Electrochemical investigation of ion implanted p-Si

The present work investigates the possibility of selective electrochemical metal deposition on ion implanted p-Si. The idea is that defects introduced into the substrate by ion implantation make it more susceptible to electrochemical reactions compared to intact Si; this increased sensitivity is to be used for selective reactions at the defect sites. It is believed that the increased reactivity is due to a lowering of the Schottky barrier breakdown potential, Ubd, of the semiconductor-electrolyte interface due to the introduction of additional energy states in the semiconductor's bandgap. These additional states may be used for facilitated charge transfer by direct tunnelling enabling spatially well defined electrochemical reactions at the implant sites. In a first step the damage created by ion implantation (Ga+ and Au2+ by focused ion beam, FIB; Ga+ by broadband ion implantation, BII) was analyzed and compared to numerical simulations. Optical microscopy, REM, AFM, and Raman spectroscopy were for characterisation. It was found that the damage created was in good agreement with the theoretical models: more damage was created for heavier ions and higher ion doses. AFM proofed to be a valuable tool to assess surface sputtering for high implant doses, while optical microscopy was more sensitive for low doses. Raman spectroscopy was used to determine the degree of amorphization of the substrate. We found that amorphization doses varied considerably depending on the implantation mode used. For BII the dose needed for amorphization was similar to values reported in the literature, while FIB implantation needed doses roughly 50 times higher. We assume that this is due to the very high current fluences in the FIB combined with short pixel dwell times during the implantation process; both these factors promote self healing of the substrate, hence a higher dose is needed for amorphization to take place. Deposition experiments at different potentials and in different electrolytes on samples implanted with varying doses lead to the following findings. Selective electrodeposition on ion implanted p-Si is possible, however, deposition conditions have to be chosen carefully. Low and high (close to or above the amorphization limit) implant doses lead to unsatisfactory deposits (sample not fully covered). We assume that in the first case, not enough defect sites are present for sufficient formation of initial nuclei, while in the second case the amorphous substrate seems to behave as an insulator making any deposition virtually impossible. Also, for the successful deposition of Cu from an acidic electrolyte, benzotriazole, a brightening agent, was added to improve deposit quality. The result were finely grained, even, and well delimited deposits with a resolution matching that of the implant process (~ 200 nm). Deposition time and potential are crucial too, as long times and more cathodic potentials promote outgrowth, while short deposition times and potentials