From mcvittie at snf.stanford.edu  Fri Aug  2 16:27:28 2002
From: mcvittie at snf.stanford.edu (Jim McVittie)
Date: Fri, 02 Aug 2002 16:27:28 -0700
Subject: Procedure for Eliminating Grass During Si Trench Etch
Message-ID: <3D4B1560.B6A4AF96@snf.stanford.edu>

P5000 users,

The following pre-conditioning process is useful for minimizing grass
from micromasking
when doing Si trench etching in the P-5000 using an oxide mask and the
HBr/NF3/O2 process in Chamber C.  These results are from the work of
Song Pang and Balaji Venkateshwaran.

1. Run Ch.C-Clean with bare Si wafer for 200s
   (90 sccm NF3/600 mT/350W/55G)
   Wafer should come out clear with no fog.

2. Run 3 patterned seasoning wafers using trench etch process with
   2 min main etch time. The oxide patterned wafers should have about
   the same pattern density as your real wafers. Although you may see
   some grass on the first wafer, the third should be mostly clean of
   grass in the etched areas. If grass is still a problem, the chamber
   most likely needs a wet clean. Note that this procedure was developed

   for wafers with 20 to 40% etched area.

3. Etch wafers.

        Jim McVittie