Week 4 Supplemental: Polysilicon Etch (Oct 27, 1995)

Objective: Starting with wafers with poly layer, unpatterned oxide hard mask layer, and patterned photoresist layer, objective is to etch the pattern into the hard mask, remove the photoresist, etch the poly using the hard mask, then remove the hard mask. We should end up with our poly layer fully defined and the wafers ready for ion implantation.

A. Oxide mask etch
Objective: etch through oxide hard mask about 1400A thick.
Using buffered HF solution, expected etch time from chart is 3.5 minutes.
Etch stop criterion is Etch to Bead +10%.
Wafer B was etched first, then wafers E and 3.
Beading observed at 3m50s, etched first wafer for 4m20s.
All wafers beaded at approximately same time, so all were etched for 4m20s.

B. Photoresist Strip
All wafers stripped for 3 minutes in ACT solution, as in Week3, step B.

C. Polysilicon etch
Objective: remove poly in areas exposed by hard mask.
Using solution of HNO3, HF, and Acetic acid.
Expected etch time: 15s.
Wafer B etched for 15s. Features appeared pretty well defined under microscope.
As we were etching, interference fringes were observed which clearly indicated the changing thickness of polysilicon. It was suggested by Chris that we could use this to determine the etch stop time.
Wafer E was etched for 17s, using the interference etch stop criterion.
After examining wafer E, we decided to etch for another 3 seconds, because patches of poly appeared to be present on the surface.
After a total of 20s of etching, the patches appeared to disappear. There was still a whitish discoloration apparent in the NE quadrant of the wafer (Wafer flat = S).
Wafer 3 was etched for 20s. Fringes disappeared after 18 s. This wafer had 3 cycles of color across its diameter before the poly etch.

D. Oxide mask removal
Using same solution as in step A for same time (since oxide thickness is the same)
All wafers etched for 4m20s, simultaneously

E. Inspection
After oxide mask removal, wafer 3 shows a lot of particulates, especially near the edge where the poly layer was peeling. Wafers B and E look nicely uniform for large areas. Some darker patches are present, however.
Wafers 3 and E were examined under the microscope. Wafer 3 contains a high concentration of particles ranging from about 1um to 100um in size. Many large poly areas appear to be peeling, but some smaller features look fine. Wafer E shows some streaks, but features are predominantly clean and crisply defined. A few transistors show some spreading of the moat area.