Demountable Superconducting Magnet Coils

Demountable Superconducting Magnet Coils

Brandon Sorbom1, Bob Mumgaard1, Joseph Minervini1, Dennis Whyte1

1MIT

Email:

Technology to be assessed

High-temperature superconducting toroidal field coils with joints for the purpose of modular, vertical maintenance inside the TF coils.

Application of the technology

While primarily envisioned for tokamaks and stellarators, this technology would be applicable to any magnetic fusion concept where it would be advantageous to temporarily break apart any of the magnetic coils. For tokamaks and stellarators, demountable toroidal field coils would allow vertical maintenance on components internal to the TF coils. In comparison to sector maintenance which requires components internal to the TF to be cut/welded to be removed/replaced, a vertical maintenance schemewould greatly simplify installation, maintenance, and replacement of internal components.

When coupled with liquid immersion blanket and a compact, high-field design, the vertical maintenance scheme enabled by demountable toroidal field coils would allow the removal and replacement of the entire vacuum vessel as a single component. An example of such a facility is the recent ARC reactor design study [1].This simplified maintenance scheme would allow the vacuum vessel in a power reactor to be considered a “consumable” component of the reactor—fabricated and tested offsite and replaced every few years. Reducing the required operational lifetime of the vacuum vessel has four principle advantages. First, it significantly lowers the survivability requirements of first wall material, both from a PMI and nuclear damage standpoint. Second, it reduces the total amount of activation experienced by a single vacuum vessel, making it easier to handle the vessel after it is taken out of the tokamak/stellarator. Third, installation of the vessel as a single piece allows all manufacturing and quality control to be performed off-site as opposed to welding pieces of the vessel together inside the TF coils. Finally, the option of modularly replacing the vacuum vessel lowers the risk of an off-normal event (such as an unexpected large disruption) permanently disabling the fusion reactor if the vacuum vessel sustains heavy damage. Demountable vacuum vessels would allow early fusion reactors to simultaneously perform as fusion nuclear science facilities by effectively moving the safety barrier out from the vacuum vessel to the blanket tank and using each replacement of the vacuum vessel as the opportunity to test new nuclear materials.

With regards to magnets, demountable TF coils would allow PF coils to be installed internal to the TF coils. Internalizing the PF coils would lower their performance requirements as well as enabling novel divertor topologies such as long leg divertors. Demountable TF coils would also allow replacement of single TF coils, reducing the risk that a TF fault would permanently disable an entire fusion reactor.

Finally, a vertical maintenance scheme as described above would have substantially faster maintenance times than a device with sector maintenance, allowing the availability of the fusion reactor to be higher and increasing its economic viability.

Critical variable(s) – variable that determines or controls the output of the technology

The goal of this technology is to allow faster, simpler maintenance to be performed on a fusion device, enabled by demountable superconducting coils. Since no reactor-type fusion devices exist yet to compare maintenance schemes to, a concrete first metric of success of this technology would be to operate a superconducting TF coilat full field without performance degradation (defined as an increase in joint resistance to unacceptable levels or degradation to the structural integrity of the magnet) after demounting and re-assembling the coil a certain number of times (determined by the estimated number of times a reactor would undergo maintenance in its lifetime). The critical variable is this case is the achievable joint resistance for a given current-capacity cable. Since HTS operating at low temperatures (e.g. 20 K) is relatively stable to temperature excursions, joint resistance primarily factors into the economics of the fusion reactor, as joint dissipation will set the size of the steady-state cooling system required (as opposed to being set by whether it is even feasible to remove the heat fast enough to avoid a quench which is the case with low-temperature superconductors).

The size of the required cooling system will not only affect the capital cost of the system but also dictate how much recirculating power from the fusion reactor is required to keep the cryoplant operational. The extent to which recirculating power affects the economics of the reactor is dependent on the output of the reactor itself, and thus will be different for each point design. However, for a fusion reactor on the order of 500-1500 MW of fusion power, joint resistances on the order of ~1 nΩ for a ~100 kA cable need to be obtained under the operating conditions of high field (5-10 T) without degradation after multiple (10-20) disassembly/reassembly cycles over the lifetime of the reactor. A joint resistance as described above would lead to modest (tens of MW) recirculating power needs for a power reactor producing 500 MW or more of fusion power [2] [3].

Recent experimental work has shown joint resistances as low as 1.8 nΩachieved in a 100 kA cable operating at 4.2 K, 0.45 T and only slightly higher for operation at 30 kA, 4.2 K, 7.5 T [4].

Design variables – parameters that can be controlled in order to optimize the critical variable. These could be qualitative.

The doctoral thesis of F. Mangiarotti [3] contains a comprehensive review of different parameters affecting joint selection and construction for fusion magnets. One of the main parameters to be explored is the geometry of the electrical joint. Among the plethora of different joint geometries are sliding joints, comb joints, lap joints, round joints, butt joints, and edge joints. Each different geometry has tradeoffs in parameters such as overall joint resistance, resistance distribution, manufacturability, and mechanical robustness. This parameter space must be more fully explored for each joint type at fusion magnet operating conditions in order to make an informed decision as to what the optimum joint type will be for a demountable fusion magnet.

