This invention relates to the production of commercial electrical power or process heat using fusion reactions. In particular, the invention relates to an inertial confinement fusion power plant which decouples life-limited components from plant availability. This allows provision of reliable electric power over periods substantially longer than the life of particular components used in the power plant itself.
The National Ignition Facility (NIF), the world's largest and most energetic laser system, is now operational at Lawrence Livermore National Laboratory in Livermore, Calif. One goal of operation of the NIF is to demonstrate fusion ignition for the first time in the laboratory. Initial experiments are calculated to produce yields of the order of 20 MJ from an ignited, self propagating fusion burn wave. The capability of the facility is such that yields of up to 150-200 MJ could ultimately be obtained. The NIF is designed as a research instrument, one in which single “shots” on deuterium-tritium containing targets are performed for research. A description of the NIF can be found in Moses et al, Fusion Science and Technology, Volume 60, pp 11-16 (2011) and references therein.
There is a rapidly growing need for power, and especially for clean power. At LLNL a project known as Laser Inertial-confinement Fusion Energy, (often referred to herein as “LIFE”) is working toward introduction of fusion based electric power plants into the U.S. economy before 2030, and in a pre-commercial plant format before that. LIFE technology offers a pathway for the expansion of carbon-free power around the world. It will provide clean carbon-free energy in a safe and sustainable manner without risk of nuclear proliferation.
One challenge with respect to LIFE, as well as any technology for generating electrical power to be distributed to large numbers of consumers, is reliability. Consumers expect to have extraordinarily high reliability in their electric power supply. The result is that utilities which provide that electrical power maintain their facilities to assure the required high reliability. Thus, among the challenges with respect to fusion power, is to provide mechanisms by which components requiring maintenance or replacement can be maintained or repaired without interrupting the availability of the plant to the grid.
In the technology described herein, a fusion power plant is provided with a fusion chamber into which capsules containing deuterium and tritium fuel are introduced multiple times per second. As the individual fuel capsules (or “targets”) reach the center of the chamber, banks of lasers fire on the targets, heating and compressing the fuel to create a fusion reaction. Heat from the fusion reaction is captured by coolant circulating through the chamber. This heat is then used to generate electricity.
We have developed an architecture for such a fusion power plant in which servicing of various components, e.g. the lasers, may be conducted without taking the power plant off-line. Furthermore, even if the fusion chamber itself requires servicing, and thereby requires the plant to be taken off-line, the architecture enables that operation to be performed relatively quickly by swapping in a new chamber.
The architecture of the power plant we provide assures its high availability and protects the capital investment of its owners. Decoupling of the life limited components from plant availability is achieved using a variety of techniques. For example, we provide the lasers which are used to initiate the fusion reaction in a preconfigured “box” which allows them to be kinematically mounted and hot swapped during plant operations. Thus laser reliability can be tolerated with a shorter mean-time-between-failure than other aspects of the power plant—where a multi-year requirement may be more appropriate. This solution also offers a short mean-time-to-replace and allows offsite, factory-based construction and maintenance of the high technology laser boxes. A dual neutron pinhole approach allows personnel to access the laser bay while the plant is operating, thereby simplifying the maintenance regime. By treating the fusion chamber as a line-replaceable-unit, emerging technologies or new materials can be used in the fusion chamber as they become available. This also protects the wider plant capital investment and enables the use of a spare chamber for continued power production during maintenance of the chamber.
A carousel architecture for the lasers is also enabled by the “laser-in-a-box approach.” This arrangement eliminates the need for a beam switchyard, while at the same time minimizing the extent to which the optics are required to be within the fusion environment. The single final optic required to be in the fusion environment can be replaced, allowing easier maintenance for this component.
The architecture also decouples the fusion chamber from the vacuum chamber. This removes the need for a vacuum seal in the highest threat environment of the “first wall” which receives most of the neutron bombardment of the fusion chamber. Our approach utilizes a segmented tubular design for the first wall, thereby providing efficient thermal coupling, low mechanical stress, and a high strength to weight ratio. The modular approach of the fusion chamber also decouples it from the optical system, allowing rapid removal and replacement of the blanket and first wall modules with only a need to make and break plumbing connections, not reconfigure accurate optical pathways.
