Strontium ion (Sr2+) separation from seawater by hydrothermally structured titanate nanotubes: Removal vs. recovery

Chemical Engineering Journal 304 (2016) 503–510

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Strontium ion (Sr2+) separation from seawater by hydrothermally structured titanate nanotubes: Removal vs. recovery Jungho Ryu a,1, Soonhyun Kim b,1, Hye-Jin Hong a, Jeongsik Hong a, Minsun Kim b, Taegong Ryu a, In-Su Park a, Kang-Sup Chung a, Jum Suk Jang c,⇑, Byoung-Gyu Kim a,⇑ a b c

Mineral Resources Research Division, Korea Institute of Geoscience and Mineral Resources, Daejeon 34132, Republic of Korea Nano & Bio Research Division, Daegu Gyeongbuk Institute of Science and Technology, Daegu 42988, Republic of Korea Division of Biotechnology, College of Environmental and Bioresource Sciences, Chonbuk National University, Iksan 54596, Republic of Korea

h i g h l i g h t s  Titanate nanotubes (TiNTs) were synthesized by a simple hydrothermal reaction.  The sorption of strontium (Sr) on TiNTs rapidly occurred, achieving Sr uptake 97 mg/g.  Na had little effect on Sr sorption despite the sorption mechanism of the Na exchange.  Ca significantly hindered Sr sorption on TiNTs among co-existing cations in seawater.  TiNTs could be easily regenerated by acid treatment and reused for repeated cycles.

a r t i c l e

i n f o

Article history: Received 8 April 2016 Received in revised form 24 June 2016 Accepted 28 June 2016 Available online 28 June 2016 Keywords: Strontium Seawater Titanate nanotubes Recovery Separation

a b s t r a c t Strontium ion (Sr2+) separation from seawater has attracted attention for radioactive pollutants removal and for Sr2+ recovery. Herein, we synthesized titanate nanotubes (TiNTs) via a simple hydrothermal reaction, characterized their physicochemical properties, and systematically evaluated Sr2+ sorption behavior under various reaction conditions corresponding to seawater environments. The synthesized TiNTs exhibited a fibril-type nanotube structure with a high specific surface area (260 m2/g). Sr2+ adsorption on TiNTs rapidly occurred following a pseudo-second-order kinetic model and was in good agreement with the Langmuir isotherm model, indicating a maximum adsorption capacity of 97 mg/g. Based on the Sr2+ uptake and Na+ release with a stoichiometric balance, the Sr2+ sorption mechanism on TiNTs was ion exchange between Na+ in the TiNT lattice and Sr2+ in the solution phase, as confirmed by XRD and Raman analysis. Among the competitive ions, Ca2+ significantly hindered Sr2+ sorption on TiNTs, whereas Na+ only slightly affected Sr2+ sorption, despite the Na+ exchange sorption mechanism. The effect of Ca2+ on Sr2+ sorption was evaluated by introducing a distribution coefficient (Kd) as a critical factor in determining the selectivity, which revealed a slightly higher selectivity for Sr2+. The Sr2+ adsorptiondesorption test in a real seawater medium enabled the determination of Kd and the concentration factor (CF) for co-existing matrix ions in seawater; these values were evaluated for Sr2+ removal and recovery from seawater. TiNTs were regenerated by acid treatment and reused through consecutive adsorptiondesorption experiments. While most studies addressing Sr2+ sorption using TiNTs aimed for extraction from wastewater and radioactive wastewater, this study elucidated Sr2+ sorption behavior under seawater conditions and provided insights into developing the removal and recovery processes from seawater. Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction Recently, the removal of strontium ions (Sr2+) from seawater has received substantial attention from an environmental perspec⇑ Corresponding authors. E-mail addresses: [email protected] (J.S. Jang), [email protected] (B.-G. Kim). 1 Equal contribution. http://dx.doi.org/10.1016/j.cej.2016.06.131 1385-8947/Ó 2016 Elsevier B.V. All rights reserved.

tive after the Fukushima plant accident, which released a large amount of radioactive Sr (90Sr) and Cs (137Cs) [1,2]. The 90Sr isotope is a beta-emitter that genetically affects seawater organisms and ultimately causes harm to humans [3]. Accordingly, the efficient separation of radioactive elements released to seawater has become a critical technological requirement, along with their removal from radioactive wastes.

