You are on page 1of 40

Accepted Manuscript

Magnetophoretic removal of microalgae from fishpond water: Feasibility of high gradient and low gradient magnetic separation Pey Yi Toh, Swee Pin Yeap, Li Peng Kong, Bee Wah Ng, Chan Juinn Chieh Derek, Abdul Latif Ahmad, JitKang Lim PII: DOI: Reference: To appear in: Received Date: Revised Date: Accepted Date: S1385-8947(12)01240-5 http://dx.doi.org/10.1016/j.cej.2012.09.051 CEJ 9812 Chemical Engineering Journal 15 August 2012 14 September 2012 17 September 2012

Please cite this article as: P.Y. Toh, S.P. Yeap, L.P. Kong, B.W. Ng, C.J.C. Derek, A.L. Ahmad, J. Lim, Magnetophoretic removal of microalgae from fishpond water: Feasibility of high gradient and low gradient magnetic separation, Chemical Engineering Journal (2012), doi: http://dx.doi.org/10.1016/j.cej.2012.09.051

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.

Magnetophoretic removal of microalgae from fishpond water: Feasibility of high gradient and low gradient magnetic separation

Pey Yi Toha, Swee Pin Yeapa, Li Peng Konga, Bee Wah Nga,b, Chan Juinn Chieh Dereka, Abdul Latif Ahmada, JitKang Lima,c,*

School of Chemical Engineering, Universiti Sains Malaysia, Nibong Tebal, Penang

14300, Malaysia.
b

School of Biological Sciences, Universiti Sains Malaysia, Minden, Penang 11800,

Malaysia.
c

Department of Physics, Carnegie Mellon University, Pittsburgh, PA 15213, USA.

*To whom correspondence should be addressed:

JitKang Lim School of Chemical Engineering Universiti Sains Malaysia Nibong Tebal, Penang 14300 Malaysia E-mail: chjitkangl@eng.usm.my Tel: 604-599-6423 Fax: 604-594-1013

ABSTRACT Microalgae blooms in commercial fish production ponds resulting in a deficit in the overall oxygen budget have posed serious challenges to aquaculture industry. In this study, we demonstrate the feasibility of rapid microalgae separation in real-time from fishpond water by magnetophoresis. By relying on the magneto-shape anisotropy of rod-liked iron oxide magnetic nanoparticle (IONPs), overall separation efficiency of microalgae cells up to 90% can be achieved in less than 3 minutes. The IONPs employed, with a saturation magnetization at 113.8 emu/g, are surface functionalized with cationic polyelectrolyte that promotes the attachment of these particles onto microalgae cells via electrostatic interaction. Kinetic of magnetophoretic separation process was monitored by suspension opacity measurements based upon a custom built light dependent resistor (LDR setup) sensor. Whereas, the overall separation efficiency of microalgae cells is determined spectrophotometrically at 685 nm wavelength. Performance of both high gradient magnetic separation (HGMS) with T/m and low gradient magnetic separation (LGMS) with T/m

were tested with varying particle concentration (50500 mg/l) and the results obtained were interpreted in term of cooperative magnetophoresis theory. Cost analysis was conducted to verify the feasibility for large scale implementation of LGMS system with the cost involved at $0.13 for every one meter cube of treated fishpond water.

Keywords: Magnetophoresis; Magnetic nanoparticle; Microalgae removal; Fishpond water; Magnetic separation.

1.

Introduction

Aquaculture and fish farming is one of the solutions for the worldwide decline of fisheries stocks of either marine or freshwater fish and also to fulfill the demand of worlds growing population [1]. One major problem that plagues the freshwater fish farm is microalgae blooms that occur in every fishpond during summer or tropical country, such as Malaysia. Microalgae will naturally grow in fishpond water because of the presence of nutrients, such as nitrogen, phosphorus, carbon source [2, 3], which originated from the fish excretion, excess fish food and decaying organic matter. Most of those nutrients promote the growth of microalgae are in organic matter form and they can be quantified as Chemical Oxygen Demand (COD) with an ideal value for fishpond at less than 50 mg/l [4]. Ironically, the growth of the microalgae naturally is beneficial for the removal of the excess nutrient in the water to avoid nutrient overloading as well as reducing the COD level. However, this benefit is diminished once the microalgae start to grow excessively. For a typical freshwater fishpond, the microalgae will keep growing as long as there are nutrient to substantiate its growth. High microalgae concentration beneficial as oxygen source through photosynthesis [5] and also provides shades for fishes from the sunlight. However, the high concentration of microalgae will be disastrous, as their huge amount will exhaust the oxygen supply through respiration and releasing carbon dioxide during nighttime. Fish may be killed overnight through suffocation [6] when dissolved oxygen (DO) is less than 2 mg/l [7]. In most cases, DO in fish pond should be maintained at least 4 mg/l all the time [4]. At extremely high nutrient level eutrophication will occur [7]. The nutrient will promote excessive growth of algae

bloom in the presence of sunlight. The sunlight will be blocked totally by the dense algae bloom to form a death zone beneath the bloom with drastic oxygen depletion and hence deteriorate the fish production. Conventionally, there are multiple well-practiced methods to maintain the microalgae population within a small fishpond [8]. Effective management of microalgae growth can be achieved naturally via few methods, such as, growing aquatic plants around the fishpond to consume the nutrient and starving the microalgae [9, 10], avoid over feeding and using high quality food to ensure complete digestion of the food [11], and using barley straw to control algae growth in pond [12]. Besides the natural treatment, by using the algaecides chemical, which contain simazine, chelated copper and potassium permanganate, is also able to kill the algae but it is harmful to living organisms and environment [13]. The quick death of all the microalgae may increase ammonium concentration and decrease dissolve oxygen in water and hence it is not favorable. Nevertheless, most of the standard practices involved for microalgae removal were labor intensive and had limited efficiency [14]. Since microalgae biomass can be employed as third generation biofuel [15] and other useful products, like nutrients in form of polyunsaturated fatty acid (PUFA) [16, 17, 18] or pigment [19, 20] with a robust removal technique without direct annihilation of microalgae might be economically more attractive. There are several microalgae separation methods which have been developed to meet high microalgae separation efficiency. The most common microalgae separation methods are filtration, centrifugation, flocculation and settling and ion exchange [21, 22]. Flocculation and settling is a versatile method, which is suitable to process large quantity of biomass, but it is time consuming [23]. Filtration method has recorded high separation efficiency, however, this method is quite costly with the 4

