M.J. Baldwin

Observations of suppressed retention and blistering for tungsten exposed to deuterium–helium mixture plasmas

Blister formation and D retention in W have been investigated for low energy (~55 ? 15?eV), high flux (~1022?m?2?s?1), high fluence (?4.5 ? 1026?m?2) ion bombardment at moderate temperature (~573?K) in mixed species D+He plasmas in the linear divertor plasma simulator PISCES-A. The amount of D retained in W is found to decrease significantly when compared with that in W exposed to pure D plasmas, as measured with high resolution thermal desorption spectroscopy. Scanning electron microscopy observations reveal the suppression of the blisters, a surface feature known to drive up retention, in the D + He mixture plasma exposed W samples. Reduced D retention is accompanied by the formation of nano-sized high density He bubbles in the near surface, observed with a transmission electron microscope (TEM). It is believed that the nano-bubbles act as a diffusion barrier to implanted D atoms and consequently reduce the amount of uptake in the W material. This newly observed effect implies that current predictions of D retention in W, in actual fusion devices, may be overestimated, since there will be He ash in fusion plasma. Toughness enhanced, fine-grained (grain size of ~1??m) W?TiC samples, exposed to pure D plasma conditions, also show little or no evidence of blistering. The measured D retention in the W?TiC samples was approximately 1 ? 1019?D?m?2 corresponding to about 2 ? 10?7 of the implanted D fluence, and is very low compared with the retention in pure stress relieved W, which exhibited surface blisters and had a D retention of about 1 ? 1021?D?m?2.

Deuterium retention in liquid lithium

Measurements of deuterium retention in samples of lithium exposed in the liquid state to deuterium plasma are reported. Retention was measured as a function of plasma ion dose in the range 6?1019?4?1022 D atoms and exposure temperature between 523 and 673?K using thermal desorption spectrometry. The results are consistent with the full uptake of all deuterium ions incident on the liquid metal surface and are found to be independent of the temperature of the liquid lithium over the range explored. Full uptake, consistent with very low recycling, continues until the sample is volumetrically converted to lithium deuteride. This occurs for exposure temperatures where the gas pressure during exposure was both below and slightly above the corresponding decomposition pressure for LiD in Li.

Tungsten nano-tendril growth in the Alcator C-Mod divertor

Growth of tungsten nano-tendrils (?fuzz?) has been observed for the first time in the divertor region of a high-power density tokamak experiment. After 14 consecutive helium L-mode discharges in Alcator C-Mod, the tip of a tungsten Langmuir probe at the outer strike point was fully covered with a layer of nano-tendrils. The thickness of the individual nano-tendrils (50?100?nm) and the depth of the layer (600???150?nm) are consistent with observations from experiments on linear plasma devices. The observation of tungsten fuzz in a tokamak may have important implications for material erosion, dust formation, divertor lifetime and tokamak operations in next-step devices.

Effect of He on D retention in W exposed to low-energy, high-fluence (D, He, Ar) mixture plasmas

W targets are exposed at fixed temperature in the range ~420–1100 K, to either pure D2, D2–δHe (0.1 < δ < 0.25), or D2–δHe–γAr (γ = 0.03) mixture plasma, or He pretreatment plasma followed by exposure to D2 plasma. A strong reduction in D retention is found for exposure temperature above 450 K and incident He-ion fluence exceeding ~1024 m−2. Reduced D retention values lie well below that measured on D2 plasma-exposed reference targets, and the scatter in retention values reported in the literature. A small level of Ar admixture to D2–0.1He plasma, leading to an Ar ion density fraction of ~3%, is found to have minimal effect on the D inventory reduction caused by He. In targets with reduced inventory, nuclear-reaction analysis reveals shallow D trapping (<50 nm), in the same locale as nanometre-sized bubbles observed using transmission electron microscopy. It is suggested that near-surface bubbles grow and interconnect, forming pathways leading back to the plasma–material interaction surface, thereby interrupting transport to the bulk and reducing D retention.

An equilibrium model for tungsten fuzz in an eroding plasma environment

A model equating the growth rate of tungsten fuzz on a plasma-exposed surface to the erosion rate of the fuzzy surface is developed to predict the likelihood of tungsten fuzz formation in the steady-state environment of toroidal confinement devices. To date this question has not been answered because the operational conditions in existing magnetic confinement machines do not necessarily replicate those expected in future fusion reactors (i.e. high-fluence operation, high temperature plasma-facing materials and edge plasma relatively free of condensable impurities). The model developed is validated by performing plasma exposure experiments at different incident ion energies (thereby varying the erosion rate) and measuring the resultant fuzz layer thickness. The results indicate that if the conditions exist for fuzz development in a steady-state plasma (surface temperature and energetic helium flux), then the erosion rate will determine the equilibrium thickness of the surface fuzz layer.

