Vitrification | Radwaste

Cold crucible retrofit

9 May 2011

Cold crucible induction melter technology (CCIM) has the potential to increase the capacity of vitrification plants by enabling higher temperature, faster operations with higher reactivity fission products and a higher concentration in glass. After five years of development, a CCIM retrofitted in the Areva La Hague facility produced its first container of vitrified waste in April 2010, ramping up to full capacity in 2011. By S Naline, F Gouyaud, V Robineau, C Girold and B Carpentier.

Countries including the US, France, the UK, Germany, Belgium, Japan and Russia use vitrification to treat high-level radioactive waste. Vitrification has three major objectives: sustainable containment of long-lived fission products, minimization of final waste volume, and operability in an industrial context.

France has been a pioneer in developing vitrification processes. High-level liquid radioactive waste from nuclear fuel treatment operations has been successfully vitrified for more than 30 years. French nuclear fuel cycle contractor Areva and nuclear research body CEA design and operate facilities and invent glass formulations which have produced more than 17,200 glass canisters to date (corresponding to more than 238x106 TBq immobilized in 6711 tons of glass). Work continues to improve the technology and matrix formulations in order to increase quality and reduce volume.

CEA has been involved in nuclear processing technology since the 1950s when it identified borosilicate glass as the most suitable containment matrix for waste from used nuclear fuel. In the 1960s it focused on vitrification technology featuring a single induction-heated, liquid-fed, batch-operated melter. Investigation of pre-treatment by calcination led to development of a the continuous process with a rotary calcinator. This technology was coupled with a Joule heated melter and integrated in the AVM (Atelier de Vitrification de Marcoule), the world’s first vitrification facility. It started active operation in 1978, vitrifying the backlog of waste stored at Marcoule, France and the high-level waste solutions resulting from the treatment of gas-cooled reactor (GCR) fuel. CEA then began developing the glass formulation suited to the waste resulting from the treatment of light water reactor (LWR) fuel. This led to the R7T7 glass composition which is used today at Areva’s La Hague vitrification lines in Beaumont-Hague, France.

In addition to the continuous two-step process with a metallic pot melter as used in France (Figure 1), some other countries vitrify waste in a relatively large ceramic furnace in a single step. Industrial continuous vitrification processes consist of two separate stages. The first stage involves calcination of the liquid fission product (FP) solutions at around 400°C. This is followed by a melting stage at around 1100°C in an induction-heated metallic vessel, in which the vitrification additive (glass frit used to speed up the process) and the calcinated FP solution (calcinate) are mixed before being poured into a metallic container. In the melting pot of a so-called Joule-heated melter, glass is heated by thermal conductivity from the walls of the pot to the core the glass bath.

Figure 1 Vitrification
Figure 1: Diagram of two-step continuous vitrification process

However with CCIM, the principle is to induce electric currents directly within the glass to raise its temperature without heating the crucible. The crucible walls are divided into chunks separated by electrical insulators to allow the electromagnetic field to pass through. The wall of the cold crucible is cooled by a water circulation system; a protective layer of cooled glass called a ‘self-crucible’ then forms, protecting the metal crucible from the effects of high temperatures and corrosion caused by the bath of molten glass (Figures 2 & 3).

Figure 2
Figure 2: Diagrammatic view of CCIM from side and above

The direct-induction heating method allows the temperature to be increased (to up to 1300°C for some new matrix formulations still being tested) making it possible to obtain new waste containment matrices beyond those which could have been produced with the Joule heated melter. Also, the CCIM technology allows the industrial vitrification capacity throughput to be significantly increased, since the higher the temperature, the faster the calcine digestion by the glass. A similar CEA prototype used to qualify the process has shown that the glass throughput of the R7 vitrification line CCIM could reach 40 kg/h, representing a potential line capacity increase of 100%.

The CEA R&D Department developed three new glass specifications for the production of vitrified waste canisters in the CCIM (see also box, ‘How to make radwaste glass’):

  • A medium level borosilicate glass for the vitrification of corrosive flushing solutions used for pre-dismantling clean-up of the UP2-400 plant at La Hague. These are known as CSD-B canisters.
  • A high level glass-ceramic for the vitrification of legacy, highly-corrosive UMo fission products (from recycled GCR fuel). These are known as CSD-U canisters.
  • A high level borosilicate glass for the vitrification of UOX fission products (fission product solutions derived from the processing of LWR fuel), with a high throughput (the capacity of the vitrification line is doubled by retrofitting a CCIM). These are known as CSD-V canisters.