problem of blocking and fouling [21, 22]. Centrifugation is the most reliable method but it is expensive and consumed large amount of energy during operation and is not suitable to handle enormous quantity of effluents [24]. In addition to all the separation techniques discussed so far, magnetic separation of microalgae from water resources is a relatively new concept which was introduced in 1970s [25, 24]. This method was recently revisited by us [26] and others [27, 28] due to its attractive advantages such as high permeation fluxes, high removal efficiency, small land area utilization and no clogging and fouling problems [29]. Moreover, magnetic separation process can also be perform directly on raw samples that contain suspended solid material due to its ease in capturing the targeted samples by using surface functionalized magnetic particles [30]. In order to achieve magnetic separation of biological cell, tagging the cell surface with a paramagnetic dipole moment is necessary since most cells are irresponsive to applied magnetic force [31]. Microalgae cells membrane surface are negatively charged because of the present of lipids, proteins and sugars, which have functional groups like -SH, -OH and -COOH. Deprotonation of those ligands will give a net negative charge on cell surface at natural pH of water [22, 32, 33, 34]. While for the magnetite, it is negative by charge when disperse in deionized water [35], with isoelectric point between 6.30-6.85 [36, 37, 38]. Under this scenario, we need a binder to immobilize the magnetic nanoparticles onto the microalgae cells. The binder that is normally employed to serve this purpose is a positively charged polyelectrolyte, where it can be adsorbed on the nanoparticle surface [26] through direct method or link to the negative charged cell surface indirectly through [30] electrostatic interaction [35]. After tagging the microalgae cells with magnetic nanoparticles, cells can be separated magnetophoretically through either low gradient 5

magnetic separation (LGMS), without magnetized matrix or high gradient magnetic separation (HGMS) [30], which contains magnetized matrix [39]. In this work, we illustrated the engineering feasibility of LGMS and HGMS in harvesting microalgae from fishpond water by direct method, which is tagging surface functionalized iron oxide nanoparticle (IONPs) onto the microalgae cells surface. Furthermore, we compared the separation efficiency between LGMS and HGMS at various IONPs concentration corresponding to different kinetic behaviors. Optical light intensity sensing system (LDR setup) is employed to measure the overall separation efficiency and quantify its kinetic, while the UV-Vis spectrometer is conducted to measure the specific cell separation efficiency. Cost analysis on HGMS system for microalgae separation from fishpond water is conducted to provide a guideline for different system setup and design preferences.

2.

Experimental Methods

2.1. Materials

Rod-liked iron oxide magnetic nanoparticle (IONPs) were obtained from Toda America, Inc. The 35 wt% very low molecular weight

poly(diallyldimethylammonium chloride) (PDDA) in water with molecular weight, Mw < 100,000 g/mol was obtained from Sigma-Aldrich, Inc. Deionized water used was obtained by reverse osmosis and further treated by the Milli-Q Plus system (Millipore) to 18 Mcm resistivity.

2.2. Characteristic of microalgae in fishpond water

Fishpond water sample was collected from fish farm of Aik Lee Fishery, which is located at Sungai Bakau, Parit Buntar, Perak, Malaysia. The samples were brought back to the laboratory for analysis. Microalgae cells species were observed under Olympus CX41RF microscope equipped with Image Pro Express 4.0.1 imaging software. Chemical Oxygen demand (COD) of the samples was measured spectrophotometrically by the DR 5000TM UV-Vis Spectrophotometer with the use of High Range Plus COD Reagent from HACH Company, USA. The pH of our fishpond water was measured by using Eutech CyberScan pH 1500.

2.3. Nanoparticles attachment to microalgae

In this work, rod-like IONPs was used with physical dimension of ~ 20 nm in diameter and 300 nm in length respectively [35]. The immobilized-on technique [26] or direct method [30] was performed with the attachment of very low molecular weight PDDA cationic polyelectrolyte onto the IONPs surface to form surface functionalized IONPs. Firstly, 3408 l of PDDA was dispersed into 25 ml of deionized water to obtain a concentration of 0.0458 g/ml, and sonicated for 1 hour. This polyelectrolyte solution was left overnight to ensure complete dissolution of PDDA. Next, 13 ml of IONPs at concentration of 0.01 g/ml was added into the polyelectrolyte solution and sonicated for 1 hour. The final surface modified particle suspension was left under mixing condition on an end-over-end rotating mixer at 37 rpm for 6 days. Electrophoretic mobility and spherical equivalent approximation of

surface functionalized IONPs was measured (Malvern Instruments Nanosizer) before and after surface functionalization to verify the successful attachment of the PDDA. After successful surface modification of IONPs, a permanent magnet was employed to collect the nanoparticles and the supernatant which contained the excess PDDA was discarded. The surface functionalized IONPs collected was further dispersed in 5 ml deionized water and this concentrated surface functionalized IONPs dispersion was then diluted to desired concentration accordingly from 50 to 500 mg/l In every separation study, 2 ml of surface functionalized IONPs suspension was added to 8 ml fishpond water sample with vigorous shaking by hand for 30 seconds to ensure uniform mixing and dispersion. The mixture was left for another 30 seconds before being used for magnetophoretic separation study.

