STS etcher question

Angie McConnell angela.mcconnell at stanford.edu
Wed May 11 13:32:37 PDT 2005


An additional problem is that the silicon process wafer is bonded to a 
quartz wafer.  This further increases the thermal resistance between the 
process wafer and the chuck since the thermal conductivity of quartz is at 
least one order of magnitude smaller than that of silicon.  Is there a 
particular reason you need to use quartz?  I have used silicon backing 
wafers glued on with photoresist dozens of times without photoresist 
burning problems.

Angie McConnell


At 12:03 PM 5/11/05, Jim McVittie wrote:
>Jim,
>
>During plasma etching there is significant heat flow into the top surface of
>your wafer from the ion bombardment. The wafer temperature will rise until 
>there
>is a balance between the heat flow into the top and out the bottom. If the 
>wafer
>temperature goes too high ( about 110C range) the resist can flow and/or burn.
>To improve heat flow out the bottom of the wafer, the space between the wafer
>and cooled chuck is fill with He gas at a pressure around 10 torr. The STS
>etcher is designed to have about a 50C temperature difference between the 
>cooled
>chuck and the wafer. When you mount a wafer on a carrier wafer, you are
>increasing the thermal resistance between the top wafer and the chuck. If the
>resist is thin and there are no gas bubbles, the increase in thermal 
>resistance
>is not too high and the resist does not burn. However, if mounting of the
>carrier wafer is not great and you have bubbles, you are going to have heating
>problems during a long etch. One approach to minimizing this problem is to use
>the IR camera and to make sure there are no gas bubbles between the wafers.
>Also have an experienced user show you the best procedure for the mounting.
>
>     Jim
>
>Steve Zhuang wrote:
>
> > Hi all,
> >
> > I have a question concerning overheating and resist burning in the STS
> > etcher. I STS etched from the back side of a silicon wafer that was bonded
> > to a quartz wafer using photoresist. The photoresist adhesive was uniform
> > and baked property. However, during the 1 hour etching time, the 
> photoresist
> > mask on the silicon wafer was partially burned, and the adhesive 
> photoresist
> > between the silicon and quartz wafers became very non-uniform. I think this
> > problem might have been caused by overheating during etching. Does anyone
> > have similar experiences? Your input is highly appreciated!
> >
> > Thanks
> >
> > Steve Zhuang




More information about the stsetch mailing list