j ourna l homepage: www.e lsev ie r .com/ locate /desa l
Magnetic core-hydrophilic shell nanosphere as stability-enhanced drawsolute for forward osmosis (FO) application
Sung Yong Park a,1, Hyo-Won Ahn a,1, Jae Woo Chung b, Seung-Yeop Kwak a,⁎a Department of Materials Science and Engineering, Seoul National University, 599 Gwanak-ro, Gwanak-gu, Seoul 151-744, Republic of Koreab Department of Organic Materials and Fiber Engineering, Soongsil University, 369 Sangdo-ro, Dongjak-gu, Seoul 156-743, Republic of Korea
H I G H L I G H T S G R A P H I C A L A B S T R A C T
• Preparation of magnetic core-hydro-philic shell nanosphere
• Core-shell magnetic nanosphere appliesto durable draw solute in FO.
• Magnetic draw solutes generate reason-able osmotic pressures.
Article history:Received 2 March 2016Received in revised form 13 June 2016Accepted 17 June 2016Available online xxxx
We have developed magnetic core-hydrophilic shell nanospheres as a draw solute with enhanced stability foruse in forward osmosis (FO) processes, which were prepared via a ligand exchange reaction. The combined re-sults of TEM, DLS, FT-IR, and TGA analysis indicated that spherical magnetic nanosphereswith 10 nm in diameterwere successfully prepared via a thermal decompositionmethod. Hydrophilic shell layers were prepared using aligand exchange reaction, and thehydrophilic siloxane agentswere robustly bound to the surfaces of themagnet-ic nanospherewithout inducingmorphological changes. The number of hydrophilic agents presented in themag-netic nanospheres increased dramatically as a result of the covalently attached hydrophilic siloxane agents. Themagnetic core-hydrophilic shell nanosphere draw solutes generated reasonable osmotic pressures due to theirhydrophilic shell layer, rendering them useful for purifying mildly brackish water. As we intended, the osmoticpressure of the magnetic core-hydrophilic shell nanosphere was well maintained during repetitive magnetic re-covery processes because the robustly bound the hydrophilic shell layer prevented particle aggregation.
Fresh water is in short supply due to the worldwide growth in de-mand for residential, agricultural, and industrial water uses. The freshwater shortage is considered one of the most pressing challenges ofthe 21st century [1–3]. Water shortage problems may potentially be
23S.Y. Park et al. / Desalination 397 (2016) 22–29
alleviated by implementing membrane processes to purify undrinkablewater resources, such as sea water, brine, or contaminated water.Among various membrane processes, forward osmosis (FO) membraneprocess was recently recognized as promising approaches to watertreatment applications [4–7] due to their low energy consumptionand low membrane fouling tendency. FO processes have been studiedin a variety of research fields, including desalination [8], wastewatertreatment [9–11], osmotic pressure generation [12], agricultural fertiliz-ing [13], and food processing [14]. However, FO processes tend to belimited in their applicability to commercial water treatment applica-tions due to their low water permeation performance and poor separa-tion of highly purified water from the permeate. The issues with FOprocesses must be addressed for the development of a suitable drawsolute.
The main requirements for draw solutes are generation of reason-able osmotic pressures and the easy recovery of draw solutes from thepermeate to obtain purified water. To satisfy these requirement, a vari-ety of draw solutes have been tried to apply FO process such as inorgan-ic salts [15], thermolytic/volatile solutes [16,17], organic solutes [18,19],polyelectrolytes [20], hydrogels [21,22], or solvents with a switchablepolarity [23,24]. These materials can generate osmotic pressures by dis-solving or dispersing the permeate, and they can be recovered from thepermeate through a variety of treatments, such as chemical precipita-tion [25], pressure control [26], thermal decomposition/evaporation[27], or membrane filtration [28,29]. However, these attempts havebeen limited to FO processes due to their low osmotic driving force, dif-ficulties associatedwith separating the particles from the permeate, thegeneration of toxic thermolytic products, the high energy requirementsof the draw solute recovery process, and the inability to reuse themate-rials [5,6].
