Session TO6: ICF: Magneto-Inertial Fusion

A conservative approach to scaling magneto-inertial fusion concepts to larger drivers

We propose a conservative approach to scaling magneto-inertial fusion (MIF) platforms to larger drivers. Using simple physics models capturing the essential elements of MIF implosions, we identify a set of dimensionless parameters whose conservation leads to self-similar implosions at all energy scales. Our approach establishes basic scaling criteria to map present-day implosions to higher energies while preserving or lessening the impact of MHD implosion instabilities and fuel energy losses (e.g., conduction, radiation, bulk flows) while steadily improving thermonuclear yield and ignition performance metrics. The same scaling rules can provide a unique mapping of an aspirational scaled-up experimental design to a platform that could be tested on present-day facilities. Our scaling strategies maintain flexibility with respect to future pulsed-power architectures by allowing arbitrary scaling of the current rise time. In summary, our analytic scaling studies provide a starting point for MIF target designers to scale targets up or down in coupled energy while maintaining as much of the underlying physics as possible, controlling and/or mitigating the impact of known physics risk factors, and shifting risk toward areas where present-day designs have exhibited sufficient margin.   

\textbf{3D HYDRA Simulations of High Performance MagLIF Experiments}

Post-shot 3D HYDRA simulations of high performing MagLIF experiments have been performed. The simulations incorporate the measured axial magnetic field strength, VISAR unfolded load current, and laser energy and spot profile. Stagnation quantities are generally in good agreement with experimental measurements, including the DD neutron yield, ion temperature, Bz*r\textunderscore stag, and liner rho*r. The simulated convergence ratio and burn averaged pressure are higher. Additional simulations with larger initial perturbations indicate the discrepancy results from the amount of magneto-Rayleigh-Taylor instability (MRTI) that develops. Simulations forced to run with very little MRTI also show that the laser deposition alone produces axial and azimuthal non-uniformity in the stagnation structure. The azimuthal asymmetry is found to be a result of non-uniform laser deposition and subsequent vorticity amplification during compression. Axial structure results from the laser driven pressure gradients combined with end loss flows. Higher frequency features in the stagnation are tied to MRTI and cause premature loss of thermal insulation.   

Learning the Link Between MagLIF Performance and Stagnation Morphology

Magnetized Liner Inertial Fusion (MagLIF) performance (measured by neutron yield) for nominally identical shots may differ by as much as an order of magnitude. This must mean that experimentally tuned parameters are not effectively controlling one or more important underlying physical quantities. In order to investigate this, we utilize a machine learning based approach to uncover the relationship between stagnation morphology and neutron yield. The morphology of a stagnated plasma produced in MagLIF is imaged using self-emission x-rays from the multi-keV plasma. By utilizing a linear regression based on the Mallat Scattering Transform of the stagnation image, we have successfully extracted morphological parameters with uncertainty at stagnation. The inferred morphological parameters and associated uncertainties are utilized to augment sparse MagLIF shot data in order to aid in learning the relationship between stagnation morphology and shot performance (neutron yield). This relationship is successfully learned using a generalized regression neural network to within an error of less than 10\%. By investigating this relationship, we search for latent physical quantities which if better controlled would yield greater repeatability.   

Magnetically-assisted ignition project on the National Ignition Facility

We are planning a project to apply an external magnetic field to a high performing DT-layered implosion on the NIF. The magnetic field can reduce hot-spot losses through reduced electron thermal conduction and increased alpha particle confinement. In addition, it may suppress mix instabilities. Two dimensional simulations show that applying a $\ge$ 30 T field to an experiment already performed on NIF can lead to hot-spot conditions significantly in the self-heating regime. Magnetizing an indirect drive target requires development of specialized hohlraums with high electrical resistivity while maintaining good x-ray conversion of laser drive. This talk will review the key physics goals and project challenges and outline the research needed to perform a test which quantifies the magnetization improvement in the hot-spot conditions of a DT-layered implosion.   