504

J. Ryu et al. / Chemical Engineering Journal 304 (2016) 503–510

Seawater contains various types of mineral resources with different concentrations depending on the elements. In the case of strategically valuable resources, such as lithium and uranium, seawater is the most attractive resource due to its large reserve, despite the very low concentrations. If the practical problem of process cost could be addressed, then extraction of these metals from seawater would be promising and beneficial compared with typical land mining [4]. Strontium, which has many industrial applications, such as ferrite magnets, ceramics, and fire-works, exists in seawater with a concentration of approximately 7 mg/L [5]. In a previous report estimating the economic potential of the recovery of various elements from seawater in terms of their commercial values and concentrations in seawater [6], Sr2+ was located above the approximate break-even line, implying that Sr2+ recovery from seawater can potentially be profitable. Nevertheless, research on the recovery of Sr2+ from seawater has received limited attention. The separation of Sr2+ from aqueous media by various methods has previously been introduced, including solvent extraction [7], adsorption by solid materials [8,9], and ion exchange [10]. Among these methods, the adsorption technique using solid adsorbents is of great interest in selectively separating Sr2+ from seawater, allowing for the concentration level of Sr2+. Many studies have been conducted to investigate Sr2+ adsorption on metal oxides, such as hydrous manganese dioxide and sodium nonatitanate, and clays, such as zeolite and titanosilicate [11–13]. However, only a few studies have reported the extraction of Sr2+ from seawater using a liquid membrane [14], commercial resin-type sorbents [15], and an alginate-based biosorbent [16,17]. Titanate-based materials have been widely investigated, primarily for the removal of heavy metals and radioactive pollutants from aqueous media due to their chemical stability and facile synthesis [18–20]. Most studies addressing Sr2+ adsorption using titanates aimed for extraction from aqueous media, such as wastewater and radioactive wastewater [12,18,21]. Similar to the cases of other adsorbents, the understanding of Sr2+ sorption behavior on titanates in a seawater medium remains insufficient. The systematic investigation of Sr2+ sorption behavior under seawater conditions is necessary to provide insights into developing new sorbent materials for the removal and recovery processes from seawater. Herein, we report Sr2+ separation from seawater using titanate nanotubes (TiNTs), focusing on an evaluation of the effect of seawater matrix ions on Sr2+ sorption behavior. This is the first study reporting Sr2+ sorption behavior on TiNTs in detail for the purpose of removing and simultaneously recovering Sr2+ from seawater. Hydrothermally structured TiNTs were prepared and systematically characterized using XRD, SEM, TEM, and BET. The overall performance of TiNTs was comprehensively evaluated in terms of the Sr2+ adsorption isotherm, adsorption kinetics, and competitive adsorption of co-existing cations in seawater. The Sr2+ sorption test in a real seawater medium provided the strategy to increase the efficiency depending on the purpose of removal and recovery. Finally, the regeneration and reusability of TiNTs were examined for practical applications and for the simultaneous recovery of Sr2+ from seawater.

ples were denoted as TiNT-S or TiNT-L, representing 200 mL or 600 mL, respectively. Then, the mixtures were transferred to Teflon-lined autoclaves and hydrothermally reacted at 120 °C for 24 h. While no stirring was applied to the reaction using the 200-mL autoclave, the suspension in the 600-mL autoclave reactor was continuously stirred during the reaction. The obtained products were washed with distilled water until the washing solution reached a pH of 7, and the precipitates were subsequently freeze-dried. 2.2. Characterization The morphology was examined using an analytical scanning electron microscope (SEM, Hitachi S-4800) with energy dispersive X-ray spectrometry (EDS) to provide the distribution of the elements, and a high-resolution transmission electron microscope (HR-TEM, Hitachi HD-2399) was also used. The phase identification of the samples was carried out by powder X-ray diffraction (XRD) using 60-kV Cu-Ka1 radiation (PANalytical, Empyrean). The surface area and pore volume analyses were performed using the BET and BJH methods with ASAP 2020, Micromeritics. Zeta potentials were measured by an electrophoretic light scattering spectrophotometer (Zetasizer, Malvern). Raman spectra of the samples before and after Sr2+ adsorption were obtained with a Raman spectrometer (UniRAM, UniNanoTech, Korea) using a 532-nm laser. The laser power was approximately 2 mW, and the signal was accumulated for 30 s. 2.3. Adsorption experiments TiNTs were dispersed in distilled water at 1 g/L (0.02 g/0.02 L) and sonicated for 30 s. A desired amount of Sr2+ and different metal cation stock solution was added to the suspension, and the solution pH was adjusted with an HCl or NaOH standard solution, which was subsequently stirred for 30 min to allow the equilibrium adsorption. The sample aliquots were intermittently withdrawn during the 30-min adsorption reaction and filtered through a 0.2-lm cellulose acetate membrane (ADVAN-TEC). The concentrations of Sr2+ and other metal cations were measured by inductively coupled plasma-atomic emission spectrometry (ICP-AES, Perkin and Elmer). The performances of TiNT samples were evaluated in terms of the uptake of Sr2+ (see SI). To investigate the effects of coexisting ions, the solutions were varied with different electrolyte ions (Na+, K+, Mg2+, Ca2+) for which the concentrations in seawater are much higher than that of Sr2+. The initial concentration of Sr2+ was fixed to 10 mg/L, and those of coexisting ions ranged from 10 mg/L to their concentrations in seawater. Competitive adsorption studies between Sr2+ and Ca2+ were performed with different molar ratios of Sr2+ to Ca2+ and equivalently increasing concentrations of Sr2+ and Ca2+. For the regeneration of the samples, TiNTs were immersed in a 0.1 M HCl solution for 30 min and washed with DI water to neutralize the samples. Regenerated TiNTs were repeatedly used for the consecutive adsorption experiments. 3. Results and discussion 3.1. Morphology of titanate nanotubes