Magnetic nanoparticles were considered as potential draw solutes[30–34] because they can be readily separated from a permeate by anexternal magnetic field without supply of chemical and thermal energy.The advantage of magnetic draw solute is easy separation from purifiedwater and reuse of draw solute without any post treatments, which canreduce FO operation cost by suppression of supplying disposable drawsolute. The osmotic pressure generated by the magnetic nanoparticlesare enhanced by a variety of surface functional agents, such as polyacrylic acid (PAA), polyethylene glycol (PEG), or diacid, applied as coat-ings onto the magnetic nanoparticles [30]. The hydrophilic modifiedmagnetic nanoparticles (hydrophilic-MNs) showed reasonable waterfluxes up to 10.4 Lm−2 h−1 [32]. However, the hydrophilic-MNs tendto be aggregated under the high strength magnetic field during thedraw solute recovery process due to a lack of binding force betweenthe hydrophilic agents and the surface of magnetic nanoparticle [35,36]. Aggregated Hydrophilic-MNs can significantly reduce the total os-motic pressure in an FO process [30], thereby hindering the utility ofthe Hydrophilic-MN draw solutes.
Here, we describe the synthesis of magnetic core-hydrophilic shellnanospheres prepared by a ligand exchange reaction between carboxyl-ate stabilizers ofmagnetic nanosphere surface and hydrophilicmodifiedsiloxane ligands to form a covalent linkage between magnetic nano-sphere core and hydrophilic shell for prevention of particle aggregation.Two hydroxyl-modified siloxane ligands were used: N-(trimethoxysilylpropyl)ethylenediamine triacetic acid, denotedSiCOOH, to introduce carboxylic acid groups into the magnetic nano-sphere, 2-[methoxy-(polyethyleneoxy)propyl] trimethoxysilane, de-noted SiPEG, to introduce the hydrophilic PEG chain. As we intended,the core-shell magnetic nanospheres preserved their particle size andtheir ability to generate an osmotic pressure after repetitive magneticrecycling processes due to the covalent bonding of hydrophilic siloxaneagents. These efforts indicate the potential of the core-shell magneticnanospheres as feasible draw solutes that are readily recovered throughmagnetic forces and offer a consistent FO performance by preventingmagnetic nanoparticle aggregation through the presence of robust co-valent bondings among the hydrophilic shell layers.
2. Experimental methods
2.1. Materials
Iron(III) acetylacetonate (Fe(acac)3, 97%), oleic acid (90%),oleylamine (70%), toluene (97%), 1,2-hexadecanediol (90%), benzylether (99%), ethanol (99.9% anhydrous), and tetramethylammoniumhydroxide (TMAH, 99%) were purchased from Sigma-Aldrich (St Louis,MO, USA). N-(trimethoxysilylpropyl) ethylenediamine triacetic acid(MW: 462.41 g mol−1, SiCOOH) and 2-[methoxy-(polyethyleneoxy)6–9propyl] trimethoxysilane (MW: 459–591 g mol−1, SiPEG) were pur-chased from ABCR GmbH & Co. (Karlsruhe, Germany). The cellulose tri-acetate (CTA) membranes used in the FO process were purchased fromHydration Technologies Inc. (HTI) (OR, USA). All chemicalswere used asreceived,without further purification. Aqueous solutionswere preparedwith deionized (DI) water with a resistivity exceeding 18 MΩ cm−1.
2.2. Preparation of magnetic core-hydrophilic shell nanosphere
Magnetic core-hydrophilic shell nanosphere draw solutes were pre-pared via a ligand exchange reaction, as reported by De Palma et al. [37].First, magnetic nanosphere (MN) stabilized with oleic acid (O-MN)wassynthesized by the thermal decomposition method [38]. 1.5 g iron(III)acetylacetonate, 5 g 1,2-hexadecanediol, 5 mL oleylamine, and 5 mLoleic acid were mixed in 40 mL benzyl ether. The mixture was heatedto 200 °C for 2 h, and the temperaturewas then raised to 300 °Cwith re-flux for 1 h under an Ar environment. Next, the solution was cooleddown to room temperature. The oleic acid stabilizer was exchangedwith the siloxane agents, by adding 2 mL SiPEG (or SiCOOH) and0.05 mL TMAH to the O-MN/toluene solution prepared by dispersing100 mg O-MN in 300 mL toluene. The mixtures were agitated in an or-bital shaker for 72 h. The magnetic core-hydrophilic shell nanospheres(black-brown colored precipitates) were obtained by washing the pre-cipitate several times with toluene. Finally, the core-shell MNs werestored in deionized water prior to use. Fig. 1 shows a schematic illustra-tion of the ligand exchange reaction on the NNs. The core-shell MNswere denoted SiPEG-MN and SiCOOH-MN, in reference to the siloxaneligands.
