

91.2

22.5

150 kDa

300 kDa

NU

Feed

(8.4)

(1.9)

98.4 (6.4) 450.9

150 kDa

24.4

MA

1,4 m

324.3

(19.8)

(2.2)

TE

D

(2.3)

347.1

169.0

(14.9)

(1.8)

150 kDa

362

773

300 kDa

357

538

1,4 m

353

693

AC

%

208.0

121.0

300 kDa

1,4 m

Increase

(1.8)

(24.5)

CE P

Retentate

IP

Chlorophyll a

SC R

Type of membrane

T

concentrations in the feed (NTU – nephelometric turbidity unit)

ACCEPTED MANUSCRIPT Table 2. Flux values for: ultrapure water (“clean membrane”); water used in the process of post-microfiltration/ultrafiltration rinsing; microalgal solution after 3 h of MF/UF and the standard deviation (SD) Permeate flux SD

SD

7.68•10-2

2.32•10-4

4.06•10-3

1.80•10-4

4.53•10-2

2.55•10-2

1.02•10-3

1.68•10-2

1.63•10-3

2.94•10-1

6.20•10-1

1.50•10-3

4.48•10-2

4.47•10-3

NU

1.58•10-2

AC

CE P

TE

D

MA

Membrane 150kDA Membrane 300 kDA membrane 1,4µm

SC R

m3/m2 · s

at the end of the MF/UF process

T

clean membrane

IP

TMP= 0.1 MPA

membrane washed after MF/UF

ACCEPTED MANUSCRIPT Table 3. Mean values and the standard deviation (SD) for concentrations of dissolved inorganic nitrogen and dissolved reactive phosphorus in the feed and in the permeate and average

150 kDa Permeate

300 kDa

CE P

150 kDa 300 kDa 1,4 m

AC

Reduction %

TE

1,4 m

0.007 (0.002) 0.009 (0.003) 0.009 (0.002) 0.007 (0.002) 0.007 (0.001) 0.002 (0.001) 22 78

0.611 (0.021) 0.380 (0.014) 0.332 (0.016) 0.472 (0.032) 0.354 (0.021) 0.276 (0.018) 22 7 17

0.034 (0.005) 0.049 (0.007) 0.046 (0.010) 0.030 (0.005) 0.022 (0.006) 0.016 (0.004) 12 55 65

TDRP

IP

1,4 m

mgN/L

TDIN

mgP/L

0.652 (0.028) 0.438 (0.022) 0.387 (0.021) 0.509 (0.040) 0.383 (0.033) 0.295 (0.024) 22 12 24

0.041 (0.004) 0.066 (0.007) 0.046 (0.005) 0.030 (0.003) 0.057 (0.004) 0.036 (0.004) 27 14 22

SC R

300 kDa

N-NH4+

NU

Feed

N-NO3-

MA

150 kDa

N-NO2-

D

Type of membrane

T

reductions of these indicators in the permeate in comparison to the feed

ACCEPTED MANUSCRIPT Highlights

- Densification of microalgae using ceramic membranes;

T

- Nutrients were efficiently removed from solutions;

IP

- Rational management of water;

AC

CE P

TE

D

MA

NU

SC R

-Outlook for further potential development of this technology was also presented