2. Experimental section 2.1. Synthesis of titanate nanotubes TiNTs were synthesized via the typical alkaline hydrothermal method. Typically, 4 and 12 g of TiO2 particles (Degussa P25) were added to 200 and 600 mL of 10 M NaOH solution, respectively. Depending on the volume of the hydrothermal reaction, TiNT sam-

Fig. 1 shows the SEM and TEM images of hydrothermally synthesized TiNT samples with different reaction volumes. The formation of nanostructures during the hydrothermal process follows a 3D ? 2D ? 1D mechanism [22,23]. The SEM images shown in Fig. 1a and b revealed that 10 M of NaOH was sufficient to induce the formation of the fibril-type nanostructure of titanates. Further observation from TEM images (Fig. 1c and d) confirmed that the

505

J. Ryu et al. / Chemical Engineering Journal 304 (2016) 503–510

(a)

(b)

(c)

(d)

Fig. 1. SEM images of (a) TiNT-S and (b) TiNT-L and TEM images of (c) TiNT-S and (d) TiNT-L.

The synthesized TiNT samples were tested as adsorbents to separate Sr2+ ions from water. It was shown that the sorption reaction of Sr2+ was fast, as the equilibrium was reached within 10 min (Fig. S2). More than 95% of the Sr2+ could be removed on TiNTs within 2 min. This rapid adsorption implies that the reaction occurs on the surface of the adsorbents in the absence of internal diffusion resistance. This type of sorption kinetics is in good agreement with a pseudo-second-order model (see SI) [25]. The obtained kinetic parameters are listed in Table S1. The correlation coefficient (R2) was very high (0.999), while the R2 value of the

120 TiNT-S TiNT-L

100 80

5 4

60

Ce / Qe (g/L)

3.2. Sorption behavior of Sr2+

intra-particle diffusion model fitting introduced for comparison was 0.851. This implies that the overall rate of the sorption process is controlled by chemisorptions rather than mass transport [26]. To evaluate the sorption capacity of TiNT samples, sorption experiments were conducted with increasing initial concentrations of Sr2+ in the range of 10 mg/L to 200 mg/L and during a fixed reaction time of 30 min according to the fast sorption kinetics. As shown in Fig. 2, the uptake of Sr2+ at a pH of 8 proportionally increased with the increasing initial concentration of Sr2+ from

Sr uptake (mg/g)

product has the tubular structure. It was also observed that the overall size of TiNT-S was smaller than that of TiNT-L, which might be attributed to the agitation during the hydrothermal reaction. The continuous stirring during the hydrothermal process can accelerate the reaction rate, resulting in the formation of larger and more stable nanofibers, even at low temperatures [24]. TiNT-S was synthesized using the 200-mL autoclave reactor in the absence of agitation, whereas TiNT-L was prepared using the 600-mL reactor with continuous stirring. Consequently, the difference in the size between the two samples prepared in this study demonstrated the effect of agitation on the alkaline hydrothermal reaction, which was in good agreement with the above-mentioned trend. The specific surface area of titanate samples was measured as 260.7 m2/g for TiNT-S and 211.2 m2/g for TiNT-L, which can also be explained by the fact that the overall size of TiNT-S is smaller than that of TiNT-L; thus, a smaller size induces a larger surface area. The large surface area of the TiNT samples arising from the tubular structure makes them very promising candidates for the adsorption of target elements.

40

3 2 2

R = 0.999 Qmax = 91.74 mg/g KL = 0.6 L/mg

1

20

0 0

100

200

300

400

500

Ce (mg/L)

0 0

50

100

150

200

[Sr]initial (mg/L) Fig. 2. Sr2+ sorption efficiency of TiNT samples as a function of the initial concentration. The inset shows the corresponding Langmuir plot of TiNT-S. ([TiNT] = 1 g/L, pHi = 8, and contact time = 30 min.)

506

J. Ryu et al. / Chemical Engineering Journal 304 (2016) 503–510

10 mg/L to 100 mg/L and reached a plateau above 100 mg/L. It was also observed that there was little difference in the Sr2+ adsorption capacities of TiNT-S and TiNT-L, which implies that the discrepancy

TiNT

Intensity (a.u.)

TiNT-L TiNT-S TiNT-S(Sr-adsorbed) TiNT-S(Regenerated)

10

20

30

40

50

60

70

3.3. Sorption mechanism of Sr2+

80

2θ (degree) Fig. 3. X-ray diffraction (XRD) patterns of pristine TiNTs, the TiNT sample after Sr adsorption ([Sr2+]0 = 100 mg/L) and the regenerated TiNT sample by HCl treatment.