2.3. Characterization of magnetic core-hydrophilic shell nanosphere
The shape and size distribution of the O-MNs and core-shell MNswere observed using energy-filtering transmission electron microscopy(EF-TEM) (Carl Zeiss LIBRA 120) and dynamic light scattering (DLS)(Otsuka Electronics ELSZ-1000) techniques. The samples used for theEF-TEM observationswere prepared by dispersing theMNs in n-hexaneto prepare O-MN or DI water to prepare the core-shell MNs. The MNs-dispersed solution was then dropped onto carbon TEM grids and driedunder vacuum overnight. The particle size and MNs distribution weremeasured by dispersing 1 mg of the MNs in 5 mL n-hexane to prepareO-MN or 5 mL DI water to prepare core-shell MNs.
The surface modifications of the core-shell MNs were characterizedusing Fourier transform infrared spectroscopy (FT-IR) (Thermo Scientif-ic Nicolet 6700 IR) and thermogravimetric analysis (TGA) (TA instru-ments Q500). The FT-IR analysis was conducted by preparing a KBrpellet containing small amounts of MNs. The FT-IR spectral range was4000 to 400 cm−1, with a 4 cm−1 spectral resolution. The TGA analysiswas conducted by drying 10 mg of each sample in a vacuum overnightat room temperature. The samples were heated from 25 °C to 800 °Cat a 20 °C/min heat rate under a N2 atmosphere. The ligand/particleratio from the combined results of the TGA and DLS analyses were cal-culated using the following equation [30].
N ¼ϖNAρ
43πR3 � 10−23
MW; ð1Þ
Fig. 1. Schematic illustration of ligand exchange reaction of magnetic nanosphere stabilized with oleic acid using hydrophilic siloxane ligands between carboxylate and siloxane ligands.
24 S.Y. Park et al. / Desalination 397 (2016) 22–29
whereN is the number of ligands on each particle (ligand/particle ratio),ω is the percentweight loss,NA is Avogadro's number, ρ is the density ofthe nanospheres, approximated as 5.1 g/cm3 [39], R is the mean radiuscalculated from the DLS analysis (cm), andMW is themolecular weightof the ligand molecules (g mol−1).
The osmotic pressure generated by the core-shell MNs was mea-sured via freezing point depression osmometry (KNAUSER Semi-microosmometer K-7400). The osmotic pressure generated by the core-shellMNs was calculated based on a rearranged expression of the van'tHoff equation.
Π ¼ iMRT; ð2Þ
where Π is the osmotic pressure, i is the van't Hoff factor,M is the mo-larity (mol L−1),R is the gas constant (0.08206 L atmK−1·mol−1), and Tis the thermodynamic temperature (K).
2.4. Evaluation of the water permeability during the FO process
The core-shell MNs were used as draw solutes in a simple FO pro-cess. A wetted FO membrane (HTI flat sheet CTA membrane, effectivearea: 2.54 cm2) was placed between equal volumes (120 mL) of the DI
Fig. 2. Schematic illustration for FO filtrat
water feed solution and the core-shell MN aqueous draw solution. Theactive layer of CTA membrane was oriented toward the feed solution.The concentration of the core-shell MN draw solution was varied be-tween 20 g L−1 and 50 g L−1. During the 90 min FO process, the drawsolution volume increase was recorded at 30 min intervals at roomtemperature.
To evaluate water flux under mild brackish conditions, we preparedfeed solution by dissolving methylene blue into DI water with variousconcentrations (500, 1000, and 5000 ppm) of which the osmotic pres-sure was under 1 atm. In case of use inorganic salt for preparation offeed solution, some of ions were penetrated over membrane, resultingin influenced osmotic gradient between feed and permeate solutions.Methylene blue does not penetrate over the TCA membrane due to itsmolecular weight (M.W.: 379.8 g mol−1). The pH condition for methy-lene blue feed solution and core-shell MN magnetic draw solutionswere recorded as 6.8 for methylene blue feed solution, 6.3 for SiPEG-MNdraw solution, and11.3 for SiCOOH-MNdrawsolution, respectively.
The water flux in the FO process was calculated according to theequation:
Jv ¼ ΔV=AΔt ð3Þ
ion and magnetic recycle processes.
25S.Y. Park et al. / Desalination 397 (2016) 22–29
where Jv is the water flux. The units of the water flux were L m−2 h−1,abbreviated LMH. ΔV is the volume change in the draw solution, Δt isthe predetermined time for permeation, and A is the effective mem-brane surface area.
