Flux dependence of helium retention in clean W(110): Experimental evidence for He self-trapping

•Clean W(110) was used to evidence the self-trapping of He in the W bulk at 300 K.•130 eV He+ ions were implanted in W at a constant fluence of 2.0 × 1021 He+m−2.•Only if the He ion flux exceeds 0.7 × 1017 He+m-2s−1, we observed He desorption peaks.•He desorption at 950 K and 1700 K are consistent with self-trapped HenV clusters.•At 5.0 × 1017 He+m-2s−1 new desorption peaks are consistent with trap-mutated HenVm>1. Helium (He) retention in tungsten (W) is a concern in fusion reactors since it could be detrimental to plasma facing components performance and influence the fusion fuel balance. He being not soluble in W, it tends to agglomerate on preexisting defects (vacancy, grain boundary), but it could in theory also self-trap (be immobilized on a non-preexisting vacancy) through the emission of a vacancy/self-interstitial W pair in the vicinity of a Hen interstitial cluster. In the present study, we prepared a pure single crystal W(110) sample with a clean surface in order to evidence the self-trapping of He in the W bulk at a sample temperature of 300 K and for a constant fluence of 2.0 × 1021 He+.m−2. At a He+ kinetic energy of 130 eV and a flux of 0.3 × 1017 He+.m−2.s−1, we only observed a small He desorption peak below 600 K. Rising the ion flux to 0.7 × 1017 He+.m−2.s−1, we observed the sudden appearance of two desorption peaks at 950 K and 1700 K. For the highest flux studied in this work, 5.0 × 1017 He+.m−2.s−1, an additional desorption peak at 1800 K and a desorption shoulder at 1900 K are observed. The temperature position of these He desorption peaks are consistent with the density functional theory literature and points to the occurrence of self-trapping once the 0.7 × 1017 He+.m−2.s−1 flux is attained at 300 K and to the possible realization of trap-mutation for the flux of 5.0 × 1017 He+.m−2.s−1. The present set of results should be used to constrain the development of He retention and He bubbles growth models based on ab initio quantities.