In the early 1990s, reports by some IC manufacturers on the use of silicon wafer polishers to reduce the topography in interlevel dielectric (ILD) films created by multiple levels of metallization began to surface. Dismissed at first as irrational, such reports eventually proved to be true as the CMP industry began gaining ground. By the end of the decade, feature densities approaching sub-100 nm demanded not only an ILD CMP but CMP steps in the front end of the line (FEOL) as well. Process designers referred to CMP reverently as an enabling technology because the planarity it afforded was required for submicron lithography and multilevel metallization. But CMP engineers and integrators regarded the new process as more of a necessary evil because of the many challenges associated with controlling the new technology in high volume manufacturing. These conflicting perceptions of CMP as an enabling technology that is difficult to manage have persisted throughout CMP's more than 15 years' history. Table 20.1 shows the technology node and the year in which the manufacturing of key CMP processes was introduced at Intel Corporation. These CMP processes, while enabling different aspects of IC technology, shared several key features. First and foremost, they were all revolutionary technologies in that they (1) significantly advanced in the state of the art, (2) required significant capital investment, and (3) required significant development efforts to make them work. An additional feature that the CMP processes shared was that they were unique solutions as, for most of them, there was no viable alternative to their use-these CMP steps were required.