CHARACTERIZATION OF THE SOLIDS WASTE IN THE HANFORD WASTE TANKS USING A COMBINATION OF XRD, SEM AND PLM *

Transcription

1 Copyright JCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume CHARACTERIZATION OF THE SOLIDS WASTE IN THE HANFORD WASTE TANKS USING A COMBINATION OF XRD, SEM AND PLM * ABSTRACT R. W. Warrant and G. A. Cooke Fluor Hanford, Technology Project Management Richland, WA * Work Performed for the US DOE Under Contract DE-AC06-96RL13200 Document HNF VA The Department of Energy s River Protection Project (RPP) is tasked with retrieving highly radioactive waste from Hanford double-shell and single-shell tanks to provide feed for vitrification for long-term storage. Approximately 330,000 metric tons of sodium-rich radioactive waste originating from separation of plutonium from irradiated uranium fuel is stored in 177 underground tanks at Hanford. Current plans call for much of this waste to be vitrified and disposed of at the Yucca Mountain waste repository. In order to do this, the contents of the tanks need to be physically and chemically characterized. INTRODUCTION The Hanford waste tanks contain various amounts of liquid and solid wastes and over twenty different waste types have been identified for the various tanks. The RPP organization needs chemical and physical data to evaluate technologies for retrieving the waste. Accurate chemical modeling of the solids in the tank waste system depends to a large extent on a proper identification of the solid phases in equilibrium with the ions in solution. Analyses of solids from Hanford Tank BY-109 were performed using the three-pronged approach of x-ray diffraction (XRD), polarized light microscopy (PLM), and scanning electron microscopy with energy dispersive x-ray spectroscopy (SEM/EDS). The tank waste at Hanford comes in three basic forms. Approximately 33% of the waste volume is liquid, typically a caustic brine containing sodium, nitrate, nitrite, hydroxide, fluoride, phosphate, and sulfate. About 42% of the waste volume is saltcake, dominantly consisting of salts precipitated out of the brines during evaporation and storage. The remaining 25% of the waste is sludge, composed of insoluble materials. In addition to standard methods of chemical and physical analysis, the characterization effort includes the determination of the chemical compounds present in the wastes. These chemical compounds provide important information about how the waste will behave during dissolution, transport, mixing and glassification activities. Knowledge of the compounds present in the waste can also aid in the effective separation of waste into high level radioactive and low level (or even clean ) waste, thus reducing the overall cost of disposal.

2 This document was presented at the Denver X-ray Conference (DXC) on Applications of X-ray Analysis. Sponsored by the International Centre for Diffraction Data (ICDD). This document is provided by ICDD in cooperation with the authors and presenters of the DXC for the express purpose of educating the scientific community. All copyrights for the document are retained by ICDD. Usage is restricted for the purposes of education and scientific research. DXC Website ICDD Website -

3 Copyright JCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume EQUIPMENT AND METHODS The characterization of Hanford tank waste at the 222S laboratory is primarily conducted using X-Ray Diffraction (XRD), Scanning Electron Microscopy (SEM) and Polarized Light Microscopy (PLM). XRD data were collected using a Rigaku 1 MiniFlex powder diffractometer (Figure 1) equipped with a 8., ,3/.439 3:4:8 ;,7, % 0 -ray tube generates copper X-rays at 30 kv and 15 ma. PLM analysis was performed on a Leitz Laborlux 2 12 POL polarized light microscope equipped with a Colorview 12 digital color camera manufactured by Soft Imaging Systems. Figure 1. Rigaku MiniFlex X-ray Diffractometer located in radiological fume hood. SEM analysis was conducted using an Aspex 3 Personal Scanning Electron Microscope, Model II, with a Noran light element energy dispersive spectrometer (EDS) for chemical analysis. The instrument was operated at an accelerating voltage of 15 kv, and the samples were mounted at a working distance of 16 mm (+/-0.5 mm). EDS spectra were acquired for 60 seconds (live time) on relatively smooth surfaces (either flat or sloping toward the detector) near the center of the image. Samples for analysis typically come directly from tank grab or core sampling, from residues left over after selective dissolution, or from precipitates generated by evaporation or mixing of liquid samples. These are prepared for PLM analysis by dispersing the particulate in a liquid that will not cause additional dissolution or precipitation. XRD sample preparation typically involved grinding the material with a small agate mortar and pestle, then filtered. The material was then fixed with collodion onto a glass slide. SEM preparation involved filtration followed by directly mounting the filter on a SEM stub or transferring the particulate from the filter to an adhesive carbon tape. The filtration stand used for XRD and SEM sample preparation is shown in Figure 2. Figure 2. Filter apparatus for sample preparation located in radiological fume hood. 1 Rigaku MiniFlex is a registered trademark of Rigaku/MSC, The Woodlands, Texas. 2 Leitz Laborlux is a registered trademark of Ernst Leitz Wetzlar GmbH 3 Aspex is a registered trademark of ASPEX Instruments, Delmont, Pennsylvania.

