REDUCING ELECTROPOLISHING TIME WITH CHEMICAL MECHANICAL POLISHING

References: [1] http://web5.pku.edu.cn/srf2007/download/proceedings/TU103.pdf [2] Metallographic polishing by mechanical methods, fourth edition, L.E. Samuels, ASM international Ed.(2003) [3] Assessment of the Origin of Porosity in Electron-Beam-Welded TA6V Plates, N. GOURET, et al., Metallurgical and material transactions A 35A, (2004), p.879 [4] Bubble formation in aluminum alloy during electron beam welding, H. Fujii et al., Journal of Materials Processing Technology 155–156 (2004) 1252–1255 [5] Proc. Of the 2nd international colloquium on EB welding and melting, Avignon, (1978) ; Proc. Of the 3rd international colloquium on welding and melting by electrons or laser beams, Lyon, (1983) ; following edition of the same colloquium... [6] E. Herms, Laboratoire d’Etude de la Corrosion Acqueuse, CEA, personnal communication It is commonly admitted that 150-200 μm need to be removed by electropolishing on the internal surface of niobium RF cavities before reaching optimal results. The proposed reason is generally the existence of a damage layer on the Nb sheet surface. Indeed recent disorientation measurement made at Cornell [1] show that hotspots exhibit higher misorientation. Damage has also been considered in the formation of pits close to the welding seam during electropolishing. Removing of 200 μm by electropolishing is a hazardous process, not only because of the dangerous chemicals involved in the process, but also because of the spread of results, probably due to the chemical mixture aging. Reducing the amount of electropolishing to a final light treatment would be a way to decrease both costs and risks of the RF cavities for large projects such as ILC. We have tried to evaluate the thickness of the damage layer after various deformations steps (mainly rolling, deep drawing and chemical mechanical polishing) by observing the density of etching figures after several light chemical etches. This provides a coarse but very rapid evaluation of the thickness of the damage layer. Complementary observations with EPSB are also presented. Finite element, orientation imaging and/or and etching figures show that the damage layer induced by rolling is noteworthy already ~150 μm thick, with a specific (001) texture that resists recrystallization. Deep drawing brings further and deeper damage in particular in the equator region where the friction against the forming dye is the highest. Welding also influences the damage distribution. Getting rid of this damage layer is possible with BCP, but it needs another 100 μm to smoothen the surface afterwards. Mechanical polishing like tumbling obviously leaves a thick damage layer, but “chemical mechanical” polishing is a way to prepare surfaces with a very thin damage layer (< 1μm?). We think that chemical mechanical polishing of half cells before welding would be a way to decrease the thickness of electropolishing necessary for the preparation of RF Nb cavities, and reduce costs and risks. REDUCING ELECTROPOLISHING TIME WITH CHEMICAL MECHANICAL POLISHING