Sandia National Laboratories Overview of Fusion at Sandia National Laboratories Keith Matzen Pulsed Power Sciences Center, Sandia National Laboratories in collaboration with many colleagues Fusion Power Associates Annual Meeting and Symposium December 2, 2009 Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. • Consider a mass of DT with radius R, density ρ, and temperature T • How does the disassembly time compare with the time for thermonuclear burn? τ_disassembly ~ R / c_s ~ R / √T τ_burn ~ 1 / (n_i ⟨σv⟩) ~ 1 / (ρ ⟨σv⟩) • The fractional burn up of the DT (for small burn up) is: f_burn ≈ τ_disassembly / τ_burn ~ ρR (⟨σv⟩ / √T) • At sufficiently high ρR and T the fractional burn up becomes significant and the energy deposited by alpha particles greatly exceeds the initial energy in the fusion fuel ("ignition") • Typical conditions are: ρR ≈ 0.6 g/cm² T ≈ 5 keV For ignition conditions: ρR ≈ 0.6 g/cm² T ≈ 5 keV ρRT ≈ 3.0 (g keV / cm²) P(Bar) = 8 • 10⁸ ρ(g/cm³) T_i(keV) PR ~ 2.4 • 10⁹ Bar - cm E ~ 3/2 PV ~ 3/2 P (4π/3 R³) ~ 1.5 • 10⁹ R²(cm)(J) E_NIF ~ 15 kJ ⇒ R ~ 30 µm ⇒ P ~ 800 GBar and ρ ~ 200 g/cm³ τ_conf ~ R / c_s ~ 30 ps Power ~ E / τ_conf ~ 0.5 • 10¹⁵ W Note for magnetic confinement fusion ignition: τ_conf ~ few seconds P ~ few Bars ρ ~ few 10⁻¹⁰ g/cm³ In either direct or indirect drive, peak drive pressures are of order ~ 50-150 MBars We need to get pressures to >1000X that for ignition Spherical implosions enable us to store energy in the fusion fuel in the form of kinetic energy, which is converted to pressure at stagnation P_stag ~ α ρ_stag^(5/3) α ρ_stag^(2/3) ~ v² ⇒ P_stag ~ v⁵ / α^(3/2) α ≡ P / P_Fermi Thin shell implosions can reach the 200-400 km/sec needed for ICF ∫ P_drive dV = 1/2 m v² m ~ 4π R² ρ δR P_drive R³ ~ R² ρ δR v² ⇒ v² ~ (P_drive / ρ) (R / δR) Double z-pinch hohlraum fusion concept (R. A. Vesey, M. C. Herrmann, R. W. Lemke et al., Phys. Plasmas 2007) • primary hohlraum w/ z-pinch • secondary hohlraum w/ capsule • symmetry control foams High yield capsule design: • DT gas 2180 µm (0.3 mg/cm³) • solid DT 280 µm • Be (0.2% Cu) 190 µm Fuel density at ignition: • 1D capsule yield: 520 MJ • 2D integrated yield: 470 MJ • Two Z-pinches, each with 9 MJ x-ray output • Symmetry control to 1% via geometry, shields • Capsule absorbs 1.2 MJ, yields 400-500 MJ • Power required (1 PW/pinch @ 20-mm-diam.) • Energy required (8-9 MJ/pinch) • Facility scale: 300 ft Because of the inefficiencies in this concept, only 0.04% of the driver energy gets to the fusion fuel. Are there more efficient concepts? Is there any way to lower the required pressure? Comparison of efficiency vs. risk (~ 1/maturity): • Double-ended hohlraum: η ~ 0.04% • Magnetized Liner Inertial Fusion: η ~ 0.5 to 1% • Pulsed power can flexibly drive many target types • Direct fuel compression and heating with the magnetic field could be greater than 20X more efficient Direct magnetically driven implosions could be over an order of magnitude more efficient than indirect radiation driven implosions. Natural geometry is cylindrical: • reduced volume compression (ρr and T_ig difficult) • implosion velocity is slow: V_imp ~ 12 cm/µs for instability-robust liners Fuel magnetizing and preheating is a potential solution: • the attainment of ignition conditions with slow implosions and modest radial convergence Magnetically-Driven Cylindrical Implosion: P = B² / (2µ₀) = 140 ( (I_MA / 30) / R_mm )² MBar 140 MBar is generated by 300 eV radiation drive. Graph shows current profiles for V = 75 kV and V = 95 kV reaching up to ~27 MA on a ~100-150 ns timescale. 