PUc

NM

h

••••

a

ARR1AA

KBMUSN

1

mmtstve

hao

0h

Journal of Hazardous Materials 267 (2014) 221– 228

Contents lists available at ScienceDirect

Journal of Hazardous Materials

jou rn al hom epage: www.elsev ier .com/ locate / jhazmat

reparation, characterization and application of NaHCO3 leached bulk(VI) imprinted polymers endowed with �-MPS coated magnetite inontaminated water

ikita Tawanda Tavengwa, Ewa Cukrowska, Luke Chimuka ∗

olecular Sciences Institute, School of Chemistry, University of Witwatersrand, Private Bag 3, Johannesburg 2050, South Africa

i g h l i g h t s

Highly selective magnetic ion imprinted polymer for uranium was prepared.Prepared polymer was studied for its binding and application to wastewater samples.Though capacity was low, it is very selective towards uranium ions.It has also fast kinetics with equilibrium reached within 45 min.

r t i c l e i n f o

rticle history:eceived 7 August 2013eceived in revised form6 December 2013ccepted 17 December 2013vailable online 3 January 2014

a b s t r a c t

A new type of ion imprinted polymer (IIP) embedded with �-methacryloxypropyltrimethoxysilane(�-MPS) coated magnetic particles for selective extraction of uranyl ions was synthesized by bulk poly-merization. The performance of the magnetic sorbent on the extraction of U(VI) was evaluated byvarious parameters which included the influence of pH, amount of the magnetic polymers, contacttime and initial U(VI) concentration. The adsorption capacity of the magnetic polymers was found tobe 1.1 and 0.95 mg g−1 for the IIP and its control ion non-imprinted polymer (NIP), respectively. The

−1

eywords:ulk polymerizationagnetic imprinted polymerraniumelectiveaHCO3

optimum amount of the sorbent was 50 mg for an initial concentration of 2.5 mg L . The preparedmagnetic imprinted sorbent displayed an uptake with a time of 45 min considered optimum. The mag-netic polymer displayed good selectivity and exhibited good reusability. Studies from binary mixturesof metal ions in aqueous solutions showed that the magnetic adsorbent selectivity following the order:U(VI) > Ni(II) > Pb(II). Experimental results illustrated the potential application of magnetic ion imprintedpolymer sorbents for selective removal of U(VI) from contaminated water.

. Introduction

The main use of uranium in the civilian sector is to fuel com-ercial nuclear power plants as it is commonly understood as aaterial storing huge amounts of energy [1]. Some uranium quan-

ities have been detected in nuclear fuel effluents, mine tailings,eawater, and other sources. Because of the potential environmen-al health threat and non-renewable resource of nuclear energy, aariety of methods have been developed for its removal and recov-ry.

Health wise, approximately 90 �g of uranium exists in the

uman body [2], however, excessive exposure to uraniumnd its compounds cause kidney toxicity. Virtually all thebserved or expected effects of uranium are from nephrotoxicity

∗ Corresponding author. Tel.: +27 11717 6703; fax: +27 11717 6749.E-mail address: luke.chimuka@wits.ac.za (L. Chimuka).

304-3894/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.jhazmat.2013.12.053

© 2013 Elsevier B.V. All rights reserved.

associated with deposition in the kidney tubules and glomerulidamage [3]. The WHO (World Health Organization), Health Canadaand Australian drinking water guidelines fixed the maximum ura-nium concentration in drinking waters to be less than 9, 20 and20 �g L−1, respectively.

The geochemical speciation of uranium influences its solubility,mobility and biological availability in the environment. Geochem-ical speciation information on uranium has been used to evaluatepotential in situ remediation approaches [4–6]. The need to deter-mine different species of uranium in environmental and biologicalmaterials is therefore important since the effects or toxicity of thiselement and its behavior depends, to a great extent on its chemicalform and concentration [7].

Many techniques have been tried and applied to the extrac-

tion of uranium and other heavy metal pollutants. These methodsinclude chemical precipitation, ion exchange, membrane tech-nologies and adsorption [8–10]. Of these methods, adsorptionis regarded as one of the most efficacious and economical way

2 zardo

beh

cMeiit[tt

i[nasamiba

2

2

iead2�fn(

swfw

2

Its(nppps(mwSwaCo

22 N.T. Tavengwa et al. / Journal of Ha

ecause of its low cost [11] and regeneration capabilities [12]. How-ver, most of the sorbents that have been used lack selectivityence, the development of molecularly imprinted polymers (MIPs).

MIPs represent a promising class of materials that have artifi-ially created receptor structures [13]. Since its discovery in 1972,IPs have attracted considerable interest from both scientists and

ngineers [14]. Instead of using molecular compounds as targets,onic species can be used and the resulting polymers are called ionmprinted polymers (IIPs) and have also drawn much attention inhe recent past as a highly selective polymeric sorbent material15]. IIPs can be used in pre-concentration and separation of thearget ion from other coexisting species or complex matrix due tohe specific interaction of ligand with the template ion [16].

