EVG bonder for GaAs?

Mahnaz Mansourpour mahnaz at stanford.edu
Mon Mar 16 16:31:37 PDT 2009

Hello Lan,

Sorry that has taken me a while to get back to you.

Here are some info/constrain that I need to bring up to your attention.
Since I have joined  Stanford and made a wide use of EV bonder, GaAs is 
one of the material I have okay once in almost 10 years I have been 
here. I think that we  do not have the capability of supporting this 
material in our bonder because of:

  concern about possible safety issues.
 The Ev bonder does not have exhausted hood or even arsenic MDA  monitors.
If  we okay it, I have to limit the individual's use on day shift and 
the hours  that the staff are here so we can supervise the  usage.

Not mentioning that we have to go with some process modification like 
sandwich the sample between two other wafers, makes bonding more 
complicated.  Last and only time they were 1 cm samples and now we are 
talking 4" wafers. So simply saying   "I will do a low temp does not 
help me".

Here are some info last time i asked expert in Stanford  to help me with 
the matter for you to read.
I am open to the idea but I need to be convinced is the right thing to 
do for our lab.
Please let me what do you think?


GaAs info:
If you're heating GaAs, you really do want to do it under a fume hood
or other exhaust.  Pure arsenic will evaporate at much lower
temperatures than GaAs desorption, even some under 200C, so you have
to make sure your wafers aren't arsenic-rich on the surface if you're
getting up that hot.  You also have to be careful about overshoot:
I've evaporated off thick arsenic caps by mistake in the MBE chamber
when the PID controller overshot 160C.

The desorption of As from GaAs happens at a higher temperature,
somewhere around 300C.  We prebake our wafers before growth at "350C"
or "400C" but this is thermocouple temperature at the heater, not the
actual wafer temperature.  If you need to prevent arsenic desorption
for device reasons, you may need another source of arsenic nearby,
like another gallium arsenide wafer on top of the first.  Of course,
that would complicate wafer bonding.

Oxide blowoff for GaAs happens at 582C.  There are other oxides
(mostly AsOx) which come off at lower temperatures, but the last layer
of gallium oxide comes off very close to 582C, and it's quite clear
from electron diffraction.  We typically grow around 450 for InGaAs,
570-580 for GaAs, and 580-600 for AlGaAs.  These are all real
temperatures (as opposed to thermocouple) but are only approximate.
InP info:
P2 desorb at 300C (I am talking about real temp)
P4 desorb at higher than 650C
In desorb at 500C

InP sublimation temp is 360C
InP oxide blowoff temp is 510C

Lan Zhang wrote:
> Hi Mahnaz,
> I just wondering if GaAs wafers are allowed in the bonder.  I think to 
> do some eutectic bonding with GaAs (~400C and low pressure).
> Thanks
> Lan

More information about the specmat mailing list