Appeal No. 2006-1663
Application No. 09/871,883
We agree with Appellants’ ultimate position that the § 103(a) rejection over
Farrar in view of Otsuka is improper. Otsuka teaches that the insulating pillars P
(i.e., dielectric pillars) are placed around a via hole to prevent large crystal growth
in the conductor layer. (Otsuka, col. 11, ll. 33-36). Otsuka discovered that when
pillars are not used, the smaller crystals with higher surface energy in the via
encourage the metal atoms to diffuse out of the via toward the larger crystals with
lower surface energy in the conductor layer thereby forming a void in the via.
(Otsuka, col. 6, l. 57 to col. 7, l. 6). In contrast, Farrar uses diffusion barrier layers
(i.e., Fig. 3K, reference numerals 314 (i.e., lower conductive liner) and 334 (i.e.,
upper conductive liner)) to prevent or slow the electromigration of copper atoms
which ultimately forms an undesirable void. (Farrar, col. 1, ll. 45-52, col. 3, ll.
52-56, col. 15, ll. 9-17, col. 16, l. 12 and col. 19, ll. 42-43). Thus, Otsuka uses a
completely different methodology and structure to control electromigration (i.e.,
insulating pillars (i.e., dielectric pillars) for controlling crystal growth in the
conductor layer to control electromigration) than Farrar (i.e., barrier layers to
prevent electromigration).
Accordingly, we agree with Appellants’ argument that including Otsuka’s
insulating pillars to control crystal growth, and thereby control electromigration,
with Farrar’s interconnect structure would be unnecessary because Farrar already
provides barrier/adhesive layers 314 and 334 to prevent electromigration. Farrar
already provides a solution to the electromigration problem (i.e., the
barrier/adhesive layers 314 and 334) which militates against combining Otsuka’s
insulating pillars to solve a non-existent electromigration problem in Farrar as
Appellants advocate. (Br. 27).
24
Page: Previous 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Next
Last modified: November 3, 2007