Deep-Learning-Enabled Assessment of Magnetic Confinement in Magnetized Liner Inertial Fusion
Abstract not provided.
Publications
Numerical study of three-dimensional instability development in solid liner dynamic screw pinches
Abstract not provided.
Deep-learning-enabled Bayesian inference of fuel magnetization in magnetized liner inertial fusion
Physics of Plasmas
Fuel magnetization in magneto-inertial fusion (MIF) experiments improves charged burn product confinement, reducing requirements on fuel areal density and pressure to achieve self-heating. By elongating the path length of 1.01 MeV tritons produced in a pure deuterium fusion plasma, magnetization enhances the probability for deuterium-tritium reactions producing 11.8−17.1 MeV neutrons. Nuclear diagnostics thus enable a sensitive probe of magnetization. Characterization of magnetization, including uncertainty quantification, is crucial for understanding the physics governing target performance in MIF platforms, such as magnetized liner inertial fusion (MagLIF) experiments conducted at Sandia National Laboratories, Z-facility. We demonstrate a deep-learned surrogate of a physics-based model of nuclear measurements. A single model evaluation is reduced from CPU hours on a high-performance computing cluster down to ms on a laptop. This enables a Bayesian inference of magnetization, rigorously accounting for uncertainties from surrogate modeling and noisy nuclear measurements. The approach is validated by testing on synthetic data and comparing with a previous study. We analyze a series of MagLIF experiments systematically varying preheat, resulting in the first ever systematic experimental study of magnetic confinement properties of the fuel plasma as a function of fundamental inputs on any neutron-producing MIF platform. We demonstrate that magnetization decreases from B ∼0.5 to B MG cm as laser preheat energy deposited increases from preheat ∼460 J to E preheat ∼1.4 kJ. This trend is consistent with 2D LASNEX simulations showing Nernst advection of the magnetic field out of the hot fuel and diffusion into the target liner.
Deep-Learning-Enabled Assessment of Magnetic Confinement in Magnetized Liner Inertial Fusion
Abstract not provided.
An overview of magneto-inertial fusion on the Z Machine at Sandia National Laboratories
Abstract not provided.
EXPLORING MAGNETO-RAYLEIGH-TAYLOR INSTABILITY DEVELOPMENT IN SOLID LINER DYNAMIC SCREW PINCHES USING THREE-DIMENSIONAL MAGNETOHYDRODYNAMIC MODELING
Abstract not provided.
Magneto-Inertial Fusion: A Promising Path Toward Burning Plasmas in the Laboratory
Abstract not provided.
Increasing stability of MagLIF targets via self-similar scaling strategies
Abstract not provided.
Performance Scaling in Magnetized Liner Inertial Fusion Experiments
Physical Review Letters
We present experimental results from the first systematic study of performance scaling with drive parameters for a magnetoinertial fusion concept. In magnetized liner inertial fusion experiments, the burn-averaged ion temperature doubles to 3.1 keV and the primary deuterium-deuterium neutron yield increases by more than an order of magnitude to 1.1×1013 (2 kJ deuterium-tritium equivalent) through a simultaneous increase in the applied magnetic field (from 10.4 to 15.9 T), laser preheat energy (from 0.46 to 1.2 kJ), and current coupling (from 16 to 20 MA). Individual parametric scans of the initial magnetic field and laser preheat energy show the expected trends, demonstrating the importance of magnetic insulation and the impact of the Nernst effect for this concept. A drive-current scan shows that present experiments operate close to the point where implosion stability is a limiting factor in performance, demonstrating the need to raise fuel pressure as drive current is increased. Simulations that capture these experimental trends indicate that another order of magnitude increase in yield on the Z facility is possible with additional increases of input parameters.
Deep Learning Enabled Bayesian Inference of Fuel Magnetization in Magnetized Liner Inertial Fusion Experiments on Z
Abstract not provided.
Bayesian Analysis Techniques Part 2: Deep Learning in the Inference Loop
Abstract not provided.
Stabilization of Liner Implosions via a Dynamic Screw Pinch
Physical Review Letters
Magnetically driven implosions are susceptible to magnetohydrodynamic instabilities, including the magneto-Rayleigh-Taylor instability (MRTI). To reduce MRTI growth in solid-metal liner implosions, the use of a dynamic screw pinch (DSP) has been proposed [P. F. Schmit et al., Phys. Rev. Lett. 117, 205001 (2016)PRLTAO0031-900710.1103/PhysRevLett.117.205001]. In a DSP configuration, a helical return-current structure surrounds the liner, resulting in a helical magnetic field that drives the implosion. Here, we present the first experimental tests of a solid-metal liner implosion driven by a DSP. Using the 1-MA, 100-200-ns COBRA pulsed-power driver, we tested three DSP cases (with peak axial magnetic fields of 2 T, 14 T, and 20 T) and a standard z-pinch (SZP) case (with a straight return-current structure and thus zero axial field). The liners had an initial radius of 3.2 mm and were made from 650-nm-thick aluminum foil. Images collected during the experiments reveal that helical MRTI modes developed in the DSP cases, while nonhelical (azimuthally symmetric) MRTI modes developed in the SZP case. Additionally, the MRTI amplitudes for the 14-T and 20-T DSP cases were smaller than in the SZP case. Specifically, when the liner had imploded to half of its initial radius, the MRTI amplitudes for the SZP case and for the 14-T and 20-T DSP cases were, respectively, 1.1±0.3 mm, 0.7±0.2 mm, and 0.3±0.1 mm. Relative to the SZP, the stabilization obtained using the DSP agrees reasonably well with theoretical estimates.
