Designing a Mirror Model for High-Temperature Superconducting Magnets

By Ramesh Gupta

Rare isotope beam capabilities under development will be able to produce a wide range of isotopes for nearly all the elements that have ever before existed on earth. Fragment Separator, a crucial element of the proposed facility to study rare isotope beam capabilities, will then select one isotope at a time. One of the most challenging aspects of the design of the Fragment Separator will be high-temperature superconductor (HTS) magnets. These magnets use the unique zero resistance properties of high-temperature superconductors to generate magnetic field and withstand the high nuclear radiation heat load with the benefit of low operating costs.

To minimize the cost of prototyping these magnets, we proposed to test our design using a magnetic mirror model that uses iron to replicate the magnetic field and forces of the four-coil assembly while only building two half coils. At low fields, iron yoke provides an ideal field perpendicular condition required for a magnetic mirror. However, iron saturates at high fields, making field perpendicular boundary condition assumptions no longer valid in the design.

OPERA-2d from Vector Fields

> > OPERA-2d model of the proposed quadrupole design.

 

These deviations from ideal conditions cannot be computed analytically. We used electromagnetic simulation to model a realistic condition that is significantly different from ideal field perpendicular boundary condition. The resulting magnetic mirror model that we built not only matched the simulation results but also validated the HTS concept. Simulation played a critical role in the magnet design by enabling us to accurately evaluate the electromagnetic performance of our design concepts.

The Beam Is Discovery’s Key

Rare isotope beam capabilities will be studied and developed at a proposed major facility in United States for research in nuclear science. The facility will produce large numbers of rare isotopes when a high-energy heavy ion beam hits the target. A fragment separator will then select a particular isotope and transport it to an experimental area.

The proposed facility studying rare isotope beam capabilities will produce up to four orders of magnitude more of some rare isotopes than have ever been generated by any existing accelerator. It will produce a wide range of isotopes for each element, including many of the nuclei that participate in the various astrophysical processes, allowing a fuller exploration of the limits of nuclear stability. Additionally, studies of nuclei of astrophysical interest will help determine what goes on in stars and in the early universe.

The key to the scientific discovery potential of the facility is its ability to provide the highest-intensity beams of stable heavy ions for the production of rare isotopes through the fragmentation and fission process. Rare isotopes will also be produced by light ions on a heavy target through an Online Separation process that leverages other separation systems.

The facility’s driver accelerator will be a flexible device capable of providing beams from protons to uranium at energies of at least 400 MeV (megaelectron volt) per nucleon, with beam power in excess of 100 kW. In comparison to competing in-flight facilities, the facility’s capability for post acceleration will allow a wider range of studies and will include the measurement of nuclear reactions at astrophysical energies and the search for new heavy elements with long lifetimes.

Critical Role of Fragment Separator

The fragment separators that will be used in the facility to collect and separate the reaction products also will have to handle the high-power primary beam and dispose of the unwanted contaminant radioactive ions. Extremely high radiation levels with accumulated doses comparable to those in nuclear reactors and very high heat loads (around 15 kW) make the quadrupole magnets for the fragment separator one of the most challenging elements of the rare isotope beam capabilities project.

OPERA-3d from Vector Fields

< < Upper half of the symmetric OPERA-3d model of the HTS quadrupole design. The lower half of the iron yoke is simulated by the field parallel boundary condition; the symmetric lower half of the coil is hidden in this picture.

 

The challenge is to make these magnets with brittle HTS materials, that can operate at an elevated temperature, without significant degradation in superconductor performance. High-temperature superconductors, such as those produced by American Superconductor, are the only materials available that can economically generate magnetic field in this hostile environment. While the materials have been successfully used for power transmission cables as well as rotating machines, such as motors and generators, this is the first time that HTS will be used in an accelerator magnet. Hence, the concept had to be modeled and tested carefully to ensure successful operation at an elevated temperature in the 20°K to 40°K range.

But research and development funding for the project has been very limited, so it was impossible to build a full-scale model within the funds allocated in the initial few years of the program. We developed the concept of testing the proposed HTS approach with a magnetic mirror model. This cost-effective approach simulates the magnetic and mechanical structure of the full magnet, including field gradients, peak fields,  Lorentz forces, etc., but is considerably less expensive because it uses only one quarter of the conductor. This approach is based on the fact that iron provides a field perpendicular boundary at low fields and acts as a mirror if placed at a location where field lines are perpendicular.

