Boston Fusion (United States)

Abstract Understanding and accounting for uncertainty is one aspect of ensuring next-step tokamaks such as SPARC will robustly achieve their goals. While traditional Plasma OPerating CONtour (POPCON) analyses guide design, they often overlook the significant impact of uncertainties in scaling laws, plasma profiles, and impurity concentrations on performance predictions. This work confronts these challenges by introducing statistical POPCONs, which leverage Monte Carlo analysis to quantify the sensitivity of SPARC’s operating points (Creely et al 2020 J. Plasma Phys. 86 5) to these crucial variables. For profiles, a physically motivated gradient-based functional form is introduced. We further develop a multi-fidelity Bayesian optimization workflow that effectively identifies operating points maximizing the probability of meeting performance goals, which gives a significant speed-up over brute force search methods. Our findings reveal that accounting for these uncertainties leads to an optimal operating point different from deterministic predictions, which balances H-mode access, confinement, impurity dilution, and auxiliary power.

Fusion energy meets the twenty-first century World Grand Challenge of sustainable, ubiquitous, and safer energy sources. However, harnessing the promise of fusion energy has proven elusive. The competing approaches to fusion power plant design include inertial confinement fusion, National Ignition Facility (ICF-NIF, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Z</i> machine, etc.,) magnetic confinement fusion (MCF-Tokamak, stellarators, etc.), and other approaches that show promise in small- (flow stabilized <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Z</i> pinches) or large-scale applications. These approaches are being accelerated with private and public funding and seek to demonstrate the feasibility of different approaches to fusion-based power plants. Yet, how can the necessary pulsed power technologies for these disruptive technology bases be accelerated with no clear “Dominant Design?” Roadmapping holds the promise to identify and develop common critical pulsed power components for laboratory, prototype, and commercial fusion, and can accelerate the commercialization of fusion reactor designs. A preroadmapping Workshop on Pulsed Power for Fusion was held at the IEEE International Pulsed Power Conference in San Antonio, TX, USA, in June 2023. The workshop had 177 attendees. Here, the common elements for many of the ICF technologies vying for dominant design were identified. The advancement of these technologies through roadmapping will enhance commercial expectations that require their rapid and innovative development in the next five years, as well as the next five to ten years. The key technologies identified that underpin and limit the advancement of fusion power include pulsed power technologies such as energy storage, high-voltage switching, additive manufacturing, and modular pulsed power circuit topologies. They are the focus of our effort in the following roadmap scenario, which will delineate potential paths to technology development.

The critical current (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">I_c</i>) of 2G REBa<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub>Cu<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub>O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">7-x</sub> (REBCO) high temperature superconductors (HTS) is highly anisotropic with respect to the applied magnetic field (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</i>), with a strong <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">I_c</i> peak expressed at <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</i>||ab-plane of the HTS crystal structure. However, as manufactured, there is typically a small-angle ab-plane tilt offset relative to the plane of the tape, and a large average offset can significantly affect magnet <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">I_c</i> if unaccounted for, especially in non-twisted coils such as the SPARC toroidal field magnets being developed by Commonwealth Fusion Systems (CFS). To address this issue, the statistics of the magnitude and orientation of the ab-offset were studied on quick time scales in production quantities for two different HTS vendors by X-Ray Diffraction (XRD) using (005) rocking curves, and the offset was observed to be vendor dependent. The data were correlated positively with offsets determined by SEM, and to <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">I_c</i> at low temperature and moderate to high field by transport and torque magnetometry.

Abstract Recent developments to the M3D-C1 code enable higher fidelity modeling of disruptions, and can be applied in the design verification of reactor-scale tokamaks. Among these new capabilities is a method to mesh conducting vessel structures such as coils and passive plates, packing of the toroidal mesh around gas injectors, as well as anisotropic resistivity inside the vessel structures. We present extended-magnetohydrodynamic (MHD) simulations of disruption mitigation via massive gas injection (MGI) in SPARC. The goal of this study is to inform the disruption mitigation layout of SPARC and aid in the design of an effective gas injector configuration. Fully three-dimensional simulations with M3D-C1 are carried out for various injector configurations with the primary goal of determining the effect of different MGI parameters on heat loads and vessel forces. The simulations include a model for impurity ionization, recombination, advection and radiation, as well as spatially resolved conducting structures around the plasma. A localized mixture of deuterium and neon with a small toroidal and poloidal width is injected in up to six locations. We demonstrate that M3D-C1 can model a rapid shutdown via MGI using narrow and more realistic gas plumes than in previous simulations. As a result of the q = 1 surface in the SPARC baseline case a sawtooth is observed early in the simulations. Despite the sawtooth and the onset of edge MHD instabilities, the impurity distribution remains localized around the injector locations, but enables a radiative shutdown of the plasma. We find that using the maximum of six gas injectors results in a lower peaking factor and leads to a more even distribution of radiation toroidally than using two injectors.

