Deep (4-5 km) boreholes are emerging as a safe, secure, environmentally sound and potentially cost-effective option for disposal of high-level radioactive wastes, including plutonium. One reason this option has not been widely accepted for spent fuel is because stacking the containers in a borehole could create load stresses threatening their integrity with potential for releasing highly mobile radionuclides like 129 I before the borehole is filled and sealed. This problem can be overcome by using novel high-density support matrices deployed as fine metal shot along with the containers. Temperature distributions in and around the disposal are modelled to show how decay heat from the fuel can melt the shot within weeks of disposal to give a dense liquid in which the containers are almost weightless. Finally, within a few decades, this liquid will cool and solidify, entombing the waste containers in a base metal sarcophagus sealed into the host rock. PACS Code 28.41 -Radioactive wastes -waste disposal.

No abstract

[1] Deep borehole disposal (DBD) is emerging as a viable alternative to mined repositories for many forms of highly radioactive waste. It is geologically safer, more secure, less environmentally disruptive and potentially more cost-effective. All high-level wastes generate heat leading to elevated temperatures in and around the disposal. In some versions of DBD this heat is an essential part of the disposal while in others it affects the performances of materials and waste forms and can threaten the success of the disposal. Different versions of DBD are outlined, for all of which it is essential to predict the distribution of temperature with time. A generic physical model is established and a mathematical model set up involving the transient conductive heat flow differential equation for a cylindrical source term with realistic decay. This equation is solved using the method of Finite Differences. A Fortran computer code (GRANITE) has been developed for the model in the context of DBD and validated against theoretical and other benchmarks. The limitations of the model, code, input parameters and data used are discussed and it is concluded that the model provides a satisfactory basis for predicting temperatures in DBD. Examples of applications to some DBD scenarios are given and it is shown that the results are essential to the design strategy of the DBD versions, geometric details and choice of materials used. Without such modeling it would be impossible to progress DBD of nuclear wastes; something that is now being given serious consideration in several countries.

The effect of ergodization, either by additional coils like in TEXTOR-dynamic ergodic divertor (DED) or by intrinsic plasma effects like in W7-X, defines the need for transport models that are able to describe the ergodic configuration properly. A prerequisite for this is the concept of local magnetic coordinates allowing a correct discretization with minimized numerical errors. For these coordinates the appropriate full metric tensor has to be known. To study the transport in complex edge geometries (in particular for W7-X) two possible methods are used.First, a finite-difference discretization of the transport equations on a custom-tailored grid in local magnetic coordinates is used. This grid is generated by field-line tracing to guarantee an exact discretization of the dominant parallel transport (thus also minimizing the numerical diffusion problem). The perpendicular fluxes are then interpolated in a plane (a toroidal cut), where the interpolation problem for a quasi-isotropic system has to be solved by a constrained Delaunay triangulation (keeping the structural information for magnetic surfaces if they exist) and discretization. All toroidal terms are discretized by finite differences.Second, a Monte Carlo transport model originally developed for the modelling of the DED configuration of TEXTOR is used. A generalization and extension of this model was necessary to be able to handle W7-X. The model solves the transport equations with Monte Carlo techniques making use of mappings of local magnetic coordinates. The application of this technique to W7-X in a limiter-like configuration is presented. The decreasing dominance of parallel transport with respect to radial transport for electron heat, ion heat and particle transport results in increasingly steep profiles for the respective quantities within the islands.

No abstract