(a)

276 288

of the surface areas for two samples was not sufficient to affect the Sr2+ sorption performance. The inset in Fig. 2 shows that the adsorption of Sr2+ fit the Langmuir model (see SI) very well according to the correlation coefficient value (R2) of 0.999, suggesting that Sr2+ sorption was achieved with homogeneous monolayer coverage [27]. Thus, the entire surface has an identical adsorption capacity and there is no significant interaction among adsorbates. The maximum adsorption capacity (Qmax) of TiNT was calculated as 91.74 mg/g, which is very close to the experimental value.

The crystal structure of TiNT samples was analyzed by XRD, and the XRD patterns are shown in Fig. 3. As seen, the main peaks of pristine titanates (TiNT-L and TiNT-S) are in good agreement with those of the primitive monoclinic Na2Ti3O7 phase (JCPDs No. 720148). Meanwhile, in the case of TiNT-S after adsorption of Sr2+ ([Sr]0 = 100 mg/L), it was observed that the overall intensity of the main peaks decreased, while a new peak did not appear, implying that no precipitate was formed on the surface of titanate. Notably, the intensity of diffractions at approximately 2h = 9.6° (100 plane) and 28° (003 plane) substantially decreased. This is because Na+ ions located in the interlayer of TiO6 octahedra were exchanged with Sr2+, followed by the change of periodicity in each plane, thus causing structural deformation, resulting in an intensity decrease [18,20]. Regarding TiNT-S after acid treatment to

447 702

Intensity (a.u.)

60 TiNT(Sr100)

TiNT(Sr50)

TiNT

200

400

600

800

Zeta Potential (mV)

908

1000

a

(a) TiNT-L TiNT-S

40 20 0

2

4

6

8

10

-20

pH

-40

-1

Raman Shift (cm )

-60

a

2.5

100

(b)

2.0

Sorption (%)

Na Release (mmol/g)

(b)

1.5

1.0

Slope = 1.92 R2 = 0.988

80 60 40

TiNT-L TiNT-S

0.5

20 0.0 0.0

0.2

0.4

0.6

0.8

1.0

Sr Uptake (mmol/g) Fig. 4. (a) Raman spectra of TiNT samples before and after Sr adsorption (Sr50 and Sr100 denote Sr-adsorbed samples where [Sr2+]0 is 50 mg/L and 100 mg/L, respectively) and (b) the correlation plot between Sr2+ uptake and the corresponding Na+ release. ([TiNT-S] = 1 g/L, pHi = 8, and contact time = 30 min.)

0 2

4

6

8

10

pHInitial Fig. 5. (a) Zeta potential of TiNT samples and (b) Sr2+ sorption efficiency as a function of the solution pH. ([TiNT] = 1 g/L, [Sr2+]0 = 10 mg/L, and contact time = 30 min.)

507

J. Ryu et al. / Chemical Engineering Journal 304 (2016) 503–510

regenerate the sample, the intensity of the diffractions at the 100 and 003 phases decreased more significantly, indicating acceleration of the structural deformation. The Raman spectra of the samples before and after ion exchange by Sr2+ are shown in Fig. 4a. For pristine TiNT-S, the spectra are virtually the same as the data reported for layered titanate prepared under low NaOH concentrations or low temperatures [28,29]. After Sr2+ adsorption (both cases of [Sr]0 = 50 and 100 mg/L), different peaks were not observed, except for a small overtone at 288 cm 1, suggesting that no new phase was formed. A small overtone at 288 cm 1 might be due to the local distortion of TiO6 octahedral structure caused by Sr2+ adsorption [30]. Meanwhile, it should be noted that the band at 908 cm 1 disappeared after Sr2+ adsorption. The peak at approximately 908 cm 1 is attributed to a four-coordinated Ti-O stretching vibration involving nonbridging oxygen atoms that are coordinated with Na ions, representing a TiO-Na stretching vibration [22,28]. Therefore, it could be concluded that Sr2+ adsorption proceeded via an exchange reaction between Na+ and Sr2+ ions, evidenced by the disappearance of the Ti-O-Na stretching vibration. To quantitatively evaluate the ion exchange, the concentration of Na+ released during the Sr2+ adsorption was monitored and correlated with the amount of Sr2+ adsorbed. As shown in Fig. 4b, the amount of Na+ released was proportional to the amount of Sr2+ adsorbed, and the slope was calculated as 1.92 (R2 = 0.988). This clearly shows that 1 mol of Sr2+ ions must be exchanged with 2 mol of Na+ ions, which confirmed that the ion exchange reaction in the titanate occurs with a stoichiometric balance.

