Actinide Electrochemistry and the PUREX Process

Colin Wessells
March 20, 2012

Submitted as coursework for PH241, Stanford University, Winter 2012

Introduction

Species (M) MO22+ MO2+ M4+ M3+
U 0.07 0.62 -0.63 -1.66
Np 1.13 0.7 0.15 -1.8
Pu 0.94 1.04 0.98 -2.0
Table 1: Reduction Potentials of Selected Actinides (V vs. H+/H2). [6]

During fission, the U-235 in nuclear fuel decays into an immense variety of daughter nuclides, including much of the periodic table, as well as most of the isotopes of each element. [1-2] Some of these fission products are dangerous for their radioactivity, or chemical toxicity and reactivity, or both. An intuitive but challenging solution to the inherent danger of spent nuclear fuel is to store it in isolation for many years, allowing the most dangerous radioisotopes to decay.

Fissile -U235 is so naturally scarce that the world's supply of fuel for light water reactors will be expended in the next few decades. [2] Reprocessing spent nuclear fuel to harvest useful nuclides such as Pu-239, and then using them as fuel could provide much more energy, allowing the nuclear power industry to continue for much longer. Reprocessing also lessens the challenges of safely storing nuclear waste, as a much larger fraction of the fuel is consumed before its final disposal.

It is worthwhile, therefore, to consider how nuclear fuel is reprocessed. This is most commonly accomplished by the "PUREX" process, which is named for "Plutonium, Uranium, Reduction, Extraction." [2-4] The PUREX method was developed to purify Pu, but later was adapted on industrial scales by the nuclear power industry. During key steps, the PUREX process relies on the remarkably different electrochemistry of U, Pu, and Np, which allows exploitation of chemical complexing agents that bind to actinides only when they are in particular electrochemical states. [2-4]

Electrochemistry of Actinides:

The reduction potentials of actinide cations were originally calculated using thermodynamic data, and have more recently been measured by numerous electrochemical techniques. [5-6] The actinides of greatest concern during the PUREX process are U, Pu, and Np. [2-4] Table 1 lists their reduction potentials in aqueous solution, with respect to the standard hydrogen reference electrode. In some cases, the reduction potentials depend on the pH or anions in solution, so the values presented here should be considered approximate. [6] As a simple example, U3+ has a reduction potential to metallic U0 of -1.66 V, so if the potential of an electrode in a solution containing U3+ is fixed below that value, then U will be electroplated onto the electrode. For species with high positive valence states, oxycations such as UO22+ (uranyl), in which the valence of uranium is 6+, form in aqueous solution. This occurs because highly charged cations will bind strongly with electronegative anions such as oxygen, and similar oxycations are observed for transition metals, as for VO2+, TiO2+, and others. [7]

The reduction potentials of PuO22+ to PuO2+ and NpO22+ to NpO2+ are both much higher than that of UO22+ to UO2+, and furthermore, so are the reduction potentials of PuO2+ to Pu4+ and Pu3+, and of NpO22+ to Np4+. [6] This means that in a solution containing all three hexavalent actinides, mild reducing agents will selectively reduce just the PuO22+ and NpO22+, while leaving behind fully oxidized UO22+. If a stronger reducing agent is used, then PuO22+, NpO22+, and UO22+ will all be reduced. However, UO2+, which is the first reduction product of UO22+, is unstable against the disproportionation 2 UO2+ → UO22+ + UIV, where the identity of latter species depends on which anions in the solution can coordinate to U4+. [4, 6] In effect, this means that the introduction of a strong reducing agent to a solution containing PuO22+, NpO22+, and UO22+ will result in the full reduction of the first two species, but not the third. The use of a chemical reducing agent to produce a solution containing fully oxidized UO22+ and partially reduced Pu3+ and Np4+ is a key step in the PUREX process. [2-4]

Electrochemistry in the PUREX Process

Detailed, step-by-step descriptions of the PUREX process are available elsewhere, but it is worthwhile to revisit the role that the electrochemical properties of U and Pu play in the success of this process. [4]

The first steps in the PUREX process rely on solvent-exchange chemistry, rather than electrochemistry. The key is that after dissolving many transition metals and actinides in nitric acid, the addition of tributylphosphate (TBP) and an organic solvent results in the selective extraction of U and Pu. [2-4] This occurs because TBP preferentially coordinates to the UO22+ and Pu4+ in the presence of nitric acid, but not with other cations such as Pu3+ or NpO22+. [4] TBP is sparingly soluble in acidic water, but highly soluble (20% wt./wt. or more) in common organic solvents such as paraffin (kerosene). Therefore, agitating a solution of dissolved nuclear fuel, nitric acid, TBP, and paraffin results in the transfer of UO2(NO3)2(TBP)2 and Pu(NO3)4(TBP)2 complexes to the paraffin. The U and Pu complexes can be transferred back to aqueous media, resulting in solutions containing U and Pu, but not the other constituents of the original nuclear fuel. [2-4]

