Laser Fusion

[rhoenix]

A multinational project led by British researchers aims to use a high-power laser to reproduce the physical reaction that occurs at the heart of the sun and every other star in the universe - nuclear fusion. If the project succeeds it has the potential to solve the world energy crisis without destroying the environment.

The scientists admit that a commercial reactor is a long way off, but they believe the laser approach to producing fusion shows great promise. The EU is considering a proposal to fund the set-up costs for a seven-year research project called HiPER - high powered laser energy research - that would build a working demonstration reactor. Preparing for the seven-year project alone, which is a collaboration of 11 nations, is expected to cost over €50m (£34m). Actually building the reactor itself will cost over half a billion euros.

The British-led project, which has been earmarked by the EU as a priority, is designed to leapfrog an American-funded project called the National Ignition Facility (Nif) in Livermore, California. When that is built in 2010, physicists are confident that the Nif laser will be powerful enough to start a fusion reaction. Experiments in the Nevada desert in the 1980s with underground explosions of nuclear weapons have already shown how much energy they will need to deliver with the laser.

Mike Dunne, director of the Central Laser Facility at a publicly funded research site in Oxfordshire that houses Vulcan, the most powerful laser in the world, said: "The world is going to take notice when this happens. Politicians are going to look around and say, 'So what are you going to do about it? What's the next step?'. This is how to take it from a scientific demonstration to a commercial reality."

Prof Dunne said that many of the details of the nuclear tests were still classified, "but the only thing that matters to us as a bunch of energy scientists is that it does work. The trick now is, can we get it to work without throwing a nuclear bomb at the thing?" That is what Nif is designed to do.

Achieving fusion on Earth in a way that will release useable energy has long been an aspiration of physicists.

The idea is to fuse two atoms of hydrogen to form helium. The reaction that powers the sun releases large amounts of energy because it turns Einstein's famous E=mc² equation on its head. A small amount of mass is lost when the hydrogen atoms combine, in the process releasing vast quantities of energy.

Unlike nuclear fission, only low-level radioactive material, no more dangerous than hospital waste, is left over afterwards. And best of all, a runaway chain reaction like the one that caused the Chernobyl meltdown is simply impossible. The fusion dream is already being pursued by a €10bn project called Iter - international thermonuclear experimental reactor - which is being built in Cadarache, France. This project aims to use powerful magnets to fuse the hydrogen atoms. But many in the laser research community see their approach of bombarding hydrogen with a high energy laser as the more promising route.

"The beauty of the laser approach is that you can divide and conquer," said Prof Dunne. There are formidable engineering challenges in building a high enough power laser, increasing its firing rate and designing the millimetre sized fuel pellets, but these can all be pursued in parallel, he said.

Others are more sceptical about the laser approach. Duarte Borba, who works at Jet, an experimental magnetic fusion reactor, said achieving ignition was not the be-all and end-all. "There is a long process still ahead before you can actually build a reactor based on laser fusion," he said.

Benefits and snags

Nuclear fusion
Process in which two isotopes of hydrogen - deuterium and tritium - are combined to produce helium, a neutron and huge amounts of energy.

Deuterium or heavy hydrogen
Conventional hydrogen is made up of a proton nucleus with an electron spinning around it. The nucleus of a heavy hydrogen atom contains a proton and a neutron.

Tritium or super-heavy hydrogen
Its nucleus contain a proton and two neutrons. It is moderately radioactive and can be manufactured from the metal lithium.

Environmental benefits
Nuclear fusion does create some low-level radioactive waste, but nothing more dangerous than you would find in a hospital. The reaction does not produce carbon dioxide so it will not contribute to the greenhouse effect and a Chernobyl-style meltdown is impossible.

Problems
The biggest challenge will be to build a powerful enough laser that can fire rapidly enough. The world's most powerful lasers need several minutes to reset for a second shot. A laser fusion reactor will need to fire several times a second. Scientists will also need to develop materials durable enough for the laser bombardment.

Very interesting stuff. I've my fingers crossed for beneficial progress.

[Sea Skimmer]

Very interesting stuff. I've my fingers crossed for beneficial progress.

I'm hoping for it to.

Of course once we have working fusion power we’ll also be well on the way to being able to make a pure fusion nuclear bomb. Just imagine, nations armed with the firepower of today’s nuclear arsenals, but without the need to worry about long term fallout! It will be peace and safety for all!