2.5. Evaluation of the particle stability during the magnetic recyclingprocess
The stability of the magnetic draw solutes were determined using arepetitivemagnetic recycling test carried out using a 50 g L−1 core-shellMNs dispersed aqueous solution. The particle stability was evaluated byvarying the pure water flux and the mean particle diameter after eachmagnetic recycling process. The water flux was recorded during the
FO process, and the core-shell MNs were separated from the permeateusing a permanent magnet with a magnetic field strength exceeding13,000 G for 15–20 min. The magnetic recovery efficiency was almost100% for both core-shell MN draw solutions. The particles were thendried under vacuum at room temperature overnight. The driedcore-shell MNs were re-dispersed in DI water and adjusted to a50 g L−1 concentration. The magnetic recycling process was repeat-ed 5–8 times. Fig. 2 shows the schematic illustration for FO filtrationand magnetic recycle processes. After each recovery process, the re-covered core-shell MNs were used in an FO process to measure thewater flux, and the mean particle size was analyzed by EF-TEM andDLS to determine the variation in the particle size distribution duringmagnetic separation.
s) and (right) particle size distribution evaluated byDLS analysis of (a)O-MN, (b) SiCOOH-
Fig. 5. TGA results formagnetic nanoparticles of (a) O-MN, (b) SiCOOH-MN and (c) SiPEG-MN.
Fig. 4. FT-IR spectra of (a) O-MN, (b) SiCOOH-MN and (c) SiPEG-MN.
26 S.Y. Park et al. / Desalination 397 (2016) 22–29
3. Results and discussion
3.1. Magnetic core-hydrophilic shell nanospheres
Magnetic nanoparticles were used in the FO process as a draw soluteto facilitate their separation from the permeate during repetitiverecycling processes through the application of an external magneticfield [30]. The typical magnetic draw solute tended to form aggregates,however, during the later magnetic separation cycles due to loss of hy-drophilic stabilizer from the surface of magnetic nanomaterial. In thisstudy, themagnetic nanospheres (MNs)were robustly bound to the hy-drophilic stabilizer through a ligand exchange reaction. The magneticcore-hydrophilic shell nanospheres (core-shell MNs) were synthesizedby replacing the oleic acid stabilizer on the MN with hydrophilic silox-ane capping agents via a ligand exchange reaction under basic condi-tion. Fig. 3 shows the TEM and DLS results obtained from the O-MNsand core-shellMNs. TheO-MNswere spherical in shapewith a diameterof 10.3 ± 2.0 nm. The core-shell MNs preserved their original sphericalshape and size during the ligand exchange reactionwithout undergoingsignificant deformations. The particle size increased slightly from10.3 ± 2.0 nm to 12.7 ± 2.5 (SiPEG-MN) and 13.6 ± 2.4 nm(SiCOOH-MN), respectively.
The chemical composition of the core-shellMNswas analyzed by FT-IR spectroscopy, as shown in Fig. 4, themajor FT-IR bands correspondingto the oleic acid groups in the O-MN, e.g., C_C stretch disappeared afterthe ligand exchange reaction had gone to completion. The siloxane IRbands were observed near 1030 cm−1 (for Si-O-Fe) and 1250 cm−1
(for Si-C) in the FT-IR spectra of core-shell MNs. As the carboxylated si-loxane bound to the MNs, the IR bands corresponding to the carboxylicacid groups appeared at 1620 cm−1, corresponding to the carbonylstretch, and at 3600–2700 cm−1, corresponding to the OH stretch. Sim-ilarly, the C\\O stretch (1040–1120 cm−1) and OH stretch (3600–3300 cm−1) were observed in the FT-IR spectra of SiPEG-MN upon ex-change of the surface stabilizers, replacing the oleic acidwith PEG-silox-ane. The FT-IR spectra demonstrated that the hydrophilic surfacemodification reactions proceeded successfully, and hydrophilic siloxaneligands thoroughly covered the surfaces of the core-shell MNs.
The assembly of the surface siloxane agents on the MNs was con-firmed by quantitatively analyzing method using TGA. As shown inFig. 5, the TGA analysis results revealed that the weight loss in theMNs from room temperature to 800 °C resulted from the decomposition
of the ligand molecules. The ligand molecule weight loss was measuredbased on the percent difference between the weights measured beforeand after the primary weight loss peaks. As the ligand exchangeproceeded, the weight loss in the core-shell MNs samples increased.The O-MN sample (Fig. 5(a)) lost about 20% of its weight whereas theSiCOOH-MN (Fig. 5(b)) and SiPEG-MN (Fig. 5(c)) samples lost 42.5%and 57.2% of their weights, respectively.