The retrofit project

Development tests for the CCIM began in the 1980s at Marcoule and several retrofitting mock-ups have been tested. In 2005, the Vitrification 2010 project began to (a) retrofit a CCIM to the operating La Hague R7 facility, (b) renovate some equipment in R7, and (c) manage the CCIM industrial commissioning. The project brought together numerous players from the CEA and Areva. Approximately 100 people were staffed in La Hague, including the project owner, the Areva La Hague Programmes Department, and the prime contractor, Areva fuel cycle engineering company SGN. A further 50 researchers from the CEA worked on the Marcoule site, with 10 members of Areva Innovation, Research and Projects Department in Paris coordinating the research by interfacing with Marcoule and the industrial project team in La Hague.

The design phase kicked off in 2005. Detailed design lasted until 2007. Manufacturing and installation took place between 2008 and mid-2009, after which tests began. The latter project deployment stage included:

  • Preparatory work to upgrade the operations equipment in the area in which the CCIM was to be installed
  • Construction of a full-scale test platform in the Beaumont Testing and Development Laboratory identical to the radioactive environment of facility R7.
  • Installation of the CCIM and its ancillaries in the radioactive contaminated zone of La Hague R7.

Throughout these different phases, between 40 and 60 experts specialising in mechanical and chemical processes, nuclear safety, electricity, I&C and equipment and piping layout worked in the engineering teams. The results of research carried out by the CEA were taken into account constantly throughout the project. To ensure that these modifications were incorporated successfully in real time, the engineering team drew on its experience and knowledge of change management and tracking. The use of this methodology in the nuclear facilities was one of the keys to the success of the project. The project owner and engineers worked closely together at all stages of the project, the industrial operating experience of the one guiding the design work of the other. The Areva group chose to keep the manufacture of the mechanical parts and equipment in-house. Two group companies were also involved: mechanical equipment integrator Mecachimie carried out the detail design work and manufacture of certain parts of the CCIM; and equipment maker Mecagest manufactured the welded parts.


The main challenge for the engineering team was to install a new process into a radioactive facility that had been in operation for 20 years. The CCIM had to be installed in place of a Joule heated melter without impacting the existing masonry structures and upstream or downstream equipment. The connections with other equipment had to comply with the dimensional constraints of the previous process. In addition, the production line had to be adapted to accommodate the CCIM power supply and cooling system. This had to be done in compliance with the dimensions of the existing wall penetrations between the highly radioactive and contaminated zone and the non-radioactive zone.

As the CCIM was installed in a highly radioactive cell into which human access was prohibited (Zone IV), the work had to be carried out remotely (using remote handling arms, cranes, etc.). This required very precise validation of the CCIM installation and maintenance operations. Every time a new piece of equipment is installed in a radioactive cell, the materials used must be qualified as being irradiation-resistant. Qualification is based on previous experience in Zone IV, on bibliographical data, and, in the case of new materials, on configured tests in irradiators. Vitrification cells are characterised by an environment which is punishing from a thermal and chemical point of view. These constraints had to be included in the design studies to ensure that the appropriate materials were used. Certain materials had to be specially qualified in their heat and/or corrosion resistance.

One of the main technological challenges of the CCIM was to ensure that all the materials in contact with the molten glass were cooled. To optimize cooling, thermal modelling of the CCIM components was undertaken, including parameters such as cooling temperature and flow rate, materials used and equipment design.

The parameters and rules used to design and manufacture the CCIM were optimized for two reasons: to ensure a service life that is sufficiently long as to be compatible with operation, and to optimize management of the technological waste volumes generated when it is dismantled in the future.

The selection of some specific materials (for example copper) brought a number of manufacturing and assembly problems. A considerable amount of research and development was performed by Mecagest to ensure that these materials could be welded and worked with precision (in strict compliance with internal and external dimensions) without losing any of their specific properties. The dimensions along the entire length of the high frequency power line were scrupulously checked, running as it did through the existing wall penetrations. The operations to assemble the crucible’s shell also had to be monitored carefully to ensure that it did not lose any of its electrical, cooling or tightness properties.

