The ITER-like Wall Project

One of the main challenges for fusion reactors is the compatibility between a reactor-grade plasma and the first wall. To suppress carbon migration which co-deposits tritium in the divertor region, the ITER design comprises a beryllium-clad first wall in the main chamber, while use of carbon (CFC) tiles is limited to the divertor strike points only, and tungsten is to be applied for the rest of the divertor. However, such a combination of materials has never been tested in a tokamak, let alone in one with ITER-relevant geometry and plasma parameters. That is why recently JET succeeded in obtaining approval for installation of an ITER-like wall in 2008. Two designs are under consideration. The reference design consists of a beryllium wall in the main chamber and an all-tungsten divertor (see figure) which matches ITER back-up/second stage design. Another option is to use CFC tiles at the divertor strike points, which would make the material choices identical to those in the ITER reference design. A final choice between these two options will be made in 2006.

Preparations for installation of the new wall have been progressing rapidly. For tungsten plasma facing components, a research programme is currently underway to determine whether tungsten-coated CFC is suitable, and what coating thickness will be required. Should coated CFC not be deemed suitable, the alternative would be to use solid tungsten tiles.Tungsten will also be used for neutral beam shine-through protection tiles. During the one year installation period, extensive use of remote handling technology will be made to implement the new first wall and divertor in 2008. Following installation, the JET experimental programme will focus on optimising operating scenarios compatible with the ITER-like wall. The level of retained tritium and its dependence on plasma scenarios will be determined, and detritiation techniques will be validated. Plasma performance will be tested to show that the level of tungsten reaching the core is acceptably low.

The lifetime of the wall will be studied with ITER-relevant power loading provided by increased neutral beam heating power. This level of heating power is expected to produce ELMs and disruptions which could cause melt damage to the first wall. The performance and life of the wall will be studied in the presence of such events, and mitigation techniques for ITER will be demonstrated and optimised. The ITER-like wall programme on JET will benefit from expertise gained at ASDEX Upgrade on tungsten coating of carbon tiles. There is also a synergy between the objectives of the two programmes: ASDEX Upgrade is exploring the viability of a tungsten first wall (tungsten is considered the long-term front runner as a material for fusion reactors), while JET will be looking at more immediate ITER needs.

High Frequency Pellet Injector Project

Another key component of the JET enhancement in support of ITER is a new High Frequency (50-60 Hz) deuterium ice Pellet Injector (HFPI) for mitigation of Edge Localised Modes (ELM) and for deep fuelling experiments. As the ELM mitigation is the main objective of this injector, the technical requirements are based on the results of recent ASDEX Upgrade experiments devoted to the study of ELM control. It has been demonstrated that the ELM frequency can be imposed by the pellet injection frequency, which leads to a significant reduction of the energy ejected during each ELM.

The new injector will be designed on the basis of the injector recently installed on Tore Supra at CEA Cadarache and installed in the Torus Hall close to the existing centrifuge pellet injector, which will be kept in place to allow a maximum of flexibility. In particular simultaneous fuelling (centrifuge) and ELM control (new injector) experiments should be possible. The new pellet injector will be connected to the vacuum vessel through an appropriate pumping line followed by a selector and by the existing pellet guide tubes. The selector system will allow selection of either the new injector or the existing one and also the track to be used to convey the pellets to the plasma (the three tracks already used with the centrifuge are still in place). The new pellet injector will be able to produce both small pellets at high frequency for ELM control (1 – 2 mm3 , up to 60 per second, 50 – 200 m/s) and large pellets at lower frequency for deep fuelling (35 – 70 mm3, up to 15 per second, 100 – 500 m/s).

Neutral Beam Enhancement Project

An upgrade of the neutral beam system on JET is planned to be installed in 2008 and subsequently deliver up to 35 MW of power for up to 20 seconds, and half this power for up to 40 seconds. This will allow exciting advances in extending ITER scenarios. With the higher power, the ELMy H-mode and Improved H-mode will be taken to higher normalised pressure βN, where Neo-classical Tearing Mode and Resistive Wall Mode control techniques will be refined under conditions closer to ITER. Furthermore, the long pulse capability of the upgraded neutral beam system will be crucial to progress advanced scenarios with full current drive sustained for an entire current diffusion time at ITER-relevant normalised pressure. In addition, greater demands will be placed on techniques for mitigation against large ELMs and disruptions. This is of particular importance in view of their potential to cause melt damage to the ITER-like first wall. The goals of NBE project will be accomplished by changing the magnetic configuration of the ion sources from present “supercusp” configuration to pure “chequerboard” configuration, replacing eight of the existing 80kV/60A High Voltage Power Supply (HVPS) units used for beam acceleration with four new 130kV/130A/20s units, and replacing critical components of each beamline, which presently rely on inter-pulse cooling (first stage neutralisers, beam duct protection etc.), with actively cooled components.

The main increase in neutral beam power will come from the conversion of the ion sources that generate the positive ions that are accelerated to form the neutral beams. This process takes place in a combined ion-source/accelerator module called a Positive Ion Neutral Injector (PINI). Chequerboard ion-sources produce larger fractions of molecular ions (D2 + and D3 +) leading to an increase of neutral beam power as a consequence of better neutralisation efficiency for the heavier (hence slower) molecular ions. The names “chequerboard” and “supercusp” derive from the arrangement of permanent magnets and the resulting magnetic field used to confine the plasma in the PINI ion-source. In order to convert it to a pure “chequerboard”, the rows of uni-directional magnets will be changed over to an alternating pattern like all the others. The plasma in the extraction region of the chequerboard ion sources is highly uniform, resulting in better optical properties of the beam and higher geometrical transmission, which will also contribute to higher power delivered to the plasma.

All 16 PINI accelerators will be modified to allow the maximum beam current to be raised from the present 60A to 65A (optimum beam optics condition at 125kV beam energy). This will further contribute to the power increase. This change requires replacement of the first grid in the accelerator structure (see figure) with a slightly modified design incorporating enlarged beam extraction apertures.

The present “supercusp” ion-source was originally developed to maximise the production of atomic ions (D+). The higher yield of slower molecular ions from the chequerboard ion-source will result in an increase of fractional (half and third) energy components in the neutral beam. In terms of the central plasma heating efficiency, normalised to total neutral beam power, the effect of the higher fractional energy components will be compensated for by the increase in acceleration voltage from 80kV to 125kV on the PINIs presently fitted to the Octant 4 injector box. This voltage increase (at 65A beam current) is only made possible by the new power supplies, which will also improve the reliability and availability of the NB system. In combination with the two similar 130kV/130A HVPS units commissioned on the Octant 8 neutral injector box in 2003, 75% of the JET PINI accelerators will be fed by new HVPS units when the enhancement is completed and brought into full operation in 2009.

Planning New Diagnostics Enhancements

A significant upgrade of the present JET diagnostics is also needed for an effective scientific programme in support of ITER. The upgrade will include diagnostics for ITER scenario development and systems required to fully exploit the JET new projects, such as ITER-like Wall and the High Frequency Pellet Injector. Final validation of diagnostic concepts for ITER, even if not indispensable for JET operation, could also be tested at JET before being installed on ITER.

According to these guidelines, new diagnostic systems were proposed for JET based on their scientific relevance at the end of 2004 by an expert working group, and is being considered by the EFDA Scientific and Technical Advisory Committee (STAC).