The TGA and DLS results were used to calculate the ligand/particleratio according to Eq. (1). The mean radii of the MNs were 5.1 nm (O-MN), 6.3 nm (SiCOOH-MN), and 6.8 nm (SiPEG-MN) based on the DLSanalysis. The calculated ligand/particle ratios of the O-MN, SiCOOH-MN, and SiPEG-MN samples were 1269, 4261, and 5250, respectively.These results indicated that greater amounts of hydrophilic siloxaneagents were bound to the surfaces of the MNs through the ligand ex-change reaction than that for the O-MN. The difference between thequantities of surface hydrophilic siloxane agents bound to the surfacesof the core-shell MNs might be related to the packing structure of thehydrophilic agents. The SiCOOH molecules form multi-branched struc-tures composed of three of carboxyl functional groups, which can intro-duce steric hindrance and electronic repulsion. These effects coulddisrupt the efficient packing of the SiCOOH ligands on the particle sur-face. By contrast, SiPEG consists of polyethylene glycol chains that caneffectively pack onto the MN surfaces.
The results obtained from HR-TEM, DLS, FT-IR, and TGA analysis in-dicated that modifying the surfaces of the magnetic nanoparticles withthe hydrophilic siloxanes provided robust interfacial bondings for thehydrophilic surface functional groups without introducingmorphologi-cal changes.
3.2. Osmotic pressure
The osmotic pressure generated by the core-shell MNs aqueous so-lution indicated that the hydrophilic shells on the MN surface werethe major components driving the osmotic pressure of the draw solu-tion. The osmotic pressure due to the core-shell MNs was measuredbased on freezing point depression osmometry techniques using a vari-ety of concentrated core-shell MN solutions (10 g L−1–50 g L−1, see Fig.6). Beyond amagnetic core-shell MN solution concentration of 50 g L−1,the MNs aggregated and precipitated in an aqueous solution. The os-motic pressure of O-MN could not be measured because it did not dis-perse in water. Fig. 6 shows that the osmotic pressures of both core-shell MN solutions increased linearly with the solution concentration.The maximum osmotic pressures of SiPEG-MN and SiCOOH-MN were7.6 and 6.3 atm, at 50 g L−1 concentrations.
These results revealed that a 50 g L−1 core-shell MN draw solutioncould be used in an FO process applied to mild brackish water with
Fig. 7.Water flux in FO process during 90minwith 20 g L−1 and 50 g L−1 concentration ofcore- shell MN draw solution.
Fig. 6. Osmotic pressure of magnetic core-hydrophilic shell nanospheres with various concentrations in aqueous solution.
27S.Y. Park et al. / Desalination 397 (2016) 22–29
ion concentration below 5000 ppm because their osmotic pressureswere under 4 atm. In particular, the SiPEG-MN solution generated alarger osmotic pressure than the SiCOOH-MN solution at comparableconcentrations. It seemed that generation of osmotic pressure wasmore influenced by number of surface hydrophilic agents than thenumber of ionic groups. The SiCOOH-MN had more ionic groups i.e.,three carboxylates, and less surface ligand ratio than that for theSIPEG-MN. In this study, the osmotic pressure of SiPEG-MN at50 g L−1 concentration was 20.6% more than that of the SiCOOH-MNat the same concentration. It was well consistence the ligand/particleratio of core-shell MNs. As mentioned in the previous section, theSiPEG-MN incorporated 23% more hydrophilic surface agents thanthat for the SiCOOH-MN. The osmotic pressuremeasurements indicatedthat theMNhydrophilic siloxane surfacemodifications generated an os-motic pressure sufficient to generate a driving force in an FO process.
3.3. The water flux in the FO process, and particle stability of the magneticcore-hydrophilic shell nanosphere draw solute during the magneticrecycling process
The pure water permeability in an FO process conducted using thecore-shell MN draw solution was measured using 20 g L−1 and50 g L−1 concentrated draw solutions. Fig. 7 plots the water flux in anFO process over 90 min. As the concentration of the core-shell MNdraw solution increased, the pure water flux increased. The purewater flux through the 20 g L−1 SiCOOH-MN and SiPEG-MN aqueoussolutions was 1.22 and 1.1 LMH, respectively. SiPEG-MN and SiCOOH-MN draw solutions containing 50 g L−1 core-shell MN yielded waterfluxes of 2.13 and 1.81 LMH. The 20 g L−1 draw solution yielded alower flux than the 50 g L−1 draw solution. Higher draw solution con-centrations were maintained during the water permeation processes,whereas low draw solution concentrations easily decreased furtherupon dilution. The osmotic pressure measurements obtained from thecore-shell MN draw solutions revealed that SiPEG-MN yielded a greaterosmotic pressure than the SiCOOH-MN solution. The pure water fluxthrough the SiPEG-MN draw solute wasmore 10% greater than that ob-tained from the SiCOOH-MN draw solution.