In literature a number of papers have been published thatnvolve ion imprinted polymers specific for uranium uptake17–22]. However, only a few studies have been reported on ura-ium magnetic ion imprinted polymers [23–25], as such, there wasn attempt in this work to prepare magnetic ion-imprinted U(VI)elective polymers. These nano-composite materials could be sep-rated easily by using external magnetic fields [26]. The addition ofagnetic nano-particles to the pre-polymerization reagents used

n this work for the synthesis of uranium IIPs have not been reportedefore. Further, synthesis was fully characterized, optimized andpplied to environmental samples.

. Experiments

.1. Chemicals

Synthesis of the magnetic polymers used the follow-ng chemicals: FeCl2·4H2O, FeCl3·6H2O, NH4OH, methanol,thylene glycol dimethacrylate (EGDMA), methacryliccid (MAA), 1,1′-azobis(cyclohexanecarbonitrile), salicylal-oxime (SALO), 4-vinylpyridine (4-VP), NaHCO3 leachant,-methoxyethanol as well as the surfactant for coating magnetite,-methacryloxypropyltrimethoxysilane (�-MPS), all purchased

rom Sigma Aldrich (Steinheim, Germany). The imprint, uranylitrate (UO2(NO3)2·6H2O) was bought from BDH Chemical Ltd.,Poole, England).

Stock solutions of U(VI), Pb(II) and Ni(II) were prepared by dis-olving an appropriate amount of salts in 1 L of deionized waterith 0.05% (v/v) HNO3. Working solutions were prepared daily

rom the stock solution through serial dilutions. The stock solutionsere stored at 4 ◦C when not in use.

.2. Equipment

ICP-OES Spectro Genesis End-on-plasma (Spectro Analyticalnstruments (Pty) Ltd., Johannesburg, South Africa) was used forhe determination of the metals’ concentration in multi-elementalolutions. AAS measurements were made on a PG-990 AAS modelLeicestershire, UK) with pyrolytically coated HGA-76 graphite fur-ace tubes used as sample holders. All pH measurements wereerformed on a 766 Calimatic pH metre equipped with a Shott N61H electrode from Knick (Berlin, Germany). Deionized water wasrepared from Millipore (Massachusetts, USA). In batch adsorptiontudies, a Laser Photo/Contact Tachometer DT-1236L from LutronTaipei, Taiwan) was used to measure the rotational speeds of the

agnetic stirrers. A Fritsch pulverizer (Idar-Oberstein, Germany)as used to crush the synthesized bulk polymers. A FEI TECNAI

PIRIT (TEM-EDS) electron microscope (Eindhoven, Netherlands)

as used for size and morphological determination of magnetite

nd magnetic particles. A LECO-932 CHNS analyser from LECOorporation (Michigan, USA) was used to determine the amountf carbon and hydrogen in the organic moiety which coated the

us Materials 267 (2014) 221– 228

magnetite. Surface area of the magnetic polymers was deter-mined with the Micromeritics Flow Prep 060 instrument (Aachen,Germany). Fourier-transform infrared spectra were recorded in thefrequency range of 400–4000 cm−1 using a Bruker FTIR spectrom-eter, Model Tensor 27 (Ettlingen, Germany). Thermogavimetricanalysis for the prepared materials was performed with PerkinElmer Pyris 1 TGA Thermogavimetric Analyser (Massachusetts,USA).

2.3. Synthesis of magnetic U(VI) polymers

Synthesis of magnetite and its coating with �-MPS was doneusing methods outlined by Thorek et al. [27] and Kan et al.[28], respectively. The polymers were synthesized according tothe method of Singh and Mishra [17] with some modifications.The imprint ion (0.270 g) was complexed with 2 mM salicylal-doxime (SALO) and 2 mM 4-vinylpyridine (4-VP) in 10 mL of2-methoxyethanol in which 2 g of �-MPS functionalized mag-netite was dispersed. The above ternary complex solution wasthen mixed with 12 mmol of methacrylic acid (MAA), 36 mmolof ethylene glycol dimethacrylate (EGDMA) and 50 mg of 1,1′-azobis(cyclohexanecarbonitrile). The pre-polymerization mixturewas then cooled to 0 ◦C and purged with N2 for 10 min, sealedand thermally polymerized in an oil bath at 80 ◦C while stirringfor 3 h. The resulting magnetic polymers were then ground andsieved to obtain the ion-imprinted polymers in the size range of27 and 53 �m. The bulk magnetic polymers were dried at 70 ◦C toremove the solvent (porogen). Finally, the magnetic polymers werethen washed with ethanol, then ethanol and water mixture withincreasing amounts of water until only water was used. MagneticNIPs were prepared likewise except that the imprint ion was notincluded.

Three grams of the bulk magnetic polymer was transferred to a250 mL volumetric flask upon which a 100 mL of 1 mol L−1 NaHCO3was added. The mixture was stirred gently for 6 h, after whichfiltration was done under vacuum. The filtrate was retained formetal analysis. A freshly prepared NaHCO3 leachant, with the sameconcentration and volume as above, was added to the magneticpolymer. This procedure was repeated for four more times and thewhole synthetic scheme is shown in Fig. 1.

