Tokamaks and alternatives

Tokamaks

In tokamaks, plasma is confined in a torus-shaped vessel by a magnetic field with two main components. The first (so-called toroidal) field is produced by a set of coils equally spaced around the plasma vessel, and keeps the plasma away from the walls. However, by itself this field is not enough to confine the plasma.

A second (so-called poloidal) field is added to counteract the natural pressure inside the plasma which tries to make it expand. In a tokamak, the poloidal field is generated by a strong electrical current which flows through the plasma, and which is also used to heat the plasma. In a large device like JET, this toroidal current is about 3-6 million ampere and is induced by transformer action - a current in a primary coil in the centre of the torus induces a current in the plasma, which acts as a secondary winding. The resulting total magnetic field has a helical shape; it winds itself around the torus. A number of other coils generate other, weaker magnetic field components, which further shape and position the plasma in the vessel.

To generate a steady current in the plasma, the current in the primary coil needs to rise continuously. As this is only possible until the maximum allowable current is reached, the plasma can only be confined for a limited amount of time, which means that a tokamak works in a pulsed mode. However, there are ways which make it possible to reach a steady state in a tokamak, which make use of additional techniques to drive the required current. Once the fusion burn is initiated, the external heating can be reduced as the plasma keeps itself hot, just like the flame from a candle. The heating systems can then be used to 'push' the plasma electrons in the right direction, creating a current independent of the transformer. In a commercial power reactor, nearly all of the heating needed to compensate plasma energy losses will come from the helium nuclei, with just a low level of external heating to fine-tune the plasma performance.

The tokamak design, developed by Sacharov and Tamm in Moscow in the 1960s, has been most successfully demonstrated in JET, the Joint European Torus experiment. JET has a plasma volume of approximately 85m³, the central part of which reaches fusion temperatures of 100-200 million °C. In 1997, over 16 MW of fusion power was generated by JET.

Figure 1:Complex magnetic fields confine the plasma

In addition to JET, the world´s largest tokamak, there are a number of medium - sized and small tokamaks in Europe being used to provide additional important experimental data.

Figure 2:Tore Supra © Image: CEA

The Tore Supra tokamak in the French Association Euratom-CEA, for example, is one of the largest tokamaks operating in Europe today and the first one to use a series of superconducting coils to generate a permanent toroidal magnetic field. The capability to run long pulse plasmas on a regular basis with Tore Supra opens the way to explore new scientific questions in ITER-relevant conditions. In 2003, Tore Supra achieved a new world record: a six and a half minutes long plasma discharge sustained by 3 megawatts of current drive power.

Stellarators

The stellarator concept, proposed as early as 1951 by Lyman Spitzer of Princeton University, was initially researched in experiments in the 1950s and 1960s in the USA and Europe. Its name suggests a "star machine", producing energy from fusion. This star machine uses strong magnetic fields to confine the plasma in a torus-shaped vessel, so in this respect, it is similar to a tokamak.

Figure 3:Simulated magnetic fields
for the Wendelstein 7-X © Image: IPP

The difference is that stellarators rely entirely on magnetic fields produced by external coils to produce their magnetic confinement. This eliminates the need for a current in the plasma, but it requires a very complex shape for the coils, as can be seen in the illustrations.

A number of problems that are caused by the large plasma current in a tokamak - such as disruptions and instabilities - either do not occur or are strongly reduced in stellarators.

Figure 4:W7-X magnets have a complex shape.
Image: IPP

A stellarator is much more complex than a tokamak to design and build - only in relatively recent times have the theoretical understanding, computing power and fabrication techniques been developed to the level needed to construct large stellarators with a performance comparable to tokamaks. This line of investigation is actively pursued, since researchers realise that plasma confinement without a large plasma current could have major long-term and economic benefits for the steady-state operation of a fusion power plant. In particular, it could substantially increase the plasma stability and thus simplify the control and monitoring of the burning fusion plasma.

In addition to offering the potential for steady-state, continuous operation, research on stellarators also helps our understanding of the physics of plasma confinement systems. In Europe, two Stellarator projects have been built, TJII in Spain and Wendelstein 7-AS in Germany. TJII is the only stellarator currently in operation but a new larger stellarator, Wendelstein 7-X, is under construction in Germany.

Spherical Tokamaks

In a spherical tokamak the plasma is confined in basically the same way as in a conventional tokamak. The main difference, as the name suggests, is in the shape of the plasma vessel and the magnetic field. A spherical tokamak looks like a sphere with a hole through the middle rather than the doughnut shape of a conventional tokamak. In technical terms, this means that the aspect ratio of a spherical tokamak - the ratio of the device´s width to its height - is low than the aspect ratio of a tokamak.

Figure 5:The plasma shape of spherical tokamaks compared
to a conventional tokamak.
© Image: CCFE

Spherical tokamak research is a relatively young subject, and is therefore much less developed than the conventional tokamak. Nevertheless, it is moving forward rapidly, and there are indications that some of the theoretically predicted advantages may be achievable. For example, it has been shown that a spherical tokamak can operate with a much higher plasma pressure (in relation to the toroidal magnetic field strength) than a conventional tokamak - a potential cost-saving advantage for a fusion reactor.

Figure 6:Spherical tokamak configuration
© Image: CCFE

Figure 7:The MAST spherical tokamak.
Image: CCFE

Reversed Field Pinches

The Reversed Field Pinch concept is a design very much like a tokamak, but the device uses a slightly different geometry of the magnetic field. The configuration gets its name from the fact that the toroidal magnetic field in the outer region of the vessel is reversed in direction with respect to its direction near the axis of the machine.

Reversed Field Pinch systems are being studied for possible low magnetic field, high plasma density confinement designs and to enhance understanding of the physics of toroidal confinement in operating regions that are outside the range of standard tokamak devices.

There are two RFP devices in the EU, RFX in Italy and EXTRAP T2 in Sweden which contribute to studies on achieving high-performance operation and controlling plasma modes.

Inertial Confinement Fusion

Figure 8:The LULI2000 experimental
chamber at Ecole Polytechnique, France

Inertial Confinement fusion, the main alternative to magnetic confinement fusion, follows a very different concept. A small pellet (~1 mm diameter) of frozen fusion fuel is flash-irradiated from all sides with a number of extremely intense laser beams. The outer layer of the pellet is blown away, which causes the inner part of the pellet to be compressed with great force: the pellet is compressed to a density of around 1000 times greater than its normal density in the solid state. This causes an inward push of hot plasma generated at the surface, which causes the temperature and pressure in the core of the pellet to rise to fusion conditions.

The main challenge in inertial fusion is achieving a powerful and homogenous irradiation of the pellet at a high repetition rate: about 10 - 20 pellets would have to be heated and burned per second to provide steady state power in a fusion power plant based on this principle.

Figure 9:Twin laser chains at LULI2000
deliver 2 kJ of energy in a few
nanoseconds

This technology has applications in nuclear weapons development and, because of the classified character of much of the research, the Council of EU Ministers have only foreseen a "keep-in-touch activity" for inertial confinement fusion within the European Fusion Programme. This monitoring activity is maintained by a number of competent laboratories supplying regular updates on progress in this field.