Highly Selective Inorganic Crystalline Ion Exchange Material for Sr

to the presence of isotopes of cesium and strontium. Similar wastes are stored at DOE facilities at Savannah River, Oak. Ridge, and Idaho. The waste i...
0 downloads 0 Views 275KB Size
Environ. Sci. Technol. 1996, 30, 3630-3633

Highly Selective Inorganic Crystalline Ion Exchange Material for Sr2+ in Acidic Solutions TINA M. NENOFF,* JAMES E. MILLER, STEVEN G. THOMA, AND DANIEL E. TRUDELL Advanced Energy Technology Center, Sandia National Laboratories, P.O. Box 5800, MS 0709, Albuquerque, New Mexico 87185-0709

Introduction We report a novel antimony titanate ion exchange material, stable in highly acidic conditions and selective to strontium against competing cations, with possible applications at Defense Waste Sites. The Hanford Reservation in Washington State contains 177 underground storage tanks that are currently being used to store approximately 65 Mgal of radioactive waste containing more than 165 million Ci of radioactivity. Much of the radioactivity in the waste is due to the presence of isotopes of cesium and strontium. Similar wastes are stored at DOE facilities at Savannah River, Oak Ridge, and Idaho. The waste in the Hanford tanks includes a mixture of sludge, salt cake, and alkaline supernatant liquids. A major planning assumption for the tank waste remediation system (TWRS) emphasized the development of enhanced sludge washing (washing and/or selective leaching of sludges) to render the sludge suitable for lowlevel waste disposal and thereby reduce the volume of waste that will require high-level waste disposal (1). One treatment scheme calls for the dissolution of washed sludge with nitric acid followed by 90Sr recovery from the acid waste. Presently, one of the best options for achieving this separation is extraction with crown ethers (2). Thus, there is a need for adsorbents that are selective for strontium yet are stable and robust in highly acidic solutions. Many different types of inorganic sorbent materials have been studied as ion exchangers for strontium cations, including natural and synthetic zeolites, zirconium phosphates, and layered sodium titanates (3-7). At lower pH, strontium exists in the unassociated 2+ oxidation state. As pH increases, Sr(OH)+ becomes more prevalent and is the sole cation at pH > 12.8 (7). The need to synthesize an ion exchange material capable of selectively sorbing a divalent cation in the substantial presence of competing monovalent and divalent cations is enormous. The lack of suitable adsorbents for removing strontium from acidic wastes has been demonstrated by Marsh et al. (8). In their study encompassing 63 adsorbents, no material exhibited a distribution coefficient for Sr in a solution simulating aciddissolved Hanford sludge (pH 0.58) of over 3.1 mL/g. In a solution simulating Hanford acidified supernate (pH 3.5), the largest measured distribution coefficient for Sr was only * Author to whom correspondences should be addressed. Telephone: (505)844-0340; fax: (505)845-9500; e-mail: tmnenof@ sandia.gov.

3630

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 12, 1996

18 mL/g. Recently, promising results were reported for the separation of Sr from an acidic Idaho National Engineering Laboratory (INEL) waste using a ligand particle web technology (9). However, applications of this technology may be limited by the inherent instability of organicbased materials at high temperatures, elevated radiation exposures, and over multiple use cycles (i.e., regeneration of material). We report herein an inorganic doped titanate that provides a complement for acid side strontium processing to the class of titanates (CSTs) developed jointly at Sandia and Texas A&M University (10, 11). Its development was based on good selectivity for Cs and Sr by the CSTs and literature information on the ion exchange properties of antimony compounds (10, 12). This new material has been tested for the selective removal of parts per million level concentrations of Sr2+ ions from solutions with a pH in the range of 1 M HNO3 to 5.7 M Na+/0.6 M OH- (with the most important results in the highly acidic regimes). This doped titanate has been characterized with an array of techniques, including equilibrium distribution coefficient (Kd) determinations over a wide pH range, powder X-ray diffraction, TEM, BET, direct-current plasma (DCP), and thermal analyses.

