Experimental particle physics at the energy frontier has been performed with colliding beam accelerators and particle detectors. Superconducting magnets for these detectors and for physics experiments including astroparticle physics have much advanced in the past years based on advancements in superconductivity. In the framework of EUCAS17, the European Conference on Applied Superconductivity, that was hosted in Geneva we review the main steps and developments in the field of superconducting magnets used in detectors.

The use of superconducting solenoids in particle physics is summarized in Table I. Noticeable developments of toroidal superconducting magnets include ATLAS Experiment at CERN and JLab’s toroidal spectrometer.  

Magnetic fields provide a convenient method to analyse the momentum of charged particles. In order to extend the energy reach of detectors, magnetic fields are required on a large scale. Moreover, the appearance of practical type-II superconductors in the early 1960s sent a shock wave through the communities of experimentalists involved in the mega-science disciplines of high energy physics. Limits on the size – imposed by the cost of power, and on the field level – imposed by the saturation of iron, of the large magnets required for the spectroscopy of charged particles would be swept away.

The reaction was immediate, and very large spectrometer magnets that took advantage of the new technology were designed, built and running within a few short years. In order to ensure cryogenic stability, a large fraction of stabilizing copper was incorporated into the matrix of the conductors used to build coils for bubble chamber magnets, at Argonne and Brookhaven National Laboratories in the US, and at CERN in Europe, in much earlier periods. Although they required a certain dexterity with the operation to eliminate problems due to eddy currents etc., the magnets achieved the desired fields and worked reliably, and the magnet for the Big European Bubble Chamber (BEBC) held the world record for stored energy of a single device for many years.

These magnets used primitive conductors with large filaments of Nb-Zr or Nb-Ti which were made until a fundamental study indicated that the twin requirements were for small filament size and twisting. This would enable the design of magnets with more compact coils using conductors requiring far less stabilizing material, thus heralding the modern era of superconductivity as applied to magnets.

Mechanical properties

The large magnetic fields and consequently the large forces induced called for an in-depth study of the mechanical properties of superconducting magnets. Different approaches have been developed to address the challenges of mechanical stability of the new powerful magnets.  Most superconducting magnets for collider detectors are solenoids. Solenoid fields have been widely used in many colliding experiments. These feature quasi-uniform field in the axial direction and benefit mechanically from being self-supporting. The coil of the solenoid is thus an integral part of the detector system through which energetic charged particles will pass. This means that an effort should be made to minimize the material used in the coil and the cryostat – especially if the coil is located in front of the electromagnetic calorimeters – to reduce the level of Coulomb scattering.

Specifically, the thin solenoid magnet has featured with the use of indirectly cooled coil using aluminium-stabilized superconductor. Pure aluminium performs much better as a stabilizing agent than copper. However, the mechanical properties of pure aluminium were rather poor, and therefore the thin solenoid development had to focus on the mechanical reinforcement while keeping the superior stabilizing feature. The length of the solenoids is generally about twice the diameter of the bore. Over the years, the technology of this type of magnet has converged on the use of indirectly cooled coils made from aluminium-stabilized superconductor.

The twin constraints of coil transparency and cost drove the designs in this direction from the outset, and led to the successful development of co-extrusion of Nb-Ti/Cu with pure aluminium. Besides being far better that copper with regard to transparency, very pure aluminium performs better as a stabilizing agent than copper, thanks to its lower resistivity at low temperature. Magneto-resistance is also less of a problem. However, the mechanical properties of pure aluminium are poor, and with the relentless striving for higher fields (and thinner coils) recent development work has concentrated on overcoming this drawback.

Perhaps, it is worth adding that aluminium stabilization of the superconductor is a key technology in modern detector magnets as it contributes to the stability of the superconductor with minimum material and weight. Moreover, the technique has been much improved in mechanical strength. One approach was to provide reinforcement of the stabilizer itself or to work with a hybrid of soft high conductivity material with strong alloy.  A more homogeneous reinforcement has been achieved by combining micro-alloying and cold-work hardening.