2.4. Low gradient magnetic separation (LGMS)

LGMS study in this work was performed in batch mode with a grade N50 NdFeB permanent magnet (Ningbo YuXiang E&M Int1 Co., Ltd). The surface magnetic field of the permanent magnet employed is ~ 6000 G (measured by an AlphaLab Model GM2 DC Magnetometer) with its magnetic field gradient < 100 T/m [26]. The kinetic behavior of magnetophoresis and cell separation efficiency under LGMS were quantified by light dependent resistor (LDR) setup and UV-Vis spectrometer for duration of 6 minutes. To quantify the kinetic behavior of magnetophoresis under LGMS, we placed a light dependent resistor (LDR) underneath the glass vial containing the sample with the incoming light transmitted through from top (Fig. 1). Under the influence of externally applied magnetic field, the sample becomes progressively more transparent

due to magnetophoretic collection, allowing more light to be detected by LDR. The degree of separation achieved was quantified by monitoring the intensity of light passing through the sample during cell separation in real time. Cell separation efficiency based on the initial fishpond water opacity was calculated with following equation:

Overall separation efficiency (%) = [V0V(t)] (V0Vcentrifuged) 100%

Eq. (1)

where V0 represents initial voltage of fishpond water sample without adding surface functionalized IONPs, V(t) represents voltage of sample at time t during magnetophoretic separation, and the Vcentrifuged represented the voltage of the centrifuged clear fishpond water sample (centrifuged for 20 minutes at ~ 4000g). Since both the binding of surface functionalized IONPs to microalgae cells and the LDR detection method is none specific, hence, this measurement will provide information on the overall separation of all negatively charged objects out from the fishpond water. To better quantify the cell separation efficiency, the absorbance of our sample was measured spectrophotometrically by UVmini-1240 Shimadzu at specific wavelength of 685 nm [40] (measured by Agilent Technologies Carry 60 UV-Vis). The cell separation efficiency was determined as

Cell separation efficiency (%) = [I0I(t)] (I0Icentrifuged) 100%

Eq. (2)

where I0, I(t) and Icentrifuged are the absorbance intensity of microalgae suspension initially, during magnetophoretic separation at time t and the clear centrifuged sample respectively. 9

2.5. High gradient magnetic separation (HGMS)

The HGMS system operated under continuous flow with magnetic field gradient > 1000 T/m were employed in this work. The HGMS column with internal diameter of 1.4 cm was packed with stainless steel wool with packing fraction of 14 vol% with packed-bed height at around 3 cm. A pairs of N50-graded NdFeB permanent magnet were employed to fully magnetize the packed-bed as shown in Fig. 2. The sample solution was pumped into the HGMS column at a flow rate of 1.25 ml/min. Cell separation efficiency was measured spectrophotometrically by using the same procedure as discussed previously and the results were analyzed according to equation (2).

2.6. Cost feasibility analysis

The costs of fishpond water treatment were estimated based on LGMS and HGMS as shown in Table 1. The water treatment systems were made up of two unit operations (Fig. 3) that were the mixer, for the mixing of fishpond water with surface functionalized IONPs, and the LGMS/HGMS magnetic separator. All the assumptions made were showed in Table 1 and further details regarding the system employed are available from the supporting documents.

3.

Results and Discussion

3.1. Characteristic of microalgae in fishpond water 10

From the microscopic observation on the fishpond water sample, there are at least 6 species of microalgae identified in the sample. They are Scenedesmus sp., Spirulina sp., Chlorella sp., Tetraedron sp., Haematococcus sp. and Dictyosphaerium sp. The fishpond water sample contains 775 mg/l COD (Table 2) and is slightly alkaline with pH ranged from 7.0 to 8.5.

3.2. Nanoparticles attachment to microalgae

The result (Table 3) showed the surface charge reversal of IONPs after addition of PDDA which confirmed the attachment of cationic polyelectrolyte onto the initially negatively charged IONPs. Furthermore, by making spherical equivalent

approximation [41], the colloidal stability of surface functionalized IONPs, monitored by dynamic light scattering (DLS) (Malvern Instruments Nanosizer ZS) showed an increment of IONPs hydrodynamic diameter from 374.0 139.3 nm to 474.4 35.1 nm after being coated with the PDDA with 1.76 0.6 nm in hydrodynamic diameter.

3.3. Microalgae cell separation by using LGMS

Fig. 4 depicts the overall and cell separation efficiency of fishpond water after LGMS collection induced by surface functionalized IONPs at various particle concentration. In all cases, the overall removal efficiency measured by LDR setup is slightly lower than the cell separation efficiency determined by UV-Vis absorbance measurement. This observation is distinctively obvious when low surface functionalized IONPs 11

concentration at 25 mg/l was employed. At this concentration, the differences between both results is 36.88%, followed with 35.84%, 23.11%, 9.37% and 13.45% for the subsequent surface functionalized IONPs concentration up to 500 mg/l. Since the present of total solid suspension contributes directly to the opacity of fishpond water, hence, the LDR measurement provides clear indication on the total removal efficiency of all negatively charged objects within the fishpond water, including the microalgae cells. The measurement obtained via the UV-Vis spectrophotometer provides a more accurate reading of chlorophyll a (and b) of the green algae occurred within 650 to 700 nm [42]. The large variation between the two measurements especially at low concentration of surface functionalized IONPs, suggested that electrostatic interaction induced particle attachment, even though is not target specific, but is very effective to promote the magnetophoretic separation of the microalgae cells from complex media. At low concentration of surface functionalized IONPs, the recorded separation efficiency of microalgae is low mainly due to the insufficient supply of surface functionalized IONPs to impart magnetic properties to the microalgae cells. This observation revealed the need to maintain high surface functionalized IONPs-to-cell ratio in order to achieve better cell separation efficiency. At high surface functionalized IONPs-to-cell ratio, there is higher tendency for more cells to be decorated by the surface functionalized IONPs and thus favor the magnetophoretic separation. This argument can be further generalized to justify the need of having colloidally stable surface functionalized IONPs before its attachment on the microalgae cells. Maintaining good dispersibility is vital to sustain high surface functionalized IONPs-to-cell ratio without losing the freely suspended particles to aggregation especially at high particles concentration [26]. Hence, by having an 12

