Appeal No. 1998-1578 Application No. 08/543,827 Neither Larson nor Scott describes a step of annealing the titanium nitride barrier layer under the condition recited in appealed claim 1 or 14 before depositing the platinum layer. However, Ho teaches as follows: Titanium nitride (TiN) is used as a barrier layer. However, TiN as sputtered suffers from two defects. First, TiN has a columnar structure. If a transmission electron micrograph were taken of the TiN, the TiN would appear as groups of columns or grains. The gaps between the columns are referred to as grain boundaries. Grain boundaries cause problems with barrier layers. The grain boundaries form a path through which the metal can migrate to reach the underlying substrate. If the metal migrates through the barrier layer, spike formation or metal diffusion into the substrate can occur. Second, sputtering TiN itself causes a problem. A simple overview of sputtering techniques will indicate how the problem develops. A sputtering chamber is comprised of the following parts: a substrate, a target, the sputtering chamber itself, gases, and a power generator. The power generated can be direct current (DC), radio frequency (RF), etc. The generator ionizes the gas to form a plasma. The plasma is directed toward the target. In this case, reactive sputtering is utilized. The nitrogen in the plasma reacts with the surface of a titanium target to form a thin layer of TiN. In addition, the plasma hits the target causing the TiN to be stripped away from the target. The TiN coats the substrate and the walls of the sputtering chamber. Sputtering has problems. If the sputtering occurs faster than the plasma reaction at the target (converting the surface titanium to TiN), some titanium will be sputtered form [sic, from] the target before it si [sic, is] converted to TiN. The 14Page: Previous 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 NextLast modified: November 3, 2007