2.4. Batch sorption experiments

Adsorption of U(VI) by magnetic IIP and NIP was performed inbatch mode in all experiments. Optimization of various parameterswas done by varying one parameter while keeping others constant.Various amounts of the magnetic sorbents (10–100 mg) and sampleconcentration of 25 mL of 0.5–10 mg L−1 of U(VI) were varied. Aftera preset contact time (10–90 min) of stirring the uranium solutionat 1500 rpm, the solution was filtrated and the unadsorbed U(VI)determined by either GFAAS or ICP-OES.

2.5. Calculation of extraction efficiency and other constants

The extraction efficiency (%) was calculated according to Eq. (1).

Extraction efficiency = C0 − Ce

C0× 100% (1)

where C0 (mg L−1) is the initial concentration and Ce (mg L−1)represents the final concentration after adsorption. Another quan-tification term for the performance of the magnetic polymers is the

N.T. Tavengwa et al. / Journal of Hazardous Materials 267 (2014) 221– 228 223

for th

a(

q

wvTittc

K

Eis

K

wvUKlpre

Fig. 1. A schematic diagram

dsorption capacity, q (mg g−1) and is given mathematically by Eq.2).

= (C0 − Ce)VW

(2)

here C0 and Ce are as described in Eq. (1), V (mL) is the sampleolume and W (g) is the amount of the magnetic polymer used.he competitive sorption experiments were conducted by prepar-ng binary-mixture of U(VI)/Pb(II) and U(VI)/Ni(II). U(VI) and thewo competitors had an initial concentration of 2 mg L−1. The dis-ribution coefficients, Kd (mL g−1) of U(VI), Pb(II) and Ni(II) werealculated using Eq. (3):

d = (C0 − Ce)VCeW

(3)

q. (4) was used to calculate the selectivity coefficients for the bind-ng of a uranyl ion in the presence of other competing ions in binaryystems:

= Kd(UO22+)

Kd(B)(4)

here B represents U(VI) competing ions mentioned above. Thealue of K gives an indication as to how selective the polymer is for(VI) ions in the presence of other competing species in solution. If

= 1, it means competing ions and uranyl ions are bound equally. A

arge value of K is therefore preferred indicating that the magneticolymer favors uranyl ions. Another important parameter is theelative selectivity coefficient, K′, which represents the enhancedffect of imprinting on selectivity and adsorption affinity for the

e synthesis of magnetic IIP.

template onto the polymer. The K′ of the magnetic-IIP against themagnetic-NIP was calculated using Eq. (5), where KIIP and KNIP arethe selectivity coefficients of the IIP and NIP, respectively.

K ′ = KIIP

KNIP(5)

3. Results and discussion

3.1. Characterization of magnetic U(VI) polymers

Uncoated magnetite particles normally have a tendency ofagglomerating. The effect of aggregation of magnetic particles isgreater when they are synthesized in aqueous medium as was thecase in this study. One of the ways to stabilize the magnetite isfunctionalizing it with coating agents and surfactants. In Fig. 2, thecoating agent was �-MPS whose hydrophilic silanol groups reactedwith the hydroxyl groups around the magnetite. The hydropho-bic end of �-MPS contained a double bond which made it possiblefor polymerization to occur. Agglomeration of the coated mag-netite, after coating, was still evident though to a lesser degree.It was also observed that the synthesized magnetite particles weremostly spherical in shape and even silica coated magnetite parti-

cles retained this spherical shape. From the micrograph of Fe3O4obtained by El Ghandoor et al. [29], it was seen that sphericallyshaped particles with a narrow size distribution with mean diame-ter of 10.6 nm were obtained. In another study by Qian et al. [30], it

224 N.T. Tavengwa et al. / Journal of Hazardous Materials 267 (2014) 221– 228

Fig. 2. Transmission electron micrograph of �-MPS coated magnetite.

F

wa

dnm�t

atR

TC

0

0.05

0.1

0.15

0.2

0.25

0.3

0

50

100

150

200

5 10 15 20 25

Iron

conc

entra

tion

(mg

L-1 )

Ura

nium

con

cent

ratio

n (m

g L-

1 ) Uranyl ionMagnetite

ig. 3. FTIR spectra of (a) uncoated magnetite and (b) �-MPS coated magnetite.

as revealed that the magnetic particles were well dispersed, withn average diameter of 10 nm.

Carbon, hydrogen, nitrogen and sulphur (CHNS) analysis wasone in order to determine the degree of coating of �-MPS on mag-etite and the results are shown in Table 1. This was achieved byeasuring the carbon and hydrogen content in the organic moiety,

-MPS. From values in Table 1, the average ligand concentration onhe magnetite was found to be 16.1 mmol g−1 for �-MPS.

Fourier-transform infrared (FTIR) results for both the uncoated

nd �-MPS coated magnetite are shown in Fig. 3. Fig. 3(a) depictshe FTIR spectrum of the synthesized magnetite. According tooonasi [31], the broad bands around 570 and 360 cm−1 correspond

able 1HNS elemental analysis of coated �-MPS magnetite.

mass (g) % C % H Ligand concentration (mmol g−1)*

2.00 20.16 2.94 16.82.00 19.25 2.92 15.82.00 18.92 2.84 15.8

* Values for the ligand concentration were all based on the carbon content.