electrosteric hindrance layer around the nanoparticles, originated from PDDA coating, led to the formation of a colloidally stable suspension which further aided the magnetophoretic separation process [43, 44]. Furthermore, the short exposure time of only 6 minutes is another factor that caused the higher cell separation efficiency unfavorable at low concentration of surface functionalized IONPs. A much higher separation efficiency of 79.12% for 25 mg/l surface functionalized IONPs has been observed if the collection time was prolonged to 20 minutes and this is consistent with our previous observation on single particle magnetophoresis [45]. The dependency of both overall and cells separation efficiency on surface functionalized IONPs concentration as witnessed in Fig. 4 is the result of magnetophoresis under low gradient magnetic field [46, 47]. The migration of microalgae cells to the magnetic field source is mainly due to the cooperative magnetophoresis of all particles attached to it. For such a rapid collection to happen (Fig. 5), the formation of large aggregate under the influence of an external applied magnetic field is necessary [48]. As the magnetically tagged microalgae cells approaches the magnetic field source, the microalgae cells collide, leading to cells chaining that further enhanced the magnetic removal rate [26]. Moreover, by increasing the particle concentration, the chances for them to stay in close proximity after attaching to microalgae cells surface will also increased. This in turn would favor the formation of aggregates on the surface of microalgae cell that contribute to rapid magnetophoresis.

3.4. Microalgae cell separation by using HGMS

13

In conventional industry practice, the removal of magnetic responsive materials, including the adsorbed biomaterials or pollutants, from its residing solution is through high gradient magnetic separation (HGMS) process [49]. In a HGMS operational unit, the magnetically responsive material is channeled through a packbed column composed of stainless steel fibers with fine mesh size. These packed fibers are responsible to generate inhomogeneous high magnetic field gradient within the column after magnetized by an external source [50]. When the sample solution flows through the magnetized stainless steel wool matrix, there are two dominating forces imposed onto the magnetically seeded microalgae cells, namely,

magnetophoretic and viscous drag force [51, 52]. If the sample solution is well mixed with low flow velocity, such as the one used in this work, diffusion force [51] and hydrodynamic resistance [53] can be neglected. Balancing all these interactions is non-trivial and becomes the main bottle neck to develop numerical and/or analytical solutions to the magnetic separation problem [54]. From Fig. 6, it is obvious that the cell separation efficiency achieved by HGMS shared a similar surface functionalized IONPs concentration dependency as LGMS system. The cell separation efficiency of HGMS increased from 66% to 91% by increasing the concentration of surface functionalized IONPs from 25 mg/l to 500 mg/l. For HGMS, the results obtained from cell counting through optical microscopy observation has verified this spectrophotometry results indicating 90% separation of microalgae cells has been achieved accompanying with 50% COD reduction (Table 2). This result is consistent with Cerff and coworkers observation in which the cell separation efficiency up to 90% can be achieved by HGMS on harvesting fresh water algae Chlamydomonas reinhardtii and Chlorella vulgaris [28].

14

At low concentration of surface functionalized IONPs (< 150 mg/l), HGMS out performed LGMS in term of cell separation efficiency mainly due to slow magnetophoresis kinetic under low field gradient. We anticipated that for longer collection time (>> 6 minutes) a much higher cell separation efficiency can be achieved with the same surface functionalized IONPs concentration. Whereas for high concentration of surface functionalized IONPs, both system performed relatively well with negligible difference ( 150 mg/l). In addition to concentration of surface functionalized IONPs, separation performance of HGMS depends on the particles size and magnetic properties of the seeding materials as well [51, 53]. Since the magnetophoretic force is directly proportional to the magnetic volume of the particle [51, 55], hence, large particle will experience a much larger force and can be collected in HGMS column much easily compared to smaller particle. However, the utilization of larger particles beyond the superparamagnetic limit, for iron oxide particles at around 50 nm [56], is not favorable as these particles has high tendency to aggregate and settle out from the solution before their attachment to the microalgae cells. After magnetic treatment the previously greenish fishpond water turned to crystal clear, especially when high particle concentration was used, indicating effective removal of microalgae (Fig. 7). By eyes inspection, a hint of blackish suspended solid can still be detected in the treated fishpond water after going through LGMS process. This is very likely due to the present of trace amount of surface functionalized IONPs in the treated fishpond water which is not fully recovered by LGMS. Since there are ample experimental evidences suggesting the toxicity of nanomaterials at cellular level [57, 58], thus, the implementation of LGMS unit for fishpond water treatment needs to be conducted with a much sophisticated design 15

which focuses on the full recovery of the particles used. Nevertheless, the surface functionalized IONPs employed in this work are coated with cationic polyelectrolyte. Thus, the positively charged particles should adsorb rapidly onto the surrounding soil surface and losing their mobility [59].

3.5. Cost feasibility analysis

In order to demonstrate the economic feasibility of magnetic separation for small to middle size fishpond water treatment, simple cost analysis was performed for LGMS system with a process involves two unit operations as shown in Fig. 3 with all the estimated expenses as shown in Table 1. This cost analysis was conducted based on the water treatment requirement of Aik Lee Fishery where all our samples were collected. The dimension of one fishpond there was around 22 m 25 m 1 m with a total of 21 fishponds. In order to cope with the cleaning requirement and water replacement rate of the fishpond during the microalgae blooming period, each fishpond need to be treated once a week with 20% of water is being treated at a rate of 48 m3/h to ensure good water quality for fish to grow. Here wet drum-type low gradient magnetic separator (Shanghai Lipu Heavy Industry Co., Ltd) is chosen due to its satisfactory capacity to treat water up to 60 m3 /h for an intermediate sized fish farm. The fish farm of this scale is not uncommon in the northern part of Malaysia. A two blades propeller agitator equipped mixer is needed here to avoid dead zones and promote better mixing of surface functionalized IONPs suspension with the fishpond water [60] before their introduction into LGMS. From our analysis, the treatment cost for fishpond water by using LGMS at the treatment rate of 48 m3/h with 300 mg/l surface functionalized IONPs (0.519 g 16

