Dear Robert,

Thanks for your advice - I'll try adding CF4 or Ar to the mix.  The ICP tool
is a Panasonic E620 R&D single-chamber, load-locked tool.  In general, does
ICP RIE yield etch profiles that are more vertical as compared to
parallel-plate RIE?  Would I be better off just switching to the ICP tool?

Thanks,

Brian

--
Brian C. Stahl
Graduate Student Researcher
UCSB Materials Research Laboratory
[email protected] / [email protected]
Cell: (805) 748-5839
Office: MRL 3117A


On Wed, Sep 9, 2009 at 1:11 PM, Robert Ditizio  wrote:

> You might consider adding CF4 to your current process to increase your
> oxide etch rate and improve the selectivity to your mask material.  The
> addition of CF4 should also help to improve the profile angle which
> presumably is not so great with CHF3 only.
>
> You will require a selectivity of >2.5:1 to get through your 100nm oxide
> film using an initial mask thickness of 40nm but >3-4:1 is preferable to
> maintain your critical dimensions.  This might not be achievable with
> the hardware configuration that you have available in the RIE etch tool
> but the CF4 additions is probably your best bet given what you have to
> work with.  You don't say what type of ICP you have available but a
> similar approach of using a CHF3/CF4 gas mixture should apply as well
> but excessive source power could result in low selectivity to the mask
> in this reactor.
>
> Adding Ar could help the process as well.  Don't bother with the Cl2 or
> BCl3.  The He and SF6 are probably not going to provide any appreciable
> benefit over the Ar and CF4 so it is best not to complicate your process
> unless you find a limitation that you cannot overcome with the CF4 and
> Ar additions.  Increasing the pressure to 5-10mT might help to improve
> the selectivity as well.
>
> Regards,
> Robert Ditizio
> Tegal Corporation