Leaching t ime (hours)

Fig. 4. Bulk magnetic IIP leached with 1 mol L−1 NaHCO3.

to stretching and bending vibrations of Fe O bonds, typical for thecrystalline lattice of magnetite. The band at 483 cm−1 from Fig. 3(a)can be assigned to the bending vibrations of the Fe O bonds. Itis associated with the torsional vibration modes of the magnetiteFe O bonds in octahedral sites. The lower band of 360 cm−1 wasnot detected because all the spectra obtained in this work werefrom 400 to 4000 cm−1. For the �-MPS coated magnetite spectrumin Fig. 3(b), the shift at 1104 cm−1 can be attributed to the stretchof Si O Si, indicating the formation of silica film on the magnetite.Functionalization can also be observed by the presence of the C Oband at 1716 cm−1 from the carboxylate of �-MPS. Ma et al. [32]also used FTIR spectroscopy for the determination of the chemi-cal structure of Fe3O4–�-MPS. They identified that both Fe3O4 andFe3O4–�-MPS had a characteristic peak of Fe O at a stretchingfrequency of 586 cm−1. Besides, they observed a strong stretchingvibration of Si O Si at 1089 cm−1 for Fe3O4–�-MPS. Their conclu-sion from the two observed bands (which was the same drawn inthis work) was that �-MPS was successfully grafted onto the surfaceof magnetite.

Fig. 4 shows the leaching of the uranyl from the magnetic poly-mer matrix. The results depicted in this figure show the expecteddecrease of the amount of leached uranyl from magnetic polymersover time. However, there was an unwanted parallel leaching of theembedded magnetic core from the magnetic polymers. 1 mol L−1

NaHCO3 showed that it was a weak leachant towards the washingaway of the magnetic core and the magnetism of the polymers wasnot reduced greatly. However, it is also observed that the leach-ing of the uranyl ion was again weak with this leachant. This isundesirable as the uranyl might bleed out when the polymers arebeing applied to real environmental water samples. More wash-ing was thus performed until no uranium was detected. Sadeghiand Aboobakri [18] synthesized aminopropyl triethoxysilane onfunctionalized silica coated Fe3O4 nanoparticles. This was done inthe presence of a uranyl ion as a template. Finally, the uranyl waswashed with 0.1 mol L−1 EDTA solution for its complete removal toform a magnetic imprinted polymer selective of uranyl. Singh andMishra [17] used exactly the same ternary monomers as those usedin this work. After cross-linking, they went further to even opti-

mize the leachants where they used various leachants in differentconcentration and found also HCl to be a strong leachant.

Brunauer, Emmett and Teller (BET) surface area measurementsare summarized in Table 2. The results demonstrated that there was

Table 2BET surface areas and constants for the bulk magnetic IIPs.

BET surfacearea (m2 g−1)

vm (cm3 g−1) c

Unleached magnetic IIP 8.8 2.09 94.86NaHCO3 leached magnetic IIP 65.2 15.43 129.6

zardous Materials 267 (2014) 221– 228 225

aoslfdistvwr

pib1hbaAtipcstTvaThaasTsbttw

3

nmss

Fig. 6. Removal efficiency obtained by varying the mass of �-MPS IIP bulk-NaHCO3

Other studies for the selective uranyl uptake have demonstratedlonger extraction times. For instance, for the prepared magnetic

N.T. Tavengwa et al. / Journal of Ha

n increase in the surface area of the magnetic polymer as a resultf leaching action. The bulk unleached magnetic IIP had a measuredurface area of 8.8 m2 g−1 whilst that of the corresponding NaHCO3eached increased to 65.2 m2 g−1. These values are single point sur-ace area measurements at p/p0 = 0.2998. Leaching with NaHCO3id not remove the uranyl as well as the magnetite completely. The

mplication of this is that the BET surface area was kept low becauseome of these two ions were still much present and embedded inhe polymer matrix of the magnetic IIPs. The monomolecular layerolume, vm and BET constant, c, for the bulk magnetic polymersere calculated from the slopes and intercepts of BET plots and the

esults are summarized in Table 2.As a comparison, imprinted polymer nanospheres for uranium

repared by complexing uranyl ion onto quinoline-8-ol functional-zed 3-aminopropyltrimethoxysilane modified silica nanoparticlesy Milja et al. [22], had BET surface areas of 188.87 and34.69 m2 g−1 for the IIP and NIP, respectively. This was evidentlyigher that the surface areas obtained in this study and was proba-ly due to the lower strength of the monomer template interactionnd/or the strength and type of the leachants used. However,hmadi et al. [19] found a BET surface area of 18.5 m2 g−1 for

he unleached polymer particles they prepared by copolymeriz-ng uranyl ions with styrene and divinylbenzene. As expected, theorosity changed after the removal of uranyl ions from the IIP parti-le to 33.2 m2 g−1 which was far less than what was obtained in ourtudies. Thermo-gravimetric (TGA) analysis was employed to quan-ify the amount of Fe3O4 encapsulated in the polymer matrix [33].he analysis was carried out in an oxygen free environment to pre-ent the oxidation of the Fe3O4. Thermo-gravimetric analysis waslso used for further structural elucidation of �-MPS–Fe3O4. TheGA spectra in Fig. 5 showed that the magnetite was stable undereating throughout the temperature range of 0–1000 ◦C. A temper-ture at 305 ◦C was enough for the decomposition of the coatinggent �-MPS. From Fig. 5, it was also observed that after the analy-is, at 1000 ◦C, the residual weight of the �-MPS–Fe3O4 was 69.17%.his meant that the other 30.83% was from the decomposed organicurface coating agent, �-MPS. It was also noted that leaching of theulk magnetic polymer reduced the residual weight significantlyo about 10%. This can be explained by the fact that the leaching ofhe uranyl resulted in the washing of away of the magnetic core asell as uranium.

