Robert,

I think you may be over-complicating this a bit.  The simplest way to
achieve the anodic bond is to place your stack as follows:

Borosilicate Glass
Silicon
Hot-plate

Assuming your hot-plate surface is a conductive metal (many are - if
not, a simple solution is aluminum foil), then by applying a negative
potential to the top glass (possibly by use of a strip of metal or
graphite).  400V is a good choice, although the pulsing you mention is
not really necessary - a straight DC potential is all that is necessary.
My primary recommendation is to use a current-limited power supply for
safety purposes (50mA is more than sufficient to accomplish the bond).
Your hot-plate becomes your ground for the negative potential,
completing the circuit (f using aluminum foil, make sure it is
grounded).

Best Regards,
Chad Brubaker


-----Original Message-----
From: Robert Dean
Sent: Friday, December 02, 2005 3:50 PM
To: [email protected]
Subject: [mems-talk] anodic bonding question

Hello,

I am trying to anodically bond a micromachined Si device to a
micromachined borosilicate glass substrate.  For this application,
alignment is not an issue.  We are using a hot plate with a high voltage
power supply.  The way we typically perform anodic bonding is to attach
Cu wires to the glass substrate and the Si substrate with conductive
epoxy, then heat the stack to 310C and apply a 400V pulse.  This usually
gives very good results.  However, for this application we cannot use
conductive epoxy on the Si device.  Does anyone have a suggestion for
how to electrically connect to the Si and glass substrates without using
conductive epoxy.  Thanks.