Nuclear Instruments and Methods in Physics Research B7/8 (1985) 71-75 North-Holland, Amsterdam 71 zyxwvuts GAS REEMISSION AND SURFACE MODIFICATION DUE TO He IMPLANTATION INTO Nb B.M.U. SCHERZER, P. BIZIRGESEN, J. EHRENBERG AND W. MijLLER Max - Planck - Institut zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA ficr Plasmaphysik, EVRA TOM Association, D - 8046 Garching Fed. Rep. Gemany The release of gas from a metal target during monoenergetic He implantation has been found to be closely correlated to blister formation in Ni. However, in earlier work on Nb and MO preblister reetnission was postulated. In the present work the surface topography of Nb is investigated close to the onset of reemission which could be determined exactly by reemission spectroscopy. Blister formation was found immediately after the start of gas memission, suggesting a similar mechanism for reemission as in Ni. The influence of target pretreatment is also described. 1. Introduction Surface modification (blistering, flaking, pm-hole formation) and gas reemission due to He implantation in metals have been studied extensively for many years [l-3]. The correlation between these two phenomena has been a topic of investigation among others by the SANDIA-group in Liver-more f4-131, our group at IPP in Garching [14-161, the JAERl-group in Tokai-mura [17,18], and the group. at the University of Quebec in Varennes [19-221. While in such cases where strong flaking occurs the number of exfoliated layers can be directly correlated to an equal number of peaks in the gas reemission [6-131, the results are less unambiguous on surfaces with unexfoliated blisters. Bauer et al. [4-61 find that the dose required for blister formation is similar to that required for the onset of the low temper- ature gas release. Contrary to this, Behrisch et al. [14] found for l-15 keV He bombardment of Nb and Saidoh et al. [18] for 200 keV He bombardment of MO, respec- tively, at ambient temperature that He reemission starts somewhat before blister formation. Most of the earlier measurements suffer from one or the other of the fol- lowing draw-backs: (1) The target surface in a reemission measurement could only be observed by eye during bombard- ment. Inspection with the microscope was only pos- sible after stopping the bombardment. (2) The homogeneity of the current density over the beam spot was not controlled. (3) The onset of gas reemission was determined rather insensitively from measurements of the gas reten- tion in the target. In a simultaneous observation of gas reemission and surface modification by means of diffuse laser scattering Ehrenberg et al. [15,16] have shown that for He im- planted into Ni blister formation and gas reemission start exactly simultaneously. It should be mentioned 0168-583X/85/$03.30 @ Elsevier Science Publishers B.V. (North-Holland Physics Publishing Division) here that also for hydrogen implantation in metals with low solubility like Cu at room temperature 123,241 or Ni, Pd, and Ta at T = 38 K [25] reemission and blistering start at almost the same fluence. The mechanism of gas reemission is not yet fully understood. The simplest assumption is that the gas contained in the cavity below a blister is released through cracks forming in the blister cap or at its circumference. This picture was modified by Ehrenberg et al. [15,16] because the total amount of gas emitted during the first stage of blister formation is larger than the gas quantity contained under the blisters themselves. It was assumed that large areas in and outside of blisters are intercon- nected laterally by a network of channels as observed by Jagger and Roth [26]. As soon as a crack is formed in one of the blister covers, gas contained in one of these interconnected areas will be released. The existence of cracks in blister covers has not been confirmed experi- mentally for low energy implantation where blisters are quite small. However, if gas reemission would start before blisters appear, other exit channels for reemission must obviously be postulated. The present work was undertaken in order to clarify whether blisters are indeed not observed in Nb im- mediately after the onset of reemission, as claimed by Behrisch et al. [14]. 2. Experiment The experimental set-up has been described in some detail previously [15,16]. The target, which was housed in a box with only small openings for beam inlet and detectors, was bombarded by 12 and 20 keV He ions. The partial pressure inside the box was monitored by means of a quadrupole mass spectrometer (QMS). The beam was swept both horizontally and vertically over a diaphragm of 1.5 mm diameter 27 mm in front of the II. METALS Hydrogen, Helium