Ion irradiation used as surrogate of neutron irradiation in graphite: Consequences on $^{14}$C and $^{36}$Cl behavior and structural evolution

International audienceGraphite has been widely used as neutron moderator, reflector or fuel matrix in different types of reactors such as gas cooled nuclear reactors (UNGG, Magnox, AGR), RBMK reactors or high temperature gas cooled reactors. Their operation produces a great quantity of irradiated graphite or other carbonaceous waste (around 250,000 tons worldwide) that requires a special management strategy. In the case of disposal, which is a current management strategy, two main radionuclides, 14C and 36Cl might be dose determining at the outlet. Particular attention is paid to 14C due to its long half-life (T∼5730 years) [1] and as major contributor to the radioactive dose. 14C has two main production routes, i) transmutation of nitrogen (14N(n,p)14C) where nitrogen is mainly adsorbed at the surfaces of the irradiated graphite; ii) activation of carbon from the matrix (13C(n,γ)14C). According to leaching tests, it was shown that even if the quantity of 14C released in the solution is low (less than 1% of the initial inventory), around 30% is in the organic form that would be mobile in repository conditions [2,3]. 36Cl is mainly produced through the activation of 35Cl (35Cl(n,γ)36Cl) which is an impurity in nuclear graphite. Its activity is low but it might be highly mobile in clay host rocks. Thus, in order to make informed decisions about the best management process and to anticipate potential radionuclide dissemination during dismantling and in the repository, it is necessary to collect information on 14C and 36Cl location and speciation in graphite, after reactor closure. The goal of the present paper is therefore to use ion irradiation to simulate neutron irradiation and to evaluate the irradiation effects on the behavior of 36Cl and 14C as well as on the induced graphite structure modifications. For that, to understand and model the underlying mechanisms, we used an indirect approach based on 13C or 37Cl implantation to simulate the respective presence of 14C or 36Cl. These isotopes were implanted into Highly Oriented Pyrolytic Graphite (HOPG) samples used as a model material system representative of the nuclear graphite coke grains which form around 80% of nuclear graphite. Nuclear graphite is manufactured from petroleum coke grains (filler) blended with coal tar pitch acting as a binder. Shaped blocks are formed by extrusion of the blend. They are heat-treated up to about 2800 °C (graphitisation treatment) and polycrystalline graphite is obtained. Blocks, intended for the moderator or reflector, may be further impregnated with pitch, re-baked and regraphitised in order to increase the density. Virgin nuclear graphites have initial densities in the range 1.6–1.8 g cm−3. The difference with graphite crystal (density = 2.265 g cm−3) is due to internal porosity. As a result of mixing of several carbon compounds, this material is structurally heterogeneous at a local scale. Nuclear graphite presents a complex multiscale organisation. It can be locally more or less anisotropic and not completely graphitised. Nuclear graphite has a polycrystalline structure and contains micrometer sized grains. The grains are formed by several more or less oriented crystallites with a size of a few hundreds nanometers. Each crystallite is formed by a triperiodical stacking of graphene planes. Nuclear graphite contains also small amounts of impurities like oxygen, hydrogen, metals and halogens, among them chlorine [4]. Ion beam irradiation was used as a surrogate for neutrons because it may produce cascades (due to ballistic interactions) that could be similar to those created by neutrons in the nuclear reactor. Ion beam (or electron beam) irradiation has been used for many years to simulate neutron irradiation. It has advantages such as for example the possibility to vary the irradiation conditions and sometimes to carry out in situ observations. Moreover, depending on the ion nature and energy, it allows covering a broad range of the neutron recoil spectrum and the rate at which atoms are displaced can be increased in comparison to reactor conditions. Dose rates can thus be much higher than under neutron irradiation allowing for higher amounts of displacements per atoms (dpa) to be reached within some days instead of months or years. Moreover, because there is no sample activation, the samples are not radioactive [[5], [6], [7], [8], [9], [10], [11]]. During neutron irradiation, the neutrons interact with the matter both by collision with the atom nuclei (i.e. ballistic damage) and by nuclear reactions. The first atoms hit by neutrons are caused to move, thus starting a cascade of atomic collisions leading to electronic excitation as they go through the matter and on the path of the atoms they displace (recoil atoms). The ballistic damage can be evaluated using the nuclear stopping power and can be denoted by the number of displacements per atom (dpa). The effect of electronic excitation can be quantified using the electronic stopping power. The experimental simulation of neutron irradiation in a reactor can be done by irradiation of the graphite samples with different ions of different energies. The choice of these parameters enables the study of the damage effects with or without electron excitation or ballistic damage. Thus, knowing that the impinging neutrons induce mainly ballistic damage into the graphite matrix but that part of the recoil carbon energy is also transferred through electronic excitation, it is interesting to use ion irradiation because both ballistic damage and electronic excitation effects can be studied coupled or decoupled according to the nature of the ion, its energy and the fluence. It is possible to cover a wide range of electronic and nuclear stopping powers by working with different particle accelerators. Thus, we simulated the effects of these different irradiation regimes using ion irradiation by varying the Sn(nuclear)/Se(electronic) stopping power ratio as well as the irradiation temperature (from room temperature up to 1000 °C). Indeed, during reactor operation, neutron irradiation leads to changes in the graphite lattice parameters depending on irradiation conditions such as flux and fluence but also temperature [12]. Finally, Secondary Ion Mass Spectrometry (SIMS) analysis was used to determine 13C and 37Cl distribution profiles and allowed us to follow the implanted isotopes behavior. The structural modifications were followed by High Resolution Transmission Electron Microscopy (HRTEM) and Raman microspectrometry