I. INTRODUCTION
Large amounts of radioactive contaminated wastewater (RCW) were generated from the nuclear accident at Fukushima NPP-1 in Japan. The RCW is reactor cooling water collected after contact with damaged nuclear fuel debris, so the main radionuclides are waste-soluble Cs (Cs-134, Cs-137), Sr (Sr-90), and I (I-131, I-129) [1]. With a half-life of 1.7×107 years, I129 is a primary long-term risk driver in shallow land disposal facilities. I131 would be an acute contaminant because of its short half-life of 8.05 days and high specific activity [2]. Releasing the untreated iodine contaminants into the environment would pollute the soil, air, and groundwater, resulting in long-term radioecological risk.
Aqueous iodine exists primarily as an iodide anion (I-) and iodate anion (IO3-), depending on the redox conditions and pH. At low to neutral pH and positive redox potentials, the iodide anion is the dominant species in a solution environment [3]. The Fukushima RCW contains various co-existent components from sea water (saline elements), corrosion products, and ground-water. Its pH ranges from 7 ∼ 8 [4]. In previous works, various methods have been developed to remove or adsorb iodide anion from aqueous solutions [5-7]. Common natural porous materials, such as alumina and hydrotalcite can absorb iodide anions through surface physical adsorption or ion exchange with an active group of the adsorbent materials [8]. However, the presence of the coexisting chloride anion in RCW can compete against iodide anion and render the natural materials without adsorption selectivity for iodide anions. Inorganic anion exchangers, such as BiPbO2(NO3) were also reported to adsorb iodide anions [9]. But such inorganic anion exchangers are usually not easy to prepare and sometimes its toxicity is also a problem. The cuprous (Cu+)-containing compounds react with iodide anions to form cuprous iodide through the participation of hydrogen anions, this adsorption process is very sensitive to the pH of the solution, so the adsorption amount is rather low in neutral and slightly acidic solutions [10]. Recently, the silver adsorbents were investigated. The adsorption mechanism mainly related to the strong chemical interaction between silver and iodide anions, and the adsorbents have good selectivity. However, the leakage of the impregnated materials could result in the lowering of the adsorbed amount [11].
In this work, we attempted to synthesize a macroporous silica-based silver loaded adsorbent (Ag(tu)3NO3/SiO2-P) by grafting the silver complexes of thiourea (Ag(tu)3NO3) into a silica-based copolymer support (SiO2-P) for adsorbing iodide anions. The SiO2-P support is a kind of inorganic macroporous material, which was prepared by impregnating the copolymer inside the macroporous SiO2 substrate [12]. Grafting the Ag(tu)3NO3 into SiO2-P has a number of advantages, such as mechanical strength, strong acid, chemical stability, radiation resistance, and the ease of solid-liquid separation. Compared to the reported iodide adsorbents, the synthesized Ag(tu)3NO3/SiO2-P is easy to prepare and can avoid silver leaching during adsorption. In this study, the adsorption kinetics, mechanisms, and dynamic adsorption behavior of iodide anions for Ag(tu)3NO3/SiO2-P were investigated.
II. EXPERIMENTAL
A. Materials
The reagents used in the experiments, such as Sulfuric acid (98%), Nitric acid (65%), hydrochloric acid (37%), Ethanol, Tin(II) chloride dehydrate, Ammonium thiocyanate, and Silver nitrate, are analytical grade or better.
The SiO2-P was synthesized based on a known method [13]. For simplicity, "P" in the SiO2-P particles was abbreviated as styrene-divinylbenzene (SDB) copolymer, which was immobilized inside the macroporous SiO2 substrate through a polymerization reaction.
B. Preparation of Ag(tu)3NO3/SiO2-P
The synthesis procedures of Ag(tu)3NO3/SiO2-P were performed as follows:
(1) SiO2-P-NO2 was prepared by nitration of the copolymer in SiO2-P. SiO2-P was mixed with nitric acid (65%) and sulfuric acid (98%) in a water bath (at 50 ℃) for 3 h.
(2) SiO2-P-NO2 was reduced to SiO2-P-NH2 in the presence of stannous chloride, hydrochloric acid (37%), and Ethanol in a water bath (at 60 ℃) for 12 h.
(3) SiO2-P-tu was prepared by mixing SiO2-P-NH2, Ethanol, and Ammonium thiocyanate into a flask with water bath heating under 70 ℃, stirring for 12 h, filtering and drying under 50 ℃.
