More robust thanks to imperfections

In most of the world’s nuclear power plants, uranium dioxide is used as fuel in the form of cylindrical fuel pellets. In a perfect lattice, uranium dioxide exhibits a ratio of two oxygen atoms for every uranium atom. Uranium and oxygen therefore form a regular lattice structure, in which every atom occupies a well-defined position. In reality, however – especially while a nuclear power plant is in operation – a different ratio between uranium and oxygen prevails in the fuel. Such deviations essentially occur in all materials in point of fact: even the purest table salt (sodium chloride, NaCl) never has the ideal ratio of one sodium atom per chlorine atom that it has in the theoretical structure. Deviations from the ideal composition also occur in actual uranium dioxide – such as if an oxygen atom is missing from one location in the structure or takes up an interstitial position, where it should not really be. Researchers have long studied the question as to whether deviations from the ideal composition and structure – whether it be due to a lack or excess of oxygen atoms – affects the behaviour of the fuel when exposed to radiation. During the nuclear reactions that take place in the nuclear reactor, for instance, the atoms in the uranium dioxide can collide with nuclei, which form as fragments of other atomic nuclei during the reactions. These collisions displace the uranium dioxide atoms from their positions in the lattice, thereby altering the lattice structure. A study conducted by the Paul Scherrer Institute (PSI) reveals that the changes caused by deviations from the ideal composition compensate for such displacements, which are triggered by the irradiation of the fuel. The resulting radiation damage is therefore smaller in structures that deviate from the ideal composition than in the perfectly structured uranium dioxide showing no imperfections. The deviations from the ideal structure thus facilitate a more effective self-healing in the uranium dioxide when exposed to radiation.

How radiation can weaken fuel

In a nuclear power plant, energy is obtained from nuclear reactions. During these nuclear reactions, the uranium nuclei are split by neutrons flying around. With the exception of normal hydrogen, neutrons are building blocks of all atomic nuclei. They are released during nuclear fission, collide with other uranium nuclei and cause them to split. Two to three additional neutrons form with every fission, which facilitate the progression of the fission reactions. The energy released during every fission event is primarily transferred to the heavy fission fragments in the form of kinetic energy. The heavy fragments move very rapidly and the majority of them are slowed down by collisions with atoms within the fuel, thus transferring a portion of their kinetic energy to these atoms. As a result, the fuel heats up extensively. The heated fuel transfers its heat to the metal cladding tube encasing the fuel (fuel rod), which in turn conducts the heat to the circulating cooling water. The heated water generates steam, which then drives a turbine and ultimately the electric generator connected to it.

Radiation damage occurs in conventional uranium oxide fuel when the quick-flying neutrons or fragments from nuclear fission collide with the fuel’s atoms and displace them. The collision can sometimes be so violent that several atoms are shifted consecutively, which can have a real knock-on effect, much like in snooker when a ball is fired quickly into a cluster of other balls. These so-called atomic displacement cascades take place extremely rapidly and on a very small length scale: a cascade initiated by a neutron or a fission fragment typically stops after just around six picoseconds (0.000000000006 seconds) and the displacements only extend over a few nanometres (one nanometre is one billionth of a millimetre). Although this makes it virtually impossible to examine these processes using direct measurements, they can be studied effectively on the computer with the aid of suitable methods. An established technique for this purpose is molecular dynamics. It describes the atomic motions roughly like those of snookers balls. Nevertheless, the fact that the forces exerted on atoms in a material are different to those that snooker balls are governed by is also taken into account.

Animation showing a collision cascade in uranium dioxide.

Structural imperfections in uranium dioxide considered for the first time

Previous calculations performed using molecular dynamics were limited to uranium dioxide in its ideal composition. While a nuclear power plant is in operation, however, uranium dioxide often deviates from its ideal structure. For instance, there is often a slight excess of oxygen at the temperatures and pressures that occur in normal operation. At even higher temperatures, on the other hand, there is a lack of oxygen atoms. Now, for the first time, the PSI researchers factored this into their study. In order to study the influence of structural deviations, the PSI researchers constructed different models of uranium dioxide on the computer – sometimes with the ideal composition, either with excess of oxygen, or with an oxygen deficit. The computer programme then calculated what happens if a uranium atom is given a very high amount of kinetic energy. This simulates the situation where it is knocked by a heavy, rapid fragment from a fission reaction. As expected, in each case the accelerated uranium atom triggered a displacement cascade, which can also be caused by a heavy nucleus fragment. Surprisingly, it turned out that the persistent damage – i.e. the number of lattice defects – is all the smaller at the end of these cascades the more the original structure of the uranium dioxide deviates from its ideal composition. This is evidently due to the deviations from the ideal structure occurring in the real material. These deviations absorb the majority of the displacements, which causes the collision itself. If a structure has less oxygen than the ideal, for instance, vacancies form within it. If excess oxygen atoms accumulate in one section of the structure due to a displacement cascade, they plug pre-existing vacancies, which ultimately results in fewer defects in the structure.

The simulations also reveal that the complete self-healing occurs with an oxygen excess of around 7.5 percent compared to the ideal structure. In other words, almost all of the deviations triggered by a displacement cascade disappear. As co-author of the study Raoul Ngayam-Happy points out, however, this result still needs to be verified via additional calculations before it can be deemed confirmed. Moreover, it is less significant for practice as such a high oxygen surplus would never occur in the actual uranium dioxide used as nuclear fuel. As Matthias Krack, head of the study and the Fuel Modelling Group at PSI, stresses: With our study, we have demonstrated for the first time that deviations from the ideal structure in uranium dioxide have a positive impact on its self-healing capability. The original structural deviations of the actual material therefore protect the fuel against the radiation damage that can occur while a nuclear power plant is in operation. Consequently, the study deepens our understanding of the crucial processes that take place in the fuel, which helps to assess the safety of the fuel rod correctly. However, the study did not examine other important factors that also influence fuel-rod safety.

Text: Paul Scherrer Institute/Leonid Leiva