Magnetic Fusion Technology

Nuclear fusion would have abundant, cheap fuel (deuterium and lithium), excellent safety, and environmental compatibility. A fusion reactor would need to heat the deuterium–tritium fuel to 10 keV (100 Million Kelvin) and confine it long enough for about 1 % of the fuel to “burn”. This can be done by using intense magnetic fields to confine the plasma electrons and ions and to provide thermal insulation between the hot plasma and the walls. Experimental “tokamaks” and “stellarators” are confining plasmas well on a small scale (plasma radius about 1 m), and a larger ITER experiment is under construction. A Demonstration Power Plant (DEMO) to generate electricity would be the next step after ITER. The final challenge will be to produce electricity that is economically competitive with other sources.

Many technologies are needed to build a fusion power plant: vacuum systems, magnet systems, plasma stability control, superconductors, radio waves, microwaves, lasers, accelerators, structural support, cryogenics systems, cryogenic pellet injectors, plasma power control, high-temperature materials that can survive radiation damage, radiation shielding, tritium control, 3D computer aided design systems, safety analyses, and diagnostic instruments. This chapter provides a brief introduction to these topics, which are discussed in the rest of the book.

Energy stored in capacitors (or inductances) can be pulsed through magnet coils to generate high magnetic fields to confine high-temperature plasma “pinches”, but such plasmas do not last long. Water-cooled copper coils can sustain magnetic fields for long times, but large coils consume hundreds of MW electrical power, so fusion reactors will probably need to use superconducting magnets. This chapter describes the methods used for calculating magnetic fields, RLC circuits, power consumed, cooling water requirements, and coil design.

Superconducting magnets are widely used in medicine, accelerators, industry, science, and fusion research. Superconducting magnets consume power mainly for refrigeration to keep them near liquid helium temperature (~4 K). To remain superconducting the wire must be kept within limiting values of temperature, magnetic field, and current density. Wires are stabilized using tiny filaments surrounded by conducting metal (such as Cu, Al, or Ag) to prevent sudden loss of superconductivity (quench). They are protected by circuitry to dump the stored energy externally in case a quench occurs. The widely used NbTi and Nb₃Sn superconductors are being augmented by less expensive MgB₂ and by high-temperature superconductors, such as REBCO, which could achieve higher fields or simplify the cryogenic requirements.

We must heat the plasma to about 10 keV to initiate fusion reactions, and, in a tokamak, provide a means to sustain the plasma current. Heating is done by the plasma current (ohmic heating) and with auxiliary heating using radio waves, microwaves, and high-energy neutral beam injection. Fusion product alpha particles also heat the plasma as they slow down. If alpha heating is powerful enough, the external auxiliary sources may be turned off (except if required for current drive in tokamaks). This chapter describes these heating and current drive systems, the hardware that they require, the current achievements, and plans for ITER.

The first wall, blanket, and shield must withstand a high heat flux, remove hundreds of MW of heat, minimize impurity flow into the plasma, survive neutron and gamma radiation damage, breed tritium fuel, maintain a low tritium inventory, sustain large temperature differences, support high stress levels, avoid safety hazards, shield the coils and external environment, avoid neutron streaming through ducts, be reliable, and be maintainable. High heat flux components, such as helium-cooled tungsten armor tiles, are being developed. Reduced activation ferritic/martensitic (RAFM) steel is the preferred structural material, and silicon carbide may be developed in the future for higher temperature operation. To achieve a satisfactory tritium breeding ratio it may be necessary to use enriched ⁶Li or a beryllium neutron multiplier. Monte Carlo computer codes can simulate 3D models of neutron and gamma transport, tritium breeding, shielding and radioisotope production. Several blanket configurations with cooling by helium, by PbLi liquid metal, or by molten salt, are being developed, and some of them will be tested in ITER tritium breeding modules. Rankine or Brayton cycle heat engines may be used to generate electricity with efficiency ≳40 %, and fusion energy may also be used for other applications, such as hydrogen production.

In addition to controlling the plasma parameters, we must control particle flow and power flow. Particle flow includes supply of DT fuel to the plasma core, removal of helium “ash”, and dealing with impurities from the walls, to maintain fuel density and avoid excessive radiation losses. Large power flows must be controlled and accommodated by the walls, limiter, and divertor; and plasma instabilities may concentrate high powers in small areas, causing material failures. Several tokamak divertor concepts are being modeled by computers and tested experimentally, including single null, double null, super X, and snowflake configurations, cooled by flowing helium in “T-tubes” or “fingers”. Liquid lithium wall coatings appear to benefit power and particle control in some experiments.