(4) Ag(tu)3NO3/SiO2-P was prepared by mixing an SiO2-P-tu and silver nitrate solution into a flask, stirring for 12 h, and filtering and drying under 50 ℃. The synthesis procedures are shown in Fig. 1. Many of the diverse structural types of Ag(I)/(x)tu (x=1 n) complexes have been reported. It was claimed that the Ag(I)/(3)tu complexes were the most common types [14-17].
C. Characterization
SEM analysis (Nova NanoSEM NPE218) was carried out to study the surface morphology of Ag(tu)3NO3/SiO2-P. The thermal stability of SiO2-P and Ag(tu)3NO3/SiO2-P were evaluated. The analyses were performed by TG-DTA equipment (Shimadzu DTG-60) with a 10 ℃/min constant heating rate from 25 to 800 ℃. The FT-IR spectrum of the Ag(tu)3NO3/SiO2-P was recorded in the range of 4000–500 cm-1 using the instrument IRAffinity-1 FT-IR spectrometer.
D. The Static Experiments
The static experiment adsorption behavior of tested iodide anions on the adsorbent was examined by batch experiments using Ag(tu)3NO3/SiO2-P. An aqueous phase (5 mL), containing varying iodide anion concentrations, was equilibrated with 0.1 g adsorbents in stopper glass tubes in a thermo state water bath (Tokyo RIKAKIKA Co.LTD) at room temperature. The concentration of the tested iodide anions before and after adsorption was measured by ICP-AES(Shimadzu 7510). The adsorption capacity at equilibrium (Qeq) can be calculated using the following formula:
where C0 and Ce are the initial and equilibrium concentrations of the iodide anions, m is the weight of the adsorbent, and V is the volume of the aqueous phase.
E. Column operation
The Ag(tu)3NO3/SiO2-P (4.05 g) was densely packed into a glass column (10 mm in diameter, 200 mm long) with a thermo jacket set at 25(1) ℃, shown in Fig. 2. A breakthrough of iodide anions was tested at a feed solution of 1 mM NaI and the flow rate was controlled to 0.5 cm3/min.
III. RESULTS AND DISCUSSION
A. SEM
The SEM micrographs of Ag(tu)3NO3/SiO2-P are shown in Fig. 3. Obvious spherical and porous particles were obtained. The practical size of Ag(tu)3NO3/SiO2-P was estimated to be 50 μm in diameter. From the smooth surface, it is found that the active ingredients were all impregnated inside the SiO2-P.
B. TG-DTA
The results of TG and DTA for SiO2-P are shown in Fig. 4. It indicates two different weight loss ranges: 280–350 ℃ and 350–550 ℃. The first and second weight losses were due to the thermal desorption of the SDB copolymer [18]. The overall weight loss of SiO2-P was estimated to be 15%, indicating that 15 wt.% SDB was impregnated inside the SiO2 substrate and the content of SiO2 was calculated to be 85 wt.%.
The results of TG and DTA for Ag(tu)3NO3/SiO2-P are shown in Fig. 5. It indicates three different weight loss ranges: the first and second weight losses denoted the burning of the SDB polymer, similar to those of Fig. 4. The third weight loss (400–550 ℃) was interpreted as the thermal decomposition of the silver complex of thiourea. The corresponding endothermic peak around 540 ℃ was also observed.
Thus, the content of the silver complex of thiourea grafted onto the copolymer was estimated to be 12%.
C. Kinetic adsorption studies
In this experiment, the effect of the contact time on the adsorption of the iodide anions by Ag(tu)3NO3/SiO2-P was studied. To find out the relationship between the time of adsorption (t) and the amount adsorbed at a time of t(qt) are plotted and shown in Fig. 6.
