Dear Lion,

Regarding your question 1), silicion dioxide is a somewhat poor choice
for ebeam evaporation, since the evaporated material will have a
poorly characterized structure and ratio of silicon to oxygen.  Consider
evaporating pure silicon, which I've found to be easier than SiO2 from a
technical standpoint.  You may get a more stable layer this way, and you
can always oxidize the deposited layer to get a silicon oxide.  Also
consider a thinner layer, to reduce stresses, if your process permits.  If
you would prefer to continue using SiO2, consider thermal evaporation.


Darren Gray
Graduate Student
Johns Hopkins University


> Message: 2
> Date: Thu, 12 Feb 2004 12:29:00 -0500
> From: "Liang Chen" 
> Subject: [mems-talk] SiO2 on GaAs by ebeam evaporator
> To: 
> Message-ID: <001401c3f18d$b3ff94a0$0e870280@sissi>
> Content-Type: text/plain;     charset="iso-8859-1"
>
> Hi,
>
> I'm trying to use SiO2 as etching mask on GaAs, so I use ebeam evaporator to
deposit a thin layer (3000A) SiO2 on it. The deposition rate is 10A/s, and very
stable. But when I dip that sample in AZ400K developer, some part of SiO2 lift
off from the substrate. The substrate is very clean, because when I took it out
from a growth chamber (we're using MBE to grow structures on GaAs substrate), I
immediately put it into evaporator chamber.
>
> So does anyone have such experience and can give me some suggestion on the
following questions?
>
> 1) Is this ebeam evaporation of SiO2 totally unsuitable for etching mask
purpose?
> 2) Do I need special treatment on GaAs surface prior to loading it into
evaporator?
> 3) Do I need post deposition treatment, like thermal curing at, for example,
400C?
>
> Thanks a lot
>
> Lion