Experimental Section Synthesis. This material is prepared hydrothermally using titanium isopropoxide (TIPT), sodium hydroxide, tetramethylammonium hydroxide (TMAOH), dopant oxides (including oxides of V, Nb, Bi, and Zr), and water as reactants in basic pH (≈12), according to the following stoichiometry: 1.0 metal oxide:2.84 Sb2O3:21 TIPT:1.42 NaOH:21.9 TMAOH:257 H2O. Two mixtures are made at room temperature; the hydroxides (TMAOH and NaOH) are mixed in one vessel, and all other reactants are mixed in another vessel. After 30 min, the hydroxide mixture is added to the other reactants mixture and stirred for 30 min. This final solution is reacted in a Parr reaction vessel at 150 °C for 3 days. The product, a fine white powder, is recovered by vacuum filtration, washed with deionized water, and airdried at ambient temperature. Characterization. Powder X-ray diffraction data were collected at room temperature on a Siemens Model D500 automated diffractometer, with Θ-2Θ sample geometry and Cu KR radiation, between 2Θ ) 5 and 60°, step size 0.05°. TEM and EDS were performed on a Philips CM30 transmission electron microscope at 300 kV. BET measurements were performed on a Quantachrome Autosorb 6B automated gas sorption system, with adsorbed and desorbed volumes of nitrogen at relative pressures in the range of 0.05-1.0. TGA measurements were collected on a DuPont 9900 system in air. The heating rate was 10 °C/ min from room temperature to 500 °C. DCP measurements were collected on a ARL Fisons SS-7 DCP. Samples were dissolved in a 2% aqua regia solution for analysis. Ion Exchange. In order to quantitatively compare the various doped sodium titanate analogs against each other and against various sorbent materials, a standard procedure was chosen for measuring batch distribution coefficients, Kd. All distribution coefficient measurements (batch Kds) reported here were obtained by contacting, with agitation,

S0013-936X(96)00533-0 CCC: $12.00

 1996 American Chemical Society

FIGURE 1. TEM (1600000×) of the mixture of titanate phases.

0.1 g of material with 10 mL of solution for approximately 24 h. At the end of the contact period, the solution was passed through a 0.2-µm Whatman syringe filter. A sample of the solution was then diluted and analyzed by DCP. Distribution coefficients (Kd) were calculated utilizing the measured mass of titanate material and the initial and final concentrations of Sr in the solution. The values are reported for as-prepared materials; no attempt to account for the volatiles content (loss-on-ignition or LOI) of the material was made. Kd was calculated as follows: Kd (mL/g) ) V(Co - C)/(wC), where V is the volume of the solution used (mL); Co is the initial concentration of the Sr (ppm); C is the concentration after equilibrium has been reached (ppm); and w is the mass of the sorbent used (g).

Results and Discussion Description of Material. The synthesis product is a dense, thermally stable (as indicated by TGA), microporous titanate material with low solubility in mild acids through basic pH and higher solubility in concentrated mineral acids. Density was measured by water displacement and determined to be approximately 1.5 g/mL. The multi-point BET surface area for this material is 28.4 m2/g. The material appears to have an “ink-bottle” pore structure, as indicated by the shape of the closed hysteresis loop. TGA data indicate an initial water weight loss at ≈100 °C, with a gradual weight

loss in the material until ≈180 °C, after which point no weight loss is observed. Diffraction data indicate that this is a very fine particle material with a primary titanate phase and a minor sodium titanate impurity phase. Powder X-ray diffraction data (see Table 1) shows that there is a layered phase present: however, the peaks are too broad to apply a crystallographic model for structure refinement. The broad peaks result from very small particle sizes. DCP data show the following elemental composition: 26% Ti, 0.1% V, 7.6% Na, and 20% Sb. TEM data is complementary to the X-ray data in discerning a two-phase system (see Figure 1). One phase is a dopant-titanate equiaxed phase whose particles are 10-30 nm. The EDS analysis of several of these particles shows strong Sb, Ti, and O peaks and a small Na peak. The other phase, whose particles are typically 5 nm × tens of nm across, is “flaked”. It has a layered structure that is easily imaged as strong curved fringes in high-resolution imaging mode. EDS analysis of several of these flakes shows strong Ti and O peaks and a smaller Na peak with essentially no Sb. This phase is likely a titanium oxide or sodium titanate phase. Hence, virtually all of the Sb is incorporated in the equiaxed phase. (Vanadium was not detectable by EDS due to the low concentration; it may be in either phase; see DCP data.)

