Control of the plasma-wall interaction is a difficult challenge to cope wit
h in controlled fusion research. A straightforward idea consists of destroy
ing the magnetic configuration at the plasma edge by field line stochastiza
tion: this can easily be achieved by applying helical magnetic perturbation
s. Interestingly enough, most of the physics properties stem directly from
the description of the chaotic nature of the field lines. This is readily e
xpected whenever parallel transport largely exceeds transverse anomalous tr
ansport, a standard property for most boundary plasmas. While quasilinear t
heory gives access to the poloidally and radially averaged values of heat a
nd particle energy transport, it can then be shown that the understanding o
f energy transport through the stochastic boundary and especially power dep
osition onto the target plates requires a non-averaged calculation. Indeed,
ergodization is limited by the bounded length of the field lines themselve
s before they hit a plasma facing component within the tokamak chamber. The
so-called ergodic layer is consequently surrounded by a generalized scrape
-off layer: the laminar zone. Towards the plasma core, the vanishing ergodi
zation is shown to give rise to an internal transport barrier.
Outstanding experimental results such as impurity screening and increased r
adiation at reduced core contamination indicate that transport monitored by
a controlled stochasticity provides a powerful means to tackle major issue
s of fusion physics. Significant advances in ergodic divertor physics give
us confidence that this open divertor concept is an alternative to the clos
ed axisymmetric divertor which relies heavily on mechanical baffling of neu
trals. Experimental evidence is given by a worldwide family of devices such
as Tore Supra, TEXT, TEXTOR and CSTN-II.