return to contents

Does the Interaction of Plutonium Oxide with Water Pose a Potential Storage Hazard?

Recent world events have resulted in downsized nuclear arsenals and focused attention on additional ways of reducing the nuclear danger. Proper disposal of the surplus weapons-grade plutonium from dismantled weapons is necessary to prevent its diversion and reuse in weapons by proliferant states or terrorists. Most nations plan to use plutonium from both commercial reactors and weapons for nuclear power generation; the United States is considering various alternatives. Since selection and completion of a disposal option will probably take many years, surplus plutonium must be stored safely for an extended interim period.

DOE standards are established for storing surplus plutonium as metal (Pu) or as dioxide (PuO2). After appropriate processing and certification, these materials are doubly confined in leak-tight stainless steel containers. According to popular belief, oxide is a stable material and the preferred form for storing plutonium. However, storage of oxide is more complicated than storage of metal. If hydrocarbons such as oils and plastics are present with the stored plutonium, whether metal or plutonium oxide, they undergo radiolytic decomposition by alpha particles from the radioactive decay of the plutonium, and this decomposition generates hydrogen gas. Whereas metal consumes the hydrogen by forming plutonium hydride, hydrogen produced in an oxide container pressurizes the vessel and may lead ultimately to rupture and the release of plutonium.

Figure 4. This failed storage container (left) containing Pu metal was packaged in air containing some moisture. Whereas one end bulged, the weld on the other end (right) failed and ruptured. The drastic failure occurred after just a few days.

The Problem

Plutonium dioxide typically exists as a fine powder, and various hydrogen-containing species adsorb on its surface. Although oxide is processed by firing in air at 950°C to remove water, hydrocarbon, and other species before storage, water readily readsorbs when the oxide is exposed to room air during packaging operations. The fate of adsorbed water during storage is a concern.

Does residual water on the oxide pose a potential storage hazard? The answer to that question is a qualified "No." If oxide is prepared, certified, and packaged according to procedures developed by NMT Division, the amount of residual water is far too small to cause problems. However, the situation would be different if oxide were to be packaged mistakenly without being fired.

Two processes by which adsorbed water might generate pressure in oxide-containing storage vessels are being investigated by members of the Applied Weapons R&D Team in NMT-5. As described by the following reactions, these processes are (1) radiolytic decomposition of adsorbed water into oxygen and hydrogen and (2) chemical reaction of dioxide and water to form a higher-composition oxide (PuO2+x) and hydrogen:


H2O (adsorbed) -----> H2 (g) + 1/2 O2 (g). (1)

PuO2 (s) + x H2O (adsorbed) -> PuO2+x (s) + x H2 (g). (2)

Reaction 1 is similar to the radiolytic decomposition of hydrocarbon compounds noted above. Reaction 2 is suggested by early studies of plutonium chemistry and by recently published results showing that a PuO2+x+ phase forms on the dioxide surface in water vapor at 250°C to 300°C. The product is a mixed-valance oxide of Pu (IV) and Pu (VI) with x .3.

A number of questions remain to be answered:

  • Does any significant reaction occur at room temperature?

  • Does some other unanticipated reaction occur?

  • How fast are these reactions?

  • Are their rates fast enough to present a potential storage problem?

  • Do O2 and H2 combine to form water?

  • Is the recombination rate faster than the rate of radiolytic decomposition?

    Experimental Approach

    With so many questions to be answered, a large research effort might be anticipated. However, essential questions relevant to the storage issue are readily addressed by simple pressure-volume-temperature (PVT) experiments. After reactants are sealed in a vessel of known volume, the temperature and pressure are monitored over time. Reaction rates are defined by changes in pressure with time. The reaction is identified by analysis of the gaseous products.

    Two PVT experiments were conducted using PuO2 prepared by reacting high-purity Pu with oxygen. In one test the oxide-containing reaction vessel was evacuated and backfilled to the saturation pressure of liquid D2O at room temperature. In the other test the vessel was backfilled with approximately 125 torr of a mixture with D2 and O2 in a 2:1 ratio.

    Analytical results for the test with D2O are the key to defining chemical behavior. Radiolytic decomposition, Reaction 1, is the important reaction if D2 and O2 appear in a 2:1 ratio. Likewise, the formation of pure D2 shows that water reacts according to Reaction 2. Occurrence of both reactions produces a mixture with a hydrogen:oxygen ratio greater than 2.

    The importance of radiolytic decomposition is most easily shown by comparing the pressure changes for the two tests. Radiolysis cannot result in pressurization of an oxide storage container if the oxygen and hydrogen recombine to form water (reverse of Reaction 1) at a faster rate than that of the forward reaction.

    Results and Conclusions

    Results of the test with D2O show that adsorbed water reacts chemically with PuO2 to form D2 according to Reaction 2. The pressure increases at a constant rate over time, and hydrogen is the only gaseous product. At the measured rate a high hydrogen pressure can be produced in a few years in a typical storage container.

    Figure 5. Time dependence of the D2+O2 pressure over plutonium oxide at 25°C. Pressure dropped rapidly the first few days and gradually slowed to a nearly constant decrease that has continued for more than a year.

    An explanation for the apparent absence of radiolytic decomposition is suggested by kinetic results for the D2+O2 mixture. As shown in Figure 5, the pressure in the mixture dropped rapidly during the first few days and gradually slowed to an almost constant decrease that has continued for more than a year. The rate at which hydrogen and oxygen combine on the catalytic surface of PuO2 is so fast that radiolytic decomposition is not observed. Occurrence of Reaction 1 is not a concern for oxide storage. The D2:O2 ratio of the gas mixture is progressively increasing over time because the D2O product is reacting with the oxide as described in Reaction 2.

    The results demonstrate that excessive pressurization of a storage container might occur if the oxide were improperly fired or if unfired oxide were mistakenly packaged. Although all questions presented above have been answered, additional studies are being conducted to determine how the rate of Reaction 2 depends on temperature and on the concentration of water adsorbed on the oxide.


    The importance of these findings extends far beyond their relevance to storage safety. Further study of the formation of PuO2+x might help to explain why environmental plutonium appears as high-oxidation-state ions in solution and why high-fired oxide tends to be insoluble. Results of the oxygen-hydrogen combination study have already been used to explain why atmospheric corrosion of plutonium metal is accelerated by moisture.

    This work was done by members of the Applied Weapons R&D Team led by John Haschke.

    Editor's note: This article brings up an interesting debate. Independent reviewers of the article point out that other researchers have tried to produce PuO2+x and higher oxides by similar methods using oxygen, ozone, nitrous oxide, and other strong oxidizing agents. No higher oxide has been produced.

    The work presented here implies that water is a better oxidizing agent than oxygen, which is difficult to prove. The question also remains whether the extra oxygen in PuO2+x is actually adsorbed on the surface or incorporated into the lattice.

    NMT | LANL | DOE
    Phone Book | Search | Help/Info

    L O S  A L A M O S & #160; N A T I O N A L   L A B O R A T O R Y
    Operated by the University of California for the US Department of Energy

    Questions? - Copyright © UC 1996 - Disclaimer 26 June 1996