VOL. 30, NO. 12, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3631

TABLE 1

Powder X-ray Diffraction Data of Mixed Titanate Phases d-spacing (Å)

relative intensity

7.3888 5.9634 3.1032 2.9807 2.5767 2.0783 1.8232 1.7999 1.7393 1.5527

100.0 53.4 56.2 99.6 36.4 37.9 46.2 31.4 25.7 40.9

FIGURE 2. Distribution coefficients, Kd, versus pH for the titanate material in various solutions.

TABLE 2

B-110 Simulant Composition (13) description

amount (M)

SiO2 HNO3 Al(NO3)3 Ba(NO3)2 Bi(NO3)3‚5H2O Ca(NO3)2‚4H2O Cr(NO3)3‚9H2O Fe(NO3)3‚9H2O La(NO3)3‚6H2O Mg(NO3)2‚6H2O NaNO3 NaH2PO4‚H2O Pb(NO3)2 Sr(NO3)2

0.029 2.3 0.013 3.3 × 10-5 0.024 5.3 × 10-3 4.0 × 10-3 0.092 0.001 1.8 × 10-3 0.19 0.072 1.3 × 10-3 7.9 × 10-4

Electron diffraction ring patterning of an aggregate of the two phases enables their distinction and identification. The d-spacings indicate that the strong intensity peak at 7.38 Å corresponds to the largest layer spacing in the titanate flakes (see Table 1). The medium intensity peak at 5.96 Å corresponds to the largest spacing measured in the equiaxed antimony titanate phase. Desorption studies indicate that the Sr2+ does not desorb from the titanate material in DI H2O. A sample was ion exchanged for 12 h in aqueous 10% Sr(NO3)2 (10 mL/g) at room temperature to give a loaded strontium titanate. This exchanged sample was then redispersed in DI H2O and stirred at room temperature for 12 h. DCP of the dissolved titanate materials indicated no noticeable change in strontium weight percent (≈10%). Adsorption of Sr in Batch Tests. This titanate phase has strong adsorption of Sr (as measured by Kd) over a broad range of pH. At neutral pH (DI H2O, 100 ppm initial Sr), Sr Kd > 99 000 mL/g. At basic pH (5.7 M Na+/0.6 M OH-, 50 ppm initial Sr), Sr Kd > 50 000 mL/g. At acidic pH (1 M HNO3 and 0.1 M HNO3, 50 ppm initial Sr) Kd ) 62 and 157 mL/g, respectively. By comparison, the sodium titanate product synthesized without dopant oxides consistantly exhibits low Sr Kds of approximately 20 mL/g in 1 M HNO3 (50 ppm initial Sr). To determine the effects of competing cations in acidic conditions, a batch Kd (run for 16 h) was measured in a Hanford B-110 dissolved sludge simulant (13) (see Table 2) and found to be 30 mL/g (Figure 2). Both calcium and magnesium are known to interfere with the selective adsorption of strontium from solution; however, results show that Kds remain high. In an acidic solution with competing cations of H+ and Ca2+ (0.05 M

3632

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 12, 1996

HNO3/45.6 ppm initial Sr/and 50 ppm initial Ca) Sr Kd ) 46 mL/g, while in an acidic solution with competing divalent cations (0.05 M HNO3/41.7 ppm initial Sr/and 43.5 ppm initial Ba, Ca, and Mg, respectively) Sr Kd ) 28 mL/g. In the latter experiment, preferential adsorption was as follows: Sr > Ca . Ba . Mg. The presence of Ca, Ba, and Mg in the B-110 simulant probably contributes substantially to the decrease in Kd value over that measured in 1 M HNO3. Ion Exchange Column Runs. The kinetic behavior of the titanate material was assessed using a laboratory-scale column. The 7 mm × 4 cm column was completely filled with +40 mesh titanate material. The feed consisted of 50 ppm Sr in 0.05 M HNO3 (resulting pH of 1.8). The experiment was carried out downflow at 10 column volumes/h flow rate and at 25°C. Effluent samples were analyzed by DCP. In HNO3, 10% breakthrough (C/Co ) 0.1) occurred at about 70 column volumes, and 50% breakthrough occurred at about 95 column volumes.

