No abstract

Three multiway calibration techniques have been applied for determining of the suppressor concentration in industrial copper electrometallization baths used in semiconductor manufacturing. Parallel factor analysis (PARAFAC) for multiway array decomposition coupled with inverse least squares (ILS) regression (PARAFAC/ ILS), direct trilinear decomposition (DTLD) coupled with ILS (DTLD/ILS), and multilinear partial least squares (N-PLS) regression were employed to develop and test calibration models based on trilinear AC voltammetric data. All techniques employed comparatively produce reliable calibration model and provide quantitative information about its robustness.

Several multiblock and hierarchical calibration techniques have been applied to the determination of the brightener concentration in industrial copper electroplating baths. Hierarchical Principal Component Regression (HPCR), Hierarchical Partial Least Squares (HPLS), Consensus PCR (CPCR), and Multiblock PLS (MBPLS) were used to build and test calibration models based on AC voltammetric data. All of the hierarchical and multiblock techniques employed produce a reliable calibration model and provide information about its robustness.

on other chemical reactions and especially on the luminescence of single pulses are necessary to elucidate the details of the chemical processes occurring during bubble nucleation and disto intensities below 3 W/cm2. Chemical effects at intensities of some IO W/cm2 have been observed by various authors, mainly by using transducers with a horn. It is not yet known quantitatively how the chemical yield responds to pulsing at these high intensities. Only two investigations at higher intensities have yet been carried out which showed that the time interval between pulses also determines the efficiency of the pulses under these condition^.^^^' .4cknowledgment. The authors thank Dr. Eberhard Janata for supported by Deutsche Forschungsgemeinschaft and Fonds der appearance in ultrasonic fields. The present studies were limited construction of the electronic equipment. This work wasIndustrie.The theory of square-wave voltammetry is extended to include the ECE mechanism in which an intermediate chemical step occurs between sequential electrode reactions. Exemplary calculations show how the thermodynamic and kinetic parameters of this mechanism affect the shapes of square-wave voltammograms. The theory is used with a nonlinear least-squares technique to obtain first-order rate constants for the chemical step of the reduction of p-nitrosophenol. This procedure discriminates against several experimental artifacts and does not require the usual separate determination of a normalization current. The rate constants obtained for the intermediate chemical step are 0.46 s-l at 25 "C, 2.40 s-' at 50 "C, and 17.3 s-I at 80 "C (12% uncertainty at 95% confidence) in good agreement with previous studies.

No abstract