[Starglider]

I'm hoping for it to.

Of course once we have working fusion power we’ll also be well on the way to being able to make a pure fusion nuclear bomb.

Magnetic confinement fusion doesn't really have anything to do with pure fusion bombs; while there's no known way to scale laser fusion initiation mechanisms down to bomb sizes, at least it's conceivable, whereas magnetic systems are inherently unsuited to weaponisation. While any type of fusion power is better than none, as far as I can see this is one of several reasons to hope that magnetic confinement is the technology that will mature and be adopted for widespread power generation. Personally I'm hoping that superconducting stellarators become practical, because they're the only reactor type that provides a true steady-state burn and while they have high geometric/magnetic complexity mechnically and electrically they're simpler (and potentially lighter) than tokamaks and multi-laser systems.

[Darth Wong]

Why is this news? Haven't they been working on laser-induced fusion for decades?

[rhoenix]

If you're asking why I posted it, it's because I'm heartened by all the nuclear fusion-based research going on right now.

[Darth Wong]

If you're asking why I posted it, it's because I'm heartened by all the nuclear fusion-based research going on right now.

Don't get too excited. We'll probably all be dead before they get a commercially viable nuclear fusion reactor working. A lot of the early physics groundwork was done on these basic principles back in the 1950s for fuck's sake, and they're still not even at the stage of developing a prototype for a commercial reactor.

The development of nuclear fusion power is the most profoundly difficult engineering project in human history: more difficult than going to the Moon was in 1969.

[rhoenix]

Don't get too excited. We'll probably all be dead before they get a commercially viable nuclear fusion reactor working. A lot of the early physics groundwork was done on these basic principles back in the 1950s for fuck's sake, and they're still not even at the stage of developing a prototype for a commercial reactor.

I didn't mean to convey irritation or anything of the sort in my reply - I was somewhat confused by your reply, and chose to reply with the more simple assumption.

The development of nuclear fusion power is the most profoundly difficult engineering project in human history: more difficult than going to the Moon was in 1969.

Exactly - this is why when I (though a layman) see more and more research facilities devoted to it's research, I feel a bit hopeful about the eventual produce of the research.

However, a hydrogen economy is being considered more and more these days. I enjoy seeing advances with both technologies.

[metavac]

Why is this news? Haven't they been working on laser-induced fusion for decades?

This is news because HiPER, along with NIF, will kick off only the second generation of large ICF experiments aimed at achieving fuel ignition.

[metavac]

Personally I'm hoping that superconducting stellarators become practical, because they're the only reactor type that provides a true steady-state burn and while they have high geometric/magnetic complexity mechnically and electrically they're simpler (and potentially lighter) than tokamaks and multi-laser systems.

Stellarators shouldn't be inherently lighter than tokamaks and definitely not more so than ICF systems. On top of that, geometric complexity is exactly what makes it hard to do physics; it's easier to talk about linear drift in a tokamak and impart current to drive a helical flow radially than to work with magnetic fields with far more complex symmetries. An even more geometrically simple configuration would be the levitated dipole (c'mon, the physics doesn't get more simple than that). Unfortunately, this idea will be stuck in the toy model stage until they tackle most of all the drift problems already conquered in tokamak research.

[Starglider]

Stellarators shouldn't be inherently lighter than tokamaks

Stellarators do not need a mechanism (i.e. coils) to drive/adjust plasma current. The steady state operation means that the thermal stresses are lower and the efficiency of the startup mechanism is less critical.

and definitely not more so than ICF systems.

Current ICF designs require a large warehouse full of lasers consisting largely of glass blocks, as well as a hugely complex multi-beam routing and focusing system. While theoretically ICF could be lighter, thus it's use in things like the Daedalus interplanetary probe design, current designs are several factors heavier and bulkier than the equivalent stellarator. I have not seen any realistic engineering design for such a lightweight ICF system; the Daedalus designers just assumed particle beams of output X, perfect focus and that dumping Y joules onto a fuel capsule would suffice for ignition.

On top of that, geometric complexity is exactly what makes it hard to do physics; it's easier to talk about linear drift in a tokamak and impart current to drive a helical flow radially than to work with magnetic fields with far more complex symmetries.

