Environ. Sci. Technol. 2001, 35, 4905-4909
Use of Iron Oxide-Coated Sand To Remove Strontium from Simulated Hanford Tank Wastes BIRGITTE Ø. HANSEN, PIERRE KWAN, MARK M. BENJAMIN,* CHI-WANG LI, AND GREGORY V. KORSHIN Department of Civil and Environmental Engineering, Box 352700, University of Washington, Seattle, Washington 98195-2700
The ability of iron oxide-coated sand (IOCS) to adsorb strontium from synthetic wastes simulating the tank wastes at the Hanford Nuclear Reservation was examined in this study. These wastes have high pH and high ionic strength, containing up to 5.5 M Na+, 3.7 M NO3-, and 1.2 M OH-. The use of IOCS in such applications may be advantageous because it is inexpensive, is nontoxic, and can be prepared from readily available materials. IOCS can selectively remove strontium from solutions, even when they contain overwhelmingly higher concentrations of potentially competing cations such as Ca2+, Al3+, and Cr3+. Sr-EDTA chelates do not adsorb to IOCS. However, the interference caused by EDTA can be substantially overcome by the addition of excess Ca to the solution since Ca displaces Sr from EDTA. The adsorbed Sr can be released, and the IOCS can be regenerated by exposure to dilute acid (pH ∼3) for short periods. The physical and adsorptive properties of the IOCS remain essentially unaltered over at least several dozens of regeneration cycles, corresponding to treatment of at least 20 000 bed volumes of influent. The only byproduct of the regeneration process is a small volume of residual waste containing essentially only strontium and dilute acid.
Introduction From the 1940s to the 1990s, approximately 210 000 m3 (55 million gal) of radioactive byproducts from plutonium processing was deposited in 177 underground tanks on the Hanford Nuclear Reservation in Washington State (1, 2). The majority of the storage tanks have exceeded their design lives by severalfold, and many of them are suspected or known to be leaking. The composition of the waste varies among the tanks and is known only approximately. However, many of the solutions are highly caustic and contain high concentrations of sodium, aluminum, nitrate, and nitrite along with smaller concentrations of organic chelating agents and precipitated metal hydroxides, phosphates, carbonates, oxides, and sulfates (2). More than 99% of the 215 million Ci of radioactivity in the waste is generated by the decay of 137Cs and 90Sr, each of which is present at only trace concentrations. The plan for tank waste cleanup involves separating and ultimately vitrifying the radioactive portion of the waste. One * Corresponding author phone: (206)543-7645; fax: (206)685-9185; e-mail:
[email protected]. 10.1021/es0108990 CCC: $20.00 Published on Web 11/03/2001
2001 American Chemical Society
option for separating Sr from the rest of the waste constituents is adsorption onto some highly selective solid. As part of the search for an acceptable adsorbent and also because of interest in the fate of any Sr that might already have leaked from the tanks, the sorption of Sr onto a variety of solids and under a wide range of conditions has been studied. Several studies have shown that Sr2+ can sorb onto inorganic solids at near-neutral pH at moderate ionic strengths. As is the case for most other cations, sorption of uncomplexed Sr onto oxides and clays typically increases with pH (3-6). However, high ionic strength and competition from other multivalent ions for surface sites can reduce Sr adsorption dramatically. For instance, Chen and Hayes (3) reported that fractional Sr sorption onto montmorillonite and illite at pH 7 decreased from >80% to 100 µg/L for approximately the next 5000 BV treated, at 0 to 30 µg/L for the 5000 BV after that, and at