D. FT-IR studies
The FT-IR spectrum of the Ag(tu)3NO3/SiO2-P is shown in Fig. 7. Since the Ag(tu)3NO3 was grafted into the holes of SiO2-P, the FT-IR spectrum of the Ag(tu)3NO3/SiO2-P shown in Fig. 7(a) was mainly dominated by the Si-O groups. In order to observe other chemical bonds clearly, partial enlarged details are shown in Fig. 7(b)–7(d). The sharp band observed at 1080 cm-1 in Fig. 7(a) was due to asymmetric vibrations of Si-O-Si. In addition, the band observed at 789 cm-1 was attributed to the symmetric stretching vibration peak of the Si-O bond [19]. After nitration, the IR spectrum of the SiO2-P showed the peaks at 1518 and 1350 cm-1, corresponding to N-O asymmetric and symmetric stretches, respectively. The asymmetric and symmetric vibrations of the NH2 groups were positioned between 3231 and 3178 cm-1, as shown in Fig. 7(d) [20]. The little shoulder observed at 1560 cm-1 in Fig. 7(c) was due to ν(C=S), indicating the co-ordination of the C=S group to the Ag atom in the complex. In addition, the band observed at 768 cm-1 was assigned to the bending vibration of the C=S group. This also supported the co-ordination of the C=S group to the Ag atom in the complex [21, 22].
The adsorption equilibrium of iodide anions (I-) on Ag(tu)3NO3/SiO2-P was attained within 10 min in pure water. The adsorbed amount of I- on Ag(tu)3NO3/SiO2-P was 8.89 mg g-1 (Fig. 6). In 0.6 M NaCl solution (simulated seawater), it was observed that the equilibrium time of adsorption on Ag(tu)3NO3/SiO2-P was similar to that in pure water. The uptake of I- for the Ag(tu)3NO3/SiO2-P adsorbent was governed by the following ion-exchange reactions [23]:
The ability of the thiourea(tu) functional group to form a stable adduct with silver is well established [16]. Ag(tu)3+ is different from Ag+, even in the NaCl solution, because the Ag(tu)3+ will not react with Cl- [24, 25]. Therefore, the Ag(tu)3NO3/SiO2-P is effective for removing the I- from the seawater.
In order to investigate the mechanism of adsorption and determine the rate-controlling step, the following kinetic model is used to test the experimental data.
The pseudo second-order kinetic model is represented by the following equation:
Integrating Eq. (3) with the boundary conditions of (1) at t=0 and qt=0 and (2) at t=t and qt=qt results in the following equation:
where K2 (mg/(g min)) is the adsorption rate constant for pseudo second-order kinetics, qt (mg/g) is the amount of iodide anions adsorbed at time t, and Qeq (mg/g) is the amount of iodide anions adsorbed at equilibrium [26].
The fit lines are shown in Fig. 8. It shows that the rate constants Qeq and the regression values for pseudo second-order kinetics are greater than 0.99, indicating that the adsorption system belongs to the second-order kinetics model. The Qeq and K2 can be determined from a plot of t/qt against t. The calculated values are summarized in Table 1. It shows that the calculated Qeq values agree with the experimental data. These results indicate that I- adsorption was a rate controlled step, which was governed by the chemisorption process.
| Kinetic model | Parameter | Pure water | 0.6 M NaCl |
|---|---|---|---|
| Pseudo second-order | Qeq (mg/g) | 8.96 | 8.98 |
| K2 (mg/(g min)) | 12.45 | 4.42 | |
| R2 | 0.9999 | 0.9988 |
E. Adsorption Isotherms
The Langmuir and Freundlich isotherm models were studied for the adsorption mechanism of iodide anions (I-) on Ag(tu)3NO3/SiO2-P. The isotherm parameters were determined using Origin software, which showed the plots of Ceq versus Qeq for different isotherms.
1. Langmuir isotherm
The Langmuir isotherm is given by the following equation:
where Qmax is the maximum monolayer adsorption capacity (mg/g), KL is the equilibrium constant related to free energy (L/mg), Ceq is the concentration of the iodide anions at equilibrium (mg/L), and Qeq is the amount of iodide anions adsorbed at equilibrium (mg/g) [27].
2. Freundlich isotherm
The Freundlich isotherm applies to multilayer adsorption and adsorption on a heterogeneous surface. It is given by the following equation:
where Qeq is the amount of iodide anions adsorbed at equilibrium (mg/g), Ceq is the concentration of the iodide anions at equilibrium (mg/L), n is the measure of deviation from linearity of adsorption, and KF is the adsorption capacity related to a multilayer (mg/g).
