Removal of Radioactive Iodide by Surfactant-modified Zeolites

373 Removal of Radioactive Iodide by Surfactant-modified Zeolites Hossein Faghihian 1 *, Akbar Malekpour 1,2 and Mohammad G. Maragheh 2 (1) Department of Chemistry, University of Isfahan, Isfahan, Iran. (2) Atomic Energy Organization of Iran, Tehran, Iran. (Received 13 December 2002; revised form accepted 26 March 2003) ABSTRACT: The removal of radionuclides such as 129 I and 131 I from radioactive liquid wastes was studied. Two natural zeolites were modified with different quaternary alkylammonium ions to replace the exchangeable cations from the zeolite surface and used as adsorbent materials. The quaternary ions used for such purpose were hexadecyltrimethylammonium, tetradecyltrimethylammonium, dodecyltrimethylammonium, tetrabutylammonium and tetramethylammonium, respectively. Some of the modified forms exhibited an adsorption capacity much higher than those of the respective natural samples. In contrast, the adsorption capacity was negligible when tetrabutylammonium and tetramethylammonium ions were used. Adsorption experiments were conducted by batch and continuous experiments, and adsorption isotherms constructed from the data obtained. The effect of interfering anions on the adsorption capacity was also investigated as were the breakthrough behaviours of radioiodide in a column charged with the various adsorbents. Desorption of iodide from the modified zeolites into different solutions was also investigated. It was concluded that, in some cases, surfactant modification was an efficient process for the uptake and immobilization of iodide. INTRODUCTION Several iodine species are produced during the fission of uranium. Of these, 129 I of 0.9% yield has the longest half-life of 1.59 10 7 y while 131 I (t 1/2 = 8.04 d) of 2.77% yield is the predominant radioisotope. Different adsorbents have been studied for the removal of iodide from radioactive wastes. The reported selectivity and capacity is usually below the desired values and immobilization of iodide by the studied adsorbent has rarely been reported. Fetter et al. (1999) studied the adsorption of radioactive iodide by the nitrate form of hydrotalcite but found that iodide replaced only a small fraction of the nitrate. -Bi 2 O 3 was also considered as a promising candidate for the removal of radioactive iodide (Kodama 1992) but unfortunately the HCO 3 anion interferes with the exchange reaction. When BiPbO 2 (NO 3 ) was used, adsorption only occurred on the adsorbent surface (Kodama 1999). Membranes constructed from anion-exchange paper have also been used to remove 125 I from radioactive wastes (Inoue and Kagoshima 2000). The uptake of radioiodide by pretreated soil has been examined under different experimental conditions (Fukui et al. 1996). Removal of radioiodide from radiopharmaceuticals has been performed using a pure metallic silver membrane (Eersels et al. 1995), while the simultaneous removal of 137 Cs and 129 I from radioactive liquid wastes by a mixture of carbon *Author to whom all correspondence should be addressed. E-mail: h.faghih@sci.ui.ac.ir.

