Experiments to determine the material removed (sputtering) from the surfaces of metal foils during irradiation with 14.8 MeV (d, t) neutrons will be described. Neutron activation analysis was used to measure the sputtered material. This is an integral technique which determines quantitatively the number of atoms on the collector. Six experiments were done using the Lawrence Livermore Laboratories rotating target (d, t) neutron source and one experiment was done at the (d, Be) source at the University of California, Davis. In each experiment a large number of collector–target assemblies were located at various distances from the neutron source. The collector–target assemblies were housed in a stainless steel ultrahigh vacuum system and all irradiations were performed in a high‐vacuum environment [∠1×10−7 torr to 5×10−9 torr (∠1.3×10−5 Pa to 6.5×10−7 Pa)]. For these experiments the mean (d, t) neutron energy was 14.8 MeV in the forward direction, and for the different beam‐on times and various source to sample distances the target fluences ranged from 1×1014 to 2.6×1016 (d, t) n/cm2. A (d, Be) source is highly anisotropic and is peaked in the forward direction. For the 20μA, 40 MeV deuteron beam used at the University of California, Davis, the average (d, Be) neutron energy was about 15 MeV, and at the point of closest approach to the Be target fluxes of ∠2×1013 n/cm2 s are produced. Neutron sputtering yields were obtained for niobium and gold. Collectors for all of the niobium and gold targets were 2.54‐cm diameter, single‐ or polycrystal wafers made from 9N pure silicon ingots. In all cases the gold targets consisted of Marz grade foils 2.54 cm in diameter and 0.0127 cm thick, vacuum annealed, and etched in aqua regia. The niobium targets were also prepared from high‐purity commercial grade material but had a wide variety of metallurgically different surface preparations. Except for the single crystals all the Nb samples were in the form of foils. Some were completely cold worked while others were vacuum annealed. Surface preparations varied from just chemical etching, to mechanical polishing and etching, and mechanical polishing and electropolishing. In this way a large range of surface topographies was produced. In all, collectors from 25 niobium samples covering 7 irradiations were examined using slow neutron activation analyses. About one‐third of the collectors had no measureable 93Nb. Analyses of the niobium yield data as a function of neutron dose gave a zero slope. This means that our data give upper bounds for the sputtering yields and our best estimate of the neutron sputtering ratio for niobium is ?1×10−5 atoms/neutron. Micrometer size niobium particles were not found even after extensive search with a scanning electron microscope equipped with energy dispersive x‐ray analysis. Table I is a summary of the secondary‐emission monitor (SEM) studies.
Statistical studies of the neutron activation analyses of the gold data show a linear relationship between the gold yield and neutron dose. Also, there is a significant gold background on the collectors. The best estimates of (d, t) neutron gold sputtering ration are ∠1.5×10−5 atoms/neutron in the forward direction and ∠1.0×10−5 atoms/neutron in the backward direction.
Details of these experiments and data analyses will be submitted for publication in the Journal of Applied Physics or Physical Review.