Abstract - Griffin - 9612157 Chemical vapor deposition (CVD) of copper thin films is being studied as a potential technology for interconnect metallization in future generations of deep sub-micron integrated circuits. As interconnect linewidths decrease and clock frequencies increase, the on-chip resistance-capacitance time delays become the major limitation in achieving high circuit speed. Smaller linewidths also lead to higher current densities, which raises concern about reliability issues such as electromigrations and stress-induced void formation. Copper offers several intrinsic property advantages over aluminum, the current material of choice for metallization. These include lower resistivity, improved electromigration resistance, and increased resistance to stress-induced voidage. This project is an inter-disciplinary, multi-investigator program of combined experimental and theoretical research aimed at further developing the process technology needed for the chemical vapor deposition of copper thin films for ULSI device metallization. This research is aimed at developing a fundamental, quantitative understanding of the reaction mechanism for deposition processes based on Cu(II) reduction chemistry. By studying this group of reactions, the PI s will be exploring an alternative to the Cu(I) disproportionation process. The higher reactivity of the Cu(I) reactants used for the latter process may eventually lead to problems with film nucleation and grain size control, as the microelectronics industry moves progressively toward higher aspect ratios for via fills. The bis(beta-diketonate) reduction chemistry can also provide a more versatile pathway for CVD of other metals (i.e., volatile beta-diketonate compounds exist for most transition metals, while the disproportionation pathway is essentially unique to copper).