The linear plot of Ceq/ Qeq vs. Ceq shows that adsorption follows a Langmuir isotherm (Fig. 9). The applicability of the Langmuir isotherm suggests monolayer coverage of iodide anions on the surface of Ag(tu)3NO3/SiO2-P. The values of Qmax and KL were calculated from the slope and intercept of the linear plots and presented in Table 2. It shows that the adsorption Qmax value for Ag(tu)3NO3/SiO2-P in 0.6 M NaCl was calculated to be 8.96 mg g-1, similar to that in pure water. On the other hand, the fitted Freundlich plots show a low R2 value around 0.6 (Fig. 10), indicating the inapplicability of the Freundlich isotherm.
| Adsorption model | Parameter | Pure water | 0.6 M NaCl |
|---|---|---|---|
| Langmuir isotherm | Qmax(mg/g) | 8.94 | 8.96 |
| KL | 6.71 | 4.16 | |
| R2 | 0.9999 | 0.9989 | |
| Freundlich isotherm | n | 5.29 | 4.98 |
| KF | 0.04 | 0.04 | |
| R2 | 0.6413 | 0.6561 |
F. The column experiments
The performance of the fixed-bed adsorption is complemented by the dynamic column studies, which can be deduced from the concept of a breakthrough curve. The breakthrough curve is usually expressed in terms of inlet concentration (C0) and outlet concentration (C), and is defined as the ratio of outlet concentration to inlet concentration (C/C0) as a function of effluent for a given bed height [16].
A breakthrough of I- was tested at a feed solution of 1 mM NaI and at a high flow rate of 0.5 cm3/min. Figure 11 illustrates the breakthrough curve of I-, which has a S-shaped profile and a steep slope, suggesting no dislodgement of Ag(tu)3NO3 from the matrix of SiO2-P. The break point of 5% breakthrough was estimated to be 235 and 231 cm3 and the column took approximately 284 and 286 cm3 before being completely exhausted with I-. The breakthrough capacity (B.T.Cap.) and the total capacity (T.Cap.) were calculated to be 7.24 mg g-1 and 7.87 mg g-1, respectively in 0.6 M NaCl solution, resulting in a relatively high column efficiency (B.T.Cap./T.Cap.) of 91.9%. The adsorption behavior of I- on Ag(tu)3NO3/SiO2-P in pure water was close to that in 0.6 M NaCl, indicating that the adsorption of I- was not affected by the presence of the Cl- anions in seawater. The column packed with Ag(tu)3NO3/SiO2-P was thus effective for removing I- from radioactive contaminated wastewater, even containing highly concentrated NaCl.
IV. CONCLUSION
The macroporous silica-based Ag(tu)3NO3/SiO2-P resin was prepared by grafting Ag(tu)3NO3 onto an SiO2 support. The uptake properties of I- were investigated by a batch method and the adsorption behaviors of I- were further studied by the column method in pure water and in 0.6 M NaCl solution.
The thermal stability of Ag(tu)3NO3/SiO2-P indicated the adsorption ability of Ag(tu)3NO3/SiO2-P can be maintained at temperatures up to 200 ℃. The adsorption of I- was fairly fast in both pure water and 0.6 M NaCl solution. The adsorption equilibrium can be attained within 10 min. The kinetic data was modeled successfully using the pseudo second-order kinetic model. The adsorption of I- fit well with the Langmuir model and the correlation coefficient (R2) was more than 0.99, suggesting a monolayer of adsorbent molecules on the surface of Ag(tu)3NO3/SiO2-P. The Ag(tu)3NO3/SiO2-P effectively adsorbed iodide anions in the presence of 0.6 M NaCl in the column operation. From the results of this study, Ag(tu)3NO3/SiO2-P is proposed to be an efficient absorbent for the separation of iodide anions from radioactive waste water.