surface functionalized IONPs/g dry microalgae biomass) to achieve 90% cell separation efficiency is estimated to be $6.03/h (= $0.13/m3 pretreatment fishpond water). Under this scenario, the magnetic materials and the polymer binder have contributed almost 33% of the total cost involved. It is very unlikely that the current market price of surface functionalized IONPs at $119/kg can be reduced significantly in near future; hence, other replacements are needed. We envisage that in order for magnetic separation to become more economically viable, magnetic materials from scrap metal or mining residues might be a better candidate. However, the environmental impact and engineering feasibility of these materials are still unexplored and need further justifications. The cost of tap water for industrial usages in Malaysia starts at $0.18/m3 [61] and is slightly higher than the LGMS treatment cost if the loss of surface functionalized IONPs is being tabulated at 0.07%. This is the targeted value which we would like to achieve by further improvement of separator efficiency involved in immediate future. Nevertheless, the charging of large amount of fish farm effluents without treatment into the nearby river has caused numerous problems to the local community and this makes magnetic separation attractive. The loss of surface functionalized IONPs for each cycle of LGMS treatment was estimated at 0.07% with the use of 0.05 mg/l surface functionalized IONPs. This value translated into ~ 0.15 mg/l of Fe (by assuming the surface functionalized IONPs is 100% magnetite) is being introduced into the fishpond and is within the ideal iron level at 0.01-0.30 mg/l to avoid bioaccumulation [4]. So, for the purposes of cost reduction and safety issue, there is a pressing need to design a more effective magnetic separation system to keep the surface functionalized IONPs leakage minimum. 17

We have also generalized our cost analysis to a HGMS system (See supporting documents for the detail HGMS cost analysis in Table SII). The purchase cost for high gradient magnetic filtration separator (HGMF) at $1,600,000 from the literature [24] was used for cost estimation without cost index normalization. This is an underestimated value and the cost involved will go up by a factor of four if the cost index was taken into consideration. The treatment cost for 165.7 m3/h of fishpond water with 50 mg/l of surface functionalized IONPs (0.0866 g surface functionalized IONPs/g dry microalgae biomass) to achieve about 90% cell separation efficiency was approximated as $85.85/h (= $0.52/m3 pretreatment fishpond water) accordance to a process rate of 207 m3/h (included feed of surface functionalized IONPs) with the same working capacity of Yadida et al., 1977 [24]. This capacity is capable to treat every fishpond in Aik Lee Fishery for three times in one week and it is comparable to the workload handled by 3 LGMS units. For HGMS system, the huge expenses arise from the operation unit (contribute 52% of total cost) with its high power consumption for magnetic power generation, pumping and flushing power. This shortcoming, perhaps, can be counter-balanced by the use of permanent magnet for the magnetization of inner matrix of HGMS unit as illustrated by Hoffman and coworkers [64].

4.

Conclusions

We have verified the feasibility of the microalgae separation from the fishpond water through the application of the surface functionalized IONPs (very low molecular weight PDDA coated TODA iron oxide) under low gradient magnetic separation

18

technique (LGMS). The feasibility of LGMS was guaranteed as long as the amount of surface functionalized IONPs introduced is sufficient to induce magnetophoresis for tagged microalgae cells. In addition, the magnetophoretic separation of microalgae from fishpond water was proven feasible by using high gradient magnetic separation technique (HGMS). For both methods, LGMS and HGMS, microalgae removal efficiency of more than 90% can be achieved depending on the amount of the surface functionalized iron oxide nanoparticles (IONPs) used. Compared to HGMS, the performance of LGMS is more sensitive toward the concentration of surface functionalized IONPs, mainly due to its cooperative magnetophoresis nature. There is a slight discrepancy in term of cell separation efficiency between HGMS and LGMS systems at surface functionalized IONPs usages below 150 mg/l. At high particle concentration, LGMS performed equally well as HGMS in microalgae separation. In our study, the key advantages of LGMS system are its low energy consumption and the ease of design by using permanent magnet arrays. These features are also generally true for magnetic separator employed for industrial applications. By monitoring the suspension opacity while undergoing magnetophoresis, we also quantify the kinetic behavior microalgae removal by LGMS for 6 minutes. We believe the removal time can be further reduced by increase the concentration of surface functionalized IONPs. From our cost analysis, LGMS system is more cost effective for microalgae separation, with an estimated cost of $0.13/m3 water treated. Here, the magnetic materials and the binding agents contribute the major expenses for LGMS technique. For the implementation of LGMS for microalgae removal from fish farm water, we envisage a simple process involved only two unit operations is good enough for the uses of small to middle fisher industry.

19

Acknowledgements

This material is based on work supported by Research University (RU) (Grant No. 1001/PJKIMIA/811165) from Universiti Sains Malaysia, Exploratory Research Grants Scheme (ERGS) (Grant No. 203/PJKIMIA/6730011) from Ministry of Higher Education of Malaysia, and International Foundation for Science (IFS) (Grant No. 304/PJKIMIA/6050232/I100). P. Y. Toh was supported by the My PhD scholarship from Ministry of Higher Education of Malaysia. We thank Dr. B. S. Ooi from School of Chemical Engineering, USM, Malaysia for providing invaluable help during the experiment. All authors are affiliated to the Membrane Science and Technology cluster USM.

Supporting Information Available Tables showing the technical data and specification of each equipment for LGMS unit, cost analysis of HGMS system based upon 90% of cell separation efficiency for 50 mg/l of surface functionalized IONPs.