NIF Hohlraum Modeling for Magnetically-Assisted Ignition

LLNL is exploring imposed magnetic fields to improve ICF performance toward ignition on NIF [L. J. Perkins et al., PoP 2017]. For current implosion experiments, an imposed axial field of 30-50 Tesla is shown in simulations to significantly reduce hotspot energy loss, due to electron conduction perpendicular to the field, and increase alpha-particle energy deposition. Here, we present radiation-magneto-hydrodynamic simulations with the Hydra code that study the effect of the imposed field on hohlraum dynamics. We expect modest effects, even though the hohlraum fill gas is well magnetized. Our short-term experimental focus is on room-temperature NIF experiments with a gas-filled capsule, based on the BigFoot campaign [C. A. Thomas, APS-DPP 2017]. Future experiments are planned using magnetized cryogenic targets. The goal is to demonstrate field compression in the capsule comparable to the MHD frozen-in law, and increased hotspot temperature and fusion yield. A significant challenge is having the field imposed by external coils diffuse through the hohlraum. Materials with lower electrical conductivity than gold are being considered, including pure uranium and gold alloys - which may have improved radiation drive due to the “cocktail "effect" [O. S. Jones, PoP 2007].   

Novel MHD features in indirect- and direct-drive magnetized ICF implosions

Imposing a 30 – 50 T B-field makes the requirements for ignition less stringent.1,2 We present several novel MHD features in magnetized implosions. (1) During the implosion, the compressed B-field in the shell diffuses rapidly into the inner gas region. The increased field is then frozen in after shock heating. Consequently, magnetic diffusion enhances the central field in the assembled configuration. (2) The B-field suppresses electron heat conduction across the field. Therefore, the ablation of the inner material surface is more pronounced in the polar regions than at the waist, resulting in a “pancake” shape in the hotspot Ti and Te contours while the x-ray contours are “sausage”. (3) During the pulse power ramp-up of the B-field, the induced E field is below the threshold for breakdown in the gas region due to high neutral collisionality. (4) We also show yield enhancement by magnetization on an indirect- and a direct-drive NIF implosion. 1. L. J. Perkins et al., PoP, 24 062708 (2017). 2. D. Ho et al., APS-DPP (2016).   

Demonstration of modified laser propagation in magnetized gas pipe experiments at the NIF

Recent MagLIF gas pipe experiments at the NIF have demonstrated modification to the axial laser propagation characteristics in the presence of an externally applied axial magnetic field. The field is supplied by the new MagNIF pulsed power system, allowing the NIF to deliver up to 30 T fields in \textasciitilde cm$^{\mathrm{3}}$ volumes. Initial measurements of plasma x-ray emission perpendicular to the gas pipe axis show that with a 12 T field, the \textasciitilde 30 kJ laser pulse propagates through the 10 mm-long, 1 atm neopentane-filled gas pipe \textasciitilde 2 ns more quickly than without a magnetic field. The morphology of the emission profile is also modified with the magnetic field, showing the hot column to be more cylindrical (than conical) with the magnetic field. These two effects both suggest that the field is reducing thermal transport in the plasma and increasing the electron temperature near the gas pipe axis, and the results are compared with Hydra simulations of the same conditions. Future experiments will measure electron temperature and density profiles with and without magnetic fields, as well as increase the field strength and the plasma density. This work was performed under the~auspices of the U.S. Department of Energy by LLNL under Contract DE-AC52-07NA27344.   

Rotation-Aided Confinement in Magneto-Inertial Fusion Schemes

We consider a standard MagLIF target, where instead of a gas-prefill that anticipates being preheated by the Z-beamlet laser, its mass is contained in a cryo-DT fiber on axis. A 1 MA pulsed power device can be used to turn the fiber into a hot plasma, which would then fill up the capsule, just before it is imploded using a 25 MA driver, such as the Z- Machine. In such a scenario, an axial $B_{z}$ would trigger an azimuthal rotation, $v_{\phi}$. One can analytically illustrate that for a given uniform $B_{z}$, and a $B_{\phi}(r,z)$, the $z$-dependence introduced by the additional driver, will yield a nontrivial $J_{r}$ that would spin the plasma via the $\vec{J}\times\vec{B}$-force. Here we investigate how much rotation is to be expected in such a target, and to what extent it may improve confinement performance over laser-driven preheating. We carry out this investigation using the multi-physics MHD AMR code, FLASH.   