Seawater contains a tremendous number of matrix ion types, making it highly complex; thus, it is highly desirable to investigate the adsorption behavior of titanates under various solution chemistry conditions prior to application in a practical seawater treatment. Accordingly, we examined the effects of pH and coexisting ions (major cations in seawater) on the adsorption of Sr2+ by the TiNT samples. Fig. 5a shows the variation of the zeta potential of titanates as a function of the solution pH. Similar profiles were observed for TiNT-S and TiNT-L, and the isoelectric point (IEP) was found to be at a pH of approximately 3.5. As shown in Fig. 5b, the variation of Sr2+ uptake at different pH values was well correlated to that of the surface charge. At a pH of 2, both samples of TiNT-S and TiNT-L displayed significantly reduced sorption efficiencies of less than 20%, while there was only a slight effect when the pH was above 4. This is due to the electrostatic repulsion between the positively charged adsorbent surface and cationic adsorbates (Sr2+) at a pH of 2. In addition, competition between Sr2+ and protons could be involved in the reduced performance because protons possess the highest priority in the ion exchange. To evaluate the effect of seawater matrix ions on Sr2+ sorption behavior, we conducted Sr2+ sorption experiments in the presence of Na+, K+, Mg2+, and Ca2+, which are representative cations dissolved in seawater. The concentration of each cation ranged from 10 mg/L to its seawater concentration, while that of Sr2+ was fixed at 10 mg/L. Fig. 6a shows that Na+ only slightly affected Sr2+ sorption, with merely a 15% reduction of sorption efficiency at the sea-

(a)

100

100

80

Sorption (%)

Sorption (%)

3.4. Effects of solution chemistry on the sorption of Sr2+

60 40 20

80 60 40 20

0

(b)

0 1

10

100

1000

10000

0

100

+

300

400

[K ] (mg/L)

100

100

80

80

Sorption (%)

Sorption (%)

[Na ] (mg/L)

60 40 20

200 +

(d)

60 40 20

(c)

0

0 0

200

400

600 2+

800

[Mg ] (mg/L)

1000

1200

0

100

200

300

400

2+

[Ca ] (mg/L)

Fig. 6. Effects of co-existing ions (a) Na+, (b) K+, (c) Mg2+, and (d) Ca2+ on Sr2+ sorption efficiency. ([TiNT-S] = 1 g/L, [Sr2+]0 = 10 mg/L, pHi = 8, and contact time = 30 min.)

J. Ryu et al. / Chemical Engineering Journal 304 (2016) 503–510

2+

3.5. Separation of Sr

3.0

(a) Ca uptake / Sr uptake

water concentration level of Na+ (10,000 mg/L). It should be noted that the presence of Na+ in the solution did not deteriorate Sr2+ sorption, although Sr2+ sorption on TiNT-S is based on the ion exchange reaction between Na+ and Sr2+. Previous studies have reported that the presence of Na+ greatly affected the Sr2+ adsorption properties, suggesting that controlling the Na+ concentration of the solution is necessary prior to conducting a sorption experiment using titanate and zeolite materials [12,18]. These contrary observations might be due to the different crystallinity of titanates arisen from different synthetic condition. The hindrance effect on Sr2+ sorption was negligible in the presence of K+ and Mg2+ up to seawater concentrations (Fig. 6b and c). However, Ca2+ significantly decreased Sr2+ sorption, as shown in Fig. 6d, displaying a 50% and 85% decrease of sorption efficiency at 100 mg/L and the seawater concentration of 400 mg/L, respectively. Sr2+ sorption was influenced by the co-existence of 4 cations, displaying a similar trend to the effect of Ca2+ only (Fig. S3). Generally, it is well known that Sr2+ and Ca2+ have similar chemical behaviors. There are three important factors affecting the selectivity of the ion exchange reaction: valence, absolute hardness, and ionic radius. Cations with a higher valence, lower hardness, and smaller radius are preferentially adsorbed on the ion-exchanger [22]. The hardness values and ionic radii of Sr2+ and Ca2+ are 16.3 (hard) and 19.7 (hard) and 1 and 1.18, respectively (Table S2) [22]. Thus, for Sr2+ and Ca2+ with the same valence, Sr2+ has priority over Ca2+ regarding hardness, whereas Ca2+ is more favorably exchanged from the perspective of ionic radius. Consequently, the significant competition between Sr2+ and Ca2+ in the ion-exchange reaction could be ascribed to the very similar chemical properties of both cations. To systematically investigate the competition effect of Ca2+ on Sr2+ sorption, we performed Sr2+ sorption experiments while varying the molar ratio of Sr2+ and Ca2+. Fig. 7a clearly shows that the molar uptake ratio is linearly proportional to the molar ratio of Sr2+ and Ca2+ added. The uptake of Sr2+ was higher than that of Ca2+ when the concentration of Sr2+ in the solution was higher than that of Ca2+ ([Ca]/[Sr] < 1), and vice versa. In addition, sorption experiments under the condition of equimolar concentrations of Sr2+ and Ca2+ were conducted to evaluate the selectivity for Sr2+ sorption in terms of the distribution coefficient (Kd) (see SI). Because the Kd value implies the ratio of the amount of metal ions adsorbed on the adsorbent to the amount of metal ions remaining in the solution, it can be the crucial factor in determining the selectivity for Sr2+ sorption to the adsorbent. Fig. 7b exhibits a linear correlation between Kd for Sr2+ and Kd for Ca2+, suggesting that the sorption behavior for Sr2+ and Ca2+ does not change if the molar ratio of Sr2+ to Ca2+ in the solution is equivalent and constant, despite increasing the concentrations of both cations. The slope of the linear correlation of Kd values was calculated as 1.16, representing a slightly higher selectivity for Sr2+ sorption than Ca2+. However, considering the concentrations of Sr2+ and Ca2+ in seawater (ca. 0.09 mM vs. 10 mM), it was natural that the advantage for Sr2+ sorption with respect to the selectivity was obviously offset in a seawater medium by the large difference of their concentrations, as shown in Fig. 7a.