1. A 10-50T axial magnetic field is applied to inhibit thermal conduction and enhance alpha particle deposition before the implosion begins (Metal/beryllium Cylindrical Liner with cold deuterium/tritium gas fuel). 2. Z Beamlet can preheat the fuel to ~100 - 1000 eV to reduce the required compression needed. 3. The Z accelerator can provide the drive current which generates an azimuthal drive field (pressure) to efficiently implode the liner (Z pinch) at 50-100 km/sec and compress the axial field by factors of 1000. Simulations indicate scientific breakeven (fusion energy out = energy deposited in fusion fuel) may be possible on Z. Ref: Basko et al. Nuc. Fusion (2000) • The ρr needed for ignition can be significantly reduced by the presence of a strong magnetic field which inhibits electron conduction and provides confinement of alpha particles. • Lower ρr means lower densities are needed (10⁻³ - 1 g/cc). • Pressure required for ignition can be significantly reduced to ~5 Gbar (<< 500 Gbar for hotspot ignition). • Large values of B/ρ are needed and therefore large values of B are needed. Graph: Yield Y (MJ/cm) vs. Current (MA) showing significant yield increase with drive current and alpha heating. Liner parameters: • Aspect Ratio, R₀/ΔR = 6 • Convergence Ratio (CR) = 20 • Initial B = 30 T • Preheat temp ~ 250 eV • Initial fuel density: 2 - 5 mg/cc Beryllium liner: • Aspect Ratio, R₀/ΔR = 6 • 60 nm surface roughness • 80 µm waves are resolved • Yield ~ 70% 1D Compressed axial magnetic field has a stabilizing effect on magneto-Rayleigh-Taylor instability growth. Parameters: Convergence ratio = 20, B-field = 30 Tesla, Maximum current = 30 MA. • In the absence of instability the liner yield would increase with aspect ratio (1D curve). • The Magneto-Rayleigh-Taylor (MRT) instability has an increasingly strong degrading effect on the yield as the aspect ratio is increased (2D curve peak around aspect ratio ~ 6-8). Approaches across institutions: • Max Planck / ITEP: Heavy Ion Beam Driver (Basko, Kemp, Meyer-ter-Vehn) • Los Alamos / Air Force Research Lab: Field Reversed Configuration, Shiva Star generator (~20 µs, 0.5 cm/µs liner implosion; Taccetti et al., Degnan et al.) • U. Rochester LLE: Direct drive laser implosion of cylinders -- shock pre-heating, high implosion velocity (Gotchev et al.) • Sandia National Laboratories: Magnetized Liner Inertial Fusion (MagLIF) -- laser preheated magnetized fuel, LASNEX simulations indicate interesting yields (Slutz et al.) We are using Lasnex to simulate MagLIF: well benchmarked, radiation hydrodynamics, includes the effect of B on alphas. Preliminary point design parameters: • Beryllium liner R₀: 2.7 mm • Liner length: 5.0 mm • Aspect Ratio R₀/ΔR: 6 • Initial fuel density: 0.003 g/cc • Final fuel density <on axis>: 0.5 g/cc • Preheat temperature: 250 eV • Peak central averaged T_ion: 8 keV • Initial B-field: 30 Tesla • Final peak B-field: 13500 Tesla • Peak current: 27 MA • 1D Yield: 500 kJ • Convergence Ratio: 23 • Peak Pressure: 3 Gbars Components being integrated: • 2D simulation of liner stability (benchmarked on Z with radiograph data) • Laser ray-trace energy deposition in 2D with applied B_z fields • 2D transport of poloidal fields (B_r, B_z) in imploding liner system • Fusion