Over time technology progresses from things that are simple to design and have low operating and manufacturing efficiencies to things that are hard to design but have high efficiencies. You only have to solve the physics once for a given design (and computation is already cheap and getting cheaper every year), after which you can build as many copies as you like, and lower mechanical and electrical complexity is greatly preferable for that (only the physicists and maybe control software programmers are affected by magnetic complexity, everyone who has to work on the thing is affected by mechanical/electrical complexity and the additional chors and challenges pulsed-mode operation). Tokamaks are almost certainly going to produce power first, but hopefully some form of stellarator will be perfected as a second-generation design.

[metavac]

Stellarators do not need a mechanism (i.e. coils) to drive/adjust plasma current. The steady state operation means that the thermal stresses are lower and the efficiency of the startup mechanism is less critical.

That doesn't reduce the mass, it just changes the geometry of the stellarator coils. And yes, magnetic confinement requires a lower gain before ignition than any feasible laser scheme.

ICF designs require a large warehouse full of lasers consisting largely of glass blocks, as well as a hugely complex multi-beam routing and focusing system. While theoretically ICF could be lighter, thus it's use in things like the Daedalus interplanetary probe design, current designs are several factors heavier and bulkier than the equivalent stellarator. I have not seen any realistic engineering design for such a lightweight ICF system; the Daedalus designers just assumed particle beams of output X, perfect focus and that dumping Y joules onto a fuel capsule would suffice for ignition.

The actual vessel and optics for NIF, with concrete included, will weigh 530 tons. By contrast, Large Helical Coil, on the other hand, is 1500 tons. I think you're absolutely right though that NIF will be significantly larger volumetrically.

Over time technology progresses from things that are simple to design and have low operating and manufacturing efficiencies to things that are hard to design but have high efficiencies. You only have to solve the physics once for a given design (and computation is already cheap and getting cheaper every year), after which you can build as many copies as you like, and lower mechanical and electrical complexity is greatly preferable for that (only the physicists and maybe control software programmers are affected by magnetic complexity, everyone who has to work on the thing is affected by mechanical/electrical complexity and the additional chors and challenges pulsed-mode operation). Tokamaks are almost certainly going to produce power first, but hopefully some form of stellarator will be perfected as a second-generation design.

I'd think the jury is still out on which is inherently a better choice, tokamak or stellarator. Still, I actually wouldn't be surprised if ICF produced net power before either of them. MCF is light-years ahead in terms of thermal stress management, coupling and MHD, but ICF is more or less an amalgamation of two different, well paved fields of healthy active research--optics and nuclear weapons design--whereas MCF is mostly about making breakthroughs in plasma physics. One thing's for sure, laser fusion's going to need MCF before it's all over.

[Starglider]

The actual vessel and optics for NIF, with concrete included, will weigh 530 tons. By contrast, Large Helical Coil, on the other hand, is 1500 tons.

That comparison is meaningless without a specification of power output. Unfortunately I don't know how to go about this, since those are both experimental installations that don't produce net power and don't include heat-exchanger/turbogenerator mass.

I'd think the jury is still out on which is inherently a better choice, tokamak or stellarator.

Regardless of which is better for stationary power generation, continuous operation is always going to be preferable to pulsed operation for vehicle applications (ICF can be high enough frequency to appear continuous to the rest of the system, but it's hell on the reaction vessel walls, even more so than being exposed to fusing plasma continuously).

[metavac]

That comparison is meaningless without a specification of power output.

Something like 1 to 10 MW for LHD, and just under 1 TW for NIF. Of course, the NIF experiment duration is orders of magnitudes shorter.

Unfortunately I don't know how to go about this, since those are both experimental installations that don't produce net power and don't include heat-exchanger/turbogenerator mass.

Absolutely, my only point is that I can't think of anything that would make a working MCF reactor inherently lighter than an ICF one.

Regardless of which is better for stationary power generation, continuous operation is always going to be preferable to pulsed operation for vehicle applications (ICF can be high enough frequency to appear continuous to the rest of the system, but it's hell on the reaction vessel walls, even more so than being exposed to fusing plasma continuously).

At first glance I'd agree, but it's ultimately going to depend on what emerges as thermocoupling for ICF and what choices we can make when it comes to the reactor vessel. Funny thing, most of the papers I've seen on fusion power for space propulsion usually talk about ICF. You seem to follow this much more closely than I do. Any idea why that is?