2.5 2.0 Slope = 0.7237 R2 = 0.9854

1.5 1.0 0.5 0.0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

[Ca] / [Sr]

(b)

80000

KdSr (mL/g)

508

60000 Slope = 1.161 R2 = 0.9998

5000

2500

0 0

2500

5000

KdCa

60000

80000

(mL/g)

Fig. 7. (a) Uptake ratio of Ca2+ to Sr2+ as a function of the molar ratio of Ca2+ to Sr2+ added, (b) the correlation of the distribution coefficient for Sr2+ and Ca2+ at an equimolar concentration of Sr2+ and Ca2+ ranging from 0.25 mM to 2 mM. ([TiNT-S] = 1 g/L, pHi = 8, and contact time = 30 min.)

ions occupied most of the sorption sites of TiNTs; thus, the ion exchange was not available for Sr2+ with a much lower concentration. The competitive sorption of Sr2+ over Ca2+ on TiNTs could occur only under the condition that the molar ratio of Sr2+ to Ca2+ was over 10% at the given loading of TiNTs, 1 g/L. Alternately, it is acceptable that Sr2+ sorption will be improved when increasing the dosage of TiNTs to sufficiently cover the Ca2+ sorption in seawater. The dosage of TiNTs was increased to 15 g/L and was applied to the sorption experiment in seawater. As summarized in Table 1, the equilibrium concentration of Sr2+ decreased from 7.7 to 3.9 mg/ L after 30 min of contact, while that of Ca2+ also decreased from

Table 1 Sr2+ separation performance of TiNTs from other competing cations in seawater.

a

from a real seawater medium 2+

TiNTs were applied to the sorption of Sr in a real seawater medium. First, because Ca2+ with a relatively high concentration in seawater significantly impedes the sorption of Sr2+ on TiNTs, Sr2+-spiked seawater was used to test the sorption performance at a 1 g/L loading of TiNTs and at Sr2+ concentrations of 25, 50, 100 and 200 mg/L. In these experiments, Sr2+ sorption occurred over an Sr2+ concentration of 100 mg/L, revealing an Sr2+ uptake of 2.4 and 9.3 mg/g for Sr2+ concentrations of 100 and 200 mg/L, respectively. This is because a large amount of competitive Ca2+

Sr Ca Mg K Na

C0 (mg/L)

Ce (mg/L)

Kd (mL/g)

b

7.7 405 1275 390 10,402

3.9 268 1223 334 10,276

61.7 33.8 0.6 11.1 0.8

2.4 97.6 58.3 43.8 297.7

Cd (mg/L)

c

CF (10

3

L/g)

20.9 16.1 3.1 7.5 1.9

a Experimental conditions with [TiNT-S] = 15 g/L, contact time = 30 min and under continuous stirring. b Concentration of metals in the desorption solution ([HCl] = 0.1 M). c Concentration factor (CF) = Qe/C0, where Qe (mg/g): metal uptake, C0 (mg/L): initial concentration in seawater. CF was calculated based on the desorption data.