Liquid lithium coolant reduces the tritium inventory in the plant and mitigates tritium confinement requirements, yet provides high thermal efficiency. By using a xenon filled fusion environment within the chamber, the first wall is protected from ion debris and x-ray heating is mitigated. This permits use of commercially available steels, yet the entire module may be replaced as a single unit when necessary. The architecture also removes the need for the entire chamber to be cleared after each target, requiring only tolerable ‘percent’ level clearing ratios.
The capability of lasers to create conditions required for inertial confinement fusion is expected to be demonstrated at NIF in 2012 or thereabouts. The fusion yield is required to be above IMJ to demonstrate the process of “ignition”, as required for establishing the physics basis of a LIFE plant. Early experiments are expected yield a ratio of fusion energy to laser energy of about 10, resulting in fusion energy yields of around 15 MJ. The first experiments to demonstrate ignition and gain use 351 nm laser light with a central hot spot ignition geometry and laser energy of around 1.5 MJ. The system described herein builds from the research on the NIF and provides an architecture for a commercial fusion power system.
Our system provides inertial confinement fusion which produces 14 million electron volt (MeV) neutrons from a deuterium-tritium (DT) fusion reaction within a fusion chamber. The fusion neutrons pass into a molten lithium cooling blanket circulating behind a “first wall” of the chamber, with the resulting heat used to drive conventional commercially available apparatus for converting that heat into electricity. The coolant will typically operate in the 500-600 degree Celsius regime. The process heat also could be used to produce hydrogen for transportation needs, to desalinize seawater, to power manufacturing processes, for oil extraction, or for other uses.
In our preferred approach, central hot spot fusion is initiated using “indirect” drive to provide the fusion reaction. Indirect drive uses energy from the lasers to heat a hohlraum within which is a capsule of the DT fusion fuel. Bombardment with high energy laser beams of the inner surface of the hohlraum causes it to emit x-rays. These x-rays compress and heat the fuel, causing fusion ignition and burn. Alternate approaches such as direct drive (no hohlraum), or fast ignition (separate compression and ignition lasers) also may be used for the fusion reaction.
Typical electric power plant availability for commercial electricity sources over the past decade is on the order of 90%, with unplanned shutdowns on the order of 3% in the US. This imposes a high bar for a technologically intensive solution such as fusion. In particular, the predictability of shutdown for gigawatt-scale power plants is important. This means that the operations must not be threatened by potential instabilities in the physics performance, or intermittent failures in the fusion technology. In this respect, inertial fusion has both benefits and challenges compared with quasi-steady-state reactors.
Herein we describe an architecture for a power plant for generating gigawatt quantities of thermal power (GWth) from a laser driven source of fusion neutrons. Our goal is to deliver commercial fusion power soon enough to make a difference to energy policy considerations for the period 2030 onwards. The approach described here provides a power plant architecture that uses the physics tested on the NIF, coupled to a driver solution using existing manufacturing technology. Our approach to plant operations also overcomes a need to wait for advanced material development. The use of a known production route for the structural materials of the plant allows rapid time-to-construction, not contingent on the development of new materials.
The separability of the sub-systems of a LIFE plant according to our architecture provides high maintainability and system availability, yet allows operational improvements based on expected advances from ongoing physics, materials and system studies. Our architecture allows trade-off decisions to be made on the wide array of possible development and risk reduction activities. It enables selection of options to enable competitive electricity production. Our architecture addresses the cost of electricity, rate and cost of capital build, licensing simplicity, reliability, availability, maintenance and inspection, predictable shutdown, quick restart after shutdown, protection of capital investment, ability to meet urban environmental and safety standards, acceptability to the public, and timely delivery.