[Sikon]

Funny thing, most of the papers I've seen on fusion power for space propulsion usually talk about ICF.

Yes. Vista, Orion, Mini-Mag Orion, ICAN, even project Daedulus, and more ... pulsed propulsion for the highest performance engines among those based on engineering calculations including heat transfer.

Regarding pulsed versus continuous operation of an engine, consider the amount of heat transfer that occurs if there is exposure to X temperature for a given length of time. Now consider the amount that occurs if there is exposure to X temperature for a millionth of the time, or even to much more than X temperature, followed by a proportionally long pause. All else being equal, heat transfer is influenced by the time of exposure, decreasing vastly if the time of exposure is less. Very high propellant temperatures can occur without the engine melting or overheating if there is only momentarily contact with expanding superhot plasma, contact so momentarily as to still be survivable, optionally helped to the degree possible with a magnetic nozzle.

An analogy is the case of fission. The NERVA design internally having continuously fissioning propellant was limited to ~ 3000 K and ~ 1000 sec Isp, yet heat transfer calcs for external pulsed propulsion concepts allow them to have vastly higher propellant temperature and a greater order of magnitude of specific impulse.

Inertial confinement fusion (ICF) and other types of external nuclear pulsed propulsion were discussed in an old thread.

There is little question about ICF being functional eventually since it is the same basic idea as the existing application of fusion energy, thermonuclear bombs, but replacing the fission trigger, a system known to be able to obtain at least a hundreds to one ratio of yield energy to driver energy, although obtaining capital costs competitive with simpler fission reactors would be a challenge and uncertain.

[metavac]

Leaving aside thermal contact with the reactor vessel, I took Starglider's concern about ICF on spacecraft to refer to oscillations in overpressure. Do you two have anything to say on that?

[Sikon]

Leaving aside thermal contact with the reactor vessel, I took Starglider's concern about ICF on spacecraft to refer to oscillations in overpressure. Do you two have anything to say on that?

The ship is designed for what it can handle. If a given mechanical impulse per pulse would be too high, having the pulses occur farther away and/or reducing the yield per pulse counters such. In the case of the fission external pulsed propulsion Orion system, the mechanical impulse or momentarily "pressure" was determined to be able to be handled by setting the detonation distance appropriately, as illustrated in a declassified portion of a General Atomics study here. For ICF, the situation is actually better in that regard since there is no minimum critical mass, no fission trigger, and no minimum efficient yield. With ICF (or even the less technologically difficult Mini-Mag Orion fission technique), detonations wouldn't have to be kiloton-range but could be even just GJ yield if desired, making the mechanical impulse still easier to handle, whether received directly or through reaction against a magnetic nozzle.

[Admiral Valdemar]

Assuming we can get net energy output via at least one of these fusion concepts, how small a powerplant could you get for propulsion, be it a rocket drive or for powering a terrestrial vehicle e.g. large bucket excavator or tanker ship? Fission is a more advanced area for obvious reasons, less technically challenging and only slightly less energy dense, but fusion requires heavy, complex equipment just to get the reaction starter, so I'm wondering if the likes of JET, for instance, would be the smallest without radical advances in technology (I'm ignoring AM initiated fusion given the idea is a bit beyond us for this century at least).

[metavac]

Assuming we can get net energy output via at least one of these fusion concepts, how small a powerplant could you get for propulsion, be it a rocket drive or for powering a terrestrial vehicle e.g. large bucket excavator or tanker ship? Fission is a more advanced area for obvious reasons, less technically challenging and only slightly less energy dense, but fusion requires heavy, complex equipment just to get the reaction starter, so I'm wondering if the likes of JET, for instance, would be the smallest without radical advances in technology (I'm ignoring AM initiated fusion given the idea is a bit beyond us for this century at least).

I can't tell you what the minimum size of a fusion reactor might be, but the size of your reactor depends on how efficiently you can address three problems, particle density, confinement time, and temperature.

If you want to write a story about it, tackle confinement time. A good sf writer will simply assume the existence of some amazing Compound X that you can use to fabricate or coat your coils with, thereby miraculously improving the pinch of the containment field. The size of your reactor will depend mostly on how much power you can prevent leaking out of the reactor in the form of neutrons, bremsstrahlung, etc. This is the big bitch, the other two relevant parameters (temperature, particle density) are somewhat easier to tackle.

Here's a cute little image from JET that shows our progress in terms of the triple product and ion temperature.

