The Tritium Breeding Blankets (TBB) are major in-vessel components of the future fusion demonstration reactor DEMO. They have three main functions: breeding tritium, a necessary fuel for the fusion reaction, extracting the energy deposited by neutrons produced in the fusion process and contributing to the shielding of the superconducting magnets. As the blankets will be exposed to high temperatures and high neutron irradiation, it is essential to investigate their performance under these extreme conditions. Prof. S.L. Dudarev of UKAEA, a co-chairman of Fusion Materials Development Topical Group, and his colleagues Prof. R. Bullough (UKAEA), Dr. P.M. Derlet (PSI, Switzerland), Dr. S.P. Fitzgerald (UKAEA), Dr. M.Y. Lavrentiev (UKAEA), and Dr. D. Nguyen-Manh (UKAEA), have made a significance advance in modelling the behaviour of steel, one of the main components of the blankets, at high temperatures.

EUROFER steel, a material based on an alloy of iron (Fe) containing chromium (Cr) as a major alloying element, is a potential material for the TBB. It will be tested in the ITER Test Blanket Modules. EUROFER is a low-activation ferritic- martensitic steel. Martensite is a very hard form of steel microstructure which can be described as having “needleshaped” crystal grains. The behaviour of any material is controlled by its chemical composition and microstructure. Under DEMO operating conditions, the martensitic microstructure of EUROFER will be affected by defects generated by high energy neutrons. Furthermore, the microstructure will be affected by high operating temperatures. The Radiation Effects Modelling & Experimental Validation (MAT-REMEV) project pursued by the fusion materials topical group has the task of understanding these crucial phenomena. In order to ensure good and safe functioning of TBB during the operation of a fusion device, the project also aims at developing modelling tools and means for their validation to predict the behaviour of EUROFER steel at high temperatures and high neutron irradiation.

Most materials exist in various stable forms (liquid, solid, gaseous), depending on temperature and pressure conditions. These stable forms are called phases and are described by the phase diagram. The stability of EUROFER is closely related to the phase diagram of the Fe-Cr alloy. Iron is a transition metal, the atoms of which are arranged in body-centred cubic (bcc) crystal lattice. It is known that pure iron undergoes three phase transitions at high temperatures. At 1185 K and at 1667 K it changes its crystal structure from α to γ, and back, whereas beyond 1811 K it melts and becomes liquid. The terms α and γ refer to body centred (bcc) and face centred (fcc) cubic crystal structures, respectively. Ferritic-martensitic steels also exhibit those transitions at similar temperatures. It is also known that on approaching the α – γ transition temperature, bcc iron, or ferriticmartensitic steels, soften. Mankind used this effect for thousands of years since it makes iron and steels forgeable. It is also what may be responsible for causing the collapse of tall buildings during fires, such as the World Trade Center fire on 11th September 2001. In the case of fusion, this effect limits the upper operating temperature of TBB in fusion reactors like DEMO to ~ 550°C. In order to develop steels that can withstand high temperatures, one needs to understand the microscopic cause of this softening effect. The new development in the modelling methodology is a step towards this goal.

So far, the phase transition points could be observed experimentally and could be predicted using thermodynamic methods. Thermodynamics alone, however, cannot identify the various effects, which drive the phase transitions. Hence it does not suffice to explain the connection between the phase transitions and various related phenomena, like softening. In the case of iron, it was already known that magnetic excitations and lattice vibrations are responsible for the phase transitions. At the same time, it was not clear how magnetic excitations and lattice vibrations affect the strength of steels. How magnetism influences the structure of defects generated by the fusion neutrons, was also a still open question. The work performed in 2008 by the above mentioned scientists provides a way of understanding and quantifying these effects. A unified model has been developed, linking the notions of phase transitions, mechanical properties of steels, and radiation damage effects.

The UKAEA group is now working on modelling the phase transitions in Fe-Cr alloys (not just pure iron), which also fully includes the effects of high temperature magnetism. They call this new method the “Magnetic Cluster Expansion” (MCE). It employs quantum mechanics to parameterize a model Hamiltonian describing magnetism of specific configurations of atoms in alloys. Using Monte-Carlo methods, it proves possible to model the statistical properties of distributions of iron and chromium atoms in an alloy. Also the random changes of direction of the magnetic moments of atoms at high temperatures can be calculated by this means. The first applications of the MCE method are promising – the group has managed to predict rather accurately the magnetic properties of iron and Fe-Cr alloys. This is especially the case for the Curie point where the macroscopic ferromagnetism vanishes.

Initially the observed α – γ phase transitions of iron could not be reproduced by the model. It was assumed one has to include the second effect, which contributes to the phase transitions of iron: lattice vibrations, or phonons. To test this assumption, the contributions of lattice vibrations were introduced, using experimental neutron diffraction data. The improved model indeed predicts that phase transitions do take place at temperatures very close to the experimentally observed ones.

It is the first time that the various contributions to the phase transformation of iron at high temperature have been identified and assessed starting from abinitio quantum mechanical calculations. This pioneering work proves that magnetism is not the only factor responsible for the α – γ phase transitions of iron: the lattice vibrations are also essential contributors. Magnetism is responsible for the stability of the α-phase, and softening of iron at high temperatures, still below the melting point. But only the combined effect of magnetic fluctuations and lattice vibrations explains the stability of the γ-phase. The next step will involve including lattice vibrations into the magnetic cluster expansion analysis. It will enable this model to describe all the aspects of phase stability of Fe-Cr alloys and EUROFER ferriticmartensitic steel.

Sergei Dudarev from UKAEA and Jean-Luis Boutard from EFDA supported the work on this article.