Long-lived materials are essential for economical fusion power. Their lifetimes may be affected by temperature, stress (gravity, thermal expansion, atmospheric pressure, coolant pressure, swelling), transmutation, embrittlement, creep, fatigue, sputtering, and corrosion. No material satisfies all the criteria, so compromises are required. Improved alloys (such as RAFM) and ceramics (such as SiC) are being developed to serve reliably in fusion reactors. A high-flux neutron source is urgently needed to test these materials, and the International Fusion Materials Irradiation Facility (IFMIF) will begin to meet this need.

Vacuum systems are required to remove air and to control the pressure of fuel gases in fusion reactors. The gas flow may be viscous (where collisions between molecules are dominant) or molecular (where collisions with the walls are dominant). The gas flow equations are similar to those for current flow in an electrical circuit. Vacuum pumping is done by reciprocating mechanical pumps, turbomolecular pumps (similar to gas turbines), jet pumps, cryosorption pumps, sublimation pumps, and cryogenic pumps. Pressure is measured by mechanical gages, manometers, thermocouple gages, and ionization gages. Vacuum chambers should be made of materials with low outgassing, and should be thoroughly polished, cleaned, and baked. ITER will use large turbomolecular and cryogenic pumps to remove helium and to maintain low pressures in the neutral beam injectors.

Fusion reactors need liquid nitrogen and liquid helium for the superconducting magnets and for vacuum pumping. Materials properties at low temperatures must be taken into account in designing cryogenic systems. Cryogenic systems for refrigeration and liquefaction are well developed, but complex and expensive. Thermal insulation is provided by multilayer reflective metallized plastic films in vacuum. The ITER cryogenic system will provide an average cooling power of 65 kW at 4.5 K, and liquid helium will be stored in a 120 m³ dewar.

Diagnostic instruments are needed to provide measurements for machine protection (magnet systems, vacuum systems, heating and cooling systems, safety systems, etc.); for plasma control; and for plasma performance evaluation. Plasma diagnostic systems include electrostatic probes, magnetic probes, measurement of particles emitted from the plasma, spectroscopy of waves emitted by the plasma, probing the plasma with injected particle beams, and probing the plasma with waves, such as microwaves and laser beams. For ITER, the next major step in magnetic fusion, diagnostics must be hardened to survive in a hot, high-radiation environment, and designed to minimize neutron streaming through ducts. Many adverse effects must be mitigated, such as radiation-induced spurious signals, erosion of windows and mirrors, and deposition of films on them. This is new territory for diagnostics and many challenges have to be overcome.

Tritium and tokamak dust are the main radioactive hazards of fusion reactors. Tritium emits a low-energy beta ray with a half-life of 12.3 years. It is hazardous if inhaled or ingested, but cannot penetrate the skin. The tritium inventory in the fuel system and walls should be well contained, minimized, and closely monitored, to keep the source term low in case of an accident. Neutron absorption will make reactor internal components radioactive, so their radioactivities will be minimized by design, with a goal of clearance or recycling most materials after a cooling period of 50–100 years. If many superconducting cables and coils are used in industry and in fusion reactors, shortages of materials such as He and Nb may occur. The ITER safety team is analyzing dozens of potential accident scenarios to prevent them or to mitigate their consequences, so that public safety will be assured without the need for an evacuation plan.

The electrical power companies have specified what features are desired for attractive power plants, with regard to economics, regulatory simplicity, and public acceptance. The plants should achieve high availability, and should have maintenance procedures that can be performed in a few months, which is very difficult for large fusion reactors. A company must borrow money to build a power plant long before it starts earning revenue, so short construction times are important, and the total capital cost is often about twice as high as the direct capital cost. Large fusion power plants (3 GWe vs. 1 GWe) have an economy of scale that reduces the cost of electricity by about 20–30 %, but grid perturbation during shutdown would be a problem. Fusion power plant design studies in Europe, Japan, China, and the USA have estimated the cost of fusion-power electricity to be higher than from fission power and fossil fuels, but fusion could become competitive under several possible scenarios.

A “Fusion-Fission Hybrid” is a fusion reactor that contains thorium, uranium or transuranic elements in its blanket. Fusion power plants usually require Q (fusion power/input power) values >10 to be economical. A hybrid could operate with Q ≲ 5, so it would be easier to build. Hybrids could be optimized for maximum power output, for fissile fuel breeding, or for incineration of radioactive wastes from fission power plants. A variety of reactors have been proposed, with attention to safety, radwaste, and nuclear nonproliferation issues. In a secure “Energy Park” one hybrid could provide fissile fuel for 6 satellite fission reactors of today’s generation or over 20 next generation reactors.