Division of systems engineering

The MADMAX project was launched in November 2016 and is led by the Max Planck Institut für Physik in collaboration with several European institutes. The objective of the project is the discovery of axions with a mass of about 100 µeV, potential candidates for dark matter, having to meet CP violation (matter/antimatter imbalance). The realization of this objective requires the development of a dedicated detector constituted by an amplifier of the electromagnetic signal corresponding to the photons (wave-corpuscle duality) induced by the conversion of the said axions. This amplifier, called dielectric haloscope, is itself made of 80 disks of 1.25 m diameter whose positioning must be ensured very finely by piezoelectric motors operating under strong magnetic fields. The haloscope must be bathed in a strong dipolar magnetic field because the power of the amplified signal is proportional to the square of the magnetic field. Moreover, the power of the signal is also proportional to the size of the amplifier, hence the size of the dielectric disks of 1.25 m in diameter, which requires a strong R&D in parallel with the development of the magnet (the state of the art is currently limited to ~30 cm in diameter). Thus, the development of a giant dipole magnet producing a magnetic field of ~9 T in an opening of 1.35 m is necessary (Figure 1). Thanks to its expertise in this field CEA IRFU was selected, in November 2017, in the framework of an innovation partnership to design this magnet with the industrialist Bilfinger Noell GmbH.

In the development strategy for the MADMAX magnet, set up by the CEA IRFU, several demonstrators must be designed in order to mitigate the various risks identified. The first level of risk concerns the conductor of the MADMAX magnet and its behavior in the event of a quench (the sudden passage of the magnet from the superconducting state to the resistive state which can lead to a degradation of the magnet). Indeed, due to market constraints, the CEA IRFU team had to define a new conductor concept for the MADMAX magnet. This concept uses the so-called “cable in conduit” structure of fusion magnets while using a copper conduit instead of a stainless steel conduit (Figure 2). In addition, the cooling of the conductor is not standard for this technology because it is achieved by filling the conduit with superfluid stagnant helium at 1.9 K rather than by a forced circulation of supercritical helium at 4.5 K as for fusion magnets. These few kelvins of difference make it possible to increase the magnetic field of the magnet and thus its performance. These novelties, at the level of the conductor, questioned the feasibility of quench detection and therefore the protection of the magnet.

In order to validate the manufacture of the conductor, its cooling concept and the detection of the quench, a demonstrator was designed, manufactured, integrated and tested between March 2020 and August 2021. It is called MACQU for MADMAX Coil for Quench Understanding (Figure 3) . The entire design, from the conductor to the support structure, including the MACQU magnet, its heat shield and the busbars, was carried out at the CEA. The conductor was manufactured by ASIPP (Institute of Plasma Physics, Chinese Academy of Science), then sent for the manufacture of the MACQU magnet to Bilfinger Noell GmbH, also in charge of the invoice for the support structure, the heat shield and busbars.

The magnet was received in Saclay in March 2021 to prepare for its integration into the JT60 test station then in place in building 126. The JT60 station had previously been modified to accommodate the magnet. Indeed, a superfluid helium production unit, named TH0, had been installed there. The quench detection/magnet protection system (MSS: Magnet Safety System) had been restructured to meet the needs of the MACQU experiment and the station's piloting automata adapted to the new configuration of the experiment. This work was entirely carried out by the DACM and DIS teams from IRFU. Upon receipt of the magnet, the work mainly focused on the production of the junction boxes of the MACQU magnet allowing the decoupling between the electrical and hydraulic circuit of the conductor at the ends of the magnet, the finalization of the instrumentation in front of allow the study of the propagation of the quench as well as the placement of the bubars (Figure 4).

The test campaign began on May 18 and ended successfully on August 27, 2021, providing the desired answers and opening up unexpected new avenues of work on the design of junction boxes, for example. The proof of concept of the cable and its cooling was demonstrated during these tests. With about sixty quenches detected on MACQU, the risk of a MADMAX magnet protection fault has been ruled out. The thermo-hydraulic behavior of the quench could be studied in greater detail and it was demonstrated for the first time that the phenomenon of thermo-hydraulic quench back also took place in superfluid helium (Figure 5). This phenomenon accelerates the quench because preheating by isenthalpic expansion of the helium in the conductor downstream of the quench ends up reducing the temperature margin of the conductor. This small margin de facto reduces the heating time to reach the critical temperature of the conductor at the level of the quench front and thus the latter accelerates.