Codeposition of deuterium with ITER materials

The levels of retention in codeposited layers of each of the three ITER materials (C, Be and W) are compared. Scaling laws, based on the conditions during the codeposition process (surface temperature, incident particle energy and ratio of the depositing fluxes), are presented to allow prediction of expected retention under ITER conditions. Retention in carbon codeposits scales inversely with incident particle energy, whereas in the metallic codeposits the retention level scales proportionally to increasing particle energy. The differing scaling of retention with incident particle energy provides insights into which material may impact the global retention in ITER depending on where it may form codeposits. In addition to the amount of retention, the release behaviour of tritium from codeposits will influence the tritium accumulation rate within ITER. The thermal release behaviour of T (or D) from codeposits can be used to evaluate the effectiveness of baking at different temperatures as a means of tritium removal. Finally, the desorption kinetics from Be and W codeposits are contrasted. In the case of W codeposits, the duration of the baking cycle is important in determining the removal efficiency, whereas with Be codeposited layers, the maximum achievable bake temperature plays the leading role in determining removal efficiency.

An empirical scaling for deuterium retention in co-deposited beryllium layers

Different mechanisms contribute to tritium retention in ITER, amongst which co-deposition with materials from the plasma-facing components is one of the main contributors. A systematic study of the influence of the deposition conditions (substrate temperature, deposition rate, energy of the incident particles) on the deuterium retention in co-deposited beryllium layers has been carried out in PISCES-B. The mechanism by which deuterium co-deposits with beryllium appears to be a combination of co-deposition and implantation, with a decreased retention for increased deposition rate and an increased retention for increased incident deuterium particle energy. A scaling equation is developed, providing a method to predict the retention in Be co-deposits formed in PISCES-B as a function of the layer formation conditions. Using this equation, previously published data on retention in Be co-deposits are re-examined and relatively good agreement is found with the prediction of the scaling equation.

Progress toward fabrication of graded doped beryllium and CH capsules for the National Ignition Facilitya)

Current ignition designs require graded doped beryllium or CH capsules. This paper reports on the progress toward fabricating both beryllium and CH capsules that meet the current design criteria for achieving ignition on the National Ignition Facility (NIF) [S. Hann et al., Phys. Plasmas 12, 056316 (2005)]. NIF scale graded copper doped beryllium capsules have been made by sputter coating, while graded germanium doped CH capsules have been made by plasma polymer deposition. The sputtering process used for fabricating graded beryllium shells was produced with a void fraction of ∼5%. Varying the deposition parameters can lead to several different beryllium microstructures, which have been tuned to reduce the void size and fraction to within specifications. In addition, polishing of beryllium-coated shells reduces the outer surface roughness of shells to ignition specifications. Transmission electron microscopy has been used to characterize void fraction and grain structure of beryllium coatings. The plasma ...

Tungsten 'fuzz' growth re-examined: the dependence on ion fluence in non-erosive and erosive helium plasma

The thickness x, of tungsten fuzz layers are measured for non-varying helium (He) plasma exposure conditions spanning four orders of ion fluence (1024–1028m−2) and flux (1019–1023 m−2 s−1), at 1000–1140 K under low energy He ion impact (50–80) eV. The data obtained are complemented by previously published data of similar growth conditions, and collectively analysed. The new analysis allows for the reconciliation of fast high flux growth with commonly observed slower growth at lower flux. It is demonstrated that the standing t1/2 time dependence is a special case of a more general expression for determining the layer thickness, , that depends on , an incubation fluence , and the growth constant m4, which is temperature dependent. The incubation fluence, which must be exceeded before the observation on the onset of fuzz surface morphology is determined to be m−2. In fuzz growth-erosion regimes, characterized by an erosion constant , that is proportional to the sputter yield, an analytic solution for has been found, by solving the growth-erosion equilibria problem of prior work with the Lambert W function. Simple limit expressions follow from the solution for determining the equilibrium fluence and fuzz thickness; the predictions of such being in good agreement with previous fuzz growth-erosion equilibria results in the literature.

Spectroscopic determination of the singly ionized helium density in low electron temperature plasmas mixed with helium in a linear divertor plasma simulator

The spectroscopic method is developed to obtain the He+ ion density nHe+ in low electron temperature, Te=5–20eV, plasmas mixed with He. Plasmas were produced in the PISCES-B linear divertor plasma simulator [R. P. Doerner et al., Phys. Scr. T111, 75 (2004)] where the electron densities are ne=(1−15)×1018m−3 and the ionization degree is ∼1–10%. In the method, the He I line intensity IHeI at λ=447.1nm is used, instead of the He II line intensity in the conventional method. The radial confinement time of He+ ions is requisite, and is measured to be at a level of the Bohm confinement time. The He+ ion concentration, nHe+∕ne, is found to be proportional to IHeI, and to weakly depend on ne and Te. Because of the higher ionization energy of He than other species (D2, Ne, and Ar), the measured nHe+∕ne becomes systematically lower than the He gas pressure fraction, and agrees with data from an omegatron mass spectrometer. The omegatron measurement and estimates of the He+ ion loss rates indicate that the influence...