Abstract Conditions similar to those at the end of the divertor leg in a tokamak were replicated in the linear plasma machine Magnum-PSI. The neutral pressure in the target chamber is then increased to cause the target to transition from an attached to a detached state. Superimposed to this steady state regime, edge localised mode (ELM)-like pulses are reproduced, resulting in a sudden increase in plasma temperature and density, such that the heat flux increases transiently by half an order of magnitude. Visible light emission, target thermography, and Thomson scattering are used to demonstrate that the higher the neutral pressure the more energy is removed from the ELM-like pulse in the volume. If the neutral pressure is sufficiently high, the ELM-like pulse can be prevented from affecting the target and the plasma energy is fully dissipated in the volume instead (ID 4 in table 1). The visible light images allow the division of the ELM-plasma interaction process of ELM energy dissipation into 3 ‘stages’ ranging from no dissipation to full dissipation (the target plasma is detached). In the second publication related to this study, spectroscopic data is analysed with a Bayesian approach, to acquire insights into the significance of molecular processes in dissipating the plasma energy and particles.

Commonwealth Fusion Systems (CFS) is pursuing the high-field approach to fusion energy, enabled by magnets composed of 4mm 2G REBCO wire. Most of these magnets will operate in the high field (20 T) and low temperature (∼ 20 K) regime and therefore require characterization of the critical current at those parameters. The industry standard is to use (77 K, self-field) data for rapid characterization of critical current. However, Ic (77 K, SF) is a poor predictor of performance at high (> 6 T) field, low temperature applications, and so more predictive correlations are extremely desirable. This study examines critical current correlations between high temperature (44-77 K) with low field (0-5 T) and low temperatures (20 K) with high field (6-20 T) from multiple manufacturers. Employing these correlations will allow for the use of cheaper and simpler critical current test systems that can still accurately predict high field, low temperature performance. Moreover, they will provide higher confidence in the predicted performance of magnets. This paper presents the predictive ability of these correlations and the associated error analysis.

The likelihood of realising tokamak power-plants will be greatly improved by the discovery of high-gain equilibria that resist the formation of small islands and hence avoid the disruptive neoclassical tearing mode. We propose a series of studies to understand how simple tokamak design can leverage aspects of tearing onset physics to maximise passive resistance to island formation. Here we investigate the variation that current profiles can bring about in preventing tearing onset through the cylindrical linear tearing stability parameter Δ′ . A database of 159148 realistic pilot-plant current profiles was generated with Monte Carlo sampling, and the distribution of Δ′ values was linked with interpretable profile characteristics. In agreement with prior theoretical and experimental studies, Δ′ was found to be strongly correlated with the existence and steepness of a local toroidal current well or hill, with the former destabilising and the latter stabilising. In the absence of these two cases, the remaining Δ′ values were linearly bounded by the toroidal current gradient at the rational surface.

Extended-MHD modeling of planned Runaway Electron Mitigation Coils (REMC) for SPARC and DIII-D is performed with the NIMROD code. A coil has been designed for each machine, with the two differing in shape and location, but both having n = 1 symmetry (with n the toroidal mode number). Compared to previous modeling efforts, three improvements are made to the simulations boundary conditions. First a resistive wall model is used in place of an ideal wall. Second, the ThinCurr code is used to compute the time-dependent 3D fields used as magnetic boundary conditions for the simulations. Third, the simulation boundary is moved from the first-wall location to the Vacuum Vessel (VV), which extends the boundary past the location of the internal REMC. To remove the 3D coil from the simulation domain, an equivalent set of 3D fields is calculated at the VV boundary that produce approximately the same field distribution at the last closed flux surface assuming vacuum between the two. Each of these three boundary condition improvements leads to an improvement in the predicted performance of the REMC for both machines. The resistive wall alone primarily effects the resonance of the coil with the plasma after the TQ, affecting the q-profile evolution in the SPARC modeling, and allowing the applied spectrum to be modified in response to the plasma in the DIII-D modeling. The movement of the simulation boundary has the most significant effect on the RE confinement overall, including in the early stages, particularly for a DIII-D inner wall limited equilibrium, where the RE loss fraction increases from 90% to > 99%, with SPARC RE losses also occurring much earlier when the boundary is placed at the VV.

