9.5 Fuel Retention in Tungsten (W)
179
the geometrical surface are saturated with H and then the surfaces of grains beneath
the saturated grains are saturated. Subsequently, the thickness of the saturated layers
of the grains beneath the saturated grains grows until whole volume of these grains
is saturated. The increase in the number of saturated grains results in the growth of
the thickness of the saturated layers as shown in Fig. 9.8. Higher H flux loading
would increase the H concentration in the saturated layers. Simultaneously, surface
temperature rise due to the H loading reduces the saturation concentration. In addition, the temperature rise would enhance grain growth and H penetration as well,
and consequently accelerates the growth of the thickness of the H saturated layers. In
SIMS depth profiles [30], hydrogen retention in form of H 2 molecules are reported,
which is quite natural as the cause of blisters and could be one of the causes of large
data scattering in the H retention data in Fig. 9.11.
It is noted that the H saturation modifies mechanical properties and oversaturated H
would precipitate as H bubbles. As a result, blistering with two different mechanisms
appears (1) accumulation of gas bubbles occurring in very near-surface layers with
smaller radius and high pressure inside and (2) exfoliation of grain boundaries at the
deeper region of the saturated layers with/without slipping of particular plane (most
probably [110] plane owing to the compressive stress in the H saturated layers) as
shown in Figs. 9.8 and 9.9. The formation of the H saturated layers might be different
from simple trapping but can be viewed as the formation of a chemical compound
and the H release is something like the decomposition of the compound. Still the
non-saturation behavior in long-term retention cannot be interpreted simply by the
formation of near-surface saturated layers and their growth. Deep penetration by
diffusion and immobilization by trapping in bulk must give significant contribution.
Neutron irradiation creates trapping sites in deeper region, which trap H penetrated
deep by diffusion to be immobilized and contribute to the non-saturation. This could
result in significant H retention in W which is hard to remove and dissipate the
advantage of W as PFW.
9.6 Comparison of Estimated Fuel Retention in a Reactor
with Full C-Wall and W-Wall
In Fig. 9.7, fluence dependences of the fuel retention rate of the C wall and the
W wall are compared, assuming surface area of 100 m
2 and particle flux of 10
25
m
−2
・t
−1 for the former, while for latter the retention data normalized in unit area
(m
−2 ). The comparison would not be straightforward because the estimation for
the W wall is just the extrapolation of the data obtained for W plates in laboratory
or probe samples set in ASDEX. Nevertheless, the comparison in Fig. 9.7 seems
reasonable because the H retention in W is dominated in bulk and little influence
on erosion and deposition. It should be noted both overlap rather well. And the fuel
retention rate could be kept 1% or less for reactor scale fluence. However, owing to
large error caused by the extrapolation, it is hard to foresee the fuel retention rate of
0.1% or below to attain the fuel self-sufficiency.
Précédent

- 183/209

Suivant