Beryllium as the plasma-facing material in fusion energy systems—experiments, evaluation, and comparison with alternative materials

AbstractThe history of plasma-facing materials in fusion systems is reviewed to explain the evolution of needs and requirements that has led to the interest in beryllium today. Critical fusion-related characteristics of beryllium, including outgassing properties, plasma-driven erosion, surface morphology modification, and post-bombardment deuterium retention have been investigated under high-flux, steady-state deuterium plasma bombardment conditions. In experiments with beryllium, hydrogen and water molecules are observed to be the major desorbed species of as-received beryllium. Different beryllium samples manufactured using sintered powder metallurgical methods show similar outgassing behavior. However, cast beryllium samples desorb 20 times less gas than sintered beryllium. Simulating the divertor plasma conditions of the International Thermonuclear Experimental Reactor (ITER), the PISCES-B plasma generating facility is used to bombard targets with a deuterium plasma. The sputtering yield of beryllium is measured using a weight loss method. The target temperature and the partial pressure of residual gases during plasma bombardment are observed to be the critical parameters affecting erosion behavior. Data on sputtering yield for room temperatures (about 40°C) and high temperatures (above 250°C) samples is reported. It is also observed that at the higher temperatures, impurities from the plasma having low atomic number (such as carbon, nitrogen, and oxygen) are deposited on the beryllium surface. An impurity layer forms which is not eroded by extended plasma exposure. In-situ emission spectroscopy shows that the Be II line intensity drops by an order of magnitude as the deposited layer begins to form, and remains low during plasma exposure. This suggests that the impurity layer acts to shield the beryllium surface, thereby suppressing the erosion of beryllium. Estimates indicate that this film reduces the net erosion yield of beryllium by up to two orders of magnitude compared to a pure unprotected beryllium surface. On the other hand, no impurity deposition is observed when the beryllium target is maintained at room temperature (∼40°C) and the impurity concentration is kept at 1%. Depth-profiling together with auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS) measurements demonstrate that the impurity layer is composed of beryllium oxide, amorphous carbon, and small amounts of beryllium carbide. Post-bombardment deuterium retention is measured using a thermal desorption spectrometry (TDS) technique. This data shows that deuterium desorption from plasma-exposed beryllium is quite different if a surface impurity layer is present. The shape of the desorption spectra of HD and D2 emitted from beryllium with the impurity layer present suggests that there are two separate release processes, one from the impurity layer and another from the bulk. Also, most deuterium is released as HD and D2 if thermal desorption is performed immediately after plasma bombardment. However, if the specimen is held in air for an extended period after plasma exposure, a significant amount of deuterium is released as D2O and CD4. Finally, a comparison of the erosion behavior and deuterium retention properties of beryllium, carbon and tungsten is presented.