Magnets for collider detectors

The two superconducting solenoids installed at the Large Hadron Collider (LHC) at CERN, for the CMS and ATLAS detectors, both rely on reinforcement of the aluminium-stabilized conductor. More specifically, the ATLAS solenoid adopted a “microalloying” approach for the mechanical reinforcement and relies on quench protection with a combination of a passive protection ensured by quickly spreading the quench to the whole coil winding using pure Al strips and an active protection using localized quench-protection heaters, resulting in the full energy absorption in the coil. CMS solenoid took an approach of hybrid configuration for the mechanical reinforcement and relies on an active protection with energy extraction into an external resistor.

A particular challenge of the ATLAS magnet system, the largest detector magnet system ever built, is to meet the physics requirements of a light and open structure that specially focus to provide precise muon particle spectroscopy. The central solenoid magnet provides an axial magnetic field of 2 T in a 2.3 m diameter warm bore in the central tracker region, providing a deflection power of 4.6 T·m. The coil is located radially in front of the liquid-argon calorimeter, so it must be both physically thin and as transparent with minimum interaction of particles resulting as possible to ensure good calorimeter performance. The design features (i) high-strength aluminium stabilized conductor, (ii) pure aluminium strip to induce uniform energy absorption in the thin coil, and (iii) a common cryostat, together with that of the liquid argon calorimeter, to minimize wall material. An extensive effort was put into achieving homogeneous reinforcement of the aluminium stabilizer while keeping sufficiently low electrical resistivity, as characterized by its residual resistivity ratio (RRR). An optimum solution has been found that uses a combined process of micro-alloying and cold-work hardening.

While these magnets must be designed to be sufficiently robust to survive quenching (i.e. transition to the normal resistive state), they are also designed with a large margin, the conductor operating at typically less than 50% (along the load line) of its short sample current at nominal field. Extra care was also taken to avoid crack-prone pockets of unfilled epoxy in the winding that could release sufficient energy to provoke a quench, so this should be rare. Passive protection is ensured by spreading the quench quickly to the whole winding using pure aluminium strips and/or inductive heating of the mandrel (quench-back), often aided by active protection with energy extraction into an external resistor. The magnets are very reliable.

Another approach refers to toroidal coils provide the unique feature of a closed magnetic field without the necessity of an iron flux return yoke. Because no field exists at the collision point and along the beam line, there is no impact on the beam – which is convenient for the accelerator. Within the toroid the field is inversely proportional to the distance from the axis. The particle momentum can be derived from measurements of the deflection angle combined with the sagitta of the trajectory. Toroids can be interesting for providing magnetic analysis in the forward direction, where solenoids are less efficient and the polar uniformity of the field gives better coverage than a dipole. However, for a number of reasons that have been analysed elsewhere they pose certain complexities related to the supporting structure – due to the forces appearing – and the efficient use of superconductors.

Finally, open axial field magnets are a variant of the solenoid, the open axial field magnet provides an axial field with completely open access from the sides and field shaping poles. Besides eliminating the coil from the trajectories of particles over a wide range of transverse angles, this geometry is very favourable for the installation and maintenance of detector equipment, as evidenced in a resistive version for an experiment at the CERN ISR. A major drawback is that the attainable field level is relatively low, and while an imposing (superconducting) evolution of the design was proposed for the GEM experiment at the SSC, regular thin solenoids are now generally preferred.

Compactness and transparency.

Compactness and transparency of a magnet are important to create a magnetic field with minimum disturbance for the particles and having maximum detector acceptance. In Table 1 that we mentioned above, X refers to the thickness of the coil in radiation lengths (X₀), which is a convenient way to express transparency. For these reasons, the ratio of stored energy to effective coil cold mass, so called the E/M ratio, is a useful parameter to scale the lightness, and compactness (or efficiency) of the magnet. The E/M ratio in the coil is approximately equivalent to its enthalpy, H, and it determines the average temperature rise of the coil after a quench as follows: E/M = H (T2) - H (T1) ≈ H (T2) where T1 is the initial temperature and T2 is the average coil temperature after full energy absorption in a quench.

In early generations of thin solenoid magnets, E/M is limited to be ~ 5 kJ/kg. Based on the development of high strength aluminium stabilizer, ~10 kJ/kg has been realized. Using similar high-strength aluminium stabilizer, the ATLAS central solenoid reached 8.1 kJ/kg at its test field of 2.1 T. The CMS solenoid achieved an E/M ratio of 12 kJ/kg at its nominal field of 4.0 T under a quench protection condition with a half energy extraction corresponding to a practical energy absorption ratio of 6 kJ/kg.