Various criteria drive fundamental design choices in the overall power plant architecture, subsystem configurations, and acceptability of certain technology options. For example, as described below, the impact of designing a laser sub-system that can be maintained while the plant remains operational is of significant importance. This allows the overall plant availability to remain high, even if the reliability of a sub-system is relatively low.
The repetitive mode of an inertial confinement fusion engine (“IFE”) ensures that the system is insensitive to occasional failures in the fusion production. That is, failed implosions impact only the plant output, rather than leading to plant shutdown. On the other hand, the component parts of an IFE engine (driver, injector, etc) must remain operational at all times at the system level—requiring a substantial level of production assurance. As described, this is mitigated by adoption of a highly parallel and modular architecture in which individual laser beamlines (and other components) can move into maintenance mode without affecting the overall plant performance. Similarly, as with other approaches to fusion, high availability requires long-lifetime materials in the high threat environments (first wall and blanket modules, vacuum/gas barriers, final optics, etc). In the LIFE design, this has been mitigated by establishing an operating regime that allows substantial protection and periodic replacement of these components. Our architecture minimizes the consequences of an accident, and to ensure the impact on the local environment is consistent with locations adjacent to the high load centers of cities and industry. In particular this means minimizing the tritium inventory and any projected waste stream.
For both safety and licensing concerns, we minimize the in-process tritium inventory, as well as to ensure tritium supply does not constrain the rate of commercial plant rollout. The scale, cost and design philosophy of the tritium processing plant is closely tied to the material content of the targets, and the clearing rate of the fusion chamber. The architecture here reduces the in-process tritium inventory to the sub-kilogram level within the fusion plant, and to a similar level for the target production factory. The high fractional burn-up in an IFE capsule (>30% for a LIFE facility) relaxes the tritium breeding requirements, while the use of only milligram quantities of fuel per shot, and choice of a pure lithium coolant, substantially reduces the amount of material entrained in the facility. Structural and coolant inventories of just tens of grams, and inventories of hundreds of grams are chosen for this architecture. This minimizes consequences of a coolant fire to an insignificant level.
The fusion reactions within the capsule of each target generate neutrons which pass into coolant surrounding the fusion chamber, causing the coolant to be heated. This heated coolant is used to generate electricity using steam turbine generators 80 (
A benefit of the approach described here is that the separability of fusion target performance from the laser drivers allows future target designs to be incorporated as long as they maintain appropriate interface characteristics to the rest of the plant, i.e. similar irradiation geometry, beamline requirements, target injection and survival constraints, acceptable threat to the first wall, and compatibility with the gas handling system, tritium plant and waste processing systems. Similarly, as laser technology improves, those advances can also be incorporated into the plant.
Also schematically illustrated in
On the right-hand side of
It is expected that the lasers in the LIFE plant will have a lifetime which is less than the lifetime of the overall power plant. By decoupling these life-limited components from plant availability, more reliable, longer duration, generation of electric power is enabled. The line replaceable unit nature of the lasers reduces or eliminates the need to take the overall plant off-line for laser maintenance. Provision of the lasers “in-a-box” allows them to be hot swapped during operations. Thus, a shorter mean time to failure of the lasers than of, e.g. the fusion chamber, can be tolerated. In addition, because each bank 200 of lasers includes a redundant laser 10, failure of a single laser or laser component in one laser of the bank does not shut down the power plant. Detailed statistical modeling of the plant operations supports the beam configuration shown, based on measured lifetimes of the critical components and reasonable calculations of maintenance timescales. Not shown in the figure are coolant loops used to provide coolant to the components in the lasers.
Diodes are used as pump amplifiers for the lasers. Helium cooling enables the amplifiers to operate with a high repetition rate. More detail about the beamline and its components can be found in our co-pending U.S. patent application Ser. No. 12/940,869, entitled “Transverse Pumped Laser Amplifier Architecture,” and U.S. patent application Ser. No. 12/913,651, entitled “Pockels Cell-Based Optical Switch with a Gas-Coupled Electrode.” Each of these patent applications is also incorporated by reference herein.