The stability of the core-shell MNs was measured during repetitivemagnetic recycling process. Fig. 8 shows the changes in the core-shellMN draw solutes, in terms of the water flux and particle size, measuredafter each magnetic recycling step. The concentrations of the core-shellMN aqueous solutions were maintained at 50 g L−1. The water fluxesthrough both of the core-shell MN solutions were preserved during 3
repetitive recycling processes. The water fluxes through the SiPEG-MN(Fig. 8(a)) and SiCOOH-MN (Fig. 8(b)) solutions were 2.01 ± 0.12LMH and 1.69 ± 0.11 LMH, respectively. The mean particle sizes forboth core-shell MNs were also preserved, and no particle aggregationwas observed during 3 repetitive magnetic recycling. However, the ag-gregation of SiCOOH-MN draw solute was observed after 4th magneticrecycling due to occurrence of particle aggregation. The fluxes werealso rapidly declined by loss of particle stability of SiCOOH-MN. TheSiCOOH-MN has three carboxylate groups which can form hydrogenbonding with other carboxylate groups in neighborhood SiCOOH-MNsunder strong magnetic recycling and drying process.
By contrast, the SiPEG-MN draw solution maintained water flux,2.13 LMH for initial flux-1.89 LMH over 8 repetitive magnetic recyclingprocesses because the SiPEG-MN did not undergo particle aggregation.Its particle diameter was changed within 10% from 10.3 nm to11.4 nm. This strong stability was derived from covalent linkage of hy-drophilic surface layer and steric hindrance effect without ionic interac-tion between magnetic particles.
These results indicated that the SiPEG-MNwere sufficiently stable toprevent particle core aggregation during themagnetic separation steps.Previous reported recycling test results obtained from non-covalentlymodified hydrophilic MN solutes [30] revealed that the water fluxtended to decrease and the mean particle size increased continually
Fig. 8. The MN stability of (a) SiPEG-MN and (b) SiCOOH-MN for during repetitivemagnetic recovery process through variation of water flux and mean particle size.
28 S.Y. Park et al. / Desalination 397 (2016) 22–29
during repetitive recycling processes. The mean particle size was 50%larger after the recycling process due to the weak bonds between thehydrophilic surface agents on theMN surface. The particle stabilitymea-surements collected during the recycling process revealed that the co-valent interactions between the hydrophilic siloxane agents and the
Fig. 9.Water flux for SiPEG-MN draw solute during FO process with various concentratedbrackish feed solution prepared by dissolution of methylene blue into deionized water.
MN surfaces were robust, endowing the particles with sufficient stabil-ity for use in magnetically driven recyclable FO processes.
The draw solute stabilities and the pure water permeabilities led usto select the SiPEG-MN as a feasible draw solutewith high osmotic pres-sure generation and superior stability during the recycling processes as-sociated with FO processes applied to brackish water. Fig. 9 shows thewater fluxes measured using 50 g L−1 SiPEG-MN draw solutions and abrackish feed solution composed of 500, 1000, or 5000 ppmmethyleneblue aqueous solution use as brackish feed water. The water fluxes ofthe low brackish water were maintained at 500 ppm or 1000 ppmmethylene blue aqueous solution. Even the highly concentrated brack-ish water, 5000 ppm methylene blue aqueous solution, yielded awater flux that was around 40% of the value obtained from purewater. These results indicated that the SiPEG-MN draw solutes couldbe used in an FO water treatment process applied to mildly brackishwater.
4. Conclusions
We demonstrated the use of magnetic core-hydrophilic shell nano-spheres as magnetic recyclable draw solutes in an FO process. Thecore-shell MNs were synthesized through a ligand exchange reactionbetween the carboxylate groups of oleic acid-stabilized MN and hydro-philic siloxane ligands to prevent aggregation during the magnetic re-covery process. The combined results of TEM, DLS, FT-IR, and TGAanalysis revealed that spherical magnetic nanoparticles with 10 nm indiameter were successfully prepared, and the hydrophilic siloxaneagents robustly bound to the surfaces of theMNwithout inducingmor-phological changes during the ligand exchange reaction. The core-shellMN draw solutes generated a reasonable osmotic pressure that can beapplied to purifymildly brackishwater. As intended in thiswork, the co-valently bound surface hydrophilic agents ensured a high level of parti-cle stability during the repetitive magnetic recovery processes. TheSIPEG-MN generated a larger osmotic pressure and provided higherparticle stability during the magnetic recycling processes compared tothe values obtained from the SiCOOH-MN due to differences in themo-lecular structures of siloxane ligands. Our approach provides a repetitiverecyclable draw solute performed well due to the covalent attachmentof hydrophilic agents to the nanoparticle surfaces.
