An impedance study of the electrochemical double layer of copper in an electro less copper plating bath was performed. Of the bath components (formaldehyde, ethylenediamine-tetraacetic acid (EDTA), CuEDTA 2 -, NaOH, Na ZS04, and NaHCOO) only the influence of hydroxide on the double layer is visible from capacitance data. The estimated surface coverage of hydroxide at the mixed potential of the electroless plating process is about 10%. On copper single crystal surfaces the OH --adsorption shows specific characteristics. Due to a surface reconstruction these characteristics, however, disappear with time after the immersion of the electrode. During the deposition process high electrode capacitances are measured (200-500 J-lF/cm 2 ). They are mainly caused by a surface roughness on the microscopic scale. When the deposition process is terminated, the surface relaxes within about 10 minutes. The addition of cyanide or Gafac (a mixture of organic esters of phosphoric acid) to the electroless plating bath markedly influences the capacitance data. However, always a correlation between the deposition rate and the electrode capacitance is observed. If, on the other hand, at potentiostatic cupric ion reduction formaldehyde in the plating bath is substituted by ethylenediamine, which also catalyzes the CuEDTA2--reduction, the capacitance values decrease with increasing deposition current. Presumably, in the first case the nucleation rate dominates the deposit growth, while in the second case the opposite is true. is of electrochemical nature [3 -5]. A mixed potential is established during the chemical deposition process. Electroless copper plating consists of the electrochemical reactions:[CuEDTA]2-+2H 2CO+40H-------+

The microstructure of copper deposits was in‐situ investigated by impedance measurements and by surface enhanced Raman scattering. The simultaneous determination of the copper deposition rate with a quartz microbalance and the mixed potential under electroless plating conditions allows to separate the two partial processes copper deposition and formaldehyde oxidation. Due to this combination of investigation techniques we are able to correlate responses of the partial processes e.g. caused by variation of the bath components on changes of the surface microstructure of the copper deposit. A microstructure with amplitudes of an atomic scale is only maintained during the electroless deposition process and is a specific catalyst for formaldehyde oxidation. The overall process is found to be a complex heterogeneously catalyzed system, which can be understood in synergetic terms as a selforganizing process.

No abstract