This approach to modularity in the laser system allows an efficient and cost effective supply chain. It allows the lasers to be shipped to the plant with conventional trucks, and for the lasers to be manufactured in conventional manufacturing facilities at locations distant from the power plant. The lasers offer a low overhead installation process using kinematic placement and minimal interfaces to the remainder of the plant. The mass markets associated with the solid-state components of the laser provide a highly competitive supply chain that now quotes diode price points consistent with a commercially viable rollout (2 to 4 ¢/W) without additional research or development of the technology. This brings solid-state lasers into an affordable range for plant construction. Adoption of a suitably large number of beamlines with the ability to enhance their output to compensate for a failed neighboring beam allows continuous plant operations to be maintained.
We estimate a mean-time-to-failure of at least 2000 hours for an individual laser. This, coupled with an ability to swap beam boxes within an operational shift of 8 hours results in high plant availability. Our approach is a dramatic shift from prior IFE plant designs, which were reliant on extremely high levels of driver reliability (years rather than weeks) to sustain an acceptable level of plant operations.
After passing through the second set of pinholes, the lasers are focused by a Fresnel lens toward the target chamber. This approach provides the advantage of having only the Fresnel lens in the fusion environment, and that lens is placed well back from the center of the chamber. These lenses are relatively inexpensive compared to the laser optics, and can be replaced as necessary.
As shown in
The architecture shown in
The first wall and blanket environment here are designed to cope with high fluences of charged particles, x-rays and neutrons while retaining mechanical integrity, low levels of activation and high levels of performance in converting thermal energy to electricity and breeding tritium. Although advanced materials are still required for some of these subsystems in a commercial setting, our architecture allows construction of a pre-commercial plant in parallel with the materials development program, rather than having to await its success. This is achieved by adopting a “line replaceable unit” (LRU) concept in combination with a gas-protected wall design to substantially reduce the threat. The modularity and scale of the target chamber, on the order of 5m radius, divided into multiple segments allows the entire cooling blanket to be replaced relatively easily.
A high-Z gas such as xenon, is introduced at sufficient density (˜4 to 6 μg/cc) to capture the ions from the exploding target within a 10 to 20 cm gas radius. This effectively eliminates the problem of ionic bombardment on the 5 m radius chamber, which was been a principal limiting factor in previous IFE designs. Similarly, the gas reduces the thermal insult from the x-ray pulse to a level consistent with using available steel materials (pulsing the chamber from 600° C. ambient to a peak of 670° C. each shot). Neutron induced damage is maintained at suitably low levels by treating the chamber modules as a line replaceable unit with a limited operational life, as will be described below. A lifetime of at least one full-power year is calculated for the demonstration plant using steels such as modified-HT-9, Eurofer-97 or F8211 (assuming 10 displacements-per-atom (dpa) can be sustained). Over 4 years full-power life is predicted for the ODS ferritic steels in the commercial plant (running at 20-25 dpa/year). Clearing of the chamber is discussed in more detail in our commonly assigned co-pending U.S. Patent Application Ser. No. 61/534,315, filed Sep. 13, 2011, and entitled “Method and System to Remove Debris from a Fusion Reactor Chamber,” which is incorporated by reference herein.
The complete fusion chamber itself is also treated as a line replaceable unit. This enables swapping in a new chamber while the previous one is maintained or upgraded. It enables the adoption of new technologies or materials as they become available. It also protects the wider plant capital investment from desired changes or upgrades in the fusion chamber design or materials.
Also shown on
Traditionally, a hermetically sealed first wall has been used to form the vacuum barrier, with ports to allow the drive radiation to enter. This imposes severe constraints on the design flexibility. In the illustrated design, the edges of the segments are beveled to protect neutrons from escaping in the joints between the segments. The modular design uses a series of tubes as the first wall, backed by a think blanket. This “chamber” is split into a set of independent modules that can be withdrawn to a maintenance bay in isolation or as a complete unit.