20

REFERENCES [1] R.L. Naylor, R.J. Goldburg, J.H. Primavera, N. Kautsky, M.C.M. Beveridge, J. Clay, C. Folke, J. Lubchenco, H. Mooney, M. Troell, Effect of aquaculture on world fish supplies, Nature 405 (2000) 1017-1024. [2] G. Huang, F. Chen, D. Wei, X.W. Zhang, G. Chen, Biodiesel production by microalgal biotechnology, Appl. Energy. 87 (2010) 38-46. [3] O. Perez-Garcia, F.M.E. Escalante, L.E. de-Bashan, Y. Bashan, Heterotrophic cultures of microalgae: Metabolism and potential products, Water Res. 45 (2011) 1136. [4] B. Santhosh, N.P. Singh, Guidelines for water quality management for fish culture in Tripura, ICAR Research Complex for NEH Region Tripura Centre, Lembucherra799 210 Tripura (West), New Manikya Press, L. N. Bari Road, Agartala, 2007. [5] A. Demirbas, Use of algae as biofuel sources, Energy Convers. Manage. 51 (2010) 2738-2749. [6] NT Fisheries Group, Fish Kill Investigation Manual. Fishery Report NO.70, ISBN 0724547053, Department of Business, Industry & Resource Development, Northern Territory Government, 2004. [7] K. Kangur, A. Kangur, P. Kangur, R. Laugaste, R., Fish kill in Lake Peipsi in summer 2002 as a synergistic effect of a cyanobacterial bloom, high temperature, and low water level, Proc. Estonian Acad. Sci. Biol. Ecol. 54 (2005) 67-80. [8] G. Borics, I. Grigorszky, S. Szabo, J. Padisak, Phytoplankton association in a small hypertrophic fishpond in East Hungary during a change from bottom-up to topdown control, Hydrobiologia. 424 (2000) 79 90. [9] R.W. Crites, D.C. Gunther, A.P. Kruzic, J.D. Pelz, G. Tchobanoglous, Design Manual Constructed wetlands and aquatic plant systems for municipal wastewater

21

treatment. U. S. Environmental protection Agency (EPA), Center for Environmental Research Information (CERI), Cincinnati, Ohio, 1988. [10] B.C. Wolverton, Aquatic plants for wastewater treatment, Nat. Sci. (1988) 382387. [11] M.P. Masser, J. Rakocy, T.M. Losordo, Recirculating Aquaculture Tank production Systems, Management of Recirculating Systems. SRAC Publication No.452 (1992). [12] M.G. Ghobrial, M.A. Okbah, S.M. Gharib, A.M. Soliman, Influence of barley straw and submerged macrophytes on fishpond wastewater quality, Egypt. J. Aquat. Res. 33 (2007) 67-87. [13] S. Graslund, B.E. Bengtsson, Chemicals and biological products used in southeast Asian shrimp farming, and their potential impact on environment - a review, Sci. Total Environ. 280 (2001) 93-131. [14] M. Weijerman, R. Most, K. Wong, S. Beavers, Attempt to control the invasive red alga Acanthophoraspicifera (Rhodophyta: ceramiales) in a Hawaiian fishpond: An assemssment of removal techniques and management options, Pacific Sci. 62 (2008) 517-532. [15] Y. Cheng, W. Zhou, C. Gao, K. Lan, Y. Gao, Q. Wu, Biodiesel production from Jerusalem artichoke (Helianthus Tuberosus L.) tuber by heterotrophic microalgae Chlorella protothecoides, J. Chem. Technol. Biotechnol. 84 (2009) 777-781. [16] Z.Y. Wen, F. Chen, F., Heterotrophic production of eicosapentaenoic acid by microalgae, Biotechnol. Adv. 21 (2003) 273294. [17] O.P. Ward, A. Singh, Omega-3/6 fatty acids: Alternative sources of production, Process Biochem. 40 (2005) 3627-3652.

22

[18] L. Brennan, P. Owende, Biofuels from microalgae - A review of technologies for production, processing, and extractions of biofuels and co-products, Renewable Sustainable Energy Rev. 14 (2010) 557-577. [19] M.A. Borowitzka, Commercial production of microalgae: ponds, tanks, tubes and fermenters, J. Biotechnol. 70 (1999) 313-321. [20] C. Curtain, Plant Biotechnology - The growth of Australia's algal B-carotene industry, AusBiotech. 10 (2000) 19-23. [21] J.P. Hoffmann, Wastewater treatment with suspended and nonsuspended algae, J. Phycol. 34 (1998) 757-763. [22] E. Molina-Grima, E-H. Belarbi, F.G. Acien Fernandez, A. Robles-Medina, Y. Chisti, Recovery of microalgal biomass and metabolites: process options and economics, Biotechnol. Adv. 20 (2003) 491-515. [23] Y. Nurdogan, W.J. Oswald, Tube settling of high-rate pond algae, Water Sci. Technol. 33 (1996) 229-241. [24] R. Yadida, A. Abeliovich, G. Belfort, Algae removal by high gradient magnetic filtration, Environ. Sci. Technol. 11 (1977) 913-916. [25] G. Bitton, J.L. Fox, H.G. Strickland, Removal of algae from Florida lakes by magnetic filtration, App. Microbio. 30 (1975) 905 908. [26] J.K. Lim, C.J.C. Derek, S.A. Jalak, P.Y. Toh, N.H.M. Yasin, B.W. Ng, A.L. Ahmad, Rapid magnetophoretic separation of microalgae, Small. 8 (2012) 16831692. [27] L. Xu, C. Guo, F. Wang, S. Zheng, C.Z. Liu, A simple and rapid harvesting method for microalgae by in situ magnetic separation, Bioresour. Technol. 102 (2011) 10047-10051.

23

[28] M. Cerff, M. Morweiser, R. Dillschneider, A. Michel, K. Menzel, C. Posten. Harvesting fresh water and marine algae by magnetic separation: Screening of separation parameters and high gradient magnetic filtration, Bioresour. Technol. 118 (2012) 289-295. [29] S. Bucak, S. Sharpe, S. Kuhn, T.A. Hatton, Cell clarification and size separation using continuous countercurrent magnetophoresis, Biotechnol. Progr. (2011). [30] M. Safarikova, I. Safarik, I., The application of magnetic techniques in biosciences, Magn. Electr. Sep. 10 (2001) 223-252. [31] J.J. Chalmers, M. Zborowski, L. Moore, S. Mandal, B.B. Fang, L. Sun, Theoretical analysis of cell separation based on cell surface marker density, Biotechnol. Bioeng. 59 (1998) 10 20. [32] S. McLaughlin, M-M, Poo, The role of electro-osmosis in the electric-fieldinduced movement of charged macromolecules on the surfaces of cells, Biophys. J. 34 (1981) 85-93. [33] T.B. Kinraide, P.R. Ryan, L.V. Kinhian, Interactive effects of Al3+, H+, and other cations on root elongation considered in terms of cell-surface electrical potential, Plant Physiol. 99 (1992) 1461-1468. [34] G.M. Gadd, Biosorption: critical review of scientific rationale, environmental importance and significance for pollution treatment, J. Chem. Technol. Biotechnol. 84 (2009) 13-28. [35] J.K. Lim, D.X. Tan, F. Lanni, R.D. Tilton, S.A. Majetich, Optical imaging and magnetophoresis of nanorods, J. Magn. Magn. Mater. 321 (2009) 1557-1562. [36] A.E. Regazzoni, G.A. Urrutia, M.A. Blesa, A.J.G. Marato, A. J. G., Some observations on the composition and morphology of synthetic magnetites obtained by different routes, J. Inorg. Nucl. Chem. 43 (1981) 1489-1493.