Laser Gate Experiment for Increasing Preheat Energy Coupling Efficiency in Magnetized Liner Inertial Fusion (MagLIF)*

In Magnetized Liner Inertial Fusion (MagLIF), a laser pulse preheats fuel inside of a cylindrical metal tube. The tube is imploded, which compresses and heats the fuel to fusion relevant temperatures. Currently, the laser ablates through a laser entrance window (LEW) that contains the fuel. To reduce absorption energy loss in the LEW, radiative loss from fuel-window mix, and laser plasma instabilities, the LEW could be removed before the laser reaches the LEW [1]. This concept is referred to as ``Laser Gate.'' In our version of Laser Gate, we weaken the LEW by driving current through a wire wrapped around the LEW perimeter. The target pressure opens the weakened window up and out of the laser path. We have imaged the LEW opening out of the laser path. Repeatability and scaling studies are ongoing. Additionally, we are integrating our setup with facilities at Sandia National Labs. [1] S.A. Slutz, et al., Phys. Plasmas 24, 012704 (2017);   

Establishing the physics design basis for dynamic screw pinch implosions on the Z Facility

Magnetically imploding cylindrical metallic shells (liners) containing preheated, premagnetized fusion fuel has proven effective at producing thermonuclear plasma conditions [1] but suffers from magneto-Rayleigh-Taylor instabilities (MRTI) that limit the attainable density, temperature, and pressure in the fuel. A novel method proposed by Schmit et al. [2] uses a helical magnetic drive field that dynamically rotates during implosion, reducing (linear) MRTI growth via a solid liner dynamic screw pinch (SLDSP) effect. Our work explores the design features necessary for successful experimental implementation of this concept. SLDSP uses a helical drive field at the liner outer surface generated via helical return-current posts, resulting in enhanced average magnetic pressure per unit drive current, mild spatial non-uniformities in the magnetic drive pressure, and augmented static initial inductance in the pulsed-power drive circuit. These topics have been investigated using transient magnetic and magnetohydrodynamic simulations and the results have led to a credible design space for dynamic screw pinch experiments on the Z Facility. [1] Gomez et al., Phys. Rev. Lett. 113, 155003 (2014). [2] Schmit et al. Phys. Rev. Lett. 117, 205001 (2016).   

Overview and Status of the new HJ1 Coaxial Plasma Gun for PLX-$\alpha$

An overview of the HJ1 coaxial plasma gun for the PLX experiment at LANL is presented, highlighting upgrades over the previous Alpha2guns and the current status of gun testing. Improvements include a faster, more robust high mass gas valve, field distortion sparkgap switches with smaller jitter, a new pre-ionization system using a self-switching glow discharge, and a re-engineered more compact 7.5kJ pfn structure. A total of 36 guns are planned, to be delivered in two groups of 18. As of July 2019, the first 18 guns have been assembled, with the first 6 HJ1 guns installed on PLX and undergoing test. The next 12 are being prepared for shipping during July/August. The 2nd set of 18 guns are mostly manufactured, with assembly and delivery scheduled for Fall 2019. Gas valve and switch lifetime testing continues, along with experimental parameter scans designed to establish the optimal operating configuration. Best repeatable shots to date are about 1 mg argon plasma, up to 67 km/s, jet lengths 10-30 cm, at $2 \times 10^{16}$ cm$^{-3}$. [1] Hsu et al., IEEE Trans. Plasma Sci.~{\bf 40}, 1287 (2012). [2] Y. C. F. Thio et al., Fus. Sci. Tech., accepted (2019); https://doi.org/ 10.1080/15361055.2019.1598736.   

