Predicting radiation damage

Materials World magazine
3 Oct 2015

Enrique Jimenez-Melero considers the materials challenges in the future of nuclear energy, and the role of energetic ions.

The design and operation of an industrial power plant relies not only on in-depth understanding of the physical and chemical processes that generate energy, but also on the selection of suitable structural materials that can withstand the increasingly demanding operating conditions. For example, the drive for higher thermal efficiencies in steam power plants in the last century has led to a gradual increase in the main operating parameters, temperature and pressure, reaching values of 600°C and 300 bar. This move towards more demanding conditions has been facilitated by the development of structural steels with improved resistance to corrosion, oxidation and creep at elevated temperatures. 

In nuclear power plants, the potential radiation-mediated degradation of components must be predicted and mitigated, so that the safe operation of the reactor is not compromised. Currently, those predictions and mitigation strategies are based upon modelling tools that rely mainly on limited data gathered either during in-service inspection of reactor components or in irradiation campaigns at small-scale test reactors. 

Prospective problems

Two scenarios are looming where reactor components will be faced with extraordinary challenges to maintain their structural integrity beyond their current design capabilities – 

The extension of the lifetime of the current fission reactors from the initial licensed period of 40 years up to 60 years and even beyond.

The industrial adoption of fusion reactors as reliable alternatives to current power plants. 

Fusion reactor technology goes even further in the demanding environment where materials will be immersed. Magnetically confined fusion reactors, such as ITER or DEMO, will be built around a plasma containing two hydrogen isotopes, deuterium and tritium. Their reaction at temperatures close to the values in the sun (~15m°C in the core) yields one neutron and one helium nucleus. The released energy of 17.6MeV is shared between the neutron (14.1MeV) and the helium particle (3.5MeV). 

The natural abundancy of tritium is so low that the required amount for the fusion reaction is generated in the tritium-breeder reactor module by neutrons reacting with lithium nuclei. The high energy of fusion neutrons of 14.1MeV (compared with the average energy of 2MeV carried by the neutrons generated in a thermal fission reactor), together with the elevated amounts of helium generated from the nuclear reaction itself, implies that the materials facing the plasma will undergo unprecedented levels of radiation damage in their internal structure. 

Despite the fact that extending the initially licensed service lifetime and the build of new reactors will impose neutron damage levels far beyond our current expertise, test reactors with equivalent neutron energies and flux are not available worldwide. In recent years, attention of the international nuclear community has been placed on the use of ion beams produced in particle accelerator facilities to perform accelerated simulation tests of neutron damage to nuclear reactor materials. 

Neutron damage 

When an energetic neutron impinges on a given material, it undergoes a series of inelastic collisions, where it progressively transfers its energy to the lattice atoms that it encounters on its path. During those collisions, the lattice atoms may gain enough energy to leave their equilibrium sites and even knock off other atoms in the lattice. The result of this displacement cascade is the formation of a dense area and a diluted area in the material, due to the formation of a relatively high density of point defects, namely self-interstitial atoms and vacancies. This whole process takes place in only a few pico-seconds of time. 

Over time, these lattice defects may evolve into larger scale defects and have a crucial impact on the material performance that reveals itself unexpectedly after several years of service. At relatively low temperatures, the defect diffusion is hindered, and those point defects created by the neutron bombardment will recombine and annihilate themselves. At relatively high temperatures, the lattice defect concentration will be mainly dominated by the thermal equilibrium. However, at intermediate temperatures, those point defects will diffuse in the lattice and form larger defects such as dislocation loops, vacancy clusters and voids, or become trapped at other defect sinks such as grain boundaries, surfaces or interfaces. 