24

[37] S. Yean, L. Cong, C.T. Yavuz, J.T. Mayo, W.W. Yu, A.T. Kan, V.L. Colvin, M.B. Tomson, Effect of magnetite particle size on adsorption and desorption of arsenite and arsenate, J. Mater. Res. 20 (2005) 3255-3264. [38] X.Q. Xu, H. Shen, J.R. Xu, M.Q. Xie, X.J. Li, The colloidal stability and coreshell structure of magnetite nanoparticles coated with alginate, App. Surf. Sci. 253 (2006) 2158-2164. [39] B.L. Hirschbein, D.W. Brown, G.M. Whitesides, Magnetic separations in chemistry and biochemistry, Chem. Tech. 12 (1982) 172-179. [40] A. Richmond, Handbook of microalgal culture: Biotechnology and applied phycology, Blackwell Science, Ameslowa, 2004. [41] D. Farrell, C.L. Dennis, J.K. Lim, S.A. Majetich, Optical and electron microscopy studies of Schiller layer formation and structure, J. Colloid Interface Sci. 331 (2009) 394 400. [42] W.T. Shoaf, B.W. Lium, Improved extraction of chlorophyll a and b from algae using dimethyl sulfoxide, Limnol. Oceanogr. 21 (1976) 926 928. [43] D. Kleshchanok, P.R. Lang, Steric repulsion by adsorbed polymer layers studied with total internal reflection microscopy, Langmuir. 23 (2007) 4332-4339. [44] J.K. Lim, S.A. Majetich, R.D. Tilton, Stabilization of superparamagnetic iron oxide core-Gold shell nanoparticles in high ionic strength media, Langmuir 25 (2009) 13384-13393. [45] J.K. Lim, C. Lanni, E.R. Evarts, F. Lanni, R.D. Tilton, S.A. Majetich, Magnetophoresis of Nanoparticles, ACS Nano. 5 (2011) 217-226. [46] M. Benelmekki, C. Caparros, A. Montras, R. Goncalves, S. Lanceros-Mendez, L.M. Martinez, Horizontal low gradient magnetophoresis behavior of iron oxide

25

nanoclusters at the different steps of the synthesis route, J. Nanopart. Res. 13 (2011) 3199-3206. [47] G. De Las Cuevas, J. Faraudo, J. Camacho, Low-Gradient Magnetophoresis through field-induced reversible aggregation, J. Phys. Chem. C 112 (2008) 945-950. [48] J. Faraudo, J. Camacho, Cooperative magnetophoresis of superparamagnetic colloids: theoretical aspects, Colloid Polym. Sci. 288 (2010) 207-215. [49] R. Oder, High gradient magnetic separation theory and applications, IEEE Trans. Mag. 12 (1976) 428 435. [50] T. Terashima, H. Ozaki, M. Sekine, Removal of dissolved heavy metals by chemical coagulation, magnetic seeding and high gradient magnetic filtration, Water Res. 20 (1986) 537-545. [51] G.D. Moeser, K.A. Roach, W.H. Green, T.A. Hatton, P.E. Laibinis, P. E., Highgradient magnetic separation of coated magnetic nanoparticles, AIChE J. 50 (2004) 2835-2848. [52] J.A. Oberteuffer, High gradient magnetic separation, IEEE Trans. Magn. (1973) 303-306. [53] T.Y. Ying, S. Yiacoumi, C. Tsouris, High-gradient magnetically seeded filtration, Chem. Eng. Sci. 55 (2000) 1101-1115. [54] J.S. Andreu, J. Camacho, J. Faraudo, M. Benelmekki, C. Rebollo, L.M. Martinez, Simple analytical model for the magnetophoretic separation of superparamagnetic dispersions in a uniform magnetic gradient, Phys. Rev. E 84 (2011) 021402. [55] A. Ditsch, S. Lindenmann, P.E. Laibinis, D.I.C. Wang, T.A. Hatton, Highgradient magnetic separation of magnetic nanoclusters, Ind. Eng. Chem. Res. 44 (2005) 6824-6836.

26

[56] D.J. Dunlop, Magnetite: Behavior near the single-domain threshold, Sci. 176 (1972) 41-43. [57] S.M. Hussain, K.L. Hess, J.M. Gearhart, K.T. Geiss, J.J. Schlager, In vitro toxicity of nanoparticles in BRL 3A rat liver cells, Toxicol. In Vitro. 19 (2005) 975983. [58] M. Mahmoudi, A. Simchi, A.S. Milani, P. Stroeve, P., Cell toxicity of superparamagnetic iron oxide nanoparticles, J. Colloid Interface Sci. 336 (2009) 510 518. [59] E. Kokufuta, K. Takahashi, Adsorption of Poly(diallyldimethylammonium chloride) on colloid silica from water and salt solution, Macromol. 19 (1986) 351354. [60] J.M. Coulson, J.F. Richardson, J.R. Backhurst, J.H. Harker, Coulson & Richardsons Chemical Engineering: Fluid Flow, Heat Transfer and Mass Transfer, vol. 1, fifth ed. Butterworth Heinemann, Ltd., Oxford, 1996. [61] Pengurusan Air Pahang Berhad, Tarif Air dan Cagaran, 2012. Available online: <http://jba.pahang.my/index.php?option=com_content&view=article&id=60&Itemid =102&lang=my>. [62] Tenaga Nasional Berhad, Pricing and Tariff, 2012. Available online: <http://www.tnb.com.my/residential/pricing-and-tariff.html>. [63] M.S. Peter, K.D. Tommerhaus, Plant Design and Economics for Chemical Engineers, fourth ed. McGraw-Hill, United States, 1991. [64] C. Hoffmann, M. Franzreb, W. H. Holl. A Novel High-Gradient Magnetic Separator (HGMS) Design for Biotech Applications, IEEE Trans. Appl. Supercon. 12 (2002) 963-966