Acknowledgments
This researchwas supportedbyBasic ScienceResearchProgramthroughthe National Research Foundation of Korea (NRF) funded by theMinistry ofScience, ICT & Future Planning (No. NRF-2015R1A2A2A01005651).
References
[1] M.A. Montgomery, M. Elimelech, Water and sanitation in developing countries: in-cluding health in the equation - millions suffer from preventable illnesses and dieevery year, Environ. Sci. Technol. 41 (2007) 17–24.
[2] R.L. McGinnis, M. Elimelech, Global challenges in energy and water supply: thepromise of engineered osmosis, Environ. Sci. Technol. 42 (2008) 8625–8629.
[3] M.A. Shannon, P.W. Bohn, M. Elimelech, J.G. Georgiadis, B.J. Marĩas, A.M. Mayes, Sci-ence and technology for water purification in the coming decades, Nature 452(2008) 301–310.
[4] T.Y. Cath, A.E. Childress, M. Elimelech, Forward osmosis: principles, applications, andrecent developments, J. Membr. Sci. 281 (2006) 70–87.
[5] C. Klaysom, T.Y. Cath, T. Depuydt, I.F.J. Vankelecom, Forward and pressure retardedosmosis: potential solutions for global challenges in energy andwater supply, Chem.Soc. Rev. 42 (2013) 6959–6989.
[6] D.L. Shaffer, J.R. Werber, H. Jaramillo, S. Lin, M. Elimelech, Forward osmosis: whereare we now? Desalination 356 (2015) 271–284.
[7] T.Y. Cath, A.E. Childress, M. Elimelech, Forward osmosis: principles, applications, andrecent developments, J. Membr. Sci. 281 (2006) 70–87.
[8] B. Van Der Bruggen, L. Lejon, C. Vandecasteele, Reuse, treatment, and discharge ofthe concentrate of pressure-driven membrane processes, Environ. Sci. Technol. 37(2003) 3733–3738.
[9] Y.J. Choi, J.S. Choi, H.J. Oh, S. Lee, D.R. Yang, J.H. Kim, Toward a combined system offorward osmosis and reverse osmosis for seawater desalination, Desalination 247(2009) 239–246.
29S.Y. Park et al. / Desalination 397 (2016) 22–29
[10] C.H. Tan, H.Y. Ng, A novel hybrid forward osmosis - nanofiltration (FO-NF) processfor seawater desalination: draw solution selection and system configuration,Desalin. Water Treat. 13 (2010) 356–361.
[11] F. Zhang, K.S. Brastad, Z. He, Integrating forward osmosis into microbial fuel cells forwastewater treatment, water extraction and bioelectricity generation, Environ. Sci.Technol. 45 (2011) 6690–6696.
[12] A. Achilli, T.Y. Cath, A.E. Childress, Power generation with pressure retarded osmo-sis: an experimental and theoretical investigation, J. Membr. Sci. 343 (2009) 42–52.
[13] S. Phuntsho, H.K. Shon, S. Hong, S. Lee, S. Vigneswaran, A novel low energy fertilizerdriven forward osmosis desalination for direct fertigation: evaluating the perfor-mance of fertilizer draw solutions, J. Membr. Sci. 375 (2011) 172–181.
[14] K.B. Petrotos, P.C. Quantick, H. Petropakis, Direct osmotic concentration of tomatojuice in tubular membrane–module configuration. II. The effect of using clarified to-mato juice on the process performance, J. Membr. Sci. 160 (1999) 171–177.
[15] A. Achilli, T.Y. Cath, A.E. Childress, Selection of inorganic-based draw solutions forforward osmosis applications, J. Membr. Sci. 364 (2010) 233–241.
[16] R.L. McGinnis, J.R. McCutcheon, M. Elimelech, A novel ammonia–carbon dioxide os-motic heat engine for power generation, J. Membr. Sci. 305 (2007) 13–19.
[17] K. Katsikaris, C. Boukouvalas, K. Magoulas, Simulation of ultrafiltration process andapplication to pilot tests, Desalination 171 (2005) 1–11.
[18] P. McCormick, J. Pellegrino, F. Mantovani, G. Sarti, Water, salt, and ethanol diffusionthrough membranes for water recovery by forward (direct) osmosis processes, J.Membr. Sci. 325 (2008) 467–478.
[19] J.O. Kessler, C.D. Moody, Drinking water from sea water by forward osmosis, Desa-lination 18 (1976) 297–306.
[20] Q. Ge, J. Su, G.L. Amy, T.-S. Chung, Exploration of polyelectrolytes as draw solutes inforward osmosis processes, Water Res. 46 (2012) 1318–1326.