Another important parameter affecting the ability of demountable joints to be used to enable effective vertical maintenance is the mechanical connection of the joint. Especially in compact, high-field tokamak designs, the TF magnet case will be required to tolerate high stresses due to JxB forces. The designed joint must be simple enough to allow for fast (and most likely remote) disassembly and reassembly while still being strong enough to hold the coil together during the normal operation of the TF coil. While there has been a small amount of work aimed at exploring mechanical joint designs[3] [5], this area is even less explored than the electrical joint selection and requires further design studies and experiments to select the optimal mechanical joint.

Risks and uncertainties

An inherent constraint on the technology is the physical properties of candidate cooling materials (i.e. liquid helium, hydrogen, or neon) affecting their ability to remove heat from the joints. However, the already high performance of REBCO conductor at relatively high temperatures (20-30K) combined with low experimentally achieved joint resistances indicates that cooling solutions will be feasible. There do not appear to be any fundamentally limiting constraints to this technology – the primary challenges will be engineering related. The main uncertainty in the calculations of steps 3 and 4 is due to the lack of experimental data which is itself simply a result of limited resources devoted to HTS magnet development (in the US and elsewhere). Another source of uncertainty is the fact that many design decisions are tied to the economics of a fusion reactor, a topic with considerable uncertainty in itself.

While there is little doubt that superconducting joints could be built, a balance between low operating cost (low joint resistance leading to low heat generation), high availability (simple mechanical joint design to enable fast maintenance), and reliability (robust joint design to prevent operational damage) will be required to justify the use of the technology.There are no inherent safety issues with this technology.

Within the US, institutional support of HTS magnet development in general has been limited, and demountable superconducting joints would only be possible with a more aggressive HTS research program. There has been little development of this technology in industry due to a.) a small (but growing) HTS market to begin with, and b.) limited benefits to current commercialsuperconducting magnet applications (e.g. NMR, generator magnets) which would not offset the cost of initial R&D. The most concentrated demountable coil work to date has been performed at NIFS in Japan in the context of the FFHR design, although the work is still in preliminary stages.There are no anticipated regulatory or societal obstacles to the development or use of this technology.

Maturity

This technology is currently at TRL4. Benchtop HTS superconducting joints built and tested at MIT and NIFS have demonstrated joint resistances that would likely be acceptable, but more experiments are required to collect performance data for all of the joint designs. Medium-sized jointed copper TF coils have been built and successfully operated in the Alcator C-Mod and DIIID tokamaks, demonstrating a mechanical joint solution at high magnet stress in the case of C-Mod, and illustrating the utility of vertical maintenance in the case of the vacuum vessel upgrade from DIII to DIIID.

Technology development for fusion applications

While demountable TF magnets would be advantageous by themselves, a simultaneous innovation in the area of liquid immersion blanketswould be strongly synergistic with demountable coil technology. To bring this technology to TRL6, further benchtop experiments on different electrical and mechanical joint concepts are required, as well as the construction of a facility which could wind and test a fusion-relevant HTS TF coil. Because much of the remaining work is engineering development as opposed to basic R&D, with adequate resources the time horizon for this technology to achieve its goal would be 3-6 years.

Most of the infrastructure to build and test benchtop-scale joints already exists, primarily at the National High Field Magnet Lab and at MIT. An HTS coil winding facility would be advantageous for compact, high-field research in general as well as benefitting joint research.

The only country besides the US which has been seriously investigating demountable joints is Japan, through the NIFS, although this program is also limited in scope. Thus, there is a large gap in global development in which the US could become world leaders in this technology.

[1] Sorbom, B. N., et al. "ARC: A compact, high-field, fusion nuclear science facility and demonstration power plant with demountable magnets." Fusion Engineering and Design 100 (2015): 378-405.

[2] Yanagi, N., Ito, S., Terazaki, Y., Seino, Y., Hamaguchi, S., Tamura, H., Sagara, A. (2015). Design and development of high-temperature superconducting magnet system with joint-winding for the helical fusion reactor. Nuclear Fusion, 55(5), 53021.

[3] Mangiarotti, Franco Julio. Design of demountable toroidal field coils with REBCO superconductors for a fusion reactor. Diss. Massachusetts Institute of Technology, 2016.

[4] Ito, S., Seino, Y., Yanagi, N., Terazaki, Y., Sagara, A., & Hashizume, H. (2014). Bridge-Type Mechanical Lap Joint of a 100 kA-Class HTS Conductor having Stacks of GdBCO Tapes, 9(3405086).

[5] Mangiarotti, F. J., & Minervini, J. V. (2015). Advances on the design of demountable toroidal field coils with REBCO superconductors for an ARIES-I class fusion reactor. IEEE Transactions on Applied Superconductivity, 25(3).