The CMS solenoid was not required to be a thin/transparent solenoid but there was a strong incentive to moderate the mass of the coil for reasons of physical size. As a record, a prototype magnet for the BESS Polar spectrometer achieved a full energy dump at a ratio of E/M ≈13 kJ/kg. The radiation thickness (X) of the detector solenoid is also determined reflecting the material properties. In the large detector magnet system, the majority of the material is aluminium stabilizer and the unit radiation thickness is 89 mm. The ATLAS solenoid has recorded 0.66 (X₀) radiation thickness including an effort to share cryostat (vacuum vessel wall) for saving the wall material.

Reaching low temperatures

Cooling of the superconducting magnets poses another major challenge. The first superconducting detector magnets were cooled by immersion in a bath of liquid. This method of cooling provides a stable temperature and is efficient for evacuating heat produced in the winding by eddy currents and other losses, particularly those associated with changing the current setting. For large magnets, however, the liquid-helium containment vessel is complicated and the quantity of helium required to fill the vessel can be very large. But detector magnets are run at constant current, and it was soon realized that it should be sufficient to cool the windings indirectly via flowing helium in a system of pipes. This leads to a much simpler design and is now universally accepted as the preferred method of cooling. The pipes must be sized to allow the cold mass to be cooled down, and insulated electrically from the winding to avoid having to use insulating breaks. In the case of solenoids the pipes are usually welded to the external mandrel which provides a large area heat sink that is bonded with epoxy resin to the coil via ground insulation composed of glass-fiber reinforced plastic ) or tape combining polyimide film with GFRP. The toroidal coils are not self-supporting: they are potted in rigid aluminium box structures, and the pipes are welded (or glued) to the boxes. Depending on the layout, the cooling is achieved by pumping either supercritical or two phase helium flow through the pipes, or (in the case of solenoids) by the thermo-siphon technique based on gravitational convection).

A bright future for superconductivity

Physics research continues to be a driving force in the quest for better conductors and they will be useful for the detectors of future colliders. The next generation of magnets planned for the Compact Linear Collider (CLIC), the International Linear Collider (ILC) and Future Circular Colliders (FCC) call for sophisticated design and advancements in the above mentioned technologies for superconducting magnets. Α new unified detector model has been developed for CLIC and the concepts explored for this detector are also of interest to the high-luminosity upgrade of the LHC as well as for a future circular electron–positron collider explored by the FCC study. In the last years, a lot of effort has been placed in developing concepts of a “general-purpose” detector for a future 100 TeV proton-proton collider. The design was originally based on a twin solenoid paired with two forward dipoles, but this have now been replaced by a simpler system comprising one main solenoid enclosed by an active shielding coil. This design achieves a similar performance while being much lighter and more compact, resulting in a significant scaling down in the stored energy of the magnet from 65 GJ to 11 GJ. The total diameter of the magnet is around 18 m, and the new design could benefit from the important lessons from the construction and installation of the LHC detectors.

A detector magnet design under consideration for a future 100 TeV proton–proton collider (top image), measuring roughly 18 m in diameter and 49 m long, as part of the CERN co-ordinated Future Circular Collider project. The single-detector concept for CLIC (lower image), which measures roughly 12 m long and 13 m high. Image Credits: FCC study (top) and CLIC collaboration (bottom) .

In the meantime the availability of reliable material has led to large-scale industrial and medical applications of superconductors, in particular for nuclear magnetic resonance (NMR) and magnetic resonance imaging (MRI). Demand for MRI magnets has created a substantial commercial market for superconductors, which can be justly regarded as a useful spinoff from the work done on the more esoteric magnets required for physics research. It is the continuing pursuit of large scale applications to magnets for physics research instruments that drives the search for better-performing superconductors – which will one day appear in more advanced commercial applications such as high field MRI. The work done on the development of superconducting magnets for particle detectors and fusion devices is not confined to the development of conductors. The very large magnets call for sophisticated mechanical analysis, for the development of materials capable of withstanding the electromagnetic forces in a hostile radiation environment, and for the development of robust electrical insulation capable of withstanding multiple heat cycling to cryogenic temperatures – development that also presents a high potential for spinoff.