.2. Effect of pH on uranium uptake

In our recent work soon to be published, a number of mag-etic ion IIPs for the selective extraction of uranium showed

aximum uptake at pHs around 4. Wang et al. [20] applied

alicylideneimine-functionalized hydrothermal-carbon-based as aolid-phase extractant for separating uranium and obtained an

Fig. 5. Thermo-gravimetric analysis of bulk magnetic polymers.

leached and �-MPS NIP bulk-NaHCO3 leached. Experimental conditions: sample pH,4; sample volume, 25 mL; uranium concentration, 2 mg L−1; contact time, 45 min;stirring speed, 1500 rpm; temperature, room temperature.

optimum pH of 4.3. A pH value of 4 was then used in all subsequentexperiments.

3.3. Effect of the amount of magnetic polymer

The dependency of the amount of uranium adsorbed on thequantity of the adsorbent used is shown in Fig. 6. It was found thatincreasing dosage of the magnetic adsorbent, the amount of ura-nium removed increased steadily to an almost constant value. Themaximum uranium uptake was found when the mass of the mag-netic polymer was 50 mg, which was then taken as the optimumdose.

3.4. Effect of contact time

The relationship between the amount of uranium adsorbed andcontact time for the magnetic polymer is shown in Fig. 7. Underthe used experimental conditions, it was observed that there wasan increase in the amount of uranium adsorbed by the magneticpolymers from 0 to 45 min, after which equilibrium was reached.A contact time of 45 min was therefore selected as the optimumtime as it was sufficient for the attainment of equilibrium. A furtherincrease in time had no effect on the adsorption of uranium ontothe magnetic polymers. As expected, the adsorption performanceof the magnetic IIP was higher than of the corresponding magneticNIP.

Schiff of N,N′-bis(3-methoxylsalicylidene)-1,2-phenylenediaminesynthesized by Zhang et al. [34], it was observed that the adsorption

0

0.2

0.4

0.6

0.8

1

0 20 40 60 80 10 0

Am

ount

ads

orbe

d (m

g g-

1 )

Contact time (min)

Magnetic IIP

Magnetic NIP

Fig. 7. Effect of contact time on the uptake of uranium by �-MPS IIP bulk-NaHCO3

leached and �-MPS NIP bulk-NaHCO3 leached. Experimental conditions: sample pH,4; sample volume, 25 mL; uranium concentration, 2 mg L−1; polymer weight, 50 mg;stirring speed, 1500 rpm; temperature, room temperature.

226 N.T. Tavengwa et al. / Journal of Hazardous Materials 267 (2014) 221– 228

Table 3Kd, K and K′ values for the magnetic IIP and NIP of �-MPS bulk-NaHCO3 in binary mixtures.

Adsorbent Kd (mL g−1) K K′ Kd (mL g−1) K K′

U(VI) Ni(II) U(VI) Pb(II)

1–

rpeoapa

3

unuTsmputawtaawtid

3

absfaIT

FNss

Magnetic IIP 3231 407 7.9

Magnetic NIP 1423 212 6.7

ate increased rapidly during the initial stages of the adsorptionrocess. After that, uptake rate slowly decreased with time andquilibrium was reached after 6 h. This was extremely slower thanur prepared magnetic IIP. On the other hand, some polymericdsorbents which were faster in the uptake of uranium(VI) wererepared by Wang et al. [23], Sadeghi and Aboobakri [18] and Fannd Wang [35] with the first two at 30 min and the last 16 min.

.5. Effect of initial uranium concentration

Fig. 8 presents the dependency of amount adsorbed from variousranium ion concentrations onto magnetic polymers for fixed mag-etic polymer dose, contact time and sample pH. As the amount ofranium ions concentration increased so was the adsorbed amount.his can be attributed to the increasing number of U(VI) ions inolution that competed for a finite number of binding sites on theagnetic polymer surface. This equilibrium loading of uranyl ions

er unit mass of the magnetic polymer increased almost linearlypon increasing the initial concentration of U(VI) ions in solution upo adsorption capacities of 0.95 and 1.1 mg g−1 for the magnetic NIPnd IIP, respectively. After reaching maximum adsorption, thereas not any further increase in the amount of U(VI) adsorbed. In

his work a combination of functional monomers of 4-VP, SALOnd MAA were used. However, simultaneous employment of 4-VPnd MAA has been recently investigated by Dakova et al. [36] andas shown to lead to poor performance of the synthesized IIPs in

erms of their analyte uptake abilities. In this case, the differencen the performance of the IIPs and NIPs becomes small as Fig. 8 canemonstrate.