burn in magnetized fuel We are building the integrated simulations needed to find self-consistent design solutions, e.g. balancing the requirements of laser heating physics with the desired preheat level for a desired implosion history and final fuel condition. Al liner target with initial perturbations: • λ = 400 µm, A = 20 µm • λ = 200 µm, A = 10 µm Analysis includes: • X-ray radiographs at 6.151 keV of Al liner. • Comparison of numerical simulations and measured amplitude for λ = 400 µm perturbation showing strong agreement. Both 1D scaling and 2D stability simulations indicate MagLIF could be an interesting path toward fusion: • Both magnetization and fuel preheat are necessary. • We propose laser preheating of the DT fuel with the Z-Beamlet laser. • Magnetized liners are expected to be robust to anomalous transport, since ωτ is modest. • 2D simulations indicate that low aspect ratio liners (5-10) are robust to the MRT instability. • The fusion yield is relatively insensitive to mixing of the liner material into the fuel (low Z liner). Sandia National Laboratories MFE Scope: • Plasma edge, plasma wall interactions, tritium retention and permeation • PFC design & development; modeling, high heat flux tests, joining, fabrication (ITER first wall, NSTX liquid lithium divertor, He-cooled refractory PFCs) • Plasma Materials Test Facility Organizational groups: • California Site / Physical & Eng. Sci. / Analytical Mat. Sci. (Dean Buchenauer, Mgr): PSI experiments, DiMES collaborations, joining metallurgy • Science & Technology / Phys., Chem. & Nano Sci. / Rad.-Solid Int. (Barney Doyle, Mgr): PSI experiments, PSI collaborations, DIII-D Edge Probes (La Jolla) • Science & Technology / Pulsed Power Center / Fusion Technology (Richard Nygren, Mgr): ITER FW R&D/Design, NSTX Liquid Li Divertor, W armor, He-cooled PFCs Collaborations include: Livermore, La Jolla, Albuquerque, NSTX PPPL, IP-ORNL, IPO/CEA France. Collaborations: JET, TEXTOR, Tore Supra, JT-60, LHD, KSTAR, DIII-D, C-MOD, TFTR, PISCES. DIII-D (General Atomics): • Sandia edge probe array • ELM control studies (Jon Watkins) NSTX (PPPL): • Li jet experiments, B field like NSTX divertor • Measurements of deposited Li (Bill Wampler) • Liquid Lithium Divertor plates & heater control (Fabrication 2008-9, Installation photo Nov 2009) Key activities: • Sandia tested Be, C, W (PFC options) • Sandia/Boeing built divertor cassette • ITER Design Reviews • US Technical lead (Mike Ulrickson) Technical Areas: • Electromagnetic analysis: forces vs. time in vertical upward disruption • Thermal-hydraulic analysis: pioneering work on coolant flow and heat transfer (Hypervaportron model / FLUENT) that established reference calculations for ITER • High heat flux tests: Plasma Materials Test Facility EB1200 Electron Beam, IR thermograph (12,000 cycles, first wall quality mockups from Japan, Russia, China & Korea) • Joining R&D: CuCrZr/316SS joint showed deleterious BCC phase formation • Thermal & stress analyses • Z facility: - Z: 26 MA in 100 to 600 ns risetime - Z-Beamlet: multi-kJ in few ns - Z-Petawatt: kJ in ps - Sophisticated diagnostics - Routinely operating at 1 shot per day • MagLIF – Magnetized Liner Inertial Fusion: - Utilize axial magnetic field and laser preheat to significantly reduce requirements for fusion ignition (P, ρR) - Greater than an order of magnitude increase in efficiency of coupling driver energy to fusion fuel