374 H. Faghihian et al./adsorption Science & Technology Vol. 21 No. 4 2003 and chabazite has also been studied (Song et al. 1997). Activated carbon was considered for the removal of iodide from radioactive wastes (Deitz 1991; Yang et al. 1993). The sorption characteristics of bentonite and some clays for iodide adsorption have also been studied (Bors et al. 1997; Rehakova et al. 1998). Because of their high selectivity, stability and low price, zeolites have frequently been used for the separation of radioactive cations. These materials are radiation-resistant and their use is compatible with cementation and vitrification. In fact, cations can interact with zeolites via ion exchange on negatively charged internal and external surfaces, but anions are not readily taken up because zeolites possess a net negative structural charge resulting from isomorphic substitution of cations in their crystal lattices. For this reason, zeolites must be modified prior to anion adsorption. Chmielewska-Horvathova and Lesny (1995, 1996) modified clinoptilolite to the silver-exchanged form and used this to remove iodide from liquid wastes. Modification of clinoptilolite and natrolite by ion-exchange processes also increased the adsorption capacity of the zeolites towards radioactive iodide (Faghihian et al. 2002). Surfactant-modified zeolite (SMZ) has been studied extensively as an effective adsorbent for the removal of inorganic anions from water (Li et al. 1998; Bowman et al. 2000). The ion exchange of zeolites with cationic surfactants can alter their surface properties and, as a consequence, their sorption capacities towards anions may be improved substantially. In the present work, the adsorption behaviour of two natural zeolites and their surfactantmodified forms towards radioactive iodide was investigated. The effect of different parameters on the adsorption process was studied in order to optimize the conditions for the removal and immobilization of iodide. EXPERIMENTAL Clinoptilolite was collected from the Absard deposit in Semnan while the natrolite was obtained from the Hasanabad region in Qom, both in Iran. Characterization of the samples has already been reported (Faghihian et al. 2002). To ensure a constant water content after characterization, the samples were allowed to equilibrate for several days over water in a closed chamber containing saturated NH 4 Cl prior to analysis. The particle size of the samples was in the range 224 400 m. The samples were then treated with the following reagents: hexadecyltrimethylammonium bromide (HD); tetradecyltrimethylammonium bromide (TD); dodecyltrimethylammonium bromide (DD); tetrabutylammonium bromide (TB); and tetramethylammonium bromide (TM). The concentration range selected ranged from below to above the critical micellar concentration (0.45 100 mm). Thus, 20 g zeolite sample and 200 ml solution were placed in 500-ml volume bottles and shaken for 24 h at 250 rpm. The supernatants were removed and the samples washed with three portions of deionized water and then air-dried. The equilibrium concentration was measured by turbidimetric methods using sodium dodecyl sulphate (SDS) as the precipitating agent. The modified samples were designated as follows: Cp-HD50 = clinoptilolite modified with 50 mm hexadecyltrimethylammonium; Nt-HD50 = natrolite modified with 50 mm hexadecyltrimethylammonium. In the same manner, the remaining samples were designated as Cp-HD0.45, Cp-HD17, Cp-HD25, Cp-HD67, Cp-HD100, Cp-TD50, Cp-DD50, Nt-HD100, Nt-TD50 and Nt-DD50. Batch adsorption experiments were conducted as follows. A known amount of zeolite sample (0.5 g) was placed in a 50 ml centrifuge tube and 10 ml iodide solution (concentration range 0.0001 0.1000 M) added. The activity of the solution was adjusted to ca. 40 000 Bq (10 6 Ci/l) by the addition of sufficient 131 I to provide enough activity in the final solution for accurate counting. The mixture was shaken for 24 h at 25 C and 250 rpm. After centrifugation, the clear supernatant

Removal of Radioactive Iodide by Surfactant-modified Zeolites 375 solution was removed for iodide measurement. The activities of the initial and supernatant solutions were measured under identical conditions using a -spectrometer (Ortec EG & G model, high pure germanium, high resolution, co-axial GMX detector with 4096 channels detector). Adsorption experiments were also conducted using a packed column to study the breakthrough behaviour of the modified zeolites. Thus, 2 g modified zeolite was packed into a 5 mm i.d., 100 mm length glass column. The iodide concentrations employed were 0.010 M and 0.001 M, respectively, with an initial specific activity of ca. 40 000 Bq. The solution was passed through the column at a constant flow rate of 0.5 ml/min using a peristaltic pump. Depending on the initial concentration employed, the iodide concentration was measured at 2 10 min intervals. RESULTS AND DISCUSSION Our previous experiments performed on zeolite-exchanged materials revealed that the adsorption capacity of some of the exchanged forms of clinoptilolite were several times higher than that of the natural variety (Faghihian et al. 2002). An alternative method of modification involves treatment of the zeolite with surfactants. In this process, the extra-structural cations of the zeolite are replaced by quaternary alkylammonium ions. Provided sufficient surfactant concentrations are employed, adsorption occurs through admicelle formation on the zeolite surface. The corresponding adsorption isotherms on clinoptilolite for hexadecyltrimethylammonium, tetradecyltrimethyl-ammonium and dodecyltrimethylammonium ions are depicted in Figure 1. The maximum extents of surfactant adsorption were 0.33, 0.34 and 0.32 mequiv/g for hexadecyltrimethylammonium, tetradecyltrimethylammonium and dodecyltrimethylammonium ions, respectively. These values are almost twice as large as the external cation-exchange capacity of clinoptilolite (0.17 mequiv/g), confirming double-layer formation of the surfactant on the zeolite surface, as suggested by Bowman et al. (2000). The adsorption capacities of the natural and different modified samples when the initial iodide concentration was 0.010 M are presented in Table 1. For the Cp-HD50 sample, where clinoptilolite treatment was performed using a 50 mm surfactant solution, iodide adsorption was 50-times higher than that on the untreated sample. No significant increase in iodide adsorption was Figure 1. Adsorption isotherms for HD, TD and DD onto clinoptilolite.

