THEA ENERGY: REINVENTING THE STELLARATOR D. A. Gates CTO, Thea Energy, Inc. Princeton, NJ USA Email: [email protected] S. Aslam¹, B. Berzin¹, A. Cote¹, D. W. Dudt¹, D. Fort¹, A. Koen¹, T. G. Kruger¹, S. T. A. Kumar¹, M. F. Martin¹, C. P. S. Swanson¹, E. Winkler¹, and the rest of the Thea Energy Team¹ ¹Thea Energy, Inc. Princeton, NJ USA Thea Energy, Inc. (formerly Princeton Stellarators Inc.) will design, construct, and operate quasi-axisymmetric stellarator fusion devices. Thea Energy’s novel design, exclusively licensed from a patent [1] developed at Princeton Plasma Physics Laboratory (PPPL), can form optimized stellarator magnetic fields without complex and costly modular coils. Unlike tokamaks, stellarators generate stable magnetic configurations with no plasma current required via pulsed inductive solenoids or high-power steady-state current drive systems. This allows stellarator power plant designs to run in steady state, with lower recirculating power. Stellarators can be designed to be immune from disruptions, the instability-driven sudden loss of plasma, which impose stringent design conditions on the tokamak. However, the complex 3-D modular coils previously required for stellarator systems have historically been a limitation in developing a practical stellarator fusion system [2,3]. While it may be possible to create a set of large, complex modular coils with sufficient precision to generate a stellarator plasma, as shown by Wendelstein 7-X (W7-X), such coils are the single largest source of technical, cost, schedule, and ultimately commercialization risks for modular coil-based stellarator architectures. Thea Energy’s novel design can form optimized stellarator magnetic fields without complex and costly modular coils. This novel stellarator architecture utilizes a combination of planar encircling coils (providing the axial field, similar to toroidal field coils in a tokamak) and an array of surface-mounted planar shaping coils to generate the complex 3D magnetic fields required in a stellarator. This design retains all the advantages of the stellarator approach and leverages the mature physics basis of the stellarator. The initial use case will be a near steady state deuterium-deuterium (D-D) stellarator neutron source, called Eos, which was also licensed from patent developed at PPPL [10]. A stellarator Fusion Pilot Plant (FPP), called Helios, will be twice the linear dimension of the Eos design. The stellarator concept has had a resurgence in interest in recent decades. The initial impetus for this interest was breakthroughs in the theoretical understanding of particle confinement in three dimensional magnetic systems [4,5]. The theoretical breakthroughs were followed by novel experimental designs including the HSX experiment at the University of Wisconsin - Madison and, more recently, the W7-X experiment at the Max-Planck Institute for Plasma Physics in Greifswald, Germany. These experiments have now convincingly demonstrated that the theoretical advances could be realized in experiment. However, these experiments, as well as the incomplete NCSX device which was partially constructed at the Princeton Plasma Physics Laboratory, all suffered from delays and cost overruns due to the highly complex nature and required high precision of the three dimensional magnet systems that were designed for these devices [2,3]. To address the issue of coil complexity, Thea Energy has developed a new two-stage approach to stellarator coil design that uses simple coils that are both planar and convex, enabling winding using traditional winding-machines. This paper describes the basic concepts that were developed to enable this radically simplified design concept which is referred to as the “All-planar Coil Stellarator”. Thea Energy’s novel approach to stellarator design was motivated by the experience of the previous generation of optimized stellarators. All three optimized stellarator designs that utilized what are frequently called “modular coils” have had serious delays in construction and faced problems achieving the precision required of these complex coils. In addition, the large out-of-plane excursions of these coils has led to reactor concepts [6] that employ port-based maintenance schemes. It is extremely difficult to achieve the required availability for a power producing fusion system using port-based maintenance. The basic method for designing coils for optimized stellarators was initially described by Merkel [7]. In this method, a