Appeal No. 1999-2115 Application No. 08/724,574 art would have been motivated to make this substitution to the process of Yu et al. because of the teaching of Rao et al. that glass reflowing smooths [sic, smoothes] the profile of the glass and leads to higher production yields." (Id.) The problem with the examiner's analysis, however, is that Yu '843 obtains a flat surface without the need for reflowing any dielectric layer. As we discussed above, Yu' 843 teaches that a flat surface is obtained by modifying the polishing rate of selected areas. Furthermore, Rao is directed to a method unrelated to a CMP process as described in Yu '843. Specifically, Rao describes controlling the profile of an interlevel dielectric prior to a reflowing step. (Column 2, lines 42-45.) According to Rao, the interlevel dielectric is deposited to a thickness significantly greater than needed, and then etched back to the desired thickness. (Column 2, lines 45- 47.) The reflowing step in Rao is said to profile the contact hole sidewalls and smooth out the profile of the dielectric interlevel over a patterned stack 24 as shown in Figure 2. Here, the examiner has not pointed to any disclosure in Rao even hinting that the reflowing step can be used in lieu of the process steps described in Yu '843 for the purpose of eliminating scratches following the CMP process, much less any 6Page: Previous 1 2 3 4 5 6 7 8 9 10 11 NextLast modified: November 3, 2007