376 H. Faghihian et al./adsorption Science & Technology Vol. 21 No. 4 2003 TABLE 1. Uptake of Iodide from a 0.01 M Iodide Solution Zeolite type Iodide uptake Zeolite type Iodide uptake (mequiv/g) (mequiv/g) Natural Cp 0.002 Cp-TD50 0.110 Cp-HD0.45 0.003 Cp-DD50 0.108 Cp-HD17 0.007 Cp-TBA50 0.003 Cp-HD25 0.098 Cp-TMA50 0.002 Cp-HD50 0.110 Natural Nt 0.001 Cp-HD67 0.113 Nt-HD50 0.003 Cp-HD100 0.116 Nt-HD100 0.003 observed when the initial surfactant concentration exceeded 50 mmol/l because of the formation of a surfactant double layer. Samples Cp-TD50 and Cp-DD50 showed almost the same adsorption capacity, but modification with tetrabutylammonium and tetramethylammonium ions did not alter the adsorption capacity significantly. It has been reported that the length of their alkyl chains influences the surface properties of surfactants. Thus, the short alkyl chains in tetrabutylammonium and tetramethylammonium ions would be incapable of forming an admicelle on the zeolite surface (Patist et al. 1997). The modified sample derived from natrolite exhibited negligible adsorption (0.003 mequiv/g) because of the very low surfactant uptake (0.01 mmol/g). Adsorption experiments were conducted via batch and column experiments. The distribution coefficients from the column experiments were then calculated using the following equation: K d = [(A i A f )/A f ] V/m where A i and A f are the initial and final activities of the solution (cps/ml), V is the volume of the initial solution (ml) and m is the mass of zeolite employed (g). A plot of K d versus the initial 129 I concentration for Cp-HD50 as an adsorbent is plotted in Figure 2. As expected, the highest K d value was obtained for dilute solutions. Obviously, the modified sample exhibited good sorption properties towards 129 I derived from radioactive liquid wastes of low iodide concentration. The breakthrough curves obtained via column experiments are depicted in Figure 3. These experiments were performed using initial iodide concentrations of 0.010 M and 0.001 M, respectively. The breakthrough points for the Cp-HD50 and Cp-HD100 samples appeared when 50 ml Figure 2. Plot of K d versus initial iodide concentration for adsorption onto Cp-HD50.

Removal of Radioactive Iodide by Surfactant-modified Zeolites 377 Figure 3. Breakthrough curves for an initial iodide concentration of (a) 0.001 M and (b) 0.01 M. of the 0.001 M effluent at a constant flow rate of 0.5 ml/min had passed through the column containing 2 g modified zeolite, i.e. ca. 30% of the batch adsorption capacity was achieved. The uptake for Cp-HD17 was negligible due to the low surfactant adsorption capacity of the sample. A breakthrough point of 10 ml was achieved for 0.01 M iodide solutions with Cp-HD50 and Cp-HD100 samples, i.e. 50% of the batch adsorption capacity was achieved. Iodide adsorption for Cp-HD100, Cp-TD50, Cp-HD50 and Cp-DD50 samples was measured by the batch method and the related isotherms are depicted in Figure 4. The removal of iodide by the modified zeolites was analyzed employing the Langmuir adsorption model using the following rearranged Langmuir equation to calculate the Langmuir constant: C e /q e = 1/Qƒb + C e /Qƒ A plot of C e versus C e /q e at different temperatures gave a straight line showing the applicability of the Langmuir isotherm (Figure 5). The values of Qƒ and b are included in Table 2. Iodide adsorption on some modified samples as a function of ph at an initial iodide concentration of 0.010 M

378 H. Faghihian et al./adsorption Science & Technology Vol. 21 No. 4 2003 Figure 4. Adsorption isotherms for iodide uptake by some modified clinoptilolite samples. Figure 5. Plots of C e /q e versus C e for various modified clinoptilolite samples. is depicted in Figure 6. The maximum uptake was observed in moderately acidic solutions at a ph value of ca. 4.0. At higher ph values, the presence of OH ions led to a significant decrease in iodide adsorption. In the lower ph region, the uptake of iodide was lower because of partial association of the iodide to form HI and dealumination of the zeolite. Adsorption of iodide was also studied in the presence of some cationic and anionic species, especially those present in nuclear wastes (Figure 7). Since the selectivity and capacity of surfactant-modified clinoptilolite towards the chromate and molybdate ion is high (Faghihian et al. 2002; Li et al. 1998), the presence of these anions in 0.01 M solutions of iodide significantly decreased the iodide uptake. The effect of the other anions studied was negligible.