.6. Selectivity studies

Competitive adsorption of the binary mixtures of U(VI)/Pb(II)nd U(VI)/Ni(II) were investigated in an equilibration–adsorptionatch systems. In this approach, the concentration of the unad-orbed ions was determined by ICP-OES. The recovery of U(VI) was

ound to be much higher than that of the competing ions (Ni2+

nd Pb2+) demonstrating that the cavities in the magnetic uranylIP had higher affinity for U(VI) because of the imprinting effect.able 3 summarizes the distribution coefficient (Kd), the selectivity

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 2 4 6 8 10

Am

ount

ads

orbe

d (m

g g-

1 )

Inintial concentration (mg L-1)

Magnetic IIPMagnetic NIP

ig. 8. Effect of initial concentration on the uptake of uranium by (a) �-MPS IIP bulk-aHCO3 leached and (b) �-MPS NIP bulk-NaHCO3 leached. Experimental conditions:

ample pH, 4; sample volume, 25 mL; polymer weight, 50 mg; contact time, 45 min;tirring speed, 1500 rpm; temperature, room temperature.

.18 2833 274 10.3 0.76 1352 100 13.5 –

coefficient (K) and the relative selectivity coefficient (K′) values ofthe competing ions with respect to the target, UO2

2+ ion.Because Ni(II) has a smaller ionic radius of 69 pm, it had an

ability to enter into the imprinted sites of the magnetic polymersfreely, hence it showed high Kd and K′ values than Pb(II) with ionicradius of 132 pm. It occupied sizeable fractions of the imprintingsites because it perfectly fitted the fabricated recognition sites toa greater extent. As shown in Table 3, the Kd values for all mag-netic IIPs are higher for UO2

2+, as compared to those of competingions. The relative order of magnitude of metal sorption followedthe order: UO2

2+ > Ni2+ > Pb2+.

3.7. Reusability of the magnetic uranyl imprinted polymers

The repeated use of the magnetic U(VI)-imprinted polymers is akey factor that determines the applicability of these adsorbents indecontamination of wastewaters. The re-usability of both the mag-netic MIP and NIP was tested by running five cycles as described inthe experiments (Fig. 9.) The extraction efficiency for the magneticMIP at first cycle was 98% while for the magnetic NIP was 96%. Forthe last cycle, the extraction efficiency for the magnetic MIP was95% and 92% for the magnetic NIP. Since the decrease after fivecycles was small, the prepared magnetic polymers can be reused.Similar polymers, though not magnetic, showed good reusabilityand stability towards uranyl ion up to the tenth cycles of repeatedexperiments [17]. For Pakade et al. [21], the stability of the poly-mers were observed up to the eighth cycle of reuse, maintaining a99% recovery of the uranyl from aqueous solution.

3.8. Application of magnetic U(VI) IIPs to real water samples

Germiston water had a pH of 2.6, oxidation reduction potentialof 436 mV and conductivity of 680 �S cm−1. The pH of the watersample was 2.6 which is very acidic and this was expected becauseof the effect of AMD. In the vicinity of the water source, there aredisused mines which were previously engaged in open cast and

deep mining. These activities led to the exposure of the pyrite rock,FeS2, to air and water which further led to the formation of H2SO4which decanted into the water bodies where the samples were col-lected. A reduction oxidation potential of 436 mV showed oxidative

60

70

80

90

100

6543210

Cycle number

Magnetic IIP

Magnetic NIP

Extra

ctio

n ef

ficie

ncy

(%)

Fig. 9. Reusability and stability of magnetic IIP and NIP of �-MPS bulk-NaHCO3

leached. Adsorption conditions: amount of materials, 50 mg; solution pH 4; solutionvolume, 25 mL; contact time, 45 min, U(VI) concentration, 2 mg L−1, desorption con-ditions: solution volume, 25 mL; contact time, 45 min, HCl concentration leachant,1 mol L−1.

N.T. Tavengwa et al. / Journal of Hazardous Materials 267 (2014) 221– 228 227

Table 4Metal composition and concentrations in real water samples.

Metal Au Co Cr Cu Fe Hg Mn Ni U Zn

Mine wastewater Mean (mg L−1) 0.273 22.95 0.836 11.25 1.8 0.675 72.05 48.3 8.5 54.75SD 0.005 0.003 0.027 0.002 0.001 0.032 0.005 0.004 0.015 0.004

Wastewater from treatment works Mean (mg L−1) 0.007 0.044 – 0.029 – 0.03 0.165 0.108 – 0.023SD 0.001 0.001 – 0.001 – 0.005 0.001 0.001 – 0.002

Table 5Application of magnetic �-MPS IIP bulk-NaHCO3 leached and the corresponding NIP on unspiked and spiked real water samples. Sample pH = 4; solution volume = 25 mL;sorbent mass = 50 mg for the magnetic IIP and NIP.