[Darth Wong]

One thing most SF writers (or people talking about real-life reactors) never talk about is the enormous wear and tear on the reactor materials. There is a huge neutron flux that would be generated by any nuclear fusion reactor, and there is no theoretical solution to this; it is part of the basic nature of nuclear fusion and cannot be altered. You have to capture the energy of the neutron radiation and convert it in order to make your reactor work, and the materials lining the reactor will be under constant bombardment. Any realistic nuclear fusion reactor would require a rigorous maintenance schedule of replacing the lining as it becomes damaged by the neutron flux.

Of course in SF I suppose you could just make a magic material that can absorb infinite amounts of neutron radiation, convert it to electrical power automagically, and never wear out.

[Starglider]

There is a huge neutron flux that would be generated by any nuclear fusion reactor, and there is no theoretical solution to this.

Aneutronic reactions release less than 1% of their energy in neutrons (strictly they release no neutrons, but there tend to be small amounts of unavoidable side reactions). There are active projects attempting to achieve hydrogen-boron fusion and many sci-fi concepts (as well as the Daedalus probe proposal as I recall) use He3-D fusion for this reason.

[Chris OFarrell]

I wonder how much progress we could have made into Fusion technology if we had taken all the money that has been dumped into the nightmare of Iraq and instead tried for an Apollo project style (if international) blast into fusion research.

I know money is only one part of the problem, but with that much cash flow and the ability to attract huge numbers of skilled scientists, engineers and so on to really focus on each part of the problem that comes up...I wonder if we could have cut the development of a commercially viable reactor down to a decade or two...

[metavac]

Of course in SF I suppose you could just make a magic material that can absorb infinite amounts of neutron radiation, convert it to electrical power automagically, and never wear out.

Or a material that you can replenish in situ. Also, reflecting neutrons or absorbing them and emitting energetic products back into the plasma would be nice as well.

[Sikon]

Assuming we can get net energy output via at least one of these fusion concepts, how small a powerplant could you get for propulsion, be it a rocket drive or for powering a terrestrial vehicle e.g. large bucket excavator or tanker ship?

Interesting question.

With nuclear fusion or nuclear fission (or antimatter), one limit can be radiation shielding. For an ordinary nuclear power plant, such may be multiple meters thick (concrete, etc.). Mobile reactors can be designed for far less weight and high-performance shielding material combinations, while accepting moderate irradiation of the surroundings, but the shield still tends to be a substantial number of centimeters thick.

Even "aneutronic" fusion like helium-3 would still produce some neutrons from secondary reactions, in addition to the high-energy gamma rays, while being more difficult to get working than deuterium-tritium, e.g. 58 KeV versus 14 KeV for minimum triple product for ignition (better to try to get the latter working first, difficult enough already). Manned areas can't be exposed to a millionth of the unshielded radiation from a megawatt-class reactor, as the cumulative total over tens of millions of seconds a year can't be allowed to add up to so much as a substantial fraction of a joule per kilogram of human tissue (1 J/kg = 100 rad).

The nuclear reactor vessel of the NASA 100,000-watt SP-100 design is just a cubic meter in volume, roughly the size of a large trashcan, with a diameter of 35-cm and shield mass of 1000-kg. And it obtains such in part by not having full radiation shielding, being kept 23 meters away from the manned mission module. The SP-100 is exceptional. For example, in comparison, the modified B-36 for the atomic-powered aircraft program of the 1950s had a 5 ton reactor with 30 tons of shielding.

Technically, as little as one or two kilograms of californium-251 would be a critical mass or thus enough for the nuclear fuel component of a minimum-weight fission reactor, although total reactor mass would tend to be much more even with relatively limited radiation shielding.

Even for a fusion reactor with no critical mass and super-advanced technology, material performance like the thickness required for eliminating nearly all of a given amount of radiation is one aspect unlikely to vastly change. Molecular materials have limitations by their nature, all within a limited density range, etc. (Though slightly off-topic, even limits like the theoretical maximum strength of a material made from carbon-carbon chemical bonds are already being approached in some cases within a single order of magnitude; all of the ordinary 92 elements are known today, no possible gaps within the periodic table there, and even additional elements like a hypothetical stable transuranic would not plausibly have orders of magnitude different properties). And non-atomic construction materials are outside the realm of scientific plausibility.