J. Ryu et al. / Chemical Engineering Journal 304 (2016) 503–510

405 to 268 mg/L, revealing that Sr2+ can be separated by TiNTs in a real seawater medium despite a high concentration of Ca2+. The distribution coefficient (Kd) of Sr2+ and Ca2+ calculated based on the metal ion uptake was determined to be 61.7 and 33.8, respectively, indicating a higher selectivity for Sr2+ than for Ca2+. To remove Sr2+ from seawater, the strategy of increasing the dosage of TiNTs can be an effective method to improve the separation efficiency of Sr2+, although it decreased the Sr2+ uptake per adsorbent to 0.25 mg/g. The concentration of desorbed ions (Cd) after acid treatment by 0.1 M of HCl was monitored, and it was employed for the calculation of the concentration factor (CF), which represents the ratio of the metal uptake to the initial metal concentration in seawater. A high CF implies a more efficient concentration of metal ions with low concentration. As listed in Table 1, the CF values for Sr2+ and Ca2+ were estimated as 20.9 and 16.1, respectively, displaying much higher values than those for other cations. The observation that the CF for Sr2+ was slightly higher than that for Ca2+ can be meaningful from the perspective of recovering Sr2+ ions from seawater. Along with the strategy of increasing the dosage of TiNTs, the use of Ca2+-removed seawater can be an alternative method to minimize the effect of Ca2+ on Sr2+ sorption. To remove Ca2+ ions from seawater, we attempted to form the Ca precipitate Ca(OH)2 by the addition of NaOH (see Fig. S4). The Sr2+ uptake for Ca2+removed seawater was observed to be 3.86 mg/g at [TiNTs] = 1 g/ L, which implies that 77% of the Sr2+ ions in pretreated seawater were adsorbed on TiNTs. Although the concentration of Na+ in pretreated seawater increased to 0.93 mol/L, which is approximately 2-fold higher than seawater due to the addition of NaOH, the Sr2+ sorption proceeded successfully. In contrast to the previous report in which the presence of Na+ significantly impeded Sr2+ sorption [12], this result reinforced that the TiNTs prepared in this work have an outstanding performance for Sr2+ sorption without the negative effect of Na+. 3.6. Regeneration of TiNTs Considering that Sr2+ sorption was noticeably reduced at a low pH, an HCl solution was used to regenerate TiNTs. Overall, the TiNTs synthesized in this study could be easily regenerated and reused several times (see SI for detailed results), guaranteeing its repetitive use for the practical application of removing and recovering Sr2+ ions from seawater. 4. Conclusions To date, many studies on adsorbent materials, including titanates, have been limited to the removal of heavy metals and radioactive isotopes in wastewater and radioactive wastewater, respectively. In this study, with the goal of removing and recovering Sr2+ from seawater, Sr2+ sorption behaviors on TiNTs were systematically evaluated under various reaction conditions. Among seawater matrix ions, Ca2+ significantly hindered Sr2+ sorption performance while there was only a slight effect of Na+. Although the selectivity of TiNTs for Sr2+ sorption was evaluated to be slightly higher than that for Ca2+, Sr2+ sorption was not favorable in a seawater medium because the concentration of Ca2+ in seawater is much higher than that of Sr2+. Based on the Sr2+ sorption behaviors observed under various conditions, it could be concluded that a high dosage of TiNTs is desirable to overcome the hindering effect of Ca2+ in the removal of Sr2+ from radio-contaminated seawater, and the use of treated seawater (e.g., Mg2+-extracted and further treated seawater by adding NaOH in the sequential recovery process of Mg2+ and Sr2+) with a reduced Ca2+ concentration can be a feasible method to recover Sr2+ by TiNTs from seawater.

509

Acknowledgments This research was supported by the Basic Research Project (GP2015-007, 16-3224) of the Korea Institute of Geoscience and Mineral Resources (KIGAM) funded by the Ministry of Science, ICT and the Future Planning of Korea and the DGIST R&D Program of the Ministry of Education, Science and Technology of Korea (14NB-03).

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2016.06.131.