The differences between the electrochemical properties of U and Pu are finally exploited once this U/Pu solution has been prepared. In this solution, Pu is present in valence states of 3+, 4+, 5+, and 6+, but U is present primarily in its 6+ state. The addition of a reducing agent results in the reduction of PuO22+ to PuO2+, and then to Pu4+, and finally to Pu3+. [3-4] Any Np still present is reduced from NpO22+ to Np4+. The most common reducing agent used for this purpose is ferrous sulfamate (FeIISO3NH2, FS), although in recent years, hydroxylamonnium nitrate (NH3OHNO3) and hydrazine nitrate (N2H5NO3) have also been used, as they do not contaminate the solution with iron. [4] The reduction potential of UO22+ is much lower, so it is not attacked by FS as strongly. [4, 6] Though some of the UO22+ is reduced to UO2+ and U4+, the resulting disproportionation of UO2+ limits the loss of UO22+. TBP selectively complexes with UO22+, but not with Pu3+ or Np4+. [2, 4] Therefore, repeating the TBP/paraffin extraction process after the selective electrochemical reduction of Pu results in solutions that contain primarily U, or primarily Pu (with trace Np). Repeating this process several times before further chemical processing results in extremely pure U and Pu that can then be used as desired. [4]

Further Comments on the Electrochemistry of Actinides

Though not exploited by the PUREX process, the different reduction potentials of U, Pu, and Np to their metallic states could be used to separate them electrochemically. As the reduction potential of U3+ is higher than those of Pu3+ and Np3+ (Table 1), it is possible to selectively electroplate U, leaving the other species behind in solution. Recent investigation of this process demonstrated successful electrodeposition of UO2. [8] Further study is needed to determine the effectiveness of this method for selective removal of U from solutions that also contain other actinides.

If not for its scarcity and radioactivity, Pu would be extremely attractive for certain types of energy storage because its unique ability to remain stable in four different valence states in aqueous solution. This property could be exploited in a flow battery. Rather than using solid electrodes like traditional batteries, flow batteries contain liquid electrode solutions. [9] Each electrode solution contains a soluble species that can be reversibly reduced and oxidized by a catalyst electrode. A proton-conducting membrane between the electrodes maintains charge neutrality while the anode solution is oxidized, the cathode solution is reduced, and energy is extracted from the battery. For example, during the discharge of a vanadium flow battery, VO2+ is reduced to VO2+ in the cathode solution, while V2+ is oxidized to V3+ in the anode solution. [9] The battery is recharged by reversing these reactions. One major limitation to flow batteries is that their volumetric and gravimetric specific energies are low because a limited amount of electrochemically active ions can be stored in solution. Using a soluble electrode species that can undergo a change in valence of more than ±1 would increase the charge capacity of the cell. For this reason, Pu would make an excellent cathode for a flow battery: each PuO22+ can accept three electrons at the same nominal voltage (about 1 V) as that of the single-electron reduction of VO2+ to VO2+. [9] The complete replacement of the vanadium IV/V cathode with Pu would therefore increase the capacity of the flow battery. Though the cost, rarity, toxicity, and radiation danger of Pu preclude its use for energy storage, the same principle holds for other species with many simultaneously stable valence states.

Summary and Conclusions

Actinide electrochemistry is a scientifically rich field that has received widespread attention from the academic, defense, and industrial communities. Differences between the electrochemical properties of U, Pu, and Np are exploited to dramatic effect at a key step in the PUREX fuel reprocessing method. They are therefore critical for the success of fuel reprocessing on industrial scales. Further advances in the control of the electrochemical and chemical properties of actinides are already improving the efficiency of the PUREX process. Actinide electrochemistry will therefore remain a subject of active research and industrial application.

© Colin Wessells. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.

References

[1] N. Kocherov, M. Lammer and O. Schwerer, "Handbook of Nuclear Data for Safeguards," International Atomic Energy Agency, INDC(NDS)-376, December 1997.

[2] W. D. Loveland, D. J. Morrissey and G. T. Seaborg, Modern Nuclear Chemistry (Wiley, 2006).

[3] M. L. Crowder and M. C. Thompson, "Studies with Ferrous Sulfamate and Alternate Reductants for 2nd Uranium Cycle," Westinghouse, WSRC-TR-2002-00389, November 2002.

[4] K. L. Nash and G. J. Lumetta, Editors, Advanced Separation Techniques for Nuclear Fuel Reprocessing and Radioactive Waste Treatment (Woodhead, 2011).

[5] S. Kihara, "Analytical Chemical Studies on Electrode Processes by Column Coulometry," Electroanalytical and Interfacial Chem. 45, 45 (1973).

[6] S. Kihara, et al., "A Critical Evaluation of the Redox Properties of Uranium, Neptunium and Plutonium Ions in Acidic Aqueous Solutions," Pure Appl. Chem. 71, 1771 (1999).

[7] A. F. Holleman and E. Wiberg, Inorganic Chemistry (Academic Press, 2001).

[8] P. Giridhar, et al, "Extraction of Uranium (VI) by 1.1 M tri-n-butylphosphate/Ionic Liquid and the Feasibility of Recovery by Direct Electrodeposition From Organic Phase," J. Alloys Compounds 448, 104 (2008).

[9] Z. Yang et al., "Electrochemical Energy Storage for Green Grid," Chem. Rev. 111, 3577 (2011).