There is going to be some minimum size, even if irradiating the surroundings was not a concern, even just for the functionality of the reactor. For example, if a fusion reaction was somehow initiated in a microscopic device but nearly all of the reaction products like neutrons, gamma rays, and high-energy nuclei just escaped through the thin walls of the device, such couldn't produce much power.

One may conclude that a reactor diameter on the order of a meter or less is possible with fission, while fusion could be either better or worse depending upon technological capabilities.

As for another limit, the minimum mass of equipment involved in the fusion reactor, that's rather uncertain. In the case of ICF fusion, there isn't any clear universal, absolute minimum driver energy requirement, since one can try to counter having less driver energy by having it hit a smaller fuel target. However, current systems have estimated needs in the low MJ range, from the Z-machine to the NIF. The NIF facility involves 1.8 MJ with 30 kJ transferred into the deuterium-tritium fuel itself. In itself, that's little energy. For perspective, 30 kJ is about 10x the energy of a rifle bullet. As a random example, a couple liter volume of some pulsed-power capacitors could nominally store 30 kJ. But there's currently multiple orders of magnitude of inefficiency. Efficient, compact drivers for delivering energy to the target in a timescale of nanoseconds are not currently available. For example, the NIF facility is hundreds of tons, though future technology might be smaller.

I don't know what would be the minimum size for magnetic confinement fusion. Even if it did someday work with enough energy gain on a large scale (not breakeven but far beyond nominal breakeven for countering the inefficiencies of the whole system), its need for containing high temperature plasma would suggest a practical minimum size. Very small plasma volume would tend to mean a high ratio of surface area to volume, a particularly high ratio of radiated power loss relative to reaction power, with a tendency for reduced energy gain if attempting to scale down to very small size, reducing the chance of practicality. That's in addition to the radiation shielding limit.

If one really wants minimum size while keeping the benefits of nuclear energy avoiding frequent refueling or recharging, radioisotopes might actually allow the smallest power sources. There are some radioisotopes that only emit alpha particles with miniscule penetration that can be stopped by less than the thickness of a sheet of paper, so such could work as a power source even for micro-robots.

(I'm ignoring AM initiated fusion given the idea is a bit beyond us for this century at least).

I'm not sure why one would think that. Largely the reason there is work today on antiproton catalyzed microfission/fusion is because it could potentially be technologically easier than pure ICF fusion for spaceship propulsion, with less driver requirements. As illustrated in detail by me in another thread, the ICAN II concept paper gives figures implying a need of only around 30 nanograms of antimatter for a craft with hundreds of tons of fuel. In their concept, more than 99.99% of the energy release is nuclear, the tiny amount of antimatter being used to help ignition like a match starting a forest fire. Current antimatter production at Fermilab and CERN is a total of 1 to 10 nanograms annually, though it isn't stored. One or two orders of magnitude increase may be obtained in the near future, more than ICAN needs:

The Fermi National Laboratory is currently constructing the Main Injector ring, which can produce 14 ng of antiprotons in one year's time. A recycling ring can boost production by a factor of 10.

From here.

More on it:

[...] Here, a small concentration of antimatter and fissionable material is used to spark a microfusion reaction with nearby material.

[...]
The AIM engine requires just 5x10^8 antiprotons per reaction; this amount can be readily obtained from Fermilab and CERN. Experimentation with such an engine can take place after methods of storing and transporting antimatter have been realized. One of the Antimatter group's chief projects in the past decade has been the design, fabrication, and testing of a portable antiproton trap (Penning trap) named "Mark I", which can store 10^10 antiprotons for one week. The experimental results from Mark I are currently being used in the development of a NASA Penning trap that can store 10^12 antiprotons, large enough to support hundreds of reactions over a 2 minute timeframe.

Storage of antimatter is a challenging task, but reaps several benefits. One of which is the generation of O15, a radioisotope used for Positron Emission Tomography (PET) of the human brain. Currently, only certain research hospitals across the world have the ability to create Oxygen-15. Due to its portability, a "radioisotope generator" antimatter trap may be transported to more remote areas for patients who cannot reach these hospitals. A second medical application concerns antiproton radiotherapy of tumors. The NASA Penning trap is being designed with these medical applications in mind.

From here.

