the lower panel, one sees that nitrogen puffing causes a strong reduction of plasma
ionization source and, correspondingly reduction of the plasma flux to the targets,
whereas the recombination sink remains small.
However, what to do with experimental data from linear divertor simulators [40–
42], which seem to show that neutrals play an important role in the reduction of
plasma flux to the target? We notice that in these experiments, the plasma was
produced by the source situated in a separate chamber and only some portion of the
generated plasma was flowing through the orifice into the working chamber. Therefore, the flux to the end target was significantly impacted by the neutral density in the
working chamber. In addition, cross-field plasma transport in these experiments was
relatively large and, for example, in [40, 41] it was concluded that cross-field plasma
transport is the main reason for the reduction of the plasma flux to the end target. As
we see, the conditions of the plasma flow to the target in linear divertor simulators
are very different from the tokamak ones, where all generated plasma particles are
supposed either to flow to the plasma-facing components or to recombine volumetrically. Therefore, even though the experimental data obtained in linear divertor
simulators on such issues as atomic physics and material erosion (see Chap. 3)
appear to be relevant for the edge plasma conditions in fusion devices (e.g. see [71–
73] and the references therein), the results on plasma detachment cannot be transferred directly to the tokamak experiments.
Even though the simple physical picture, boiled down to Eq. (9.14), for the
plasma flux to the target in cold divertor plasma allows explaining the key experimental observations, it only describes the integral plasma flux to the plasma-facing
components. However, from Fig. 1.5b one sees that the divertor plasma detachment
process does not happen uniformly over the entire divertor target. Instead, it starts
from some particular flux tubes. Therefore, we need to find some local conditions for
the onset of detachment, which we define as the beginning of the rollover of the
specific plasma flux j d . Following [74] we will use the same concept of the “closed
box” for some particular magnetic flux tube, which we used for the analysis of the
self-sustained tokamak divertor plasma oscillations. We notice that such an approach
might be not directly suitable for the onset of divertor detachment in stellarators
having a complex divertor magnetic geometry.
In the previous subsection, we found that for high recycling conditions, the
plasma in the flux tube is sustained largely by the plasma recycling processes in
the divertor region. However, the total pressure P
tot
ft within the flux tube, which can
be supported by recycling, is limited by the power needed for hydrogen ionization
and is determined by Eq. (9.7), which can be expressed as follows
P
tot
ft
e
< P
tot
ft
À Á
max
¼ q recycl
ffiffiffiffiffiffiffiffiffiffi
M
γE
H
ion
s
,
ð9:15Þ
where q recycl ¼ q ft À q imp . In [74] it was shown that any further increase of P
tot
ft would
result in a sharp reduction of the plasma temperature in the vicinity of the divertor
246
9 Physics of Some Edge Plasma Phenomena
ionization source and, correspondingly reduction of the plasma flux to the targets,
whereas the recombination sink remains small.
However, what to do with experimental data from linear divertor simulators [40–
42], which seem to show that neutrals play an important role in the reduction of
plasma flux to the target? We notice that in these experiments, the plasma was
produced by the source situated in a separate chamber and only some portion of the
generated plasma was flowing through the orifice into the working chamber. Therefore, the flux to the end target was significantly impacted by the neutral density in the
working chamber. In addition, cross-field plasma transport in these experiments was
relatively large and, for example, in [40, 41] it was concluded that cross-field plasma
transport is the main reason for the reduction of the plasma flux to the end target. As
we see, the conditions of the plasma flow to the target in linear divertor simulators
are very different from the tokamak ones, where all generated plasma particles are
supposed either to flow to the plasma-facing components or to recombine volumetrically. Therefore, even though the experimental data obtained in linear divertor
simulators on such issues as atomic physics and material erosion (see Chap. 3)
appear to be relevant for the edge plasma conditions in fusion devices (e.g. see [71–
73] and the references therein), the results on plasma detachment cannot be transferred directly to the tokamak experiments.
Even though the simple physical picture, boiled down to Eq. (9.14), for the
plasma flux to the target in cold divertor plasma allows explaining the key experimental observations, it only describes the integral plasma flux to the plasma-facing
components. However, from Fig. 1.5b one sees that the divertor plasma detachment
process does not happen uniformly over the entire divertor target. Instead, it starts
from some particular flux tubes. Therefore, we need to find some local conditions for
the onset of detachment, which we define as the beginning of the rollover of the
specific plasma flux j d . Following [74] we will use the same concept of the “closed
box” for some particular magnetic flux tube, which we used for the analysis of the
self-sustained tokamak divertor plasma oscillations. We notice that such an approach
might be not directly suitable for the onset of divertor detachment in stellarators
having a complex divertor magnetic geometry.
In the previous subsection, we found that for high recycling conditions, the
plasma in the flux tube is sustained largely by the plasma recycling processes in
the divertor region. However, the total pressure P
tot
ft within the flux tube, which can
be supported by recycling, is limited by the power needed for hydrogen ionization
and is determined by Eq. (9.7), which can be expressed as follows
P
tot
ft
e
< P
tot
ft
À Á
max
¼ q recycl
ffiffiffiffiffiffiffiffiffiffi
M
γE
H
ion
s
,
ð9:15Þ
where q recycl ¼ q ft À q imp . In [74] it was shown that any further increase of P
tot
ft would
result in a sharp reduction of the plasma temperature in the vicinity of the divertor
246
9 Physics of Some Edge Plasma Phenomena