Assessment of Laser Induced Ablation Spectroscopy (LIAS) as a method for quantitative in situ surface diagnostic in plasma environments

In this work Laser Induced Ablation Spectroscopy (LIAS) is investigated as an in
situ plasma surface interaction diagnostic for fusion reactors and fusion experiments.
In LIAS an intensive laser pulse is used to ablate the material under investigation
during plasma operation. Ablation products penetrate into the edge region of the
plasma and are excited and ionized. In case of molecules and clusters additionally
dissociation occurs. The emitted line radiation is observed by radiometric calibrated
spectroscopy.
Results from LIAS of W/C/Al/D–mixed layers and amorphous hydrocarbon
layers are presented. Using a fast camera system time resolved measurements of the
LIAS–process could be performed, allowing investigation of the temporal behavior
of excitation, dissociation and ionization processes. For Tungsten, 90% of the LIAS
light is observed within 10±3 μs after the laser pulse. In case of carbon within 20±
3 μs. Additionally separation in time of LIAS emission and the LIBS emission caused
by the laser pulse at the surface within single measurements was demonstrated. This
allows the separate analysis of both processes in a coaxial setup which is foreseen
for future experiments.
The inverse photon efficiency of the Balmer Dα –emission from LIAS of a-C: D–
layers was found to be
D
XB
a-C: DL→IASD
Dα
= 71 ± 7.
The plasma perturbation due to LIAS was investigated by laser energy density
variation when ablating W/C/Al/D–mixed layers. Local plasma perturbation is
found to increase with laser energy density. Balmer Hγ/Hδ – line intensity ratio
measurements only show for ohmic discharges and the case of the lowest central
density signs of local plasma perturbation in LIAS of graphite samples.
A simple analytical model for local plasma perturbation during LIAS is introduced
and evaluated. Qualitative agreement between the model and the above reported
experimental observations is found; a stronger influence on local conditions
is found by tungsten than by carbon ablation, with ohmic discharges more susceptible
to perturbation than neutral beam injector heated ones. Limitations and possible
improvements of the model are discussed.
A Monte Carlo code developed in the framework of this thesis is used for modeling
the measured neutral atom emission profiles. The model is in good agreement
with the analytical solution in case of a homogeneous plasma. With the best estimate
input parameters no agreement between observed and modeled emission profiles is
found. Thus, a three dimensional parameter space describing the plasma profile is
defined by density and temperature at the last closed flux surface and the density
decay length. In this parameter space the surface on which measured and simulated
profile emission maxima agree is found for both Tungsten and Carbon. In case
of Tungsten, agreement between measured and simulated emission profile shape is
found for λne = 13mm. In contrast, for carbon no match for the emission shapes
can be found. Taken together with the spectroscopic observation this suggests that
non-atomic species significantly contribute to the observed light emission, creating
the need for extension of the model.
In the concluding discussion the results are discussed and further investigations
are proposed.