Environ. Sci. Technol. 2001, 35, 547-551
Approaching Zero Discharge in Uranium Reprocessing: Photochemical Reduction of Uranyl T. MARK MCCLESKEY,* TRUDI M. FOREMAN, ERIN E. HALLMAN, CAROL J. BURNS, AND NANCY N. SAUER Chemical Science and Technology-18, MS J514, Los Alamos National Laboratory, Los Alamos, New Mexico 87545
We have studied the photochemical reduction of uranyl to generate UO2 without hydrogen reduction. Formate and oxalate were examined as potential reductants that only lead to CO2 production as a side product. Despite the similar nature of the two reductants, the mechanism for quenching the uranyl excited-state changes drastically and leads to dramatically different chemistry at low pH. Oxalate quenches by unimolecular electron transfer and formate quenches by H-atom abstraction. Because of the change in mechanism, photochemical reduction of uranyl with formate works with high efficiency at low pH while photolysis in the presence of oxalate leads to the generation of CO and no net uranyl reduction. Photochemical reduction of uranyl with formate at low pH leads to U(IV) in solution that can then be precipitated as UO2 by simply raising the pH with yields as high as 99.992%.
Introduction The development of novel waste minimization technologies is being driven in both the chemical and nuclear industries by increasingly stringent environmental regulations, escalating costs for waste disposal, and growing public concern over lifecycle management of hazardous materials. Within both these industries, major institutions have established zero discharge goals as a portion of their environmental stewardship programs (1). Clearly, enhancing efficiencies of current processes will not meet these goals. Dramatic changes in traditional processing methods will be required. The need for new methodologies is exemplified by the fact that uranium recovery schemes have not been enhanced since the late 1960s. The steps in current uranium reprocessing include the following: dissolution of uranium materials in nitric acid, purification by solvent extraction, recovery of uranyl nitrate, denitration to give UO3 with release of NOx, and subsequent reduction to UO2 under a hydrogen atmosphere at 500 °C. The UO2 is converted to UF4 by reaction with HF and reduced to metal with calcium. Direct low-temperature reduction of uranyl to give UO2 as a precipitated product represents an attractive alternative for minimizing waste, eliminating NOx generation, and avoiding the potentially hazardous hydrogen reduction step. Most of the recent literature on the reduction of uranyl has focused on the removal of uranyl from water or soil at contaminated sites (2-5). Citrate (2, 3), TiO2 particles (4), and zero-valent iron (5) have been examined as potential reductants for uranyl in environmental conditions. It has also been demonstrated that uranyl can be reduced electrochemically to give U4+ in solution (6-9), but the precipitate * Corresponding author e-mail:
[email protected]; phone: (505)667-5636; fax: (505)667-9905. 10.1021/es001078i CCC: $20.00 Published on Web 12/30/2000
2001 American Chemical Society
formed on the electrode surface yielded only the partially reduced U3O8 as opposed to UO2 (9). We report here on a photochemical, room temperature synthesis of UO2 directly from a solution of uranyl nitrate at low pH. When uranium materials for recovery are dissolved in nitric acid, the uranyl ion (UO22+) is the dominant species in solution. Uranyl has the unique characteristic of exhibiting luminescence at room temperature in aqueous solution even in the presence of oxygen. Uranyl absorbs light at 415 nm as an electron moves from an oxygen-based orbital to a more energetic uranium-based orbital. This excited state then decays by emitting green light centered at 510 nm. The excited state of uranyl, UO22+*, is known to be a potent oxidant (E° ) 2.6 eV) with a lifetime of 6 µs (10). The long lifetime allows bimolecular oxygen transfer reactions to occur with a variety of organic substrates (11-16). The efficiencies of the uranyl photoreactions can be extremely high with quantum yields as large as 0.6 in the case of the uranyl oxalate system at low pH, which has been studied in depth for use as an actinometer (17-19). We have studied the photoreduction of uranyl from U(VI) to U(IV) in the presence of oxalate from pH 2 to pH 7 and in the presence of formate from pH 1.5 to pH 4. Oxalate and formate were chosen as the organic reductants because they are water-soluble, two-electron reductants that can yield CO2 and H2O as the only side products under proper conditions. Previous studies on the photochemistry of uranyl in the presence of oxalate and formate have examined the gaseous products of CO and CO2 in detail. We have focused on the reduction of uranium to U(IV) and the subsequent conversion to UO2. The challenges in this area are to develop a system that can work in the presence of nitrate and preferably at low pH (when U(IV) is soluble) to avoid the complications of generating a precipitate during photolysis. Dramatic differences in photolysis at low pH with formate and oxalate prompted us to investigate the mechanism of quenching of the uranyl excited state. The results presented here demonstrate that under the correct conditions photolysis has great potential as a means to generate UO2 in a uranium reprocessing cycle and the quenching mechanism is critical for generating U(IV) at low pH.