Progress Toward the Formation of Fully Spherical Imploding Plasma Liners on PLX

The Plasma Liner Experiment (PLX) at LANL is being upgraded to form spherically imploding supersonic plasma liners from the merger of up to 36 individual plasma jets. Previous work has studied the individual merging of multiple supersonic plasma jets [1,2], and has shown that the resulting structure may exhibit shock formation or a diffuse interpenetration, depending strongly on the relative velocity of the jets and the plasma ionization state. These experiments will investigate the liner structure beyond the initial merging phase, through the convergence and stagnation of the formed plasma liner. Diagnostic measurements will characterize the evolution of the plasma structure, density profile, temperature, ionization state, and ram/stagnation pressure, with application to the use of plasma liners as a standoff driver for magneto-inertial fusion. Measurements results will be used to benchmark fluid simulation predictions and to understand scaling to increased liner energies.\\ $^1$ A. Moser et al., Physics of Plasmas 22.5 (2015): 055707.\\ $^2$ S. J. Langendorf et al., Physical review letters 121 (18), 185001.\\   

High resolution studies of MHD instabilities in magnetic anvil cells

Magnetic anvil cells allow to reach relatively high material pressures (several Mbar) using modest pulsed-power drivers (1MA), providing that the current rise time is shorter than 300 ns and the compression is axially symmetric. The work presented here explores the degree of symmetry obtained from the isentropic compression of a uniform aluminum sample using the 3D extended MHD code PERSEUS. Simulations show that instabilities have little impact on the symmetry of the sample, providing that the sample is heavy enough to start with. While this geometry generates density and temperature gradients during compression, Abel inversion techniques can be used to recover the local mass density and temperature under axisymmetric conditions. But what about the pressure? The cylindrical symmetry has another major advantage. The measurement of the magnetic field at the edge of the sample at full compression allows to infer directly the pressure of the warm dense matter sample on axis. In this case, the magnetic anvil cell resembles its mechanical counter-part, the diamond anvil cell, where the force applied externally on the diamond is directly related to the pressure in the diamond cell gap.   

Liner Compression of an FRC: Implosion vs Diffusion Times

To achieve the cost-minimum regime near a megagauss for a controlled fusion power reactor [1] requires repetitive operation of the liner implosion. Such operation is possible with piston-driven liquid metal liners, stabilized by rotation against growth of Rayleigh-Taylor modes. A common complaint, however, is that these implosions may be too slow to provide energy and temperature gain in compressing FRCs. The present paper compares times for liner compression and for loss by diffusion of flux, particles and energy. The condition on ideal contraction of FRC length with radius, h/r$^{\mathrm{0.4}}=$ constant, used in earlier work [1] is relaxed here for a self-consistent model. Successful compression again appears possible, even for implosion time half the loss time, which may be satisfied for pneumatically-driven, compressible liners and neo-classical transport. Issues of mirror effects in the open-field region are also discussed in regard to extrapolating experimental studies to reactor conditions. [1] P.J. Turchi, S.D. Frese, M.H. Frese, ``Stabilized Liner Compressor for Low-Cost Controlled Fusion at Megagauss Field Levels,'' IEEE Trans. on Plasma Science, 45, 10, 2800 (2017).   

Scaling of a compact multi-beam ion accelerator to higher beam power for plasma heating

Reducing the size, power, and cost of accelerators opens new opportunities in mass spectrometry, ion implantation and ultimately plasma heating for fusion. Our technology is based on wafer-based components where beam transport is in the direction of the surface normal to the wafer. This allows stacking of these units to increase beam energy while limiting the peak applied voltage to several kilovolts. The wafer-based implementation allows us to operate multiple beamlets on a single wafer in parallel for increased current per wafer compared to a single beam with one large aperture. We've demonstrated a compact multi-beam RF ion accelerator in an array of 3x3 beams, and integration of all accelerator components (matching section, focusing elements and acceleration stages) [1], and an energy gain of 2.6 KeV per gap using a near board RF driver [2]. We now report our effort of scaling to higher beam power using an array of 112 beamlets and an energy gain of up to 10 keV per acceleration gap. We will also discuss the opportunities to scale this technology further to reach hundreds of KeV beam energy. [1] Persaud, A. et al., Rev. Sci. Instrum. 88, 063304 (2017); Seidl, P. A. et al., Rev. Sci. Instrum. 89, 053302 (2018). [2] Seidl, P. et al., https://arxiv.org/abs/1809.08525 (2018)