To further reduce the cost, the coil length was reduced from 1125 mm in the full-scale model to 300 mm in the mirror model. The complete magnet will use 24 HTS coils, each consisting of 175 turns, enclosed in two cryostats. The six coils that were built for the mirror model will be recycled later for use in the first short model magnet, which requires a total of 24 coils. The mirror model was designed so that the initial test could be performed within existing facilities and without building a special cryostat.

Electromagnetic Model Simulation with a Field Perpendicular Boundary

The upper half of the full design was modeled using OPERA-2d and OPERA-3d software from Vector Fields. The primary purpose was to simulate the geometry of the mirror model magnet in a magnetic model with a field perpendicular boundary condition.

Physicists graphically generated a model of the magnet by defining a two-dimensional cross-section through the model and extruding the third dimension. The source currents were specified using the library of coil shapes provided with the program. The analysis provided graphical output including graphs and histograms of the solution and contour plots that showed the magnetic field values superimposed on the surfaces of the model. The simulation played a critical role in the design process.

OPERA-3d from Vector Fields

> > OPERA-3d model of the complete magnetic mirror design.

 

The HTS coils were wound with HTS wire from American Superconductor. HTS wire offers significant advantages over LTS conductors because it can be operated at much higher temperatures, typically 20°K to 40°K compared to 4°K with LTS. The ability to operate at higher temperatures makes it possible to use a much less-complex cryogenic system to cool the magnet,  which substantially reduces the required capital investment.

At the same time, HTS conductors deliver lower operating costs because less power is required for cooling. HTS conductors also provide greater stability and thermal headroom for enhanced reliability, flexibility, and uptime through increased tolerance to thermal spikes and cryocooler degradation. Finally, another advantage of HTS conductors that is not utilized in this design, is that they can achieve much higher magnetic fields due to the fact that the critical field of HTS is much higher than that of LTS.

Fabrication and Testing of Magnetic Mirror Model

The HTS coils were fabricated first with a manual machine and later with a computer-controlled winding machine. The new winding machine is more efficient and provides better control, requires less labor, and maintains a record of winding parameters. The magnet was assembled from a series of flat spirals fabricated by co-winding stainless steel reinforced HTS ribbon with an additional stainless steel insulating tape. Two such coils connected at their inner turns form a double pancake, and the magnet is built up of such units mounted in an aluminum fixture that supports the magnetic forces and locates the winding on the iron pole piece.

The magnetic mirror model was tested to validate the technology and measure the current carrying capacity of the coil pack as a function of temperature. The vertical test facility at BNL is equipped to do measurements in the temperature range of 4.2°K to 80°K by cooling the magnet and allowing it to warm adiabatically. The process is slow enough that the difference in temperature between various sections of the coil is small during the measurements.

In the first series of measurements, the current carrying capacity of the entire six-coil package was measured as a function of temperature. The design operating current of the quadrupole is about 125 A at 30°K. In the short magnetic mirror model, the peak field on the conductor is somewhat lower than in the full-length quadrupole, so the equivalent current in this design is about 150 A. The magnet reached the intended current at 30°K with some margin.

Up to now, no such accelerator or beam line magnet has been made with HTS. The challenging magnet requirements of the fragment separator and the recent advances in HTS offered a unique opportunity to seriously evaluate this approach. The successful test of this magnet is the first significant step toward demonstrating that HTS-based magnets can provide a good technical solution for one of the most critical items of the rare isotope beam capabilities proposal.

The large thermodynamic savings in the operating costs are the primary motivation for using HTS magnets in the facility. HTS magnets reduce the effective heat load by a factor of ten and bring an enormous savings in the operating cost of the 400 kW end of the fragment separator. Operating at 30°K offers a factor of 10 decrease in the cooling requirements compared to 4°K operation. The performance target of 150 A at 30°K was met with some margin. One of the keys to this successful application was the ability to accurately simulate the magnet well in advance of prototyping.

Note: The author appreciates the assistance of Structured Information and Vector Fields.

Ramesh Gupta is a Scientist with the Superconducting Magnet Division at the Brookhaven National Laboratory. Send your comments about this article through e-mail by clicking here. Please reference “HTS magnets, November 2006” in your message. You can also visit the author’s Web site by clicking here.

Contacts

American Superconductor
Westborough, MA

Brookhaven National Laboratory
Upton, NY

Vector Fields
Aurora, IL