XPS, Auger study of Cu3Si and its reaction with oxygen

Carbon mediated reduction of silicon dioxide and growth of copper silicide particles in uniform width channels

“…We therefore ascribe the striking uniformity in average width of the inner channels etched in the silicon oxide to the uniform rate of reduction of silicon oxide by copper silicide over the wafer surface as previously observed. 16 Copper silicide is known to catalyse the reverse reaction of oxidation of silicon to silicon oxide even at room temperature 17 -a process which we believe explains our observation of an indistinct 2-3 lm layer of silicon oxide with a fractal boundary beneath the particle visible in Figure 2(f), similar to previous reports. 18 …”

Reaction of nanometer-sized Cu particles with a SiO2 substrate

“…Strongly bonded Cu may catalyze the Si-O bond breaking, since Cu also catalyzes the reverse reactions of Si-O formation. 13,14,[16][17][18] Metal evaporation as applied by others 9,21 may result in a weaker metal-support interaction 4 and, therefore, not result in a reaction between Cu and SiO 2 during UHV annealing. We note that also impurities on the sample surface or constituents of the residual gas may help to reduce SiO 2 .…”

“…13,14,16 Roomtemperature oxidation of Si is catalyzed by Cu 3 Si and a thick amorphous SiO 2 layer grows spontaneously beneath the Curich layer. 17,18 So the ease of regeneration of the annealed Cu/SiO 2 samples as described above can be understood from the oxidation behavior of a silicide and is another piece of evidence for the presence of Cu silicide after annealing in UHV.…”

“…Cu silicide can be distinguished by the splitting of the Si LVV peak at 92 eV into two peaks at 90 and 94 eV. 13,14 In Fig. 2, AES spectra of the Si LVV region of our Cu/SiO 2 system are presented after various treatments.…”

SEM-EDX and SAM-AES Investigations on Rochow Contact Masses

“…4 Moreover, a high spatial resolution power is required, because the catalytic activity seems to be restricted to distinct micrometer-sized areas on the silicon surface. Several authors have applied modern surface methods to surfaces of massive silicon or massive copper silicides, [5][6][7][8][9] which were expected to model the surface of real powder-like contact masses. However, it is not always certain whether or not such specimens really simulate the respective properties of a real contact mass.…”