References [1] K. Hirose, Fukushima Dai-ichi nuclear power plant accident: summary of regional radioactive deposition monitoring results, J. Environ. Radioact. 111 (2012) (2011) 13–17. [2] P.P. Povinec, K. Hirose, M. Aoyama, Radiostrontium in the western North Pacific: characteristics, behavior, and the Fukushima impact, Environ. Sci. Technol. 46 (2012) 10356–10363. [3] A. Sachse, A. Merceille, Y. Barré, A. Grandjean, F. Fajula, A. Galarneau, Macroporous LTA-monoliths for in-flow removal of radioactive strontium from aqueous effluents: application to the case of Fukushima, Microporous Mesoporous Mater. 164 (2012) 251–258. [4] U. Bardi, Extracting minerals from seawater: an energy analysis, Sustainability 2 (2010) 980–992. [5] E.E. Angino, G.K. Billings, N. Andersen, Observed variations in the strontium concentration of sea water, Chem. Geol. 1 (1966) 145–153. [6] M.J. Driscoll, F.R. Best, Progress Toward the Recovery of Uranium From Seawater, MIT, USA, 1982. [7] C. Xu, J. Wang, J. Chen, Solvent extraction of strontium and cesium: a review of recent progress, Solvent Extr. Ion Exch. 30 (2012) 623–650. [8] A.D. Ebner, J.A. Ritter, J.D. Navratil, Adsorption of cesium, strontium, and cobalt ions on magnetite and a magnetite silica composite, Ind. Eng. Chem. Res. 40 (2001) 1615–1623. [9] S. Chegrouche, A. Mellah, M. Barkat, Removal of strontium from aqueous solutions by adsorption onto activated carbon: kinetic and thermodynamic studies, Desalination 235 (2009) 306–318. [10] D.V. Marinin, G.N. Brown, Studies of sorbent/ion-exchange materials for the removal of radioactive strontium from liquid radioactive waste and high hardness groundwaters, Waste Manage. (Oxford) 20 (2000) 545–553. [11] Y.D. Noh, S. Komarneni, K.J.D. Mackenzie, Titanosilicates: giant exchange capacity and selectivity for Sr and Ba, Sep. Purif. Technol. 95 (2012) 222–226. [12] A. Merceille, E. Weinzaepfel, Y. Barre, A. Grandjean, The sorption behaviour of synthetic sodium nonatitanate and zeolite A for removing radioactive strontium from aqueous wastes, Sep. Purif. Technol. 96 (2012) 81–88. [13] T.P. Valsala, A. Joseph, N.L. Sonar, M.S. Sonavane, J.G. Shah, K. Raj, V. Venugopal, Separation of strontium from low level radioactive waste solutions using hydrous manganese dioxide composite materials, J. Nucl. Mater. 404 (2010) 138–143. [14] S.-K. Jeong, C.-S. Ju, Extraction of strontium ion from sea water by contained liquid membrane permeator, Korean J. Chem. Eng. 19 (2002) 93–98. [15] V.A. Avramenko, V.V. Zheleznov, E.V. Kaplun, T.A. Sokol’nitskaya, A.A. Yukhkam, Sorption recovery of strontium from seawater, Radiochemistry 43 (2001) 433–436. [16] H.-J. Hong, J. Ryu, I.-S. Park, T. Ryu, K.-S. Chung, B.-G. Kim, Investigation of the strontium (Sr (II)) adsorption of an alginate microsphere as a low-cost adsorbent for removal and recovery from seawater, J. Environ. Manage. 165 (2016) 263–270. [17] D. Song, S.-J. Park, H.W. Kang, S.B. Park, J.-I. Han, Recovery of lithium (I), strontium (II), and lanthanum (III) using Ca–alginate beads, J. Chem. Eng. Data 58 (2013) 2455–2464. [18] D. Yang, Z. Zheng, H. Liu, H. Zhu, X. Ke, Y. Xu, D. Wu, Y. Sun, Layered titanate nanofibers as efficient adsorbents for removal of toxic radioactive and heavy metal ions from water, J. Phys. Chem. C 112 (2008) 16275–16280. [19] S.-S. Liu, C.-K. Lee, H.-C. Chen, C.-C. Wang, L.-C. Juang, Application of titanate nanotubes for Cu (II) ions adsorptive removal from aqueous solution, Chem. Eng. J. 147 (2009) 188–193. [20] T. Wang, W. Liu, N. Xu, J. Ni, Adsorption and desorption of Cd (II) onto titanate nanotubes and efficient regeneration of tubular structures, J. Hazard. Mater. 250 (2013) 379–386. [21] D. Yang, H. Liu, Z. Zheng, S. Sarina, H. Zhu, Titanate-based adsorbents for radioactive ions entrapment from water, Nanoscale 5 (2013) 2232–2242. [22] N. Li, L. Zhang, Y. Chen, M. Fang, J. Zhang, H. Wang, Highly efficient, irreversible and selective ion exchange property of layered titanate nanostructures, Adv. Funct. Mater. 22 (2012) 835–841.

510

J. Ryu et al. / Chemical Engineering Journal 304 (2016) 503–510

[23] Y. Wang, G. Hu, X. Duan, H. Sun, Q. Xue, Microstructure and formation mechanism of titanium dioxide nanotubes, Chem. Phys. Lett. 365 (2002) 427– 431. [24] D.V. Bavykin, A.N. Kulak, F.C. Walsh, Metastable nature of titanate nanotubes in an alkaline environment, Cryst. Growth Des. 10 (2010) 4421–4427. [25] Y.-S. Ho, G. McKay, Pseudo-second order model for sorption processes, Process Biochem. 34 (1999) 451–465. [26] E.-J. Kim, C.-S. Lee, Y.-Y. Chang, Y.-S. Chang, Hierarchically structured manganese oxide-coated magnetic nanocomposites for the efficient removal of heavy metal ions from aqueous systems, ACS Appl. Mater. Interfaces 5 (2013) 9628–9634.

[27] H.-J. Hong, I.-S. Park, J. Ryu, T. Ryu, B.-G. Kim, K.-S. Chung, Immobilization of hydrogen manganese oxide (HMO) on alpha-alumina bead (AAB) to effective recovery of Li+ from seawater, Chem. Eng. J. 271 (2015) 71–78. [28] X. Sun, Y. Li, Synthesis and characterization of ion-exchangeable titanate nanotubes, Chem. Eur. J. 9 (2003) 2229–2238. [29] W. Liu, P. Zhang, A.G.L. Borthwick, H. Chen, J. Ni, Adsorption mechanisms of thallium (I) and thallium (III) by titanate nanotubes: ion-exchange and coprecipitation, J. Colloid Interface Sci. 423 (2014) 67–75. [30] W. Hu, L. Li, G. Li, Y. Liu, R.L. Withers, Atomic-scale control of TiO6 octahedra through solution chemistry towards giant dielectric response, Sci. Rep. 4 (2014) 6582.