The studies carried out by the engineering team involved cold tests which supplemented those already carried out by CEA’s R&D team. These tests were run on a full-scale prototype in Areva’s Beaumont Testing and Development Laboratory. A team of 15 including members of engineering’s Testing division and Areva teleoperators, were involved.

The prototype accurately reproduced the facility in several ways:

  • The cold crucible itself
  • The cooling loops
  • The operating system
  • The electrical power supply provided by the high-frequency generator and the high-frequency power line
  • The dimensions and the configuration of the industrial vitrification cell (remote operation).

The tests were conducted to:

  • Validate the process
  • Qualify the equipment prior to its installation in Cell Z4
  • Train die operators in an environment and conditions similar to those of industrial operation.

The key figures for this test phase are:

  • 25 weeks of testing
  • 240 elementary tests and 70 integrated tests
  • 580 hours of training spread over 18 months
  • 70 non-radioactive glass canisters produced
  • 45 operators and remote operators trained.


It took almost six months to dismantle the old equipment, remove it, and clean up the area in which the CCIM was to be installed. Most of the work was carried out remotely. The cleaning operations ended with chemical treatment of the cell, restoring it to a level of cleanliness that was very close to its original level when it was brand new. A new calciner and dust scrubber were installed in the cleaned-up zone along with the CCIM.

At the same time, other pieces of equipment that had been used for 20 years in the highly radioactive atmosphere were cleaned, upgraded and/or repaired. More than 1000 hours of work were carried out in compliance with the ALARA principle. The clean-up operations extended the length of time maintenance personnel can safely work in the radioactive zone by a factor of 10.

Thanks to all of these preparations and studies, the CCIM was installed smoothly in the radioactive cell of R7 in just a few days. Commissioning tests took place without shutting down the other vitrification lines in the plant. The operator was kept thoroughly informed at all times. Areva operating personnel joined the testing teams to benefit from their experience in industrial operation of a vitrification line in a radioactive facility, and to continue their training in the new CCIM technology.

To put a new process, new equipment or a new glass formulation into practice, the La Hague plant has to meet two requirements:

1. Glass specification approval has to be obtained from the French Nuclear Safety Authority (ASN), based on the specification file which includes details of the process qualification and the glass performance qualification. The latter includes the chemical and radiological formulation, as well as a description of the long-term behaviour of the glass made. These qualifications are made by the R&D laboratory, that is, the CEA. Finally, the specification file is compiled by Areva and sent to the safety authority to be approved.

2. Authorization for industrial start-up to be granted by ASN, based on the safety analysis of the facility involved in the engineering.

All of the changes in process require considerable preparation before the industrial plant can be put into service. This preparatory stage involves about ten different technical sectors, including: the vitrification producer, the analysis laboratories, the process and glass quality inspection sector, the glass product canisters management sector, industrial maintenance, IT, the plant programme sector, and the safety sector.

Apart from the technical preparation required for the physical installation of the CCIM, a certain number of organizational details had to be finalized in preparation for commissioning. Work was needed to draw up personnel training plans, a maintenance and spares programme, and procedures and operating documents. This complex, intensive, three-year process that involved a large number of specialized areas was started in 2007. It included:

  • A total of 8000 hours training provided to 300 people to ensure top-quality glass produced to the highest occupational and nuclear safety standards. The training programmes also covered the process, the new heating technology and CCIM maintenance
  • New maintenance programmes assembled and 140 spare parts files prepared for hot cell equipment
  • All documents included in the operation or maintenance reference base upgraded. About 560 documents were revised, for example calculation methodology, analysis protocols, process control
  • The software used by the operating support department was upgraded or redeveloped. For example, glass composition calculation involves several sectors, from the producer to the canister return department, via quality control and the analysis laboratory
  • Preparation for qualification of the procurement of new raw materials (glass frit and reagents)
  • Compilation and approval of the operating quality reference base by an external auditor.

When the CCIM is up and running on an industrial scale, it will be possible to produce glass using several types of solution, a variety of fission products and flushing effluents from facilities being cleaned up and dismantled. This offers considerable operating flexibility but requires full mastery of the transition phases, as well as meticulous management and monitoring of raw materials, the process and the equipment used. Campaign changes are possible and could take place several times a year.