winding surface is defined as an offset from the plasma boundary of an optimized stellarator equilibrium. A sheet current is then found on this surface that minimizes the normal component of magnetic field on the plasma boundary. Then by the virtual casing theorem, this sheet current will provide the appropriate magnetic field to support the equilibrium (subject to the additional constraint of matching the magnitude of the axial field to avoid the trivial B=0 solution). A current potential is then calculated, and the contours of this current potential define an initial guess for the shapes of the centerlines of the modular coils. The solution is then iterated to minimize the errors with finite-build coil representations. Many similar methods have since been developed for designing modular coils [8,9]. Thea Energy has developed a new coil optimization scheme, which will be referred to as the Thea method. It splits the coil optimization process into two steps by defining two different types of coil arrays, which are called encircling coils and shaping coils (alternately, large coils and small coils). The encircling coil array provides the axial field, and the shaping coils provide the remaining components of the normal field on the plasma boundary. The first optimization step is to minimize the normal component of the field on the boundary due to the large coils, which are subject to the constraints that the coils be both planar and convex. The geometry constraints ensure that the large encircling coils can be wound using traditional winding machines. The second step is to define the location of the shaping coils to be consistent with sector maintenance. In particular, the coil centerlines are used to define cut planes through a winding surface which is offset by a fixed distance from the plasma boundary. The cut-planes define toroidal sectors that are then further subdivided in equal arc segments defining rings. These rings are then divided poloidally in equal length segments making a grid on the winding surface. A planar rounded rectangle coil is placed in each element of the grid. The current in each coil is then optimized to minimize the residual normal component of the field on the plasma boundary. The first use case that is envisioned for the All-Planar Coil Stellarator is the Stellarator Neutron Source [10] called Eos. The purpose of this device is to produce radioisotopes, particularly tritium, in large quantities using neutrons from D-D fusion reactions in a device that is roughly half the linear dimensions of a full-scale (deuterium-tritium) D-T FPP. This device involves a unique combination of new technologies: 1) A quasi-symmetric All-Planar Coil Stellarator able to operate for very long pulses (>24hours), 2) optimized for fast-particle confinement, 3) featuring high temperature superconducting (HTS) coils, 4) integrating 1 MeV negative ion neutral beams (NNBI), and 5) incorporating high frequency (170GHz) gyrotrons capable of heating at high field (~6 Tesla on axis). Each of these technologies is required to optimize D-D neutron production. The All-Planar Coil Stellarator provides a device that has modular, replaceable blanket modules. The optimization for fast particle confinement is required to prevent fast ion loss from the injected neutral beams. The high current density, high critical temperature, and high field of the HTS coils is required to make the coils small enough to fit in a device of this size. The high energy NNBI is required to be in the plateau region of the D-D neutron cross-section. The high frequency gyrotrons are necessary to elevate the electron temperature to increase the slowing down time of the injected fast ions. This unique combination of technologies allow operation in a high-electron-temperature regime that is optimized to produce D-D neutrons in near steady state. Early estimates indicate the ability to achieve ~2x10¹⁷ neutrons/second from the neutron producing branch of the D-D reaction which will be used to produce tritium in the neutron blanket. The tritium producing branch produces another ~2x10¹⁷ tritons/second which roughly doubles the tritium production rates. The estimate is a total production rate of 0.2 grams of tritium per day in a plasma approximately one quarter of the size of the Tokamak Fusion Test Reactor (TFTR) (R~3 m, B~6 T, A~6). This tritium production rate is comparable to that of a CANDU reactor [11]. In addition to tritium production, this device can also be used to produce other