Radioactive liquid waste treatment at Fukushima Daiichi
. J Chem Technol Biot, 2013, 88: 1592-1596. DOI: 10.1002/jctb.4141Sorption and transport of iodine species in sediments from the Savannah River and Hanford Sites
. J Contam Hydrol, 2005, 78: 185-205. DOI: 10.1016/j.jconhyd.2005.05.007Development and testing of a silver chloride-impregnated activated carbon for aqueous removal and sequestration of iodide
. Environ Technol, 2005, 26: 1255-1262. DOI: 10.1080/09593332608618595Macroporous LTA-monoliths for in-flow removal of radioactive strontium from aqueous effluents: Application to the case of Fukushima
. Microporous Mesoporous Mater, 2012, 164: 251-258. DOI: 10.1016/j.micromeso.2012.07.019Prediction of iodide adsorption on oxides by surface complexation modeling with spectroscopic confirmation
. J Colloid Interf Sci, 2009, 332: 309-316. DOI: 10.1016/j.jcis.2008.12.037Extraction and quantitative analysis of iodine in solid and solution matrixes
. Anal Chem, 2005, 77: 7062-7066. DOI: 10.1021/ac050972vEfficient and selective extraction of iodide through a liquid membrane
. Microchem J, 2001, 69: 45-50. DOI: 10.1016/S0026-265X(00)00188-0Adsorption and desorption of iodine by various Chinese soils: II. Iodide and iodate
. Geoderma, 2009, 153: 130-135. DOI: 10.1016/j.geoderma.2009.07.020Reactivity of inorganic anion exchanger BiPbO2(NO3) with fluoride ions in solution
. Solid State Ionics, 2001, 141-142: 603-607. DOI: 10.1016/S0167-2738(01)00775-5Sorption of iodide on cuprite (Cu2O)
. Langmuir, 2000, 16: 4519-4527. DOI: 10.1021/la9903999Adsorption of iodide ions on a calcium alginate-silver chloride composite adsorbent
. Colloid Surface A, 2011, 386: 166-171. DOI: 10.1016/j.colsurfa.2011.07.014Study on selective separation of cesium from high level liquid waste using a macroporous silica-based supramolecular recognition absorbent
. J Radioanal Nucl Ch, 2012, 293: 13-20. DOI: 10.1007/s10967-012-1738-6Studies on the separation of minor actinides from high-level wastes by extraction chromatography using novel silica-based extraction resins
. Nucl Technol, 2000, 132: 413-423.Structural and infrared spectroscopic studies of some novel mechanochemically accessed adducts of silver(I) oxoanion salts with thiourea
. Inorg Chem, 2009, 48: 3185-3197. DOI: 10.1021/ic802312jStability of silver-thiourea complexes in montmoril-lonite clay
. J Chem Soc, 1975, 71: 256-264. DOI: 10.1039/F19757100256Silver(I) complexes of thiourea
. Transit Metal Chem, 2002, 27: 782-785. DOI: 10.1023/A:1020350023536Leaching of gold and silver from Miller Process dross through non-cyanide leachants
. Hydrometallurgy, 1996, 42: 27-33. DOI: 10.1016/0304-386X(95)00049-MEquilibrium and kinetic studies of selective adsorption and separation for strontium using DtBuCH18C6 loaded resin
. J Nucl Sci Technol, 2012, 49: 320-327. DOI: 10.1080/00223131.2012.660022Sorption of Np(V) onto TiO2, SiO2, and ZnO: An in situ ATR FT-IR spectroscopic study
. Environ Sci Technol, 2009, 43: 7665-7670. DOI: 10.1021/es901256vAuto-deposition of gold on chemically modified polystyrene beads
. Colloid Surface A, 2005, 269: 119-124. DOI: 10.1016/j.colsurfa.2005.06.073Spectroscopic, thermal and mechanical behavior of allylthiourea cadmium chloride single crystals
. Mater Res Bull, 2007, 42: 2040-2047. DOI: 10.1016/j.materresbull.2007.02.002Structural and infrared spectroscopic studies of some novel mechanochemically accessed adducts of silver(I) oxoanion salts with thiourea
. Inorg Chem, 2009, 48: 3185-3197. DOI: 10.1021/ic802312jIodide-selective electrodes based on the silver(I) complex of a novel N-thiocarbamoylimine-dithioether derivative
. Anal Chim Acta, 1996, 329: 247-252. DOI: 10.1016/0003-2670(96)00115-8Ferric chloride, thiourea and brine leach recovery of Ag, Au and Pb from complex sulfides
. JOM, 1986, 38: 18-22. DOI: 10.1007/BF03257810PVC membranes based on silver(I)-thiourea complexes
. Anal Chim Acta, 1999, 402: 129-135. DOI: 10.1016/S0003-2670(99)00562-0Parameter optimization of activated carbon production from Agave sisalana and Punica granatum peel: adsorbents for C.I. reactive orange 4 removal from aqueous solution
. Clean, 2013, 41: 797-807. DOI: 10.1002/clen.201100719Kinetics and isotherm analysis of Tropaeoline 000 adsorption onto unsaturated polyester resin (UPR): a non-carbon adsorbent
. Environ Sci Pollut R, 2013, 20: 1493-1502. DOI: 10.1007/s11356-012-0994-x