The industrial commissioning of a CCIM on an existing vitrification line in the R7 facility is the successful outcome of an innovative, complex and ambitious five-year project, preceded by intense R&D. It is the result of very close collaboration between the CEA and Areva. Areva successfully project managed the complex operations carried out in highly radioactive cells which came in on schedule and within budget. Since April 2010, almost 50 canisters have been produced (corresponding to about 20 tons of glass). A full-capacity industrial campaign was scheduled to begin in January 2011.

The project is a world first. Not only is it the first time a CCIM has been commissioned to vitrify high-level solutions, it is also unique in that an innovative technology was successfully integrated into an existing cell that had not changed in 20 years of high-level solution vitrification. Installation of the CCIM makes possible the vitrification of larger volumes of waste. In future, more compact equipment and a more efficient process will be possible by feeding the solutions directly into the vitrification furnace. This will lead to a considerable increase in vitrification capacity and flexibility.

Author Info:

This article is based on the paper ‘Vitrification 2010-A Challenging French Vitrification Project to Retrofit a Cold Crucible Inductive Melter at the La Hague Plant’ - 10382 - presented at the Waste Management conference in Phoenix, Arizona, 7-11 March 2010.

S. Naline, Areva NC, Tour Areva, 1 place Jean Millier, 92084 Paris La Défense, France; F. Gouyaud & V. Robineau, Areva NC, 50344 Beaumont La Hague Cedex, France; C. Girold, CEA Marcoule, BP 171, 30207 Bagnols-sur-Cèze, France; B. Carpentier, SGN, 25 ave. Tourville, 50120 Equeurdreville Hainneville, France

How to make radwaste glass

The containment matrix composition is first determined through laboratory tests. During this stage, the leaching properties of the glass and the physical properties of the molten material (thermal conductivity, electrical conductivity, viscosity, crystallisation, etc.) have to be taken into account. The target is to maximize waste loading while ensuring that acceptable leaching behaviour and appropriate physical properties of the glass. Preliminary tests are performed on a full-scale prototype to check feasibility.

The qualification programme is used to demonstrate the feasibility of producing glass with the same properties as those determined in the laboratory. Different types of tests are used to:

- Specify the necessary operating conditions, such as temperature range in the calciner and the melter, molten bath stirring and sparging parameters

- Establish the process data (decontamination factors of equipment, material balance, thermal balance)

- Establish the recovery procedure for returning to nominal conditions in the event of a disturbance in the process.

- The physical properties are modelled and the results compiled.

Five different types of tests are performed on the full scale prototype:

- Tests to determine the nominal operating parameters (for example maximum and minimum temperature, maximum and minimum sparging, maximum and minimum stirring velocity) guaranteeing the quality of the material produced in the full-scale pilot by final characterization of its physical and chemical properties compared with the same material produced in the laboratory

- Sensitivity tests on chemical composition and operating conditions

- Transient mode tests to determine melter control parameters, process startup and shutdown, and calciner standby periods

- Degraded mode tests to identify procedures for offsetting or mitigating the impact of incidents on safety, on the process equipment, and on the material

-Extended long-term testing (500 hours) to demonstrate that process operation is stable, that the operating conditions specified for nominal operation and transient phases are applicable, and that the material properties remain constant over time.

From the time waste composition is known to a completed material and process specification to produce a glass, at least four years of R&D are required in the most simple case. Even then, several dozen lab glasses are produced and characterized and around 30 full scale tests are carried out.


[1] Chauvin E., Do Quang R., Drain F., Pereira Mendes F. - "French Industrial Vitrification Plant - 30 years old, robust and still innovating", GLOBAL, Paris (paper 9076 - 2009).

[2] Do Quang R., Petitjean V, Hollebecque JF, Pinet O., Flament T., Prod'homme A. "Vitrification of HLW produced by uranium/molybdenium fuel reprocessing in COGEMA's cold crucible melter", WM'03 Conference, Tucson.

CCIM installation CCIM installation
Figure 1 Vitrification Figure 1 Vitrification
Figure 2 Figure 2
Figure 3 Figure 3

Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.