The SPARC Toroidal Field Model Coil (TFMC) Program was a three-year effort between 2018 and 2021 that developed novel Rare Earth Barium Copper Oxide (REBCO) superconductor technologies and then successfully utilized these technologies to design, build, and test a first-in-class, high-field (∼20 T), representative-scale (∼3 m) superconducting toroidal field (TF) coil. The program was executed jointly by the MIT Plasma Science and Fusion Center (PSFC) and Commonwealth Fusion Systems (CFS) as a technology enabler of the superconducting high-field pathway to fusion energy, and, in particular, as a risk retirement program for the no insulation (NI) TF magnet in the SPARC net-energy fusion tokamak. The TFMC achieved its programmatic goal of experimentally demonstrating a large-scale high-field REBCO magnet, achieving 20.1 T peak field-on-conductor with 40.5 kA of terminal current, 815 kN/m of Lorentz loading on the REBCO stacks, and almost 1 GPa of mechanical stress accommodated by the structural case. Fifteen internal demountable pancake-to-pancake joints operated in the 0.5 to 2.0 nΩ range at 20 K and in magnetic fields up to 12 T. The DC and AC electromagnetic performance of the magnet predicted by new advances in high-fidelity computational models was confirmed in two test campaigns while the parallel, single-pass, pressure-vessel style coolant scheme capable of large heat removal was validated. In the test facility, a feeder system composed of REBCO current leads and cables was experimentally qualified up to 50 kA, and a liquid-free cryocooler-based helium cryogenic system provided 600 W of cooling power at 20 K with mass flow rates up to 70 g/s at a maximum design pressure of 2 MPa for the test campaigns. Finally, the feasibility of using passive, self-protection against a quench in a fusion-scale NI TF coil was experimentally assessed. While the TFMC was intentionally not optimized for quench resiliency – and suffered localized thermal damage in response to an intentional open-circuit quench at 31.5 kA terminal current – the extensive data and validated models that it produced represent a critical step towards this important objective.

The SPARC Toroidal Field Model Coil (TFMC) is the first large-scale (∼3 m), high-field (∼20 T) superconducting fusion magnet based on Rare Earth Yttrium Barium Copper Oxide (REBCO). Its objective was to retire risk for the toroidal field magnet in the SPARC tokamak, a burning plasma class magnetic confinement fusion energy device. Weighing 10,058 kg and utilizing 270 km of REBCO, the TFMC is a non-insulated, stack-in-plate style superconducting magnet. It has three main components: (1) the winding pack; (2) the structural case; and (3) the case extensions, or plena. The winding pack is composed of sixteen single pancakes with two termination plates top and bottom. The pancakes are Nitronic 40 radial plates machined with spiral channels on one side for the REBCO tape stack and single-pass channels on the opposite side for supercritical helium coolant. After assembly, each pancake undergoes a vacuum-pressure impregnation solder process to provide good mechanical protection of the REBCO tape stack and efficient thermal and electrical connectivity within each pancake. The pancakes are bolted along the inner and outer perimeter to provide mechanical and thermal connectivity while inter-pancake joints provide low resistance current transfer between pancakes. The top and bottom termination plates facilitate electrical connection to a superconducting feeder system. Embedded throughout the winding pack are 211 voltage taps, 34 temperature sensors, 34 helium flow monitors, 4 Hall probes, and 4 resistive surface heaters. The winding pack is contained within a structural case, a “trough and lid” style design composed of two Nitronic 50 forgings machined to shape and bolted together. The case reacts the large electromechanical stresses approaching 1 GPa during operation and serves as a pressure vessel that enables 20 bar supercritical helium flow that cools the winding pack and case. Two case extensions or “plena” are attached to the case with unique high-pressure feedthroughs to provide winding pack access for current, cooling, and instrumentation, completing the magnet assembly.