Sample Uranium concentration (mg L-1) % Recovery

Spiked Determined Magnetic Magnetic

Magnetic IIP Magnetic NIP IIP NIP

AMD – 8.50 (0.02) 8.50 (0.02) – –5.00 13.23 (0.01) 11.34 (0.03) 78 6710.0 17.58 (0.05) 15.17 (0.04) 76 66

WWTP – <DL <DL – –1.00 0.97 (0.08) 0.81 (0.02) 78 655.00 4.75 (0.05) 4.20 (0.01) 76 67

<

csosfltcrsf

4

saArtwims

A

(p

R

[

[

[

[

[

[

[

[

[

[

[

[and efficient ion imprinted polymer for recovery of uranium from environmen-

10.0 9.80 (0.03)

DL: Below detection limit and SD values in parenthesis.

onditions. This meant that most metals in the AMD water wereubjected to an oxidative environment. A high conductivity valuef 680 �S cm−1 indicated that most metals which constituted theample, some of which are shown in Table 4, were in their ionicorm. This resulted from the action of the highly acidic water whicheached out metallic ions as it flowed down to water bodies. Whenhe method was applied to real water samples, extraction efficien-ies of 77% and 66% were obtained for the magnetic IIP and NIP,espectively, as shown in Table 5. This was a demonstration of theuitability of the magnetic sorbent to selectively remove uraniumrom complex aqueous matrix.

. Conclusions

Magnetic polymers for the selective extraction of uranium wereuccessfully synthesized. From the selectivity studies, U(VI) wasdsorbed more as compared to its competitors, Ni(II) and Pb(II).verage recoveries of 77% and 66% for the magnetic IIP and NIP,espectively, from environmental wastewaters were recorded. Dueo the imprinting effect, adsorption performance of magnetic IIPas always higher than its corresponding magnetic NIP. Stabil-

ty and reusability of the magnetic polymers up to the sixth cycleeant that they could be applied repeatedly without losing sub-

tantial extraction ability.

cknowledgements

The authors would like to thank Water Research CommissionWRC) of South Africa and the University of Witwatersrand forroviding financial support.

eferences

[1] A.K. Sarangi, K.K. Beri, Uranium mining by in-situ leaching, in: Presentedand published in the Proceedings of the International conference on Tech-nology management for Mining, Processing and Environment, IIT, Kharagpur,2000.

[2] WHO, Depleted Uranium, Sources, Exposure and Health Effects, WHO, Geneva,2001.

[3] Radiation Effects – Effects by Radionuclides, 2010, Available from:http://hps.org/publicinformation/ate/q754.html (accessed 02.12.10).

[4] D.R. Lovley, E.J.P. Phillips, Y.A. Gorby, E.R. Landa, Microbial reduction of ura-nium, Nature 350 (1991) 413–416.

[

8.30 (0.07) 78 66

[5] U.S. DOE, From Cleanup to Stewardship, U.S. D.O.E. Office Environ. Manage-ment, 1999.

[6] J.K. Fredrickson, J.M. Zachara, D.W. Kennedy, M.C. Duff, Y.A. Gorby, S.M.W. Li,K.M. Krupka, Reduction of U(VI) in goethite (�-FeOOH) suspensions by a dis-similatory metal-reducing bacterium, Geochim. Cosmochim. Acta 64 (2000)3085–3098.

[7] A. Kot, J. Namiesnèik, The role of speciation in analytical chemistry, TrAC 19(2000) 69–79.

[8] A. Mellah, S. Chegrouche, M. Barkat, The precipitation of ammonium uranylcarbonate (AUC): thermodynamic and kinetic investigations, Hydrometallurgy85 (2007) 163–171.

[9] B. Gu, Y. Ku, G.M. Brown, Sorption and desorption of perchlorate andU(VI) by strong-base anion-exchange resins, Environ. Sci. Technol. 39 (2005)901–907.

10] D. de Agreda, I. Garcia-Diaz, F.A. López, F.J. Alguacil, Supported liquid mem-branes technologies in metals removal from liquid effluents, Rev. Metal 47(March–April) (2011) 146–168.

11] P. Hadi, J.P. Gao, Barford, G. Mckay, Novel application of the non-metallic frac-tion of the recycled printed circuit boards as a toxic heavy metal adsorbent, J.Hazard. Mater. 252–253 (2013) 166–170.

12] I.S. Ahamed, A.K. Ghonaim, A.A. Abdel Hakim, M.M. Moustafa, A.H. Kamal El-Din, Synthesis and characterization of some polymers for removing of someheavy metal ions of industrial wastewater, J. Appl. Sci. Res. 4 (12) (2008)1946–1958.

13] S.A. Piletsky, E.V. Piletskaya, I.A. Nicholls, M. Ulbricht, Receptor and transportproperties of imprinted polymer membranes – a review, J. Membr. Sci. 157 (2)(1999) 263–278.

14] P.C.L. Lai, A. Namvar, D.H. Hawkins, S.M. Reddy, Molecular imprinted poly-mers for biorecognition of biogents, in: K. Warriner, M. Zourob, S. Elwary, A.P.F.Turner (Eds.), Principles of Bacterial Detection: Biosensors, Recognition Recep-tors, and Microsystems, Springer Science + Business Media, LLC, 233 SpringStreet, New York, NY 10013, USA, 2008.

15] Y. Mingdi, O. Ramstrom (Eds.), Molecular Imprinted Materials (Science andTechnology), Marcel Dekker, New York, USA, 2004.

16] T.P. Rao, R. Kala, S. Daniel, Metal ion-imprinted polymers-novel materials forselective recognition of inorganics, Anal. Chim. Acta 578 (2006) 105–116.

