Why do mechanisms matter in radioactive waste management? Or: Know your enemy Francis Livens Centre for Radiochemistry Research University of Manchester With thanks to: Rick Parkman, Rebeca Alvarez, Laura Butchins, Emmanuelle Chardon, Jo Renshaw, Helen Steele, Mike Wharton, Kath Morris, Jon Lloyd, David Vaughan, Steve Parry, Richard Pattrick, John Charnock Contents What do we mean by mechanism? Two examplesMicrocrystalline materials- iron oxides and sulfides Biological transformations- bioreduction Legacy wastesPu behaviour in storage ponds Why Do We Need to Know Mechanisms? We often need to extrapolate, e.g. scaling up to process plant or evaluating long term wasteform performance. Understanding physical and chemical mechanism allows realistic description and gives greater confidence in these extrapolations A Simple Example- Radioactivity Measurements 100 95 Count rate C ount ra te 100 90 85 80 60 40 20 80 0 0 1 Time (days) 2 0 5 10 15 20 25 Time (days) Count rate = 100 - 0.08 x Time Count rate = (Count rate)t=0 e –k x Time Decrease to 50% after 6.25 days Decrease to 50% after 8 days But the real world is complicated Problems- low concentrations, complex mixtures, fine particle size, heterogeneity, variability, biological activity Optical High Resolution Optical 1 mm 1 cm SEM 100 µm TEM SEM image of aquatic sediment 100 nm Concrete is heterogeneous on scales from cm to nm How can we study mechanism in these complex systems? Use laboratory models to reduce variability, provide greater control, and allow the use of higher concentrations of radioelements. This allows use of a wider range of techniques and characterisation at the molecular and near-molecular scale. But, because we are working with simplified systems, we need to interpret the results with caution NpO2+ NpO22+ Microcrystalline Materials The X-ray Absorption Spectroscopy Experiment Use synchrotron X-ray sourcemonochromatic, intense (106 x X-ray tube) Eject core electron from absorber atom Outgoing photoelectron wave reflected back from neighbouring atoms Interference pattern contains information on number, type, distance of backscatterers Double crystal monochromator Sample White beam Monochromatic I0 It X-rays If Imon XANES (X-ray absorption near edge spectroscopy)fingerprints oxidation state 4 3 2 1 0 21.0 21.1 21.2 21.3 XANES Spectrum 4 3 2 15.0 1 0 20.5 10.0 5.0 20.7 20.9 21.1 21.3 21.5 21.7 21.9 22.1 0.0 -5.0 Raw Spectrum -10.0 -15.0 0 EXAFS (extended X-ray absorption fine structure)- quantitative modelling of coordination environment (n ± 20%; r ± 0.02 Å) 2 4 6 8 10 12 14 16 Isolated EXAFS 80 60 40 20 0 0 2 4 6 8 Fourier Transform 10 Example- Uranyl Ion Reacting with Hydrous Fe Oxides Proposed coordination environment of uranyl on the iron oxide surface EXAFS spectroscopy gives: Number Type Distance (Å) 2 O 1.79 3 O 2.35 2 O 2.52 1 Fe 3.42 From: Waite et al., Geochimica et Cosmochimica Acta 58, 5465-5478 (1994) Why Iron Sulfides? Important and widespread mineral phases in anaerobic conditions such as aquatic sediments Microcrystalline with high surface area and redox active surface Originates from use of alternative electron acceptors in bacterial metabolism Fe(III)→ Fe(II) SO42- → S2Fe2+ + S2- → FeS Microbiology and Radioactivity (eds M J Keith-Roach and F R Livens), Elsevier Dec 2001 Removal of UO22+ and NpO2+ from Solution by FeS Uranium uptake is almost quantitative and is independent of solution concentration 120 100 UO22+ 80 60 40 20 NpO2+ 0 0 2 4 Initia l S o lution Con c (mM) 6 Neptunium uptake is relatively low but independent of solution concentration Reactions with mineral