Radiation damage is characterised by the number of displacements per atom (dpa) – 1dpa denotes that, on average, every atom in the irradiated volume has been displaced once from its equilibrium position in the lattice. For example, fuel cladding in water-cooled reactors can experience a total dose of 15dpa after 4–5 years, and can approach 80dpa after 40 years of service. As a comparison, reactor pressure vessel materials are shielded from the fuel radiation by the cladding and the water coolant, and therefore experience significantly lower damage levels of 0.05–0.1dpa after 40 years of operation. First wall materials in a fusion reactor may even reach dose levels of 200dpa. 

One route to assess neutron damage is the use of fission test reactors, such as the Advanced Test Reactor in the US, the Petten Nuclear Reactor in the Netherlands and the Jules Horowitz Reactor currently being built in France. Despite the value of those neutron damage experiments, the access to test reactors is very limited and costly. 

A neutron surrogate 

The University of Manchester has a radiation science facility in Cumbria, the Dalton Cumbrian Facility (DCF). DCF was officially opened in 2013 and hosts a 5MV ion tandem accelerator. This is the only accelerator facility in the UK specifically devoted to simulating nuclear reactor environments. DCF now forms part of the National Nuclear User Facility and grants access to the nuclear industry and academic community. 

The tandem accelerator is attached to two ion sources –

A discharge-based TORoidal volume ion source capable of generating intense light ion beams (up to 100mA H and 15mA He beams)

A source of negative ions by cesium sputtering that produces heavy ion beams with a current of ~10mA.

The negative ion beam generated by either of these two sources is pre-accelerated and injected into the accelerator tank, with a terminal potential that can be set up to 5MV. Inside the tank, the negative ions pass through Argon stripper gas, so that they become partially depleted of their electrons and gain a second acceleration, now as positive ions. This leads to maximum beam energies of 10MeV (H+) and 15MeV (He2+). The energy of the heavy ion beam depends on the charge state of the ions after passing through the stripper gas. 

The use of a bending magnet allows the desired charge state of the ions to be selected, and to direct those ions to the beam line dedicated to radiation damage experiments. A series of beam line components allows us to measure the beam current and profile, and to focus the beam on the sample position. The sample is mounted on a stage that possesses heating/water-cooling capabilities and therefore allows us to control the sample temperature during the experiment, and to compensate for the additional heating due to the beam itself on the sample. By continuous measurement of the sample temperature and charge deposited on the sample, it is possible to perform accelerated irradiation experiments under controlled conditions of temperature, radiation dose and dose rate. Total dose levels that are reached in service reactor conditions after a few years of operation can now be achieved under controlled experimental conditions in just a few hours or days. 

When the charged particles bombard the material, they will gradually lose their energy by inelastic collisions with the lattice atoms. However, due to their charge they will also cause a relatively high density of excited/ionised atoms along their path. The consequence is that the stopping power of the material for charged particles is higher than for neutrons, and those particles will have a limited penetration depth in the material (a few hundred micrometres for protons and just 2-3 micrometres for heavy ions, as compared to potentially a few centimetres for neutrons). This means that the damage material available for post-irradiation examination is limited to a thin layer on top of the non-irradiated sample material. Luckily, we now have the expertise to prepare adequately proton-irradiated samples by traditional electro-polishing techniques, or heavy ion-irradiated samples by focus ion beam-based methodologies. 

The damage rate (dpa/s) in ion irradiation experiments is 2–4 orders of magnitude higher than the rates experienced by in-core materials in current water-cooled fission reactors. To compensate for the increased damage rate, the temperatures at which ion irradiation experiments are performed are higher than the neutron irradiation temperatures of materials in nuclear reactors. This shift in temperature allows us to achieve similar diffusion kinetics of the radiation-induced defects, and therefore generates equivalent damaged microstructures. 

These results highlight the correlation between the damaged structures of equivalent materials produced using different types of radiation, and herald a promising future where extensive ion irradiation campaigns can support predictive modelling development and also safety cases concerning both current and future fission and fusion reactors. 

Dr Enrique Jimenez-Melero is a Lecturer in Radiation Material Science, at The University of Manchester’s Dalton Nuclear Institute.