Experimental Section Reagents. Formic acid, sodium formate, sodium D-formate, 50% sodium hydroxide in water, oxalic acid, phosphoric acid, and nitric acid were ACS reagent grade and used as received (Aldrich). Uranyl nitrate was used as received from Strem. Water was deionized by a Millipore filtration system (>18 Ω conductivity). Absorption Spectra. Absorption spectra of aqueous solutions with 5 mM uranyl nitrate were measured from pH 3 to pH 5.5, in the presence of 2 equiv of oxalate from pH 3 to pH 9 and in the presence of 2 equiv of formate from pH 3 to pH 5 using a Perkin-Elmer Lambda-19 spectrometer. The pH was adjusted using sodium hydroxide and nitric acid. The absorption spectra of uranyl nitrate in the presence of formate at pH 4 and at pH 5 were examined to determine if a uranyl formate complex was evident from the absorption spectra. For these experiments, solutions were prepared from uranyl nitrate (with hydrated uranyl as the dominant species in solution (2)) and sodium formate and adjusted to either pH 4 or pH 5. Aliquots of the concentrated formate solution were then added to the uranyl solution to introduce equivalents of formate without altering the pH. Absorption spectra were also used to monitor the low pH photolysis experiments. For these reactions, the absorption spectra were VOL. 35, NO. 3, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
547
FIGURE 1. UV/Vis spectra of uranyl in the presence of 2 equiv of oxalate (A) at various pH values. Close-up of the oxo-metal chargetransfer region at pH 3.5. taken on an HP 8453A diode array spectrometer. A spectrum of U(IV) was prepared by dissolution of UCl4 in water that had been purged with argon to avoid any oxidation to uranyl. Photolysis Experiments. Photolysis experiments were done using either a quartz or a glass, water-cooled, immersion well with a 450-W mercury lamp. The entire apparatus was contained within a box to prevent UV exposure. A nitrogen purge was used during photolysis to avoid unwanted oxidation by dissolved oxygen. For experiments at low pH, when no precipitate was observed, aliquots of the reaction mixture were taken at time intervals and monitored by UV/ Vis with an HP 8453A diode array spectrometer. The pH was also monitored during the course of the reaction and adjusted with nitric acid so that it never increased more than one pH unit. After photolysis reached completion, the solution pH was adjusted to pH 9 with 50% sodium hydroxide and stirred for 10 min with a slow nitrogen purge into the solution to prevent any oxidation of the UO2 from air. The resulting solution was then poured into bottles and centrifuged, and the supernate decanted. The solid was then dried in a vacuum oven overnight. The final supernate was analyzed for uranium by ICP-AES using a Varian Liberty 200 instrument. In the studies with oxalate at near neutral pH, a precipitate formed in situ, and aliquots of the supernate were taken and analyzed for uranium concentration by ICP-AES. After photolysis, the workup to obtain the final product precipitate was carried out in the same manner as described for the low pH experiments including an adjustment of the pH to 9 with 50% sodium hydroxide. The final supernate was analyzed for uranium by ICP-AES. A 300-W quartz tungsten halogen lamp was used in one photolysis experiment to demonstrate the ability to use different types of lamps. Photodegradation Analysis. Powder XRD measurements of the solid black precipitates formed during the photolysis experiments were performed on an INEL XRD spectrometer and compared to a known sample of high-fired UO2 measured on the same spectrometer. Samples were also analyzed by FT-IR using a Nicolet Avatar. Elemental analysis for C, H, 548
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 3, 2001