surfaces studied with XANES spectroscopy Uranyl reacted with Fe phases Neptunyl reacted with FeS Surface reactions with FeS studied by EXAFS Uranium Aqueous Increasing [U] Neptunium 60 50 0.27 mM Np soln. -1 (650 µ g g ) 50 40 Amplitude 20 3 k χ(k) (arbitrary units) Increasing [Np] 40 0.68 mM Np soln. -1 (1 625 µ g g ) 30 0.27 ppm Np soln. -1 (650 µ g g ) 30 0.68 ppm Np soln. -1 (1 650 µ g g ) 20 10 2.74 mM Np soln. -1 (6 500 µ g g ) 10 0 2.74 ppm Np soln. -1 (6 500 µ g g ) 0 -10 3 5 7 9 -1 k (Å ) 11 13 0 2 4 6 r (Å) 8 10 EXAFS Fitting Parameters Concentration 4500 ppm 24000 ppm 113000 ppm Uranium Concentration Number,Type Distance (Å) 2O 1.81 4O 2.40 2O 1.81 2O 2.14 4O 2.36 2O 1.83 1O 2.07 5O 2.31 Number,Type Distance (Å) 4O 2.25 3S 2.63 4O 2.25 3S 2.61 6500 ppm 4O 2.26 Neptunium 3S 2.64 650 ppm 1625 ppm Proposed coordination environment of Np on FeS surface L N Moyes et al., Environmental Science and Technology, 34, 10621068 (2000); Environmental Science and Technology, 36, 179-183 (2002) Microbiological Transformations Bacterial Redox Processes- Geobacter sulfurreducens Known transformationsTcO4- (soluble) TcO2.2H2O (insoluble) UO22+ (soluble) UO2.2H2O (insoluble) So how does Np (NpO2+ is potentially mobile) behave? Effect of G. sulfurreducens on NpO2+ 0.14 0.12 Absorbance % Remaining in Solution 0.1 100 0.08 0.06 0.04 0.02 0 800 90 900 1000 1100 Wavelength (nm) With acetate No e- donor Dead cells Control (no cells) With hydrogen 80 70 0 100 Time after Inoculation (h) 200 1200 Chemical Transformations • Assumed mechanism: 2e- reduction U(VI) Æ U(IV) • Alternative mechanism: 1 e- reduction U(VI) Æ U(V) – U disproportionates Æ U(VI) + U(IV) – Cycle of U(VI) reduction until nearly all U(IV) UO22+ 2 eU4+ UO22+ eUO2+ Æ UO22+ + U4+ Identification of U(V) by EXAFS Spectroscopy 100% 80% 60% U(IV) U(V) 40% U(VI) 20% 0% 0 2 4 8 24 Hours Abundance of Different U Species U(VI) U(V) U(IV) Plutonium in Ponds Observations- Discharges of Pu from ponds may be difficult to predict These effluents need to be carefully managed Useful Stuff and Important Questions What is in the ponds? Corroded Magnox sludge (CMS) What happens to the effluents? Water NaOH to bring pH to 11 Mg + 2 NaOH NaOH + CO2 Mg(OH)2 + CO32- Mg(OH)2 + 2 Na+ NaHCO3 MgCO3 + 2 OH- Effluent pH 11 Sand bed filter Adjust to pH 7 with CO2 gas Key QuestionsIs Pu in solution or particle associated? What is the nature of the particles? Zeolite ion exchange What is the nature of Pu solution species? What are the effects of effluent treatments? AlsoHow do we immobilise the Magnox sludges? Discharge to sea ESEM Imaging of CMS Simulant 1600 x 3200 x 9400 x Two morphologies Wide distribution of particle sizes XRD- brucite and artinite (MgCO3.Mg(OH)2.3H2O) Solubility, filterability, interconversion? What influences Pu solubility? % of Pu(IV) retained on a 0.22 µm filter Use filtration (0.22 µm) to define “solubility” Factorial experiment allows quantitative assessment of different controls on Pu 100 100 75 75 Analysis of variance 50 50 25 25 0 + CMS pH 11 - CMS pH 7 + CMS pH 7 + Si pH 7 + CMS +Si pH 7 Conclusions We need to understand the mechanisms underlying waste behaviour. There are complexities in defining the behaviour of the actinides, arising both from the chemical complexity of the actinides and from the complexity of heterogeneous waste systems. Using modern spectroscopic and analytical techniques it is possible to make progress. We have a good deal of work still to do, especially for the transuranic elements. It would be very helpful to have robust, fundamental models (e.g quantum) of these difficult systems