In general, the biggest difficulty with bonding of glass to silicon
(especially if we are talking Borofloat glasses rather than fused
silica) is going to be roughness.

For fusion bonding to work at all the surface roughness has to be below
2 nm RMS.  For a hermetic seal, we're talking <0.5 nm(5 A) rms.  This is
typically the polish level of prime silicon. In most cases that I have
seen, the polish level of glasses is lower than this.  We have been
successful bonding Borofloat glass to silicon (and fused silica to
silicon, for that matter), but in most cases, the glass needed to be
CMP'd first.


Best Regards,
Chad Brubaker

EV Group
invent * innovate * implement
Senior Process Technology Engineer - Direct: +1 (480) 305 2414, Main: +1
(480) 305 2400 Fax: +1 (480) 305 2401
Cell: +1 (602) 321 6071
E-Mail: [email protected], Web: www.EVGroup.com

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Joseph Grogan
Sent: Wednesday, February 11, 2009 12:00 PM
To: General MEMS discussion
Subject: Re: [mems-talk] Silicon_Glass Fusion Bonding

Tied in with the clean surface tip is the idea that the surfaces must be

in intimate contact in order to bond. Surface contaminants will of
course screw that up. But even with perfectly clean surfaces you need to

make sure that you are pushing your wafers together in order to make up
for stuff like bow/warp, surface roughness, etc, so that the bonding
surfaces are actually touching. Try putting some significant weight on
your sample or clamping it somehow. The fact that some spots bonded and
others didn't tells me that either you have surface contaminants
interfering, or you aren't getting good contact between the wafers.

Good luck.
Joe Grogan

Shay Kaplan wrote:
> Make sure the substrates are perfectly clean before you attach them
>
> Shay
>