And more commentary:

Dr. Robert Forward has shown, based on his findings in a study of antimatter production, that if a dedicated antimatter factory were built now, it could be approximately 6000 times more efficient than Fermilab's and CERN's antimatter production facilities (bringing it up to a grand 0.01% efficiency).

From here

The key is one is talking about nanogram amounts of antimatter for a special application. If the figures had been kilograms of antimatter instead, a billion times as much, the near-term possibility wouldn't be so. But the terrible inefficiency and expense of producing antimatter per gram isn't as bad if one only needs a more miniscule amount, to supply a very tiny portion of total energy.

[...] e.g. large bucket excavator [...]

Though this is a little off-topic, some excavating equipment actually already is indirectly nuclear-powered.

The world's largest equipment is often powered directly from power lines because fuel costs would otherwise be huge, powered from an electrical grid supplied in part by nuclear reactors. To quote a general description:

A large dragline system used in the open pit mining industry costs approximately US$50-100 million. A typical bucket has a volume ranging from 30 to 60 cubic metres, though extremely large buckets have ranged up to 168 cubic metres.[1] The length of the boom ranges from 45 to 100 metres. In a single cycle it can move up to 450 metric tonnes of material.

Most mining draglines are not fuel powered like most other mining equipment. Their power consumption is so great that they have a direct connection to the high-voltage grid at voltages of between 6.6 to 22kV. A typical dragline, with a 55 cubic metre bucket, can use up to 6 Megawatts during normal digging operations. Because of this, many (possibly apocryphal) stories have been told about the blackout-causing effects of mining draglines. For instance, there is a long-lived story that, back in the 1970s, if all seven of Peak Downs (a very large coal mine in central Queensland, Australia) draglines turned simultaneously, they would blackout all of North Queensland.

In all but the smallest of draglines, movement is accomplished by "walking" using feet or pontoons, as caterpillar tracks place too much pressure on the ground, and have great difficultly under the immense weight of the dragline. Maximum speed is only at most a few hundred metres per hour since the feet must be repositioned for each step.

From here

Here's an example, in this case of the biggest single-bucket digging machine ever built:

* Weight: 27 million lbs., or 13,500 tons
* Bucket Capacity: 220 cubic yards, 325 tons (12 car garage)
* Height: 222 ft., 6 in.
* Length of the boom: 310 feet
* Length of machine with boom down: 487 ft., 6 in.
* Empty bucket weight: 230 tons
* Width: 151 ft., 6 in., compare to an 8 lane highway!
* Cable diameter: 5 in
* Electrically powered: 13,800 volts
* Mobility: hydraulically driven walking feet

In her working lifetime, the Big Muskie removed over 608,000,000 cubic yards of overburden (twice the earth moved during the construction of the Panama Canal), uncovering over 20,000,000 tons of clean coal. She cost $25 million in 1969.

From here

[Darth Wong]

There is a huge neutron flux that would be generated by any nuclear fusion reactor, and there is no theoretical solution to this.

Aneutronic reactions release less than 1% of their energy in neutrons (strictly they release no neutrons, but there tend to be small amounts of unavoidable side reactions). There are active projects attempting to achieve hydrogen-boron fusion and many sci-fi concepts (as well as the Daedalus probe proposal as I recall) use He3-D fusion for this reason.

The reason we want to use some kind of hydrogen radioisotope fusion is the availability of hydrogen. Hence, we're talking about D-T fusion.

[Starglider]

The reason we want to use some kind of hydrogen radioisotope fusion is the availability of hydrogen. Hence, we're talking about D-T fusion.

Did you research this claim at all before making it? Hydrogen-boron fusion is even better than D-T fusion in this regard, because it works with ordinary hydrogen (you don't even need deuterium) and the most common isotope of boron (~80% frequency, currently refining ~2 million tonnes of boron a year). Tritium on the other hand is extremely rare, which is why practical D-T power station concepts require lithium blankets that 'breed' tritium via neutron capture (with the associated mechanical complexity of capturing and storing it). However hydrogen-boron requires roughly an order of magnitude more ion energy than D-T and has only a third of the reaction rate at the same density. He3 is rare on earth, but it's a lot more common than tritium or even lithium to breed tritium with in the universe in general, so He3 fusion wins for sci-fi concepts (D-D fusion beats He3-He3 for reactant abundancy, but it's much more difficult - pure hydrogen fusion would be the best but AFAIK it's essentially impossible to use in a reasonably compact reactor).