No abstract

High-resolution scanning electron microscopy has been used to observe the effects of self-ion bombardment on the topography of germanium surfaces. Holes (mean separation 400 Å) appeared in the surface at doses above 2×1015 ions/cm2 (ion energy 50 keV). These enlarged with increasing ion dose and developed into a complex cell structure. This structure underwent no further gross change for doses above 1×1017 ions/cm2, where the mean cell diameter was ∼1200 Å. The surface preparation was found to have no influence on the appearance of this cellular structure. Some specimens were fractured after bombardment to observe a section of the bombarded surface. A porous layer of thickness 2.5 times the projected range (Rp) was observed at doses just below those where changes in surface topography were first observed. At the highest dose (4×1017 ions/cm2) the thickness of this layer was 6Rp. The cellular surface structure was observed at all ion energies chosen so long as the energy deposited in the surface by the ion beam was kept below 0.5 W cm−2. The observed effects can be qualitatively explained by the formation of large voids. When these intersect the surface the effects of sputter etching and redeposition combine to enlarge the cell structure. At surface loadings above 0.6 W cm−2 different topographies were observed, as a consequence of the surface temperature exceeding the amorphous/crystalline transition temperature during bombardment.

No abstract

No abstract

A TeV-range e e ÿ linear collider has emerged as one of the most promising candidates to extend the high energy frontier of experimental elementary particle physics. A high accelerating gradient for such a collider is desirable to limit its overall length. Accelerating gradient is mainly limited by electrical breakdown, and it has been generally assumed that this limit increases with increasing frequency for normal-conducting accelerating structures. Since the choice of frequency has a profound influence on the design of a linear collider, the frequency dependence of breakdown has been measured using six exactly scaled single-cell cavities at 21, 30, and 39 GHz. The influence of temperature on breakdown behavior was also investigated. The maximum obtainable surface fields were found to be in the range of 300 to 400 MV=m for copper, with no significant dependence on either frequency or temperature. Introduction.-The feasibility of a compact e e ÿ linear collider (CLIC) [1] which aims for a center-ofmass energy in the TeV range is studied at CERN within an international collaboration. CLIC is characterized by the choice of a very high accelerating gradient of 150 MV=m, a high operating frequency of 30 GHz, and a two-beam accelerator scheme to produce the necessary rf power. The high-power testing of rf structures is currently being carried out in the CLIC Test Facility (CTF II) [2], a two-beam accelerator providing up to 280 MWof 30 GHz power at a pulse length of 16 ns. The discovery two years ago of substantial damage due to electrical breakdowns in prototype 30 GHz structures at accelerating fields of about 60 MV=m obliged the CLIC team to undertake a more systematic study of the phenomenology of rf breakdown. To complement the rather expensive development and testing of traveling wave CLIC-type accelerating structures, a series of experiments using simple single-cell standing-wave cavities, directly driven by a high-charge electron beam, were performed. This test setup enables very high surface fields to be obtained in the cavity at a well defined location. These tests assume that the maximum electrical field on the surface is the key parameter for breakdown initiation, and enable fundamental questions such as frequency and temperature dependence of the breakdown behavior to be investigated in a relatively simple way.Cavity design and fabrication.-A total of six cavities, two at each of three different frequencies (21, 30, 39 GHz) were made. These high-gradient single-cell cavities have a pillbox-type geometry exactly scaled for the different frequencies for operation with the transverse-magnetic (TM 010 ) mode (see Fig. 1). The scaling factor s 30 GHz =f 0 GHz was applied to the beam-pipe diameter, the cavity length, the cavity diameter, and the radius at the beam-pipe opening. The location of the maximum surface fieldÊ E S in these cavities is also indicated in the figure. A small coupling aperture (about 1 mm wide) was