Appeal No. 2005-0123
Application No. 10/114,759
inconsistent. High sheet resistivity makes it
difficult to maintain an equipotential surface.
Thus, in view of appellant's admissions, the skilled artisan
would have realized that Simpson's seed layers of titanium
nitride and tantalum nitride would not work particularly well for
electroplating copper. We note that appellant further states
(specification, page 2) that because the metal nitrides do not
work as seed layers, a copper seed layer is typically used,
deposited by PVD. Thus, given the teachings of Simpson and
appellant's admissions, the skilled artisan would have been left
with a choice of titanium, tantalum, or copper for the seed
layer, with no suggestion to select titanium or tantalum over the
copper.
However, Xu discloses (column 6, lines 47-65) that a prior
art method involves depositing titanium and titanium nitride,
then moving the wafer to a separate chamber for annealing, and
then moving the wafer to yet another chamber for sputtering the
aluminum. Xu states (column 6, lines 63-65) that "[i]t is
greatly desired to provide a simpler process for filling plugs of
high aspect ratios." Xu thus suggests the need for a method
which reduces the number of times that the wafer is moved.
8
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 Next
Last modified: November 3, 2007