A Trip to 'Fusion Island'

William Nuttall, Bartek Glowacki and Richard Clarke

Text of article published in The Engineer, 31 October 2005 (by permission)

Energy is regularly creating headline news, and much of it makes grim reading. Rising petrol prices and an increasingly unstable geo-political situation in the planet's oil-producing regions are hardly reassuring.

Anxiety over CO2 emissions and a growing, though by no means universal consensus that 'something must be done' poses more questions than answers. On a more positive note we are witnessing the first, tentative steps towards a clean, sustainable 'hydrogen economy' and real progress towards fusion energy as a viable technology.

As three specialists in energy technology, science and policy, we have a 'big idea' that links all the above, and which we hope will prompt comment and suggestions.

As well as departing from the conventional wisdom that fusion energy's role will be as a generator of electricity, we believe we have identified a neat solution to one of the major technical issues facing the nascent fusion industry.

Our proposal is to plan for nuclear fusion energy not for electricity generation, but as a hydrogen production technology. We are not the only ones to take this position - General Atomics, the US nuclear technology group, has highlighted nuclear fusion as a production source for a hydrogen economy.

But we go further by proposing a fusion energy system with no connection to national electricity grids whatsoever. Fusion does not need to be embedded in complex national electricity infrastructures. Instead, it could make a huge contribution as a self-contained 'island' of activity.

On 'Fusion Island', hydrogen would be:

  • The product sold commercially;
  • Potentially the cryogenic liquid used to cool the fusion reactor magnets;
  • The source of energy to fire-up the fusion machine; and
  • A link to an industrial sector that is willing and able to fund the high capital cost of a fusion system.

Currently, our main energy carriers are fossil fuels and electricity. The former is economically storable, the latter is not. Hydrogen, like electricity, is not an energy resource (or fuel) but rather a pure energy carrier. The hydrogen economy will need manufactured hydrogen yielding a fluid energy product that could be transported and stored in a manner similar to fossil fuels.

For hydrogen to be a truly environmentally benign energy carrier, it will be important to produce it efficiently and without the combustion of fossil fuels.

General Atomics has noted the efficiency benefits arising from the direct thermochemical production of hydrogen using high temperature catalytic reactions such as the Sulfur-Iodine cycle. This is currently seen as the most efficient hydrogen production process. It does not rely on wasteful intermediation using electricity for electrolysis, but does require high temperatures - above the 750C mark (and preferably around 900C).

Conventional approaches to fusion dedicated to electricity production plan for fusion blanket temperatures (where steam or hot helium would be produced to drive thermal turbines) of around 500C. Fusion designs could surely be shifted into the 750-1000C range, the so-called 'high temperature' approach.

Importantly, however, environmental factors are not the only policy-driver for the hydrogen economy. Over the timescale of the development of fusion energy systems, energy security issues are likely to motivate a shift towards hydrogen, fusion and more sophisticated energy storage and distribution.

The fusion generator described above would produce large quantities of hydrogen needing safe storage before transportation. It seems likely that the energy density benefits of cryogenic hydrogen storage and distribution will outweigh the higher efficiency of compressed gas storage.

Part of the concept is that the stored cryogenic hydrogen product might itself underpin the operation of the superconducting magnets needed to maintain the magnetic 'tokamak' confinement of the fusion plasma. Conventionally, and at great cost, fusion designs envisage Niobium-Tin (Nb3Sn) superconducting windings cooled with liquid helium.

Our vision is for magnet windings of Magnesium Diboride cooled to approximately 20K by liquid hydrogen. MgB2 has the pleasing characteristic that, as an intermediate temperature superconductor, it is straightforward to produce flexible wires for magnet windings. It is expected that a full-scale fusion reactor will require fields of approximately 7T in the heart of the plasma. In a conventional tokamak geometry this corresponds to a field of 13T at the windings themselves. Fields of 13T are achievable using MgB2 even if the substantial mechanical reinforcement required by composite superconductors, such as Nb3Sn were to be used.

In fact, things are likely to be easier with MgB2 as it will provide greater intrinsic mechanical strength greatly simplifying the engineering challenges. As a result of its mechanical attributes and continuing improvements in MgB2 superconducting performance it is likely that fields substantially greater than 13T will be achievable for Fusion Island.

MgB2 has the benefit that if it is placed in a moderate neutron flux, its superconducting properties actually improve owing to the creation of nanostructure defects.