The entire unit is calculated to weigh approximately 1000 metric tons, and could be transported on rails within a suitable cask to the hot cell decommissioning/maintenance area. A replacement chamber would be kept pre-assembled in preparation for operation. By removing the need to disconnect and reconnect any vacuum pipe work, and by decoupling the chamber from the optical infrastructure, a relatively rapid exchange can be achieved. If the entire process of chamber removal and insertion required as much as a month, then a commercial system with a wall lifetime of at least 4 years would only impact the plant availability by 2%. Periodic maintenance at this level is almost certainly required for other plant operations anyway.
Also illustrated in
For each segment of the chamber, the blanket coolant is introduced into the annular narrow regions around the periphery of the blanket bulk region (between the closely spaced lines which surround the blanket bulk region. This coolant flows to cool the outer walls of the trapezoidal chamber. The coolant flows downward to the bottom of that segment, then returns to the top through the large trapezoidal region labeled “blanket bulk.” This flow pattern places the fastest moving coolest coolant closest to the walls of the segment. In a preferred implementation, there is about a 1 cm gap between the first wall tubes and the exterior surface of the blanket containment. There is preferably about a 5 cm gap between the two 1 cm thick walls surrounding the trapezoidal region which contains the bulk flow.
After ignition on NIF, the “next step” is a pre-commercial LIFE power plant generating on the order of 400 MW fusion power. Estimates of the technology development program requirements, along with manufacturing and construction timescales indicate that this plant could be commissioned and operational by the mid-2020s. Such a step would demonstrate all the required technologies and materials certification needed for the subsequent rollout of fusion-based electric power at commercial power plant levels beginning in the 2030s.
Importantly, the performance characteristics and system design of the initial facility is an enabling step for subsequent gigawatt electrical power level pure fusion systems, and for the implementation of gain ˜5-10 fissile/fertile blankets, assuming licensing approval. The latter would offer near term GWe net output on the basis of NIC-level gain and aconservative system design, which could be tested as a direct upgrade of the initial pre-commercial plant.
The timeliness requirements for commercial delivery are compelling. Rollout from the 2030s would remove 90-140 Gigatonnes of CO2-equivalent carbon emissions by the end of the century (assuming US coal plants are displaced and the doubling time for roll-out is between 5 and 10 years). Delaying rollout by just 10 years removes 30-35% of the carbon emission avoidance, which at $50-100/MT translates to a net present value of 70-260 B$. If IFE is to be a meaningful component of the solution, a focused delivery program is essential. Material availability and tritium production calculations show that the rate and scale of market penetration can meet 25% of new build plants by 2050 and significantly more thereafter. Estimates of the capital and operational cost of this approach are competitive with new nuclear and proposed solutions for sequestered carbon fossil-fuel plants.
The LIFE approach is based on the ignition evidence emerging from NIF, and utilizes a line-replaceable unit approach to ensure high plant availability and to allow evolution from available technologies and materials. Utilization of a proven physics platform for the ignition scheme is an essential component of an acceptably low risk solution. An integrated design has been presented here that could lead to an operating power plant by the mid-2020s. The pre-commercial plant described here could further qualify and certify subsystems and components for the subsequent commercial fleet.
We have described an architecture for a fusion power plant operating with principles of laser inertial confined fusion. While we have described our architecture in detail with regard to a preferred implementation, it will be appreciated that the detailed systems described could be implemented using variations of the technology described here. Accordingly, the scope of the invention is defined by the appended claims.
This United States Patent Application is related to, and claims priority from, two earlier filed United States Provisional Patent Applications entitled “Inertial Confinement Fusion Power Plant which Decouples Life-Limited Components from Plant Availability,” filed Nov. 8, 2010, as Ser. No. 61/411,390 and from a provisional application having the same title and filed Feb. 1, 2011, as Ser. No. 61/425,198.
The United States Government has rights in this invention pursuant to Contract No. DE-AC52-07NA27344 between the United States Department of Energy and Lawrence Livermore National Security, LLC.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/US11/59820 | 11/8/2011 | WO | 00 | 5/1/2014 |
Number | Date | Country | |
---|---|---|---|
61411390 | Nov 2010 | US | |
61425198 | Feb 2011 | US |