17] D.K. Singh, S. Mishra, Synthesis and characterization of UO22+-ion imprinted

polymer for selective extraction of UO22+, Anal. Chim. Acta 644 (2009)

42–47.18] S. Sadeghi, E. Aboobakri, Magnetic nanoparticles with an imprinted polymer

coating for the selective extraction of uranyl ions, Microchim. Acta 178 (2012)89–97.

19] S.J. Ahmadi, O. Noori-Kalkhoran, S. Shirvani-Arani, Synthesis and characteri-zation of new ion-imprinted polymer for separation and preconcentration ofuranyl (UO2

2+) ions, J. Hazard. Mater. 175 (2010) 193–197.20] H. Wang, L. Ma, K. Cao, J. Geng, J. Liu, Q. Song, X. Yang, S. Li, Selective solid-phase

extraction of uranium by salicylideneimine-functionalized hydrothermal car-bon, J. Hazard. Mater. 229–230 (2012) 321–330.

21] V.E. Pakade, E.M. Cukrowska, J. Darkwa, G. Darko, N. Torto, L. Chimuka, Simple

tal samples, Water Sci. Technol. 65 (4) (2012) 728–736.22] T.E. Milja, K.P. Prathish, R.T. Prasada, Synthesis of surface imprinted

nanospheres for selective removal of uranium from simulants of Sambhar saltlake and ground water, J. Hazard. Mater. 188 (2011) 384–390.

2 zardo

[

[

[

[

[

[

[

[

[

[

[

[

[of metal coordination imprinted polymers, J. Appl. Polym. Sci. 116 (2010)

28 N.T. Tavengwa et al. / Journal of Ha

23] J. Wang, R. Peng, J. Yang, Q. He, Y. Liu, Selective adsorption of uranium(VI)on U(VI) ion-imprinted chitosan composite magnetic microspheres, in: Com-puter Distributed Control and Intelligent Environmental Monitoring (CDCIEM),International Conference, 2011, pp. 1714–1717.

24] C. Shang, L. Zhou, Adsorption properties for uranium(VI) ions using ion-imprinted magnetic chitosan microspheres, Polym. Mater. Sci. Eng. 29 (2013)56–59.

25] L. Zhou, C. Shang, Z. Liu, G. Huang, A.A. Adesina, Selective adsorption of ura-nium(VI) from aqueous solutions using the ion-imprinted magnetic chitosanresins, J. Colloid Interface Sci. 366 (2012) 165–172.

26] O. Philippova, A. Barabanova, V. Molchanov, A. Khokhlov, Magnetic polymerbeads: recent trends and developments in synthetic design and applications,Eur. Polym. J. 47 (2011) 542–559.

27] D.L.J. Thorek, A.K. Chen, J. Czupryna, A. Tsourkas, Superparamagnetic iron oxidenanoparticle probes for molecular imaging, Ann. Biomed. Eng. 34 (2006) 23–38.

28] X.W. Kan, Q. Zhao, D.L. Shao, Z.R. Geng, Z.L. Wang, J. Zhu, Preparation andrecognition properties of bovine hemoglobin magnetic molecularly imprintedpolymers, J. Phys. Chem. B 114 (2010) 3999–4004.

29] H. El Ghandoor, H.M. Zidan, M.M.H. Khalil, M.I.M. Ismail, Synthesis and somephysical properties of magnetite (Fe3O4) nanoparticles, Int. J. Electrochem. Sci.7 (2012) 5734–5745.

30] Z. Qian, Z. Zhang, Y. Chen, A novel preparation of surface-modifiedparamagnetic magnetite/polystyrene nanocomposite microspheres by

[

us Materials 267 (2014) 221– 228

radiation-induced miniemulsion polymerization, J. Colloid Interface Sci. 327(2008) 354–361.

31] P. Roonasi, Adsorption and Surface Reaction Properties of SynthesizedMagnetite Nano-particles, Division of Chemistry Department of Chemical Engi-neering and Geosciences, Luleå University of Technology, S-971 87 Luleå,Sweden, 2007.

32] W. Ma, S. Xu, J. Li, Y. Lin, C. Wang, Hydrophilic dual-responsive magnetite/PMAAcore/shell microspheres with high magnetic susceptibility and pH sensitivityvia distillation–precipitation polymerization, J. Polym. Sci. A: Polym. Chem. 49(2011) 2725–2733.

33] Y. Zhan, X. Luo, S. Nie, Y. Huang, X. Tu, S. Luo, Selective separation of Cu(II) fromaqueous solution with a novel Cu(II) surface magnetic ion-imprinted polymers,Ind. Eng. Chem. Res. 50 (2011) 6355–6361.

34] X. Zhang, C. Jiao, J. Wang, Q. Liu, R. Li, P. Yang, M. Zhang, Removal of uranium(VI)from aqueous solutions by magnetic Schiff base: kinetic and thermodynamicinvestigation, Chem. Eng. J. 198–199 (2012) 412–419.

35] P. Fan, B. Wang, Regulatory effects of Zn(II) on the recognition properties

258–266.36] I. Dakova, I. Karadjova, V. Georgieva, G. Georgiev, Synthesis and application of

vinylpyridine containing ion-imprinted copolymer gel microbeads for Cu(II)solid-phase extraction, J. Sep. Sci. 35 (2012) 2805–2812.