4 Copyright JCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume RESULTS To date, we have synthesized and examined the simple salts: NaNO 3, NaNO 2, Na 2 SO 4 (anhydrous), Na 2 CO 3. H 2 O, NaF, Na 2 C 2 O 4 and Na 3 PO 4. 12H 2 O, the double salts: Na 7 F(PO 4 ) 2. 19H 2 O, Na 3 FSO 4, Na 3 NO 3 SO 4. H 2 O, Na 6 CO 3 (SO 4 ) 2 and the aluminous compounds: NaAl(OH) 4, Al(O)OH (Boehmite) and Al(OH) 3 (Gibbsite). These standard materials were examined because they have all been found in tank waste materials. Each of these standards was examined by PLM, SEM, and XRD, (Figure 3), and the resulting images and spectra were compiled in an image database. Saltcake waste has been the major focus of waste characterization studies to date. Saltcake from tank 241BY109 is the most thoroughly studied tank to date. The three methods, in concert, identified Na 7 F(PO 4 ) 2. 19H 2 O, Na 3 FSO 4, Na 2 C 2 O 4, NaF, Na 3 AlF 6, and Aluminum Hydroxide. One of these phases, the Na 3 AlF 6 (Cryolite) Figure 3. Each standard and sample were analyzed by XRD, was only identified during the SEM/EDS, and PLM. evaporation of solutions generated during saltcake dissolution. Figure 4 is the XRD pattern showing the original BY109 saltcake with the four phases identified below. Figure 4. Diffractogram of BY-109 original sample containing: Na7F(PO 4 ) 2. 19H 2 O, Na 3 FSO 4, NaF and Na 2 C2O 4.

5 Copyright JCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume The SEM revealed the presence of Alumino-silicates and heavy element phases, including U, Cr, Bi/Pd and Ca/Sr/Cr- rich phases. Closer examination of some of these phases in the insoluble residue of Tank 241S112 yielded XRD data that identified one uranium-bearing phase (shown in figure 5) as sodium diuranate (Na 2 U 2 O 7 ). A calcium/chromium bearing phase that seems to Figure 5. SEM image and EDS spectra of particle in the BY-109 residue sample which is thought to contain Na 2 U 2 O 7. allow significant substitution of strontium for calcium was also found by SEM analysis in both tanks (Figures 6 and 7). This phase has been tentatively identified through an XRD slow Figure 6. SEM image and EDS spectra of particle in the BY-109 residue sample thought to be a hydrouvarovite, (Ca, Sr) 3 (Cr, Al) 2 (OH) 12 scan spectrum as a garnetoid phase (Figure 8). The chemical composition of this phase is consistent with hydrouvarovite (Ca, Sr) 3 (Cr,Al) 2 (OH) 12. This phase has not been previously described in the open literature.

6 Copyright JCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume Figure 7. Solid from S-112 sample similar to unidentified phase found in BY-109 sample thought to be hydrouvarovite, (Ca, Sr) 3 (Cr,Al) 2 (OH) 12. Figure 8. S-112 sample with phases identified as Al(OH) 3, AlO(OH), Na 2 U 2 O 7, Ca 3 Al 2 (OH) 12 or Ca 3 Al 1.54 Fe 0.46 [(OH) 4 ] 3. The phase identified as Katoite is a hydrousgarnet. In this sample, the chemical data is consistent with the hydrouvarovite phase (Ca, Sr) 3 (Cr, Al) 2 (OH) 12.

7 Copyright JCPDS - International Centre for Diffraction Data 2003, Advances in X-ray Analysis, Volume CONCLUSION High fluoride content in BY-109 leads to presence of sodium fluoride as well as fluoride phosphate and fluoride sulfate double salts. The identification of the double salts such as Na 3 FSO 4 and Na 7 F(PO 4 ) 2. 19H 2 O has proven to be very important in predicting the solubilities during dilution and retrieval of tank wastes. Heavy element phases, observed by PLM, are characterized by SEM/EDS, and are identified by XRD. A proposed new Hydrogarnet phase, hydrouvarovite, with an approximate composition of (Ca, Sr) 3 (Cr,Al) 2 (OH) 12 has been identified in tank residues. This phase appears to be responsible for the low solubility of strontium. ACKNOWLEDGMENTS Dr. Dan Herting pioneered the phase analysis of Hanford tank waste at the 222S laboratory. He has been principal investigator of a number of studies using differential dissolution and PLM analysis, primarily in the characterization of saltcake. Dr. William S. Callaway, D. Wayne Edmonson, John R. Smith, and Kristi Brault have all made contributions to the analytical efforts.