Deep Learning Enabled Bayesian Inference of Fuel Magnetization in Magnetized Liner Inertial Fusion Experiments on Z
Abstract not provided.
A conservative approach to scaling magneto-inertial fusion concepts to larger pulsed-power drivers
Physics of Plasmas
The Magnetized Liner Inertial Fusion (MagLIF) experimental platform [M. R. Gomez et al., Phys. Rev. Lett. 113, 155003 (2014)] represents the most successful demonstration of magneto-inertial fusion (MIF) techniques to date in pursuit of ignition and significant fusion yields. The pressing question remains regarding how to scale MIF concepts like MagLIF to more powerful pulsed-power drivers while avoiding significant changes in physical regimes that could adversely impact performance. In this work, we propose a conservative approach for scaling general MIF implosions, including MagLIF. Underpinning our scaling approach is a theoretical framework describing the evolution of the trajectory and thickness of a thin-walled, cylindrical, current-driven shell imploding on preheated, adiabatic fuel. By imposing that scaled implosions remain self-similar, we obtain a set of scaling rules expressing key target design parameters and performance metrics as functions of the maximum driver current I max. We identify several scaling paths offering unique, complementary benefits and trade-offs in terms of physics risks and driver requirements. Remarkably, when scaling present-day experiments to higher coupled energies, these paths are predicted to preserve or reduce the majority of known performance-degrading effects, including hydrodynamic instabilities, impurity mix, fuel energy losses, and laser-plasma interactions, with notable exceptions clearly delineated. In the absence of α heating, our scaling paths exhibit neutron yield per-unit-length scaling as Y ? [I max 3, I max 4.14] and ignition parameter scaling as χ ? [I max, I max 2.14]. By considering the specific physics risks unique to each scaling path, we provide a roadmap for future investigations to evaluate different scaling options through detailed numerical studies and scaling-focused experiments on present-day facilities. Overall, these results highlight the potential of MIF as a key component of the national ignition effort.
Design of dynamic screw pinch experiments for magnetized liner inertial fusion
Physics of Plasmas
Magnetic implosion of cylindrical metallic shells (liners) is an effective method for compressing preheated, premagnetized fusion fuel to thermonuclear conditions [M. R. Gomez et al., Phys. Rev. Lett. 113, 155003 (2014)] but suffers from magneto-Rayleigh-Taylor instabilities (MRTI) that limit the attainable fuel pressure, density, and temperature. A novel method proposed by Schmit et al. [Phys. Rev. Lett. 117, 205001 (2016)] uses a helical magnetic drive field with a dynamic polarization at the outer surface of the liner during implosion, reducing (linear) MRTI growth by one to two orders of magnitude via a solid liner dynamic screw pinch (SLDSP) effect. Our work explores the design features necessary for successful experimental implementation of this concept. Whereas typical experiments employ purely azimuthal drive fields to implode initially solid liners, SLDSP experiments establish a helical drive field at the liner outer surface, resulting in enhanced average magnetic pressure per unit drive current, mild spatial nonuniformities in the magnetic drive pressure, and augmented static initial inductance in the pulsed-power drive circuit. Each of these topics has been addressed using transient magnetic and magnetohydrodynamic simulations; the results have led to a credible design space for SLDSP experiments on the Z Facility. We qualitatively assess the stabilizing effects of the SLDSP mechanism by comparing MRTI growth in a liner implosion simulation driven by an azimuthal magnetic field vs one driven with a helical magnetic field; the results indicate an apparent reduction in MRTI growth when a helical drive field is employed.
Establishing the physics design basis for dynamic screw pinch implosions on the Z Facility
Abstract not provided.
A conservative approach to scaling magneto-inertial fusion concepts to larger drivers
Abstract not provided.
Design and simulation of dynamic screw pinch experiments on the Z Facility
Abstract not provided.
Enhancing performance of magnetized liner inertial fusion at the Z facility
Physics of Plasmas
The Magnetized Liner Inertial Fusion concept (MagLIF) [Slutz et al., Phys. Plasmas 17, 056303 (2010)] is being studied on the Z facility at Sandia National Laboratories. Neutron yields greater than 1012 have been achieved with a drive current in the range of 17-18 MA and pure deuterium fuel [Gomez et al., Phys. Rev. Lett. 113, 155003 (2014)]. We show that 2D simulated yields are about twice the best yields obtained on Z and that a likely cause of this difference is the mix of material into the fuel. Mitigation strategies are presented. Previous numerical studies indicate that much larger yields (10-1000 MJ) should be possible with pulsed power machines producing larger drive currents (45-60 MA) than can be produced by the Z machine [Slutz et al., Phys. Plasmas 23, 022702 (2016)]. To test the accuracy of these 2D simulations, we present modifications to MagLIF experiments using the existing Z facility, for which 2D simulations predict a 100-fold enhancement of MagLIF fusion yields and considerable increases in burn temperatures. Experimental verification of these predictions would increase the credibility of predictions at higher drive currents.
Stagnation Morphology in Magnetized Liner Inertial Fusion Experiments
Abstract not provided.
Characterizing the degree of fuel magnetization for MagLIF using neutron diagnostics
Abstract not provided.
A more stable more uniform Magnetized Liner Inertial Fusion configuration
Abstract not provided.
A Path to Increased Performance in Magnetized Liner Inertial Fusion
Abstract not provided.
Neutron Diagnostic Development fro the Z Accelerator
Abstract not provided.
Modifying MagLIF Stagnation Conditions and Morphology by Changing Liner Initial Conditions
Abstract not provided.
Results 1–25 of 74