Conclusions The doped titanate material is a microporous phase that potentially offers significant advantages for use in strontium separations from wastes such as acid-washed sludge. In comparison to organic ion exchangers, titanates will be chemically, thermally, and radiation stable. The current doped titanate phase formulation exhibits high distribution coefficients for Sr in acidic solutions and in a preliminary ion exchange column study exhibited rapid kinetics. Considering the distribution coefficient, the column packing density, and the results from the column ion exchange experiment, we estimate that, in an ion exchange column test with the B-110 sludge simulant, 50% breakthrough would occur at between 10 and 20 column volumes of feed. Although this is competitive with the web technology (2, 9), we are attempting to optimize the performance of the material in the presence of competing cations by refining the synthesis and dopant chemistry. Alternately, competing cations could be used to regenerate the material for multi-cycle operations. Ongoing experiments include full characterization of the antimony titanate phase and ion exchange column studies with B-110.

Acknowledgments This work was performed at Sandia National Laboratories and was supported by the U.S. Department of Energy under Contract DE-AC04-94AL85000. We thank Tom Headley for the TEM work and evaluation.

Literature Cited (1) Colton, N. G. Westinghouse Hanford Company Report TWRSPP94-053. 1994. (2) Izatt, R. M.; Bradshaw, J. S.; Bruening, R. L.; Tarbet, B. J.; Krakowiak, K. E. In Emerging Separation Technologies for Metals and Fuels; Lakshmanan, V. I., Bautista, R. G., Somasundraran, P., Eds.; The Minerals, Metals & Materials Society: Warrendale, PA, 1993; pp 67-75. (3) Isirikyan, A. A.; Dubinin, M. M. In Occurrence, Properties and Utilization of Natural Zeolites; Kallo, D., Sherry, H. S., Eds.; Akademiai Kiado: Budapest, 1988; p 553. (4) Zhdanov, S. P.; Shubaeva, M. A.; Andreeva, N. R. Izv. Akad. Nauk SSSR, Ser. Khim. 1989, 10, 2208. (5) Komarneni, S.; Roy, R. Nature 1982, 299, 707. (6) Heinonen, O. J.; Lehto, J.; Miettinen, J. K. Radiochim. Acta 1981, 28, 93. (7) Lehto, J.; Clearfield, A. J. Radioanal. Nucl. Chem. Lett. 1987, 118 (1), 1.

(8) Marsh, S. F.; Svitra, Z. V.; Bowen, S. M. Los Alamos National Laboratory Report LA-12654. 1994. (9) Bray, L. A.; Brown, G. N. Pacific Northwest Laboratory Report PNL-10283. 1995. (10) Anthony, R. G.; Philip, C. V.; Dosch, R. G. Waste Manage. 1993, 13, 503. (11) Anthony, R. G.; Dosch, R. G.; Gu, D.; Philip, C. V. Ind. Eng. Chem. Res. 1994, 33, 2702. (12) Benvenutti, E. V.; Gushikem, Y.; Davanzo, C. U.; de Castro, S. C.; Torriani, I. L. J. Chem. Soc. Faraday Trans. 1992, 88 (21), 3193. (13) Lumetta, G. J.; Wagner, M. J.; Barrington, R. J.; Rapko, B. M.; Carlson, C. D. Pacific Northwest Laboratory Report PNL-9387. 1994.

Received for review June 19, 1996. Revised manuscript received August 26, 1996. Accepted September 6, 1996. ES9605331

VOL. 30, NO. 12, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3633