article posted 04 April 2016
Damien Perret PhD, is a research and development scientist at the French alternative energies and atomic energy commission (CEA) where he
works on glass formulation for nuclear waste conditioning. In this field, Damien also develops and applies statistical methods for designing mixture experiments,
developing property-composition predictive models and analyzing big data. Prior to joining the CEA, Damien worked on photoresist materials for advanced microlithography a
pplications as a scientist at Dow Chemical Electronic Materials. In 2003, he received a PhD in materials science, studying acoustic and mechanical properties of oxide
and metallic glasses.
Glass formulation approach for mixed nuclear waste treatment using PIVIC incineration-vitrification In Can process
Damien Perret*, Isabelle Hugon, Olivier Pinet, Patrice Charvin, Florent Lemont
CEA, DEN, DTCD - Marcoule, F-30207 Bagnols-sur-Cèze, France
A new development project is being conducted by the CEA, in collaboration with AREVA and Andra, for the treatment and conditioning of solid technological waste,
a mixture of metals, organic and mineral matter, coming from MOX producing plants.
Although much less radioactive than spent fuel waste, this a-contaminated waste must nevertheless be stored in suitable packages. This is where an innovative
process, called PIVIC, could be used to treat and condition the waste in a single step. The process involves the following technological elements: plasma incineration
of organics, vitrification of resulting ashes and fusion of metallic pieces, all being carried out by using an In Can process.
For this application, evolutions of traditional sodium borosilicate glass formulations are needed, in order to meet the challenging requirements which are specific
to the process. First, vitrification is operated at high temperature, in the presence of a liquid metal phase, and in a crucible made from a silicon carbide-based
inner layer. Resulting process environment is highly reducing and volatility of alkaline and boron elements may occur. Moreover, due to the presence of metal and
glass liquid phases in contact, aluminium from the metal phase can be oxidized, leading to an increase of glass alumina content. Lastly, in the configuration where
the waste is introduced in silica-based fibber bags, the silica content of the glass phase will increase, which has to be taken into account when evaluating potential
Among the glass properties that have to be mastered, viscosity is certainly one the most important for the PIVIC application. The glass melt has to be fluid enough
in order to enable mixing movement and efficient incorporation of the ashes. But at the same time, low viscosity of the glass melt may exacerbate corrosion and volatility
phenomena which have to be avoided as much as possible.
It is well known that silica, alumina and alkaline contents play a key role on the glass melt viscosity. In this work, potential glass compositions for the PIVIC
process are presented. The effect of glass composition on viscosity was investigated and volatility tests were conducted. In parallel, a statistical approach was
developed for fitting the glass formulation to the wide range of waste composition.
Figure 1. The PIVIC process at CEA Marcoule, France
Figure 2. Simultaneous fusion of metal (yellow, in the middle) and glass (orange) in the fusion unit of the PIVIC process
Figure 3. Cross section of a PIVIC can showing the metallic and the glass layers