27

28

Highlights Rapid magnetophoretic separation of mixed strains of microalgae from fishpond water is feasible. The concentration of surface functionalized IONPs affect the microalgae removal efficiency. LGMS and HGMS systems achieve high separation efficiency. Kinetic behavior of both LGMS and HGMS systems are compared. LGMS system more cost effective for small scale fishfarm water treatment.

29

Table(s)

Table 1. Cost involved for microalgae removal by LGMS with 90% of cell separation efficiency and 300 mg/l of surface functionalized IONPs used.

LGMS (Rate of fishpond water being treated is 48 m 3/h per unit LGMS) LGMS Separator (Wet Drum-type Magnetic Separator)b
($ 3230) over 15 years

Cost ($)/ha 0.09 0.20 0.86

Mixerb
($ 7177) over 15 years

Storage Tankb
(Pretreatment surface functionalized IONPs& Sludge after treatment) ($ 30818) over 15 years

Feed Pumpb
(Fishpond water & surface functionalized IONPs) ($ 2700) over 15 years

0.08 0.08 0.74

Power (LGMS separator, mixer and pumps)b, c


(4.85 kW)

Installationd
(Instrallation cost, piping, instrumentation cost, electrical installation, yard improvement and maintenance)

Labore Raw Materialf

2.00 1.98 Total 6.03 3 ($ 0.13 /m pretreatment fishpond water)

Cost estimation based on the continuous treatment of fishpond water for 8 hours per day (6 working days/week; 50 working weeks/year). There was 20% of water (110 m3) of each pond (Based on a unit size of fishpond in fishing farm of Aik Lee Fishery) will be treated. 21 fishponds will be treated in 6 working days.
b

See supporting documents for technical data and preference of each equipment showed in Table SI.

Cost estimation based on Malaysia electrical rate of Tenaga Nasional Berhad (2012) [62]. Reference: Book of Peter and Timmerhaus (1991) [63].

Cost estimation based on Malaysia labor rate.

Cost estimation based on cost of Fe3O4 (Six C USA Co., Ltd, China) and cationic polyelectrolyte of chitosan (Weifang Union Biochemistry Co. Ltd, Shandong, China) that are abundantly available in economic price and normally use in industrial application. It was assumed that there was 0.07% lost of surface functionalized IONPs by washout for each batch of water treatment and the surface functionalized IONPs was being recycled.

Table 2. Chemical Oxygen Demand (COD) and cell count of fishpond water before and after the microalgae removal by HGMS.

Before COD (mg/l)a Cell count (x104 cells/ml)b


a b

After 364c 825

Decreasing (%) 53.03 90.02

775 8267

Measured by the DR 5000 UV-Vis Spectrophotometer.. Counted on the Neubauer Improved Heamocytometer. c COD can further reduced by increase the fishpond water treatment rate up to appropriate capacity to meet the desired value.

Table 3. Electrophoretic mobility and spherical equivalent approximation of averaged hydrodynamic diameter of IONPs, very low molecular weight PDDA and surface functionalized IONPs.

Electrophoretic mobility [mcm/Vs] IONPs PDDA Surface functionalized IONPs -1.757 0.001 4.196 0.094 5.775 0.065

Hydrodynamic diameter [nm] 374.2 139.3 1.76 0.6 474.4 35.1

Figure(s)

Fig. 1. Schematic diagram of LGMS setup employed in this study. The LDR sensor was employed to measure the light transmitted through the fishpond water sample. Magnified image showed the magnetic field induction generated through the medium by a NdFeB magnet with surface magnetization ~ 6000 Gauss measured by Alphalab, Inc. DC Magnetometer Model GM2. This measurement verified the magnetic field working range of our system.

Fig. 2. Experimental HGMS setup with a cylinder column packed with stainless steel wool matrix and expose to NdFeB permanent magnets. This arrangement is chosen to resemble our LGMS setup for the ease of comparison.

Fig. 3. Block flow diagram of the fishpond water microalgae cells separation for fish farm application.

Fig. 4. Separation efficiency achieved by LGMS as a function of surface functionalized IONPs concentration measure by (i) LDR setup, and, (ii) UV-vis spectrometer respectively.

Fig. 5. Real time overall separation efficiency of magnetically tagged microalgae cells from fishpond water based on the initial fishpond sample after mixed with 150 mg/l surface functionalized IONPs, presented in graph together with images, in the LGMS separation after 6 minutes exposure to NdFeB permanent magnet.

Fig. 6. The separation efficiency of the microalgae cells from the fishpond water through (i) LGMS, and, (ii) HGMS as a function of surface functionalized IONPs concentration, measure by UV-Vis spectrometer.

(i)
Fishpond water 25 mg/l IONPs 50 mg/l IONPs 150 mg/l IONPs 300 mg/l IONPs 500 mg/l Centrifuged IONPs fishpond water

(ii)
Fishpond water 25 mg/l IONPs 50 mg/l IONPs 150 mg/l IONPs 300 mg/l IONPs 500 mg/l IONPs Centrifuged fishpond water

Fig. 7. The images of the fishpond water sample, treated fishpond water by different surface functionalized IONPs concentration, and centrifuged fishpond water for the (i) LGMS and (ii) HGMS.

You might also like