Given the high neutron fluxes in parts of the fusion reactor, such materials' resilience could be of great importance. Some cryogenic helium will be required for cryo-pumping to maintain the cleanliness of the reactor vessel.

Importantly, however, the demands for helium in the Fusion Island concept could be dramatically lower than under conventional approaches to fusion energy.

Moving to hydrogen-based fusion cryogenics may raise safety concerns and the need to ensure that flammable hydrogen is at all times safely isolated from any oxidants. Of course, safety will be paramount in every component of the hydrogen economy, and a fusion plant will be no exception. There would need to be a very careful examination of all the safety implications. If taken literally, the Fusion Island concept would be implemented on a real island, with all the associated safety benefits.

Conventional plans for fusion energy involve electricity on two levels, most obviously as the end-product for sale. Second, however, we must not forget that a fusion power plant will require substantial electrical energy to re-start. At first glance, this could be seen as a big argument in favour of fusion's role in an integrated electricity system.

On further reflection, this becomes less obvious. For example, a fusion power plant contributing 2GW to the electricity grid on a winter evening could be providing several percentage points of total supply at a time of a very low-capacity surplus in the system. If for any reason the fusion power station were to trip out, then the grid supply/demand balance would be hard to maintain. The system operator might call upon open cycle gas turbine and pumped-water storage systems to cover the shortfall. If the fusion power plant operators were then to call the system operator asking for 500MW to restart their machine, it is far from clear that they would receive a positive response.

However, in our Fusion Island concept the fusion reactor would be supported by its own dedicated large-scale hydrogen fuel cell park. The capacity of the park would be determined as part of the island's integrated black-start capacity involving fuel cells, energy storage flywheels (as used on the existing JET/EFDA fusion research machine for roughly half of its start-up power needs) and possibly supercapacitors for the final big electrical push.

In normal operations Fusion Island would have on-site temperatures ranging from 20K to more than 1000K. As hydrogen fuel cells require raised temperatures in order to operate, there is a conceivable benefit in integrating the fuel cell system into the fusion heat extraction system, or even the fusion blanket itself. Even if truly cold and dark, the plant could combust sufficient hydrogen to warm the fuel cells to an operational level.

Who, it is fair to ask, will pay for all this? After lengthy negotiations an international consortium recently announced plans to build an experimental fusion reactor called ITER at Cadarache in France. ITER is designed to demonstrate 500MW of net power generation. One of our trio, William Nuttall, has previously argued in the pages of this magazine that this experimental system should be the last large-scale tokamak fusion machine to be built using public money.

Realistically, it is unlikely that a riskaverse, competitive electricity industry will be keen to invest the large sums necessary to construct prototype and early-stage commercial fusion power plants. Adding to their nerves would be the fear that the first generation of fusion plants might suffer from poor operational reliability. Synchronising a burning plasma system to the national grid is going to be a major challenge for fusion electricity.

As electricity is not an economically storable commodity, and supply contracts involve stiff penalties for failure to generate, any such reliability failings could be expensive. But if as we have argued here, the early fusion energy systems are dedicated to thermochemical hydrogen production primarily serving the needs of the transport sector, then intermittency is not a concern, because hydrogen storage is an integral part of the system. Furthermore, hydrogen as an energy carrier has the possibility of becoming the dominant fluid energy product sold by the oil majors once oil itself becomes depleted or environmentally unacceptable later this century.

At the outset of this article we spoke of the environmental and geo-political pressures that are set to converge to the advantage of the Fusion Island concept.

These inevitably lead to the conclusion that the source of private capital for the development of hydrogen fusion energy systems should be the oil majors, aiding their transition from oil companies to more broad-based energy providers. Thermochemical hydrogen production and distribution from a Fusion Island would require many of the skills and competencies the oil majors already possess. Furthermore, these businesses still retain an adventurous culture of exploration and risk taking well suited to developing a difficult, high-stakes technology such as fusion.

About the authors

  • William Nuttall is senior lecturer, technology policy, Judge Business School, University of Cambridge and author of Nuclear Renaissance (CRC Press).
  • Bartek Glowacki is reader and head of Applied Superconductivity and Cryoscience, Department of Materials Science and Metallurgy, University of Cambridge.
  • Richard Clarke is section leader JET/EFDA Cryogenics, UKAEA Culham Science Centre, Abingdon.