High magnetic fields are the Holy Grail of high-energy accelerators. The strength of the dipole field determines the maximum energy the beam can achieve on a given orbit, and large-aperture, high-gradient quadrupoles, with high peak field, govern the beam collimation at the interaction points. This is why, this September, members of the CERN Magnet Group in the Technology Department had big grins on their faces when the RMC racetrack test magnet attained a peak field of 16.2 T, twice the nominal field of the LHC dipoles, and the highest field ever reached in this configuration.

This result was achieved thanks to a different superconductor –the intermetallic and brittle compound Nb3Sn – used for the coils and the new "bladder-and-keys" technology developed at LBNL to withstand the extremely powerful electromagnetic forces.

The beginning of this success story dates back more than 10 years, when experts started to realise that Nb–Ti alloy, the workhorse of the LHC (and of all superconducting accelerators until then), and the conventional collar structure enclosing the superconducting coils in a locked, laminated assembly, would soon run out of steam. A technological quantum leap was needed.

Seeds sown

The first seeds of a European programme were sown in 2004, when a group of seven European laboratories and universities (CCLRC RAL, CEA, CERN, CIEMAT, INFN, Twente University and Wrocŀław University), under the co-ordination of CEA Saclay, decided to join forces to develop the technologies for the next-generation high-field magnets. Initially conceived to develop a 15 T dipole with a bore of 88 mm, the NED JRA EU-funded programme subsequently became an R&D programme to develop a new conductor. Its main result was an industrial Nb3Sn powder-in-tube (PIT) conductor with high current densities, designed to reach fields up to 15 T.

Three of the NED JRA partners – CEA, CERN and RAL – saw the importance of exploiting the new technology and continued the R&D beyond the NED JRA programme. Inspired by programmes at neighbouring laboratories, in particular LBNL, they started to develop a sub-scale model magnet with racetrack coils: the short model coil (SMC). This intermediate step led the partners to learn the basic principles of Nb3Sn coil construction. In fact, the SMC became a fast-turnaround test-bed for medium-sized cables, and is still in use at CERN. In 2011, the second SMC assembly successfully achieved 12.5 T. In a subsequent SMC assembly in 2012, the field went up to 13.5 T. With these results, CERN and its European partners demonstrated that they were on track to master Nb3Sn magnet technology.

Towards high fields

Since 2009, CERN and CEA have continued work on the technology, initially under the FP7-EuCARD project activities, and today within the scope of the CEA/CERN magnet collaboration. The focus of the FP7-Eucard high-field magnet (HFM) work package became the construction of a 13 T dipole magnet with a 100 mm aperture, which will be used to upgrade the FRESCA facility at CERN: FRESCA2. To achieve the 13 T objective, the CERN-CEA team designed the magnet for a field of 15 T using the state-of-the-art Nb3Sn technology: a 1 mm wire supplied by the only two manufacturers in the world capable of meeting the critical current specification, one in Germany (powder-in-tube, or PIT wire) and one in the US (rod restack process, or RRP wire). The cable for FRESCA2, was designed to have 40 strands and to carry nearly 20 kA at 1.9 K for a magnetic field of 16 T. This is an impressive set of values compared with the LHC, where the dipole cables can carry 13  kA at 1.9 K for a magnetic field of 9 T.

In spite of the engineering margins in the design, FRESCA2 proved to be a challenging goal. CERN therefore decided to design and construct an intermediate step, consisting of two racetrack, flat coils and no bore, made with the same 40 strand cable and fabrication procedures as for FRESCA2. This magnet was named the "racetrack model coil" (RMC).

Two initial assembly configurations were built using either RRP and PIT cables, and then a third one – called RMC_03 – was trained up to a maximum current of 18.5 kA at a temperature of 1.9 K. Based on the calculation of the field, this current corresponds to peak fields of 16 T in the coil wound with PIT cable and 16.2 T in the coil wound with RRP cable. With this result – a new record in this configuration – CERN has reached LBNL in the domain of high dipole fields.

Nb3Sn will be used to build the IR QXF quadrupoles and the 11 T dispersion suppressor dipoles for the high-luminosity upgrade of the LHC (CERN Courier January/February 2013 p28). The RMC record paves the way for a promising demonstration of this technique for future developments. In particular, cables of the same type as for FRESCA2 are also being considered for the Future Circular Collider (FCC) studies (CERN Courier April 2014 p16).

A lot of hard work remains before CERN and its collaborating partners will be able to achieve a 16 T field inside a beam aperture with the required field quality for an accelerator, so the development work on FRESCA2 continues. The coils are under construction and a test station is being built in SM18 to host the giant magnet, which should be ready for testing by next summer.