Removal of Radioactive Iodide by Surfactant-modified Zeolites 379 TABLE 2. Langmuir Data for the Adsorption of Iodide by Some Modified Clinoptilolite Samples Zeolite form Q o (mmol/g) b (1/mmol) Cp-HD50 0.116 0.742 Cp-TD50 0.113 0.623 Cp-DD50 0.105 0.633 Cp-HD100 0.120 0.725 Figure 6. Effect of ph on iodide adsorption onto various modified clinoptilolite samples. Figure 7. Effect of ion interference on iodide uptake by various modified clinoptilolite samples.

380 H. Faghihian et al./adsorption Science & Technology Vol. 21 No. 4 2003 Figure 8. Desorption of iodide into various ionic solutions from modified zeolites. The desorption of iodide into different solutions was studied at a concentration of 0.10 M for the surrounding solution. The corresponding results are depicted in Figure 8 from which it may be concluded that the desorption sequence was NaOH > NH 4 OH > NaCl > HNO 3 > HCl. Minimal desorption was observed for the Cp-HD50 sample, whereas the Cp-HD100 sample showed maximum desorption. It may be concluded that the Cp-HD50 sample may be used for the removal and stabilization of radioactive iodide whereas Cp-HD100 can be considered as an alternative material for the separation of radioactive 131 I from nuclear wastes. CONCLUSIONS The following conclusions may be drawn from the results obtained: 1. Treatment of zeolites by surfactants significantly improved their iodide adsorption capacity through the formation of admicelles on the zeolite surface. The surfactant concentration must be at least 30-times the critical micellar concentration of the zeolite. 2. Modified clinoptilolite displays a good selectivity and capacity towards iodide. On the other hand, natrolite exhibited negligible uptake towards iodide due to its compact structure. 3. The studied samples may be divided into two groups: (a) modified samples with low desorption properties that can be used for adsorption and immobilization of radioiodide; and (b) modified samples with a high iodide uptake but high desorption values, which could be used for the separation and preconcentration of radioiodide. 4. It was concluded from column experiments that a greater number of adsorption sites were involved as the effluent concentration increased. 5. The adsorption capacity was much less than values obtained via precipitation experiments (Faghihian et al. 2002), but surfactant modification provided a fast, cheap and convenient method.

Removal of Radioactive Iodide by Surfactant-modified Zeolites 381 REFERENCES Bors, J., Gorny, A. and Dultz, S. (1997) Radiochim. Acta 78, 117. Bowman, R.S., Sullivan, E.J. and Li, Z. (2000) Natural Zeolites for the Third Millennium, Colella, C., Mumpton, F.A., Eds, De Frede, Naples, Italy, p. 287. Chmielewska-Horvathova, E. and Lesny, J. (1995) J. Radioanal. Nucl. Chem. 200, 351. Chmielewska-Horvathova, E. and Lesny, J. (1996) J. Radioanal. Nucl. Chem. 214, 209. Deitz, V.R. (1991) Chem. Phys. Carbon 23, 233. Eersels, J.L.H., Custers, F.G.L., Ensing, G.J. and Herschid, J.D.M. (1995) Eur. J. Nucl. Med. 22, 1185. Faghihian, H., Maragheh, M.G. and Malekpour, A. (2002) J. Radioanal. Nucl. Chem. 254, 545. Fetter, G., Olguin, M.T., Bosch, P., Lara, V.H. and Bulbulian, S. (1999) J. Radioanal. Nucl. Chem. 241, 595. Fukui, M., Fujikawa, Y. and Satta, N. (1996) J. Environ. Radioact. 31, 199. Inoue, H. and Kagoshima, M. (2000) Appl. Radiat. Isotop. 52, 1407. Kodama, H. (1992) Bull. Chem. Soc. Jpn. 65, 3011. Kodama, H. (1999) Czech. J. Phys. 49, 971. Li, Z., Anghel, I. and Bowman, R.S. (1998) J. Disp. Sci. Technol. 19, 843. Patist, A., Chhabra, R., Pagidipati, R., Shah, R. and Shah, D.O. (1997) Langmuir 13, 432. Rehakova, M., Casciola Massinelli, L., Anderson, I.G. and Bastl, Z. (1998) J. Incl. Phenom. Mol. Recog. Chem. 31, 131. Song, K.C., Lee, H.K., Moon, H. and Lee, K.J. (1997) Sep. Purif. Technol. 12, 215. Yang, O.B., Kim, J.C., Lee, J.S. and Kim, Y.G. (1993) Ind. Eng. Chem., Res. 32, 1692.