Abstract

This thesis focuses on copper surface chemistry in response to various additives used in the chemical mechanical planarization (CMP) process. Chemical mechanical planarization (CMP) is an important process in the manufacturing of integrated circuits. Polishing slurries are used to planarize each layer of the circuit using many different components including oxidizers, electrolytes, chelating agents, abrasives, and corrosion inhibitors. CMP is a process that has been developed empirically in industry and few details are known about the actual mechanisms involved.

The first project investigated the effect of different anions in solutions containing benzotriazole (BTA) on copper removal rate. Corrosion inhibitor films thicknesses containing different anions were measured utilizing atomic force microscopy (AFM). Films grown from halide-containing solutions were found to be thicker than those grown from other anions. Ellipsometry was used as a secondary method of measuring film thicknesses and showed trends similar to AFM data. Films were also characterized with surface-enhanced Raman spectroscopy, which showed close association of BTA with the copper surface, and mass spectrometry showed halide inclusion in the Cu-BTA polymer. Open circuit potential measurements were also made to monitor copper oxide formation in halide and non-halide containing solutions. It was concluded that films grown in halide solutions have halide-inclusion and are thicker than those grown in other electrolytes. The thicker films correlated well with reduction in copper removal rates.

The second project studied the electrochemical reduction of hydrogen peroxide on a copper surface in acidic sulfate solutions using cyclic voltammetry, rotating disk electrode experiments, surface-enhanced Raman spectroscopy, and density functional theory calculations. Hydrogen peroxide is a commonly used oxidizing agent in copper CMP. The spectroscopy revealed that the hydrogen peroxide molecule was reduced at negative potentials to form a Cu-OH surface species in acidic solutions, a result consistent with the insight from Tafel slope measurements. Tafel slope plots were derived from data from the rotating disk electrode experiments and suggested that electron transfer is not the rate determining step for peroxide reduction. Density functional theory calculations support the instability of peroxide relative to the surface-coordinated hydroxide on both Cu(111) and Cu(100) surfaces.

A third project investigated the effect of structure of corrosion inhibitors benzotriazole and 1,2,4-triazole (TAZ) on copper removal rate during CMP. Removal rates were higher for solutions containing TAZ than solutions containing BTA. The corrosion inhibitor films were characterized using AFM, cyclic voltammetry, impedance spectroscopy, surface-enhanced Raman spectroscopy, and mass spectrometry. Inhibitor films formed from TAZ were thicker, more permeable, and rougher than films formed from BTA. The addition of glycine to these solutions showed an increase in corrosion for the TAZ-covered surface, little change to the BTA-covered surface, and an increase in removal for both systems. This study correlated removal rate with the physical properties of these two different corrosion inhibitors.