medical isotopes by swapping out the blanket modules. Thea Energy has done preliminary calculations on a breeding blanket for the Stellarator Neutron Source, Eos. Initial investigations have explored the use of lithium carbonate as a combination moderator/tritium breeder. These studies indicate that a 40 cm water cooled conformal blanket with steel containment can reduce the neutron flux at the outer blanket boundary to acceptable levels. The criteria for acceptable neutron fluence were determined using a 30 year magnet lifetime and the damage threshold as determined by reference [12]. Only natural lithium isotope concentrations are required to achieve this level of performance. Upon completion of the Eos device, a follow-on D-T Fusion Pilot Plant called Helios, will be constructed. The device will have roughly twice the linear dimension (R~6m) of the Eos device with the same magnetic field on axis (6T). Blanket technology for Helios has not yet been fully determined but possibilities include dual-coolant helium/lead-lithium and ceramic solid breeders. NNBI will not be used for heating on this device, only Electron Cyclotron Resonance Heating (ECRH). To the extent possible the device will be a direct scale-up of Eos. Major changes will include the inclusion of large scale heat removal from the blanket. Thea Energy has successfully demonstrated that physically viable magnetic solutions exist for the All-Planar Coil Stellarator. This new concept enables the design of a stellarator FPP that is maintainable, controllable, and consistent with the requirements imposed by a neutron blanket. There is a large-port sector maintenance scheme. The first use case will be a Stellarator neutron source that will use D-D beam target fusion to generate 2x10¹⁷ n/s which is capable of generating 0.2g of tritium per day. Initial neutronics calculations indicate a <50 cm thick blanket system is sufficient for neutron moderation, generating the desired levels of tritium, and protecting the magnet systems from neutron damage. The follow on device will be a fusion pilot plant that is roughly twice the linear dimension of the neutron source [1] International Patent Application No. PCT/US2023/063949, filed March 8, 2023 (entitled “Planar Coil Stellarator” and published as WO/2023/178004) [2] Bosch, H.-S., T. Andreeva, R. Brakel, T. Bräuer, D. Hartmann, A. Holtz, T. Klinger, et al. 2018. “Engineering Challenges in W7-X: Lessons Learned and Status for the Second Operation Phase.” IEEE Transactions on Plasma Science 46 (5): 1131–40. https://doi.org/10.1109/TPS.2018.2818934. [3] Chrzanowski, James H, Thomas G Meighan, Steven Raftopolous, Lawrence Dudek, and Paul J Fogarty. 2009. “Lessons Learned During the Manufacture of the NCSX Modular Coils.” PPPL-4442. Princeton Plasma Physics Lab. (PPPL), Princeton, NJ (United States). https://www.osti.gov/biblio/963972. [4] D. A. Garren and A.H. Boozer, Phys. Fluids B 3, 2822–2834 (1991) [5] Boozer, Allen H. 1983. “Transport and Isomorphic Equilibria.” The Physics of Fluids 26 (2): 496–99. https://doi.org/10.1063/1.864166. [6] Najmabadi, F., A. R. Raffray, S. I. Abdel-Khalik, L. Bromberg, L. Crosatti, L. El-Guebaly, P. R. Garabedian, et al. 2008. “The ARIES-CS Compact Stellarator Fusion Power Plant.” Fusion Science and Technology 54 (3): 655–72. https://doi.org/10.13182/FST54-655. [7] Merkel, P. 1987. “Solution of Stellarator Boundary Value Problems with External Currents.” Nuclear Fusion 27 (5): 867. https://doi.org/10.1088/0029-5515/27/5/018. [8] Pomphrey, N., L. Berry, A. Boozer, A. Brooks, R. E. Hatcher, S. P. Hirshman, L.-P. Ku, et al. 2001. “Innovations in Compact Stellarator Coil Design.” Nuclear Fusion 41 (3): 339. https://doi.org/10.1088/0029-5515/41/3/312. [9] Landreman, Matt. 2017. “An Improved Current Potential Method for Fast Computation of Stellarator Coil Shapes.” Nuclear Fusion 57 (4): 046003. https://doi.org/10.1088/1741-4326/aa57d4. [10] International Patent Application No. PCT/US2023/064025, filed March 9, 2023 (entitled “Systems and Method for Stellarator Neutron Source” and published as WO/2023/178009) [11] Abdou, Mohamed, Marco Riva, Alice Ying, Christian Day, Alberto Loarte, L. R. Baylor, Paul Humrickhouse, Thomas F. Fuerst, and Seungyon Cho. 2020. “Physics and Technology Considerations for the Deuterium–Tritium Fuel Cycle and Conditions for Tritium Fuel Self Sufficiency.” Nuclear Fusion 61 (1): 013001. https://doi.org/10.1088/1741-4326/abbf35. [12] D X Fischer, et al., 2018 Supercond. Sci. Technol. 31 044006