FIGURE 2. Overlaid spectra of uranyl and uranyl in the presence of 2 equiv of formate at pH 4.0 (A) and pH 5.0 (B). and N was done using a Perkin-Elmer 2400 elemental analyzer. Quenching Measurements. Excited-state lifetimes of uranyl were determined using a Quanta Ray DCR-3 Nd:YAG laser with a R-955 PMT detector and a HP 54111D digitizing oscilloscope. These studies were done at low pH to avoid the complication of precipitate formation during the experiment. The goal was to identify the initial mechanism of quenching of the uranyl species in the presence of formate and oxalate at low pH. Quenching studies were done in a quartz cuvette fitted with a septum. Experiments at pH 2.2 were done in aqueous solution with 0.5 M H3PO4 and 5 mM uranyl nitrate. Under the reaction conditions, the dominant uranium species in solution is a uranyl bisphosphate complex (20) with a lifetime in solution of 50 µs. The use of phosphate has advantages in terms of signal intensity and greater insensitivity to small fluxuations in the pH. Solutions were deaerated by a nitrogen sparge for 5 min, and the cuvette was sealed with a septum. Sodium formate and sodium D-formate were added from a stock solution of 1.5 M formate titrated to pH 2.2 with phosphoric acid. Aliquots of the formate solution were added via a syringe, the resulting solution was stirred with a magnetic stirrer for at least 1 min, and then a lifetime measurement was taken. This process was continued until the lifetime was 0.990. For quenching with oxalic acid, oxalic acid was added in aliquots as a solid by removing the septum and adding the solid under a flow of nitrogen. In the case of oxalic acid, no change in the lifetime (50 µs) was observed; hence, a Stern-Volmer analysis was not done.
FIGURE 3. Photochemical reduction with oxalate at pH 4.5. % uranium in solution as a function of time for initial concentrations of 0.100 and 0.010 M uranium.
Results Absorption Spectra. The oxo to metal absorption band for uranyl is centered at 420 nm and gains intensity as the pH increases. At pH >4, low energy bands begin to appear as the uranyl hydroxide dimer [UO2(OH)]22+ begins to form (20). When oxalate is added to a uranyl solution, the UV/Vis absorption spectra indicate that the oxalate binds strongly to uranyl in agreement with previous studies (13, 17-19, 22). The uranyl bisoxalate complex has a red-shifted absorption with a smaller energy vibrational coupling (new fine structure appears with a spacing of 550 cm-1) as compared to uranyl. The uranyl bisoxalate complex is very stable and highly soluble up to pH 9. The UV/Vis spectra remain unchanged between pH 2 and pH 7, but low energy bands begin to appear at higher pH. The spectra of uranyl remains unchanged in the presence of 2 equiv of formate at pH 4 and pH 5 (Figure 2). We chose to examine pH values greater than the pKa of formic acid (3.75) to be sure that free formate would be available for binding. Peaks at 475 and 490 nm previously attributed to a uranyl formate may have been due to the formation of the uranyl hydroxide dimer if the pH was allowed to increase by adding formate directly as opposed to adding formate from a stock solution adjusted to the same pH as the uranyl solution. Photolysis Experiments. To avoid the complication of speciation changes in the oxalate system at pH >7.5, we performed the initial photolysis with oxalate at pH 7.0. Initial experiments were run in a quartz reactor. Results of photolysis with uranyl oxalate and uranyl nitrate as the uranyl source at pH 7 with a total of 4 equiv of oxalate present demonstrated that nitrate inhibited the formation of the photoprecipitate. Nitrate in solution has an absorbance at 300 nm and can be photoactivated to generate highly oxidizing NOx species. The use of a glass reactor, which effectively cuts out light