[21] D. Li, X. Zhang, J. Yao, G.P. Simon, H. Wang, Stimuli-responsive polymer hydrogels asa new class of draw agent for forward osmosis desalination, Chem. Commun. 47(2011) 1710–1712.
[22] D. Li, X. Zhang, J. Yao, Y. Zeng, G.P. Simon, H.Wang, Composite polymer hydrogels asdraw agents in forward osmosis and solar dewatering, Soft Matter 7 (2011)10048–10056.
[23] A.D. Wilson, F.F. Stewart, Structure-function study of tertiary amines as switchablepolarity solvents, RSC Adv. 4 (2014) 11039–11049.
[24] M.L. Stone, C. Rae, F.F. Stewart, A.D.Wilson, Switchable polarity solvents as draw sol-utes for forward osmosis, Desalination 312 (2013) 124–129.
[25] R. Alnaizy, A. Aidan, M. Qasim, Copper sulfate as draw solute in forward osmosis de-salination, J. Environ. Chem. Eng. 1 (2013) 424–430.
[26] A. Razmjou, G.P. Simon, H. Wang, Effect of particle size on the performance of for-ward osmosis desalination by stimuli-responsive polymer hydrogels as a drawagent, Chem. Eng. J. 215 (2013) 913–920.
[27] R.L. McGinnis, N.T. Hancock, M.S. Nowosielski-Slepowron, G.D. McGurgan, Pilotdemonstration of the NH 3/CO 2 forward osmosis desalination process on high sa-linity brines, Desalination 312 (2013) 67–74.
[28] S. Zhao, L. Zou, D. Mulcahy, Brackish water desalination by a hybrid forward osmo-sis–nanofiltration system using divalent draw solute, Desalination 284 (2012)175–181.
[29] J. Su, T.-S. Chung, B.J. Helmer, J.S. de Wit, Enhanced double-skinned FO membraneswith inner dense layer for wastewater treatment and macromolecule recycle usingsucrose as draw solute, J. Membr. Sci. 396 (2012) 92–100.
[30] Q. Ge, J. Su, T.S. Chung, G. Amy, Hydrophilic superparamagnetic nanoparticles: syn-thesis, characterization, and performance in forward osmosis processes, Ind. Eng.Chem. Res. 50 (2011) 382–388.
[31] H. Bai, Z. Liu, D.D. Sun, Highly water soluble and recovered dextran coated fe 3o 4magnetic nanoparticles for brackish water desalination, Sep. Purif. Technol. 81(2011) 392–399.
[32] M.M. Ling, K.Y. Wang, T.S. Chung, Highly water-soluble magnetic nanoparticles asnovel draw solutes in forward osmosis for water reuse, Ind. Eng. Chem. Res. 49(2010) 5869–5876.
[33] M.M. Ling, T.-S. Chung, X. Lu, Facile synthesis of thermosensitive magnetic nanopar-ticles as “smart” draw solutes in forward osmosis, Chem. Commun. 47 (2011)10788–10790.
[34] M.M. Ling, T.-S. Chung, Novel dual-stage FO system for sustainable protein enrich-ment using nanoparticles as intermediate draw solutes, J. Membr. Sci. 372 (2011)201–209.
[35] Q. Ge, J. Su, T.-S. Chung, G. Amy, Hydrophilic superparamagnetic nanoparticles: syn-thesis, characterization, and performance in forward osmosis processes, Ind. Eng.Chem. Res. 50 (2010) 382–388.
[36] M.M. Ling, K.Y. Wang, T.-S. Chung, Highly water-soluble magnetic nanoparticles asnovel draw solutes in forward osmosis for water reuse, Ind. Eng. Chem. Res. 49(2010) 5869–5876.
[37] R. De Palma, S. Peeters, M.J. Van Bael, H. Van den Rul, K. Bonroy, W. Laureyn, J.Mullens, G. Borghs, G. Maes, Silane ligand exchange to make hydrophobicsuperparamagnetic nanoparticles water-dispersible, Chem. Mater. 19 (2007)1821–1831.
[38] S. Sun, H. Zeng, D.B. Robinson, S. Raoux, P.M. Rice, S.X. Wang, G. Li, MonodisperseMFe2O4 (M = Fe, Co, Mn) nanoparticles, J. Am. Chem. Soc. 126 (2004) 273–279.
[39] L. Amirkhani, J. Moghaddas, H. Jafarizadeh, Effect of surface modification agent andcalcination process on the preparation of the hydrophobic magnetic silica aerogelsby ambient pressure drying method, Bulg. Chem. Commun. 47 (2015) 82–88.
